PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
J Med Chem. Author manuscript; available in PMC 2013 January 26.
Published in final edited form as:
PMCID: PMC3335202
NIHMSID: NIHMS350953

Wnt inhibition correlates with human embryonic stem cell cardiomyogenesis: A Structure Activity Relationship study based on inhibitors for the Wnt response

Abstract

Human embryonic stem cell-based high content screening of 550 known signal transduction modulators showed that one “lead” (1, a recently described inhibitor of the proteolytic degradation of Axin) stimulated cardiomyogenesis. Because Axin controls canonical Wnt signaling, we conducted an investigation to determine whether the cardiogenic activity of 1 is Wnt dependent, and developed a structure activity relationship to optimize the cardiogenic properties of 1. We prepared analogs with a range of potencies (low nanomolar to inactive) for Wnt/β-catenin inhibition and for cardiogenic induction. Both functional activities correlated positively (r2 = 0.72). The optimal compounds induced cardiogenesis 1.5-fold greater than 1 at 30-fold lower concentrations. In contrast, no correlation was observed for cardiogenesis and modulation of TGFβ/SMAD signaling, that prominently influences cardiogenesis. Taken together, these data show that Wnt signaling inhibition is essential for cardiogenic activity and that the pathway can be targeted for the design of drug-like cardiogenic molecules.

Keywords: Human Embryonic Stem Cell, Wnt inhibition, SAR, cardiomyogenesis, IWR-1, inhibitors

Introduction

The proposal to produce human heart muscle cells from stem cells has attracted much attention because of the potential for promising applications in medicine and drug discovery, ranging from cell transplantation to in vitro pharmacological testing. Use of human stem cell-derived cardiomyocytes as tools for drug discovery and development (i.e., high throughput assays, specific disease models, target identification and validation or toxicity assessment) has many advantages over current assays that rely on a non-cardiomyocyte setup13. Cardiomyocytes, the key cells for cardiac safety evaluation at the preclinical stage, are hard to come by. Scientists are usually limited to animal-derived cells or tissues or human engineered cells (cell lines heterologously expressing human cardiac ion channels, cardiac cell cultures, isolated tissue preparations, and perfused animal hearts) that have limited predictivity in human. For example, overexpression of the hERG channel in fibroblasts is commonly used in drug development as a model to evaluate cardiotoxicity of novel drugs. One major advantage of cardiomyocytes derived from hES cells is that they are of human origin and can be maintained in culture for extended time periods without losing their characteristics. Theoretically, they represent an unlimited source for human cardiomyocytes for in vitro testing. Moreover, the information gained from developing small molecules for stem cell-derived cardiomyocyte differentiation in vitro may lead to the development of drugs capable of mobilizing endogenous cardiac progenitor cells to regenerate damaged muscle in the adult heart4.

Cell differentiation is a complex and still poorly understood process. As for other tissue types, the development of human myocardial cells requires close temporal control of inducing factors to stimulate the stepwise progression from pluripotent cells to uncommitted progenitor, to committed precursor and finally to myocardial cells including cardiomyocytes4,5. Current approaches to stimulate stem cell differentiation have included naturally occurring factors, including the introduction of lentiviral vectors carrying transcription factors, the addition of growth factors and the use of small molecule signaling pathway modulators1. The approach described herein focuses on the latter, because use of small molecules as differentiation reagents overcomes the inherent high cost of biological factors or reagents for in vitro applications and can be developed into drug candidates for in vivo applications for therapeutic development.

We recently described a human Embryonic Stem Cell (hESC)-based high content screen (HCS) of about 550 known pathway modulators (i.e., InhibitorSelect and StemSelect, both from EMD Chemicals Inc.) and used these compounds to identify key signaling pathway(s) that control differentiation of uncommitted cardiac progenitors to form cardiomyocytes6. Only one small molecule (i.e., 1 (IWR-1)7,8, Fig. 1) was identified as a “lead” from this library screen. The abbreviation IWR stands for Inhibitor of the Wnt Response. Wnt is a hybrid name of Int, a gene active in mouse mammary tumors, and Wingless, a gene essential for wing development in Drosophila. The Wnt pathway has been studied in a wide range of organisms and is implicated in mammalian development and cancer. Compound 1 was recently reported as an inhibitor of the oncogenic canonical Wnt response at the Axin level7,8. Besides being involved in cancer9, Wnt and Wnt inhibition regulate crucial processes during embryonic development10. Wnt signaling plays a multiphasic role in heart development and Wnt inhibition is critical to form committed progenitor cells11,12. This has led to the use of the natural Wnt inhibitor Dickkopf-1 (DKK-1) to enhance cardiogenesis in hESC culture13 and the discovery of small molecule tool compounds2.

Figure 1
Screening “lead” 1. Regions A, B, C indicate the different areas of the molecule that were systematically modified for lead optimization.

Herein, we examined the effect of the previously reported Wnt inhibitors 15 on human ESC-mediated cardiomyocyte-differentiation in parallel with a step-wise process to improve the pharmacological properties of 1. Compound 1 was optimized for its ability to inhibit the Wnt pathway and stimulate cardiogenesis. Newly designed compounds 10, 29 and 34 showed decreased IC50 values for inhibition of the Wnt pathway as well as increased cardiomyogenesis potency. Compared to 1, compounds 10, 29 and 34 showed improved functional activity and improved physiochemical propertied, thus rendering them more attractive for in vitro and possibly, in vivo studies.

Chemical Synthesis

The synthesis of 1 and certain analogs was recently described in papers reporting new Wnt inhibitors for cancer7,8. From a structural perspective and as previously described by Lu et al.,8 compound 1 can be divided into three regions (i.e., A, B, C, Fig. 1). The SAR of all three regions of 1 was explored.

Regions A and C were modified using synthetic protocols described by Chen et al.,7 (Schemes 1 and and2).2). Briefly, the desired amino acid 6 was heated with anhydride 7 in DMF and afforded intermediates 8a–e in good yields. The carboxylic acid moiety of 8a–e was then activated in the presence of thionyl chloride and treated with the appropriate amine (i.e., R3NH2) to give compounds 9–39, 53–56 (Scheme 1). Compounds 41, 42c–f, 43g–k, 44l–p were prepared by treating the acid chloride of para-nitrobenzoic acid with the appropriate amine, R3NH2. After hydrogenation of the nitro group, intermediate 40 was subjected to a reductive amination step or acylation or treatment with an anhydride to give the desired products 4144 (Scheme 2).

Scheme 1
Synthesis of analogs 939, 5356.
Scheme 2
Synthesis of compounds 4144.

The SAR of Region B was explored using various synthetic approaches. Compound 45 (Table 1) was prepared following the same sequence of steps as in Scheme 1 but by replacing 6 by 3-aminobenzoic acid. Compound 46 (Table 1) was prepared following Scheme 2 by replacing 4-nitrobenzoic acid with 5-nitro-2-furoic acid. Compound 48, was prepared in three steps from 4-nitrophthalic anhydride (Scheme 3). Condensation of 5-nitroisobenzofuran-1,3-dione with 8-aminoquinoline followed by reduction of the nitro group afforded 47. Condensation of 47 with cis-endo-dihydrocarbic anhydride using conditions described in Scheme 1 afforded 48. For the reduced version 50 (Table 1), intermediate 49 was obtained in two steps by reductive amination of 8-aminoquinoline with 4-nitrobenzaldehyde followed by reduction of the nitro group with sodium dithionite. Compound 49 was treated with cis-endo-dihydrocarbic anhydride under conditions described in Scheme 1 to afford compound 50 (Scheme 4). Finally, the reverse amide 52 (Table 1) was synthesized using intermediate 51. Treatment of 4-nitroaniline with carbic anhydride followed by hydrogenation gave 51 and following addition of freshly prepared quinoline-8-carbonyl chloride, treatment with aniline provided the desired reverse amide (52, Scheme 5).

Scheme 3
Synthesis of 48.
Scheme 4
Synthesis of 50.
Scheme 5
Synthesis of 52.
Table 1
SAR for the Central Region (B) of the molecule for Wnt inhibition.

Results and Discussion

Because 1, the lead compound from screening, was reported to inhibit canonical Wnt signaling at the Axin level7, we compared 1 to four mechanistically different Wnt inhibitors (2 (XAV939), 3 (IWP-3), 4 (iCRT-5) and 5 (pyrvinium), Fig. 2A) in a Wnt inhibition assay (Fig. 2B) and a hESC cardiogenesis assay (Fig. 2C) to examine the relation between Wnt inhibition and cardiogenesis and to narrow down and identify the part of the pathway blocked in the signaling cascade that influenced cardiogenic activity. To ensure that the SAR analysis was not confounded by cell toxicity, the five Wnt inhibitors (15) and certain analogs of 1 were tested in parallel for cytotoxicity (see Supporting Information, Fig. S2). Compound 1 blocks the Wnt pathway by stabilizing the Axin protein complex via a direct interaction with Axin7. Compound 2 stabilizes Axin by inhibiting the poly-ADP-ribosylating enzymes tankyrase 1 and 214. Compound 3 inhibits the production of Wnt by preventing the palmitoylation of Wnt proteins by Porcupine7. Compound 4 has been reported to inhibit the interaction between β-catenin and TCF4 within the cell nucleus15. Compound 5 activates CK1α activation in the Wnt pathway and promotes degradation of β-catenin and inhibited Axin degradation16. In our hands, 5 was a very modest Wnt inhibitor (IC50 = 592 nM) and also quite cytotoxic, so it was not tested in the cardiac assay. Compounds 3 and 4 were similarly modest Wnt inhibitors (i.e., IC50 = 538 and 728 nM, respectively), but were not toxic to HEK293T cells at concentrations up to 20 μM (See Suppl. Info Fig. S2). Compound 2, described to induce cardiac differentiation in mouse ESCs17, had an IC50 of 51 nM in the Wnt assay and did not induce cell toxicity. Compound 1 had a similar profile (i.e., IC50 = 26 nM and was not toxic to HEK293T cells). As we showed previously6, 1, 2 and 3 induced cardiogenesis by ca. 1000-fold at their maximum efficacious concentrations (i.e., 4, 2 and 2.5 μM respectively, Fig. 2C) over a DMSO control but differed in their potencies. Compound 2 was about 1.5-fold less potent than 1 and 3. Compound 4 was weakly cardiogenic but only at a higher concentration (50 μM). Compound 3 was not selected as a lead candidate because its mode-of-action in inhibiting Wnt production is reported to be at an upstream position compared to 1. Taken together, these data suggested that 1 was the most appealing structure as a lead candidate. It was potent both as a Wnt inhibitor and as a cardiogenic agent in the hESC studies and was not cytotoxic to the cells examined.

Figure 2
Inhibition of the Wnt pathway with reported inhibitors 15. A) Structures of compounds 15; B) Dose-response curves and IC50 values for Wnt inhibition for 15 representing different classes of Wnt inhibitors. Data is the mean ± ...

Wnt Inhibition

Initially, a structurally diverse library of ca. 90 analogs of 1 was synthesized and tested for Wnt inhibition at a standard concentration (i.e., 1 μM). Compounds that showed inhibition above 50% at 1 μM were re-evaluated for dose-dependent inhibition. Next, more focused libraries were prepared to improve potency and physicochemical features of promising analogs.

Using synthetic strategies depicted in Schemes 15, variations on the central region (Region B) were prepared and evaluated for Wnt inhibition (Table 1). Lu et al., previously showed that the phenyl ring of 1 could be replaced by a trans-cyclohexyl moiety and still maintain Wnt inhibition potency8. As shown in Table 1, an increase in Wnt inhibition potency was observed with the completely saturated trans-cyclohexyl analog 10 (i.e., IC50 = 4 nM). Because a cyclohexyl ring can adopt both cis and trans configurations it was of interest to examine the stereoselectivity of Wnt inhibition. The cis analog 11 was prepared and found to be potent (i.e., IC50 = 57 nM). It is possible that the potency of the cis-cyclohexyl moiety can be explained by 11 adopting a boat conformation. Next, modifications of the amide function revealed important information about Wnt inhibition SAR. Reduction of the amide function (50, 45% inhibition at 1 μM), changing the amide position to the meta-position (45, 0% inhibition at 1 μM) or replacing the central aryl substituent by a furan group (46, 0% inhibition at 1 μM) decreased the potency of the Wnt inhibition and showed that both the amide bond and the 1,4-substitution pattern for the central B region were essential for maximal inhibitory potency. A rigidified version (48) and a reverse-amide version (52) were also prepared. Reversing the amide as in 52 was tolerated but showed some loss of Wnt inhibitory potency (i.e., IC50 = 119 nM) while the rigid analog 48 was inactive (i.e., 15% inhibition at 1 μM). Compounds 9 (i.e., central phenyl ring) and 10 (i.e., central trans-cyclohexyl ring) possessed similar IC50 values (i.e., 24 and 4 nM, respectively) that suggested that the central ring itself likely seems to be a simple placeholder and does not contribute to a discernible biological interaction.

A systematic study of the effect of the position of the nitrogen atom in the quinoline heterocycle on Wnt inhibition was done with compounds possessing a trans-cyclohexyl ring in the central position. The results summarized in Tables 2 and and33 showed that the 8-aminoquinoline isomer was the most potent quinoline (i.e., 10, IC50 = 4 nM). Potency decreased when the distance between the quinoline nitrogen and the amide function increased suggesting either an intramolecular interaction between the amide hydrogen and the quinoline nitrogen or alternatively, a specific interaction with a biological target. Lu et al., showed that by replacing the quinoline with a phenyl ring substituted at various positions by halogens or a trifluoromethyl group usually led to a significant loss of potency8. The optimal aryl substitution pattern was for a 4-bromophenyl and 2-methoxyphenyl (i.e., IC50 = 100 nM for both compared to 200 nM for 1)8. Incorporation of electron donating groups (2-hydroxyl compound 19, 2- and 4-methoxy compounds 20 and 21, 4-dimethylamino compound 22, with 33% inhibition at 1 μM, 50% inhibition at 1 μM, IC50 = 215 nM, 25% inhibition at 1 μM, respectively) or electron withdrawing groups (4-cyano compound 23, 4-acetyl compound 24, with IC50 values of 317 nM and 41% inhibition at 1 μM, respectively) didn’t lead to a clear picture of the effect of electronic substituents on Wnt inhibition (Table 3). On the other end, the presence of a methoxy group at the ortho-position (compound 21 with an IC50 of 215 nM) or a para-methoxy substituent (22 with 50% inhibition at 1 μM) suggested that the electronic nature of the substituent at the aryl ring was not critical for Wnt inhibition. It is likely that shape rather than electronic properties and the position of the substituent affected potency. A carbonyl-containing functionality at the ortho position afforded the most potent analogs examined and apparently mimicked the role of the quinoline nitrogen to afford potent Wnt inhibition. A methyl ester at the 2-position (27) had an IC50 of 67 nM and the 2-methylketone (28) had an IC50 of 6 nM. The tetralone analog 29 was similarly potent (i.e., IC50 = 7 nM). When increased steric bulk was introduced using 9H-fluoren-9-one (i.e., 30, IC50 = 194 nM), potency decreased suggesting steric hindrance in this region may have a negative effect on Wnt inhibition (Table 3).

Table 2
SAR for Region C: A systematic study of the effect of the nitrogen position of the quinoline heterocycle on potency for Wnt inhibition.
Table 3
SAR of Region A of the molecule: Effects of phenyl and pyridinyl substituents on the potency for Wnt inhibition.

A number of unsubstituted heterocycles (i.e., 2-pyrimidine, 2-pyrazine, 3-(1,2,4-triazole), 2-thiazole, 2-pyridine, 3-pyridine, 4-pyridine) were synthesized as replacements for the quinoline substituent. While the unsubstituted 2-, 3- and 4-pyridines (i.e., compounds 3133, Table 3) showed IC50 values in the 100 nM range (i.e., IC50 = 109, 120 and 80 nM, respectively), none of the other heterocycles described above were functionally active (i.e., % inhibition below 50% at 1 μM, data not shown). Substituted pyridines were then prepared by combining the optimal substituents identified from the data of Table 3 with 3- and 4-pyridine derivatives. Compounds combining the 3-pyridine with an acetyl group at the 2- or 6-positions were prepared and tested. Remarkably, the 2-acetyl-3-pyridine analog 34 gave an IC50 value of 7 nM while the 6-acetyl-3-pyridine isomer 35 was significantly less potent (IC50 = 590 nM).

As reported (but not exemplified) by Lu et al.,8 Region A (Fig. 1) was very sensitive to chemical modifications. In our hands, any attempts to modify this region either by replacing the bicycle by bulky R groups (i.e., compounds 41, 42c–f, Scheme 2), or replacing the entire carbic moiety by various groups (i.e., compounds 43g–k and 44l–p, Scheme 2) led to a complete loss of potency (i.e., inhibition < 32% at 1 μM). Only minor structural alterations such as saturation of the double bond (i.e., 9, IC50 = 24 nM) were well-tolerated and two-carbon-bridge analogs (i.e., 36, IC50 = 412 nM and 37, IC50 = 314 nM, Table 4) and oxa-analog (i.e., 39, IC50 = 605 nM) were >10-fold less potent. Adding a methyl group to the carbic moiety (i.e., 38, IC50 =1176 nM) also resulted in significant loss of potency highlighting the sensitivity of this region to substitution.

Table 4
SAR for Region A of the molecule for Wnt inhibition.

Chemical Stability

Certain selected compounds, as well as 1 for comparison purposes, were evaluated for chemical stability in buffer before being tested for their cardiogenic properties. hESC cardiogenesis studies were done in a cell-based assay at pH 7.4 and biological activity was presumably required for at least 24 hours after compound addition four days after initiating hESC differentiation6. Therefore, we determined the aqueous stability of key compounds at pH 7.4 for at least 24 hours. Representatives of the various subseries, (i.e., compounds 1, 10, 28, 32 and 52), were incubated at 10 μM in phosphate buffer (pH 7.4) for up to 30 hours. No degradation was observed as determine by HPLC for any of the compounds except for 52 whose half-life was 3.8 hours (Supplemental Information, Table S1).

Induction of Cardiogenesis

From the 180 synthetic analogs tested in the Wnt inhibition assay, 26 were selected to examine cardiogenic induction in the presence of hESCs. This subset of analogs comprised a full range of inhibitory potency in the Wnt assay and contained the most structurally diverse scaffolds for evaluation of hESC cardiogenic activity. As shown in Figure 3, analogs from the following series were selected as representative members to investigate: a) variation of the N-position of the quinoline C region, including phenyl- versus pyridinyl-substituents and ortho-carbonyl groups, b) variation of the central B region linker including aromatic versus aliphatic and cis- versus trans-configured linkers, and c) variation in the norbornene A region moiety. Figure 3A shows a list of values for the potency of Wnt inhibition in HEK293T cells and induction of cardiogenesis in hESCs. Three classes of compounds stimulated hESC cardiogenesis. The first group of compounds (i.e., 32, 21, 35, 33, 31, 39, 37, 23, 52, 12) was functionally inactive as inducers of cardiogenesis and possessed uniformly large IC50 values in the Wnt inhibition assay (> 200 nM). A subset of these compounds (i.e., 31, 32 and 33), however, had IC50 values in the 100 nM rangeas Wnt inhibitors. Further studies need to be done to explain their biological properties. Compound 52 contained a “reverse amide” and was a relatively potent Wnt inhibitor (i.e., IC50 = 119 nM) but inactive as a cardiogenic inducer. The chemical half-life of 52 has been estimated to be less than five hours in the presence of assay buffer (see Supporting Information, Table S1) and this could explain its low cardiogenesis functional activity. While lack of aqueous stability would not be expected to compromise functional activity in the short timeframe of the Wnt assay, the actual concentration of 52 in the hESC assay may have declined compared to the time necessary to induce cardiogenesis and thus could explain its lack of potency as a cardiogenic stimulant. The second class (i.e., 14, 30, 1, 13, 9, 56, 11, 55, 54, 27, 18, 53) contained compounds that elicited a dose-dependent induction of cardiogenesis functional activity, without a decrease in cardiogenesis at higher doses. These compounds were also potent Wnt inhibitors based on their IC50 values in the Wnt assay (20–200 nM). Compound 30 exhibited a modest Wnt inhibition (i.e., IC50 = 194 nM) and solid cardiac induction (104% cardiogenesis). We suspect that the high lipophilicity of the 9-fluorene substituent causes low aqueous solubility and extensive protein binding. This could possibly explain the observed differences in these two distinct assays. The third class of compounds (i.e., 29, 34, 10, 28) were potent Wnt inhibitors and potent stimulators of cardiogenesis, even at low concentrations (i.e., < 0.15 μM). They were distinguishable from the second group of compounds by their dose-response curves (Fig. 3C). They showed a very strong cardiogenesis at the lowest concentrations followed by a constant decrease of percent cardiogenesis at increasing compound concentration in the cell-based assay. This effect could be due to anti-cardiogenic, anti-proliferative, or toxic cellular effects caused by strong Wnt inhibition. Compared to other analogs, compound 52 was substantially less stable at pH 7.4 than other analogs (see Supporting Information, Table S1) and thus its potency in the hESC assay may be possibly underestimated. Compound 18 had a negative Z score at 625 nM and thus appeared as an outlier our correlation, but it was cardiogenic at higher concentrations and reached 24% maximum cardiogenesis at 10 μM (Fig 3A).

Figure 3
Cardiogenic activity and Wnt inhibition of selected analogs. A) Table shows data set for analogs tested in cardiac and Wnt assays; IC50 (nM) were measured in the Wnt inhibition assay. Red highlighted are “outlier” compounds. B) Maximum ...

Based on the correlation analysis (Fig. 3D), hESC cardiomyocyte induction apparently strongly paralleled Wnt inhibition and that is apparent in SAR studies. Even minor variations in the norbornene part of the molecule (Region C) led to a loss of functional activity for both Wnt inhibition and hESC cardiogenesis. For example, the two carbon-bridge analog 37 and oxa-bridge analog 39 had IC50 values of 314 and 605 nM, respectively, in the Wnt inhibition assay and stimulated cardiogenesis only to the extent of less than 5%. The more potent compounds 9 (one carbon bridge analog) and 10 (saturated version) had IC50 values of 24 and 4 nM, respectively, in the Wnt assay and induced cardiogenesis of 75% and 145% respectively, compared to 1. The effect of stereochemistry of the central portion of the molecule (Region B) on the stereoselectivity of the biological endpoints was examined. The cis- and trans-configuration of the saturated cyclohexyl linker (Region B) had a similar effect on cardiogenesis and Wnt inhibition. For example, 28 (trans) and 56 (cis) had percent cardiogenesis of 78% and 72%, respectively (Fig. 3B), and were relatively potent at Wnt inhibition (i.e., IC50 values of 6 and 23 nM respectively, Table 3). A number of alternate positions of the nitrogen in the quinoline heterocycle was tolerated for functional biological activity, but complete omission of the nitrogen atom led to a dramatic decrease in induction of cardiogenic potency (i.e., 12, 41% Wnt inhibition at 1 μM and 8% cardiogenesis). Finally, different phenyl- and pyridinyl-substituted analogs gave comparable functional activity for induction of cardiogenesis and this showed that the quinoline could be replaced while retaining cardiogenesis induction. In particular, ortho-substitution of phenyl- and pyridinyl-residues with carbonyl-containing groups afforded highly potent derivatives and those analogs (i.e., 28, 29, and 34) were among the most potent and effective compounds tested.

Finally, we tested potent Wnt inhibitors for selective inhibition of TGFβ signaling to investigate possible off-target effects. TGFβ interacts with Wnt to modulate stem cell fate18, and inhibition of TGFβ signaling can also be cardiogenic in mouse19 and human ESCs20. Therefore, ten representative structurally distinct compounds (i.e., 1, 10, 28, 32, 34, 35, 29, 53, 34 and 35) were incubated at concentrations varying from 0.01 to 5 μM (same dose-range used in the Wnt assay) in a SMAD response element-reporter assay in HEK293T cells stimulated with human recombinant TGFβ2. No significant inhibition effect was observed compared to DMSO vehicle controls even at the highest concentrations examined (i.e., % inhibition of SMAD response <20% at 0.01, 0.1, 1 and 5 μM, respectively, see Supplemental Information Fig. S3), indicating that TGFβ inhibition is not involved in the cardiac phenotype of the IWR-1 analogs examined. Furthermore, because the functional biological activity of 1 has been attributed to Axin stabilization, these data also indicate that Axin stabilization does not inhibit TGFβ/SMAD signaling.

Conclusion

Enhancing the production of human cardiomyocytes offers a wealth of opportunities for drug research and toxicological assessment of new chemical entities. The Wnt pathway has been associated with cardiogenesis and using 1 as a “lead”, we obtained small molecule Wnt inhibitors with IC50 values ranging from low nanomolar (4 nM) to the micromolar range. We showed a direct (i.e., titratable) correlation between Wnt inhibitory potency as determined by IC50 values in mammalian cells and cardiogenic functional activity in a human ESC-based assay. We were also able to improve the percent of cardiogenesis to 176% with compound 29 compared to 100% for the initial “lead” 1 at much lower concentrations (i.e., 150 nM for 29 compared to 5000 nM for 1, equivalent to 30-fold greater potency) rendering 29 more amenable to in vitro and in vivo applications. Because adult hearts contain cardiogenic progenitor cells that resemble the hESC-derived progenitor cells used in assays described herein in terms of gene expression profile and developmental potential4, it is possible that Wnt inhibition could promote differentiation of adult human cardiac progenitor cells. Therefore, an important next step will be to explore their utility for ex vivo expansion of adult heart-derived progenitors and to nominate and evaluate optimal compounds for their potential as in vivo regenerative compounds in animal models of myocardial infarction.

Experimental Section

General

Reagents and solvents were used as received from commercial sources. Compound 2 was purchased from Cayman Chemicals, 4 from Princeton BioMolecular Research and 5 from Sigma-Aldrich. Synthetic products were isolated using flash column chromatography system (Teledyne ISCO, CombiFlash Rf) with UV detection at 254 nm or PTLC (Preparative Thin Layer Chromatography) with UV indicator. NMR (Nuclear Magnetic Resonance) spectra were recorded at 300 MHz (1H) on a Varian Mercury 300 or at 125 MHz (13C) on a Bruker AMX-500 II (NuMega Resonance Lab, San Diego, CA). Chemical shifts were reported as ppm (δ) relative to the solvent (CDCl3 at 7.26 ppm, CD3OD at 3.31 ppm, d6-DMSO at 2.50 ppm and 3.52 ppm). S stands for singlet, d for doublet, t for triplet, q for quadruplet, m for multiplet and br for broad. Low resolution mass spectra were obtained using Hitachi M-8000 mass spectrometer with an ESI source.

Purity of final products was determined with a Hitachi 8000 LC-MS (Hitachi) using reverse phase chromatography (C18 column, 50 × 4.6 mm, 5 μm, Thomson Instrument Co., Oceanside, CA). Compounds were eluted using a gradient elution of 95/5 to 5/95 A/B over 5 min at a flow rate of 1.5 mL/min, where solvent A was water with 0.05% TFA and solvent B was acetonitrile with 0.05% TFA. For purity analysis, peak area percent for the TIC (Total Ion Count) at 254 nm and retention time (tR in minutes) were provided. Purity of final products was ≥ 95%.

Biological Assays

Wnt Inhibition

Compounds were tested for their ability to inhibit the β-catenin-dependent canonical Wnt pathway. A Wnt assay was adapted from Chen et al.,7 in our labs using HEK293T cells in a 96-well format. Briefly, the commercially available Super(8x)TOPflash vector driven by a (7x)TCF-firefly luciferase response element was transiently transfected into HEK293T cells together with a TK-driven Renilla luciferase plasmid as an internal control to normalize the luminescence signal and a Wnt3A-expressing vector as the source of pathway activation. Because of the transient nature of the assay and, consequently, variations between independent experiments, 1 was included as a positive control in parallel with untreated and DMSO-treated vehicle controls in experiments conducted. Maximum inhibition of Wnt response in the assay format was around 90% with the most potent inhibitors at the highest doses examined. Nonlinear regression analysis was performed using the log(inhibitor)/normalized response equation of the Prism 5 software.

hESC Cardiogenesis

Human embryonic stem cell H9 lines carrying MYH6-mCherry reporters were used in the hESC assay as previously described in detail6 and is summarized in the Supplemental Information. In brief, embryoid bodies (EB) were grown until day 4, dissociated gently to single cells and transferred to 384-well plates. Concomitantly, different dilutions of small molecules were added. At day 10, media was exchanged for a serum free media (SFM) containing the thyroid hormone analog triiodothyronine T3 to increase the red signal driven by the MYH6 promoter for improved image analysis.21 Cells were changed to PBS at day 14 for imaging and red fluorescence images were collected on a high throughput microscope. For quantifying the level of cardiac induction, the total area and intensity of the MYH6-mCherry reporter was measured in each well. Cardiac activity was reported by stating either percent cardiogenesis normalized to 1 (set as 100% at its maximum potency) or Z scores (relative to DMSO control) (Fig. 3A and B).

Chemistry

Compounds 1 and 3 were synthesized following the procedure described in reference 7 and their 1H NMR compared to the literature spectra7.

Scheme 1: Compounds 9–39, 53–56

General Procedure for Intermediates 8

Triethylamine (0.92 mL, 6.6 mmol) and the desired anhydride 7 (6 mmol) were added to a solution of 4-amino acid 6 (6 mmol) in 5 mL DMF. The solution was heated for 16 hours at 120 °C. After returning to room temperature, the solvent was evaporated. The residue was then dissolved in ethyl acetate (100 mL) and washed with 1N HCl (20 mL). The organic layer was washed with brine (25 mL), dried with anhydrous magnesium sulfate. The solution was filtered to yield the desired intermediate 8 that was then used directly or purified by liquid chromatography.

4-endo-Dihydronorbornyl benzoic acid (8a)

beige solid; 84% yield; DCM/MeOH 9/1 Rf 0.5; 1H NMR (300 MHz, CD3OD): 1.41 (d, J = 8.4 Hz, 2H), 1.61–1.69 (m, 4H), 2.89 (br s, 2H), 3.26 (br s, 2H), 7.41 (d, J = 8.7 Hz, 2H), 8.04 (s, OH), 8.19 (d, J = 8.7 Hz, 2H).

4-(endo-Dihydronorbornyl)-trans-cyclohexanecarboxylic acid (8b)

white solid; 35% yield; DCM/MeOH 9/1 Rf 0.8; LC-MS (1000 (+)-5.5-254-95:5) tR 3.49 min; 246.02 [M−COOH], 291.75 [M+1H]; 1H NMR (300 MHz, CDCl3): 1.25 (d, J = 8.4 Hz, 2H), 1.46–1.69 (m, 8H), 2.13 (d, J = 11.7 Hz, 2H), 2.21–2.44 (m 3H), 2.74 (br s, 2H), 2.99 (br s, 2H), 3.98 (tt, J = 12.1 and 3 Hz, 1H), 8.01 (s, OH).

4-(endo-Dihydronorbornyl)-cis-cyclohexanecarboxylic acid (8c)

white solid; 73% yield; Hex/EtOAc 1/1 Rf 0.4; LC-MS (1000 (+)-5.5-254-95:5) tR 3.49 min; 246.02 [M−COOH], 291.75 [M+1H]; 1H NMR (300 MHz, CDCl3): 1.23 (d, J = 8.4 Hz, 2H), 1.46–1.62 (m, 8H), 2.28–2.45 (m, 5H), 2.71 (bs, 2H), 2.97 (bs, 2H), 3.96 (tt, J = 12.1 and 3 Hz, 1H), 8.01 (s, 1H).

4-cis-endo-(3a,4,7,7a-Tetrahydro-1H-4,7-ethanoisoindole-1,3(2H)-dion-2-yl)-benzoic acid (8d)

off-white solid; 99% yield; DCM/MeOH 9/1 Rf 0.2; 1H NMR (300 MHz, DMSO): 1.61-1.38 (m, 4H), 2.16 (s, 2H), 2.92 (s, 2H), 7.41 (d, J = 8.4 Hz, 2H), 8.05 (d, J = 8.7 Hz, 2H), 13.15 (br s, 1H).

4-cis-endo-(Hexahydro-1H-4,7-ethanoisoindole-1,3(2H)-dion-2-yl)-benzoic acid (8e)

white solid; 12% yield; DCM/MeOH 9/1 Rf 0.3; 1H NMR (300 MHz, CDCl3): 1.68-1.61 (m, 4H), 2.16 (s, 2H), 2.92 (s, 2H), 3.25-3.23 (m, 4H), 8.06-8.03 (m, 2H), 7.29-7.26 (m, 2H).

General procedure for compounds 9–39, 53–56

Compounds were prepared in a library fashion. One gram of the appropriate intermediate 8 was heated for 16 hours in 5 mL thionyl chloride at 70 °C. TLC (DCM/MeOH 9/1) of a reaction aliquot diluted in methanol showed complete conversion to the methyl ester. Excess thionyl chloride was removed to afford the acid chloride. The acid chloride was dissolved in 20 mL of DCE. 0.3 mL of the DCE solution was added to 30 mg of the appropriate amine and 0.1 mL of pyridine. Solutions were heated at 50 °C for 16 hours and the crude material was purified by liquid chromatography.

4-(cis-endo-1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(quinolin-8-yl)benzamide (9)

white solid; 51% yield; Hex/EtOAc 3/7 Rf 0.6; LC-MS (1000 (+)-5.5-254-95:5) tR 2.79 min; 411.95 [M+1H] 96.3% at 254 nm; 1H NMR (300 MHz, CDCl3): 1.49 (d, J = 8.4 Hz, 2H), 1.58–1.79 (m, 4H), 2.90 (br s, 2H), 3.29 (br s, 2H), 7.47–7.63 (m, 5H), 8.16–8.23 (m, 3H), 8.84 (dd, J = 4.2 and 1.5 Hz, 1H), 8.92 (dd, J = 6.9 and 1.8 Hz, 1H), 10.79 (br s, NH).

4-(cis-endo-1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(quinolin-8-yl)-trans-cyclohexyl carboxamide (10)

off white solid; 55% yield; Hex/EtOAc 1/1 Rf 0.4; LC-MS (1000 (+)-5.5-254-95:5) tR 5.43 min; 418.02 [M+1H] 96.3% at 254 nm; 1H NMR (300 MHz, CDCl3): 1.28 (d, J = 8.4 Hz, 2H), 1.53–1.64 (m, 4H), 1.71–1.85 (m, 4H), 2.21 (br d, J = 12.3 Hz, 2H), 2.42 (dr q, J = 12.6 and 3.6 Hz, 2H), 2.56 (tt, J = 12 and 3.3 Hz, 1H), 2.76 (br s, 2H), 3.01 (br s, 2H), 4.09 (tt, J = 12.3 and 3.9 Hz, 1H), 7.43–7.55 (m, 3H), 8.15 (dd, J = 8.1 and 1.5 Hz, 1H), 8.76 (dd, J = 6.9 and 2.1 Hz, 1H), 8.81 (dd, J = 3.9 and 1.5 Hz, 1H), 9.92 (br s, 1H).

4-(cis-endo-1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(quinolin-8-yl)-cis-cyclohexyl carboxamide (11)

transparent oil; 50% yield; Hex/EtOAc 1/1 Rf 0.4; LC-MS (1000 (+)-5.5-254-95:5) tR 3.35 min; 418.02 [M+1H] 97.4% at 254 nm; 1H NMR (300 MHz, CDCl3): 1.26 (d, J = 8.4 Hz, 2H), 1.46–1.59 (m, 4H), 1.71–1.85 (m, 4H), 2.18 (br d, J = 12.3 Hz, 2H), 2.34–2.75 (m, 5H), 2.94 (br s, 1H), 3.00 (br s, 1H), 3.98–4.09 (m, 1H), 7.43–7.55 (m, 3H), 8.13 (dd, J = 8.1 and 1.5 Hz, 1H), 8.72–8.82 (m, 2H), 9.89+10.07 (br s, 1H).

4-(cis-endo-1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(naphtalen-1-yl) benzamide (12)

off white solid; 56% yield; Hex/EtOAc 1/1 Rf 0.6; LC-MS (1000 (+)-5.5-254-95:5) tR 3.81 min; 410.88 [M+1H] 95.3% at 254 nm; 1H NMR (300 MHz, CDCl3): 1.41–1.51 (m, 2H), 1.58–1.79 (m, 4H), 2.89 (br s, 2H), 3.28 (br s, 2H), 7.46–7.61 (m, 5H), 7.77 (d, J = 8.7 Hz, 1H), 7.86–7.95 (m, 2H), 8–8.12 (m, 3H), 8.24 (br s, 1H).

4-(cis-endo-1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(isoquinolin-1-yl)-trans-cyclohexylcarboxamide (13)

white solid; 51% yield; DCM/MeOH 9/1 Rf 0.7; LC-MS (1000 (+)-5.5-254-95:5) tR 4.84 min; 418.02 [M+1H] 99.6% at 254 nm 1H NMR (300 MHz, CDCl3+CD3OD): 1.22 (d, J = 8.4 Hz, 2H), 1.51–1.61 (m, 4H), 1.65–1.77 (m, 4H), 2.16 (d, J = 11.7 Hz, 2H), 2.21–2.44 (m 3H), 2.70 (br s, 2H), 2.98 (br s, 2H), 4.03 (tt, J = 12.1 and 3 Hz, 1H), 7.43 (br d, J = 5.4 Hz, 2H), 7.53–7.59 (m, 1H), 7.67 (dt, J = 6.9 and 0.9 Hz, 1H), 7.76 (d, J = 8.1 Hz, 1H), 8.06 (br s, 1H).

4-(cis-endo-1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(isoquinolin-4-yl)-trans-cyclohexylcarboxamide (14)

yellow solid; 16% yield; DCM/MeOH 9/1 Rf 0.4; LC-MS (1000 (+)-5.5-254-95:5) tR 4.10 min; 418.02 [M+1H] 99.9% at 254 nm; 1H NMR (300 MHz, CDCl3+CD3OD): 1.25 (d, J = 8.4 Hz, 2H), 1.51–1.61 (m, 4H), 1.65–1.77 (m, 4H), 2.15 (br d, J = 12.3 Hz, 2H), 2.25–2.43 (m, 3H), 2.74 (br s, 2H), 2.99 (br s, 2H), 3.99 (tt, J = 12.3 and 3.9 Hz, 1H), 7.58–7.73 (m, 3H), 7.82 (d, J = 8.4 Hz, 1H), 7.94 (d, J = 8.1 Hz, 1H), 8.04 (br s, 1H), 8.74 (s, NH).

4-(cis-endo-1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(quinolin-4-yl)-trans-cyclohexyl carboxamide (15)

off-white solid; 5% yield; LC-MS (1000 (+)-5.5-254-95:5) tR 2.75 min; 417.88 [M+1H] 99.4% at 254 nm.

4-(cis-endo-1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(quinolin-5-yl)-trans-cyclohexyl carboxamide (16)

white solid, 40% yield; DCM/MeOH 9/1, Rf 0.4; LC-MS (1000 (+)-5.5-254-95:5) tR 4.20 min; 418.02 [M+1H] 99.5% at 254 nm; 1H NMR (300 MHz, CDCl3): 1.25 (d, J = 8.4 Hz, 2H), 1.53–1.64 (m, 4H), 1.71–1.85 (m, 4H), 2.17 (br d, J = 12.3 Hz, 2H), 2.21–2.49 (m, 3H), 2.72 (br s, 2H), 2.99 (br s, 2H), 4.08 (tt, J = 12.3 and 3.9 Hz, 1H), 7.42 (dd, J = 8.4 and 3.9 Hz, 1H), 7.66–7.83 (m, 2H), 7.98 (d, J = 7.5 Hz, 1H), 8.17 (d, J = 8.4 Hz, 1H), 8.92 (d, J = 4.5 Hz, 1H), 8.82 (br s, 1H).

4-(cis-endo-1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(isoquinolin-5-yl)-trans-cyclohexylcarboxamide (17)

yellow glassy solid; 72% yield; DCM/MeOH 9/1, Rf 0.7; LC-MS (1000 (+)-5.5-254-95:5) tR 3.27 min; 418.02 [M+1H] 92.9% at 254 nm; 1H NMR (300 MHz, CDCl3): 1.28 (d, J = 8.4 Hz, 2H), 1.53–1.64 (m, 4H), 1.71–1.85 (m, 4H), 2.22 (br d, J = 12.3 Hz, 2H), 2.24–2.58 (m, 3H), 2.71 (br s, 2H), 2.99 (br s, 2H), 4.08 (tt, J = 12.3 and 3.9 Hz, 1H), 7.55–7.64 (m, 2H), 7.79 (d, J = 8.4 Hz, 1H), 7.95 (s, 1H), 8.13 (d, J = 7.8 Hz, 1H), 8.53 (d, J = 6.3 Hz, 1H), 9.24 (s, 1H).

4-(cis-endo-1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(isoquinolin-8-yl)-trans-cyclohexylcarboxamide (18)

yellow solid; 16% yield; DCM/MeOH 9/1, Rf 0.8; LC-MS (1000 (+)-5.5-254-95:5) tR 4.04 min; 418.02 [M+1H] 90.1% at 254 nm; 1H NMR (300 MHz, CDCl3): 1.27 (d, J = 6.9 Hz, 2H), 1.53–1.69 (m, 4H), 1.71–1.88 (m, 4H), 2.20 (br d, J = 12.4 Hz, 2H), 2.24–2.58 (m, 4H), 2.73 (br s, 2H), 3.00 (br s, 2H), 4.01–4.16 (m, 1H), 7.57–7.64 (m, 2H), 8.00–8.06 (m, 1H), 8.20 (br s, 1H), 8.53 (d, J = 5.8 Hz, 1H), 9.43 (s, NH).

4-(cis-endo-1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(p-tolyl)benzamide (19)

off white solid; 14% yield; Hex/EtOAc, 1/1; Rf 0.6; LC-MS (1000 (+)-5.5-254-95:5) tR 4.55 min; 374.88 [M+1H] 98.5% at 254 nm; 1H NMR (CDCl3): 1.46 (d, J = 8.4 Hz, 2H), 1.53–1.74 (m, 4H), 2.34 (s, 3H), 2.89 (br s, 2H), 3.27 (br s, 2H), 7.18 (d, J = 8.1 Hz, 2H), 7.41 (d, J = 8.7 Hz, 2H), 7.51 (d, J = 8.1 Hz, 2H), 7.79 (br s, 1H), 7.95 (d, J = 8.4 Hz, 2H).

4-(cis-endo-1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(4-methoxyphenyl) benzamide (20)

grey solid; 20% yield; Hex/EtOAc, 1/1; Rf 0.4; LC-MS (1000 (+)-5.5-254-95:5) tR 3.73 min; 390.82 [M+1H] 93.4% at 254 nm; 1H NMR (CDCl3): 1.19–1.26 (m, 2H), 1.45–1.74 (m, 4H), 2.89 (br s, 2H), 3.27 (br s, 2H), 3.82 (s, 3H), 6.91 (d, J = 8.4 Hz, 2H), 7.41 (d, J = 8.4 Hz, 2H), 7.54 (d, J = 8.4 Hz, 2H), 7.78 (s, NH), 7.95 (d, J = 8.4 Hz, 2H).

2-cis-endo-(1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(2-methoxyphenyl) benzamide (21)

white solid; 91% yield; Hex/EtOAc 3/7, Rf 0.6; LC-MS (1000 (+)-5.5-254-95:5) tR 4.12 min; 390.35 [M+1H] 100% at 254 nm; 1H NMR (300 MHz, CDCl3): 1.47 (d, J = 8.4 Hz, 2H), 1.58–1.79 (m, 4H), 2.89 (br s, 2H), 3.28 (br s, 2H), 3.92 (s, 3H), 6.92 (dd, J = 7.8 and 1.2 Hz, 1H), 6.99–7.13 (m, 2H), 7.41–7.45 (m, 2H), 7.97–8.02 (m, 2H), 8.51 (dd, J = 7.8 and 1.2 Hz, 1H), 8.52 (br s, 1H).

2-cis-endo-(1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(4-(dimethylamino)phenyl) benzamide (22)

dark green solid; 36% yield; (DCM/MeOH 9/1, Rf 0.2; LC-MS (1000 (+)-5.5-254-95:5) tR 2.17 min; 403.95 [M+1H] 99.6% at 254 nm; 1H NMR (CDCl3): 1.46 (d, J = 8.4 Hz, 2H), 1.59–1.76 (m, 4H), 2.88 (br s, 2H), 2.94 (s, 6H), 3.27 (br s, 2H), 6.74 (d, J = 9.3 Hz, 2H), 7.38 (d, J = 8.1 Hz, 2H), 7.48 (d, J = 9.3 Hz, 2H), 7.74 (s, NH), 7.94 (d, J = 8.1 Hz, 2H).

2-cis-endo-(1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(4-cyanophenyl)benzamide (23)

white solid; 48% yield; DCM/MeOH, 95/5, Rf 0.1; LC-MS (1000 (+)-5.5-254-95:5) tR 3.48 min; 347.68 [M+1H] 90%; 1H NMR (CDCl3+CD3OD): 1.33 (d, J = 8.4 Hz, 2H), 1.55–1.69 (m, 5H), 2.77 (br s, 2H), 3.20 (br s, 1H), 7.26 (d, J = 8.2 Hz, 2H), 7.53 (d, J = 9.1 Hz, 2H), 7.77 (d, J = 9.1 Hz, 2H), 7.90 (d, J = 8.8 Hz, 2H).

2-cis-endo-(1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(4-acetylphenyl)benzamide (24)

off white solid; 35% yield; DCM/MeOH, 95/5, Rf 0.4; LC-MS (1000 (+)-5.5-254-95:5) tR 2.98 min; 402.84 [M+1H] 97.9% at 254 nm.

4-(cis-endo-1,3-dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(2-hydroxyphenyl)-trans-cyclohexylcarboxamide (25)

orange solid; 43% yield; Hex/EtOAc 1/1, Rf 0.4; LC-MS (1000 (+)-5.5-254-95:5) tR 3.01 min; 418 [M+1H] 99% at 254 nm; 1H NMR (300 MHz, CD3OD): 1.23 (d, J = 8.4 Hz, 2H), 1.53–1.71 (m, 4H), 1.75–2.01 (m, 4H), 2.06 (br d, J = 12.3 Hz, 2H), 2.28–2.35 (m, 3H), 2.72 (br s, 2H), 2.98 (br s, 2H), 3.98–4.14 (m, 1H), 6.76–6.87 (m, 2H), 6.96 (dd, J = 7.98 and 1.5 Hz, 1H), 7.58 (dd, J = 7.98 and 1.5 Hz, 1H).

4-(cis-endo-1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(2-ethoxyphenyl)-trans-cyclohexyl carboxamide (26)

white solid; 73% yield; Hex/EtOAc 1/1, Rf 0.6; LC-MS (1000 (+)-5.5-254-95:5) tR 3.46 min; 418 [M+1H] 99% at 254 nm; 1H NMR (300 MHz, CDCl3): 1.23 (d, J = 8.4 Hz, 2H), 1.53–1.71 (m, 10H), 2.06 (br d, J = 12.3 Hz, 2H), 2.28–2.35 (m, 4H), 2.72 (br s, 2H), 2.98 (br s, 2H), 3.98–4.14 (m, 3H), 6.83 (dd, J = 7.5 and 1.5 Hz, 1H), 6.87–7.11 (m, 2H), 7.85 (br s, NH), 8.34 (dd, J = 7.5 and 1.5 Hz, 1H).

Methyl 2-(4-(cis-endo-1,3-dioxooctahydro-2H-4,7-methanoisoindol-2-yl))-N-trans-cyclohexanecarboxamido)benzoate (27)

white solid; 58% yield; DCM/MeOH 95/5 Rf 0.8; LC-MS (1000 (+)-5.5-254-95:5) tR 3.58 min; 418 [M+1H] 92.4% at 254 nm; 1H NMR (300 MHz, CDCl3): 1.24 (d, J = 8.4 Hz, 2H), 1.51–1.72 (m, 8H), 2.14 (br d, J = 11.7 Hz, 2H), 2.26–2.41 (m, 3H), 2.73 (br s, 2H), 2.99 (br s, 2H), 3.91 (s, 3H), 4.03 (t, J = 12.6 Hz, 1H), 7.05 (t, J = 7.8 Hz, 1H), 7.51 (t, J = 8.7 Hz, 1H), 8 (dd, J = 7.8 and 1.2 Hz, 1H), 8.71 (d, J = 8.7 Hz, 1H), 11.15 (br s, 1H).

4-(cis-endo-1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(2-acetylphenyl)-trans-cyclohexyl carboxamide (28)

brown solid; 38% yield; Hex/EtOAc 1/1 Rf: 0.4; LC-MS (1000 (+)-5.5-254-95:5) tR 4.40 min; 408.28 [M+1H] 94.7% at 254 nm; 1H NMR (300 MHz, CDCl3): 1.26 (d, J = 8.4 Hz, 2H), 1.53–1.85 (m, 8H), 2.16 (br d, J = 12.3 Hz, 2H), 2.28–2.49 (m, 3H), 2.67 (s, 3H), 2.75 (br s, 2H), 3.01 (br s, 2H), 4.08 (tt, J = 12.3 and 3.9 Hz, 1H), 7.11 (t, J = 7.5 Hz, 1H), 7.55 (t, J = 8.7 and 1.5 Hz, 1H), 7.90 (dd, J = 7.8 and 1.2 Hz, 1H), 8.76 (d, J = 8.4 Hz, 1H), 11.84 (s, NH).

4-(cis-endo-(1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl))-N-(8-oxo-5,6,7,8-tetrahydro naphthalen-1-yl)-trans-cyclohexylcarboxamide (29)

brown oil; 51% yield; Hex/EtOAc 1/1 Rf 0.6; LC-MS (1000 (+)-5.5-254-95:5) tR 3.73 min; 434.48 [M+1H] 97.5% at 254 nm; 1H NMR (300 MHz, CDCl3): 1.21 (d, J = 8.4 Hz, 2H), 1.53–1.67 (m, 8H), 2.07–2.15 (m, 4H), 2.19–2.54 (m, 3H), 2.61–2.78 (m, 4H), 2.88-3.-02 (m, 4H), 3.98–4.14 (m, 1H), 6.91 (dd, J = 7.5 and 1.5 Hz, 1H), 7.42 (t, J = 7.5 Hz, 1H), 8.60 (dd, J = 7.5 and 1.5 Hz, 1H), 12.23 (br s, 1H).

4-(cis-endo-(1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl))-N-(9H-fluoren-9-on-1-yl)-trans-cyclohexylcarboxamide (30)

yellow solid; 46% yield; DCM/MeOH 95/5 Rf 0.7; LC-MS (1000 (+)-5.5-254-95:5) tR 4.38 min; 469 [M+1H] 99.4% at 254 nm; 1H NMR (300 MHz, CDCl3): 1.26 (d, J = 8.4 Hz, 2H), 1.47–1.74 (m, 8H), 2.18 (br d, J = 11.7 Hz, 2H), 2.26–2.48 (m, 3H), 2.75 (br s, 2H), 3.00 (br s, 2H), 4.04 (t, J = 12.6 Hz, 1H), 7.15 (t, J = 7.2 Hz, 1H), 7.25–7.31 (m, 1H), 7.38–7.47 (m, 3H), 7.58 (d, J = 7.5 Hz, 1H), 8.34 (d, J = 8.1 Hz, 1H), 10.17 (br s, NH).

4-(cis-endo-(1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl))-N-(pyridin-2-yl)-trans-cyclohexyl carboxamide (31)

off white solid; 6% yield; DCM/MeOH 9/1 Rf 0.7; LC-MS (1000 (+)-5.5-254-95:5) tR 1.27 min; 367.88 [M+1H] 97.5% at 254 nm; 1H NMR (300 MHz, CDCl3): 1.26 (d, J = 8.4 Hz, 2H), 1.51–1.74 (m, 9H), 2.09 (br d, J = 11.7 Hz, 2H), 2.26–2.39 (m, 2H), 2.75 (br s, 2H), 3.01 (br s, 2H), 4.01–4.17 (m, 1H), 7.01–7.05 (m, 1H), 7.69 (dt, J = 9 and 2.1 Hz, 1H), 7.99 (br s, NH), 8.21 (d, J = 8.7 Hz, 1H), 8.24–8.27 (m, 1H).

4-(cis-endo-(1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl))-N-(pyridin-3-yl)-trans-cyclohexyl carboxamide (32)

white solid; 32% yield; DCM/MeOH 9/1 Rf 0.8; LC-MS (1000 (+)-5.5-254-95:5) tR 2.46 min; 367.88 [M+1H] 97.9% at 254 nm; 1H NMR (300 MHz, CDCl3+1 drop CD3OD): 1.28 (d, J = 8.4 Hz, 2H), 1.53–1.64 (m, 4H), 1.71–1.85 (m, 4H), 1.86 (br d, J = 12.3 Hz, 2H), 2.12–2.38 (m, 3H), 2.76 (br s, 2H), 3.01 (br s, 2H), 4.08 (tt, J = 12.3 and 3.9 Hz, 1H), 7.21 (dd, J = 8.4 and 3.6 Hz, 1H), 8.14 (br d, J = 3.6 Hz, 1H), 8.20 (br d, J = 8.4 Hz, 1H), 8.42 (d, J = 2.1 Hz, 1H)

4-(cis-endo-(1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl))-N-(pyridin-4-yl)-trans-cyclohexyl carboxamide (33)

yellow solid; 8% yield; DCM/MeOH 9/1 Rf 0.7; LC-MS (1000 (+)-5.5-254-95:5) tR 1.68 min; 367.89 [M+1H] 97.3% at 254 nm; 1H NMR (300 MHz, CDCl3): 1.25 (d, J = 8.4 Hz, 2H), 1.46–1.75 (m, 8H), 2.07 (br d, J = 12.3 Hz, 2H), 2.23–2.37 (m, 3H), 2.75 (br s, 2H), 3.02 (br s, 2H), 4.08 (tt, J = 12.3 and 3.9 Hz, 1H), 7.31 (br s, NH), 7.48 (d, J = 6.3 Hz, 2H), 8.50 (br d, J = 6.3 Hz, 2H).

4-(cis-endo-(1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl))-N-(2-acetyl-pyridin-3-yl)-trans-cyclohexylcarboxamide (34)

off-white solid; 17% yield; Hex/EtOAc 7/3 Rf 0.2; LC-MS (1000 (+)-5.5-254-95:5) tR 4.35 min; 409.28 [M+1H] 96.4% at 254 nm; 1H NMR (300 MHz, CDCl3): 1.26 (d, J = 8.8 Hz, 2H), 1.51–1.78 (m, 8H), 2.18 (br d, J = 11.8 Hz, 2H), 2.31–2.52 (m, 5H), 2.76 (br s, 1H), 2.80 (s, 3H), 3.03 (br s, 1H), 3.97–4.15 (m, 1H), 7.43–7.48 (m, 1H), 8.35 (dd, J = 4.4 and 0.9 Hz, 1H), 9.09 (dd, J = 8.8 and 0.9 Hz, 1H), 11.62 (br s, NH).

4-(cis-endo-(1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl))-N-(6-acetyl-pyridin-3-yl)-trans-cyclohexylcarboxamide (35)

off-white solid; 18% yield; Hex/EtOAc 7/3 Rf 0.2; LC-MS (1000 (+)-5.5-254-95:5) tR 3.46 min; 409.48 [M+1H] 98.6% at 254 nm; 1H NMR (300 MHz, CDCl3): 1.28 (d, J = 8.4 Hz, 2H), 1.49–1.78 (m, 8H), 2.17 (d, J = 11.7Hz, 2H), 2.29–2.49 (m, 5H), 2.71 (s, 3H), 2.77 (br s, 1H), 2.99 (br s, 1H), 3.98–4.11 (m, 1H), 7.64 (d, J = 5.23 Hz, 1H), 8.47 (d, J = 5.53 Hz, 1H), 10.05 (s, 1H), 11.20 (s, NH).

4-cis-endo-(3a,4,7,7a-Tetrahydro-1H-4,7-ethanoisoindole-1,3(2H)-dion-2-yl)-N-(quinolin-8-yl) benzamide (36)

yellow solid; 10% yield; Hex/EtOAc 1/1 Rf 0.3; LC-MS (1000 (+)-5.5-254-95:5) tR 3.87 min; 423.68 [M+1H] 96% at 254 nm; 1H NMR (300 MHz, CDCl3): 1.70-1.45 (m, 4H), 3.06 (s, 2H), 3.29 (br s, 2H), 6.33-6.31 (m, 2H), 7.60-7.41 (m, 5H), 8.21-8.14 (m, 3H), 8.85-8.83 (m, 1H), 8.93-8.90 (m, 1H), 10.74 (s, 1H)

4-cis-endo-(Hexahydro-1H-4,7-ethanoisoindole-1,3(2H)-dion-2-yl)-N-(quinolin-8-yl) benzamide (37)

white solid; 51% yield; Hex/EtOAc 3/7 Rf 0.6; LC-MS (1000 (+)-5.5-254-95:5) tR 2.79 min; 411.95 [M+1H] 96.3% at 254 nm; 1H NMR (300 MHz, DMSO): 1.48–1.53 (m, 4H), 1.66–1.68 (m, 4H), 2.08–2.10 (m, 2H), 3.08–3.10 (m, 2H), 7.55 (d, J = 8.8 Hz, 2H), 7.66–7.70 (m, 2H), 7.77 (dd, J = 8.4 and 1.5 Hz, 1H), 8.17 (d, J = 8.8 Hz, 2H), 8.48 dd, J = 8.4 and 1.5 Hz, 1H), 8.73 (dd, J = 8.4 and 1.5 Hz, 1H), 8.98–8.99 (m, 1H), 10.70 (br s, NH).

4-(cis-endo-8-Methyl-1,3-dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(quinolin-8-yl) benzamide (38)

orange solid; 64% yield; Hex/EtOAc 1/1 Rf 0.4; 1H NMR (300 MHz, DMSO): 1.47 (d, J = 8.4 Hz, 1H), 1.58–1.79 (m, 1H), 1.83 (s, 3H), 3.28 (br s, 2H), 3.92 (s, 2H), 5.87 (s, 1H), 7.49–7.55 (m, 2H), 7.52–7.77 (m, 3H), 8.13 (d, J = 8.8 Hz, 2H), 8.45 (dd, J = 8.53 and 0.9 Hz, 1H), 8.69 (dd, J = 8.53 and 0.9 Hz, 1H), 8.95 (dd, J = 4.4 and 0.9 Hz, 1H).

4-(Hexahydro-1H-4,7-epoxyisoindole-1,3(2H)-dion-2-yl)-N-(quinolin-8-yl)benzamide (39)

7-Oxabicyclo[2.2.1]heptane-2,3-dicarboxylic anhydride was prepared according to literature procedures22 beige solid; 68% yield; DCM/MeOH 9/1 Rf 0.8; LC-MS (1000 (+)-5.5-254-95:5): tR 3.23 min, 414 [M+1H] >99% at 254 nm; 1H NMR (300 MHz, CDCl3): 1.68–1.75 (m, 2H), 1.94–1.98 (m, 2H), 3.11 (s, 2H), 5.05 (m, 2H), 7.47–7.63 (m, 5H), 8.16–8.19 (m, 2H), 8.21 (br d, J = 1.67 Hz, 1H), 8.85 (dd, J = 1.67, 4.21 Hz, 1H), 8.91 (dd, J = 1.89, 7.08 Hz, 1H), 10.75 (s, 1H).

4-(cis-endo-1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(4-bromo-2-methoxy phenyl)-trans-cyclohexylcarboxamide (53)

off-white solid; 11% yield; DMC/MeOH 9/1 Rf 0.8; LC-MS (1000 (+)-5.5-254-95:5) tR 3.98 min; 474.48-476.48 [M+1H] 90.3% at 254nM; 1H NMR (300 MHz, CDCl3): 1.11 (d, J = 8.4 Hz, 2H), 1.45–1.77 (m, 6H), 2.04 (br d, J = 12 Hz, 4H), 2.08–2.41 (m, 4H), 2.74 (br s, 2H), 2.99 (br s, 2H), 3.87 (s, 3H), 4.04 (t, J = 12.6 Hz, 1H), 7 (s, 1H), 7.07 (d, J = 7 Hz, 1H), 7.87 (br s, NH), 8.24 (d, J = 7 Hz, 1H).

4-(cis-endo-1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(5-hydroxyl-pyridin-3-yl)-trans-cyclohexylcarboxamide (54)

beige solid, 21% yield; DCM/MeOH, 9/1; Rf 0.7; LC-MS (1000 (+)-5.5-254-95:5) tR 1.47 min; 383.88 [M+1H] 97.2% at 254nM; 1H NMR (300 MHz, CDCl3+CD3OD): 1.06 (d, J = 8.4 Hz, 2H), 1.39–1.59 (m, 6H), 1.82 (br d, J = 11.7 Hz, 2H), 2.06–2.38 (m, 3H), 2.58 (br s, 2H), 2.88 (br s, 2H), 4.04 (t, J = 12.6 Hz, 1H), 7.59 (s, 1H), 7.65 (s, 1H), 7.87 (s, 1H).

4-(cis-endo-1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(2-chloropyridin-3-yl)-trans-cyclohexylcarboxamide (55)

pale yellow solid, 35% yield; DCM/MeOH, 9/1; Rf 0.6; LC-MS (1000 (+)-5.5-254-95:5) tR 2.91 min; 401.68 [M+1H] 98.7% at 254nM; 1H NMR (300 MHz, CDCl3): 1.26 (d, J = 8.4 Hz, 2H), 1.51–1.74 (m, 6H), 2.18 (br d, J = 11.7 Hz, 2H), 2.26–2.48 (m, 3H), 2.73 (br s, 2H), 3.00 (br s, 2H), 3.92–4.02 (m, 1H), 7.25 (dd, J = 8.4 and 4.9 Hz, 1H), 7.69 (br s, NH), 8.10 (dd, J = 4.9 and 1.8 Hz, 1H), 8.73 (dd, J = 8.4 and 1.8 Hz, 1H).

4-(cis-endo-1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(2-acetylphenyl)-cis-cyclohexyl carboxamide (56)

beige wax; 40% yield; Hex/EtOAc 1/1 Rf 0.5; LC-MS (1000 (+)-5.5-254-95:5) tR 4.06 min; 408.42 [M+1H] 96.2% at 254nM; 1H NMR (300 MHz, CDCl3): 1.24 (d, J = 8.4 Hz, 2H), 1.48–1.79 (m, 7H), 2.29–2.46 (m, 5H), 2.64 (s, 3H), 2.72 (br s, 2H), 2.98 (br s, 2H), 3.97–4.08 (m, 1H), 7.07–7.13 (m, 1H), 7.54 (tt, J = 8.7 and 1.5 Hz, 1H), 7.89 (td, J = 8.1 and 1.5 Hz, 1H), 8.76 + 8.87 (dd, J = 8.7 and 0.9 Hz, 1H), 11.80 + 11.96 (br s, NH).

3-(cis-endo-1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(quinolin-8-yl)benzamide (45)

Compound 45 was synthesized according to procedures described for compounds 939, 5356 but replaced 4-aminobenzoic acid by 3-aminobenzoic acid. white solid; 96% yield; Hex/EtOAc 1/1; Rf 0.4; LC-MS (1000 (+)-5.5-254-95:5) tR 3.91 min; 411.82 [M+1H] 98.9% at 254 nm; 1H NMR (500 MHz, d6-DMSO): 1.32–1.37 (m, 2H), 1.57–1.64 (m, 3H), 1.70–1.72 (m, 1H), 2.70–2.72 (m, 2H), 3.31–3.33 (m, 2H, overlaps with the residual H2O signal), 7.53–7.55 (m, 1H), 7.66–7.70 (m, 2H), 7.75–7.79 (m, 2H), 7.88–7.89 (m, 1H), 8.10–8.12 (m, 1H), 8.47 (dd, J = 8.4 and 1.5 Hz, 1H), 8.69 (dd, J = 8.4 and 1.5 Hz, 1H), 8.96–8.97 (m, 1H), 10.68 (br s, NH, 1H).

5-(cis-endo-1,3-Dioxooctahydro-2H-4,7-methanoisoindol-2-yl)-N-(quinolin-8-yl)furan-2-carboxamide (46)

Compound 46 was synthesized according to procedures described for compounds 4144 but replaced 4-nitrobenzoic acid by 5-nitrofuroic acid. orange oil; 20% yield; DCM/MeOH 9/1 Rf 0.7; LC-MS (1000 (+)-5.5-254-95:5) tR 4.12 min; 401.82 [M+1H]; 90% at 254 nm; 1H NMR (CDCl3): 1.2-1.82 (m, 5H), 2.92 (br s, 2H), 3.33 (br s, 2H), 7.36 (d, J = 3.6 Hz, 1H), 7.43–7.48 (m, 3H), 7.54–7.58 (m, 1H), 8.17 (dd, J = 8.25 and 1.65 Hz, 1H), 8.76 (dd, J = 6.9 and 2.1 Hz, 1H), 8.87 (dd, J = 3.9 and 1.5 Hz, 1H), 10.74 (br s, NH).

Scheme 5, Step a: N-(4-Nitrophenyl)-(3aR,4S,7R,7aS)-1,3,3a,4,7,7a-hexahydro-1,3-dioxo-4,7-methano-2H-isoindol (51)

Compound 51 was synthesized according to the procedure described for intermediate 8 but starting from 1.64 g carbic anhydride (10 mmol), 1.67 mL triethylamine and 1.52 g 4-nitroaniline. Compound 51 was isolated as a yellow solid with 52% yield (DCM/MeOH, 9/1; Rf 0.6). LC-MS (1000 (+)-5.5-254-95:5): tR 3.35 min; 96.8 % at 254 nm. 1H NMR (300 MHz, CDCl3): 1.62–1.68 (m, 1H), 1.80–1.86 (m, 1H), 3.49 (dd, J = 1.63, 3.09 Hz, 2H), 3.55 (m, 2H), 6.27 (br t, J = 1.80 Hz, 2H), 7.40–7.44 (m, 2H), 8.26–8.31 (m, 2H).

Scheme 5, Step b: N-(4-Aminophenyl)-(3aR,4S,7R,7aS)-1,3,3a,4,7,7a-hexahydro-1,3-dioxo-4,7-methano-2H-isoindol

300 mg of 51 was dissolved in 30 mL of EtOH and 40 mg of Pd/C (10%) added. The mixture was stirred at room temperature for 16 hours under a hydrogen atmosphere. The solution was filtered through a pad of Celite and concentrated to dryness. The crude product was purified by flash chromatography (Hex/EtOAc 1/1 Rf 0.2) to afford 180 mg (67% yield). LC-MS (1000 (+)-10.0-254-95:5): tR 3.14 min, 257 [M+1H]; 98.8% at 254 nm. 1H NMR (300 MHz, CDCl3): 1.43–1.72 (m, 6H), 2.85 (m, 2H), 3.21 (m, 2H), 3.81 (br s, 2H, NH2), 6.71–6.75 (m, 2H), 6.98–7.01 (m, 2H).

Scheme 5, Step c: N-[4-((3aR,4S,7R,7aS)-1,3,3a,4,7,7a-Hexahydro-1,3-dioxo-4,7-methano-2H-isoindol-2-yl)phenyl]-quinoline-8-carboxamide (52)

Quinoline-8-carbonyl chloride was freshly prepared by treating 135 mg (0.78 mmol) of quinoline-8-carboxylic acid with an excess of thionyl chloride (622 μL) for three hours at 60–70 °C. After returning to room temperature, excess thionyl chloride was removed by evaporation. The mixture was added dropwise to a solution of 100 mg of the aniline precursor from step b (0.39 mmol) and 5 equivalents of triethylamine (1.95 mmol) in 7 mL of dry acetonitrile. The mixture was stirred at room temperature for three days, concentrated in a vacuum and purified by flash chromatography (toluene/EtOAc/acetone 6/3/1 Rf 0.5). The product was obtained as an off-white solid with 87% yield (140 mg). LC-MS (1000 (+)-5.5-254-95:5): tR 3.45 min, 412 [M+1H]; >99% a 254 nm; 1H NMR (300 MHz, CDCl3): 1.48–1.75 (m, 6H), 2.88 (m, 2H), 3.25 (m, 2H), 7.27–7.32 (m, 2H), 7.56 (dd, J = 4.3, 8.3 Hz, 1H), 7.74 (dd, J = 7.45, 7.50 Hz, 1H), 7.98–8.03 (m, 2H), 8.02 (br d, J = 1.6 Hz, 1H), 8.34 (dd, J = 1.8, 8.4 Hz, 1H), 8.96 (dd, J = 1.6, 7.4 Hz, 1H), 9.01 (dd, J = 1.8, 4.3 Hz, 1H).

Supplementary Material

1_si_001

Acknowledgments

We thank Alyssa Morgosh and Claire Johns at HBRI for their technical assistance with compound preparation during their internship at HBRI, and Drs. Zebin Xia and Marcia I. Dawson (SBMRI) for kindly providing 3 for biological testing. We thank the NIH (R01HL059502 and R33HL088266 to MM) and the California Institute for Regenerative Medicine (CIRM, RC1-000132 to MM) and the T Foundation (to JRC) for their generous financial support. We gratefully acknowledge postdoctoral fellowships from the German Research Foundation (DFG) (to DS), and CIRM (TG-0004) and American Heart Association (to EW).

Abbreviations

hESC
Human Embryonic Stem Cell
HCS
High Content Screen
HEK293T
Human Embryonic Kidney 293T
IWR
Inhibitor of Wnt Response

Footnotes

Supporting Information Available The contents of Supporting Information includes the following: (1) analytical and spectral characterization data for all compounds obtained from Schemes 24, (2) detailed assay procedures for the hESC assay, (3) cell viability/toxicity data for selected compounds, (4) data for TGFβ inhibition for selected compounds, (5) chemical stability data for select compounds. This material is available free of charge via the Internet at http://pubs.acs.org.

References

1. Murata M, Tohyama S, Fukuda K. Impacts of recent advances in cardiovascular regenerative medicine on clinical therapies and drug discovery. Pharmacol Ther. 2010;126(2):109–118. [PubMed]
2. Willems E, Lanier M, Forte E, Lo F, Cashman J, Mercola M. A chemical biology approach to myocardial regeneration. J Cardiovasc Transl Res. 2011;4(3):340–350. [PMC free article] [PubMed]
3. Sartipy P, Bjorquist P, Strehl R, Hyllner J. The application of human embryonic stem cell technologies to drug discovery. Drug Discov Today. 2007;12(17–18):688–699. [PubMed]
4. Mercola M, Ruiz-Lozano P, Schneider MD. Cardiac muscle regeneration: lessons from development. Genes Dev. 2011;25(4):299–309. [PubMed]
5. Noseda M, Peterkin T, Simões FC, Patient R, Schneider MD. Cardiopoietic factors: Extracellular signals for cardiac lineage commitment. Circulation Research. 2011;108:129–152. [PubMed]
6. Willems E, Spiering S, Davidovics H, Lanier M, Xia Z, Dawson M, Cashman J, Mercola M. Small molecule inhibitors of the Wnt pathway potently promote cardiogenesis in human embryonic stem cell-derived mesoderm. Circulation Research. 2011;109:360–364. [PMC free article] [PubMed]
7. Chen B, Dodge ME, Tang W, Lu J, Ma Z, Fan CW, Wei S, Hao W, Kilgore J, Williams NS, Roth MG, Amatruda JF, Chen C, Lum L. Small molecule-mediated disruption of Wnt-dependent signaling in tissue regeneration and cancer. Nat Chem Biol. 2009;5(2):100–107. [PMC free article] [PubMed]
8. Lu J, Ma Z, Hsieh JC, Fan CW, Chen B, Longgood JC, Williams NS, Amatruda JF, Lum L, Chen C. Structure-activity relationship studies of small-molecule inhibitors of Wnt response. Bioorg Med Chem Lett. 2009;19(14):3825–3827. [PMC free article] [PubMed]
9. Luu HH, Zhang R, Haydon RC, Rayburn E, Kang Q, Si W, Park JK, Wang H, Peng Y, Jiang W, He TC. Wnt/beta-catenin signaling pathway as a novel cancer drug target. Curr Cancer Drug Targets. 2004;4(8):653–671. [PubMed]
10. Tian Y, Cohen ED, Morrisey EE. The importance of Wnt signaling in cardiovascular development. Pediatr Cardiol. 2010;31(3):342–348. [PubMed]
11. Schneider VA, Mercola M. Wnt antagonism initiates cardiogenesis in Xenopus laevis. Genes Dev. 2001;15(3):304–315. [PubMed]
12. Marvin MJ, Di Rocco G, Gardiner A, Bush SM, Lassar AB. Inhibition of Wnt activity induces heart formation from posterior mesoderm. Genes Dev. 2001;15(3):316–27. [PubMed]
13. Yang L, Soonpaa MH, Adler ED, Roepke TK, Kattman SJ, Kennedy M, Henckaerts E, Bonham K, Abbott GW, Linden RM, Field LJ, Keller GM. Human cardiovascular progenitor cells develop from a KDR+ embryonic-stem-cell-derived population. Nature. 2008;453(7194):524–528. [PubMed]
14. Huang SM, Mishina YM, Liu S, Cheung A, Stegmeier F, Michaud GA, Charlat O, Wiellette E, Zhang Y, Wiessner S, Hild M, Shi X, Wilson CJ, Mickanin C, Myer V, Fazal A, Tomlinson R, Serluca F, Shao W, Cheng H, Shultz M, Rau C, Schirle M, Schlegl J, Ghidelli S, Fawell S, Lu C, Curtis D, Kirschner MW, Lengauer C, Finan PM, Tallarico JA, Bouwmeester T, Porter JA, Bauer A, Cong F. Tankyrase inhibition stabilizes axin and antagonizes Wnt signalling. Nature. 2009;461(7264):614–620. [PubMed]
15. Gonsalves FC, Klein K, Carson BB, Katz S, Ekas LA, Evans S, Nagourney R, Cardozo T, Brown AM, DasGupta R. An RNAi-based chemical genetic screen identifies three small-molecule inhibitors of the Wnt/wingless signaling pathway. Proc Natl Acad Sci U S A. 2011;108(15):5954–5963. [PubMed]
16. Saraswati S, Alfaro MP, Thorne CA, Atkinson J, Lee E, Young PP. Pyrvinium, a potent small molecule Wnt inhibitor, promotes wound repair and post-MI cardiac remodeling. PLoS One. 2011;5(11):e15521. [PMC free article] [PubMed]
17. Wang H, Hao J, Hong CC. Cardiac induction of embryonic stem cells by a small molecule inhibitor of Wnt/beta-catenin signaling. ACS Chem Biol. 2011;6(2):192–197. [PMC free article] [PubMed]
18. Letamendia A, Labbe E, Attisano L. Transcriptional regulation by Smads: crosstalk between the TGF-beta and Wnt pathways. J Bone Joint Surg Am. 2001;83-A(Suppl 1Pt 1):S31–39. [PubMed]
19. Lim JY, Kim WH, Kim J, Park SI. Involvement of TGF-beta1 signaling in cardiomyocyte differentiation from P19CL6 cells. Mol Cells. 2007;24(3):431–436. [PubMed]
20. Kattman SJ, Witty AD, Gagliardi M, Dubois NC, Niapour M, Hotta A, Ellis J, Keller G. Stage-specific optimization of activin/nodal and BMP signaling promotes cardiac differentiation of mouse and human pluripotent stem cell lines. Cell Stem Cell. 8(2):228–240. [PubMed]
21. Gustafson TA, Markham BE, Morkin E. Analysis of thyroid hormone effects on myosin heavy chain gene expression in cardiac and soleus muscles using a novel dot-blot mRNA assay. Biochem Biophys Res Commun. 1985;130(3):1161–1167. [PubMed]
22. Sprague PW, Heikes JE, Gougoutas JZ, Malley MF, Harris DN, Greenberg R. Synthesis and in vitro pharmacology of 7-oxabicyclo[2.2. 1]heptane analogues of thromboxane A2/PGH2. J Med Chem. 1985;28(11):1580–1590. [PubMed]