Expression and characterisation of human and rat CRF2alpha receptors.

Rat and human CRF2alpha receptors were expressed in CHO-pro5 cells and stable cell lines generated. Each receptor was characterised using [125I][tyr0]sauvagine and results compared to CRF1 receptors expressed in the same parental cell line. Under identical assay conditions, [125I][tyr0]sauvagine labelled both CRF1 and CRF2alpha receptors with high affinity. The level of expression varied from 103 fmol/mg membrane protein to 1842 fmol/mg membrane protein (rat CRF1 receptors and rat CRF2 receptors, respectively). It was possible to establish robust scintillation proximity assays (SPA) using wheat germ agglutinin (WGA) SPA beads to trap membrane protein. The success of the SPA assay format was dependent on the level of receptor expression observed. The rank order of affinities of a series of peptide CRF receptor agonists and antagonists was similar to that described in the literature for the two receptor subtypes as determined using radioligand binding and cAMP accumulation. No pharmacological differences were apparent between rat and human cloned receptors with the exception of alpha-helical CRF-(9-41). This peptide exhibited 10-fold higher affinity for rat CRF2alpha receptors as compared to human CRF2alpha receptors. PD 173307, PD 173602 and PD 174239 exhibited high affinity and selectivity for human CRF1 receptors, and as such represent useful tools for probing CRF receptor function.

Cloning and deletion mutagenesis of the alpha2 delta calcium channel subunit from porcine cerebral cortex. Expression of a soluble form of the protein that retains [3H]gabapentin binding activity.

The anti-epileptic, anti-hyperalgesic, and anxiolytic agent gabapentin (1-(aminomethyl)-cyclohexane acetic acid or Neurontin) has previously been shown to bind with high affinity to the alpha2delta subunit of voltage-dependent calcium channels (Gee, N. S. , Brown, J. P., Dissanayake, V. U. K., Offord, J., Thurlow, R., and Woodruff, G.N. (1996) J. Biol. Chem. 271, 5768-5776). We report here the cloning, sequencing, and deletion mutagenesis of the alpha2delta subunit from porcine brain. The deduced protein sequence has a 95.9 and 98.2% identity to the rat and human neuronal alpha2 delta sequences, respectively. [3H]Gabapentin binds with a KD of 37.5 +/- 10.4 nM to membranes prepared from COS-7 cells transfected with wild-type porcine alpha2 delta cDNA. Six deletion mutants (B-G) that lack the delta polypeptide, together with varying amounts of the alpha2 component, failed to bind [3H]gabapentin. C-terminal deletion mutagenesis of the delta polypeptide identified a segment (residues 960-994) required for correct assembly of the [3H]gabapentin binding pocket. Mutant L, which lacks the putative membrane anchor in the delta sequence, was found in both membrane-associated and soluble secreted forms. The soluble form was not proteolytically cleaved into separate alpha2 and delta chains but still retained a high affinity (KD = 30.7 +/- 8.1 nM) for [3H]gabapentin. The production of a soluble alpha2delta mutant supports the single transmembrane model of the alpha2 delta subunit and is an important step toward the large-scale recombinant expression of the protein.

Identification of novel ligands for the gabapentin binding site on the alpha2delta subunit of a calcium channel and their evaluation as anticonvulsant agents.

As part of a program to investigate the structure-activity relationships of Gabapentin (Neurontin), a number of alkylated analogues were synthesized and evaluated in vitro for binding to the Gabapentin binding site located on the alpha2delta subunit of a calcium channel. A number of other bridged and heterocyclic analogues are also reported along with their in vitro data. Two compounds showing higher affinity than Gabapentin were selected for evaluation in an animal model of epilepsy. One of these compounds, cis-(1S,3R)-(1-(aminomethyl)-3-methylcyclohexyl)acetic acid hydrochloride (19), was shown to be effective in this model with a profile similar to that of Gabapentin itself.

A summary of mechanistic hypotheses of gabapentin pharmacology.

Although the cellular mechanisms of pharmacological actions of gabapentin (Neurontin) remain incompletely described, several hypotheses have been proposed. It is possible that different mechanisms account for anticonvulsant, antinociceptive, anxiolytic and neuroprotective activity in animal models. Gabapentin is an amino acid, with a mechanism that differs from those of other anticonvulsant drugs such as phenytoin, carbamazepine or valproate. Radiotracer studies with [14C]gabapentin suggest that gabapentin is rapidly accessible to brain cell cytosol. Several hypotheses of cellular mechanisms have been proposed to explain the pharmacology of gabapentin: 1. Gabapentin crosses several membrane barriers in the body via a specific amino acid transporter (system L) and competes with leucine, isoleucine, valine and phenylalanine for transport. 2. Gabapentin increases the concentration and probably the rate of synthesis of GABA in brain, which may enhance non-vesicular GABA release during seizures. 3. Gabapentin binds with high affinity to a novel binding site in brain tissues that is associated with an auxiliary subunit of voltage-sensitive Ca2+ channels. Recent electrophysiology results suggest that gabapentin may modulate certain types of Ca2+ current. 4. Gabapentin reduces the release of several monoamine neurotransmitters. 5. Electrophysiology suggests that gabapentin inhibits voltage-activated Na+ channels, but other results contradict these findings. 6. Gabapentin increases serotonin concentrations in human whole blood, which may be relevant to neurobehavioral actions. 7. Gabapentin prevents neuronal death in several models including those designed to mimic amyotrophic lateral sclerosis (ALS). This may occur by inhibition of glutamate synthesis by branched-chain amino acid aminotransferase (BCAA-t).

Isolation of the [3H]gabapentin-binding protein/alpha 2 delta Ca2+ channel subunit from porcine brain: development of a radioligand binding assay for alpha 2 delta subunits using [3H]leucine.

The novel antiepileptic agent gabapentin (Neurontin) binds with high affinity to the alpha 2 delta subunit of a voltage-dependent Ca2+ channel. We report here a simple purification scheme for detergent-solubilized alpha 2 delta subunits from porcine brain. This involves sequential chromatography on Q-Sepharose, Cu(2+)-charged iminodiacetic acid-Sepharose, wheat germ lectin-agarose, and Mono Q. The purified protein was essentially homogeneous by SDS-polyacrylamide gel electrophoresis with a subunit Mr of 145,000. Using [3H] gabapentin as the radiolabeled tracer and (S)-3-isobutyl gamma-aminobutyric acid to define nonspecific binding, the overall purification factor was 2760-fold and the apparent yield 26.6%. We also developed and validated a novel binding assay for alpha 2 delta Ca2+ channel subunits using the ligand pair L-[3H]leucine/L-isoleucine. Even in binding assays of crude brain membrane fractions, [3H]leucine proved to be remarkably stable and specific for the alpha 2 delta Ca2+ channel subunit. [3H]Leucine offers several advantages over custom-labeled [3H]gabapentin: it has a higher specific activity, is relatively inexpensive, and is available from commercial sources.

Spermine modulation of specific [3H]-gabapentin binding to the detergent-solubilized porcine cerebral cortex alpha 2 delta calcium channel subunit.

1. Recent studies have identified the [3H]-gabapentin-binding protein, purified from porcine cerebral cortical membranes, as the alpha 2 delta subunit of voltage-sensitive calcium channels (Gee et al., 1996). The present study investigates the influence of the polyamine spermine on specific [3H]-gabapentin binding to detergent-solubilized porcine cerebral cortical membranes. 2. Spermine, spermidine, 1,10 diaminodecane, Mg2+ and Zn2+, all divalent cations, displaced [3H]-gabapentin binding to detergent-solubilized membranes in a concentration-dependent manner with a maximal inhibition of 65-75%. Radioligand binding studies showed that spermine did not directly interact with the [3H]-gabapentin-binding site. Spermine inhibited [3H]-gabapentin binding by interacting with a polyamine-sensitive allosteric site on the membrane protein. The steep concentration-dependence of spermine inhibition of [3H]-gabapentin binding may suggest multi-site co-operativity. 3. Prolonged dialysis of cerebral cortical membranes and Tween 20-solubilized membranes resulted in a > 2.0 fold increase in [3H]-gabapentin binding. The increase in binding was due to the removal of a heat stable, low molecular weight (< 12,000Da) endogenous molecule which influences [3H]-gabapentin binding competitively. 4. Dialysis of detergent-solubilized cerebral cortical membranes also resulted in a decrease in the maximum inhibition of [3H]-gabapentin binding by spermine. Since the rates of the increase in [3H]-gabapentin binding and the loss of the ability of spermine to inhibit [3H]-gabapentin binding on dialysis were different it was inferred that a second endogenous ligand was removed during dialysis. 5. During initial steps of purification of the [3H]-gabapentin-binding protein there was a decrease in the maximum inhibition of [3H]-gabapentin binding by spermine. The loss of the second endogenous molecule during initial purification would reasonably explain the reduction in inhibition of binding by spermine. However, spermine stimulation of [3H]-gabapentin binding to material that eluted from the gel-filtration column later in the purification scheme does not appear to be due to removal of a dialysable endogenous factor or to the dissociation of other calcium channel subunit(s). 6. Adding back dialysate, before or after boiling, to detergent solubilized membranes resulted in a dose-dependent restoration of the inhibition of [3H]-gabapentin binding and of the maximal inhibition [3H]-gabapentin binding by spermine. This result is consistent with the re-addition of two endogenous heat stable ligands. 7. The findings that [3H]-gabapentin binding to the pure alpha 2 delta subunit was stimulated by spermine indicates that the alpha 2 delta subunit of voltage-sensitive calcium channels bears a modulatory spermine site. Such a spermine site has not been identified before. Spermine stimulation of [3H]-gabapentin binding to the purified protein was reversed to inhibition after adding back dialysate. Thus the inhibitory spermine effect in membranes is also probably due to one or more modulatory sites on the alpha 2 delta subunit.

The novel anticonvulsant drug, gabapentin (Neurontin), binds to the alpha2delta subunit of a calcium channel.

Gabapentin (1-(aminomethyl)cyclohexane acetic acid; Neurontin) is a novel anticonvulsant drug, with a mechanism of action apparently dissimilar to that of other antiepileptic agents. We report here the isolation and characterization of a [3H]gabapentin-binding protein from pig cerebral cortex membranes. The detergent-solubilized binding protein was purified 1022-fold, in a six-step column-chromatographic procedure, with a yield of 3.9%. The purified protein had an apparent subunit Mr of 130,000, and was heavily glycosylated. The partial N-terminal amino acid sequence of the Mr 130,000 polypeptide, EPFPSAVTIK, was identical to that reported for the alpha2delta subunit of the L-type Ca2+ channel from rabbit skeletal muscle (Hamilton, S. L., Hawkes, M. J., Brush, K., Cook, R., Chang, R. J., and Smilowitz, H. M. (1989) Biochemistry 28, 7820-7828). High levels of [3H]gabapentin binding sites were found in membranes prepared from rat brain, heart and skeletal muscle. Binding of [3H]gabapentin to COS-7 cells transfected with alpha2delta cDNA was elevated >10-fold over controls, consistent with the expression of alpha2 delta protein, as measured by Western blotting. Finally, purified L-type Ca2+ channel complexes were fractionated, under dissociating conditions, on an ion-exchange column; [3H]gabapentin binding activity closely followed the elution of the alpha2 delta subunit. [3H]Gabapentin is the first pharmacological agent described that interacts with an alpha2delta subunit of a voltage-dependent Ca2+ channel.

[3H]gabapentin may label a system-L-like neutral amino acid carrier in brain.

The ability of large neutral amino acids to interact with a site in mouse and pig brain labelled by [3H]gabapentin was examined. As previously described for rat tissue, [3H]gabapentin bound to synaptic plasma membranes prepared from mouse or pig cerebral cortex with high affinity (Kinetically derived KD = 14 and 17 nM for mouse and pig, respectively). Equilibrium binding in each species was inhibited by gabapentin and a range of large neutral amino acids. L-leucine (IC50 = 80 nM), L-isoleucine (IC50 = 72 nM), L-norleucine (IC50 = 40 nM) and L-methionine (IC50 = 50 nM) were the most potent of those tested. Binding was also inhibited by L-phenylalanine (IC50 = 380 nM), L-valine (IC50 = 310 nM) and the selective system-L substrate 2-amino-2-carboxy-bicycloheptane (IC50 = 420 nM) but not by the sodium-dependent System-A substrate methylaminoisobutyric acid. The presence of a submaximal concentration of leucine reduced [3H]gabapentin binding affinity but did not affect the maximum number of binding sites, suggesting a competitive interaction between leucine and the binding protein. The results suggest [3H]gabapentin may label a site in brain that resembles the large neutral amino acid transporter described in other tissues.

The effects of substrates, products and other ligands on the susceptibility of inositol monophosphatase to proteolysis by endoprotease lys-C.

Inositol monophosphatase is cleaved by endoprotease lys-C at a single site (Lys36-Ser37). The rate of proteolysis is greatly reduced in the presence of substrate (D,L-Ins(1)P) and Mg2+, and less so in the presence of Pi and Mg2+, consistent with protection of the susceptible bond in the E-P or E-Pi states of the enzyme. Potentiation by Li+ of the protection afforded by a substrate analogue, 1S-phosphoryloxy-2R,4S-dihydroxycyclohexane, and Mg2+ supports the idea that Li+ binds to the E-P state.

Limited proteolysis and 'in vitro' mutagenesis of bovine brain inositol monophosphatase identifies an N-terminal region important for activity.

Bovine brain inositol monophosphatase is rapidly cleaved by endoprotease lys-C at a single site in the absence of SDS. Further sites are revealed only after prolonged incubation with high concentrations of protease. The initial cleavage occurs near one end of the enzyme, generating an N-terminally-derived 36-residue peptide, which is blocked, and a large 28 kDa fragment bearing a free N-terminus. The start sequence of this fragment was found to be Xaa-Ser-Pro-Ala-Asp-Leu-Val, consistent with the cDNA sequence, and Lys-36-Ser-37 was identified as the cleavage site. The activity of the cleaved enzyme was markedly decreased to 3% of that of the native enzyme, although its dimeric structure was preserved. The 36-residue peptide was not covalently associated with the large fragment after proteolytic cleavage, although the possibility of non-covalent association could not be excluded. Finally, the epitope for the inhibitory monoclonal antibody G-2A4 [Gee, Howell, Ryan & Ragan (1989) Biochem J. 264. 793-798] was found to lie proximal to the endoprotease lys-C cleavage site. In vitro mutagenesis further mapped the epitope for monoclonal antibody G-2A4 to residues around Cys-8 of the enzyme. These results suggest that the N-terminal region of the enzyme is important for activity.

Modification of myo-inositol monophosphatase by the arginine-specific reagent phenylglyoxal.

myo-Inositol monophosphatase is inhibited by the arginine-specific reagent phenylglyoxal. The rate of inactivation is decreased in the presence of Pi, a competitive inhibitor of the enzyme. The effect of Pi is dependent on the presence of Mg2+, but is unaffected by Li+, an uncompetitive inhibitor. In the absence of Mg2+, the substrate, Ins(1)P, binds to the enzyme but is not converted into products, and affords only a small degree of protection against inactivation by phenylglyoxal. Li+ had no further effect under these conditions, but in the presence of Mg2+ caused a marked potentiation of the protective effect of substrate alone. In the absence of substrate, Li+ had no effect on activation by phenylglyoxal. Incorporation of 14C-labelled phenylglyoxal showed that inactivation was associated with modification of a single arginine residue per monomer in the dimeric enzyme. These findings support a mechanism in which Li+ inhibits monophosphatase by trapping a phosphorylated enzyme intermediate and preventing its hydrolysis.

A monoclonal antibody to bovine brain inositol monophosphatase. Immunoaffinity purification of the brain and kidney enzymes and evidence for their structural identity.

A monoclonal IgG2b(K) antibody, G-2A4, has been generated against bovine brain myo-inositol monophosphatase (EC 3.1.3.25). The identity of the antigen recognized by the antibody was established by using e.l.i.s.a. and Western blotting procedures, and by immunoprecipitation of enzyme activity from crude brain supernatant. In addition, the hydrolysis of Ins1P by crude brain extract was inhibited by up to 83% by the pure antibody. Under identical conditions, the hydrolysis of Ins(1,4)P2 was unaffected. An immunoadsorbent column containing monoclonal antibody G-2A4 covalently attached to CNBr-activated Sepharose 4B has been used for rapid purification of the brain enzyme. Elution conditions have been optimized to allow isolation of the enzyme in high yield (54%) with full retention of column-binding capacity. The enzyme was electrophoretically homogeneous, Mr 30,000 and of higher specific activity than that purified conventionally. Chromatography of the pure enzyme on high resolution ion-exchange columns revealed some charge heterogeneity, possibly indicative of some type of post-translational modification. The immunoadsorbent column has also been used to purify the bovine kidney cortex enzyme to homogeneity. Partial proteolytic fragmentation patterns of the brain and kidney enzymes using endoprotease glu-C were identical, suggesting that they are almost certainly products of the same gene.

Evidence for at least four different inositol bisphosphatases in bovine brain.

Bovine brain supernatant contains at least four enzymes capable of hydrolysing inositol bisphosphates. These activities may be distinguished on the basis of their metal, salt and pH dependence, sensitivity to Li+ ions and thiol-modification reagents, and on their molecular sizes. In addition to Li+-sensitive Ins(1,4)P2/Ins(1,3,4)P3 1-phosphatase [Gee et al. (1988) Biochem. J. 253, 777-782] which has an absolute requirement for Mg2+, two (Li+-insensitive and Mg2+-independent) phosphatases, capable of hydrolysing Ins(3,4)P2 and Ins(1,3)P2, respectively, have been identified. Both enzymes were inhibited by only moderate concentrations of salt, although for the former there was no obvious correlation between inhibitory potency and either the nature of the anion/cation or the ionic strength of the buffer. Ins(3,4)P2 phosphatase had a pH optimum of 7.6 and this activity could be resolved on gel-filtration columns into a two overlapping peaks of molecular mass 170 kDa and 450 kDa. Mg2+-independent Ins(1,3)P2 phosphatase had a pH optimum of 7.1 and displayed a single broad activity peak on gel-filtration columns. However, if assays were performed in the presence of Mg2+, a second Ins(1,3)P2 phosphatase was revealed (35 kDa), which had a pH optimum of 8.8 Ins(1,4)P2/Ins(1,3,4)P3 1-phosphatase, Ins(3,4)P2 phosphatase, Mg2+-independent Ins(1,3)P2 phosphatase and inositol monophosphatase were all inhibited by 5,5'-dithiobis(2-nitrobenzoic acid) with IC50 values of 34 microM, 65 microM and 560 microM and 1100 microM, respectively. The metabolism of Ins(1,3,4)P3 by brain supernatant was also examined. Product specificity was shown to be entirely dependent on the buffer conditions employed. In Mg2+-containing buffers, Ins(1,3,4)P3 was hydrolysed predominantly to Ins(3,4)P2, consistent with hydrolysis by Ins(1,4)P2/Ins(1,3,4)P3 1-phosphatase. In the presence of EDTA, Ins(1,3,4)P3 was degraded exclusively by a 4-phosphatase enzyme generating Ins(1,3)P2. Under these conditions, high concentrations of Ins(3,4)P2 blocked the hydrolysis of Ins(1,3,4)P3.

Purification and properties of inositol-1,4-bisphosphatase from bovine brain.

Inositol-1,4-bisphosphatase has been purified 13,000-fold from bovine brain supernatant. The enzyme is monomeric, with an apparent subunit Mr of 40,000. Maximal hydrolytic rates were observed in Tris buffer, pH 7.8, in the presence of 9 mM-Mg2+. The enzyme acted as a 1-phosphatase, hydrolysing both inositol 1,4-bisphosphate [Ins(1,4)P2] (Km 0.04 mM) and inositol 1,3,4-trisphosphate [Ins(1,3,4)P3] (Km 0.5 mM) to inositol 4-phosphate and inositol 3,4-bisphosphate respectively. Li+ inhibited the hydrolysis of both substrates in an uncompetitive manner, with apparent Ki values of 9.63 mM and 0.46 mM for Ins(1,4)P2 and Ins(1,3,4)P3 respectively.

The purification and properties of myo-inositol monophosphatase from bovine brain.

1. An inositol monophosphatase was purified to homogeneity from bovine brain. 2. The enzyme is a dimer of subunit Mr 29,000. 3. The enzyme hydrolyses both enantiomers of myo-inositol 1-phosphate and both enantiomers of myo-inositol 4-phosphate, but has no activity towards inositol bisphosphates, inositol trisphosphates or inositol 1,3,4,5-tetrakisphosphate. 4. Several non-inositol-containing monophosphates are also substrates. 5. The enzyme requires Mg2+ for activity, and Zn2+ supports activity to a small extent. 6. Other bivalent cations (including Zn2+) are inhibitors, competitive with Mg2+. 7. Phosphate, but not inositol, is an inhibitor competitive with substrate. 8. Li+ inhibits hydrolysis of inositol 1-phosphate and inositol 4-phosphate uncompetitively with different apparent Ki values (1.0 mM and 0.26 mM respectively).

The dephosphorylation of inositol 1,4-bisphosphate to inositol in liver and brain involves two distinct Li+-sensitive enzymes and proceeds via inositol 4-phosphate.

1. Hydrolysis of both enantiomers of inositol 1-phosphate and both enantiomers of inositol 4-phosphate to inositol is inhibited by LiCl in liver and brain. 2. The phosphatase activity is predominantly soluble. 3. Inositol 1,4-bisphosphate is also hydrolysed by the soluble fraction of liver and brain. 4. Bisphosphatase activity is inhibited by LiCl, but is less sensitive than monophosphatase activity. 5. The product of bisphosphatase in liver and brain is inositol 4-phosphate.

Proteins of the kidney microvillar membrane. Enzymic and molecular properties of aminopeptidase W.

Aminopeptidase W is a newly discovered enzyme of the renal and intestinal brush borders, having been first isolated as a 130 kDa glycoprotein recognized by a monoclonal antibody [Gee & Kenny (1985) Biochem. J. 230, 753-764]. It is particularly effective in the hydrolysis of dipeptides, Glu-Trp (Km 0.57 mM; kcat. 6770 min-1) being a favoured substrate. Dipeptides with tryptophan, phenylalanine or tyrosine in the P1 position were rapidly hydrolysed, but the requirements in respect of the P1 residue were not stringent. The activity of aminopeptidase W is markedly influenced by ionic conditions. The highest activity was observed in 100 mM-Tris/HCl, pH 8; phosphate ions were strongly inhibitory. Activity was also greatly affected by bivalent metal ions, and the magnitude and direction of the effects depended on the nature of the buffer anions and on pH. The most effective inhibitors were amastatin and bestatin. Some thiols also inhibited, but other chelating agents, EDTA and 1,10-phenanthroline, had no effect over the concentration range 1-10 mM. Other group-specific inhibitors, for cysteine, serine or aspartic peptidases, were also ineffective. Some molecular properties were studied. Deglycosylation by treatment with N-glycanase diminished the apparent subunit Mr from 130,000 to 90,000. The enzyme contained zinc, 1.2 atoms/subunit, and in spite of the atypical properties of this enzyme in respect of chelating agents, a zinc-catalysed mechanism is the most probable. Its roles in digestion and in renal function are not yet clear.

Proteins of the kidney microvillar membrane. The 130 kDa protein in pig kidney, recognized by monoclonal antibody GK5C1, is an ectoenzyme with aminopeptidase activity.

The hybridoma GK5C1, secreting a monoclonal IgG1 antibody, was generated after immunizing a mouse with pig kidney microvillar membranes. An immunoradiometric assay showed that only kidney and intestine contained detectable amounts of the antigen recognized by the antibody, the highest concentration being observed in the ileum. Immunocytochemistry confirmed this observation and revealed that the antigen was associated with renal and intestinal brush borders. By 'Western' blotting, the antigen in kidney microvilli was shown to be a 130 kDa polypeptide. Papain treatment of the membrane before blotting converted the antigen to a 125 kDa polypeptide, no longer associated with membrane. Immunoaffinity chromatography of detergent-solubilized kidney membranes yielded a pure 130 kDa protein. When one purification was monitored by the immunoradiometric assay, the yield was 3.5% and the purification factor was 1000-fold. The antigen constituted about 0.8% of the microvillar membrane protein. The protein could be reconstituted into liposomes, where electron microscopy revealed an asymmetric orientation, similar to that of ectoenzymes in this membrane. The stalk length was about 3 nm. In electron micrographs the purified protein appeared to be dimeric. A search for enzymic activity was rewarded when L-leucyl-L-tryptophan was observed to be hydrolysed. Failure to hydrolyse N-blocked peptides and the ability to release the N-terminal residue from extended peptides, including Leu-Trp-Leu and Leu-Trp-Met-Arg, showed that the activity was that of an aminopeptidase. The enzyme was maximally active at pH 7.5 and irreversibly inactivated outside the range pH 6-10. This activity could not be attributed to trace contamination with aminopeptidase N. The best substrates so far identified for the 130 kDa protein were those with tryptophan in the P1', position. This protein is a new microvillar enzyme and it is proposed that it be called aminopeptidase W.