Enkephalin dipeptidyl carboxypeptidase (enkephalinase) activity: selective radioassay, properties, and regional distribution in human brain

Effect of neutral endopeptidase inhibitors on 3H-substance P binding in rat ileum

To determine whether neutral endopeptidase regulates the binding of substance P to the receptors, and if so, what the mechanism is, we determined the effect of neutral endopeptidase inhibitors, thiorphan and phosphoramidon, on specific binding of 3H-substance P to homogenates of rat ileum. Specific binding was of high affinity and was saturable (dissociation constant, KD = 2.4 +/- 0.17 nM and number of maximal binding sites, Bmax = 101.1 +/- 5.5 fmol/mg protein), and the receptor subtype was substance P-P type. Neutral endopeptidase inhibitors increased the specific binding to up to 160% of control (P less than 0.005). Neutral endopeptidase inhibitors prevented the degradation of 3H-substance P during the binding assay and increased the amount of 3H-substance P remaining in the assay system to up to 4.5-fold of control (P less than 0.005), but did not significantly change the KD or Bmax values of specific binding. Protease inhibitors of kininase II, serine proteinases, or thiol proteinases did not significantly change either specific binding or the amount of 3H-substance P remaining in the assay system. We conclude that neutral endopeptidase regulates the binding of substance P to the receptors and that it does so by decreasing the amount of substance P available to the receptors, without significantly changing the affinity or the number of receptors.

Labelling and exploration of the active site of enkephalinase (EC 3.4.24.11) in kidney membranes with [3H]thiorphan as ligand

[3H]Thiorphan, a potent inhibitor of enkephalinase (EC 3.4.24.11), was used to label the enzyme in membranes from rat kidney cortex an to explore its specificity at the active site. [3H]Thiorphan binding occurred reversibly, with low non-specific binding and to a single class of sites. The dissociation constant, determined by either kinetics or saturation studies was approximately 0.4 nM. The ratio of the maximal velocity of enkephalinase with enkephalins as substrates to the maximal binding of [3H]thiorphan was consistent with the catalytic constant of the enzyme. Enkephalinase inhibitors competed with [3H]thiorphan and had inhibitory constants in agreement with the corresponding values derived from measurement of the enzyme catalytic activity, whereas inhibitors of other metallopeptidases were ineffective. The inhibitory potencies of a series of systematically varied oligopeptides regarding [3H]thiorphan binding and enkephalinase activity were also highly correlated. Structure-activity relationships among competitors indicated that the main subsites of enkephalinase are: (1) the hydrophobic pocket in P'1, the requirements of which are best satisfied by aromatic amino acid side chain residues (2) the P'2 subsite, the requirements of which are best satisfied by amino acids with a short, uncharged side chain and a free terminal carboxyl group. This novel binding assay should facilitate the exploration of the active site of enkephalinase and the development of new inhibitors.

Toluene diisocyanate increases airway responsiveness to substance P and decreases airway neutral endopeptidase

Substance P and related tachykinins contribute to the airway hyperresponsiveness caused by toluene diisocyanate (TDI) in guinea pigs. Neutral endopeptidase (NEP) is an important modulator of substance P-induced responses. To test the hypothesis that exposure to TDI would increase responsiveness to substance P by inhibiting activity of this enzyme, we determined the dose of substance P required to increase pulmonary resistance by 200% above baseline (PD200) before and after administration of the pharmacologic inhibitor phosphoramidon in guinea pigs studied 1 h after a 1-h exposure to air or 3 ppm TDI. TDI exposure increased responsiveness to substance P significantly. However, phosphoramidon caused a significantly greater leftward shift of the substance P dose-response curve in air-exposed animals than it did in TDI-exposed animals, so that after phosphoramidon, mean values of PD200 in animals exposed to air or TDI did not differ. Tracheal NEP activity was significantly less after exposure to TDI than after exposure to air, whereas activity in the esophagus was the same in both groups. These results suggest that TDI exposure increases the bronchoconstrictor responsiveness of guinea pigs to substance P, in large part through inhibition of airway NEP.

Enkephalins and learning and memory: a review of evidence for a site of action outside the blood-brain barrier

A series of studies indicate that enkephalins exert dramatic influences on learning and memory in rats and mice, when studied with conditioning tasks that are both negatively and positively motivated. Pharmacological analysis of these enkephalin actions on conditioning suggests that the [leu]enkephalin acts through a delta opioid receptor which is located outside the blood-brain barrier. Control studies indicate that enkephalins do not simply affect the performance of a conditioned response through actions on shock sensitivity or locomotor activity. Characterization of the peripheral enkephalin mechanism that affects behavior suggests an action through an enzymatic system that controls the concentrations of enkephalin present at its receptors in the periphery. This enzymatic mechanism is sensitive to experience, since its activity changes following conditioning, which suggests that it may be a regulatory mechanism for behavior.

Molecular cloning and amino acid sequence of human enkephalinase (neutral endopeptidase)

We have isolated a cDNA clone encoding human enkephalinase (neutral endopeptidase, EC 3.4.24.11) in a lambda gt10 library from human placenta, and present the complete 742 amino acid sequence of human enkephalinase. The human enzyme displays a high homology with rat and rabbit enkephalinase. Like the rat and rabbit enzyme, human enkephalinase contains a single N-terminal transmembrane region and is likely to be inserted through cell membranes with the majority of protein, including its carboxy-terminus, located extracellularly.

Studies of the in vitro human plasma degradation of methionine-enkephalin

1. Incubation of [3H]tyrosine methionine-enkephalin (6 x 10(-9) M final concentration) with human platelet-poor plasma (1:9 ratio to Trizma Base buffer, pH 7.4) results mostly (greater than 95%) in hydrolysis of the tyrosyl-glycine peptide bond. This enzymatic reaction is essentially completed within 90 min, showing a half-life, Km and Vmax of 12.8 +/- 2.5 min, 0.70 +/- 0.01 mM and 17.90 +/- 1.05 mumol/L/min, respectively. These values are comparable to those previously reported for the human plasma degradation of leucine-enkephalin. 2. As expected hydrolysis of the methionine-enkephalin tyrosyl-glycine peptide bond was blocked by the known aminopeptidase inhibitors bestatin and puromycin (IC50 1.2 +/- 0.4 and 4.3 +/- 2.4 microM, respectively) but not by either thiorphan or captopril. 3. Neither the storing (up to 60 days) nor the freezing and thawing (up to ten times during a 60 days periods) significantly changed the above kinetic parameters, showing the stability of the plasma methionine-enkephalin degrading aminopeptidase.

Enkephalinase (EC 3.4.24.11) is highly localized to a striatonigral pathway in rat brain

Autoradiographic visualization of enkephalinase (membrane metalloendopeptidase, EC 3.4.24.11) in sagittal sections of rat brain using a 125I-labelled monoclonal antibody showed the presence of a dense immunoreactivity in a tract joining the striatum to the substantia nigra. Unilateral kainate injections into the striatum elicited a strong ipsilateral decrease in enkephalinase activity and immunoreactivity in both the injected area and substantia nigra, particularly its pars compacta. This demonstrates the presence of enkephalinase all along fibers of a striatonigral pathway.

Regional distribution of mu, delta and kappa opioid receptors in human brains from controls and parkinsonian subjects

The binding properties of mu and delta opioid receptors were investigated in several areas of human brain by using [3H]Tyr-D-Ala-Gly-(Me)Phe-Gly-ol and [3H]Tyr-D-Thr-Gly-Phe-Leu-Thr as respective selective ligands, while the totality of opioid receptors was measured by using [3H]etorphine as a non-selective agonist. Receptor densities were highest in cerebral cortex, amygdala and striatum, and lowest in the substantia nigra (pars compacta). In the different brain areas of patients with Parkinson's disease, the density and the proportion of the various opioid receptors were not significantly different from control subjects.

Molecular cloning and amino acid sequence of rat enkephalinase

cDNA clones encoding rat enkephalinase (neutral endopeptidase, EC 3.4.24.11) have been isolated in lambda gt10 libraries from both brain and kidney mRNAs and the complete 742 amino acid sequence of rat enkephalinase is presented. The enzyme possesses a single transmembrane spanning domain near the N-terminal of the molecule but lacks a signal sequence. Because enkephalinase has it active site located extracellularly and is thus an ectopeptidase, we suggest that the N-terminal transmembrane region of the enzyme anchors the protein in membranes and that the majority of the protein, including the carboxy terminus, is extracellular. Enkephalinase, a zinc-containing metallo enzyme, displays homology with other zinc metallo enzymes such as carboxypeptidase A, B and E, suggesting enzymatic similarities in these enzymes.

New substrates for enkephalinase (neutral endopeptidase) based on fluorescence energy transfer

Novel fluorescent substrates for enkephalinase (neutral endopeptidase; EC 3.4.24.11) have been developed. These new assays are based on the disappearance of energy transfer between a tryptophan or a tyrosine residue and the 5-dimethylaminonaphthalene-1-sulfonyl group (dansyl) in the substrates dansyl-Gly-Trp-Gly or dansyl-Gly-Tyr-Gly upon hydrolysis of their Gly-Trp or Gly-Tyr amide bond by enkephalinase. No significant difference in Km or kcat values were found for dansyl-Gly-Trp-Gly and dansyl-Gly-Tyr-Gly, indicating that, in contrast to thermolysin, the active site of enkephalinase easily accommodates tryptophan residues. Both tryptophan and tyrosine-containing substrates can be used for continuous recording of enkephalinase activity and should prove useful for detailed study of the substrate specificity of this enzyme.

Characterisation of two probes for the localisation of enkephalinase in rat brain: [3H]thiorphan and a 125I-labeled monoclonal antibody

We studied the binding of two radioactive probes, i.e. [3H]thiorphan and a 125I-labeled monoclonal antibody raised against the rabbit kidney enzyme, to enkephalinase (EC 3.4.24.11, membrane metalloendopeptidase) from rat cerebral membranes. [3H]Thiorphan binding at equilibrium to striatal membranes was monophasic with a KD (0.7 nM) and a pharmacology consistent with a selective labeling of the enzyme. The ratio of Vmax/Bmax was in the same range as the Kcat of the enzyme purified from peripheral tissues. The monoclonal antibody immunoprecipitated to a similar extent the solubilised enkephalinase activity and [3H]thiorphan binding sites from striatum. The regional distributions of binding sites for the two probes established either on isolated membranes or autoradiographic sections were highly heterogeneous and similar to that of enkephalinase activity. Hence the two probes appear to label membrane-bound enkephalinase in rat brain but, from a technical point of a view, the 125I-monoclonal antibody is a more sensitive and flexible tool.

Steady-state level and turnover rate of the tripeptide Tyr-Gly-Gly as indexes of striatal enkephalin release in vivo and their reduction during pentobarbital anesthesia

Tyr-Gly-Gly (YGG) was recently shown to be an extraneuronal metabolite of opioid peptides derived from proenkephalin A, formed in brain by the action of "enkephalinase" (membrane metalloendopeptidase, EC 3.4.24.11) and degraded by aminopeptidases. The dynamic state of YGG in mouse striatum was studied by evaluating the changes in its level elicited by inhibitors of these peptidases. Inhibition of YGG synthesis by Thiorphan or acetorphan reduced YGG levels with a t1/2 (mean +/- SEM) of 12 +/- 2 min, indicating an apparent turnover rate (mean +/- SEM) of 18 +/- 2 pmol/mg of protein per hr. An apparent turnover rate of 18 +/- 2 pmol/mg of protein per hr was derived from the rate of YGG accumulation elicited by the aminopeptidase inhibitor bestatin. In addition, accumulation of Tyr-Gly-Gly-Phe-Met (YGGFM) in an extrasynaptosomal fraction after blockade of its degradation by Thiorphan and bestatin occurred at a rate of 18 +/- 3 pmol/mg of protein per hr, which is likely to reflect the rate of enkephalin release in vivo. Hence, the three series of data suggest that striatal enkephalins rapidly turn over--e.g., with a t1/2 in the 1-hr range. Pentobarbital anesthesia reduced by about 60% the rate of YGG accumulation elicited by bestatin and the extrasynaptosomal YGGFM accumulation elicited by Thiorphan and bestatin. This suggests that the activity of striatal enkephalin neurons is depressed during anesthesia. Pentobarbital (and chloral hydrate) did not affect the steady-state level of YGGFM but rapidly reduced that of YGG. Hence, the steady-state levels of YGG seem a reliable index of changes in enkephalin release, and measuring levels of characteristic fragments might therefore provide a general means of evaluating neuropeptide release in vivo.

An immunohistochemical study of endopeptidase-24.11 ("enkephalinase") in the pig nervous system

Endopeptidase-24.11, a plasma membrane ectoenzyme with the ability to hydrolyse a variety of neuropeptides, has been localized in the pig nervous system by an immunoperoxidase technique. The endopeptidase was mapped in cryostat sections of the fore and mid-brain to the following structures: caudate-putamen, globus pallidus, olfactory tubercle, nucleus interpeduncularis and substantia nigra. Endopeptidase-24.11-like immunoreactivity was also found in the pia mater, choroid plexus and ependymal lining of the central canal. In the spinal cord, weak staining was observed in the dorsal horn, but strong staining was found in the dorsal root ganglia and nerve roots. Within the central nervous system, endopeptidase immunoreactivity was confined to gray matter and within the positive areas of the striatum densely staining areas, corresponding to striosomes, were discernible. These well-defined structures were exploited in serial sections to examine the alignment of the enzyme-rich patches of neuropil with correspondingly strong staining for other antigens. A consistent match was observed with a monoclonal antibody to neurofilament protein, but there was a poor correlation with a polyclonal antibody to glial fibrillary acidic protein. Substance P-like and [Leu]enkephalin-like immunoreactivity were also studied in sections adjacent to those stained for the endopeptidase. Good matching between enzyme-rich and peptide-rich areas was observed, but some enkephalin-rich areas did not align with enzyme staining and indeed endopeptidase-rich areas were not necessarily matched with areas rich in either peptide. These findings suggest a neuronal rather than an astrocytic location for endopeptidase-24.11 in the CNS and lend support to the view that it plays a central role in neuropeptide metabolism at membrane surfaces. In the peripheral nervous system, the endopeptidase was located in Schwann cell membranes surrounding dorsal root ganglion cells and nerve fibres, while in the pituitary the main concentration was in the adenohypophysis, where only a proportion of the endocrine cells were found to be immunoreactive.

Enkephalin hydrolysis in homogenates of various absorptive mucosae of the albino rabbit: similarities in rates and involvement of aminopeptidases

The systemic delivery of peptides and proteins from the nasal, rectal, vaginal, and buccal mucosae has been the subject of active investigation. The objective of this study was to determine the pathway and rate of hydrolysis of methionine enkephalin (TGGPM), leucine enkephalin (TGGPL), and [D-Ala2] met-enkephalinamide (TAGPM) in homogenates of these non-oral mucosae relative to the ileal mucosa. Aminopeptidases appeared to contribute over 85% to the hydrolysis of TGGPM and TGGPL, while dipeptidyl peptidase and dipeptidyl carboxypeptidase contributed much less. Overall, TGGPM was somewhat more susceptible to hydrolysis than TGGPL but was 10 times more so than TAGPM. These enkephalins were most rapidly hydrolyzed in the rectal and buccal homogenates, followed by the nasal and then the vaginal homogenates, but the differences in hydrolytic rates were small. Indeed, these rates did not differ substantially from the ileal mucosa, suggesting that the same enzymatic barrier to enkephalin absorption possibly exists in both the oral and the non-oral mucosae.

Enkephalinase activity in both physiological and pathological conditions in man

Among all the enzymes involve din enkephalin degradation "enkephalinase", cleaving the Gly3-Phe4 amide bond, is considered for its specificity with endogenous enkephalins and their receptors. This enzyme, first identified in membrane bound form, has been recently characterized in a soluble one by a new sensitive fluorimetric method substituting the radiometric technique. The possibility to evaluate "enkephalinase" activity in human plasma, amniotic fluid and cerebro spinal fluid (CSF) allows us to investigate its behavior in various physiological and pathological conditions in which alterations of the endogenous opioid system are hypothesized. Our studies were focused on pregnancy, the first period of life, idiopathic headache and opioid addiction. In these conditions "enkephalinase" activity (EKA) generally results increased. In some cases the activity is proportional to the increased amount of substrate, in other cases no correlation seems apparent.

Binding of the bidentate inhibitor [3H]HACBO-Gly to the rat brain neutral endopeptidase "enkephalinase"

The synthesis and binding properties to rat brain tissue of the enkephalinase inhibitor [3H] N-[(R,S)-3-hydroxyaminocarbonyl-2-benzyl-1-oxopropyl]-glycine ([3H]HACBO-Gly, 45 Ci/mmole) is reported. [3H]HACBO-Gly binding to membranes from various rat brain tissue is saturable (KD = 0.4 +/- 0.05 nM) and linearly related to the amount of tissue. Non specific binding is less than 15% of total binding at the KD concentration. The regional distribution of [3H]HACBO-Gly binding and enkephalinase activity are closely correlated with highest levels in striatum and substantia nigra. The efficiency of inhibitors of various peptidases (thiorphan, captopril, bestatin ...) to inhibit [3H]HACBO-Gly binding or enkephalinase activity are similar. These results indicate that [3H]HACBO-Gly binds selectively to enkephalinase. This compound should help to clarify the localization of the enzyme in the CNS.

Comparison of dipeptidyl carboxypeptidase and endopeptidase activities in the three enkephalin-hydrolysing metallopeptidases: "angiotensin-converting enzyme", thermolysin and "enkephalinase"

Angiotensin-converting enzyme (ACE), thermolysin and "enkephalinase", three metallopeptidases cleaving the Gly3-Phe4 amide bond of enkephalins, were compared regarding substrate specificity and effects of butanedione, an arginyl-directed reagent. The hydrolysis of enkephalins and analogues was more affected by the nature of P1 and P2 residues in the case of thermolysin than in those of ACE or "enkephalinase"; amidation of the C-terminal carboxylate decreased drastically the hydrolysis by ACE but only marginally by thermolysin and the effect was intermediate for "enkephalinase". With adequate model substrates, the ratio of dipeptidylcarboxypeptidase to tripeptidylcaroxypeptidase (endopeptidase) activities were of 25 for ACE, 3 for "enkephalinase" and only 0.3 for thermolysin. Finally a butanedione treatment increased thermolysin activity, but abolished ACE activity; it reduced "enkephalinase" activity by 80% when measured with a free C-terminal carboxylate enkephalin analogue but only slightly with the corresponding amidated derivative. A critical role of an Arg residue in ACE and, to a lesser extent, in "enkephalinase" (but not in thermolysin) is suggested to be responsible for the preferential dipeptidylcarboxypeptidase activity of these two enzymes.

Synthesis and biological activity of linear and cyclic enkephalins modified at the Gly3-Phe4 amide bond

As part of our continuing effort to define structure-activity relationships for enkephalin and design enzymatically resistant analogs, we report the synthesis and biological activities of linear and cyclic enkephalin analogs modified at the Gly3-Phe4 amide bond. The partial retro-inverso enkephalin analog Tyr-D-Ala-gGly-(R,S)-mPhe-Leu-NH2 and its cyclic counterpart, Tyr-cyclo[D-A2 bu-gGly-(R,S)-mPhe-Leu-], were synthesized as diastereomeric mixtures using solution methodology. The racemic benzylmalonate allowed the linear analog to be synthesized by fragment coupling at the reversed bond. Cyclization of the second analog was carried out at high concentration, eliminating formation of polymer by the use of an insoluble base. All gem-diaminoalkyl residues were prepared by conversion of peptidyl amides with benzene iodonium bis(trifluoroacetate). Diastereomers of both compounds were separable by reverse phase HPLC but those of the linear compound racemized rapidly under conditions of testing and were therefore tested together. All analogs tested had activities ranging from 6 to 14% of the activity of Leu enkephalin, indicating that the Gly3-Phe4 amide bond is important, though not crucial, for receptor binding.

Characterization of membrane-bound aminopeptidases from rat brain: identification of the enkephalin-degrading aminopeptidase

Rat brain aminopeptidase activity was solubilized from membranes by incubation with thiols. This novel procedure resulted in the release of the same two aminopeptidases (MI and MII) previously shown to be solubilized by the nonionic detergent Triton X-100. The solubilized aminopeptidases MI and MII were resolved by ion-exchange chromatography and further purified by hydroxylapatite chromatography. Aminopeptidase MI was shown to hydrolyze only the beta-naphthylamides of arginine and lysine whereas aminopeptidase MII exhibited a broad specificity with respect to amino acid beta-naphthylamides. Only aminopeptidase MII hydrolyzed Leu-enkephalin at a significant rate, indicating that this enzyme can account for the membrane-bound enkephalin aminopeptidase activity. The enkephalin-degrading aminopeptidase is potently inhibited by opioid (alpha-neo-endorphin and dynorphin) as well as nonopioid (substance P, somatostatin, and angiotensin I) peptides in the range of 0.2-2.0 microM. The regional distribution of aminopeptidases MI and MII in rat brain are rather different, with aminopeptidase MII distribution more closely paralleling the distribution of opiate receptors.

Assessment of the role of "enkephalinase" in cholecystokinin inactivation

Cholecystokinin octapeptide and the C-terminal tetrapeptide are hydrolysed by a highly purified preparation of "enkephalinase" (EC 3.4.24.11). In both cases the Asp-PheNH2 bond is hydrolysed and the Gly4-Trp5 bond of the octapeptide is also cleaved, though more slowly. Evaluated from the appearance of Phe-NH2, the Km for the hydrolysis of the octapeptide by the purified peptidase is 57 microM and that for the tetrapeptide 65 microM. The apparent affinities of these peptides for the enzyme in striatal membranes are similar. The importance of this hydrolysis in the inactivation of endogenous cholecystokinin was assessed by studying the fate of cholecystokinin immunoreactivity released from slices of rat cerebral cortex and striatum by depolarization with potassium. In the absence of any peptidase inhibitor only 16% of the peptide released from the tissue was recovered in immunoreactive form in the medium, indicating that endogenous cholecystokinin octapeptide is, like other neuropeptides, rapidly and extensively hydrolysed following release. Selective inhibition of "enkephalinase" by Thiorphan (DL-3-mercapto-2-benzylpropanoyl glycine) did not significantly alter the recovery from slices of cerebral cortex and had only a very slight effect in the case of striatal slices. This suggests that, while cholecystokinin octapeptide is a substrate for "enkephalinase", this enzyme plays a less important (if any) role in the inactivation of endogenous cholecystokinin than for the opioid peptides.

Enkephalinergic markers in substantia nigra and caudate nucleus from Parkinsonian subjects

Marked reductions in opiate receptor binding (-42%), "enkephalinase" activity (-39%), and Met5-enkephalin levels (-72%) accompanied the well-established dopamine depletion in the substantia nigra pars compacta of Parkinsonian subjects. In contrast, enkephalinergic markers were not significantly modified in caudate nucleus.

Bidentate peptides: highly potent new inhibitors of enkephalin degrading enzymes

Three series of bidentates bearing an hydroxamic or an N-Acyl-N-hydroxy amino group on structures related to Phe-Gly or Phe-Ala exhibit strong inhibitory potency against purified enkephalinase with IC50 values in the 4 to 15 nM range. As with thiol-containing inhibitors, such as thiorphan, the most active compounds are those in which a methylene spacer separates the benzyl P1' moiety from the Zn coordinating residue. Formation of a bidentate complex with the metal enzyme is clearly demonstrated by a loss of potency of three order of magnitude following the removal of one component of the bidentate group. All the compounds studied are unable to interact with angiotensin converting enzyme (IC50 greater than 10,000 nM). Moreover, compounds of the general formula HONHCO-CH2-CH(CH2 phi)-CONH-CH(R)-COOH belonging to the most active series of enkephalinase blockers (IC50 approximately 4 nM) behave also as highly potent and competitive inhibitors (IC50 approximately 10 nM) of a Tyr-Gly releasing dipeptidylaminopeptidase purified from rat brain. The pure steroisomer [(R)-3-(N-hydroxy)carboxamido-2-benzylpropanoyl]-L-alanine designated kelatorphan, exhibits also a relatively good inhibitory potency against aminopeptidases (IC50 approximately 10 microM) and can be considered as the first virtually complete inhibitor of enkephalin metabolism. This very interesting property of inhibiting all three enzymes of enkephalin metabolism could enhance the required selectivity for a possible clinical use of these inhibitors as new analgesic and psychoactive drugs.

A highly sensitive fluorometric assay for "enkephalinase," a neutral metalloendopeptidase that releases tyrosine-glycine-glycine from enkephalins

A fluorogenic peptide, dansyl-D-Ala-Gly-Phe(pNO2)-Gly (DAGNPG), was synthesized as a selective substrate for the neutral metalloendopeptidase (EC 3.4.24.11) involved in enkephalin metabolism. This enzyme, designated "enkephalinase," cleaves the Gly-Phe(pNO2) peptide bond of DAGNPG (V = 0.65 mumol/mg protein/min and Km = 45 microM) leading to a fluorescence increase related to the disappearance of intramolecular quenching of the dansyl fluorescence by the nitrophenyl residue. This change was used for quantitative measurements of "enkephalinase" activity in different tissues and determination of inhibitory potency of various compounds. The substrate is not cleaved by aminopeptidase or dipeptidylaminopeptidase activities and the assay itself is rapid, convenient, and sensitive.

Reaction of opioid peptides with neutral endopeptidase ("enkephalinase")

The kinetics of the reactions of nine opioid peptides with the neutral endopeptidase ("enkephalinase") activities of human kidney, rat kidney, and rat brain have been determined. These opioid peptides can be divided into two classes, those that are good inhibitors of Leu5-enkephalin hydrolysis (Ki less than 75 microM) and good substrates for the enzyme, and those that are poor inhibitors (Ki greater than 500 microM) and are not substrates for the enzyme. The former group includes Leu5-enkephalin, Met5-enkephalin, Met5-enkephalin-Arg6-Phe7, beta-lipotropin, and gamma-endorphin, while the nonreactive opioid peptides include alpha-neo-endorphin, beta-neo-endorphin, dynorphin, and beta-endorphin. These results suggest that those peptides containing the Met5-enkephalin sequence are more reactive than those containing the Leu5-enkephalin sequence. The lack of specificity of this neutral endopeptidase indicates that it may function in the degradation of a variety of biologically active peptides.

Analgesic effects of kelatorphan, a new highly potent inhibitor of multiple enkephalin degrading enzymes

Kelatorphan, [(R)-3-(N-hydroxy)-carboxamido-2-benzylpropanoyl]-L-alanine, represents the first virtually complete inhibitor of enkephalins metabolism with KI = 1.4 nM against enkephalinase, KI = 2 nM against the Gly2 -Gly3 cleaving dipeptidylaminopeptidase and KI = 7 microM on aminopeptidase activity. The analgesic effect of [Met5]enkephalin was potentiated 50000 times (ED50 approximately 10 ng) by intracerebroventricular co-administration in mice of kelatorphan (50 micrograms). This effect was significantly higher than that produced by bestatin (50 micrograms) + thiorphan (50 micrograms). Kelatorphan alone was at least two-fold more potent as analgesic than the above mixture of inhibitors.

Proteases of human brain

Growing appreciation of the multiple functions of proteolytic enzymes in intracellular protein degradation and post-translational modification, in the release of biologically active macromolecules and peptides from precursors and in cellular protein regulation and quality control has stimulated interest in proteases in neurobiology and neuropathology. In this article, the proteinases and peptidases thus far studied in the human central nervous system are reviewed with respect to their enzymology, anatomical and cytological distributions and contributions to neurological and psychiatric disease states. Though information concerning brain proteases in man is fragmentary, it suffices to establish the importance of these complex systems for advancing knowledge of human cerebral function in health and disease.

Differences in the structural requirements for selective interaction with neutral metalloendopeptidase (enkephalinase) or angiotensin-converting enzyme. Molecular investigation by use of new thiol inhibitors

Despite the similarities in their mechanism of action, the structural requirements for selective interaction with angiotensin-converting enzyme or enkephalinase are different. Inhibitory potency of a series of new mercaptoalkanoyl amino acids were determined on pure angiotensin-converting enzyme (EC 3.4.15.1) from porcine plasma and on neutral metalloendopeptidase (EC 3.4.24.11) purified from rat brain. This latter enzyme, first designated as enkephalinase, seems to be synaptically involved in the degradation of enkephalins. All tested compounds, whose design was based on the classical active-site model of metallopeptidases, are reversible and competitive inhibitors of both enzymes. Owing to the remarkable similarity in the general topology of metallopeptidases, the differences in optimal binding requirements to enkephalinase and angiotensin-converting enzyme were interpreted from crystallographic studies on related enzymes such as thermolysin and carboxypeptidase A. The large size of the S'1 subsite of enkephalinase allows efficient binding (Ki approximately equal to 2-30 nM) of aromatic and bulky hydrophobic residues such as a cyclohexyl ring. In contrast, a methyl group in position P'1 favors inhibitory potency against angiotensin-converting enzyme while a cyclohexyl ring leads to a complete loss of activity. This feature could mean that optimal binding of the Zn atom present in the catalytic site is a more stringent requirement in angiotensin-converting enzyme than in enkephalinase. An increase in the size of the P'2 component of thiol inhibitors potentiates the affinity for angiotensin-converting enzyme without a significant change on enkephalinase. Finally, methylation of the ultimate amide bond of inhibitors produces a 30-fold decrease in potency towards enkephalinase but does not affect the binding of angiotensin-converting enzyme. These findings allow a rational design of selective inhibitors of enkephalinase, an essential prerequisite for their possible clinical use as new analgesic and psycho-active agents.

Processing and degradation of met-enkephalin by peptidases associated with rat brain cortical synaptosomes

The generation of met-enkephalin (Tyr1-Gly2-Gly3-Phe4-Met5) from met-enkephalin-Arg6-Phe7 and subsequent degradation of the liberated peptides to the free amino acids by rat brain cortical synaptosomes in vitro was demonstrated by HPLC and amino acid analyses. Kinetic measurements of the individual steps of met-enkephalin processing and degradation upon incubation with synaptosomes revealed the following sequence of cleavage: 1. Hydrolysis of the Met5-Arg6 peptide bond, generating met-enkephalin and the dipeptide Arg-Phe. Captopril and EDTA inhibit this reaction. 2. Hydrolysis of the Tyr1-Gly2 peptide bond, generating Tyr and a tetrapeptide. Puromycin (ID50 = 5 X 10(-5) M) and parahydroxymercuribenzoate (ID50 = 5 X 10(-4) M) inhibit this reaction. 3. Hydrolysis of the Gly3-Phe4 peptide bond. Parahydroxymercuribenzoate (ID50 = 5 X 10(-4) M) inhibits this reaction completely. 1 mmol liter-1 Puromycin does not inhibit this reaction. 4. Hydrolysis of the Phe4-Met5 peptide bond. 5. Hydrolysis of the Gly2-Gly3 peptide bond. The pH optimum of all cleavage reactions was found to be around 7.8.

A dipeptidyl carboxypeptidase in brain synaptic membranes active in the metabolism of enkephalin containing peptides

A dipeptidyl carboxypeptidase activity has been localized in synaptic plasma membranes which have been prepared from isolated rat brain cortical synaptosomes. The specificity of this proteolytic activity towards various synthetic and biological active peptides is compared to the peptidase activities of intact synaptosomes. In contrast to the synaptosomal peptidases which are capable of cleaving all peptide bonds of Met-enkephalin-Arg6-Phe7 the peptidase activity associated with the synaptic plasma membrane exclusively hydrolyses a dipeptide from the carboxyl terminus of all hepta- and hexapeptides tested. The fact that this dipeptidyl carboxypeptidase does not cleave the Gly3-Phe4 peptide bond of Met-enkephalin suggests that this enzyme is different from "enkephalinase". The synaptic membrane dipeptidyl carboxypeptidase is inhibited by metal chelating agents and thiols but is not affected by compounds known to inhibit serine proteases, thermolysin and "enkephalinase".

Complete differentiation between enkephalinase and angiotensin-converting enzyme inhibition by retro-thiorphan

Thiorphan, N-[(R,S)-3-mercapto-2-benzylpropanoyl]glycine is a highly potent inhibitor (Ki = 3.5 nM) of "enkephalinase," a metalloendopeptidase cleaving the Gly-Phe bond (positions 3 and 4) of enkephalins in brain tissue. In accordance with this property, thiorphan displays antinociceptive activity after systemic administration. However, thiorphan also inhibits to a lesser extent (Ki = 140 nM) the widely distributed angiotensin-converting enzyme, a carboxydipeptidase implicated in blood pressure regulation. Therefore, in view of an eventual clinical use of enkephalinase inhibitors, it was very important to develop fully specific compounds. Such derivatives were obtained taking into account that N-methylation of the ultimate amide bond of dipeptides strongly decreases enkephalinase affinity without affecting angiotension-converting enzyme recognition, whereas retro-inversion of the amide bond leads to the inverse effect. Thus, the retro-inverso dipeptide (R)-H2N-CH(CH2 phi)-NHCO-CH2-CO2H exhibits an inhibitory potency on enkephalinase (IC50 approximately equal to 12 muM) close to that of the natural dipeptide L-Phe-Gly (IC50 approximately equal to 3 muM). This result shows the topological analogy between the crucial components involved in enkephalinase recognition both in active dipeptides and structurally related retro-inverso isomers. Taking into account these observations, retro-thiorphan, (R,S)-HS-CH2-CH-(CH2 phi)-NHCO-CH2-COOH, was prepared. As compared to thiorphan, the retro isomer is 50% as potent (Ki = 6 nM) on enkephalinase but displays a drastic loss of potency on angiotension-converting enzyme (IC50 greater than 10,000 nM). This specificity was interpreted as a consequence of differences in the stereochemical constraints involving enzyme-inhibitor hydrogen bonding. This hypothesis is supported by reported crystallographic studies on related enzymes such as thermolysin and carboxypeptidase A. As expected, retro-thiorphan exhibits about the same analgesic potency as thiorphan on the hot plate and writhing tests in mice. Therefore, the topological concept of retro-inverso isomers could be extended to other enkephalinase inhibitors, allowing the design of potent and highly selective compounds occurring as new classes of analgesic and psychoactive agents.

Tyrosine hydroxylase and methionine-enkephalin in the human mesencephalon. Immunocytochemical localization and relationships

The immunocytochemical localization of tyrosine hydroxylase (TH) and methionine-enkephalin (met-enkephalin) was determined at two representative caudal and rostral levels of the human mesencephalon. Four main groups of catecholaminergic neurons were delineated, situated in the substantia nigra and the lateral, ventromedial and dorsomedial tegmentum, extending over several cytoarchitectonic divisions. They matched fairly well the dopaminergic cell groups described in monkey midbrain. TH-like immunoreactivity and neuromelanin were closely related in neurons of substantia nigra, but less so in the other groups. A widespread met-enkephalinergic innervation was observed in most areas containing catecholaminergic neurons. It followed a characteristic pattern: homogeneous and very dense in the lateral and posterior portions of substantia nigra; patchy and less dense in the other areas, the medio-ventral and periaqueductal gray being only sparsely innervated, in contrast to observations in rodents. Dopaminergic cell bodies surrounded by met-enkephalinergic varicosities were seen in most groups, particularly in the lateral substantia nigra and medioventral tegmentum. The topography of met-enkephali-like immunoreactive terminals in the substantia nigra was reminiscent of the distribution of neostriatal and pallidal afferents.