Studies on the alkylation of phenolate in an organofluorine solvent and its application to the synthesis of myrsinoic acids A and E

Article abstract of DOI:10.1080/00397911.2017.1288823

Among the distinct features of organofluorines, solubility properties to molecules often enable unique applications. More importantly, organofluorines can be compatible with chemical reactions in general, even under harsh conditions. In this context, a study on the alkylation of phenolate using an organofluorine solvent were performed for the preparation of ortho-alkyl phenols. Herewith, the study on the alkylation of phenolate in an organofluorine solvent as well as its application to the synthesis of bioactive terpenylated phenolic natural products, myrsinoic acids A and E, will be reported. Our study suggested that organofluorine solvents could be better alternatives to the conventional nondissociating solvents as organic reaction media to suppress the nondesirable and nonselective side reactions such as in the ortho-alkylation of phenolate.

Full text of DOI:10.1080/00397911.2017.1288823

Studies on the Alkylation of Phenolate in an Organofluorine Solvent and Its
Application to the Synthesis of Myrsinoic Acid A and E
Miseon Ryu1, Minjun Kim1, Minseon Jeong1, Jaebong Jang2,3, Minyoung Lee1, Hyo-Eon
Jin4, Jong-Wha Jung1
1College of Pharmacy, Research Institute of Pharmaceutical Sciences, Kyungpook
National University, Daegu, Korea, 2Department of Cancer Biology, Dana-Farber Cancer
Institute, Boston, MA, USA, 3Department of Biological Chemistry and Molecular
Pharmacology, Harvard Medical School, Boston, MA, USA, 4College of Pharmacy, Ajou
University, Suwon, Korea
Correspondence to: J.-W. Jung (jungj@knu.ac.kr), H.-E. Jin (hjin@ajou.ac.kr).
†These authors equally contributed to this work.
Abstract
Among the distinct features of organofluorines, solubility properties to molecules often
enables unique applications. More importantly, organofluorines can be compatible with
chemical reactions in general, even in harsh conditions. In this context, a study on the
alkylation of phenolate using an organofluorine solvent were carried out for the
preparation of ortho-alkyl phenols. Herewith, the study on the alkylation of phenolate in
an organofluorine solvent as well as its application to the synthesis of bioactive
terpenylated phenolic natural products, myrsinoic acid A and E, will be reported. Our
study suggested that organofluorine solvents could be better alternatives to the
conventional non-dissociating solvents as organic reaction media to suppress the non-
desirable and non-selective side reactions such as in the ortho-alkylation of phenolate.
1

GRAPHICALABSTRACT
KEYWORDS: Organofluorine solvent; alkylation of phenolate; terpenylated phenol;
myrsinoic acid
INTRODUCTION
C-prenylated phenols are found in diverse natural sources, especially in plant species.
Bioactive C-prenylated phenols such as benzoic, cinnamic and coumaric acid derivatives
as well as prenylated ubiquinones/naphthaquinones, flavonoids and xanthones have
attracted considerable scientific attentions due to its unique biological and
pharmacological properties which are intimately associated with the presence of phenolic
hydroxyl groups1-4. Nevertheless, no single synthetic method encompassed the
preparation of C-prenylated phenols, and some synthesis of prenylated phenols are
reported to be challenging5-6. Various synthetic methods has been developed so far for the
preparation of ortho-alkyl phenols: Lewis acid catalyzed ortho-alkylation of phenol
including the Friedel-Crafts type, Claisen rearrangement, metal-mediated coupling via
metal-halogen exchange or directed ortho-metalation are popular strategies among them5-
6. In particular, the ortho-alkylation of phenolate is a favored strategy for the preparation
of terpenylated phenols especially for the prenylated. Although ortho-alkylation of
2

phenolate is intuitive and tempting, due to the advantages in terms of atom- and step-
economy, it is often challenging because of the strong preference of competing O-
alkylation and non-selective side reactions such as regioisomeric or multiple alkylation
depending on the structure. Non-dissociating organic solvents have been utilized to
facilitate selective alkylation for the preparation of ortho-alkyl phenols7-9, and in this
context, we became interested in using an organofluorine solvent as a novel non-
dissociating solvent during our synthetic program for the preparation of bioactive natural
products.
Myrsinoic acid A (MAA, 1), a prenylated phenol, was isolated in 1991 from the leaves of
a Brazilian shrub, Rapanea umbellata (Figure 1). The structure of MAA has been
identified as a terpeno-p-hydroxybenzoic acid on the basis of spectroscopic analysis of its
methylated derivative10. The structure of MAA was later confirmed by an extensive NMR
study during a bioassay-guided isolation of anti-inflammatory compounds from the
methanolic extract of Myrsine seguinii11. MAA was also reported to exhibit antibacterial
activities12. In addition, the anticancer effect of MAA through a novel mechanism of
action, i.e. the inhibition of heparin-binding epidermal growth factor-like growth factor,
attracted attentions13-14. Among the structural features of MAA, the carboxylate
functionality was found to be of importance for the activity of MAA, and structurally
related terpeno-benzoic acids, myrsinoic acid B, C, E (MAE, 2), and F were also isolated
to possess similar activities15 (Figure 1). All those biological properties of myrsinoic
acids made it clear that an efficient and scalable method for the synthesis is urgently in
need. In particular, the synthesis of MAA has never been reported yet. Herewith, we
3

would like to report studies on the alkylation of phenolate in an organofluorine solvent
and its application to the synthesis of MAA and MAE via a sequential alkylation strategy.
RESULTS AND DISCUSSION
We envisioned that a straightforward synthesis of MAA and MAE could be achieved via
sequential ortho-alkylation or bis-alkylation of metal phenolate, respectively (Figure 2).
Unlike conventional alkylation of enolates, however, C-alkylation of phenolates are often
challenging. The preference of phenolate toward O-alkylation over C-alkylation, which is
less likely for the conventional alkylation of enolates, can be understood by the disruption
of aromaticity during the C-alkylation of phenolates. In spite of that, the composition of
the products is known to be determined by the relative nucleophilicity of each atom in
relation to the electrophile in a kinetically controlled condition16, and attempts to achieve
an efficient ortho-alkylation has been continued. Several pioneers successfully provided
conditions by exploring counterion, additives, concentration, temperature, pressure,
leaving group, and the structure of substrate or electrophile7-8. For the countercation of
phenolates, for example, [Na+] is known to favor C-alkylation whereas [K+] and [Li+]
favor O-alkylation and bis-alkylation7. Among the factors influencing the alkylation of
phenolate, interestingly, the regioselectivity is strongly dependent on the solvent.
Considering that the cation is coordinated to the atom with the maximal electron density,
the preferential C-alkylation over O-alkylation is expected in a poorly cation-solvating
solvent. In fact, toluene and benzene are representative solvents of choice when it comes
to C-alkylation of phenolate.9 Thus, our study commenced with the screening of reaction
4

conditions for the ortho-alkylation of metal phenolate to provide methyl 4-hydroxy-3-(3’-
methyl-2’-butenyl)benzoate, especially focused on the potential non-dissociating solvents.
Looking for a better non-dissociating solvent alternative to the reported such as diethyl
ether or toluene, we came up with the idea of using organofluorines. Among the distinct
features of organofluorines, solubility properties to molecules often enables unique
applications: they are found in various applications including anesthetics,
pharmaceuticals, agrichemicals, fluoropolymers, refrigerants, oil and water repellents,
surfactants, reagents in catalysis, and many more. In addition, organofluorines can be
compatible with chemical reactions in general, even in harsh conditions. It is noteworthy
that the utilization of organofluorines as reaction media has been scarcely reported.
As shown in Table 1, the examined reactions in diethyl ether, toluene and tetrahydrofuran
seemed far less than ideal with commercially available sodium phenolate 3’.(Entry 1-3)
Although there has been several successful reports for the C-prenylation of phenolates6,
only a small amount of the desired product 4 was isolated in both diethyl ether and
toluene, in spite of the attempt to optimize conditions. (Entry 1-4). It is noteworthy that
electron-withdrawing groups in a phenolate substrate is known be detrimental for the C-
alkylation of phenolate as it lowers electron density17. Using tetrahydrofuran as the
reaction solvent afforded O-prenylated product 5 in a relatively high yield, and which is
in consistent with the reported7 (Entry 5). The preference of O-prenylation in
tetrahydrofuran might be attributed to the strong basicity of tetrahydrofuran to make
stronger coordination with the counter-cation. Thus, we decided to explore the reaction in
5

organofluorines for the prenylation of phenolate. Commercially available, and less
expensive organofluorines, i.e. trifluorotoluene, hexafluorobenzene, and trifluoroethanol
were selected as non-dissociating organofluorine solvents. To our delight, the reactions in
trifluorotoluene afforded the desired product 4 in higher yield than the reactions in diethyl
ether or toluene (Entry 6). An elevated reaction temperature of 40 °C was proven to be
beneficial (Entry 7), while even higher temperature was not (Entry 8). It was also
interesting to find bis-prenylated phenol 6 as a major product at 80 °C. Surprisingly, the
reactions in hexafluorobenzene, a perfluoroaromatic compound, produced O-prenylated
product 5 in high yield (Entry 9, 10). The reaction in trifluorobenzene provided C-
prenylated product 4 as a major product (Entry 11), while the reaction in
perfluoro(methylcyclohexane) also provided O-prenylated product 5 as a major product
(Entry 12). Interestingly, the reactions in trifluoroethanol, a polar organofluorine,
produced bis-prenylated product 6 as a major (Entry 13, 14).
The optimized reaction condition in hands, a sequential protocol of prenylation and
geranylation was successfully utilized for the synthesis of MAA (Scheme 1). The
prenylated phenol 4, prepared from sodium phenolate 3’ as above, was metallated with
sodium hydride, and then reacted with geranyl bromide in trifluorotoluene to afford the
desired methyl ester 9 in high yield. The yield of C-alkylation reaction of phenolate
increases as the length of electrophilic isoprene units increases in organic solvents7-9, and
which seems to be in consistent with the alkylation in an organofluorine solvent. The
methyl ester 9 was shortly hydrolyzed to complete the synthesis of MAA. Next, the
synthesis of MAE could also be achieved with a slight modification of our protocol.
6

Sodium phenolate 3’ was reacted with a slight excess amount of geranyl bromide for the
bis-geranylation, and which successfully provided the methyl ester 10 along with a small
amount of O-geranylated byproduct. The methyl ester 10 was hydrolyzed to complete the
synthesis of MAE. The synthesized MAA and MAE were identical to the reported, based
on the comparison of spectral data (See, Supplementary Information.).
CONCLUSION
In conclusion, a study on the alkylation of phenolate using an organofluorine solvent
were carried out. C-Alkylation reaction of sodium phenolate in trifluorotoluene was
found to be superior to the reaction in conventional non-dissociating solvents including
diethyl ether and toluene. This study demonstrated organofluorine solvents can provide
non-dissociating environment to facilitate the C-alkylation of phenolate while
suppressing the non-desirable O-alkylation. In addition, the developed protocol using
trifluorotoluene was also successfully applied to the synthesis of bioactive terpenylated
phenolic natural products, MAA and MAE, via sequential alkylation and bis-alkylation
respectively. Considering that MAA and MAE are promising bioactive natural products
as well as potential precursors for the preparation of other myrsinoic acids, our protocol
would expedite further studies related in the biological properties of myrsinoic acids and
their derivatives. We also believe our protocol would provide an intuitive and efficient
synthetic route for the preparation of other terpenylated phenols which display a wide
range of significant biological activities.
EXPERIMENTAL
7

Typical Procedure For The Mono-C-Alkylation
Preparation Of Methyl 4-Hydroxy-3-(3’-Methylbut-2’-En-1’-Yl)Benzoate (4)
Sodium phenolate 3’ (174 mg, 1 mmol) was suspended in trifluorotoluene (4 mL), and 1-
bromo-3-methyl-2-butene (179 mg, 1.2 mmol) was added dropwise to the suspension.
The suspension was stirred for a 24 h at a given temperature (40 °C). The resulting
reaction was then quenched with aqueous NH4Cl solution, and extracted with EtOAc.
The organic layer was dried over anhydrous MgSO4, and volatiles were removed under
the reduced pressure. The residue was purified by SiO2 chromatography. Yield 51% (112
mg); Rf 0.3 (EtOAc : n-hexane = 1 : 5); White wax; 1H-NMR (500 MHz, CDCl3) 7.82
(2H, m), 6.82 (1H, d, J = 8.9 Hz), 5.63 (1H, bs), 5.31 (1H, t, J = 7.3 Hz), 3.88 (3H, s),
3.39 (2H, d, J = 7.2 Hz), 1.79 (6H, s); 13C-NMR (125 MHz, CDCl3) 167.7, 158.9, 134.8,
131.8, 129.6, 127.3, 122.1, 121.2, 115.4, 52.0, 29.2, 25.8, 17.9; IR (ATR-IR) max 3254,
1676, 1596 cm-1; HRMS (EI) m/z [M]+ calc. for C13H16O3 220.1099, found 220.1100.
Typical Procedure For The Di-C-Alkylation
Preparation Of Methyl 3,5-Bis((E)-3,7-Dimethylocta-2,6-Dien-1-Yl)-4-
Hydroxybenzoate (10)
Sodium phenolate 3’ (174 mg, 1 mmol) was suspended in trifluorotoluene (4 mL), and
geranyl bromide (0.476 ml, 2.4 mmol) was added dropwise to the suspension. The
suspension was stirred for a 24 h at rt, and the reaction was then quenched with aqueous
NH4Cl solution. The mixture was extracted with EtOAc. The organic layer was dried
over anhydrous MgSO4, and volatiles were removed under the reduced pressure. The
residue was purified by HPLC. Yield 45% (191 mg); Rf 0.7 (CH2Cl2); Colorless oil; 1H-
8

NMR (500 MHz, CDCl3) 7.70 (2H, s), 5.85 (1H, s), 5.31 (2H, m), 5.08 (2H, m), 3.86
(3H, s), 3.38 (4H, d, J=7.2), 2.09 (8H, m), 1.76 (3H, s), 1.68 (3H, s), 1.60 (3H, s); 13C-
NMR (125 MHz, CDCl3) 166.9, 162.7, 141.8, 131.9, 131.6, 123.7, 122.5, 118.9, 114.3,
65.1, 51.8, 39.5, 26.3, 25.7, 17.7, 16.7; IR (ATR-IR) max 3408, 2921, 2858, 2347,1562,
1413 cm-1; HRMS (EI) m/z [M]+ calc. for C28H40O3 424.2977, found 424.2981.
ACKNOWLEDGEMENT
This work was supported by a grant of the Korea Healthcare Technology R&D Project
through the Korea Health Industry Development Institute (KHIDI), funded by the
Ministry of Health & Welfare, Republic of Korea (HI14C0209) and a grant from the
Research Programs (NRF-2015M3A9E7029176) of the National Research Foundation of
Korea (NRF) funded by the Ministry of Science, ICT, and Future Planning.
SUPPORTING INFORMATION
Full experimental detail including 1H and 13C NMR spectra for this article can be found
via “Supplementary Content” section of this article’s webpage.
REFERENCES
[1] Cheng, W.; Li, W. Science 2014, 343 (6173), 878-881.
[2] Kuzuyama, T.; Noel, J. P.; Richard, S. B. Nature 2005, 435 (7044), 983-987.
[3] Botta, B.; Vitali, A.; Menendez, P.; Misiti, D.; Monache, G. D. 2005, 12 (6), 713-739.
[4] Yazaki, K.; Sasaki, K.; Tsurumaru, Y. Phytochemistry 2009, 70 (15), 1739-1745.
[5] Hoarau, C.; Pettus, T. R. Synlett 2003, 1, 127-137.
9

[6] Talhi, O.; Silva, A. Curr. Org. Chem. 2013, 17 (10), 1067-1102.
[7] Le Noble, W. J. Synthesis 1970, 1, 1-6.
[8] Yamada, S.; Ono, F.; Katagiri, T.; Tanaka, J. Nippon Kagaku Kaish 1980, 5, 733-737.
[9] Pereda-Miranda, R.; Bernard, C. B.; Durst, T.; Arnason, J. T.; Sánchez-Vindas, P.;
Poveda, L.; San Román, L. J. Nat. Prod. 1997, 60 (3), 282-284.
[10] Januário, A. H.; Vieira, P. C.; da Silva, M. F. d. G. F.; Fernandes, J. B.
Phytochemistry 1991, 30 (6), 2019-2023.
[11] Dong, M.; Nagaoka, M.; Miyazaki, S.; Iriye, R.; Hirota, M. Biosci. Biotechnol.
Biochem. 1999, 63 (9), 1650-1653.
[12] Cruz, A. B.; Kazmierczak, K.; Gazoni, V. F.; Monteiro, E. R.; Fronza, L. M.; Martins,
P.; Yunes, R. A.; Bürger, C.; Tomio, T. A.; Freitas, R. A. J. Med. Plants Res. 2013, 7 (19),
1323-1329.
[13] Lee, S. W.; Mandinova, A. U.S. Pat. 2012, 8,232,318.
[14] Miyamoto, S.; Yagi, H.; Yotsumoto, F.; Kawarabayashi, T.; Mekada, E. Cancer Sci.
2006, 97 (5), 341-347.
[15] Hirota, M.; Miyazaki, S.; Minakuchi, T.; Takagi, T.; Shibata, H. Biosci. Biotechnol.
Biochem. 2002, 66 (3), 655-659.
[16] Reichardt, C.; Welton, T. Solvents and solvent effects in organic chemistry; John
Wiley & Sons: 2011.
[17] Bigi, F.; Casiraghi, G.; Casnati, G.; Sartori, G. Synthesis 1981, 4, 310-312.
10

Table 1. Comparison of alkylation in various organofluorine solventsa
Entry Solvent
Temp. Time Yieldb
3c
4
5
6
7
8
5
8
3
3
1
Diethyl ether
rt
rt
24h 36 26 <1 6
24h 34 19 <1 4
27
35
2
Toluene
3
Toluene
40 oC 24h 37 16 <1 5
80 oC 24h 34 18 <1 2
39
43
4
Toluene
5
Tetrahydrofuran
Trifluorotoluene
Trifluorotoluene
Trifluorotoluene
Hexafluorobenzene
Hexafluorobenzene
Trifluorobenzene
rt
rt
24h <1 61 <1 4 <1 35
6
24h 43 18 <1 7
4
5
28
22
7
40 oC 24h 51 18 <1 4
8d
9
80 oC 24h <1 3 39 2 <1 56
rt
40 oC 24h
40 oC 24h 29 9 11
24h <1 70 <1 4 <1 26
10
11
12
13
14
3
63 <1 7
1
26
40
43
69
51
3
8
Perfluoro(methylcyclohexane) 40 oC 24h
6
2
41 2
4
4
3
6
Trifluoroethanol
Trifluoroethanol
rt
24h
1 22 3
40 oC 24h <1 5 33 4
aRefer to the Experimental Section. Briefly, sodium salt of methyl 4-hydroxybenzoate (1
eq.) and 1-bromo-3-methyl-2-butene (1.2 eq.) were subjected to the given reaction
condition unless noted otherwise; bDetermined by isolated yields or HPLC analysis;
cMethyl 4-hydroxybenzoate; d2.4 eq. of 1-bromo-3-methyl-2-butene was used.
11

Scheme 1. Synthetic strategy for the synthesis of myrsinoic acid A and E
12

Scheme 2. Synthesis of myrsinoic acid A and E
13

Figure 1. The structures of natural myrsinoic acids
14