Discovery and characterization of a small‐molecule enteropeptidase inhibitor, SCO‐792

Abstract Enteropeptidase, localized into the duodenum brush border, is a key enzyme catalyzing the conversion of pancreatic trypsinogen proenzyme to active trypsin, thereby regulating protein digestion and energy homeostasis. We report the discovery and pharmacological profiles of SCO‐792, a novel inhibitor of enteropeptidase. A screen employing fluorescence resonance energy transfer was performed to identify enteropeptidase inhibitors. Inhibitory profiles were determined by in vitro assays. To evaluate the in vivo inhibitory effect on protein digestion, an oral protein challenge test was performed in rats. Our screen identified a series of enteropeptidase inhibitors, and compound optimization resulted in identification of SCO‐792, which inhibited enteropeptidase activity in vitro, with IC 50 values of 4.6 and 5.4 nmol/L in rats and humans, respectively. In vitro inhibition of enteropeptidase by SCO‐792 was potentiated by increased incubation time, and the calculated K inact/KI was 82 000/mol/L s. An in vitro dissociation assay showed that SCO‐792 had a dissociation half‐life of almost 14 hour, with a calculated k off rate of 0.047/hour, which suggested that SCO‐792 is a reversible enteropeptidase inhibitor. In normal rats, a ≤4 hour prior oral dose of SCO‐792 effectively inhibited plasma elevation of branched‐chain amino acids in an oral protein challenge test, which indicated that SCO‐792 effectively inhibited protein digestion in vivo. In conclusion, our new screen system identified SCO‐792 as a potent and reversible inhibitor against enteropeptidase. SCO‐792 slowly dissociated from enteropeptidase in vitro and inhibited protein digestion in vivo. Further study using SCO‐792 could reveal the effects of inhibiting enteropeptidase on biological actions.


| INTRODUC TI ON
Proteins are pivotal macronutrients for various cellular activities, as well as whole-body metabolism. Amino acids show many functions within the body. For example, amino acids serve as the only source of nitrogen in mammals. 1 Amino acid-derived nitrogen is a critical element in the synthesis of the precursors (purine and/or pyrimidine) of major energy molecules, such as adenosine triphosphate, adenosine diphosphate, and/or nucleic acids. 1 In addition, nitrogen is incorporated into compounds that can regulate major biochemical signaling pathways, such as nitric oxide. 1 Furthermore, amino acid deamination in the body's proteins generates a carbon skeleton rich in oxygen and hydrogen suitable for subsequent biochemical transformation. 1 This carbon skeleton can be used by the liver to generate glucose through gluconeogenesis and other macromolecules, such as lipids. 1 The carbon skeleton derived from amino acids is also relevant in producing intermediaries fueling the Kreb's cycle that are thereafter transformed into energy and/or other metabolic molecules. 1 Taken together, amino acids can be considered as biochemical molecules that can be converted into energy, carbohydrates, lipids, and biochemical intermediates depending on the specific bodily metabolic situation.

Degradation of dietary proteins and subsequent amino acid
absorption is a key step in maintaining protein homeostasis in mammals. [2][3][4] Activation of pancreatic enzymes is essential for digestion of proteins to give amino acids that are subsequently absorbed in the gut. During food digestion in the gut, enteropeptidase (EC 3.4.21.9) serves as a critical upstream molecule in the process of protein digestion. 5 Enteropeptidase is a serine protease localized on the intestinal brush border. The enzyme catalyzes the conversion of inactive trypsinogen, which is secreted from the pancreas into the gut, to active trypsin 6 . The activated trypsin in turn activates downstream digestive enzymes, such as chymotrypsinogen, proesterase, procarboxypeptidases A and B, and prolipase, which allow the absorption of amino acids and triglycerides in the gut. 5 Congenital enteropeptidase deficiency in humans has resulted in intestinal malabsorption and a lean phenotype, which suggest the pivotal role of this enzyme in regulating body homeostasis. 7,8 Interestingly, a recent seminal study suggested that inhibiting gut enteropeptidase may be a novel strategy for correcting obesity. 9 Concomitantly, accumulating reports suggest that the strong connection of plasma amino acid change with obesity and insulin resistance. 10 Considering that enteropeptidase is an upstream key molecule in a protein degradation process, 5  Thus, the current study was conducted to identify new enteropeptidase inhibitors and characterize their in vitro and in vivo biological activity profiles. We first constructed a new high-throughput screening (HTS) system to identify compounds inhibiting enteropeptidase in vitro. After screening, an optimized compound, 11 was further characterized by in vitro and in vivo studies.  (N-({(3S)-6-[(4-carbamimidamidobenzoyl)oxy]-2,3-dihydro-1-benzofuran-3-yl}acetyl)-L-aspartic acid hydrate) was synthesized by Takeda Pharmaceutical Company Limited. The substrates QSY21-Gly-Asp-Asp-Asp-Lys-Ile-Val-Gly-Gly-Lys (Cy5) and 5FAM-Abu-Gly-Asp-Asp-Asp-Lys-Ile-Val-Gly-Gly-Lys (CPQ2)-Lys-Lys-NH 2 were purchased from CPC Scientific (Sunnyvale, CA

| Enteropeptidase enzyme assay
In the HTS, enzyme and substrate were dissolved in the enteropepti- where t is the time, F is the fluorescence, and F 0 is the fluorescence at time t = 0.
The k inact /K I,app value was determined as the slope of the [I] vs k obs plot, and the corrected k inact /K I value was also estimated according to the following equation: where [I] is the concentration of inhibitor, [S] is the concentration of substrate, and K m is the Michaelis-Menten constant.
All enteropeptidase enzyme assay and compound evaluation were conducted at pH 8 because the optimal pH of enteropeptidase was 8 as previously reported 12 ; Magee et al. 13

| Dissociation assay
For the dissociation assay, compounds were dissolved in DMSO and then diluted in the enteropeptidase assay buffer. Ten microliters of compound solution was added to a 96-well plate, and then 10 μL of 100 mU/mL human recombinant enteropeptidase solution was added to the plate and incubated at room temperature for 120 minutes. The concentration of the compound was equal to 10-fold of the IC 50 value upon incubation for 120 minutes. After this incubation, 2 μL of an compound-enzyme mixture was transferred to a 96-well black plate, and then 200 μL of substrate solution [3 μmol/L 5FAM-Abu-Gly-Asp-Asp-Asp-Lys-Ile-Val-Gly-Gly-Lys(CPQ2)-Lys-Lys-NH 2 ] was added to the well. By its rapid dilution, the concentration of the inhibitor dropped from 10-fold above the IC 50 to 10-fold below it. The fluorescence was measured every 60 minutes at an excitation wavelength of 485 nm and an emission wavelength of 520 nm using an EnVision multilabel plate reader. The progress curves were fitted to the following equations to determine the values for k off and dissociation half-life, t 1/2 .

| Animals
All animals were housed in a room with controlled temperature (23°C), humidity (55%), and lighting (lights on between 7:00 am and 7:00 pm). All animals were allowed free access to standard

| Pharmacokinetic study in rats
Male Sprague-Dawley rats were obtained from Charles River Laboratories Japan, Inc. (Yokohama, Japan). A pharmacokinetic study was conducted when the animals were 8 weeks old. For oral administration, SCO-792 was suspended in a 0.5% (w/v) methylcellulose solution. For intravenous administration, SCO-792 was dissolved in DMSO and added with saline to prepare a dosing formulation at a concentration of 0.2 mg/mL (DMSO:saline = 2:8 (v/v)). The oral dosing formulations were mixed well and given to fed-male rats at single doses of 10 mg/5 ml/kg using polypropylene syringes with gavage needles. The intravenous dosing formulation was injected into the femoral vein of fed-male rats at a dose of 0.2 mg/1 ml/kg using polypropylene syringes with needles under anesthesia with isoflurane.
The blood was collected at the indicated time points, and the plasma concentrations of SCO-792 were determined by high-performance liquid chromatography/tandem mass spectrometry.

| Oral protein challenge in rats
Male Sprague-Dawley rats were obtained from CLEA Japan Inc.
(Tokyo, Japan). Eight-week-old rats were randomized into three groups on the basis of body weight and body weight change before the experiment (n = 4). SCO-792 (10, 30 mg/kg in a 0.5% (w/v) (2)∕k off methylcellulose solution) was then orally administered 1, 2, 4, and 6 hours before an oral whey protein load (2.5 g/kg; SAVAS whey protein 100, Meiji, Japan), and blood samples were collected at indicated time points for the measurement of plasma branched-chain amino acids (BCAA).

| BCAA measurement
Plasma BCAA concentration was measured using an enzymatic spectrophotometric assay as described by Beckett. 14 Briefly, bacterial leucine dehydrogenase was used to catalyze the oxidation

Cy5-substrate (µM)
of BCAA, and the production of NADH was measured using a spectrophotometer (excitation 355 nm, emission 460 nm). L-leucine was used to generate a standard curve to determine sample concentrations.

| Statistical analysis
Statistical significance was first analyzed using Bartlett's test for homogeneity of variances, followed by the Williams' test for dosedependent studies. The Williams' test was performed using a onetailed significance level of 2.5% (0.025). All data are presented as means ± standard deviations (SDs). The dose-response data were fitted to a four-parameter logistic curve using GraphPad Prism ver. 5 to determine the half-maximal inhibitory concentration (IC 50 ) values and 95% confidence intervals.

| Development of an enteropeptidase assay system
Enteropeptidase is produced as a proenteropeptidase in enterocytes, the activation of which requires digestion to a light chain and a heavy chain. The light chain contains the catalytic active site, whereas the disulfide-linked heavy chain provides the anchor to the brush border membrane. Thus, we used recombinant light chain of human enteropeptidase to discover its inhibitors. To identify small-molecule enteropeptidase inhibitors, we designed two specific dual-labeled substrates and developed enzyme assay systems. In the assay, the energy of the donor of an uncleaved substrate is transferred to the proximal acceptor at the opposite side of the substrate. When protease cleaves the substrate, the pair separates and the donor no longer transfers its emission to the acceptor, which leads to an increase in the fluorescence intensity observed from the donor. 15 Enteropeptidase activity was measured with our designed substrates, Cy5-labeled ( Figure 1C) and 5FAM-labeled substrates ( Figure 1E), and commercially available substrate, H-Gly-Asp-Asp-Asp-Asp-Lys-βNA (GDDDDK-βNA), naphthylamine ( Figure 1A). In our assay system, the fluorescence signal from the enzyme reaction increased in a time-dependent manner. The calculated K m value for the 5FAM-labeled substrate was 0.7 μmol/L ( Figure 1F), which is almost 300 times lower than that of a commercially available substrate, GDDDDK-βNA, whose K m value was 230 μmol/L ( Figure 1B). Regarding Cy5-labeled substrate, we could not calculate the K m value because the enzyme activity was suppressed at a high concentration of substrate ( Figure 1D), which may have been caused by energy transfer not only intramolecularly but also intermolecularly at a high concentration.

| Identification of small-molecule inhibitors of enteropeptidase
We searched for compounds harboring an amidine or guanidine moiety in Takeda's compound library because such compounds are supposed to be good binders to proteases, such as enteropeptidase, that cleave after a basic amino acid residue. As a result, 1116 compounds were screened by enzyme assay using Cy5-labeled substrate. Among these compounds, 164 that showed >45% inhibition at 30 μmol/L were selected as candidates for enteropeptidase inhibitors. Next, the compounds that met the following conditions were removed:  Figure 2B) at 120-minute incubation. Further kinetic study revealed that the inhibition by T-0046812 was time-dependent ( Figure 2B and C), and the k inact /K I value was 5300 (/mol/L s) ( Figure 2D). The structure of T-0046812 is expected to undergo hydrolysis at its ester moiety to form a covalent bond with the catalytic serine of enteropeptidase.

| Biochemical characterization of the investigated drug, SCO-792
In the optimization process, we improved the chemical stability by  (Table 1). Further kinetic study revealed that the inhibition of and the k inact /K I value was 82 000 (mol/L s) ( Figure 3D).
A dissociation assay revealed that inhibition by SCO-792 was reversible and release of the compound from the enteropeptidasecompound complex should occur very slowly, with a calculated dissociation t 1/2 and K off rate of 14 hour and 0.047/hour ( Figure 4A).
Based on the structure of SCO-792, the ester moiety of the compound is expected to be subjected to quick nucleophilic attack by the catalytic serine of enteropeptidase; moreover, based on the result of the dissociation assay, the covalent complex of guanidinobenzoate and enteropeptidase is expected to be hydrolyzed slowly after the formation of covalent bonds ( Figure 4B). Because of the slowly reversible nature, SCO-792 is considered to exhibit the desired ac-

| Pharmacokinetic profiles of SCO-792 in rats
After oral administration of SCO-792 to rats at a dose of 10 mg/kg, the plasma concentrations of SCO-792 reached 6.60 ng/mL (C max ) at

| In vivo inhibition of enteropeptidase in rats
Given that SCO-792 was found to be a potent inhibitor against duodenal enteropeptidase activity, which is essential for dietary protein diges- were tested in rats. This study revealed that a ≤4 hour prior oral dosing of SCO-792 effectively and dose-dependently inhibited plasma BCAA elevations induced by oral protein dosing in rats ( Figure 6).

| D ISCUSS I ON AND CON CLUS I ON S
In this study, we reported the discovery and pharmacological profiles The most common way to detect protease activity is based on specifically labeled substrates. 15 In this approach, the cleaved substrates contain a chromophore or fluorophore, which enables light energy from the fluorophore in an uncleaved state. We used cyanine Cy5 dye as a donor and QSY21 as an acceptor for conducting HTS, and since the emission wavelength of Cy5 is longer than that for the usual sources from the compound itself, a Cy5-labeled substrate is less susceptible to false-positive/negative results. We also developed a system using CPQ2 and 5FAM in combination for a long assay period because Cy5 is subjected to oxidation by environmental ozone or other oxidants, which results in a decrease in fluorescence intensity. 16 Moreover, both Cy5-and 5FAM-labeled substrates cover both prime and non-prime sides, which enable us to evaluate both prime side-binding and non-prime side-binding inhibitors. In addition, the new method using a Cy5-labeled or 5FAM-labeled substrate requires only 30 mU/mL or 8 mU/mL of enzyme each, which corresponds to approximately 0.3% or 0.08%, respectively. By comparison, the assay using GDDDDK-βNA needs a large amount of enzyme (typically 10 U/mL). As a result, we achieved a reduction of enzyme concentration in the assay, so we could evaluate stronger inhibitors kinetically using the highly sensitive substrates. 17 F I G U R E 5 Plasma concentration of SCO-792 after oral (A) or intravenous (B) administration in rats. SCO-792 was administrated at the dose of 10 mg/kg orally and 0.2 mg/kg intravenously. Values are presented as the mean ± SD (n = 3). Calculated pharmacokinetic parameters are shown in Enteropeptidase is a specific protease that recognizes and cleaves after a basic amino acid residue. 18  In the process of compound optimization, we noticed that slow dissociation of compound from enteropeptidase is very important for inhibition of enteropeptidase activity in vivo. In fact, the drug-target residence time has been reported to be an important parameter for in vivo efficacy. 29,30 In particular, the importance of drug-residence time for enteropeptidase inhibitors probably depends on their mechanism of action site. Since enteropeptidase inhibitors act in the intestinal tract without circulating in the blood, Recently, in vivo enteropeptidase inhibition by SCO-792 was demonstrated to be highly effective in improving the disease status of diabetes and obesity in mouse models. 31 In addition, inhibition of intestinal trypsin, which is a downstream molecule of enteropeptidase, was shown to decrease body weight and improve metabolism in leptin-deficient and DIO mice. 32 Taken these together, protein digestion inside the gut likely to have a significant role in regulating body weight and metabolism and inhibiting this step may be a rational strategy to improve obesity and diabetes. inhibitors are attracting attention because they exhibit sufficiently long efficacy without the potential for side effects resulting from permanent bonding between the compound and target protein. 35 Accordingly, the reversible covalent inhibitor character of SCO-792 has an advantage for demonstrating in vivo biological effects because such a compound effectively inhibits the proteolysis of trypsinogen and following digestive signals with reduced side effects over long time periods.
In summary, our new screen identified SCO-792 as a potent and reversible enteropeptidase inhibitor against enteropeptidase.
When orally dosed to rats, SCO-792 effectively inhibited protein digestion. Taken together, SCO-792 was found to be an effective enteropeptidase inhibitor in vitro and in vivo. Further study using SCO-792 could demonstrate the effects of inhibiting enteropeptidase on biological actions.

ACK N OWLED G EM ENTS
The authors are grateful to Ms. Yumi Zama for preparing the protein for rat enteropeptidase, Ms. Miyako Shibazaki for executing the enzyme assay, and Mr. Tsutomu Henta for executing the HTS and Dr.
Hideyuki Oki for insightful comments.

CO N FLI C T O F I NTE R E S T
This study was conducted with the financial support of Takeda

DATA ACCE SS I B I LIT Y S TATE M E NT
The data that support the findings of this study are available from the corresponding author upon reasonable request.