2.1. Assays

2.2. Probe Chemical Characterization

CID 53364507
SID 125269079
ML256

The probe structure was verified by ¹H NMR (see section 2.3) and high resolution MS (m/z calculated for C₂₆H₂₄N₄O₂ [M+H]⁺: 425.1972, found 425.1974). Purity was assessed to be 96% by NMR. Solubility (room temperature) was determined to be 16.3 μM in PBS, 7.5 μM in DMEM medium, and >100 μM in DMEM medium containing 10% fetal calf serum. Stability in PBS was determined by LC-MS to be >48 hours.

2.3. Probe Preparation

2-methoxy-5-phenylpyridine: A mixture of 5-bromo-2-methoxypyridine (5.641 g, 30 mmol, 1.0 eq), phenylboronic acid (4.389 g, 36 mmol, 1.2 eq), Na₂CO₃ (9.539 g, 90 mmol, 3.0 eq) and Pd(Ph₃P)₄ (1.74 g, 5 mmol%) in toluene/H₂O (v/v 5:3, 240 mL) was heated to 90°C under N₂ overnight. The product was extracted with EtOAc (3 × 300 mL) and purified by silica gel column chromatography (1–10% ethyl acetate/hexanes) to afford 2-methoxy-5-phenylpyridine (5.0571g, 27 mmol, 90%).

5-phenylpyridin-2-ol: To a solution of a 2-methoxy-5-phenylpyridine (3.13 g, 17 mmol, 1.0 eq) in dry CH₂Cl₂ (85 mL) was slowly added BBr₃ (5 g, 20 mmol, 1.2 eq) at −20°C. Upon warming to room temperature, the reaction mixture was stirred overnight. 1N HCl (34 mL) was added to quench the reaction at 0°C. The product was extracted with ethyl acetate (3 × 170 mL). The combined organic phases were dried over NaSO₄, filtered and concentrated under reduced pressure. The product was purified by silica gel column chromatography (1–10% methanol/CH₂Cl₂) to recover unreacted starting material and afford 5-phenylpyridin-2-ol (0.0825 g, 0.5 mmol, 3%).

5-phenylpyridin-2-yl 4-(2-methylquinolin-4-yl)piperazine-1-carboxylate (Probe ML256): To a solution of triphosgene (0.0549 g, 0.185 mmol, 0.33 eq) in CH₂Cl₂ (5.55 mL), was slowly added 5-phenylpyridin-2-ol (0.0951 g, 0.555 mmol, 1.00 eq) and DIEA (97 μL, 0.0716 g, 0.555 mmol, 1.00 eq) as a solution in THF (5.55 mL). The reaction was stirred at room temperature for 30 min to generate the intermediate chloroformate. In a separate flask containing CH₂Cl₂ (5.55 mL), 2-methyl-4-piperazinoquinoline (0.1892 g, 0.833 mmol, 1.50 eq) was mixed with DIEA (146 μL, 0.1078 g, 0.833 mmol, 1.5 eq). The mixture was stirred for 5 minutes at room temperature and then dropped into the flask containing the chloroformate. The reaction mixture was stirred at room temperature for another 30 minutes and partitioned between CH₂Cl₂ (28 mL) and saturated NH₄Cl (28 mL). The organic layer was washed one more time with saturated NH₄Cl (17 mL), and dried over MgSO4. The solvent was removed by rotary evaporation and the product was purified by silica gel column chromatography (1:1 ethyl acetate/hexanes) to afford 5-phenylpyridin-2-yl 4-(2-methylquinolin-4-yl)piperazine-1-carboxylate (Probe ML256) (0.0903 g, 0.212 mmol, 38%). ¹H NMR (400 MHz, CDCl₃) δ 8.59 (dd, J = 2.5, 0.7 Hz, 1H), 8.04–7.91 (m, 3H), 7.65 (ddd, J = 8.3, 6.8, 1.4 Hz, 1H), 7.60–7.35 (m, 6H), 7.22 (dd, J = 8.4, 0.7 Hz, 1H), 6.78 (s, 1H), 4.02 (t, J = 4.9 Hz, 2H), 3.90 (t, J = 4.8 Hz, 2H), 3.2 (t, J = 5.2 Hz, 4H), 2.70 (s, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 159.71, 157.84, 156.62, 153.10, 149.50, 146.76, 138.30, 137.19, 135.33, 129.56, 129.50, 129.32, 128.35, 127.34, 125.17, 123.20, 121.95, 116.31, 110.07, 52.20, 52.08, 45.08, 44.30, 25.87; HRMS ESI-TOF high-acc (M + H⁺) calculated for C₂₆H₂₄N₄O₂ 425.1972, found 425.1974; purity estimate by NMR: 96%.

3.1. Summary of Screening Results

In the primary fluopol-ABPP pPAFAH HTS Assay (AID 463082), ~326K compounds were tested for inhibition of enzyme labeling by the SH-specific activity-based probe FP-Rh [18]. The assay was conducted in singlicate at 3.39 μM compound concentration using purified, recombinant human enzyme. A total of 4,934 compounds (1.5%) were active, passing the set threshold (mean + 3× standard deviation) of 24.55% inhibition. A total of 2,500 compounds were selected for testing in the confirmation assay (AID 463230), in which compounds were re-tested at the same concentration in triplicate. Of the 2,341 available compounds, 1,675 (71.6%) confirmed as active (Figure 3.1-1).

Figure 3.1-1

Flow chart describing HTS results.

Because the fluopol-ABPP HTS assay was conducted with purified, recombinant enzyme, we selected a subset of active compounds for secondary gel-based screening to assess pPAFAH inhibition in a complex proteome. Compounds were cherry-picked from among the top HTS hits using the following criteria: in the confirmation HTS assay, a total of 407 compounds inhibited the target enzyme by at least 60%. Of those 407 compounds, 87 were removed due to reactivity with other select enzymes screened by fluopol-ABPP: GST01 (68, AID 1974), PME1 (5, AID 2130), PREPL (13, AID 2751), and PAD4 (1, AID 485272). Of the remaining 320 compounds, 92 exhibited more than 5% activity in all bioassays tested and were removed. Out of this group of 228 compounds, just over half were cherry picked for secondary screening based on lack of undesired functional groups (esters, hydrazines, oximes, etc.) and medchem potential. Because the carbamate scaffold was observed to feature prominently among the top HTS hits, an additional 25 carbamates were also selected for a total of 153 compounds.

For the gel-based competitive ABPP assay, a complex membrane proteome lysate prepared from 293T Hek cells overexpressing recombinant mouse pPAFAH was incubated with test compound (5 μM) followed by reaction with the FP-Rh activity-based probe. The reaction products were separated by SDS-PAGE and visualized in-gel using a flatbed fluorescence scanner. Test compounds that act as pPAFAH inhibitors prevent enzyme-probe interactions, thereby decreasing the fluorescence intensity of the protein bands. As reported in AID 588474, only twelve compounds (Table 3.1-1) inhibited the target enzyme by ≥75%, eight of which were carbamates (See Figure S1 and Table S1 for gel images and full tabulated results, respectively). Most of the carbamate compounds showed good selectivity vs. PAFAH2, the closest sequence homolog of pPAFAH, in the counterscreen HTS fluopol-ABPP assay, indicating that medchem optimization should be able to deliver a selective pPAFAH inhibitor.

Table 3.1-1

Most promising pPAFAH inhibitor leads by gel-based competitive ABPP.

We were excited to discover promising carbamate leads for pPAFAH, as we and others have had previous success developing potent and selective serine hydrolase inhibitors based on the carbamate scaffold [28–33] (Table 3.1-2). The carbamate scaffold readily lends itself to diversity-oriented synthesis and medchem manipulation. Moreover, as a class, carbamates are typically active in vivo and have the ability to penetrate the nervous system [28, 33, 34]. We have confirmed by click chemistry-ABPP that members of this chemotype show negligible reactivity with proteins outside the SH class [35]. Within the SH class, we have previously identified carboxylesterases (CESs) as common anti-targets for carbamate inhibitors [28, 35]. In this regard, the carbamate is not unique, as the CES enzymes appear to be common targets for a variety of SH-directed inhibitors [36], likely a reflection of the role they play in xenobiotic metabolism. However, CESs are mostly restricted in their expression profile to liver [37], thus their presence as anti-targets should not affect SH characterization in most cells and tissues. For example, both the FAAH inhibitor URB597 [31] and the MAGL inhibitor JZL184 [33] (see structures Table 3.1-2) have proven to be highly useful chemical probes [31, 33, 38–40] despite having some CES inhibitory activity [35, 41]. Unless the target of interest is a liver enzyme, the key selectivity requirement for successful carbamate probes is to show high selectivity among all non-CES SHs. In the case of pPAFAH, we are interested in studying the effects of enzyme inhibition in the brain, so CES inhibitory activity should not be a significant concern. We have also previously addressed the issue of imperfectly selective probes by generating control “anti-probes” (see ML211 and ML226 Probe Reports), which are structurally and mechanistically similar to the optimized inhibitor but lack activity against the SH of interest, for use in parallel biological experiments to confirm functional assignments.

Table 3.1-2

Representative carbamate SH inhibitors.

Also included in Table 3.1-2 is a carbamate lead inhibitor for pPAFAH, WWL153, which we identified from a small, in-house carbamate screening library [28]. The compound exhibited decent potency for the target enzyme, but poor selectivity, inhibiting anti-targets fatty acid amide hydrolase (FAAH) and monoacylglycerol lipase (MAGL) with only a marginal dosing window (Table 3.4-1). Fortuitously, the HTS carbamate hits gave us new medchem leads to further develop this scaffold (detailed in Section 3.4) ultimately leading us to the optimized pPAFAH probe ML256 (SID 125269079).

Table 3.4-1

Target SAR Analysis.

3.2. Dose Response Curves for Probe

IC50 values for both the human (6 nM; AID 588766) and mouse (31 nM; AID 588770) isoforms of pPAFAH were obtained from gel-based competitive-ABPP data (Figure 3.2-1).

Figure 3.2-1

IC50 curves for probe ML256 (SID 125269079) as determined by gel-based competitive-ABPP with FP-Rh against purified, recombinant human pPAFAH (left) and overexpressed mouse pPAFAH in a complex proteome lysate (293T Hek membrane) (right).

3.3. Scaffold/Moiety Chemical Liabilities

The probe compound was determined to covalently modify the catalytic serine (Ser273) of pPAFAH by LC-MS/MS analysis (AID 588788). The observed mass shift of the active site peptide corresponds to carbamoylation of the enzyme upon serine nucleophilic attack at the carbonyl followed by loss of the hydroxyl moiety (Figure 3.3-1), as expected of this chemotype [35].

Figure 3.3-1

Covalent modification of pPAFAH by ML256 (SID 125269079) as determined by LC-MS/MS analysis.

The probe compound showed no reactivity with glutathione (100 μM), indicating that it has a tempered electrophilicity and specific structural elements that direct reactivity towards pPAFAH. Gel-based and LC-MS/MS (MudPIT)-based ABPP selectivity profiling also confirmed high target specificity (Section 3.6).

We [42], and others [43–45], have found that irreversible inhibitors such as ML256 offer many advantages over reversible inhibitors as pharmacologic probes for enzyme characterization in biological systems. Perhaps most importantly, it is technically straightforward to confirm the irreversible inhibition of enzymes in living systems (including mice) using competitive ABPP and click chemistry ABPP; as such, we can define the precise quantity of inhibitor and treatment time required to selectively inactivate an enzyme of interest for biological studies. This information is extremely valuable for guiding the functional characterization of SHs. Additionally, required dosing is often lower, irreversible compounds are not as sensitive to pharmacokinetic parameters, and administration can induce long-lasting inhibition [42]. In the case of the EGFR inhibitor PD 0169414, its irreversibility and high selectivity were credited with producing prolonged inhibition of the target, alleviating concerns over short plasma half-lives and reducing the need for high peak plasma levels, thus minimizing potential nonspecific toxic effects [46].

Indeed, over a third of enzymatic drug targets are irreversibly inhibited by currently marketed drugs [47]. Examples of covalent enzyme-inhibitor pairs include serine type D-Ala-D-Ala carboxypeptidase, which is covalently modified by all B-lactam antibiotics, acetylcholinesterase, whose active site serine undergoes covalent modification by pyridostigmine, prostaglandin-endoperoxide synthase, which is the target of the ubiquitously prescribed aspirin, aromatase, which is irreversibly modified by exemestane, monoamine oxidase, which is covalently modified by L-deprenyl, thymidylate synthase, which is covalently modified by floxuridine, H⁺/K⁺ ATPase, which undergoes covalent modification by omeprazole, esomeprazole, and lansoprazole, and triacylglycerol lipase, whose serine nucleophile is targeted by orlistat [47].

3.4. SAR Tables

The SAR Table 3.4-1 includes 21 compounds: the top 8 carbamates identified by HTS (entries 1–8), our in-house carbamate lead WWL153 (entry 9), and 12 new synthetic compounds (entries 10–21). The analogs are variable at the left hand binding group and right hand leaving group substituents. To investigate whether or not the carbamate was the best electrophile for pPAFAH inhibition, we also generated a series of five triazole urea analogs (Table 3.4-2). The triazole urea reactive group is another “privileged” scaffold for SH inhibition from which we have derived several potent, selective, and in situ-active SH probes: ML211 as a dual inhibitor of LYPLA1&2, ML226 for ABHD11, and ML225 for the pPAFAH homolog PAFAH2 (see also ref. [48]).

Table 3.4-2

Target SAR Analysis: Triazole Ureas.

All SAR compounds were subject to gel-based competitive ABPP profiling [19] to assess both potency and selectivity against several dozen FP-sensitive SHs. Because endogenous pPAFAH is difficult to visualize by gel due to low abundance and/or overlapping SHs, potency was assessed against recombinant mouse pPAFAH in the membrane proteome of 293T Hek cells engineered to overexpress the enzyme. Given our interest in exploring pPAFAH biochemistry in the brain and owing to the brain’s rich diversity of anti-target SHs, selectivity was assessed using a mouse brain membrane proteome preparation. For both the potency and selectivity analysis, compounds were tested at three concentrations: 10 μM, 1 μM, and 100 nM, and the gel profiling results are included in Section 3.6 (see Section 2.1 and AIDs 588773 and 588774 for protocol details). Proteins are listed as anti-targets for a given compound concentration if at least 50% inhibition was observed relative to the DMSO only (no compound) control. For comparison, we included the four non-carbamate HTS hits from Table 3.1-1 in our investigation of potency and selectivity, the results (none showed pronounced inhibition of the target enzyme) are tabulated in Table S2.

J11/WWL153 Analogs: The development of a pPAFAH inhibitor began by combining structural elements of the top HTS hits, chiefly J11 (SID 26727296, entry 1) and P09 (SID 4257433, entry 2), with our in-house lead WWL153 (SID 99205832, entry 9) [28]. WWL153 displays an IC50 on the order of ~200 nM towards pPAFAH with some inhibition of FAAH and MAGL at higher (1 μM and 10 μM) concentrations, which we wanted to avoid with the final compound. We hypothesized that the 2-methyl-4-piperazinoquinoline motif was providing the majority of the potency and selectivity for our compound, especially given the predominance of piperidine-like motifs in the structures of the 12 HTS hit compounds (all but J11, Table 3.1-1). In contrast, the leaving group moieties of the HTS carbamates, while all aromatic with an emphasis on bis/bi aromatic systems (e.g., 4-hydroxybiphenyl for P09 and P03, 4-phenoxyphenol for A07; Table 3.4-1, entries 2, 6, and 7, respectively) evinced considerably more diversity. An especially intriguing motif was the 6-chlorobenzo[d]isoxazol-3-ol leaving group of compound J11 (entry 1). Out of all the HTS carbamates, J11 displayed the best inhibition of pPAFAH in gel-based analysis (Tables 3.1-1 and 3.4-1). However, it also showed a strong potency towards the anti-targets FAAH and ABHD6, with persistent inhibition of the former even at low (100 nM) concentration (Table 3.4-1). We hypothesized that this lack in selectivity could be coming from the very simple dimethyl-amino binding group and that a more complex amine may provide better selectivity.

Based on this analysis, we synthesized a series of first generation analogs by coupling the 6-chlorobenzo[d]isoxazol-3-ol leaving group of J11 with the 2-methyl-4-piperazinoquinoline binding group of WWL153 to create JMN5, and with two prevalent amines among the HTS hits: piperidine to generate JMN6, and morpholine to give JMN10 (Table 3.4-1, entries 10–12). During synthesis, it was found that urea structural isomers of the 6-chlorobenzo[d]isoxazol-3-ol carbamates were formed as minor byproducts of the reaction; thus JMN7 (isomer of JMN5) and JMN11 (isomer of JMN10) were also synthesized and isolated for testing (Table 3.4-1, entries 13 and 14). As a group, these 6-chlorobenzo[d]isoxazol-3-ol carbamates and ureas (entries 10–14), show high potency for pPAFAH (>80% inhibition of pPAFAH at 0.1 uM; Table 3.4-1) but they also inhibited numerous other SHs, including FAAH, MAGL, and ABHD6, with ~equal potency. This anti-target activity suggests that the 6-chlorobenzo[d]isoxazol-3-ol leaving group overly activates the carbonyl, and that less activating leaving groups may provide better selectivity.

WWL153/Triazole Urea Analogs: Recently, our lab has investigated 1,2,3-triazole ureas as SH inhibitors (See ref. [48] and ML211, ML225 and ML226 Probe Reports). Given the potency of the urea JMN7 (entry 13) and the fact that we have generated selective triazole urea SH inhibitors in the past, we thought investigation of the triazole urea electrophilic scaffold for pPAFAH inhibition was merited. As such, we synthesized a series of triazole urea analogs using the 2-methyl-4-piperazinoquinoline binding group in combination with and a series of different biaromatic triazole leaving groups (Table 3.4-2). These compounds showed great potency towards pPAFAH (near complete inhibition observed at 0.1 μM compound concentration for all); however, they also inhibited FAAH, MAGL, and numerous other SH anti-targets with high potency as well. Due to the lack of selectivity, the triazole urea was deprioritized as scaffold for generation of a pPAFAH inhibitor.

P09/WWL153 Analogs: As mentioned before, one leaving group motif that was prevalent among the HTS carbamate hits was the 4-hydroxybiphenyl (P09 and P03, Table 3.4-1, entries 2 and 6). These compounds were intriguing because they not only have decent potency for inhibiting pPAFAH, but also avoid inhibiting FAAH at low concentration. The 4-hydroxybiphenyl leaving group was thus combined with the 2-methyl-4-piperazinoquinoline binding group to form JMN4 (Table 3.4-1, entry 15), which showed excellent potency for pPAFAH (IC50 = 91) and the highest selectivity coefficient (>109-fold) of any synthetic compounds tested thus far.

Since the 4-hydroxybiphenyl leaving group seemed important for avoiding FAAH inhibition, presumably due to its steric bulk, we proceeded to synthesize several JMN4 analogs with biaryl or aryl-cyclo leaving groups (JMN18-JMN23, Table 3.4-1, entries 16–21). With the exception of JMN21 (entry 19), all analogs displayed reduced potency (and reduced selectivity in the case of JMN20, entry 18) in comparison to JMN4. Compound JMN21 was ~3× more potent (IC50 = 31 nM) than JMN4 and also exhibited no anti-target effects at the highest concentration (10 μM) tested. The only difference between JMN21 and JMN4 is the replacement of a carbon in the phenol group with a nitrogen atom. This modification slightly activates the carbonyl, thus increasing potency. However, the biaryl system of JMN21 prevents promiscuous binding, thus restricting this heightened reactivity to the target enzyme. This analog, JMN21, was designated as pPAFAH probe ML256.

Summary: As compared to our initial HTS (J11 and P09) and in-house (WWL153) lead pPAFAH inhibitors, probe ML256 (IC50 31 nM) represents a 5-fold improvement in potency and two orders of magnitude improvement in selectivity, showing no evidence of FP-sensitive SH anti-target reactivity up to 10 μM compound concentration by gel-based ABPP.

3.5. Cellular Activity

In Situ Inhibition: Probe ML256 (SID 125269079) is active against pPAFAH in situ (AID 588817), inhibiting enzymatic activity by 58% at 100 nM compound concentration after 4 hours as assayed by gel-based competitive ABPP with FP-Rh (Figure 3.5-1). This result indicates that ML256 is free to cross cell membranes and inhibit its target in cells.

Figure 3.5-1

Inhibition of pPAFAH activity in situ by probe ML256 (SID 125269079). Transiently transfected 293T Hek cells overexpressing mouse pPAFAH (serum-free DMEM medium) were treated with ML256 for 4 hours, washed, harvested and the membrane fraction was isolated (more...)

In Vivo Inhibition: We also assessed the ability of probe ML256 (SID 125269079) to inhibit pPAFAH in vivo (AID 588823). Brain tissue was harvested from mice administered ML256 (50 mg/kg, oral gavage, 4 hour timecourse) and the membrane fraction isolated and profiled by gel-based competitive ABPP using either HT-01 (Figure 3.5-2, left panel) or FP-Rh (Figure 3.5-2, right panel). HT-01 is a triazole urea activity-based probe [27] that labels a handful of SHs in the brain, allowing enhanced visualization of pPAFAH, which is otherwise obscured by overlapping SH bands with FP-Rh profiling. In probe-treated animals, pPAFAH activity is reduced by more than 95%.

Figure 3.5-2

Inhibition of pPAFAH activity in vivo by probe ML256 (SID 125269079). Mice (n=2 per group) were administered ML256 (50 mg/kg via oral gavage) or vehicle only. After 4 hours, mice were sacrificed and their brain tissue removed. The membrane fraction was (more...)

There is one observed CES anti-target, CES1, which is inhibited by ML256 under these conditions (Figure 3.5-2A, left panel); however, CES1 is a liver enzyme [28, 37], and its presence in the membrane proteome is an artifact of blood contamination. We therefore do not believe that ML256’s anti-target activity against CES1 inhibition will unduly affect biochemical profiling of pPAFAH in the brain. Regardless, we have identified other carbamate inhibitors that also inhibit CES1, but do not affect pPAFAH (e.g., WWL47, Table 3.1-1 [28], or the in vivo-active dual KIAA1363/CES1 inhibitor JW480 [29]), and therefore could be used as “anti-probes” for biological studies. We have previously employed a similar anti-probe strategy for the LYPLA1/2 dual inhibitor ML211, which showed inhibitory activity against anti-target ABHD11, for which we developed a selective, control “anti-probe”, ML226. Other than CES1, no other anti-targets in the brain membrane proteome are observed (Figure 3.5-2A, right panel), a clean selectivity profile that is further substantiated by more in depth LC-MS/MS based profiling as detailed in Section 3.6.

Demonstrated oral bioavailability, penetration of the blood-brain barrier to deliver near complete inhibition, and a largely clean selectivity profile strongly support the utility of this probe for in vivo functional and biochemical characterization of pPAFAH. It should be noted that the irreversible nature of ML256 greatly facilitates these kinds of detailed potency and selectivity analysis in living systems.

Cytotoxicity: The probe ML256 (SID 125269079) and analogs JMN4 (SID 125269060) and WWL153 (SID 99205832) were evaluated for cytotoxicity (AID 588768) in 293T Hek cells cultured in both serum-free and serum-supplemented medium. As shown in Figure 3.5-3, ML256 had significantly higher CC50 values than either analog, affording a large dosing window (50-fold) about the in situ IC50 value of <100 nM (Figure 3.5-1).

Figure 3.5-3

Cytotoxicity anaysis of probe ML256 and analogs JMN4 and WWL153 in serum-free (A) and serum-supplemented (B) medium (AID 588768). 293T Hek cells cultured in DMEM medium with and without 10% FCS were treated with test compound and viability was assessed (more...)

3.6. Profiling Assays

HTS Carbamates: To date, the HTS hit carbamates (Table 3.1-1) have all been tested in several hundred other cell-based and non-cell based bioassays deposited in PubChem, and have shown an average hit rate of 2.1%. The most reactive carbamate, A06, with a hit rate of 3.5%, also has a Michael acceptor functionality, which could explain its increased reactivity. This low hit rate indicates that this compound class may not be generally active. No HTS activity data are yet available for probe ML256 or any of the synthetic analogs, nor has ML256 been submitted for commercial or non-commercial broad panel screening.

Gel-based Competitive ABPP: Probe ML256 (SID 125269079) and analogs listed in Tables 3.4-1 and 3.4-2 have been subject to gel-based competitive ABPP screening to assess SH reactivity against more than 20 FP-sensitive SHs (including lipases, esterases, proteases, and uncharacterized hydrolases) visible by 1D SDS-PAGE separation and fluorescent detection. This medium-throughput proteome-wide screening technique was instrumental in our medchem optimization of the probe compound (Section 3.4), allowing rapid assessment of potency and selectivity, as visualized by disappearance of bands in compound-treated lanes relative to the DMSO-only control.

For comparative assessment of compound potency (AIDs 588773 [HTS hits] and 588774 [synthetic compounds]), Figure 3.6-1 shows inhibition of recombinantly-expressed mouse pPAFAH in membrane proteome preparation from transiently transfected 293T Hek cells overexpressing the enzyme at three compound concentrations (10 μM, 1 μM, and 0.1 μM). All data are summarized in Tables 3.4-1, 3.4-2, and S2. As a class, the triazole ureas were the most potent compounds (showing most complete inhibition at 0.1 μM compound concentration), but, as discussed in Section 3.4, they exhibited poor selectivity profiles (see Figure 3.6-3, lower left panel).

Figure 3.6-1

Potency of ML256 (SID 125269079) and analogs against recombinant mouse pPAFAH in a complex proteome (293T Hek cell membranes overexpressing the enzyme). For this gel-based competitive ABPP assay, proteome was incubated with test compound (10 μM, (more...)

Figure 3.6-3

Selectivity of synthetic SAR compounds as assessed by gel-based competitive ABPP in the mouse brain membrane proteome. See Tables 3.4-1 and 3.4-2 for summary of selectivity results. pPAFAH is not visible in these gels due to overlapping SH bands. See (more...)

For selectivity analysis, the competitive ABPP experiment was repeated in mouse brain membrane proteome (AIDs 588773 [HTS hits] and 588774 [synthetic compounds]). This proteome was chosen due to its relevance for future study of pPAFAH in the nervous system as well as its diversity of SHs. pPAFAH is not visible due to overlapping SH bands. Compounds were tested at the same three concentrations (10 μM, 1 μM, and 0.1 μM) for determination of fold selectivity as reported in the tables in Section 3.4. Proteins are listed as anti-targets (Tables Section 3.4 and S2) if at least 50% inhibition is observed (as quantified relative to the DMSO control) for a given compound concentration. HTS hits are shown in Figure 3.6-2 and synthetic analogs are shown in Figure 3.6-3. Overall, this gel-based competitive ABPP analysis revealed that ML256 (Figure 3.6-3, top left panel) shows no anti-target inhibition at 10 μM, corresponding to a large (>322-fold) selectivity window over all (>20) FP-sensitive SHs resolvable by 1D SDS-PAGE.

Figure 3.6-2

Selectivity of HTS hit compounds as assessed by gel-based competitive ABPP with FP-Rh in the mouse brain membrane proteome. See Tables 3.4-1 and S2 for summary of selectivity results. pPAFAH is not visible in these gels due to overlapping SH bands. See (more...)

Selectivity vs. PAFAH2: Because PAFAH2 is the closest sequence homolog to pPAFAH, we wanted to explicitly test the reactivity of ML256 (SID 125269079) towards this enzyme by gel-based competitive ABPP (AID 588767). For this assay, the soluble proteome of BW5147-derived murine T cells, which endogenously express PAFAH2, was treated with ML256 (10 μM, 1 μM, and 0.1 μM) or DMSO for 30 minutes followed by reaction with FP-Rh. Treatment with AA39-5 (SID 103913575), a compound known to inhibit PAFAH2 (see ML225 Probe Report), is shown as a control. No inhibition of PAFAH2 is observed at 10 μM ML256 concentration, giving fold selectivity of >322 for pPAFAH vs. PAFAH2.

Figure 3.6-4Selectivity of ML256 (SID 125269079) vs. PAFAH2, the closest sequence homolog of pPAFAH, as assessed by gel-based competitive ABPP with FP-Rh in the murine T-cell soluble proteome

No inhibition of anti-target PAFAH2 is observed at the highest concentration (10 μM) tested relative to the DMSO control, giving a fold selectivity of >322. Known PAFAH2 inhibitor AA39-5 (SID 103913575) is shown as a control. Fluorescent images shown in grey scale.

ABPP-MudPIT: To more comprehensively identify potential anti-targets, we utilized a quantitative LC-MS/MS-based platform termed competitive ABPP-MudPIT (Figure 3.6-5A) [49]. ABPP-MudPIT allows for robust quantitation of inhibited enzymes via comparison of the spectra counts peptides from control vs. inhibitor-treated samples. Paired samples of mouse brain membrane (AID 588785) and soluble (AID 588787) proteomes were treated with 1 μM of ML256 or DMSO (30 minutes) followed by reaction with the affinity-tagged FP-biotin activity-based probe (5 μM, 2 hours). Reactions were enriched with streptavidin, digested on-bead with trypsin, and analyzed by MudPIT [25, 26] using an LTQ Thermo Scientific instrument. Protein identifications and spectra counts were obtained by MS2 searching using the Prolucid (http://fields.scripps.edu/researchtools.php) search algorithm and filtered using DTASelect [50]. The depicted bar graphs show the relative spectra counts between inhibitor- and DMSO-treated samples for each SH identified. Enzymes susceptible to inhibition upon compound treatment would be expected to have significantly reduced spectra counts in the inhibitor-treated sample. The results (Figure 3.6-5B and C) demonstrate that ML256 exhibits significant selectivity for pPAFAH, blocking 100% of activity while not affecting any of the other 40+ SHs detected in the mouse brain proteome, with the exception of the liver enzyme CES1, consistent with gel-based profiling results (see Section 3.5). The other PAFAH family enzymes PAFAH1b2 and PAFAH1b3 also showed no evidence of ML256 inhibition.

Figure 3.6-5

Potency and selectivity of ML256 assessed by ABPP-MudPIT. A) Overview of the ABPP-MudPIT method. B and C) Inhibition profiles of ML256 (1 μM, 2 hour) in the mouse brain membrane (B) and soluble (C) fractions shows significant inhibition of only (more...)

Of note, we have previously determined that we could detect the SH fatty acid amide hydrolase (FAAH) at concentrations as low as 0.0005% of the total proteome (~200 copies per cell) by gel-based ABPP [18]. By comparison, we estimate that the sensitivity of a standard ABPP-MudPIT assay is at least 10-fold higher (0.00005% of the total cell proteome, or 20 copies per cell). As such, both gel-based and LC-MS/MS-based methods offer detection of low abundance SHs for a comprehensive selectivity analysis.

4.1. Comparison to existing art and how the new probe is an improvement

There were two existing pPAFAH inhibitors when we initiated our HTS campaign for inhibitor discovery: the carbamate WWL153 (Table 3.4-1, entry 9) and darapladib, a highly potent (IC50 270 pM) reversible inhibitor of pPAFAH (Figure 4.1-1) currently in phase III clinical trials for treatment of atherosclerosis [21, 51–55].

Figure 4.1-1

Structure of darapladib, a GlaxoSmithKline drug in phase III clinical trials for the treatment of atherosclerosis.

WWL153 was our best lead inhibitor for pPAFAH inhibition, and ML256 is a significant improvement in potency (>5-fold), selectivity (two orders of magnitude), and cytotoxic properties (>100-fold). We have also demonstrated that ML256 is active both in situ and in vivo. The inhibitory properties of WWL153 in living systems have not been investigated.

Darapladib is a highly potent inhibitor of pPAFAH that operates by a reversible (non-covalent) mechanism, in contrast to the irreversible, covalent mechanism of ML256. As demonstrated in Section 3.5, a key advantage of ML256 is that its irreversible mechanism of action enables, when combined with ABPP methods, direct confirmation of pPAFAH inhibition in vivo. Thus, persistent questions, such as the endogenous substrates of pPAFAH and the enzyme’s function in the mammalian nervous system, can be assessed in mouse models with ML256. Such studies are more difficult to perform with a reversible inhibitor like darapladib, given that there are no known biomarkers that can be used to confirm the reversible inhibition of pPAFAH in living systems. As such, we view the irreversible probe ML256 as a critical addition to the chemical tools available for manipulation of pPAFAH activity for biochemical investigation.

Additionally, while darapladib is not commercially available (and very difficult to obtain from GlaxoSmithKline in sufficient quantities for comprehensive biological studies), we will make ML256 available to any investigator interested in studying pPAFAH biology. Unfortunately there is also no straightforward way to circumvent darapladib commercial unavailability through in-house synthesis. There is no detailed synthetic route published for the compound, and the synthetic overview provided in Bioorg. Med. Chem. Lett. [21] was not sufficiently detailed to allow, in our hands, successful resynthesis of the compound. In contrast, ML256 is easily synthesized in three steps from commercially available starting materials and a detailed protocol is included in Section 2.3.

Finally, one key aim of our HTS inhibitor discovery campaign was to find new scaffolds for co-crystallization studies of pPAFAH-inhibitor complexes to further define the active site and substrate specificity of pPAFAH. Attempts to crystallize pPAFAH with darapladib have been unsuccessful; however, the medchem campaign to optimize a lead carbamate inhibitor for pPAFAH has generated to a diversity of potent analogs that may lend themselves to crystallographic study (in progress).

4.2. Mechanism of Action Studies

As determined from LC-MS/MS analysis (AID 588788), the probe is an activity-based inhibitor that covalently labels the active site serine nucleophile, Ser273, of pPAFAH (section 3.3). The observed mass shift of the active site peptide is consistent with carbamoylation of the enzyme (Figure 3.3-1).

4.3. Planned Future Studies

We plan to comprehensively establish parameters for in situ and in vivo use and explore the target specificity of ML256 with application of alkyne analogs via click chemistry ABPP. For biological application, we plan to use ML256 as a probe for investigation of the potential role of pPAFAH in inflammatory processes by global proteomic and metabolomic profiling of pPAFAH in in vivo (mouse) and in situ (cultured cell) models of oxidative stress to identify potential metabolic pathway involvement. We will also continue to investigate ML256 and its analogs as tools for structural studies of pPAFAH active site architecture.