A single-step synthesis of xyridins A and B, metabolites from Xyris indica

Article abstract of DOI:10.1007/s10593-005-0262-9

A facile single-step synthesis of the title isocoumarins isolated from Xyris indica has been elaborated. Condensation of butanoyl chloride and 2-oxobutanoyl chloride with 3,4-methylenedioxyhomophthalic acid afforded xyridin A and xyridin B, respectively. Xyridin A was saponified to the corresponding keto acid, which on reduction gave the (±)-3,4-dihydro-6,7- methylenedioxy-3-propylisocoumarin in which diastereotopy of the methylenic protons around the stereogenic center was observed. A mass fragmentation mechanism for xyridins has also been suggested.

Full text of DOI:10.1007/s10593-005-0262-9

Chemistry of Heterocyclic Compounds, Vol. 41, No. 8, 2005
A SINGLE-STEP SYNTHESIS OF XYRIDINS A AND B,
METABOLITES FROM Xyris indica
Aamer Saeed
A facile single-step synthesis of the title isocoumarins isolated from Xyris indica has been elaborated.
Condensation of butanoyl chloride and 2-oxobutanoyl chloride with 3,4-methylenedioxyhomophthalic
acid afforded xyridin A and xyridin B, respectively. Xyridin A was saponified to the corresponding keto
acid, which on reduction gave the (±)-3,4-dihydro-6,7-methylenedioxy-3-propylisocoumarin in which
diastereotopy of the methylenic protons around the stereogenic center was observed. A mass
fragmentation mechanism for xyridins has also been suggested.
Keywords: isocoumarins, xyridin A, xyridin B, Xyris indica, diastereotopy, mass fragmentation.
In 1995, Ruangrungsi et al. isolated from the nonpolar fraction of the chloroform extract of the flowering
heads of the weed Xyris indica two new isocoumarins, which they named xyridins A and B [1]. Xyris indica L.
(tall yellow-eyed grass) is one of the five species of the genus Xyris found throughout Thailand and is known
locally as "Kra thin thung". In Bengal the plant has been used in folk medicine as a cure for ring worm, itch, and
leprosy. The structures of xyridin A and B were established by modern spectroscopic techniques as
6,7-methylenedioxy-3-propylisocoumarin (1a) and 6,7-methylenedioxy-3-(1-oxopropyl)isocoumarin (1b),
respectively.
X
O
O
O
O
1a,b
1 a X = H2, b X = O
Although xyridins A and B, like the majority of other naturally occurring isocoumarins derived
biogenetically from acetate via the acetate-polymalonate pathway, possess a C(8) or C(6) and C(8) oxygenation,
xyridins A and B are unique in being 3-alkyl/acyl-substituted 6,7-methylenedioxyisocoumarins. In the literature
unsubstituted 4,5-, 6,7-methylenedioxyisocoumarins and 4,5-, 5,6-, 6,7-methylenedioxy-3,4-dihydro-
isocoumarins are known along with 3-methyl-6,7-methylenedioxy-3,4-dihydroisocoumarin as the only example
of a 3-alkyl-substituted 6,7-methylenedioxyisocoumarin [2-4]. Some important examples of other natural
products bearing a 6,7-methylenedioxydihydroisocoumarin moiety are hippeastrine [5], tetrabenzylcorricidin [6],
lycoricidin [7], pancratistatin [8, 9], and tazettine [10]. Peshawarine contains an 3-aryl-7,8-methylenedioxy-3,4-
dihydroisocoumarin skeleton [11].
__________________________________________________________________________________________
Department of Chemistry, Quaid-i-Azam University 45320, Islammabad Pakistan; e-mail:
aamersaeed@yahoo.com. Published in Khimiya Geterotsiklicheskikh Soedinenii, No. 8, pp. 1140-1145, August,
2005. Original article submitted July 12, 2002.
0009-3122/05/4108-0967©2005 Springer Science+Business Media, Inc.
967

The synthesis of xyridins A and B was undertaken as a continuation of our efforts towards synthesis of
naturally occurring isocoumarins and dihydroisocoumarins [12-16] which exhibit a wide variety of biological
activities despite their relatively simple structure. The limited quantities available from natural sources together
with the possibility of preparing analogues with improved biological activities show the imperative need for the
total synthesis. The condensation of acid chlorides with homophthalic acid has recently emerged as a standard
and authentic route for the preparation of the 3-substituted isocoumarin skeleton [17-19]. Herein a short and
efficient synthesis of xyridins A and B achieved employing this method and the conversion of the former into
corresponding the (±)-3,4-dihydroisocoumarin will be described.
RESULTS AND DISCUSSION
4,5-Methylenedioxyhomophthalic acid was prepared in two steps from 2-bromo-4,5-
methylenedioxybenzoic acid according to the literature procedure [2]. Commercially available 2-oxobutanoic
acid was converted into acid chloride by treatment with oxalyl chloride in dry benzene at room temperature;
butanoyl chloride was the commercial product. Initial efforts to condense 4,5-methylenedioxyhomophthalic acid
with acid chlorides led to opening of the methylenedioxy bridge due to its sensitivity towards acids. Eventually
the condensation was successfully achieved in the presence of a few drops of pyridine. Normally, a yield of 60-
85% is obtained in these condensations, which are carried out at 200°C, but in the case of xyridin A, due to the
relatively lower boiling point of 1-butanoyl chloride (102°C) and the sensitivity of the methylenedioxy bridge to
acids, only a moderate yield of 45% could be obtained despite a longer reflux time. In the case of xyridin B the
limiting factor was the susceptibility to decomposition of 2-oxobutanoyl chloride, resulting in 58% yield.
However, the relatively lower yields are more than compensated by the remarkable simplicity of one-pot
synthesis (Scheme 1).
Thus, direct condensation of 4,5-methylenedioxyhomophthalic acid with butanoyl chloride at elevated
temperatures afforded the xyridin A (1a). It has the characteristic 1H singlet of the isocoumarin moiety at
δ 6.15 ppm for H(4), and a 2H singlet at 6.09 ppm for the methylenedioxy group in the 1H NMR and signals at
δ 103.20 (C(4)), 157.41 (C(3)), and 102.2 ppm (OCH2O) in the 13C NMR spectrum. The IR spectrum showed the
lactonic carbonyl absorption at 1722 cm-1. In the case of xyridin B, due to extension of conjugation, the
characteristic H(4) singlet was found further downfield in the aromatic region at δ 7.30 H(4) and that of the
1
methylenedioxy group at δ 6.20 ppm in the H NMR, and at δ 109.1 (C(4)), 149.41 (C(3)), and 195.3 ppm
13
(C=O) in the C NMR spectrum. DEPT 90° and DEPT 135° experiments confirmed these assignments. The IR
spectrum showed the lactonic and ketonic carbonyl absorptions at 1728 and 1685 cm-1, respectively.
Alkaline hydrolysis of xyridin A (1a) furnished 4,5-methylenedioxy-2-(2-oxopentyl)benzoic acid (3) in
good yield. The keto acid existed partially in its cyclic tautomeric lactol form (3-hydroxy-6,7-methylenedioxy-3-
propyl-3,4-dihydroisocoumarin) as evidenced by the 1H NMR. Thus, in addition to the 2H singlet at δ 3.94 ppm
(H(4), open chain form) each proton of ArCH2 showed a broad signal at δ 2.91 and 3.10 ppm (lactol form). In
the 13C NMR spectrum C(4) resonated at δ 51.80 ppm. The IR spectrum showed the carbonyl absorptions at
1680 and 1705 cm-1.
Sodium borohydride reduction of the keto acid–lactol tautomeric mixture 3 afforded the corresponding
racemic hydroxy acid which underwent spontaneous cyclodehydration on standing for a few minutes (as
monitored by TLC) to (±)-6,7-methylenedioxy-3-propyl-3,4-dihydroisocoumarin (4) [20] without any
dehydrating agent. The protons of methylene groups on either side of the newly created stereogenic center are
nonequivalent, exhibiting a phenomenon called the diastereotopic effect [21, 22]. The motion of methylene at
C(4) is restricted owing to its incorporation into the heterocyclic ring, whereas methylene at C(1') may undergo
free rotation, thus corresponding to ABX and ABMNX systems, respectively. Thus, typical ABX splitting
1
(dddd) of prochiral benzylic protons H(4) (AB) and H(3) (X) protons in H NMR spectrum was observed. The
AB part at δ 2.85-2.96 ppm showed that the hydrogen atom trans to the side chain is located slightly upfield and
968

Scheme 1
Br
O
O
O
O
COOH
COOH
COOH
O
2
O
Cl
Cl
a
a
O
O
O
O
O
O
O
O
O
O
1a
1b
b
H
O
O
O
O
O
c
O
COOH
3
O
4
OH
O
O
O
O
a 200 oC, 4–20 h, 45–58%;
c NaBH4, EtOH, 4 h, r. t., 85%
b 5% KOH/EtOH, 4 h reflux, 70%;
appeared as a doublet of doublets at δ 2.85 ppm (Jgem = 12.4, Jtrans = 3.52 Hz), while the cis hydrogen resonated
slightly downfield at 2.96 ppm (Jgem = 17.08, Jcis = 13.04 Hz) with a chemical shift difference (∆ν = H4A–H4B) of
about 0.1 ppm, which is a measure of the diastereotopy extent. The methine proton H(3) (X) showed a dddd
pattern at δ 4.25 ppm, which is evidence of the presence of diastereotopic methylenes on both C(4) and C(1').
Thus, H(1'A) and H(1'B) and each of the protons of methylene at C(1') also showed a dddd at δ 1.61 and
1.80 with a chemical shift difference of 0.19 ppm, almost double the extent of diastereotopy compared to that
between H(4A) and H(4B). It is quite interesting to note that the diastereotopic effect extends up to the C(2')
methylenic protons, corresponding to an ABMNX2 system with two sets of multiplets at δ 1.45 and 1.54,
respectively; in this case ∆ν is reduced to ≈ 0.1 ppm due to the higher separation from the chiral center. The
13C NMR spectrum showed signals at δ 77.63 and 35.01 ppm for C(3) and C(4), respectively. The lactonic
carbonyl absorption appeared at 1720 cm-1 in the IR spectrum.
Scheme 2 delineates the mass fragmentation pattern of the EIMS spectra of xyridins A and B, which is
consistent with the general mass fragmentation mechanism for isocoumarins. The major peaks correspond to
α-cleavage, γ-H rearrangement, and retro-Diels–Alder cleavage products.
Thus, a facile single-step synthesis of the medicinally important isocoumarins xyridins A and B has been
elaborated and their structures were unambiguously confirmed, which makes the substances available for
biological evaluation.
969

Scheme 2
.
.
+
+
X
O
O
O
O
X
O
O
– CO
– CO
O
1a–b; m/z 176 (I 76%)
1b–b; m/z 218 (I 46%)
1a–a; m/z 204 (I 14%)
1b–a; m/z 218 (I 16%)
.
+
O
– H2C=CH2
O
O
1a–c; m/z 162 (I 23%)
X
1b–c;
(I 19%)
.
+
–
+
.
+
X
O
CH
C
O
O
O
O
O
O
O
1a
1a X = H2, m/z 232 (I 79%)
1b X = O, m/z 246 (I 68%)
1a–d; m/z 161 (I 12%)
1b–d; (I 16%)
+ . X
– CO
O
O
O
– CO
– CO
1a–g; m/z 133 (I 37%)
1b–g; (I 100%)
+
O
O
+
O
O
1a–f; m/z 161 (I 12%)
1b–f; (I 16%)
1a–e; m/z 189 (I 17%)
1b–e;
(I 20%)
EXPERIMENTAL
General: 2-Oxobutanoic acid and 1-butanoyl chloride were commercial products from Aldrich.
2-Bromo-4,5-methylenedioxybenzoic acid was prepared from piperonal according to the reported procedure
13
[23].1H and C NMR spectra (δ) were recorded in CDCl3 at 400 and 100 MHz, respectively (Bruker AM-100).
IR spectra were recorded on a Bruker Vector 22, mass spectra (EI, 70eV) on a MAT 312 instrument, and
elemental analyses with a CHN-Rapid Heräus. Flash column chromatography was carried out on Merck
Kieselgel 60 (230-400 mesh).
6,7-Methylenedioxy-3-propylisocoumarin (1a, Xyridin A). A stirred mixture of 4,5-methylene-
dioxyhomophthalic acid 2 (0.25 g, 1.11 mmol) and butanoyl chloride (0.46 ml, 4.46 mmol) containing a few
drops of dry pyridine was heated on an oil bath at 200°C for 20 h. Flash column chromatography of the residue
(petroleum ether–ethyl acetate, 8:1) afforded 1a (0.11 g, 45%) as yellow scales; mp 64-65°C (lit. [1] 67-88°C).
970

1
IR spectrum (KBr), ν, cm-1: 3072, 2913, 2849, 1722, 1685, 1645, 1575, 1260, 1120, 935, 810. H NMR
spectrum, δ, ppm (J, Hz): 0.95 (3H, t, J = 7.2, H(3')); 1.67 (2H, tq, J = 7.2, J = 7.24, H(2')); 2.44 (2H, t, J = 7.0,
13
H(1')); 6.09 (2H, s, OCH2O); 6.15 (1H, s, H(4)); 6.70 (1H, s, H(5)); 7.59 (1H, s, H(8)). C NMR spectrum,
δ, ppm: 163.03 (C(1)), 157.41 (C(3)), 153.7 (C(7)), 148.3 (C(6)), 135.8 (C(8a)), 114.65 (C(4a)), 104.01 (C(5)),
103.20 (C(4)), 102.2 (OCH2O); 35.71 (C(1')), 20.86 (C(2')), 13.51 (C(3')). EIMS, m/z (I, %): 232 [M]+ (79), 204
(14), 203 (77), 189 (17), 176 (76), 162 (23), 161 (12), 133 (37), 75 (24). Found, %: C 67.26; H 5.19. C13H12O4.
Calculated, %: C 67.23; H 5.21.
6,7-Methylenedioxy-3-(1-oxopropyl)isocoumarin (1b, Xyridin B). A stirred mixture of compound 2
(0.25 g, 1.11 mmol) and 2-oxobutanoyl chloride (0.53 g, 4.46 mmol) containing a few drops of dry pyridine was
heated on an oil bath at 200°C for 4 h. Flash chromatography of the residue (petroleum ether–ethyl acetate, 15:1)
afforded 1b (0.16 g, 58%) as scales; mp 188-190°C (lit. [1] 198-199°C). IR spectrum (KBr), ν, cm-1: 3084, 2913,
1
2849, 1728, 1685, 1645, 1575, 1260, 1120, 935, 810. H NMR spectrum, δ, ppm (J, Hz): 1.21 (3H, t, J = 7.2,
H(3')); 3.01 (2H, q, J = 7.2, H(2')); 6.20 (2H, s, OCH2O); 7.10 (1H, s, H(5)); 7.30 (1H, s, H(4)); 7.38 (1H, s,
H(8)). 13C NMR spectrum, δ, ppm: 195.3 (C(1'), C=O), 161.03 (C(1)), 149.41 (C(3)), 154.0 (C(7)), 151.0 (C(6)),
132.8 (C(8a)), 118.65 (C(4a)), 109.10 (C(4)), 106.41 (C(5)), 102.8 (OCH2O), 31.86 (C(2')), 7.51 (C(3')). EIMS,
m/z (I, %): 246 [M]+ (68), 189 (20), 162 (19), 161 (16), 133 (100), 75 (24). Found, %: C 63.29; H 4.16.
C13H10O5. Calculated, %: C 63.42; H 4.09.
4,5-Methylenedioxy-2-(2-oxopentyl)benzoic Acid (3). A stirred solution of xyridin A (1a) (0.095 g,
0.40 mmol) in ethanol (10 ml) was treated with 5% KOH (20 ml) and the mixture refluxed for 4 h. After cooling
the reaction mixture, cold water (10 ml) was added and the mixture acidified with diluted hydrochloric acid and
immediately extracted with dichloromethane (2 × 30 ml). The organic phase was dried (MgSO4) and the solvent
evaporated under vacuum to leave 3 as a yellowish solid. Recrystallized from petroleum ether–ethyl acetate
(0.066 g, 70%); mp 104-106°C. IR spectrum (KBr), ν, cm-1: 3011, 2949, 1715, 1694, 1601, 1202, 1162.
1H NMR spectrum, δ, ppm (J, Hz): 0.87 (3H, t, J = 6.02, H(3')); 1.44 (2H, br. m, H(2')); 1.75 (2H, br. m, H(1'));
2.38 ( 2H, br. m, H(1') open chain); 2.91 (1H, br. m, H(4) lactol); 3.10 (1H, br. m, H(4) lactol); 3.94 (2H, br. m,
H(4) open chain); 6.17 (2H, s, OCH2O); 6.73 (1H, s, H(5)); 7.65 (1H, s, H(8)); 11.22 (1H, br. s, COOH).
13C NMR spectrum, δ, ppm: 195.54 (C(3), C=O), 168.30 (COOH), 57.70 (C(4)), 153.9 (C(7)), 148.3 (C(6)),
104.43 (C(8a)), 112.90 (C(5)), 102.2 (OCH2O), 42.99 (C(1')), 19.73 (C(2')), 13.80 (C(3')). EIMS, m/z (I, %): 250
[M]+ (11.45), 232 (80.1), 203 (11.61), 179 (41.43), 134 (100). Found, %: C 62.23; H 5.67. C13H14O5. Calculated,
%: C 62.39; H 5.64. (In order to avoid confusion and for direct comparison C/H numbering is the same as in
isocoumarins 1a,b.)
(±)-6,7-Methylenedioxy-3-n-propyl-3,4-dihydroisocoumarin (4). Sodium borohydride (0.67 g,
18 mmol) was added portionwise to a stirred solution of compound 3 (0.05 g, 0.21 mmol) in ethanol (10 ml) and
water (30 ml). The reaction mixture was stirred for 4 h at room temperature, diluted with water (50 ml), acidified
with conc. HCl, and stirred for a further 2 h. It was then saturated with ammoniunm sulfate and extracted with
EtOAc (3 × 30 ml). The layers were separated and the organic layer dried (MgSO4) and concentrated. Flash
chromatography (petroleum ether–ethyl acetate, 7:1) afforded 4 as light yellow prisms (0.16 g, 85%);
1
mp 54-57°C. IR spectrum (KBr), ν, cm-1: 2960, 2853, 1720, 1604, 1583, 1464, 1198, 1085, 840. H NMR
spectrum, δ, ppm (J, Hz): 0.96 (3H, t, J = 7.3, H(3')); 1.64 (1H, m, H(2A')); 1.71 (1H, m, H(2B')); 1.71 (1H, dddd,
J = 13.5, J = 10.2, J = 7.4, J = 5.2, H(1A')); 1.87 (1H, dddd, J = 13.5, J = 10.8, J = 5.0, J = 5.0, H(1B')); 2.85 (1H,
dd, Jgem = 16.3, Jtrans = 4.0, H(4)); 2.96 (1H, dd, Jgem = 16.5, Jcis = 11.11, H(4)); 4.48 (dddd, J = 10.5, J = 7.6,
J = 5.4, J = 4.6, H(3)); 6.14 (2H, s, OCH2O); 6.74 (1H, s, H(5)); 7.59 (1H, s, H(8)). 13C NMR spectrum, δ, ppm:
162.46 (C(1)), 76.71 (C(3)), 36.03 (C(4)), 114.92 (C(4a)), 106.05 (C(5)), 147.7 (C(6)), 154.2 (C(7)), 107.30
(C(8)), 134.90 (C(8a)), 32.71 (C(1')), 18.10 (C(2')), 13.80 (C(3')). EIMS, m/z (I, %): 234 [M]+ (36), 232 (41),
163 (31), 162 (100), 191 (29), 134 (65). Found, %: C 66.59; H 6.09. C13H14O4. Calculated, %: C 66.66; H 6.02.
971

The author gratefully acknowledges a Georg Forster research fellowship from the Alexander von
Humboldt-Foundation Germany.
REFERENCES
1.
N. Ruangrungsi, T. Sekine, T. Phadungcharoen, S. Suriyagan, and I. Murakoshi, Phytochemistry, 38,
481 (1995).
2.
3.
4.
V. H. Belgaonkar and R. N. Usgaonkar, J. Chem. Soc., Perkin Trans. 1, 702 (1977).
J. N. Srivasta and D. N. Chaudhury, J. Org. Chem., 27, 4337 (1962).
B. H. Bhide and K. K. Shah, Indian J. Chem., 19B, 9 (1980).
5.
J. Katakawa, H. Meguri, and H. Irie, Heterocycles, 22, 2213 (1984).
6.
R. C. Thompson and J. Kallmerten, J. Org. Chem., 55, 6076 (1990).
7.
H. Paulsen and M. Stubbe, Liebigs Ann. Chem., 535 (1983).
8.
T. K. Park and S. J. Danishefsky, Tetrahedron Lett., 36, 195 (1995).
9.
G. E. Keck, S. F. McHardy and J. A. Murry, J. Am. Chem. Soc., 117, 7289 (1995).
M. M. Abelman, L. E. Overman and V. D. Tran, J. Am. Chem. Soc., 112, 6959 (1990).
J. J. Fitzgerald, A. R. Pagano, V. M. Sakoda and R. A. Olofson, J. Org. Chem., 59, 4117 (1994).
N. H. Rama, A. Saeed and C. W. Bird, Liebigs Ann. Chem., 1331 (1993).
N. H. Rama, A. Saeed and C. W. Bird, Liebigs Ann. Chem., 711 (1995).
A. Saeed and N. H. Rama, J. Chem. Soc. Pak., 16, 220 (1994).
A. Saeed and N. H. Rama, Arabian J. Sci. Eng., 21, 403 (1996).
N. H. Rama, R. Iqbal, Kh. Zamani, A. Saeed, M. Z. Iqbal, and M. I. Chaudary, Indian J. Chem., 37B,
365 (1998).
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
K. Nozawa, M. Yamada, Y. Tsuda, K. I. Kawai, and S. Nakajima, Chem. Pharm. Bull., 29, 2491 (1981).
K. Nozawa, M. Yamada, Y. Tsuda, K. I. Kawai, and S. Nakajima, Chem. Pharm. Bull., 29, 3486 (1981).
H. Kaji, M. Yamada, K. Nozawa, K. I. Kawai, and S. Nakajima, Org. Prep. Proced. Int., 18, 253 (1986).
T. Kanda, S. Kato, T. Sugino, N. Kambe, A. Oganawa, and N. Sonoda, Synthesis, 1102 (1995).
J. K. Kendall, T. H. Fisher, H. P. Schultz, and T. P. Schultz, J. Org. Chem., 54, 6076 (1989).
D. H. Williams and I. Fleming, Spectroscopic Methods in Organic Chemistry, 4th Ed., McGraw-Hill,
UK, 1987, p. 88.
23.
D. Becker, L. R. Hughes, and R. A. Raphael, J. Chem. Soc., Perkin Trans. 1, 1674 (1977).
972