Materials and Reagents

• Steady-Glo® Luciferase Assay System was purchased from Promega (Catalog No. E2550; Madison, WI)
• CellTiter-Glo® Luminescent Cell Viability Assay was purchased from Promega (Catalog No. G7573; Madison, WI)
• Cells to CT Bulk Lysis reagents purchased from Ambion (Catalog No. 4391851C; Grand Island, NY)
• Cells to CT Bulk RT reagents purchased from Ambion (Catalog No. 4391852C; Grand Island, NY)
• Light Cycler 480 Probes Master purchased from Roche (Catalog No. 4887301001)
• Human GAPD (GAPDH) Endogenous Control VIC/MGB probe/primer limited purchased from Applied Biosystems (Catalog No. 4326317E; Grand Island, NY)
• Human MITF FAM probe/primer set purchased from Applied Biosystems (Catalog No. 4331182 Hs01117294_m1; Grand Island, NY)
• Human TRPM1 probe/primer set purchased from Applied Biosystems (Catalog No. 4331182 Hs00170127_m1; Grand Island, NY)
• Human CDK2 probe/primer set purchased from Applied Biosystems (Catalog No. 4331182 Hs01548894_m1; Grand Island, NY)
• Human DCT probe/primer set purchased from Applied Biosystems (Catalog No. 4331182 Hs01098278_m1; Grand Island, NY)
• Human MLANA probe/primer set purchased from Applied Biosystems (Catalog No. 4331182 Hs00194133_m1; Grand Island, NY)

Cell Lines

The following cell lines were used in this study:

• TRPM-1:luc is a SK-MEL-5 melanoma cell line that expresses firefly luciferase under the control of the melastatin (TRPM-1) promoter. This cell line was used in the primary HTS campaign and generated by the Fisher Lab.
• SK-MEL-5 is the parental cell line to SKMEL5 TRPM-1:luc cells, and does not contain the luciferase reporter. This cell line was obtained from ATCC (Catalog Number HTB-70; Manassas, VA).
• A375 obtained from ATCC (Catalog Number CRL-1619; Manassas, VA) is a melanoma cell line characterized to be independent of MITF for growth and survival
• MALME-3M was obtained from ATCC (Catalog Number HTB-64; Manassas, VA) is a melanoma cell line dependent upon MITF activity for growth and survival.

2.1. Assays

A summary listing of completed assays and corresponding PubChem AID numbers is provided in Appendix A (Table A1). Refer to Appendix B for the detailed assay protocols.

2.2. Probe Chemical Characterization

The probe was prepared as described in Section 2.3 and Appendix C. ¹H NMR, ¹³C NMR and HRMS were used to characterize this compound and the results are consistent with the proposed structure. UPLC purity of the material studied was 100% at 214 nm.

2.3. Probe Preparation

The probe (ML329) was synthesized in one step from commercially available 1,4-naphthoquinone and sulfanilamide using cerium(III) chloride heptahydrate as a Lewis acid catalyst as shown in Scheme 1. The reaction was allowed to stir at 75 °C for three days (unoptimized) then dilute citric acid was added to the reaction suspension and the insoluble material was collected by filtration. The filter cake was washed with water, dried, then purified by preparative RPLC.

Scheme 1

Synthesis of the Probe (ML329).

Experimental procedures for the synthesis of the probe can be found in Appendix C.

2.4. Additional Analytical Analysis

ML329 was tested for stability in PBS/1% DMSO (Figure 1A), human serum and mouse serum. ML329 was also tested for solubility and GSH/DTT adduct formation (Figures 1B/1C). All results indicate that ML329 is very stable in PBS/1% DMSO, the GSH/DTT adduct assays, and in the different serums. Experimental procedures for the analytical assays are provided in Appendix D.

Figure 1

Stability of ML329 in PBS Buffer & GSH and DTT Stability Assays. MLS004556029 (CID 12387471, SID 144221520) was tested over a time course in a PBS stability assay (A), GSH stability assay over 6 hours (B), DTT stability assay over 48 hours (C), (more...)

Probe attributes

• Has an average IC₅₀ value of 1.2 μM (+/- 0.1 μM) in the TRPM-1 reporter assay
• Shows activity in two MITF-dependent melanoma cell lines (SK-MEL-5 & MALME-3M)
• No apparent cytotoxicity in A375, a MITF-independent cell line (EC₅₀= >35 μM).
• Shows dose-dependent inhibition of the expression of multiple MITF target in multiple qPCR assays (TRPM-1, MITF, CDK2, DCT, MLANA)

To identify novel inhibitors of MITF, TRPM-1 promoter activity was measured with an engineered cell line that allows for luminescence detection via the TRPM-1 promoter in a MITF-dependent melanoma cell line, SK-MEL-5. The TRPM-1 promoter responds dynamically to alterations of MITF function and is more sensitive than a standard cell cytotoxicity assay (17). This primary cell-based assay was used in a pilot screen that identified parthenolide, the positive control, as a compound that could decrease the luminescence signal. The assay was optimized for automation and a high throughput screen was pursued. The automated assay was first tested with a validation set of 2,240 compounds. These compounds were tested in duplicate with in-plate neutral (DMSO) and positive controls (18 μM parthenolide). Robustness, reproducibility, and variability parameters were analyzed before initiating the full HTS. The HTS was run over the course of several weeks. Data were normalized relative to controls, and plate patterns were corrected using a multiplicative algorithm in Genedata Screener Assay Analyzer. For each compound, the average of the two replicates was determined and used for subsequent analysis.

Determination of hits required several criteria: only assay plates with a Z′ greater than 0.5 were accepted for analysis, compounds needed to reach 60% inhibition relative to 18 μM parthenolide, and score in fewer than 10% of HTS assays listed in PubChem. In total, 331,578 compounds were screened. Of these, 3,206 compounds were considered active (a hit rate of 0.96%), 613 were inconclusive, and 315,055 were inactive. 1,064 compounds scored in over 10% of assays listed in Pubchem and were excluded from further study. Compounds were clustered based upon chemical structure using a customized script in Pipeline Pilot. Clusters were rated based upon structural liabilities and ranked accordingly. Substructures were analyzed and compared to inactive compounds to identify inactive analogs. Representatives from the more desirable clusters were selected for retest. In addition, a small number of inactive analogs were chosen to provide initial structure/activity relationship (SAR) data during the retest studies. Of these, 1,379 compounds were retested over a range of concentrations to validate activity, and 583 compounds showed dose-dependent inhibition of TRPM-1 expression below IC₅₀ of 10 μM. 261 compounds were below IC₅₀= 5 μM.

Since active compounds produce a decrease in signal in the primary assay, confirmation was required that the reduction in luminescence was not a result of general cytotoxicity by the compound or by inhibition of luciferase itself. Compounds were tested in several melanoma cell lines, including A375, a human melanoma cell line that proliferates despite very low levels of MITF protein (17). In recent studies, siRNA-mediated knockdown of MITF does not alter A375 viability or growth rates and seems to persist independent of MITF regulation (4; Fisher lab, unpublished data). In addition, compounds were tested for cytotoxicity in the parental cell line SK-MEL-5 and another MITF dependent cell line, MALME-3M, to verify dose-dependent inhibition of cell proliferation. Of the compounds that retested in the primary assay, 48 compounds showed the appropriate activity in the SK-MEL-5, MALME-3M and A375 assays.

All available dry powder samples of the remainder qualifying compounds were procured from commercial sources and purity was determined. Only compounds over 90% pure were tested. The compounds were retested in the primary assay and in the cytotoxicity assays. 11 compounds showed an IC₅₀ value of less than 35 μM in A375 cells and were excluded from further consideration. Approximately 44 compounds had an IC₅₀ value below 10 μM in the primary assay. Fifteen chemical scaffolds showed potencies below 10 μM, and three of these were prioritized for follow-up assays. MITF is known to regulate the expression of a number of lineage-specific melanogenesis genes and cell cycle regulators. Therefore, compounds were screened for inhibition of gene expression in qPCR assays for the following genes: MITF, TRPM-1, MLANA, BCL2A, and CDK2. Three scaffolds were put forward as candidates for medicinal chemistry but only one scaffold met all of the assay cutoffs and was amenable to analog production. The lead compound was CID 1716436 (Figure 2).

Figure 2

The structure of MLS000680589 (CID 1716436).

As of 01 November 2012, the original hit (SID 22416871, CID 1716436, MLS000680589, BRD-K45681478-001-06-9) is listed as active in 48 of 617 PubChem assays (6 of which are part of the MITF project). It registered an IC₅₀ value in the primary assay of 4.1 μM (Figure 3A, AID 493177), and showed no cytotoxicity after 24 h in A375 cells (Figure 3B, AID 540335). In addition, it showed potency close to the IC₅₀ cutoff of 10 μM in the SK-MEL-5 and MALME-3M cytotoxicity assays (Figure 3C and 3D, respectively). With its good potency and lack of cytotoxicity in A375 cells, it was prioritized for further development. Analogs were designed and synthesized, eventually leading to the more potent and more soluble probe, ML329 (see Section 3.4).

Figure 3

Dose response curves for initial hit, MLS000680589. MLS000680589 (SID 22416871, CID 1716436) was tested across a range of concentrations up to 35 μM in the primary assay and several secondary assays. Concentration response curves were generated (more...)

Table 1Probe and Selected Analogs Assay Performance (Average IC₅₀, μM)

View in own window

CID / SID	Broad ID / KU ID	Structure	TRPM-1 promoter assay	A375 CTG	SK-MEL-5 CTG	MALME-3M CTG	TRPM-1 qPCR
12387471 / 144221520	BRD-K73037408-001-01-5 / KUC111774N-03		1.2	70	0.1	0.7	0.15
81124 / 144221523	BRD-K29842115-001-07-4 / KUC111337N-02		5	60	2.7	3.4	0.6
72909 / 144221522	BRD-K75502546-001-01-9 / KUC111359N-02		3.2	64	8.5	7.3	2.4
279009 / 144221524	BRD-K48101260-001-01-8 / KUC111761N-02		6.6	70	1.2	13.5	IA
56951846 / 144221521	BRD-K93744531-001-01-2 / KUC111358N-02		7.7	70	14	18.6	6
56928031 / 135611193	BRD-K16604218-001-01-2 / KUC111109N		41.7	64	70	63	NT

NT= not tested, IA=inactive (IC₅₀ >35 μM), Values in bold fail to meet probe criteria

3.1. Summary of Screening Results

Refer to subsections for a detailed description of the results.

In the primary HTS, compounds were active if they decreased TRPM-1 expression as measured by luminescence. The positive control, parthenolide (18 μM), caused a decrease in expression that led to over a 6-fold reduction in signal. Compounds with greater than 60% activity of the positive control were considered actives and chosen for confirmation studies (see Section 3.4 for details). Figure 4 displays the critical path for probe development. To explore SAR, numerous analogs were synthesized and tested. Selected results are shown in Tables 2 to 8 in Section 3.4.

Figure 4

Critical Path for Probe Development.

Table 2

Nascent SAR derived from the HTS results.

Table 8

Round 2 SAR and Hydrogen-substituted Naphthoquinone compounds.

3.2. Dose Response Curves for Probe

Figure 5Dose-dependent Activity of the Probe (ML329)

MLS004556029 (CID 12387471, SID 144221520) was tested across a range of concentrations up to 35 μM in the primary assay and several secondary assays. Concentration response curves were generated with Genedata Screener Condeseo and show normalized percent activity for the individual doses. TRPM-1 promoter assay (AID 651588), IC₅₀= 1.2 μM (A); A375 cytotoxicity assay (AID 651591), IC₅₀>35 μM (B); SK-MEL-5 CellTiter-Glo (AID 651586), IC₅₀= 0.75 μM (C) and MALME-3M CellTiter-Glo (AID 651585), IC₅₀ = 0.88 μM (D). ○=replicate 1, △=replicate 2, ◻=replicate 3, ◇=replicate 4

3.3. Scaffold/Moiety Chemical Liabilities

The solubility of the probe (ML329) was experimentally determined to be 0.4 μM in phosphate buffered saline (PBS) solution (pH 7.4). The probe is stable in PBS buffer solution, with 100% remaining after 48 hours (see Figure 1A). While the probe could be imagined to be reactive toward a thiol nucleophile, the probe was found to be quite stable to glutathione (GSH), with 80% remaining after 6 hours (see Figure 1B), and dithiothreitol (DTT), with 100% remaining after 48 hours (see Figure 1C). The probe is stable in human plasma with approximately 100% remaining after a 5 hour incubation. See the Appendix D for the experimental procedures for the solubility, PBS stability, GSH/DTT stability and plasma stability measurements.

3.4. SAR Tables

The naphthoquinone core was chosen based on potency and selectivity for MITF inhibition, while remaining non-toxic to MITF independent melanocytes. Our focus was on improving the biological activity and chemical stability through analog synthesis. To begin, a more detailed analysis of the primary high throughput (AID 488899) and associated initial screens (AID 493073, AID 493191, AID 493240 and AID 540346) was performed (Table 2). Of the 341,348 compounds reported as tested in AID 488899, there were 347 compounds that contained the benzoquinone substructure without containing the 2-chlorobenzoquinone substructure. This list was further culled by only allowing benzoquinones with a single fusion to another ring (173 compounds after filter). A final structural filter was imposed where structurally complex and/or synthetically challenging compounds were excluded to allow for efficient analog production. The last two filters were based on compound biological promiscuity data: 140/154 compounds had PubChem Active_AID_ratios of ≤ 15%, where the Active_AID_ratio is the ratio of the number of PubChem bioassays where the compound tested active divided by the total number of bioassays in which the compound was tested, as a rough surrogate for general bioactivity promiscuity; 37/140 had a RankScore (a measure of activity) in AID 488899 of at least 60. All of these compounds except one (SID 92764352; CID 386492; RankScore of 66 in AID 488899; not shown in Table 2) had a 1,4-naphthoquinone (or quinoline-5,8-dione) core.

Some of the compounds from the high throughput screen had been tested in the confirmatory and secondary assays and these results are shown in Table 2. Therefore, analysis of the compounds centered around the RankScore and confirmatory activities, where available. Since there was not a strong correlation between these two numbers within the activity ranges of interest (≥60 RankScore and < 15 μM AC₅₀), we used the structures in Table 2 to guide us in forming a small matrix-type library including benzamide, sulfonamide, cyclic amine, and aniline pharmacophores. A selection of these compounds was synthesized and tested, with some additional compounds added to probe the structural requirements and tolerances of the system.

The biological assay data and physical properties of these analogs are presented in Tables 3-8. Characterization data (¹H NMR spectra and UPLC chromatograms) of these analogs are provided in Appendix F.

Table 3

Round 1 SAR and Anilino-substituted Naphthoquinone Compounds.

In Table 3, direct chloro substitution on the naphthoquinone core (Table 3, entry 1) gave desirable activity, but these analogs were found to react with glutathione and were removed from further consideration as probe candidates. Maintaining the anilino groups (Table 3, entries 2-11) consistently provided active compounds, although the best anilino compounds were appended with piperidine, piperazine, methoxy or hydrogen substituents (Table 3, entries 2-5). Direct methyl substitution on the naphthoquinone core (Table 3, entry 15) attenuated target potency, but removing the anilino functionality (Table 3, entry 13) maintained desirable potency and selectivity. Appending the naphthoquinone with tertiary nitrogens (Table 3, entry 11) also provided active compounds though potency was lost. Two potential probe candidates were noted from this particular series (i.e., Table 3, entries 4 and 13), after the first round SAR feedback.

Table 4 illustrates that replacing the anilino moiety with N-methylacetamides provided an increase in potency, but at the expense of increasing toxicity shown by the antitarget activity (Table 4, entries 1-6). N-methylacetamides were removed from further consideration due to their associated toxicity.

Table 4

Round 1 SAR and N-methylacetamide-subsituted Naphthoquinone Compounds.

Next, we evaluated a series of benzoyl substitutions on the naphthoquinone ring, and the results are summarized in Table 5. Direct comparison of anilino compounds (e.g., Table 3, entries 2, 6, and 9) with their benzoyl counterparts (e.g., Table 5, entries 1, 5, and 6) illustrates that the benzoyl series gave comparable target potencies, while the antitarget potencies often diminished below the advantageous absolute cut-off limits (e.g., Table 5, entries 1-3 and 5). Appending the benzoyl moiety with electron-withdrawing groups resulted in similar activity (Table 5, entry 2) as did extending the alkyl chain pendant to the piperazine ring (Table 5, entry 3). However, a phenyl substitution on the piperazine ring led to a significant loss in potency (Table 5, entry 6; Table 3, entry 9).

Table 5

Round 1 SAR and Benzoyl-substituted Naphthoquinone compounds.

Table 6 contains data from benzene- and thiophene-substituted sulfonamide analogs. Halogen substitutions improved the activities of thiophene analogs (Table 6, entry 3 and 4) but were detrimental to other sulfonamides (Table 6, entry 1 and 2). Naphthoquinones with -H, -OMe, and –NH₂ substitutions proved equipotent to the piperazine analogs (Table 6, entries 5, 6, 8, and 10) but piperidine substitution attenuated potency (Table 6, entry 7).

Table 6

Round 1 SAR Benzene- and Thiophene-substituted Naphthoquinone Compounds.

Appending electron-withdrawing groups to the anilino moiety, as shown in Table 7, gave a slight increase in potency (Table 7, entries 1 and 2) compared to a slight loss of activity when electron-donating groups were used (Table 7, entries 1 and 5). Potencies were improved further by exchanging the N-methylpiperazines for N-ethyl- or N-isopropylpiperazines (e.g., Table 7, entries 5-7). The piperidine and pyrrolidine substituents were generally more potent towards the antitarget than the piperazines (Table 7, entries 8-11) and the morpholino-like analogs lost target potency when compared in the SKMEL5 and MALME-3M assays (Table 7, entries 12 and 13). Anilino analogs presented in Table 7 had enhanced potencies over those presented in Table 3, but still did not satisfy the probe criteria.

Table 7

Round 2 SAR and Anilino- and Nitrogen-heterocycle-substituted Naphthoquinone Compounds.

Table 8 shows the SAR results of the hydrogen-substituted naphthoquinones, which were designed and synthesized based on the two potential probe candidates from the first round SAR (Table 8, entries 1 and 3). Adding heteroatoms into the naphthoquinone ring potentiated the antitarget potency (Table 8, entry 4). Ortho- and para-electron-withdrawing groups pendant to the anilino moiety were well tolerated (Table 8, entries 6, 9, and 11) but meta- groups attenuated potencies (Table 8, entries 7 and 10). para-Chloro substitution also attenuated target potency (Table 8, entry 8). Extending the aromatic ring with a methylene linker was a significant detriment to activity (compare Table 8, entries 8 and 12). Substitution at the para-position with a sulfonamide resulted in the most potent compound that retained selectivity and therefore, was nominated as the probe (Table 8, entry 9).

3.5. Cellular Activity

The primary assay and all of the secondary assays were cell-based, all lead compounds showed activity in cells. In the medicinal chemistry phase of our studies, ML329 and its analogs were tested in multiple cell-based experiments. ML329 consistently showed low micromolar activity in multiple cell-based assays and has good permeability and solubility properties. ML329 also showed potent activity in primary human melanocytes (IC₅₀= 7 μM, AID 651920) (Appendix I). The counterscreen with A375 cells helped to remove compounds that possessed non-specific cellular toxicity.

3.6. Profiling Assays

ML329 was run in a panel of assays with PanLabs/Eurofins and showed little to no inhibition with the exception of human Adenosine A_2A where it showed 50% inhibition at 10 μM. See Appendix H for more details.

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

Investigation into relevant prior art entailed searching the following databases: SciFinder and the Thomson Reuters Integrity. The search terms applied and hit statistics for the prior art search are provided in Table A2 (Appendix G). Abstracts were obtained for all references returned and were analyzed for relevance to the current project. For all references that were deemed relevant, the articles were analyzed and the results are summarized below. The searches are current as of October 9, 2012.

Table A2

Search Strings and Databases Employed in the Prior Art Search.

Figure 7Reported inhibitors of MITF

Since the start of the project, a modest amount of small-molecule prior art has been reported in the scientific literature, especially in the year 2012. However, it is not known if these small-molecule MITF inhibitors recapitulate genetic manipulation or deletion of MITF. Prior to the start of this probe project, the pilot screen identified a compound, parthenolide, that decreased TRPM-1 expression but was inadequate in the other cell-based assays because of a lack of killing in MITF-dependent cell lines. It was used as the positive control for the primary assay for this project. This compound does not show selectivity to MITF and potently kills some MITF-independent melanoma and other cancer cell lines. 2-amino-3H-phenoxazin-3-one (APO) was shown to reduce melanogenesis and the expression of MITF (18). We tested APO and found that it potentially inhibited TRPM-1 promoter activity (IC₅₀= 2 μM) but also killed A375 cells (IC₅₀= 9 μM) suggesting a mode of action nonspecific to MITF. Another report described “compound-17”, a small molecule that inhibited MITF binding to E-box DNA motifs that are found in the promoters of several pigment cell-specific genes like TRPM-1 (19). Compound-17 showed activity with 1 mM of compound in an electrophoretic mobility shift assay (EMSA) but did not alter TRPM-1 protein levels with up to 50 μM of compound after 72 hours of treatment. Although we have not measured TRPM-1 protein levels, we have measured promoter activity in the luciferase assay and mRNA transcript level by qPCR. We detect changes with low micromolar amounts of ML329 after 24 hours of treatment in both the luciferase and qPCR TRPM-1 assays (Figure 5A and 8A). One lead compound (CID 9616928) was eliminated because of lack of potency in SK-MEL-5 cell cytotoxicity, but notably, is similar to the HDAC inhibitor, panobinostat/LBH589 (CID 6918837), which has been shown to reduce MITF expression levels and reduce melanoma tumor growth in vivo (20). LBH589 potently reduced TRPM-1 expression in the luciferase reporter assay but failed to meet the other probe criteria for the remaining cell-based assays. In particular, LBH589 was not potent enough in SK-MEL-5 cells and led to over 50% reduced viability in A375 cells. A number of compounds have been reported to inhibit melanogenesis or MITF but none have been shown to directly interact with MITF and so were not tested against ML329. Several of these substances with low micromolar activity are listed in Table 10.

Figure 8

qPCR for MITF and several target genes. MLS004556029 (CID 12387471, SID 144221520) was tested across a range of concentrations up to 35 μM in SK-MEL-5 melanoma cells for multiple qPCR assays. Concentration response curves were generated with Genedata (more...)

Table 10

Compounds reported to inhibit MITF.

4.2. Mechanism of Action Studies

The assays utilized in this project were phenotypic, cell-based assays that leave some ambiguity to how ML329 works in cells. The assays used in this project only show circumstantial evidence of ML329 regulating MITF and its molecular pathways. The compound could be acting on an upstream activator of MITF since we see a reduction of MITF transcription and several of its target genes by qPCR (Figure 8B, AID 651773). MITF expression is regulated by a number of transcriptional regulators including: MITF itself, BRAF, WNT signaling, SOX10, CREB, and Pax3 (2). ML329 reduced the expression of multiple MITF target genes suggesting a mechanism that impacts a broader profile of MITF targets, including both melanogenesis and cell cycle genes such as CDK2 (Figure 8C). ML329 consistently lowered the expression of several MITF target genes including: melastatin (TRPM-1), dopachrome tautomerase/tyrosinase-related protein-2 (DCT), the cell cycle regulator cyclin-dependent kinase-2 (CDK2) and melan-A (MLANA/MART1). The primary screen was carried out in a BRAF(V600E) mutated melanoma cell line, SK-MEL-5. It is quite clear that ML329's activities are very different from those of BRAF inhibitors, something studied in the Fisher lab quite extensively. BRAF-MEK-MAPK lead to proteolysis of MITF, and thus MAPK pathway suppression leads to MITF stabilization/up-regulation. We have observed (and published) that multiple MITF target genes are up-regulated following BRAF inhibitor treatments. In contrast we find that ML329 suppresses both MITF and multiple of its transcriptional target genes. We believe that MITF up-regulation following BRAF suppression represents a survival mechanism that limits efficacy of BRAF targeted therapies. Therefore, it is plausible that concurrent use of MITF antagonists (like ML329) may offer significant benefit in combination with BRAF inhibitors. Proper determination of mechanism of action will require a number of different studies, some of which are outlined in Section 4.3.

4.3. Planned Future Studies

Figure 9Suggested medicinal chemistry to improve cellular activity

Our recommendations for future SAR expansion around the probe (ML329) involve the following: (1) substitute the anilino nitrogen with various alkyl or acyl groups, (2) append the sulfonamide nitrogen with mono- and bis-substituted amines (i.e., methyl, morpholine, piperidine, or pyrrolidine), (3) move the sulfonamide around the ring to which it is currently attached, (4) add substituents around the ring system adjacent to the sulfonamide, (5) replace the aniline-benzenoid moiety with other heterocyclic rings, (6) substitute the C3-position of the naphthoquinone ring system with methyl and methoxy, (7) replace the benzenoid ring of the naphthoquinone core with various heterocyclic rings (i.e., pyridine and piperidine), and (8) substitute the benzenoid ring of the naphthoquinone core with mono- and multi- substituted electron-withdrawing and/or electron-donating groups, as well as groups to probe steric requirements.

ML329 was tested with three melanoma cell lines, primary melanocytes, the TRPM-1 reporter cell line and A375 cells. Only one of these cell lines, A375, does not rely upon MITF for proliferation. There are a number of other melanoma cell lines available (e.g. M14, UACC62, UACC257, 501mel, SK-MEL-28, & SK-MEL-2) and the probe will be tested in these other melanoma cell lines. In addition, MITF has been implicated in clear cell sarcoma and ML329 will be tested in tumor-derived sarcoma cell lines to determine if it maintains activity in a different disease context (4). ML329 was tested with primary melanocytes from a single human donor and will be tested with other donor-derived melanocytes to verify consistency across several genetic backgrounds.

For mechanism of action studies, ML329 could be tested using biophysical experiments to determine direct binding of the compound to the MITF protein. Differential scanning fluorimetry (DSF) or thermal shift is a means of detecting binding of a ligand to purified protein. This technique is routinely used at the Broad Institute and could be applied to MITF. ML329 leads to a decrease in the expression of MITF and a number of target genes. It is not known if the DNA binding capacity of MITF is impaired. ML329 could prevent binding of MITF to E-box domains within promoter elements. An EMSA experiment could address whether binding of MITF to E-box DNA sequences are impacted by the presence of the compound. A weak MITF inhibitor has reported activity in an MITF/E-box EMSA (19). A recurrent mutation in MITF has been associated with familial and sporadic melanoma. This mutation correlates with impaired sumoylation (20). We can test the compound in the context of this mutation to determine if the potency is altered or if inhibition is maintained with this mutant version of MITF.

A gene-expression profiling experiment of SK-MEL-5 cells in the absence or presence of the ML329 will be performed. We plan to test ML329 in a Luminex-based platform called L1000 performed at the Broad Institute where gene expression of 1000 genes is measured after 6 and 24 hours of compound treatment in a panel of 20 cell lines as part of the LINCS program (https://commonfund.nih.gov/LINCS/). Tools developed at the Broad Institute, such as GenePattern and Gene Set Enrichment Analysis (GSEA), will be applied to determine the comparative markers responsible for the greatest difference between the different states. A computational method also developed at the Broad called the Connectivity Map, or cmap, that facilitates target identification through the analysis of small-molecule signatures will also be applied (43). This tool uses gene-expression profiles in a systematic approach to enable the discovery of functional connections between diseases, genetic perturbations, and small-molecule perturbations.

A number of melanoma models that utilize xenografts on rodents are used to determine the capacity of a drug or probe on melanoma (44). These studies can be done by topical administration of the test substance onto the tumor or intraperitoneal injection. We can evaluate ML329 in an analogous tumor graft model.

SK-MEL-5 TRPM-1 Luciferase Reporter Assay (2084-01)

SK-MEL-5/TRPM1Luc Culture Medium

• DMEM (High Glucose, HEPES, Phenol Red), Invitrogen Catalog No. 12430-047
• Fetal Bovine Serum (10%), Thermo-Hyclone Catalog No. SH30071.03
• Pen-Strep-Glutamine (1%), Invitrogen Catalog No. 10378-016
• Hygromycin (250 ug/mL), Invitrogen 10687-010

SK-MEL-5/TRPM1Luc Plating Medium: DMEM (High Glucose, no Phenol Red), Invitrogen Catalog No. 31053-036, Fetal Bovine Serum (10%), Thermo-Hyclone Catalog No. SH30071.03, Pen-Strep-Glutamine (1%), Invitrogen Catalog No. 10378-016

Steady Glo Promega Catalog No. E2550

Parthenolide Enzo Catalog No. BML-T113-0250

SK-MEL-5/TRPM1Luc cells were maintained in DMEM (10%FBS, 1% Pen-Strep-Glutamine, 250 ug/mL Hygromycin). Cells were fluid changed every 3 days and/or split upon reaching 100% confluency. For the primary HTS, cells were thawed at 4 million cells per Falcon T175 flask. After 3 days, the cells were fluid changed. After 3 more days, the cells were passed to a Corning triple flask (10-15 million cells) and plated after 3 days in the triple flask.

Day 1

1. Plate TRPM-1 luc/SKMEL5 cells at 2000 per well in 30 L media (phenol red free DMEM/10% Fetal Bovine Serum/Penicillin/Streptomycin/L-Glutamine)
2. Use Corning white 384-well, square, opaque-bottomed plates (Corning Catalog No. 8867BC)

Day 2

3.

Pin 100 nL compound/DMSO solution (Cybi Well) into assay plates. (For HTS, required sentinel pinning with the positive control, parthenolide (6 mM))

4.

Incubate 24 hours at 37°C in Liconic incubator.

Day 3

5.

Add 20 μL 100% Promega SteadyGlo per well with Thermo Combi fluid transfer apparatus.

6.

Shake 15 seconds on “big bear” plate shaker and incubate at room temperature for 5 minutes.

7.

Read on the Perkin-Elmer EnVision plate reader with ultra-sensitive luminescence (US LUM) settings for 0.5 sec per well

SK-MEL-5 Cytotoxicity Assay (2084-02)

SK-MEL-5 Cells ATCC Catalog No. HTB-70, lot 58483232, passage 28

SK-MEL-5 Culture Medium

• DMEM (High Glucose, HEPES, Phenol Red), Invitrogen Catalog No. 12430-047
• Fetal Bovine Serum (10%), Thermo-Hyclone Catalog No. SH30071.03
• Pen-Strep-Glutamine (1%), Invitrogen Catalog No. 10378-016

SK-MEL-5 Plating Medium

• DMEM (High Glucose, no Phenol Red), Invitrogen Catalog No. 31053-036
• Fetal Bovine Serum (10%), Thermo-Hyclone Catalog No. SH30071.03
• Pen-Strep-Glutamine (1%), Invitrogen Catalog No. 10378-016

SK-MEL-5 cells were maintained in DMEM (10%FBS, 1% Pen-Strep-Glutamine). Cells were fluid changed every 3 days and/or split upon reaching 100% confluency. For secondary assays, cells were thawed at 4 million cells per Falcon T175 flask. After 3 days, the cells were fluid changed, after 3 more days cells were passed to a Corning Triple flask (10-15 million cells) and plated after 3 days in the triple flask.

Day 1

1. Plate SK-MEL-5 cells at 3,000 per well in 30 μL media (phenol red free DMEM/10% Fetal Bovine Serum/Penicillin/Streptomycin/L-Glutamine)
2. Use Corning white 384-well, square, opaque-bottomed plates (Corning Catalog No. 8867BC)

Day 2

3.

Pin 100 nL compound/DMSO solution (Cybi Well) into assay plates. (For HTS, required sentinel pinning with the positive control, parthenolide (6 mM))

4.

Incubate 24 hours at 37°C in Liconic incubator.

Day 3

5.

Add 20 μL 100% Cell Titer GLO per well with Thermo Combi fluid transfer apparatus.

6.

Shake 15 seconds on “big bear” plate shaker and incubate at room temperature for 5 minutes.

7.

Read on the Perkin-Elmer EnVision plate reader with luminescence (LUM) settings for 0.1 sec per well.

A-375 Cytotoxicity Assay (2084-03)

A-375 Cells ATCC Catalog No. CRL-1619, Lot No. 58463364, passage 166

A-375 Culture Medium

• DMEM (High Glucose, HEPES, Phenol Red), Invitrogen Catalog No. 12430-047
• Fetal Bovine Serum (10%), Thermo-Hyclone Catalog No. SH30071.03
• Pen-Strep-Glutamine (1%), Invitrogen Catalog No. 10378-016

A-375 Plating Medium

• DMEM (High Glucose, no Phenol Red), Invitrogen Catalog No. 31053-036
• Fetal Bovine Serum (10%), Thermo-Hyclone Catalog No. SH30071.03
• Pen-Strep-Glutamine (1%), Invitrogen Catalog No. 10378-016

A-375 cells were maintained in DMEM (10%FBS, 1% Pen-Strep-Glutamine). Cells were fluid changed every 3 days and/or split upon reaching 90% confluency. For secondary assays, cells were thawed at 2 million cells per Falcon T175 flask. After 3 days, cells were passed to a Corning Triple flask (6-8 million cells) and plated after 3 days in the triple flask.

Day 1

1. Plate A-375 cells at 3,000 per well in 30 μL media (phenol red free DMEM/10% Fetal Bovine Serum/Penicillin/Streptomycin/L-Glutamine)
2. Use Corning white 384-well, square, opaque-bottomed plates (Corning Catalog No. 8867BC)

Day 2

3.

Pin 100 nL compound/DMSO solution (Cybi Well) into assay plates. (For HTS, required sentinel pinning with the positive control, parthenolide (6 mM))

4.

Incubate 24 hours at 37°C in Liconic incubator.

Day 3

5.

Add 20 μL 100% Promega Cell Titer GLO per well with Thermo Combi fluid transfer apparatus.

6.

Shake 15 seconds on “big bear” plate shaker and incubate at room temperature for 5 minutes.

7.

Read on the Perkin-Elmer EnVision plate reader with luminescence (LUM) settings for 0.1 sec per well.

MALME-3M Cytotoxicity Assay (2084-04)

MALME-3M Cells ATCC Catalog No. HTB-64, Lot No. 58483222, passage 26

MALME-3M Culture Medium

• IMDM (High Glucose, Phenol Red), ATCC Catalog No. 30-2005
• Fetal Bovine Serum (20%), Thermo-Hyclone Catalog No. SH30071.03
• Pen-Strep-Glutamine (1%), Invitrogen Catalog No. 10378-016

MALME-3M Plating Medium

• IMDM (no Phenol Red), Gibco Catalog No. 21056-02
• Fetal Bovine Serum (10%), Thermo-Hyclone Catalog No. SH30071.03
• Pen-Strep-Glutamine (1%), Invitrogen Catalog No. 10378-016

MALME-3M cells were maintained in IMDM (20%FBS, 1% Pen-Strep-Glutamine). Cells were fluid changed every 3 days and/or split upon reaching 100% confluency. For secondary assays, cells were thawed at 6 million cells per Falcon T175 flask. After 3 days, cells were fluid changed, after 3 more days cells were passed to a Corning Triple flask (15-18 million cells) and plated after 3 days in the triple flask.

Day 1

1. Plate MALME-3M cells at 3,000 per well in 30 μL media (phenol red free IMDM/10% Fetal Bovine Serum/Penicillin/Streptomycin/L-Glutamine)
2. Use Corning white 384-well, square, opaque-bottomed plates (Corning Catalog No. 8867BC)

Day 2

3.

Pin 100 nL compound/DMSO solution (Cybi Well) into assay plates, including the positive control.

4.

Incubate 24 hours at 37°C in Liconic incubator.

Day 3

5.

Add 20 μL 100% Promega Cell Titer GLO per well with Thermo Combi fluid transfer apparatus.

6.

Shake 15 seconds on “big bear” plate shaker and incubate at room temperature for 5 minutes.

7.

Read on the Perkin-Elmer EnVision plate reader with luminescence (LUM) settings for 0.1 sec per well.

Cell proliferation assay with primary human melanocytes (2084-06)

Primary human neonatal melanocytes were isolated from discarded foreskins by gentle dispase treatment and grown in TIVA media (Ham's F10 media supplemented with 7% FBS, penicillin/streptomycin/glutamine, 0.1mM IBMX, 50ng/mL TPA, 1μM Na₃VO₄ and 1μM dbcAMP). Cells were passaged using Accutase (Sigma Catalog #A6964-100ML) for gentle treatment and generation of a single cell suspension.

Day 1

1. Plate primary melanocytes at 3,000 per well in 30 μL media (TIVA media)
2. Use Corning white 384-well, square, opaque-bottomed plates (Corning Catalog No. 8867BC)

Day 2

3.

Pin 100 nL compound/DMSO solution (Cybi Well) into assay plates. (pinning with the positive control, parthenolide (18 μM final concentration))

4.

Incubate 24 hours at 37°C in Liconic incubator.

Day 3

5.

Add 20 μL 100% Promega Cell Titer GLO per well with Thermo Combi fluid transfer apparatus.

6.

Shake 15 seconds on “big bear” plate shaker and incubate at room temperature for 5 minutes.

7.

Read on the Perkin-Elmer EnVision plate reader with luminescence (LUM) settings for 0.1 sec per well.

qPCR Assay for Target Gene Expression (MITF: 2084-05, TRPM1: 2084-09, CDK2: 2084-11, DCT: 2084-12, MLANA: 2084-13)

SK-MEL-5 Cells ATCC Catalog No. HTB-70, lot 58483232, passage 28

SK-MEL-5 Culture/Plating Medium

• DMEM (High Glucose, HEPES, Phenol Red), Invitrogen Catalog No. 12430-047
• Fetal Bovine Serum (10%), Thermo-Hyclone Catalog No. SH30071.03
• Pen-Strep-Glutamine (1%), Invitrogen Catalog No. 10378-016

Parthenolide Enzo Catalog No. BML-T113-0250

Cells to CT Bulk Lysis Solution Ambion Catalog No. 4391851C

Cells to CT Bulk RT Reagents Ambion Catalog No. 4391852C

Light Cycler 480 Probes Master Roche Catalog No. 4887301001

Human GAPD (GAPDH) Endogenous Control VIC/MGB probe/primer limited Applied Biosystems Catalog No. 4326317E

Target Gene FAM probe/primer sets

• Human MITF FAM probe/primer Applied Biosystems Catalog No. 4331182 Hs01117294_m1
• Human TRPM1 probe/primer Applied Biosystems Catalog No. 4331182 Hs00170127_m1
• Human CDK2 probe/primer Applied Biosytems Catalog No. 4331182 Hs01548894_m1
• Human DCT probe/primer Applied Biosystems Catalog No. 4331182 Hs01098278_m1
• Human MLANA probe/primer Applied Biosystems Catalog No. 4331182 Hs00194133_m1

SK-MEL-5 cells were maintained in DMEM (10%FBS, 1% Pen-Strep-Glutamine). Cells were fluid changed every 3 days and/or split upon reaching 100% confluency. For secondary assays, cells were thawed at 4 million cells per Falcon T175 flask. After 3 days, the cells were fluid changed, after 3 more days cells were passed to a Corning triple flask (10-15 million cells) and plated after 3 days in the triple flask.

Day 1

1. Plate SK-MEL-5 cells at 4,000 per well in 30 μL media (DMEM/10% Fetal Bovine Serum/Penicillin/Streptomycin/L-Glutamine)
2. Use Corning white 384-well, square, opaque-bottomed plates (Corning Catalog No. 8867BC)

Day 2

3.

Pin 100 nL compound/DMSO solution (Cybi Well) into assay plates. (in plate positive control, parthenolide (6 mM))

4.

Incubate 24 hours at 37°C in Liconic incubator

Day 3

• Cell Lysis

  • 5.

    The medium is aspirated from assay plates and the cells are washed twice (100 μL PBS) using the ELX405 Plate Washer (Biotek).

    6.

    The assay plates are flipped upside down and centrifuged at 1000 rpm for 2 minutes to remove the excess liquid.

    7.

    10 μL of Lysis solution with DNase I (Ambion, from Cell to CT Lysis Mix) is added to each well using the MultiDrop Combi/Standard tube dispensing cassette (Thermo Scientific).

    8.

    Each assay plate is then shaken for 2 minutes and incubated for an additional 8 minutes at room temperature.

    9.

    1 μL of stop solution (Ambion, from Cell to CT Lysis Mix) is added with the Multidrop Combi-nL (Thermo Scientific) and the assay plate is centrifuged at 1000 rpm for 2 minutes.

• Reverse Transcription (RT) Mix

  • Table 1

    View in own window

    Component	Amount per reaction
    2X RT Buffer	5 μL
    20X RT Enzyme Mix	0.5 μL
    Nuclease-Free Water	2.5 μL

  • 10.

    8 μL of RT mix is dispensed into each well of a RT assay plate (Axygen, PCR-384 RGD C).

    11.

    2 μL of the lysed cells are transferred into RT assay plate using Vario transfer unit (Cybi Well).

    12.

    The RT assay plates are incubated at 37°C for 1 hour and the reverse transcriptase is inactivated by incubating the plates for 1 minute at 95°C.

    13.

    cDNA is stored at -80°C until ready for qPCR analysis

Day 4

• qPCR Master Mix

  • Table 2

    View in own window

    Component	Amount per reaction
    2X Roche Master Mix	2.5 μL
    20X FAM Target Gene Taqman probe/primer	0.125 μL
    20X VIC GAPDH Taqman probe/primer	0.125 μL
    PCR water	1.25 μL

  • 14.

    4 μL/well of qPCR master mix is dispensed in PCR plate (Roche Light Cycler 480 Multiwell Plate 384, Catalog No. 04 729 749 001) using the Multidrop Combi-nL (Thermo Scientific).

    15.

    1 μL/well of RT DNA is transferred in the 4 μL/well PCR plate.

    16.

    The PCR plates are centrifuged for 2 minutes at 1000 rpm.

    17.

    PCR is performed using Thermo Cycler (Roche Light Cycler 480 II) with Macro Protocol:

  • View in own window

    Step	Temperature	Time
    1.	95°C	10 minutes
    2.	95°C	10 seconds
    3.	60°C	30 seconds
    Step 2 and 3 (55 cycles)
    4.	40°C	30 seconds

Data Analysis

For the primary screen and other assays, negative-control (NC) wells and positive-control (PC) wells were included on every plate. The raw signals of the plate wells were normalized using the ‘Stimulators Minus Neutral Controls’ or the ‘Neutral Controls’ method (when no positive control was available) in GeneData Screener Assay Analyzer (v7.0.3 & v10.0.2). The median raw signal of the intra-plate NC wells was set to a normalized activity value of 0, while the median raw signal of the intra-plate PC wells was set to a normalized activity value of 100. Experimental wells were scaled to this range, resulting in an activity score representing the percent change in signal relative to the intra-plate controls. The mean of the replicate percent activities were presented as the final ‘PubChem Activity Score’. The ‘PubChem Activity Outcome’ class was assigned as described below, based on an activity threshold of 70%:

• Activity_Outcome = 1 (inactive), less than half of the replicates fell outside the threshold.
• Activity_Outcome = 2 (active), all of the replicates fell outside the threshold, OR at least half of the replicates fell outside the threshold AND the ‘PubChem Activity Score’ fell outside the threshold.
• Activity_Outcome = 3 (inconclusive), at least half of the replicates fell outside the threshold AND the ‘PubChem Activity Score did not fall outside the threshold.

Study Objective

To evaluate, in radioligand binding assays, the activity of probe compound ML329 (KUC111774N-02) across a panel of 67 receptors.

Methods

Methods employed in this study have been adapted from the scientific literature to maximize reliability and reproducibility. Reference standards were run as an integral part of each assay to ensure the validity of the results obtained.

Where presented, IC₅₀ values were determined by a non-linear, least squares regression analysis using MathIQ™ (ID Business Solutions Ltd., UK). Where inhibition constant (K_i) are presented, the Ki values were calculated using the equation of Cheng and Prusoff (Cheng. Y., Prusoff, W.H., Biochem. Pharmacol. 22:3099-3108, 1973) using the observed IC₅₀ of the tested compound, the concentration of radioligand employed in the assay, and the historical values of the KD of the ligand (obtained experimentally at Eurofins Panlabs, Inc.). Where presented, the Hill coefficient (n_H), defining the slope of the competitive binding curve, was calculated using MathIQ™. Hill coefficients significantly different than 1.0, may suggest that the binding displacement does not follow the laws of mass action with a single binding site. Where IC₅₀, K_i, and/or nH data are presented without Standard Error of the Mean (SEM), data are insufficient to be quantitative, and the values presented (Ki, IC₅₀, n_H) should be interpreted with caution.