(Z)-3-butylidenephthalide from Ligusticum porteri, an α-glucosidase inhibitor

Article abstract of DOI:10.1021/np100447a

An extract from the roots of Ligusticum porteri, orally administered to groups of normal and diabetic mice, showed significant hypoglycemic and antihyperglycemic effects. Experimental type-II DM was achieved by treating mice with streptozotocin 15 min after an injection of β-nicotinamide adenine dinucleotide. (Z)-6,6′,7,3′α-Diligustilide (1), (Z)-ligustilide (2), 3-(Z)-butylidenephthalide (3), myristicin (4), and ferulic acid (5) were isolated from the active extract. When tested In Vivo, compounds 1-3 showed antihyperglycemic activity, with 3 being the most active. Compound 3 (56.2 mg/kg) decreased blood glucose levels in NAD-STZ-diabetic mice after an oral sucrose load, suggesting that its antihyperglycemic effect is due to inhibition of α-glucosidase at the intestinal level. Furthermore, 3 inhibited the activity of yeast-α-glucosidase (IC₅₀ 2.35 mM) in a noncompetitive fashion with a K_i of 4.86 mM. Docking analysis predicted that 3 binds to the enzyme in a pocket close to the catalytic site, but different from that for acarbose, with a K_i of 11.48 mM. Compounds 1 and 2 did not affect α-glucosidase In Vivo, but altered glucose absorption by a mechanism yet to be determined. The stimulatory effect of 5 on insulin secretion, present in high amounts in the extract, has been demonstrated in previous investigations. The present study provides scientific support of the use of L. porteri in Mexican folk medicine for the treatment of diabetes.

Full text of DOI:10.1021/np100447a

ARTICLE
pubs.acs.org/jnp
(Z)-3-Butylidenephthalide from Ligusticum porteri, an r-Glucosidase
Inhibitor
Fernando Brindis,†,§ Rogelio Rodríguez,† Robert Bye,‡ Martín Gonzꢀalez-Andrade,† and Rachel Mata*,†
†Facultad de Química and ‡Instituto de Biología, Universidad Nacional Autꢀonoma de Mꢀexico, Mꢀexico DF 04510, Mꢀexico
S Supporting Information
b
ABSTRACT: An extract from the roots of Ligusticum porteri,
orally administered to groups of normal and diabetic mice,
showed significant hypoglycemic and antihyperglycemic effects.
Experimental type-II DM was achieved by treating mice with
streptozotocin 15 min after an injection of β-nicotinamide
adenine dinucleotide. (Z)-6,60,7,30R-Diligustilide (1), (Z)-
ligustilide (2), 3-(Z)-butylidenephthalide (3), myristicin (4),
and ferulic acid (5) were isolated from the active extract. When
tested In Vivo, compounds 1-3 showed antihyperglycemic activity, with 3 being the most active. Compound 3 (56.2 mg/kg)
decreased blood glucose levels in NAD-STZ-diabetic mice after an oral sucrose load, suggesting that its antihyperglycemic effect is
due to inhibition of R-glucosidase at the intestinal level. Furthermore, 3 inhibited the activity of yeast-R-glucosidase (IC50 2.35 mM)
in a noncompetitive fashion with a Ki of 4.86 mM. Docking analysis predicted that 3 binds to the enzyme in a pocket close to the
catalytic site, but different from that for acarbose, with a Ki of 11.48 mM. Compounds 1 and 2 did not affect R-glucosidase In Vivo,
but altered glucose absorption by a mechanism yet to be determined. The stimulatory effect of 5 on insulin secretion, present in high
amounts in the extract, has been demonstrated in previous investigations. The present study provides scientific support of the use of
L. porteri in Mexican folk medicine for the treatment of diabetes.
ype-II diabetes mellitus (DM) is a chronic disease associated
with abnormally high blood glucose levels (hyperglycemia)
families used as hypoglycemic agents. Ligusticum porteri Coult. &
Rose (Apiaceae) is one of these.4 Roots and leaves from this herb,
commonly referred to as “osha”, Porter’s licorice-root, or “chu-
chupate”, are employed to treat inflammatory, respiratory, gas-
trointestinal, and infectious complaints.5-7 More recently, its use
as an antidiabetic drug has been described.4 The essential oil
(100 μg/mL) prepared from the roots increased the antimicro-
bial activity of norfloxacin against a norfloxacin-resistant strain of
Staphylococcus aureus.8 Ethanolic and aqueous extracts did not
affect the growth of human MCF-7/AZ breast or HCT8/E11
colon cancer cells.9 A CH2Cl2-MeOH (1:1) extract of roots of
L. porteri exhibited a significant antinociceptive effect, as detected
by the writhing test in mice.10 On the other hand, toxicity studies
of the same extract according to the Lorke11 and Ames12 pro-
cedures revealed that the plant was devoid of genotoxic effects
and allowed calculation of a medium lethal dose (LD50) of 1 g/kg
for mice.13
T
due to defects in insulin secretion, or action, or both. Hypergly-
cemia provokes protein glycation, which cause tissue damage and
devastating complications including heart and peripheral vascu-
lar complaints, retinopathy, nephropathy, and neuropathy.1,2
According to the World Health Organization, at least 2.8% of
the world population suffered from type-II DM in 2000. The
incidence of this illness is increasing rapidly, and it is estimated
that by 2030 this figure will almost double. Type-II DM occurs
throughout the world, but is more common in developed coun-
tries. The greatest increase in prevalence is, however, expected to
occur in developing countries, where most patients will probably
be found by 2030.1
There is currently no cure for diabetes, but the condition can
be managed so that patients can live a relatively normal life.
Treatment of diabetes focuses on two goals: keeping blood
glucose within the normal range, thus preventing the develop-
ment of long-term effects, and changes in diabetic patients’ life-
style. Oral medications are available to lower blood glucose in
type-II diabetics, but most of these products possess side effects
after prolonged use.3 In consequence, the search for new ther-
apeutic agents for treating type-II DM, including plants used in
folk medicine, has increased notably in recent years.4
Chemical investigation of L. porteri led to the isolation and iden-
tification of secondary metabolites such as phenylpropanoids, ter-
penoids, and several phthalides including (Z)-6,60,7,30R-diligustilide
(1), (Z)-ligustilide (2), and (Z)-3-butylidenephthalide (3).14-19
Special Issue: Special Issue in Honor of Koji Nakanishi
Mexico is rich in medicinal plants highly prized by the pop-
ulation for the treatment of diabetes, and according to a recent
review, there are about 306 species from 235 genera and 93
Received: July 2, 2010
Published: September 29, 2010
r
Copyright
2010 American Chemical Society and
314
dx.doi.org/10.1021/np100447a J. Nat. Prod. 2011, 74, 314–320
American Society of Pharmacognosy
|

Journal of Natural Products
ARTICLE
Table 1. Effect of Compound 3 and Crude Extract of the Roots of L. porteri on Blood Glucose Levels in STZ-Diabetic Mice after an
Oral Load of Sucrose (2 g/kg)a
blood glucose concentration
test sample
dose, mg/kg
0 h
0.5 h
1 h
1.5 h
2 h
vehicle
acarbose
extract
3
155 ( 4.7
153 ( 7.4
143 ( 3.7
152 ( 8.7
149 ( 5.0
236 ( 9.4(56)
152 ( 8.3(-0.4)b
194 ( 6.8(35)b
214 ( 12.0(41)
184 ( 7.3(23)b
169 ( 14.0(11.4)
157 ( 7.9(0.5)b
173 ( 2.1(21)
180 ( 13.8(19)
161 ( 6.1(7.6)
163 ( 4.9(6.1)
159 ( 6.5(6.2)
170 ( 6.0(19)
173 ( 11.9(16)
153 ( 6.9(2.5)
145 ( 6.0(-4.3)
158 ( 6.3(6.8)
161 ( 3.9(12)
163 ( 15.4(6.7)
159 ( 6.8(6.4)
5
100
10
3
56.2
a Each value is the mean ( SEM for six mice in each group. Percent variations of glycemia are in parentheses. b p < 0.05 significantly different ANOVA
followed by Dunnett’s t test for comparison with respect to control group.
Since the pharmacological investigation of plants used in folk
medicine could lead to the discovery of new antidiabetic agents
and/or to the proposal of therapeutic alternatives for correcting
impaired glucose homeostasis, the objectives of the present study
were to ascertain if L. porteri had blood glucose lowering activity
and to identify new R-glucosidase inhibitors that could effectively
control postprandial glucose levels. Such control has been shown
to be essential in preventing the progress of impaired glucose
tolerance toward type-II DM by decreasing the incidence of
hyperglycemia.
Supporting Information) and diabetic (Table 1) mice, was carried
out. This test is usually performed to evaluate if the antihyper-
glycemic action of a drug involves inhibition of intestinal
R-glucosidase; this enzyme reduces the rate of digestion of
carbohydrates by preventing their immediate breakdown into
monosaccharides, which then would be absorbed quickly into the
bloodstream. Slowing the absorption of carbohydrates gives the
β-cells in the pancreas more time to secrete adequate insulin to
cover the meal.21 The extract (31.6 and 56.2 mg/kg) lowered
glucose levels in both diabetic and normal mice, revealing that
inhibition of R-glucosidase was mediating the observed antihy-
perglicemic effect. The activity was compared with that of
acarbose (5 mg/kg), a natural R-glucosidase inhibitor currently
used to control blood glucose levels.
’ RESULTS AND DISCUSSION
The use of L. porteri in Mexican folk medicine prompted us
to determine its efficacy as an antidiabetic agent using well-
known animal models. First, an organic extract (56.2, 100, and
316 mg/kg) of the roots was tested using the oral glucose tole-
rance test (OGTT) in two sets of animals: normal and diabetic.
Experimental type-II DM was achieved by treating mice with
streptozotocin (STZ, 100 mg/kg) 15 min after an injection of β-
nicotinamide adenine dinucleotide (NAD, 40 mg/kg).20 This
preliminary treatment with NAD provokes partial protection
against the cytotoxic action of STZ by scavenging free radicals
and causes only minor damage to pancreatic β-cell mass, creating
a diabetic syndrome close to type-II DM. This model has recently
proven to be a valuable tool for investigation of new antidiabetic
agents.20 In normal mice, the crude extract (Table S1, Supporting
Information) did not affect the postprandial peak at all doses
tested, but when assessed in NAD-STZ-diabetic mice, a signifi-
cant decrease (p < 0.05) of the postprandial peak was observed
(Figure S1 and Table S2, Supporting Information). The glucose-
lowering effect (∼45% at all doses tested) was noticed after
30 min and persisted throughout the experiment, suggesting that
the extract was not only antihyperglycemic but also hypoglyce-
mic. Next, in order to confirm the hypoglycemic action of the
extract, its acute effect was evaluated in normal (Table S3, Sup-
porting Information) and diabetic animals (Table S4, Supporting
Information). The results revealed that the extract (56.2, 100,
and 316 mg/kg) caused significant decreases in blood glucose
levels only in NAD-STZ-diabetic mice when compared with
vehicle-treated groups (p < 0.05). The effect of the extract was
attained after 1.5 h and was maintained all through the experi-
ment; in the case of glybenclamide (Gly), used as positive control,
the initial effect was observed after 3 h of treatment and lasted
throughout the experiment. Furthermore, the action of the lowest
dose of extract tested was comparable to that of Gly. Finally, an
oral sucrose tolerance test (OSTT), both in normal (Table S5,
In order to determine the antidiabetic principle(s) of the plant,
the active extract was subjected to chromatographic fractionation
to yield compounds 1-5. Phthalides 1-3 were identified by
comparison of their spectroscopic and spectrometric data with
those described previously,16,22,23 and compounds 4 and 5 were
identified by comparison with commercial authentic samples.
In the OGTT with normal mice, administration of com-
pounds 1 (31.2 and 56.2 mg/kg) and 3 (10 and 20 mg/kg) dis-
played significant (p < 0.05) antihyperglycemic activity, provok-
ing a decrease of the postprandial peak in comparison with the
vehicle, whereas 2 (10 and 20 mg/kg), the most abundant
315
dx.doi.org/10.1021/np100447a |J. Nat. Prod. 2011, 74, 314–320

Journal of Natural Products
ARTICLE
Table 2. Effect of Compounds 1-3 from the Roots of L. porteri on Blood Glucose Levels in Normal Mice after an Oral Load of
Glucose (1 g/kg)a
blood glucose concentration
test sample
dose mg/kg
0 h
0.5 h
1 h
1.5 h
2 h
vehicle
108 ( 5.3
121 ( 7.0
93 ( 6.8
172 ( 10.6(59)
71 ( 7.1(-42)b
97 ( 8.3(3.8)b
135 ( 8.7(41)b
151 ( 3.0(42)
187 ( 10.0(40)
110 ( 10(-9)b
153 ( 3.0(20)b
109 ( 15.8(1.5)
43 ( 8.0(-65)b
66 ( 9.5(-28)b
92 ( 8.8(-3.8)
130 ( 5.1(25)
150 ( 3.6(14)
93 ( 6.9(-24)b
117 ( 3.7(-6.9)
110 ( 13.2(2.5)
37 ( 3.5(-69)b
61 ( 8.9(-35)b
84 ( 13.1(-13)b
113 ( 1.8(7)
94 ( 7.7(-12)
39 ( 6.2(-68)b
46 ( 8.1(-50)b
69 ( 9.0(-27)b
96 ( 1.7(-8)
Gly
1
10
31.2
56.2
10
1
95 ( 4.3
2
109 ( 2.9
134 ( 4.9
122 ( 4.0
129 ( 8.6
2
20
126 ( 5.6(-3.5)
70 ( 10(-41)b
103 ( 2.6(-18)b
109 ( 6.9(-16)
71 ( 9.4(-42)b
98 ( 7.3(-23)b
3
10
3
20
a Each value is the mean ( SEM for six mice in each group. Percent variations of glycemia are in parentheses. b p < 0.05 significantly different ANOVA
followed by Dunnett’s t test for comparison with respect to control group.
Table 3. Effect of Compounds 2 and 3 from Roots of L. porteri on Blood Glucose Levels in STZ-Diabetic Mice after an Oral Load of
Glucose (1 g/kg)a
blood glucose concentration
test sample
dose mg/kg
0 h
0.5 h
1 h
1.5 h
2 h
vehicle
218 ( 50.5
163 ( 16.4
172 ( 6.0
185 ( 10.5
149 ( 1.6
225 ( 15.4
284 ( 52.1
429 ( 59.4
320 ( 59.3(51.9)
250 ( 31.1(42)
192 ( 5.4(13)b
217 ( 14.2(17)b
186 ( 2.6(25)b
295.6 ( 17.4(35)
356 ( 63.7(27)b
467 ( 52.8(12)b
229 ( 39.0(10.3)
182 ( 27.7(3.8)
146 ( 6.0(-14)b
170 ( 7.0(-6.3)b
173 ( 2.1(17.1)
229 ( 16.6(3.5)
292 ( 53.1(3.3)
413 ( 62.3(-5.2)
217 ( 49.5(5 ꢀ 10-3
147 ( 17.6(-10)
138 ( 6.7(-19)b
160 ( 16.6(-12)b
151 ( 2.2(2.1)
)
193 ( 49.4(-12.9)
130 ( 19.0(-21.5)
143 ( 8.6(-15.6)
165 ( 14.0(-10.0)
157 ( 1.0 (6.5)
Gly
2
10
10
2
31.2
56.2
10
2
3
214 ( 12.4(1.5)
290 ( 58.3(-0.3)
387 ( 62.4(-12)
206 ( 15.8(-7.4)
259 ( 48.4(-9.1)
352 ( 62.1(-21.7)
3
31.2
56.2
3
a Each value is the mean ( SEM for six mice in each group. Percent variations of glycemia are in parentheses. b p < 0.05 significantly different ANOVA
followed by Dunnett’s t test for comparison with respect to control group.
Table 4. Effect of 1 from the Roots of L. porteri on Blood Glucose Levels in STZ-Diabetic Mice after an Oral Load of Glucose
(1 g/kg)a
blood glucose concentration
test sample
dose mg/kg
0 h
0.5 h
1 h
1.5 h
2 h
vehicle
218.5 ( 50.5
163.5 ( 16.4
330.1 ( 77.0
176.1 ( 11.5
208 ( 18.4
320 ( 59.3(51.9)
250.8 ( 31.1(42)
391.6 ( 64.8(32)
248.1 ( 13.7(44)
344 ( 50.2(57)
229.8 ( 39.0(10)
182 ( 27.7(3.8)
331.1 ( 67.1(5.7)
218.5 ( 8.9(32)
320 ( 38.1(52)
217 ( 49.5(0.005)
147.5 ( 17.6(-10)
334.6 ( 68.0(5.7)
205.3 ( 9.4(23)
193 ( 49.4(-13)
130.6 ( 19.0(-21)
304.1 ( 65.2(-5.5)
179.1 ( 7.5(8.8)
Gly
1
10
10
1
31.2
1
56.2
265 ( 31.7(25)
237 ( 28.2(8.9)
a Each value is the mean ( SEM for six mice in each group. Percent variations of glycemia are in parentheses. p < 0.05 significantly different ANOVA
followed by Dunnett’s t test for comparison with respect to control group.
compound in the organic extract, did not affect blood glucose
levels after the glucose challenge (see Table 2). In NAD-STZ-
diabetic mice, compound 2 (10, 31.2, and 56.2 mg/kg) induced a
significant (p < 0.05) and dose-dependent decrease of the post-
prandial peak, suggesting that its mode of action did not involve
inhibition of glucose absorption, but that it had a different mech-
anism yet to be demonstrated (Table 3). In contrast, compound
1 (10, 31.2, and 56.2 mg/kg), which affected the glucose level in
normal mice, did not lower the postprandial peak in NAD-STZ-
diabetic mice, indicating that its effect is mediated by an insulin-
dependent mechanism. Finally, compound 3 (10, 31.2, and
56.2 mg/kg) inhibited (Table 4) the postprandial peak after the
glucose load in NAD-STZ-diabetic mice at the two highest doses
tested. The latter result was consistent with compound 3 being
an inhibitor of glucose absorption. Since compound 3 was the
only one lowering the postprandial peak in both normal and
diabetic animals, its effect on glucose absorption In Vivo using
the OSTT was investigated. The assay was performed only in
NAD-STZ-diabetic mice due to scarcity of sample. Thus, admin-
istration of 3-(Z)-butylidenephthalide (3) (10 and 56.2 mg/kg)
to NAD-STZ-diabetic mice provoked a decrement of 50% of the
postprandial peak (p < 0.05) at a dose of 56.2 mg/kg (Figure 1).
316
dx.doi.org/10.1021/np100447a |J. Nat. Prod. 2011, 74, 314–320

Journal of Natural Products
ARTICLE
Figure 1. Effects of the crude extract of the roots of L. porteri and compound 3 on blood glucose levels in NAD-STZ-diabetic mice using the OST test.
Each value is the mean ( SEM for 6 rats in each group. *p < 0.05 significantly different ANOVA followed by Dunnett’s t-test for comparison with respect
to negative control values at the same time.
Figure 2. Plots of R-glucosidase inhibition at different concentrations of substrate: (A) acarbose; (B) (Z)-3-butylidenephthalide (3).
These results clearly revealed that compound 3 is able to inhibit
R-glucosidase at the intestinal level with consequent decrease of
the postprandial peak in NAD-STZ-diabetic mice.
using p-nitrophenyl-R-D-glucopyranoside (pNPG) as substrate
and yeast R-glucosidase.29 The results were also compared with
those of acarbose (Figure 2), revealing that 3 inhibited the activ-
ity of yeast R-glucosidase in a concentration-dependent manner.
The IC50 calculated was 2.35 mM (acarbose = 0.42 mM). As
expected from the In Vivo experiments, compounds 1 and 2 were
not R-glucosidase inhibitors and neither were compounds 4 and 5.
In order to obtain further evidence of the nature of inhibition
exerted by compound 3, a kinetic analysis of the enzyme inhi-
bition activity was assessed using different amounts of compound
3 and acarbose. The results showed that 3 inhibited activity of the
enzyme with a Ki of 4.86 mM vs Ki of 0.46 mM for acarbose.
Dixon plots30 for R-glucosidase from yeast in the presence of
acarbose and 3-(Z)-butylidenephthalide (3) at different concen-
trations (Figure 2) revealed typical curves for competitive and
mixed-type noncompetitive inhibitors, respectively. According to
Figure 2, the increment in acarbose concentration resulted in
lowering the slope of the lines, in agreement with a reversible
inhibitor. On the other hand, compound 3 displayed an inhibi-
tion profile for a mixed-type inhibitor since the corresponding
graphic showed a series of lines intersecting to the left of the
vertical and above the horizontal axes.28
Compounds 4 and 5 were not tested because their antidiabetic
properties In Vivo and in vitro have been previously demon-
strated. Ferulic acid (5) at a concentration of 25 μM causes a
2.98-fold increase in 2-deoxyglucose transport into the L6 myo-
tubes; its performance was comparable to commercial oral hypo-
glycemic drugs.24 Expression of GLUT4 and PI3K transcripts
was also elevated in ferulic acid (5)-treated cells. In addition, 5
possesses stimulatory effects on insulin secretion in vitro in a
concentration-dependent manner.25 In other tests, ferulic acid (5)
administered orally at low dosage (0.1% or 0.01%) suppressed
hyperglycemia associated with induced diabetes in the model of KK-
Ay mice.26 Finally, ferulic acid (5) was the most potent insulin-
secreting agent among several cinnamic acid derivatives, and its i.v.
administration at a dose of 5 mg/kg significantly decreased plasma
glucose and increased insulin concentration in normal rats.27
Myristicin (4) and some analogues from Myristica fragrans
fruit are claimed as adiponectin production stimulators and
health foods for prevention and treatment of type-II diabetes,
hyperlipidemia, hypertension, obesity, and arteriosclerosis.28
The effects of compounds 1-5 on the activity of R-glucosi-
dase was also tested in vitro according to a well-known procedure
Altogether these results suggest that compound 3 does not
bind to the same catalytic site as acarbose in R-glucosidase.
317
dx.doi.org/10.1021/np100447a |J. Nat. Prod. 2011, 74, 314–320

Journal of Natural Products
ARTICLE
Figure 3. Docking results using the structural model of the R-glucosidase: (A) site of binding of acarbose, which comprises the catalytic site of the
enzyme; (B) binding conformation of (Z)-3-butylidenephthalide (3).
Indeed, the docking analysis predicts that 3 binds to the enzyme
in a different pocket than acarbose; the binding site for 3 is close
to the catalytic site and is formed by residues Ser299, Thr287,
Val297, Trp340, His302, Ala341, Thr342, Ile334, and Tyr344
(Figure 3), with a predicted Ki of 11.48 μM. As indicated in the
Experimental Section, the docking was performed with a model
built by homology with Bacillus cereus (1UOK.pdb) R-glucosi-
dase, which possessed the highest sequence identity (38.9%)
with R-glucosidase from yeast,31 thus conserving the catalytic
residues in most R-glucosidases (His111, Asp205, Glu276,
His348, and Asp349). The docking study was validated with
acarbose.
In conclusion, the information generated hitherto indicates
that L. porteri extract and compounds have potent antihyper-
glycemic effect. One of its metabolites, (Z)-3-butilidenephtha-
lide (3), displayed significant antihyperglycemic effect by
inhibiting the activity of intestinal and yeast R-glucosidases
In Vivo and in vitro. According to the in vitro studies, this
compound acts as a noncompetitive, therefore allosteric, in-
hibitor of this important enzyme. This type of inhibitor has
several advantages over competitive ones since their effects
lasted for longer periods of time.32 The related compounds,
(Z)-6,60,7,30R-diligustilide (1) and (Z)-ligustilide (2), however,
did not affect R-glucosidases In Vivo but affected glucose
absorption by a mechanism yet to be determined. Finally,
ferulic acid (5), present in high amounts in the extract, has
demonstrated in previous investigations noted stimulatory
effects on insulin secretion In Vivo and in vitro. Therefore,
the extract of L. porteri represents a good phytotherapeutic
agent with active principles having different modes of action. In
addition, (Z)-3-butylidenephthalide (3) is a new R-glucosidase
inhibitor potentially useful for the development of new anti-
diabetic, antiobesity, and antiviral agents. The present study
provides scientific support of the use of L. porteri in Mexican
folk medicine for the treatment of diabetes.
’ EXPERIMENTAL SECTION
Plant Material and Extract. The roots of L. porteri were collected
in Basigochi (Chihuahua, Mexico) in October 2003. A voucher speci-
men (No. 31733) is deposited at the National Herbarium (MEXU),
Mexico City. The air-dried and pulverized (250 g) roots were macerated
with CH2Cl2-MeOH (1:1; 5 L ꢀ 2) at room temperature during 7
days, with subsequent evaporation of the solvent in vacuo to yield 87.5 g
of a yellow residue.
Isolation of Compounds 1-5. The extract (87.5 g) was frac-
tionated by column chromatography on silica gel (1 kg, 70-230 mesh,
Merck) eluting with hexane, hexane with increasing amounts of CH2Cl2
(8:2 f 2:8), CH2Cl2, and CH2Cl2-MeOH (9:1) to yield 10 primary
fractions. Fraction 3 (3 g), eluted with hexane, was further separated by
TLC on silica gel (hexane-CH2Cl2, 8:2) to yield 400 mg of myristicin
(4). Fraction 4 (3.5 g), eluted with hexane-CH2Cl2 (1:1), contained
the major components (Z)-6,60,7,30R-diligustilide (1), (Z)-ligustilide
(2), and (Z)-3-butylidenephthalide (3). Compounds 1-3 were sepa-
rated by extensive TLC on silica gel (hexane-CH2Cl2, 1:1) to yield 5.5,
0.5, and 0.6 g, respectively. HPLC-UV analysis revealed the purity of the
components [silica gel Hibar RT LiChrospher100 RP C-18, 5 μm, 4 ꢀ
250 mm, Merck; ACN-H2O (0.5% AcOH) 20:80 f 100:0, 0-40 min;
flow rate 1 mL/min; detecting at 260 and 280 nm]. The structures of
compounds 1-3 were characterized as (Z)-6,60,7,30R-diligustilide, (Z)-
ligustilide, and (Z)-3-butylidenephthalide by spectroscopic and spectro-
metric analyses and comparison with literature data.16,22 Finally, 7.5 g of
ferulic acid (5) precipitated from fraction 9 (20 g), eluted with CH2Cl2,
which was identical to a reference sample.
Experimental Animals. Male ICR normoglycemic mice, weigh-
ing 20-25 g, 20-25 days old, were obtained from Centro UNAM-
Harlan (Harlan Mꢀexico, SA de CV). Procedures involving animals and
their care were conducted in conformity with the Mexican Official Norm
for Animal Care and Handling (NOM-062-ZOO-1999) and in com-
pliance with international rules on care and use of laboratory animals.
Furthermore, clearance for conducting the studies was obtained from
the Ethics Committee for the Use of Animals in Pharmacological and
318
dx.doi.org/10.1021/np100447a |J. Nat. Prod. 2011, 74, 314–320

Journal of Natural Products
ARTICLE
Toxicological Testing, Facultad de Química, UNAM. For the pharma-
cological studies, groups of six animals were used. All doses are in mg/kg
of body weight. The animals were housed in groups of eight under
standard laboratory conditions (12 h light-dark cycle under controlled
temperature, 22 ( 1 °C) and maintained on a standard pellet diet and
water ad libitum.33
Statistical Analysis. Data are expressed as the means ( SEM of
the number (n = 6) of animals in each group. ANOVA was used to
analyze the changes in blood glucose. Dunnett range posthoc compar-
isons were used to determine the source of significant differences where
appropriate; p e 0.05 was considered statistically significant. Sigma Stat
software was used for statistical analyses.
Preparation of Test Samples. All samples were suspended in
0.05% Tween 80 in saline solution and administered by intragastrical
route. Gly (Sigma-Aldrich Corp., St. Louis, MO, USA) was used as a
hypoglycemic model drug. Acarbose (Sigma-Aldrich) was used as an
antihyperglycemic drug. Glucose and sucrose (Sigma-Aldrich) were
used as carbohydrates to carry out the glucose and sucrose tolerance
tests. Control group mice received only the vehicle (0.05% Tween 80 in
saline solution) in the same volume (0.2 mL of vehicle/10 g of body
weight) by the same route.33
Induction of Experimental Diabetes. Diabetes mellitus type-
II was induced by a single intraperitoneal injection of freshly prepared
STZ (100 mg/kg, Sigma-Aldrich) dissolved in 0.1 M citrate buffer, pH
4.5, in a volume of 10 mL/kg, 15 min after an injection of NAD (40 mg/
kg; Sigma-Aldrich) dissolved in distilled water, in a volume of 10 mL/kg.
After 7 days of STZ administration, blood glucose levels of each mouse
were determined. Mice with blood glucose levels higher than 140 mg/dL
were considered diabetic and included in the study.33
Collection of Blood Samples and Determination of Blood
Glucose Levels. Blood samples were collected from the caudal vein
by means of a small incision at the end of the tail. Blood glucose levels
(mg/dL) were estimated by the enzymatic glucose oxidase method using a
commercial glucometer (One Touch Ultra 2, Johnson & Johnson, CA,
USA). The percentage variation of glycemia for each group was
calculated with respect to the initial (0 h) level, according to the
following equation, where Gi is initial glycemia value and Gt is the
glycemia value after treatment administration:
In Vitro r-Glucosidase Inhibition Study. The R-glucosidase
inhibition assay was performed according to Oki et al., with modifica-
tions of our own.29 R-Glucosidase (EC 3.2.1.20) from baker’s yeast was
purchased from Sigma-Aldrich. The inhibition was measured spectro-
photometrically at pH 7.0 and 37 °C using 0.25 mM p-nitrophenyl-R-D-
glucopyranoside (pNPG) as a substrate and 0.6 units/mL of enzyme, in
100 mM sodium phosphate buffer. Acarbose was dissolved in phosphate
buffer, and serial dilutions from 0.1 to 1.0 mg/mL were made and used as
positive control. The increments in absorption at 405 nm due to the
hydrolysis of pNPG by R-glucosidase were determined spectrophoto-
metrically on a microplate reader (Bio Instruments Incorporated, USA).
A 10 μL amount of acarbose or 3-(Z)-butylidenephthalide (3) solution
was incubated for 5 min with 20 μL of enzyme stock. After incubation, 10
μL of substrate (5 mM) was added and further incubated for 35 min at
30 °C. The concentration required to inhibit activity of the enzyme by
50% (IC50) was calculated by regression analysis, using the following
equation, where v is the percentage of inhibition, A100 is the maximum
inhibition, I is the inhibitor concentration, IC50 is the concentration
required to inhibit activity of the enzyme by 50%, and s is the cooperative
degree:
A100
v ¼
S
1 þ ðI=IC50Þ
Kinetics of Enzyme Inhibition. Kinetic parameters were deter-
mined using the Dixon plot method at increasing concentrations of
substrates and inhibitors, plotting slopes versus the reciprocal of the
substrate concentration. All results are expressed as the mean of at least
six experiments ( EM. Kinetic parameters such as Vm, Km, and Ki were
evaluated using the nonlinear regression method based on the following
inhibition equation, where v is the initial velocity in the absence and
presence of the inhibitor, s and I respectively are the concentration of
substrate and inhibitor, Vm is the maximum velocity, Km is the
Michaelis-Menten constant, Ki is the competitive inhibition constant,
and K0i is the uncompetitive inhibition constant:
% Variation of glycemia ¼ ½ðGt - GiÞ=Giꢁ ꢀ 100
Acute Hypoglycemic Assay in Normal and Diabetic Ani-
mals. Normal or NAD-STZ-diabetic mice, placed in acrylic boxes in
groups of six animals, were deprived of food for 12 h before experi-
mentation, but allowed free access to tap water throughout. The crude
extract of L. porteri was freshly prepared before experimentation and
tested at doses of 56.2, 100, and 316.2 mg/kg. Blood samples were
collected at 0, 1.5, 3, 5, 7, and 9 h after treatment administration. Gly was
used as positive control.33
Oral Glucose Tolerance Test (OGTT). Normal or NAD-STZ-
diabetic mice were placed in groups of six animals each. Mice were put
on fast for 12 h before the experiment, with water ad libitum. In the
normal and diabetic animals, the crude extract of L. porteri was tested at
doses of 56.2, 100, and 316 mg/kg. Compound 1 was assayed at doses of
31.2 and 56.2 mg/kg in the case of normal mice, whereas diabetic
animals were tested at doses of 10, 31.2, and 56.2 mg/kg. Compounds 2
and 3 were tested at doses of 10 and 20 mg/kg in normal animals, and in
diabetic animals they were tested at doses of 10, 31.2, and 56.2 mg/kg.
Gly (10 mg/kg) was suspended in the same vehicle. Time 0 min was set
before treatment with the extract or compounds; 30 min later a glucose
load (1 g/kg) was given to the animals. Blood samples were obtained 30,
60, 90, 120, and 180 min after the carbohydrate load.33
VmS
v ¼
I
I
Kmð1 þ =Ki Þ þ sð1 þ =Kir Þ
Homology Modeling of Yeast r-Glucosidase. Although the
X-ray crystal structures of a few bacterial R-glucosidases have been
reported, structural information is still unavailable for most eukaryotic
R-glucosidases commonly used in biological assays, such as that from
baker’s yeast. Therefore, we carried out homology modeling of R-
glucosidase from baker’s yeast to obtain its structure. This homology
modeling started with the retrieval of the amino acid sequence of R-
glucosidase MAL12 (EC 3.2.1.20) from baker’s yeast that comprises 584
amino acid residues from the UniProt (http://www.uniprot.org/uni-
prot/). Using the program CPH models (automated neural-network
based protein modeling server; http://www.cbs.dtu.dk/services/CPHmo-
dels/), the structural model was built based on the structure of oligo-1,6-
glucosidase from Bacillus cereus (1UOK.pdb),34 which revealed the highest
sequence identity (38.9%) with the target. This model was then opti-
mized geometrically with the program HyperChem 8 release, with the
intention of correcting all the links and angles of the model and validated
using the ProCheck (stereochemical quality analysis software),31 Verify-3D
(structure evaluation software),35 and Whatcheck (protein verification
tools software)36 computer programs.
Oral Sucrose Tolerance Test (OSTT). The assay was performed
exactly as described for the OGTT but using sucrose as the carbohy-
drate. In this experiment only the extract was tested in both normal and
diabetic animals. In the case of normal mice the doses were 31.2, 56.2,
and 100 mg/kg. For diabetic animals, however, the extract was tested at a
single dose of 100 mg/kg. Finally, compound 3 was assayed at doses of
10 and 56.2 mg/kg, but only in diabetic animals. Acarbose was used as
positive control.
319
dx.doi.org/10.1021/np100447a |J. Nat. Prod. 2011, 74, 314–320

Journal of Natural Products
ARTICLE
Docking of r-Glucosidase Inhibitor. Blind docking was car-
ried out with AutoDock4 software (http://autodock.scripps.edu/)37,38
using the default parameters, except for the number of GA runs (100)
and the Lamarkian genetic algorithm with local search and 25 million
energy evaluations (Long. Evals.) per run, on a parallel supercomputer of
distributed memory; it contains 1368 processors AMD Opteron, around
3 terabytes of memory, and 160 terabytes of storage (http://www.super.
unam.mx/). The target in each docking run was obtained from homol-
ogy modeling of R-glucosidase. The starting conformation of the ligand
was an energy-minimized form with geometric optimization using the
program HyperChem 8 release. Acarbose and 3-(Z)-butylidenephtha-
lide (3) molecules were first prepared by AutoDockTools 1.5.4 (http://
mgltools.scripps.edu/), adding polar hydrogen atoms and Kollman
charges to R-glucosidase and computing Gasteiger charges for acarbose
and 3-(Z)-butylidenephthalide (3). The protein was held rigid during
the docking process while the ligands were allowed to be flexible. The
grid box size was 126 Å ꢀ 126 Å ꢀ 126 Å in the x, y, and z dimensions,
with the center of the grid corresponding to the protein.
(8) Cegiela-Carlioz, P.; Bessiere, J. M.; David, B.; Mariotte, A.
M.; Gibbons, S.; Dijoux-Franca, M. G. Flavour Fragrance J. 2005,
20, 671–675.
(9) Daniels, A. L.; Van Slambrouck, S.; Lee, R. K.; Arguello, T.
S.; Browning, J.; Pullin, M. J.; Kornienko, A.; Steelant, W. F. Oncol. Rep.
2006, 15, 1327–1331.
(10) Deciga-Campos, M.; Gonzalez-Trujano, E.; Navarrete, A.;
Mata, R. Proc. West. Pharmacol. Soc. 2005, 48, 70–72.
(11) Lorke, D. Arch. Toxicol. 1983, 54, 275–287.
(12) Ames, B. N.; Durston, W. E.; Yamasaki, E.; Lee, F. D. Proc. Natl.
Acad. Sci. U. S. A. 1973, 70, 2281–2285.
(13) Deciga-Campos, M.; Rivero-Cruz, I.; Arriaga-Alba, M.;
Castaneda-Corral, G.; Angeles-Lopez, G. E.; Navarrete, A.; Mata, R.
J. Ethnopharmacol. 2007, 110, 334–342.
(14) Delgado, G.; Garcia, P. E.; Roldan, R. I.; Bye, R.; Linares, E. Nat.
Prod. Lett. 1996, 8, 145–150.
(15) Delgado, G.; Reza-Garduno, R. G.; Rios, M. Y.; Del Rio, F.
Planta Med. 1992, 58, 570–571.
(16) Delgado, G.; Rezagarduno, R. G.; Toscano, R. A.; Bye,
R.; Linares, E. Heterocycles 1988, 27, 1305–1312.
(17) Gillespie, S. G.; Duszynski, J. N. Planta Med. 1998, 64, 392–392.
(18) Huang, Y.; Pu, F. Zhongguo Zhong Yao Za 1990, 15, 422–423.
(19) Zschocke, S.; Liu, J. H.; Stuppner, H.; Bauer, R. Phytochem.
Anal. 1998, 9, 283–290.
(20) Masiello, P.; Broca, C.; Gross, R.; Roye, M.; Manteghetti, M.;
Hillaire-Buys, D.; Novelli, M.; Ribes, G. Diabetes 1998, 47, 224–229.
(21) Duran, T. A.; Derojas, J. F. Diabetes 1973, 22, 318–318.
(22) Beck, J. J.; Stermitz, F. R. J. Nat. Prod. 1995, 58, 1047–1055.
(23) Gijbels, M. J. M.; Scheffer, J. J. C.; Svendsen, A. B. Planta Med.
1982, 44, 207–211.
(24) Prabhakar, P. K.; Doble, M. Phytomedicine 2009, 16, 1119–1126.
(25) Nomura, E.; Kashiwada, A.; Hosoda, A.; Nakamura, K.;
Morishita, H.; Tsuno, T.; Taniguchi, H. Bioorg. Med. Chem. 2003,
11, 3807–3813.
(26) Ohnishi, M.; Matuo, T.; Tsuno, T.; Hosoda, A.; Nomura, E.;
Taniguchi, H.; Sasaki, H.; Morishita, H. Biofactors 2004, 21, 315–319.
(27) Adisakwattana, S.; Moonsan, P.; Yibchok-Anun, S. J. Agric. Food
Chem. 2008, 56, 7838–7844.
’ ASSOCIATED CONTENT
S
Supporting Information. Figure S1 and the Tables
b
S1-S5 are available free of charge via the Internet at http://
pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*Tel: 525-556225289. Fax: 525-556225329. E-mail: rachel@
servidor.unam.mx.
Notes
§This paper is taken in part from the Ph.D. thesis of F.B.
(28) Kono, Y.; Yamazaki, Y.; Maruyama, S. JP patent JKXXAF JP
2007261993 A, 2007.
’ ACKNOWLEDGMENT
This work was supported by grants from DGAPA-UNAM
(IN218110-3) and CONACyT (99395). We thank G. Duarte,
M. Guzmꢀan, N. Esturau, M. Gutierrez, I. Rivero, and A. Pꢀerez for
their valuable technical assistance. F.B. acknowledges a fellow-
ship from CONACYT; M.G.-A. acknowledges a postdoctoral
fellowship by DGAPA-UNAM. We are indebted to Direcciꢀon
General de Servicios de Cꢀomputo Acadꢀemico (DGSCA), UNAM,
for providing the resources to carry out computational calcula-
tions through the KanBalam System.
(29) Oki, T.; Matsui, T.; Osajima, Y. J. Agric. Food Chem. 1999,
47, 550–553.
(30) Kim, M. J.; Lee, S. B.; Lee, H. S.; Lee, S. Y.; Baek, J. S.; Kim, D.;
Moon, T. W.; Robyt, J. F.; Park, K. H. Arch. Biochem. Biophys. 1999,
371, 277–283.
(31) Laskowski, R. A.; MacArthur, M. W.; Moss, D. S.; Thornton,
J. M. J. Appl. Crystallogr. 1993, 26, 283–291.
(32) Christopoulos, A. Nat. Rev. Drug Discovery 2002, 1, 198–210.
(33) Cristians, S.; Guerrero-Analco, J. A.; Perez-Vasquez, A.;
Palacios-Espinosa, F.; Ciangherotti, C.; Bye, R.; Mata, R. J. Nat. Prod.
2009, 72, 408–413.
(34) Watanabe, K.; Hata, Y.; Kizaki, H.; Katsube, Y.; Suzuki, Y.
J. Mol. Biol. 1997, 269, 142–153.
(35) Luthy, R.; Bowie, J. U.; Eisenberg, D. Nature 1992, 356, 83–85.
(36) Hooft, R. W.; Vriend, G.; Sander, C.; Abola, E. E. Nature 1996,
381, 272.
’ DEDICATION
This paper is dedicated to Dr. Koji Nakanishi from Columbia
University for his pioneering work on bioactive natural products.
(37) Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart,
W. E.; Belew, R. K. J. Comput. Chem. 1998, 19, 1639–1662.
(38) Huey, R.; Morris, G. M.; Olson, A. J.; Goodsell, D. S. J. Comput.
Chem. 2007, 28, 1145–1152.
’ REFERENCES
(1) Wild, S.; Roglic, G.; Green, A.; Sicree, R.; King, H. Diabetes Care
2004, 27, 1047–1053.
(2) Kerner, W.; Bruckel, J. Diabetol. Stoffwechs. 2009, 4, S115–S117.
(3) Crepaldi, G.; Carruba, M.; Comaschi, M.; Del Prato, S.; Frajese,
G.; Paolisso, G. J. Endocrinol. Invest. 2007, 30, 610–614.
(4) Andrade-Cetto, A.; Heinrich, M. J. Ethnopharmacol. 2005,
99, 325–348.
(5) Bye, R. A. Econ. Bot. 1986, 40, 103–124.
(6) Linares, E.; Bye, R. A., Jr. J. Ethnopharmacol. 1987, 19, 153–183.
(7) Appelt, G. J. Ethnopharmacol. 1985, 13, 51–55.
320
dx.doi.org/10.1021/np100447a |J. Nat. Prod. 2011, 74, 314–320