Regioselectivity Amination of Usnic Acid by Ammonia in Water

Abstract

Usnic acid is a well-known secondary lichen metabolite exhibiting a broad spectrum of biological activity. Previously it was shown that the reaction of usnic acid with various amines resulted in enamine-bond formation instead of the C(11)=O carbonyl group. Enamines obtained have a pronounced biological activity. In this work, we have shown that the reaction of usnic acid with ammonia can be regioselective if the solvent is replaced by water. The regioselectivity of that reaction depends on temperature and ammonia quantity. The C-1 enamine as only product formation has been obtained by the usnic acid reaction with an excess of ammonia (20 eq.) in water with cooling (+9 °C).

1. Introduction

Usnic acid 1 (Scheme 1), a well-known secondary lichen metabolite, shows a wide spectrum of biological activity: antibacterial, analgesic, immunomodulatory, antitumor, antimalarial, anti-tuberculosis, photoprotective, etc. [1,2,3,4,5]. The reaction of usnic acid with primary amines, as a rule, proceeds via attack to the carbonyl group C(11)=O, accompanied by the double bonds rearrangement and leading to compounds of the enamine type [6]. Often such compounds have a pronounced biological activity. For example, enamines obtained with different amines leading to functionalized enamines exhibit hypoglycemic, antimicrobial, anti-tuberculosis, antiproliferative, anti-inflammatory, and antimalarial properties [7,8,9,10,11].

Previously, it was shown that the reaction of (+)-usnic acid with an excess of aqueous ammonia in a benzene–ethanol mixture with reflux leads to C(11)-enamine (compound 2, Scheme 1) as the only reaction product with a yield of 82% [12]. The structure of compound 2 was confirmed by full structural characterization, including X-ray diffraction analysis [12,13]. Different substituted enamines can usually be obtained by the usnic acid reaction with primary amines in boiling ethanol [7,8,9,10,11].

2. Results and Discussion

We verified that the reaction of (+)-usnic acid with aqueous ammonia (12.5 or 50 eq.) in methanol, ethanol, DMSO, DMF, acetonitrile, or benzene led to the formation of only the ordinary reaction product C(11)-enamine, compound 2.

However, it was observed that using water as solvent led to the formation, in addition to the previously described product 2, of a new compound 3 as a minor product that contains an enamine group according to atom C(1) (Scheme 2).

Additionally, it was noticed that in the reaction with an increase of the ammonia amount, an increase of the C(1)-enamine amount occurs. Reactions were carried out with aqueous ammonia with reflux in water (

Table 1. The enamine products ratio (according to NMR data) in the mixture, depending on the ammonia amount.

Ammonia Amount	Compound 2	Compound 3
12.5 eq.	72	28
62.5 eq.	53	47
125 eq.	34	66

). The ratio of products was calculated using ¹H NMR spectra (Figures S12–S14). These signals of H-4 (5.79 for compound 2 and 5.75 for compound 3) were integrated separately.

An increase in the amount of C(1)-enamine 3 with an increase in the amount of ammonia may be associated with the kinetic features of the reaction. Probably, the order of the reaction that resulted in C(1)-enamine 3 is higher than the order of reaction that led to C(11)-enamine 2.

The reaction products can be easily separated by stepwise precipitation. When the mixture was acidified to a pH = 7, a yellow precipitate of C(11)-enamine was formed, which was filtered out. Then, with subsequent acidification to a pH < 5, a gray-green precipitate of C(1)-enamine was formed.

It was noticed that the ratio of the product depends on the temperature (

Table 2. The enamine products ratio (according to NMR data) in the mixture, depending on the reaction temperature.

Temperature *	Compound 2	Compound 3
9 °C	trace	100
23 °C	23	77
100 °C	34	66

* Temperature was measured in water or sand bath.

). During reactions in aqueous ammonia (125 eq.), the amount of C(1)-enamine increased when the temperature decreased. However, when carried out in other solvents (methanol, ethanol, DMSO, DMF, acetonitrile, or benzene), reactions with different ammonia amounts (12.5, 62.5, 125 or gaseous ammonia) resulted in the formation of C(11)-enamine as the only product independently on temperature. An increase in the quantity of C(11)-enamine with an increase in temperature may be due to its greater thermodynamic stability compared to compound 3.

Thereby, the reaction was carried out by stirring (+)-usnic acid in aqueous ammonia with cooling (+9 °C), and it led to the only product—C(1)-enamine 3. After completion of the reaction (TLC control), the mixture was neutralized with dilute hydrochloric acid (1M aqueous solution). The precipitated product was filtered off and dried in air. Compound 3 was obtained with a 90% yield.

The structure of new C(1)-enamine 3 was confirmed using NMR ¹H and ¹³C (see Supplementary Materials, Figures S5, S6, S11 and S12). In the NMR spectra, the signals of the atoms of the A ring in enamines 3 and previously described compound 2 practically coincide, but the signals of the atoms in the C ring differ markedly (

Table 3. NMR ¹H and ¹³C of enamines 2 and 3.

Comparison of PMR			Comparison of 13C NMR
Atom *	Signals of 2	Signals of 3	Atom *	Signals of 2	Signals of 3
H-4	5.79	5.75	C-1	197.55	174.52
CH3-10	1.93	2.02	C-2	100.79	103.67
CH3-12	2.51	2.42	C-3	188.46	184.32
CH3-14	2.59	2.63	C-4	102.48	102.79
CH3-15	1.59	1.72	C-4a	172.91	171.18
OH-7	13.37	13.51	C-9a	105.05	107.11
OH-9	12.26	—	C-9b	56.02	51.13
NH2-1	—	9.41 and 11.35	C-11	175.80	198.75
NH2-11	9.82 and 11.53	—	C-12	24.47	32.54
			C-15	30.94	33.80

* The atom numbers in these compounds are different from the numeration in the nomenclature name.

). The ¹H NMR spectra provide differences for the methyl group (signals on 1.6–2.6 ppm). The signals of the C-15 methyl group differ most noticeably. In the area of -OH and -NH signals (9.5–13.5 ppm), four signals are observed for compound 2, while three signals are for derivative 3. The signal of the OH-9 group probably disappears for enamine 3. The ¹H NMR spectrum of product 3 clearly indicates the formation of a new product. In ¹³C NMR spectra, the change in the signal of the C-12 methyl group from 24.47 to 32.54 ppm is significantly noticeable, which may indicate that a carbonyl group is located at the C-11 atom. The signal of the C-4 atom remains in place (102 ppm), which may indicate that the C-3 carbonyl group is unchanged. A change in the C-9b signal from 56.02 to 51.13 ppm may indicate the transformation of C-1 carbonyl into the enamine group. Therefore, we proposed structure 3.

The data on melting points, specific rotation, and ultraviolet and infrared spectra of compounds 2 and 3 are presented in

Table 4. Melting point, specific rotation, UV-data, and IR data of compounds 2 and 3.

	Mp (°C)	[α]D24.6 * (deg)	Ultraviolet Data * λmax (nm) and log10ε (in Parentless)	IR (cm−1)
2	271–273	+638	227 (4.40), 293 (4.51)	2500–3400, 1697, 1626, 1549
3	269–271	+599	215 (4.33), 226 (4.33), 284 (4.43), 323 (4.08)	2500–3600, 1700, 1624, 1579

* The ethanol used as solvent.

. The melting point of products 2 and 3 is practically identical. The specific rotation of compound 3 is less than that of derivative 2. Ultraviolet spectra of these enamines are similar. However, there is a new absorption maximum at 215 nm and a more distinct signal at 323 nm for compound 3 (see Figures S5 and S10). The infrared spectra provide some differences. Absorption of -OH and -NH groups changes from 2500 to 3400 for enamine 2 up to 2500–3600 for compound 3 (Figures S3 and S8). The carbonyls and enamines absorptions remain the same.

The structure of compound 3 was confirmed by an X-ray diffraction analysis (Figure 1). According to the XRD data, the compound is crystallized as hydrate (1:1) with two independent molecules of compound 3.

The molecular geometry of compound 3 is close to the analogous of isoxazole derivative 4 (Figure 1) [14]. In a crystal, molecules of compound 3, with the participation of water molecules, are linked by OH…O (H…O 1.69(4) ÷ 2.05(7) Å) intermolecular hydrogen bonds into ribbons along the a axis.

3. Materials and Methods

The analytical and spectral studies were conducted at the Chemical Service Center for the collective use of the Siberian Branch of the Russian Academy of Science.

The ¹H and ¹³C-NMR spectra for solutions of the compound in DMSO-d₆ were recorded on a Bruker AV-400 spectrometer (Bruker Corporation, Hanau, Germany; operating frequencies 400.13 MHz for ¹H and 100.61 for ¹³C). The residual signals of the solvent were used as references (δ_H 2.50, δ_C 39.52). The mass spectra (ionizing-electron energy 70 eV) were recorded on a DFS Thermo Scientific high-resolution mass spectrometer (Thermo Fisher Scientific, Waltham, MA, USA). Electron impact ionization was used in the measurement of mass spectra. The UV and IR spectra were measured on a Spectrophotometer Cary-5000 (Varian, Palo Alto, CA, USA) and Fourier IR-spectrometer 640-IR (Varian, Palo Alto, CA, USA). The melting points were measured using a Kofler heating stage. The specific rotation was determined on a PolAAr 3005 (Optical Activity Ltd., Huntingdon, UK) and provided in (deg × mL) × (g × dm)⁻¹, whereas the concentration of the solutions is shown in g × (100 × mL)⁻¹. Thin-layer chromatography was performed on TLC Silica gel 60F₂₅₄ (Merck KGaA, Darmstadt, Germany). X-ray diffraction data for 4 were collected on a Bruker Kappa Apex II diffractometer using Mo Kα radiation. The structure was solved with the ShelXT [15] solution program and the model was refined with ShelXL 2018/3 [16] using full-matrix least-squares minimization on F². Synthetic starting materials and reagents were acquired from Reachem (Moscow, Russia). (+)-Usnic acid was obtained from Zhejiang Yixin Pharmaceutical Co., Ltd. (Lanxi, China). All chemicals were used as described unless otherwise noted.

The atom numbers in the compound provided for the assignment of signals in the NMR spectra correspond to the traditional numeration for usnic acid.

3.1. Procedure for the Usnic Acid C(1)-Amination Reaction

Usnic acid (1 mmol) was added to the ammonia in water (125 mmol, 25% water solution). The reaction mixture was a string with cooling (+9 °C) for 4 days. The conversion was monitored by TLC analysis (silica gel: CHCl₃), and the disappearance of the starting reagent was observed. After that, 1 M HCl solution was added until the pH became less than 5, and a precipitate was formed. The mixture was filtered and air-dried.

(2S)-4,10-diacetyl-3-amino-11,13-dihydroxy-2,12-dimethyl-8-oxatricyclo[7.4.0.0^2,7] trideca-1(9),3,6,10,12-pentaen-5-one (3): Light-brown powder with a 90% yield. M.p.: 269–271 °C. NMR ¹H (DMSO-d₆, δ): 1.72 (3H, s, H-15), 2.02 (3H, s, H-10), 2.42 (3H, s, H-12), 2.63 (3H, s, H-14), 5.75 (1H, s, H-4), 9.41 (1H, s, NH2), 11.35 (1H, s, NH2), 13.51 (1H, s, OH-7). NMR ¹³C (DMSO-d₆, δ): 8.48 (C-10), 31.33 (C-14), 32.54 (C-12), 33.80 (C-15), 51.13 (C-9b) 101.38 (C-6), 102.79 (C-4), 103.67 (C-2), 106.20 (C-8), 107.11 (C-9a), 156.71 (C-9), 156.90 (C-5a), 162.72 (C-7), 171.1 (C-4a), 174.52 (C-1), 184.32 (C-3), 198.75 (C-11), 201.36 (C-13). [α]_D^24.6 +599 (c 0.1980, EtOH). IR (cm−1): 2500–3600 (OH, NH, chelated), 1700 (C=O, ketoenamine system), 1624 (C=O chelated acetyl group), 1579 (C=O chelated ketoenamine system). HRMS: m/z 343.1049 [M]+ (calcd. for (C₁₈H₁₇O₆N₁)+: 343.1050). UV (EtOH) λ_max (nm) and log₁₀ε (in parentless): 215 (4.33), 226 (4.33), 284 (4.43), 323 (4.08).

Crystal data for compound 3: C₁₈H₁₇NO₆ + H₂O, M = 361.34 g·mol⁻¹, crystal size 0.50 × 0.60 × 0.70 mm³, orthorhombic, space group P2₁2₁2₁: a = 11.6754(6) Å, b = 15.9925(11) Å, c = 18.5031(12) Å, V = 3454.9(4) ų, Z = 8, D_calc = 1.389 g/cm³, T = 296 K, 30802 reflections measured (4.32° ≤ 2θ ≤ 56.00°), 8330 unique (R_int = 0.0606). These data were used in all calculations. The final R₁ = 0.0546 (I > 2σ(I)) and wR₂ = 0.1506 (all data), GOF = 1.004. Largest diff. peak/hole/e Å⁻³ 0.27/−0.26. Data were deposited at the Cambridge Crystallographic Data Centre as CCDC 2248817 (Supplementary Materials). The data can be obtained free of charge from the Cambridge Crystallographic Data Centre via http://www.ccdc.cam.ac.uk/structures.

3.2. General Procedure for the Usnic Acid Amination Reaction

Usnic acid (1 mmol) was added to the mixture of the solvent (water, EtOH, MeOH, DMSO, DMF, CH₃CN, or benzene; 5 mL) and ammonia (12.5, 62.5, or 125 mmol, 25% water solution or gaseous ammonia). The reaction mixture was refluxed or stirred at room temperature. The conversion was monitored by TLC analysis (silica gel: CHCl₃), and the disappearance of the starting reagent was observed. After that, 1 M HCl solution was added until the pH became less than 5, and a precipitate was formed. The mixture was filtered and air-dried.

4. Conclusions

It was shown that the regioselectivity of the reaction between usnic acid and ammonia depends on the solvent, reaction temperature, and ammonia amount. So, the reaction in methanol, ethanol, DMSO, DMF, acetonitrile, or benzene led to the formation of ordinary C(11)-enamine, whereas the reaction in water resulted in C(1)-enamine formation. A decrease in the reaction temperature, as well as an increase in ammonia amount, led to an increase of C(1)-enamine amount. An increase in the amount of C(1)-enamine with an increase in the amount of ammonia may be associated with the kinetic features of the reaction. Probably, the order of the reaction resulting in C(1)-enamine is higher than the order of reaction led in C(11)-enamine. An increase in the quantity of C(11)-enamine with an increase in temperature may be due to its greater thermodynamic stability compared to compound 3. However, further study is required to confirm these hypotheses.

The research of this reaction allows us to find conditions for the selective C-1 amination with 90% yield by usnic acid interaction with an excess of ammonia (20 eq.) in water with cooling (+9 °C). Additionally, it was observed that the C(11)- and C(1)-enamines can be easily separated from the reaction mixture by stepwise precipitation.