2-Furanylmethyl N-(2-propenyl)carbamate

Abstract

The overexpression of protein phosphatase 5 (PP5) has been correlated to tumor cell reproduction, making it a candidate for small molecule drug therapy. Prior work has focused on functionalized and decorated scaffolds that maximize contacts within and around the active site. The assembly and testing of cantharidin derivatives decorated with functionalized attachments has been our focus in order to affect the optimal binding of PP5. Condensation of 2-hydroxymethylfuran with allyl isocyanate meets the metrics of the rapid installment of functionality, as part of the core scaffold. Once condensed, cycloaddition followed by hydrogenation produces the desired derivative of norcantharidin in three synthetic steps.

1. Introduction

In the last two decades, a significant amount of cancer research has been devoted to the study of protein kinases [1]. Protein kinases play a crucial role in the activation and inactivation of proteins important in the regulation of cell cycle progression, and thus the connection to tumorigenesis. With the recent developments of inhibitory activities on antiproliferative pathways, a novel approach to antitumor therapy has, as a focus, the counterpart of kinases, protein phosphatases [2]. Protein phosphatases reverse the actions of kinases by dephosphorylating substrates. Of the phosphatases which act on serine and threonine residues of proteins, phosphoprotein phosphatase 5, or PP5, is a unique serine/threonine phosphatase whose genetic coding sequence is found on chromosome 19 in humans.

Not surprisingly, PP5 is in many protein complexes that contribute to signaling networks that regulate cellular proliferation and apoptosis. An analysis of human breast cancer cells revealed a correlation between the overexpression of PP5 and cancer cell development [3]. Researchers who performed a follow-up study in order to determine if overexpression of PP5 existed, found a direct and positive effect on tumor growth. The experiment compared tumor size development in mice that were injected with differing amounts of PP5 and concluded that PP5 overexpression was directly proportional to tumor size.

The catalytic activity of PP5 generally occurs as part of a complex of proteins, with its tetratricopeptide (TPR) domain used in protein–protein interactions that make the complex possible [4]. As studies have shown that a PP5 mutant consisting only of its TPR domain was sufficient in inhibiting GR-mediated transcription, we propose that the inhibition of the GR-mediated antiproliferative pathway should not only be limited to the phosphatase activity of PP5 but also to actions that include its binding properties and actions.

As the GR pathway is also used in the phosphorylation and subsequent activation of the guardian of the genome p53, the phosphorylation of p53 increases its effectiveness in activating its downstream pathways via multiple mechanisms [5]. The phosphorylation of p53 decreases the binding affinity to the ubiquitin ligase MDM2 as well, preventing p53 from being marked for degradation. Due to its inhibitory activity on antiproliferative pathways, the overexpression of PP5 promotes cell proliferation. These results indicate that PP5 downregulates cellular functions essential for antitumor activity, supporting the targeting of PP5 in cancer research and potential therapy.

2. Background

Cantharidin is a naturally occurring inhibitor found in blister beetles and has been used in Chinese medicine for centuries for a wide range of ailments [6]. The fatty, odorless, naturally occurring toxin has been found to be a semi-selective protein phosphatase inhibitor (PP5C IC50 = 0.2 ± 0.03 μM). As the synthetic assembly of the bicyclic core can occur in one step and involves the cycloaddition of a diene and dienophile, resulting in the exo adduct (vide infra), this system is an ideal template for the assembly and testing of next generation derivatives, having goals of both potency and selectivity (Scheme 1).

Co-crystal structures of PP5C and derivatives of cantharidin’s demethylated cousin norcantharidin have been published (Figure 1) [7,8]. Close inspection of the catalytic pocket when docked with a norcantharidin derivative revealed the necessity of close proximity between the oxygen bridgehead and the carboxylate oxygens of the inhibitor with manganese catalytic ions of PP5. These findings offered a key structural and retrosynthetic perspective in that any synthetic approach involving the assembly of next generation inhibitors specific to PP5C must advance the exo isomer of the core cantharidin/norcantharidin scaffold.

Complementing the data obtained from the co-crystals of PP5C and the derivatives of norcantharidin were IC50 data against serine/threonine protein phosphatases [3]. Within the serine/threonine family of protein phosphatases, cantharidin and derivatives of norcantharidin have shown particular activity within the domains of PP1CA, PP2CA, PP5C, and PP6C [7]. Notable, however, was the activity (or lack thereof) found with protein phosphatase 5 and derivatives of norcantharidin, having substituents bound at C-5 of the bicyclic scaffold (Chart 1).

As shown above in Chart 1, potency was documented with both norcantharidin (PP5 IC50 1.0 μM) and the propoxymethyl derivative of norcantharidin (PP5 IC50 0.7 μM). For example, where the C-5 attachment extended beyond five atoms, potency dropped as evident with the assembly and testing of the decoxymethyl derivative of norcantharidin (PP5 IC50 14.3 μM). This trend was consistent with prior work, which examined shorter, longer, and branched alkyl derivatives, covalently linked around the bicyclic core.

Additionally, prior work has revealed a higher level of selectivity when comparing the IC50 of PP1 of the propoxymethyl derivative of norcantharidin (PP1 IC50 3.7 μM) to the parent compound norcantharidin (PP1 IC50 8.9 μM). Even though these levels of inhibition do not meet the threshold of a viable candidate for potential therapy, the trendline offers clear evidence of our hypothesis that decorated scaffolds of the bicyclic core, offering an opportunity to assemble inhibitors exhibiting both potency and selectivity.

The rationale, based on computational work and the co-crystal structures, highlights favorable interactions through supplementary hydrogen bonding with neighboring amino acid residues while avoiding steric clashes within the catalytic site.

The exploitation of neighboring grooves of a catalytic site is not new and is arguably best-illustrated with fostriecin and PP2A. Fostriecin (Figure 2) is a phosphate monoester that acts as a potent and selective inhibitor of protein phosphatase 2A [9]. Fostriecin inhibits PP2A through covalent bonding at a nucleophilic cysteine residue within its active catalytic grooves. The lactone moiety (cyclic ester) is believed to be responsible for the covalent interaction between the inhibitor and active site. When circling back to cancer research and potential therapy, fostriecin entered phase 1 of clinical trials based upon, in part, the antitumor activity in mouse leukemia cells, which were founded on evidence of protein phosphatase inhibition negatively affecting tumor proliferation. Although this does not directly relate to PP5, this novel system serves as the gold standard for selective and potent protein phosphatase inhibition, as well as a model for future inhibitor development.

3. Results and Discussion

As the bicyclic core, upon its assembly, via a [4 + 2] cycloaddition using derivatives of furan and maleic anhydride, installs all the salient features found to be essential as it relates to the key interactions discussed above [10], the opportunity exists to develop an efficient synthetic scheme that explores and then assays synthetically to decorate derivatives. With the systems we have prepared and tested to date, our focus has been potency and selectively.

Now that we have prepared and tested the propoxymethyl derivative, which has not diminished potency relative to the baseline values of norcantharidin, our focus has shifted. Our goal now is the rapid and efficient assembly of functionalized furan derivatives as function follows structure [11]. The key is not on the actual synthesis but the development of a rapid and efficient methodology having the assembly of derivatives of norcantharidin so that we can probe both potency and selectivity. Although furan derivatives are at the center of our study, it should be noted that the reason we are not exploring derivatives of maleic anhydride is that carboxylic acid derivatives of the anhydride and modifications to the alkene of the dienophile have been prepared and have shown insignificant gains.

Figure 3 illustrates our current approach, which we feel addresses the metrics of a rapid and efficient assembly of next generation inhibitors of PP5. In one step, functionalized furan derivatives in high yield are obtained upon condensation of isocyanate functionality with a primary alcohol to generate a carbamate. This, combined with the Diels–Alder cycloaddition followed by hydrogenation, yields the target compound in three synthetic steps.

We have been successful in generating carbamate-functionalized furans with either allyl alcohol or furoic acid via the Curtius rearrangement using diphenylphosphoryl azide (DPPA) or hydroxymethylfuran (both at positions 2 and 3) with allyl isocyanate. When working with 2-hydroxymethylfuran and allyl isocyanate (Scheme 2), the isolated yields of 1 after chromatography and bulb-to-bulb distillation ranged from 67 to 73% yield (2.4 mmol scale). Access to all the spectra acquired can be found within the Supplementary Materials. What is important to note with this high yielding transformation is the inclusion of heteroatomic functionality within the alkyl chain and use of the alkene functionality at the terminus. Although the heteroatom within the alkyl chain has been shown to computationally offer additional points of contact within the groove of the active site, the installment of alkene functionality is new and allows one to explore further derivatization at multiple points of this synthetic scheme, hence, proof of principle in the assembly of furan derivatives bearing carbamate functionality. Work toward the cycloaddition and testing of these next generation derivatives at C-2 and C-3 using heteroatomic and functionalized alkyl chains is ongoing and will be reported in due course.

4. Materials and Methods

The NMR that generated all the spectra as part of this submission was a JEOL ECA-500 spectrometer (JEOL Ltd., Tokyo, Japan). The software used to process all the corresponding data was JEOL Delta^TM Version 6.1.0 (MAC). ¹H NMR (500 MHz) and ¹³C NMR (125 MHz) spectra were obtained as solutions in CDCl₃. Chemical shifts were reported in parts per million (ppm) and referenced to δ 7.27 (¹H NMR) and δ 77.00 (¹³C NMR). Infrared spectra were recorded using a JASCO FT/IR-4100 (JASCO, Tokyo, Japan) and were reported in wavenumbers (cm⁻¹). For the synthetic procedure performed, additional considerations consisted of the following: TLC analyses were performed on flexible aluminum backed TLC plates with a fluorescent indicator. Detection was conducted by UV absorption (254 nm) followed by charring with 10% KMnO₄ in water. Solutions were concentrated in vacuo using a rotary evaporator, and the residue was purified by column chromatography followed by bulb-to-bulb distillation. Crude reaction mixtures were purified using a silica gel column (70–230 mesh, 60 Å). Bulb-to-bulb distillations were performed under vacuum using an Aldrich Kugelrohr short-path distillation apparatus model Z24, 860-6 with air bath oven. Gas chromatography and mass spectroscopy were performed on a Shimadzu GC-17A/GCMS-QP5000 (Kyoto, Japan) (GC: 100 °C (2 min), 10 °C/min, 280 °C (5 min); MS: 70 eV, LRMS). The HRMS data were generated at the University of South Alabama Mass Spectrometry Core Facility. The chemicals used for the synthetic procedure (allyl isocyanate, ethyl acetate, hexanes, 2-hydroxymethylfuran, triethylamine, and toluene) were reagent grade or better.

5. Experimental

2-Furanylmethyl N-(2-propenyl)carbamate (1)

The 2-hydroxymethylfuran (furfuryl alcohol) (0.21 mL, 2.4 mmol, 1 equiv.) was added to a 50 mL round-bottomed flask (RBF) equipped with a magnetic stir bar and water-jacketed condenser. Furfuryl alcohol was then dissolved in toluene (10 mL). After adding one drop of triethylamine, allyl isocyanate (0.20 g, 2.4 mmol, 1 equiv.) was added to the reaction mixture by syringe at room temperature. After placing the setup under a blanket of argon, the reaction was allowed to warm to an external temperature of 65 °C and stir overnight. Upon cooling to room temperature, the reaction mixture was concentrated in vacuo, chromatographed, and distilled (bulb-to-bulb). The column chromatography (SiO₂) gradient system started with hexanes (one column equivalent) then increased in polarity using the following EtOAc/hexanes mixture, all representing three column equivalents: hexanes, 1:32, 1:16, 1:8, 1:4, 1:1. The residue was then distilled (bulb-to-bulb (133–135 °C (2 Torr))), resulting in isolated yields of a clear and colorless oil, ranging between 67–73%. Mixed fractions caused the discrepancy in yield, and no changes in outcome were observed if the order of addition was reversed (allyl isocyanate first and furfuryl alcohol last) or the reaction was allowed to react longer. Access to all the spectra acquired can be found within the Supplementary Materials. TLC; SiO₂, 1:1 EtOAc:hexanes, R_f = 0.79. ¹H NMR (500 MHz, CDCl₃); δ 7.42 (d, 1H, CH (furan), J = 1.5 Hz), 6.41 (d, 1H, CH (furan), J = 3.0 Hz), 6.36 (q, 1H, CH (furan), J = 1.5 Hz), 5.88–5.79 (m, 1H, CH (alkene)), 5.19 (d, 1H, CH (alkene), J = 17.0 Hz), 5.12 (d, 1H, CH (alkene), J = 10.5 Hz), 5.07 (s, 2H, OCH₂, 4.80 (br s, NH), 3.82 (t, 2H, CH₂ (allylic), J = 5.5 Hz). ¹³C NMR (CDCl₃); δ 155.9, 150.0, 143.2, 134.3, 116.2, 110.5, 110.4, 58.6, 43.5. IR (neat); 3334, 3148, 3122, 3083, 3013, 2986, 2944, 1701, 1524, 1246 cm⁻¹. GCMS (70 eV, LREI); t_R = 9.0 min, m/z 181 (M⁺). HRMS (LC-MS/MS) m/z: [M + H]⁺ calcd for C₉H₁₁NO₃ 182.0817; found 182.0814.

6. Conclusions

As molecular structure impacts function, our focus on the development of potent and selective inhibitors of PP5 has shifted toward the assembly of functionalized derivatives of furan. Once prepared, cycloaddition with maleic anhydride followed by hydrogenation yields a decorated norcantharidin derivative in three synthetic steps. Our approach involves the use of a terminal alkene tethered to carbamate functionality, which will allow for the advancement and testing of the next generation inhibitors of PP5C. The key is the rapid access to the target modalities with high efficiency.