PHYTOTHERAPY RESEARCH

Phytother. Res. 26: 317–324 (2012)

Published online 17 November 2011 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/ptr.3664

REVIEW
A Review of the Efficacy and Safety of Banaba
(Lagerstroemia speciosa L.) and Corosolic Acid

Sidney J. Stohs,1* Howard Miller2 and Gilbert R. Kaats3

1
Creighton University, Medical Center, Omaha, NE 68178, USA
2
Nutratech Inc., West Caldwell NJ 07006, USA
3
Integrative Health Technologies, San Antonio TX 78209, USA

Banaba (Lagerstroemia speciosa L.) extracts have been used for many years in folk medicine to treat diabetes,
with the first published research study being reported in 1940. This review summarizes the current literature
regarding banaba and its constituents. The hypoglycemic effects of banaba have been attributed to both corosolic
acid as well as ellagitannins. Studies have been conducted in various animal models, human subjects and in vitro
systems using water soluble banaba leaf extracts, corosolic acid-standardized extracts, and purified corosolic acid
and ellagitannins. Pure corosolic acid has been reported to decrease blood sugar levels within 60 min in human
subjects. Corosolic acid also exhibits antihyperlipidemic, antioxidant, antiinflammatory, antifungal, antiviral,
antineoplastic and osteoblastic activities. The beneficial effects of banaba and corosolic acid with respect to
various aspects of glucose and lipid metabolism appear to involve multiple mechanisms, including enhanced
cellular uptake of glucose, impaired hydrolysis of sucrose and starches, decreased gluconeogenesis and the regula-
tion of lipid metabolism. These effects may be mediated by PPAR, MAP K, NF-kB and other signal transduction
factors. No adverse effects have been observed or reported in animal studies or controlled human clinical trials.
Banaba extract, corosolic acid and other constituents may be beneficial in addressing the symptoms associated with
metabolic syndrome, as well as offering other health benefits. Copyright © 2011 John Wiley & Sons, Ltd.
Keywords: banaba; Lagerstroemia speciosa L.; corosolic acid; ellagitannins; hypoglycemic; safety; efficacy.

et al., 2009). Many of these plants are native to Asia,

INTRODUCTION
although corosolic acid has also been isolated from
European and South American plants. A discussion of
Banaba (Lagerstroemia speciosa L., crepe myrtle) has the pharmacological effects of these plant species is
been used as a folk medicine to treat diabetes in various beyond the scope of this review.
parts of the world, primarily Southeast Asia. The This review summarizes studies that have been
hypoglycemic affect of aqueous (hot water) and methanol conducted in animals, humans and in vitro systems on
extracts have been demonstrated in several animal models the antihyperglycemic, antihyperlipidemic, antioxidant,
as well as a number of human studies. Most studies have antiinflammatory, antifungal, antiviral and antineoplastic
focused on corosolic acid (Fig. 1) which is isolated with activities of banaba extracts, corosolic acid-standardized
an organic solvent from the leaves of the plant, and banaba extracts, and isolated and structurally character-
corosolic acid is used to standardize banaba extracts ized corosolic acid and ellagitannins. Safety and mech-
(Ulbricht et al., 2007; Park and Lee, 2011). Some studies anistic studies conducted to date on these various
indicate that ellagitannins in water soluble fractions may preparations are also summarized.
be responsible for at least some of the insulin-like activity
of banaba, and the antioxidant, antiinflammatory and
glucose regulatory properties of tannins in general have
been reviewed by Klein et al. (2007). HUMAN STUDIES
Corosolic acid has also been isolated from a number of
other plant species including but not limited to Vaccinium
macrocarpon (cranberry) (Kim et al., 2011), Ugni molinae A human clinical study of banaba was reported by Ikeda
(Aguirre et al., 2006), Eriobotrya japonica (Hu et al., 2006; et al. (1999). A proprietary product called Banabamin in
Hou et al., 2009; Lu et al., 2008, 2009; Rollinger et al., tablet form containing an aqueous extract of banaba was
2010), Perilla frutescens (Banno et al., 2004), Campsis used. This product also contained extracts of green tea,
grandiflora (Kim et al., 2005), Symplocos paniculata green coffee and Garcinia. Twenty-four human subjects
(Na et al., 2006), Weigela subsessilis (Thuong et al., 2006; with mild type 2 diabetes were given three tablets three
Lee and Thuong, 2010), Glechoma longituba (Yang times daily. A 13.5% average decrease in blood glucose
et al., 2006), Potentilla chinensis (Shen et al., 2006), Rubus levels was reported, and no adverse effects were
bioflorus (Kang et al., 2008) and Phlomis umbrosa (Liu observed. The constituents in the product responsible
for the antidiabetic effect were not determined.
* Correspondence to: Dr Sidney J. Stohs, 4967 Stillwater Trail, Frisco,
Ikeda et al. (2002) also conducted a 1 year open label
TX 75034, USA. safety and efficacy study on 15 subjects, administering
E-mail: sstohs@yahoo.com 100 mg tablets daily of a water soluble banaba extract.
Received 03 May 2011
Revised 29 August 2011
Copyright © 2011 John Wiley & Sons, Ltd. Accepted 29 August 2011
318 S. J. STOHS ET AL.

not clear if the effect was due entirely to the corosolic

acid or a combination of the corosolic acid with tannin
components.
In an unpublished study by Xu (2008. Action of helping
lower blood glucose level-clinical test. Chinese Center for
Disease Control and Prevention, Beijing Hospital) using
the same soft gel product containing 10 mg corosolic acid
as described by Tsuchibe et al. (2006), 100 subjects, with
Figure 1. Corosolic acid. pre-diabetes or type 2 diabetes were enrolled. Half the
subjects were given one soft gel containing the corosolic
acid-standardized banaba extract and the other half
The extract was not standardized, and the constituent(s) received a placebo for 30 days. Both fasting and 2 h post-
responsible for the antidiabetic effects was not deter- prandial blood glucose levels in the treated group
mined. However, due to the low aqueous solubility of decreased by 10% relative to the control (placebo) group.
corosolic acid, the amount of corosolic acid in the water Also reported was an improvement in diabetic symptoms
soluble preparation that was used would be expected to including a decrease in thirst, drowsiness and hunger.
be low, and the hypoglycemic effects of the product may Furthermore no adverse effects were observed, with no
have been primarily due to water soluble ellagitannins. changes in blood pressure, liver or kidney function, blood
A significant decrease (16.6%) in fasting blood glucose cell count or hemoglobin.
levels was observed in individuals with fasting blood glu- Fukushima et al. (2006) published a study involving 31
cose levels greater than 110 mg/dL (Ikeda et al., 2002). subjects in a double blind cross-over design that were
After both 6 months and 1 year, significant improvements given a capsule containing 10 mg corosolic acid or a pla-
were observed with respect to glucose tolerance and cebo 5 min before a 75 g oral glucose tolerance test.
glycated albumin following treatment with the banaba Blood glucose levels were measured at 30 min intervals
extract. The banaba extract did not cause hypoglycemia. for 2 h. The authors reported that corosolic acid treat-
No changes in hematological or biochemical characteris- ment resulted in lower blood glucose levels from 60 to
tics and no adverse effects were observed over the 1 year 120 min compared with controls, the difference being sta-
course of the study. tistically significant (p < 0.05) at the 90 min time point.
The antidiabetic activity of a banaba extract standar- According to the authors, the corosolic acid that was
dized to 1% corosolic acid in a soft gel capsule formula- used was 99% pure, thus indicating that the blood sugar
tion has been examined (Judy et al., 2003). Ten type 2 lowering effect was due specifically to the corosolic acid.
diabetic subjects were given 32 mg or 48 mg of the prod- A single report has suggested that corosolic acid may
uct (0.32 and 0.48 mg corosolic acid, respectively) daily have been involved in nephrotoxicity and lactic acidosis
for 2 weeks. A 30% decrease in blood glucose levels in a diabetic patient with impaired kidney function who
was reported after the 2 weeks. It is not clear whether was also taking diclofenac for joint pain (Zheng et al.,
the observed effect was due to corosolic acid, the tannin 2010). Diclofenac is a non-steroidal antiinflammatory
components or a combination thereof. drug, and this class of drugs is known to cause renal
A 12 week lifestyle intervention study involving 56 damage and failure. The role of corosolic acid, if any, is
subjects was conducted by Lieberman et al. (2005) that not clear. The use of a drug known for its nephrotoxicity
included the use of dietary supplements, diet and exer- in conjunction with impaired kidney function readily
cise. The dietary supplements were taken prior to each explains the resulting kidney failure. The ability of coro-
meal, and contained 16 mg banaba extract, 100 mg bitter solic acid to inhibit gluconeogenesis could favor lactic acid
melon extract, 133 mg Gymnema extract, 1500 mg production. If corosolic acid impaired the metabolism of
Garcinia cambogia extract (60% hydroxycitric acid), diclofenac, it could theoretically have exacerbated the
2.6 mg bioperine (from black pepper), 10 mg wheat known nephrotoxicity of the drug. No evidence was
amylase inhibitor, 167 mcg elemental chromium, 50 provided to specifically demonstrate this possible effect,
mcg elemental vanadium and 50 mg elemental magne- and no controlled clinical studies have reported nephro-
sium. At the end of 12 weeks, the subjects had lost an toxicity in diabetic subjects receiving corosolic acid.
average of 6.29 kg (13.8 lb) including 3.72 kg (8.2 lb) The above clinical studies demonstrate that banaba
body fat as determined by a bioelectric impedance body extract, banaba extract standardized to corosolic acid
fat analyser. Approximately 73% of subjects completed and corosolic acid itself decrease fasting as well as post-
the study. The amount of corosolic acid and ellagitan- prandial blood glucose levels in humans. A decrease in
nins in the banaba extract was not reported, and it is blood glucose levels has been observed within 2 h of
not clear what contribution to the weight loss was dosing, and the decrease is typically in the range of
provided by each constituent in the product. 10–15%, although a decrease of 30% has been reported.
In a study published by Tsuchibe et al. (2006), 12 non- No adverse effects have been observed or reported in
diabetic subjects with a baseline blood glucose level of any studies involving human subjects receiving banaba,
104 mg/dL were given a soft gel capsule daily for including one study involving 15 subjects who were
2 weeks containing 10 mg corosolic acid as a banaba given banaba extract daily for up to 1 year.
extract standardized to 18% corosolic acid. A 12%
decrease in fasting as well as 60 min postprandial blood
glucose levels was observed after 2 weeks of administer-
ing the product. The authors also reported an average ANIMAL STUDIES
three pound weight loss after the 2 weeks. No adverse
effects were observed during or after the trial. Although The first published research on the hypoglycemic,
this product contained a high level of corosolic acid, it is insulin-like activity of banaba was reported by Garcia
Copyright © 2011 John Wiley & Sons, Ltd. Phytother. Res. 26: 317–324 (2012)
 EFFICACY AND SAFETY OF BANABA EXTRACTS AND COROSOLIC ACID 319

(1940, 1941). An aqueous extract equivalent to 1–2 g of in the diet increased the expression of peroxisome pro-
dried leaves per kg body weight given orally to rabbits liferator-activated receptor-alpha (PPARa) in the liver
lowered blood sugar for 4–6 h. and PPARg in white adipose tissues, thus providing a
Various animal studies have subsequently shown that mechanistic explanation for the loss in body weight and
banaba extracts of unknown composition, banaba the decrease in hepatic steatosis in these mice. These
extracts standardized to corosolic acid, and highly puri- results indicate that corosolic acid may be beneficial in
fied corosolic acid exert beneficial effects with respect addressing various aspects of the metabolic syndrome
to blood glucose and lipid regulation. Kakuda et al. that consists of hyperlipidemia, obesity, hypertension
(1996) fed genetically diabetic (KK-AY) mice diets con- and insulin resistance. The aspects of metabolic syndrome
taining 5% of a hot water extract and 2% of a methanol which may be addressed by corosolic acid include
extract of banaba leaves for 5 weeks. The elevation of obesity, insulin resistance and hypertriglyceridemia and
blood glucose levels was significantly suppressed by hypercholesterolemia.
feeding either of the two extracts. The levels of serum Yamada et al. (2008a) also investigated the mechanism
insulin, plasma total cholesterol and amount of urinary of action of corosolic acid on gluconeogenesis in rat
glucose were all lowered by feeding the extracts, with liver using perfused livers and isolated hepatocytes.
somewhat greater activity with the water extract that Corosolic acid (20–100 mM) in a dose-dependent manner
was given at a higher concentration. The specific consti- decreased gluconeogenesis by increasing production of
tuents in the extracts responsible for these effects were fructose-2,6 diphosphate by lowering cyclic AMP levels
not determined. and inhibiting protein kinase A activity. In addition,
A hot water extract of banaba leaves suppressed corosolic acid increased glucokinase activity without
blood glucose elevation following starch administration affecting glucose-6-phosphatase activity, suggesting an
but not after glucose administration to rats (Suzuki increase in glycolysis. The results provide additional mech-
et al., 2001). These investigators demonstrated that the anistic information regarding the antidiabetic actions of
extract inhibited the activities of various hydrolytic corosolic acid.
enzymes including a-amylase, glucoamylase, isomaltase, Deocaris et al. (2005) administered various banaba
maltase and sucrase. The constituents in banaba respon- leaf extracts, of unknown and unstandardized compos-
sible for these enzyme inhibitory activities were not ition prepared with 80% ethanol, subcutaneously to al-
determined. loxan-induced diabetic mice. Banaba leaf extract had
Yamaguchi et al. (2006) fed 0.072% corosolic acid in the minimal effects on blood glucose levels, but when com-
diet to spontaneously hypertensive rats for 14 weeks. The bined with insulin, the activity was synergistically
investigators reported a significant decrease in blood enhanced. Gamma irradiation of banaba leaves led to
pressure, serum free fatty acids and oxidative stress extracts with higher hypoglycemic activity when mixed
markers relative to the diet containing no corosolic acid. with insulin than unirradiated extracts. Irradiation
However, they observed no effect of the corosolic acid appeared to lead to improved extraction efficiency of
in the diet on body weight gain or blood glucose levels. the active component (s).
A study by Matsuura et al. (2004) examined the abilities A study was conducted involving genetically diabetic
of various teas (aqueous decoctions), including from (db/db) mice fed a diet supplemented with a water
banaba leaves, to suppress the elevation of blood glucose soluble extract of banaba at a 0.5% concentration (Park
in rats from continuous intragastric infusion of sucrose or et al., 2005). At the end of 12 weeks, animals receiving
maltose. Banaba had no significant effect on glucose the banaba extract exhibited significantly reduced blood
levels. The composition of the banaba extract was not glucose, insulin, triglycerides and hemoglobin A1C
reported. The reason for the lack of effect of banaba on levels. Furthermore, increased expressions of liver
blood glucose levels in these two rat studies is not known. PPAR-a mRNA and lipoprotein lipase (LPL) mRNA
These results are in sharp contrast to the results presented as well as adipose tissue PPAR-g mRNA were observed.
below that primarily involved mouse studies as well as The results suggest that the banaba extract increased
streptozotocin-induced diabetic rats, and the human insulin sensitivity and blood sugar regulation by regulat-
studies reported above. ing PPAR-mediated lipid metabolism. However, the
In a study involving genetically diabetic mice given an identity of the responsible factor(s) in banaba was not
extract of banaba (0.8 mg/kg body weight) for 12 weeks determined.
orally, no effect was observed on fasting blood sugar Miura et al. (2004) conducted several studies in genet-
levels, hemoglobin A1C content, body weight or insulin ically diabetic (KK-AY) mice to which corosolic acid
levels (Hong and Maeng, 2004). However, kidney was administered. At a single dose of 10 mg/kg, coroso-
glucose-6-phosphatase activity was significantly lower lic acid significantly reduced blood sugar levels. This ef-
than in control animals. The most likely explanation fect was shown to be associated with an increase in the
for the poor antidiabetic effect is that the dose of the muscle glucose transporter (GLUT4). In a subsequent
extract was too low. In addition, the extract had not study, they showed that a single dose of 2 mg/kg coroso-
been standardized or analysed with respect to potential lic acid reduced blood sugar levels for up to 2 weeks
active constituents. (Miura et al., 2006), supporting the hypothesis that
Yamada et al. (2008b) examined the effects of feeding corosolic acid improves glucose metabolism by reducing
mice a high fat diet for 9 weeks with and without 0.023% insulin resistance.
corosolic acid (not a standardized extract of banaba). In a study in mice, Takagi et al. (2008) demonstrated
Corosolic acid treatment reduced fasting plasma levels that an oral dose of 10 mg/kg corosolic acid suspended
of glucose, insulin and triglycerides by 23%, 41% in water inhibited the intestinal hydrolysis of sucrose
and 22%, respectively. A 10% decrease in body weight but not maltose or lactose, thereby at least in part facili-
and a 15% loss in fat total mass were also observed rela- tating the lowering of blood glucose levels since sucrose
tive to control animals. In addition, the corosolic acid is a disaccharide composed of glucose plus fructose. In
Copyright © 2011 John Wiley & Sons, Ltd. Phytother. Res. 26: 317–324 (2012)
320 S. J. STOHS ET AL.

this study, a sugar solution was administered 30 min Oleanolic acid is a pentacyclic terpene acid that is
orally after the corosolic acid, and blood samples were structurally related to corosolic acid, has been isolated
drawn at 30, 60 and 120 min. These results agree with from banaba leaves, and exhibits a-glucosidase activity
the observations of Suzuki et al. (2001) who showed that (Hou et al., 2009). Oleanolic acid and insulin were
an extract of banaba leaves inhibited sucrase activity, shown to decrease blood glucose levels in control and
and exerted hypoglycemic effects through multiple streptozotocin-induced diabetic rats given a glucose
mechanisms. load after an 18 h fast (Musabayane et al., 2010).
Takagi et al. (2010) have shown that when genetically Furthermore, daily treatment for 5 weeks with oleanolic
diabetic (KK-Ay) mice are fed a high cholesterol diet acid significantly decreased blood glucose levels in
with and without 0.023% corosolic acid for 10 weeks, diabetic animals with concomitant restoration of hepatic
the corosolic acid significantly decreased blood choles- and muscle glycogen stores to near normal levels, and
terol and liver cholesterol content by 32% and 46%, in combination with insulin provided even greater
respectively. Furthermore, the diabetic mice were given antihyperglycemic activity. Oleanolic acid may act
corosolic acid (10 mg/kg body weight) orally in water via a mechanism distinct from insulin including its
followed by the oral administration of a high cholesterol a-glucosidase activity, and the two may exert a synergistic
cocktail 30 min later. Corosolic acid significantly inhib- effect in the regulation of hyperglycemia.
ited mean blood cholesterol levels 4 h after administra- The antiinflammatory activity of various pentacyclic
tion of the cholesterol relative to control animals. The triterpene acids including corosolic acid was assessed
effect was believed to be due to inhibition of cholesterol by Aguirre et al. (2006) in vivo using a mouse ear assay,
absorption via inhibition of the enzyme cholesterol using arachidonic acid and 12-O-tetradecanoylphorbol-
acyltransferase. 13 acetate as the inflammation-inducing agents. Coroso-
Several studies have examined the effects of aqueous lic acid was effective against both inflammatory agents.
banaba extracts on streptozotocin-induced diabetic rats. In summary, various studies have demonstrated that
In none of these studies were the active constituents banaba extracts as well as corosolic acid significantly
determined. A hot water (75–90  C) extract of banaba decrease blood glucose levels in genetic as well as strep-
leaves was shown to depress the elevated blood glucose tozotocin- and alloxan-induced diabetic animals. Better
levels in streptozotocin-diabetic rats by about 43%, responses appear to occur in mice compared with rats.
while increasing the activity of glucose-6-phosphate In these studies, corosolic acid has also been shown to
dehydrogenase as well as glutathione content, and improve insulin sensitivity, increase cellular uptake of
decreasing the activity of the gluconeogenic enzymes glucose, decrease serum triglycerides and cholesterol,
glucose-6-phosphatase and fructose-1,6-diphosphatase facilitate weight loss and improve oxidative stress
(Saha et al., 2009). The dose of the extract used and markers without production of adverse or toxic effects.
the duration of the study were not reported. The results In addition, the animal studies suggest that banaba
suggest that the hypoglycemic activity occurs through extracts and corosolic acid may have additional health-
suppression of gluconeogenesis and stimulation of enhancing benefits that extend beyond the effects
glucose oxidation via the pentose phosphate pathway. observed to date in human subjects. The animal
Thuppia et al. (2009) prepared an extract of banaba studies also support the safety findings reported in
leaves by boiling for 2 h which was subsequently human studies, and have provided extensive informa-
freeze-dried. The extract was administered orally at tion on the mechanisms of action of banaba extract
doses up to 2000 mg/kg body weight for 12 days to and corosolic acid.
streptozotocin-diabetic rats. A dose of 1000 mg/kg
decreased fasting blood glucose levels by approximately
43% on day 12, which returned to pretreatment levels
by day 3 after cessation of treatment. The banaba IN VITRO STUDIES
extract produced no changes in blood glucose levels in
non-diabetic rats. The high doses of the extract needed The antioxidant and free radical scavenging activities of
to produce a hypoglycemic effect may be reflected in banaba were demonstrated for an aqueous extract in
the extraction procedure where boiling may have in vitro free radical generating systems in a concentra-
destroyed some of the active constituents. tion dependent manner (Unno et al., 1997). The banaba
An aqueous extract (150 mg/kg body weight) given to extract was shown to have potent radical scavenging ac-
streptozotocin-induced diabetic mice for up to 15 days tivity on 1,1-diphenyl-2-picrylhydrazyl (DPPH) radical
significantly decreased not only blood glucose levels and superoxide radicals generated by a hypoxanthine–
but also exhibited a potent antioxidant effect (Saumya xanthine oxidase system. The extract also inhibited lipid
and Basha, 2011). Administration of the extract reduced peroxidation in a rat liver homogenate system. The tan-
streptozotocin-generated reactive oxygen species nin content of the extract was about 37% of dry weight.
(superoxide anion, hydrogen peroxide and nitric acid) The antifungal activities of 29 plant extracts were
by up-regulating superoxide dismutase, catalase and determined against two fungal strains, and a methanol
glutathione-S-transferase activities as well as reduced extract of banaba demonstrated high inhibitory activity
glutathione levels. at a concentration of 10 mg/mL (Sato et al., 2000). A
Administration of a spray-dried extract of banaba hot water extract exhibited lower activity. The banaba
(100 mg/kg body weight) for 28 days by gavage to extract was one of five plant extracts to demonstrate
alloxan-induced diabetic mice resulted in significantly significant antifungal activity. The components respon-
lower blood and urine glucose levels (Tanquilut et al., sible for the antifungal activity and their concentrations
2009). In addition, lower food and fluid intakes as well in the banaba extracts were not determined. Further
as lower body weights were observed in response to studies are required to determine whether banaba is a
the banaba extract. viable source of antifungal agents.
Copyright © 2011 John Wiley & Sons, Ltd. Phytother. Res. 26: 317–324 (2012)
 EFFICACY AND SAFETY OF BANABA EXTRACTS AND COROSOLIC ACID 321

Liu et al. (2001) demonstrated that both water and acid. The a-amylase inhibiting activity of corosolic acid
methanol extracts of banaba stimulated glucose uptake was not determined, and may not have been present in
by 3T3 adipocytes. The extracts also inhibited adipocyte large amounts since water was used to make the initial
differentiation induced by insulin. The active compo- leaf extracts.
nents in these banaba extracts were not identified. The corosolic acid content of banaba leaves, banaba
Tanaka et al. (1992) isolated and structurally charac- methanol extracts and various commercial dosage
terized various ellagitannins from the fruit and leaves forms have been determined using high performance
of banaba. Lagerstannins A and B together with five liquid chromatography (HPLC) as well as high
known tannins including lagerstroemin were isolated performance thin layer chromatography (HPTLC) by
from the fruit, while lagerstannin C was isolated and Vijaykumar et al. (2006). Several banaba leaf samples
characterized from banaba leaves. The three lagerstan- were shown to contain 0.31–0.38 mg corosolic acid/
nins possess a gluconic acid core which is rarely found 100 mg, while methanol extracts of leaves contained
in the plants. up to 11.3 mg/100 mg.
Three ellagitannins, lagerstroemin, flosin B and regi- Using a Chinese hamster ovary cell system, Shi et al.
nin were extracted from banaba and shown to increase (2008) demonstrated that corosolic acid stimulated
glucose uptake by isolated rat adipocytes (Hayashi glucose uptake via enhancing insulin receptor phosphor-
et al., 2002). Lagerstroemin exhibited glucose transport ylation. Furthermore, corosolic acid inhibited several
stimulation with a 50% effective concentration (EC50) diabetes-related non-receptor protein tyrosine phos-
of 80 mM, with a maximum effect of approximately phatase enzymes. These studies provide supporting
54% of that of insulin. However, it is doubtful that this and mechanistic information regarding the ability of cor-
concentration of lagerstroemin can be achieved in vivo osolic acid to exert a hypoglycemic effect.
following oral administration of the doses commonly The ability of an aqueous banaba extract to block the
used for banaba extracts. Furthermore, tannic acid, activation of nuclear factor (NF)-kB by tumor necrosis
which is commercially available and widely distributed factor (TNF) in a dose and time dependent manner
in plants, has been shown to exhibit glucose transport was demonstrated using the cardiomyocyte cell line
activity with an EC50 of 17 mM, approximately five times H9c2 (Ichikawa et al., 2010). The authors suggested that
more potent than lagerstroemin (Li et al., 2005). Thus, it this antiinflammatory action might explain the ability of
is not plausible to ascribe the glucose regulatory activity banaba extract to inhibit diabetes-induced cardiomyo-
of banaba extracts specifically to lagerstroemin. cyte hypertrophy. This study provides mechanistic
Subsequent studies with lagerstroemin demonstrated insight into the antiinflammatory activity demonstrated
that it exhibited insulin-like activities including increasing by corosolic acid and other pentacyclic terpene acids in
glucose uptake and decreasing isoproterenol-induced mice (Aguirre et al., 2006).
glycerol release in rat adipocytes, and increasing extracel- Six pentacyclic triterpene acids (oleanolic acid, arju-
lular signal-related kinase (Erk) activity in Chinese nolic acid, asiatic acid, maslinic acid, corosolic acid and
hamster ovary cells (Hattori et al., 2003). It should be 23-hydroxyursolic acid) were isolated from banaba
noted that these studies were conducted in vitro, and leaves by ethyl acetate extraction (Hou et al., 2009),
similar activities have not been demonstrated in animal and their abilities to inhibit a-amlyase and a-glucosidase
or human systems. activities were determined. Corosolic acid exhibited the
More recently, Bai et al. (2008) have isolated and greatest inhibitory activity against a-glucosidase with an
structurally characterized seven ellagitannins and four IC50 of 3.53 mg/mL, while all of the six pentacyclic triter-
methyl ellagic acid derivatives from banaba leaves. A penes exhibited weak or no inhibitory activity against
number of polyphenolic compounds including corosolic a-amylase. These results provide additional information
acid and quercetin were also isolated. The ellagitannins regarding the mechanisms of action of banaba with
all exhibited the ability to stimulate insulin-like glucose respect to its antidiabetic activity.
uptake as well as to inhibit adipocyte differentiation in An interesting study regarding the anabolic effects of
3T3-L1 adipocyte cells in culture. Furthermore, the corosolic acid on osteoblastic bone formation was
methyl ellagic acid derivatives exhibited inhibitory reported by Shim et al. (2009). Concentrations up to
activity with respect to glucose transport. These studies 5 mM corosolic acid significantly stimulated differenti-
clearly demonstrate the antihyperglycemic activity of ation of mouse osteoblasts. This effect was shown to
well characterized ellagic acid derivatives from banaba, be mediated by activation of mitogen activated protein
show that this activity is not restricted to a single com- kinase (MAPK), NF-kB and activator protein-1. These
pound and indicate that multiple mechanisms of action results suggest that corosolic acid may be useful in con-
are involved. junction with bone diseases such as osteoporosis and
Hosoyama et al. (2003) conducted a quantitative periodontitis.
analysis of an a-amylase inhibitor in aqueous banaba Several recent studies have examined the effects of
leaf extracts. The extracts were hydrolysed with hydro- corosolic acid on various human tumor cells. Fujiwara
chloric acid, extracted with an organic solvent and sub- et al. (2011) demonstrated that corosolic acid inhibited
jected to high performance liquid chromatography. the proliferation of glioblastoma cells by suppressing
Using bioassay-guided analysis of various fractions, the activation of signal transducer and activator tran-
the polyphenolic valoneaic acid lactone was isolated scription-3 (STAT3) and NF-kB in both glioblastoma
and identified as an a-amylase inhibitor with an IC50 cells and tumor-associated macrophages. These authors
of approximately 108 mg/mL. However, no standard concluded that corosolic acid may have potential for
a-amylase inhibitor was used. The authors measured tumor prevention and therapy. Xu et al. (2009) showed
the valoneaic acid content of eight banaba leaf decoc- that corosolic acid induced apoptotic cell death in
tions, and showed that the a-amylase inhibiting activities human cervical adenocarcinoma HeLa cells through
were correlated with the content of this polyphenolic the activation of caspases via a mitochondrial pathway.
Copyright © 2011 John Wiley & Sons, Ltd. Phytother. Res. 26: 317–324 (2012)
322 S. J. STOHS ET AL.

They concluded that corosolic acid had strong potential derivatives have also been isolated from banaba and
for clinical application. shown to exhibit antihyperglycemic activity. Further-
Lee et al. (2010a, 2010b) have conducted several stud- more, much information and insight regarding the
ies on the ability of corosolic acid to induce cell cycle ar- mechanisms of action of banaba extracts, corosolic acid
rest and programmed cell death in human gastric cancer and ellagitannins have also been obtained using these
cells. In one gastric cancer cell line, corosolic acid was in vitro systems.
shown to act by activation of AMP-activated protein
kinase and caspase-3, leading to programmed cell death.
In the other report involving another gastric cancer cell
line, corosolic acid dramatically inhibited the expression
of human epidermal growth factor receptor (HER2) SUMMARY AND CONCLUSIONS
oncogene which in turn promotes programmed cell
death. These studies demonstrate the clinical potential A growing body of evidence involving animal and
of corosolic acid in gastric cancers. human studies as well as in vitro systems indicates that
The isoflavonoid compound orobol-7-O-D-glucoside banaba leaf extracts exert antidiabetic and antiobesity
has been isolated from banaba, and was shown to have effects. There is strong evidence to indicate that coroso-
broad spectrum antiviral activity against human rhino- lic acid as well as ellagitannins are responsible for these
viruses (Choi et al., 2010). The 50% inhibitory concen- effects. Other polycyclic terpene acids such as oleanolic
tration (IC50) ranged from 0.58 to 8.80 mg/mL. The acid and valoneaic acid may also contribute to the anti-
authors suggest that this substance may be useful in hyperglycemic effects. With the development of techni-
treating human rhinoviruses which are commonly ques to purity various components of banaba, studies
associated with respiratory infections. are now being conducted with more highly purified
The antinociceptive activity of a chloroform extract of and structurally characterized materials. As a conse-
banaba bark has been assessed in mice, based on folk quence, information is being obtained regarding the
medicinal uses for the treatment of pain (Morshed specific effects of the various constituents, particularly
et al., 2010). A dose of 500 mg of the extract/kg body with respect to corosolic acid.
weight produced a 51% reduction in the pain incidence A number of studies in animals and human subjects
compared with a 38% reduction when 200 mg aspirin/kg using highly purified corosolic acid and corosolic acid-
was administered. The authors did not determine the standardized preparations indicate that this component
identity of the active constituent(s). of banaba exhibits properties that are beneficial in
Cytochrome P450 3A is the most prominent drug addressing various factors involved in glucose regula-
metabolizing enzyme system in the body. Corosolic acid tion and metabolism, including the enhanced cellular
isolated from cranberry was tested for its ability to uptake of glucose, improved insulin sensitivity, decreased
inhibit this enzyme system (Kim et al., 2011) in human gluconeogenesis, and inhibited intestinal hydrolysis of
intestinal microsomes. The concentration of corosolic sucrose, thereby lowering blood glucose levels. Further-
acid needed to provide 50% inhibition (IC50) was about more, decreased serum cholesterol and triglycerides have
4.3–8.8 mM, and the authors concluded that it exhibited been observed in response to corosolic acid.
the potential to contribute to drug–dietary substance Based on the studies conducted to date, no adverse
interactions. However, this concentration of corosolic effects have been reported in animals using either coro-
acid cannot be achieved in the blood with a dose as high solic acid or standardized banaba extracts, nor have
as 10 mg, although at the intestinal level this concentra- adverse events been observed or reported in controlled
tion might be achieved. human clinical studies. However, no animal studies have
Another study that looked at the potential influence been designed specifically to assess toxicity or LD50
of a banaba extract to influence the bioavailability of values for corosolic acid or banaba extracts standardized
drugs was conducted by Nagai et al. (2009). They exam- to specific concentrations of corosolic acid.
ined the abilities of various herbal extracts to inhibit the The above studies indicate that corosolic acid and
activity of human sulfotransferase 1A3 (SULT1A3), a corosolic acid standardized banaba extracts may be
prominent detoxifying enzyme in the intestinal epithe- beneficial in addressing issues associated with elevated
lium. An extract of banaba was shown to inhibit the blood sugar levels and obesity. Furthermore, corosolic
sulfation of dopamine and ritodrine at IC50 concentra- acid exhibits antiinflammatory, antihyperlipidemic,
tions of 16.0 and 7.5 mg/mL, respectively. The authors antiviral and antitumor promoting effects. Standardized
concluded that extracts of banaba and tea may increase banaba extracts, corosolic acid and/or ellagitannins in
the bioavailability of drugs whose bio-availabilities are combination with other ingredients may be useful in
limited by the action of this enzyme in the intestine. dealing with symptoms associated with metabolic
The extracts were non-standardized and of unknown syndrome. Corosolic acid and standardized banaba
composition, and the method of extraction was not extracts may also be highly effective either as stand-
described. The factor or factors responsible for the alone products or in combination with other natural
enzyme inhibition are not known, and the affect of products possessing hypoglycemic, antihyperlipidemic
corosolic acid on this enzyme was not assessed. As and appetite suppressant activities.
a consequence, the clinical significance of this study Additional human efficacy and safety studies are
is unclear. warranted, particularly studies assessing the dose and
In summary, various in vitro studies involving cell-free time dependent effects of corosolic acid or corosolic
systems and cell cultures have demonstrated antioxidant, acid-standardized banaba extracts and ellagitannins
antifungal, antiviral, antineoplastic and osteoblastic alone or in combination with other ingredients on blood
activities of banaba extracts and corosolic acid. In lipids (triglyceride and cholesterol), insulin and glucose
addition, various ellagitannins and methyl ellagic acid levels as well as weight loss and weight management.
Copyright © 2011 John Wiley & Sons, Ltd. Phytother. Res. 26: 317–324 (2012)
 EFFICACY AND SAFETY OF BANABA EXTRACTS AND COROSOLIC ACID 323

Investigations are needed to clearly define and under- Conflict of Interest

stand the roles and importance of corosolic acid and
related pentacyclic terpene acids relative to the ellagi- The SJS has served as a consultant for Nutratech Inc., West Caldwell,
tannins present in banaba. Finally, additional acute NJ, a supplier of banaba extracts. HM is an employee of Nutratech,
and subchronic animal safety studies are needed. Inc. GRK reported no conflicts of interest.

REFERENCES

Aguirre MC, Delporte C, Backhouse N, et al. 2006. Topical anti- speciosa L. leaves in genetically diabetic KK-AY mice. Biosci
inflammatory activity of 2-alpha-hydroxy pentacyclic triter- Biotechnol Biochem 60: 204–208.
pene acids from the leaves of Ugni molinae. Bioorg Med Chem Kang SH, Shi YQ, Yang CX. 2008. Terpenoids and steroids of
14: 5673–5677. root of Rubus biflorus (In Chinese). Zhong Yao Cai 31:
Bai N, He K, Roller M, et al. 2008. Active compounds from 1669–1671.
Lagerstroemia speciosa, insulin-like glucose uptake-stimulatory/ Kim DH, Han KM, Chung IS, et al. 2005. Triterpenoids from the
inhibitory and adipocyte differentiation-inhibitory activities in flower of the Campsis grandiflora K. Schum as human acyl
3T3L1 cells. J Agric Food Chem 56: 11668–11674. CoA:cholesterol acyltransferase inhibitors. Arch Pharmacol
Banno N, Akihisa T, Tokuda H, et al. 2004. Triterpene acids from Res 28: 550–556.
the leaves of Perilla frutescens and their anti-inflammatory Kim E, Sy-Cordero A, Graf TN, Brantley SJ, Paine MF, Oberlies NH.
and anti-tumor promoting effects. Biosci Biotechnol Biochem 2011. Isolation and identification of intestinal CYP3A inhibi-
68: 85–90. tors from cranberry (Vaccinium macrocarpon) using human
Choi HJ, Bae EY, Song JH, Baek SH, Kwon DH. 2010. Inhibitory intestinal microsomes. Planta Med 77: 265–270.
effects of orobol-7-O-D-glucoside from banaba (Lagerstroemia Klein G, Kim J, Himmeldirk K, Cao Y, Chen X. 2007. Antidiabetes
speciosa L.) on human rhinoviruses replication. Lett Appl and anti-obesity activity of Lagerstroemia speciosa. Evid
Microbiol 51: 1–5. Based Complement Alternat Med 4: 401–407.
Deocaris CC, Aguinaldo RR, dela Ysla JL, Asencion AS, Mojica Lee MS, Cha EY, Thuong PT, Kim JY, Ahn MS, Sul JY. 2010a.
ERE. 2005. Hypoglycemic activity of irradiated banaba Down-regulation of human epidermal growth factor receptor
(Lagerstroemia speciosa L.) leaves. J Appl Sci Res 1: 95–98. 2/neu oncogene by corosolic acid induces cell cycle arrest
Fujiwara Y, Komohara Y, Ikeda T, Takeya M. 2011. Corosolic acid and apoptosis in NCI-N87 human gastric cancer cells. Biol
inhibits glioblastoma cell proliferation by suppressing the acti- Pharm Bull 33: 931–937.
vation of signal transducer and activator of transcription-3 and Lee MS, Lee CM, Cha EY, et al. 2010b. Activation of AMP-activated
nuclear-factor kappa B in tumor cells and tumor-associated protein kinase on human gastric cancer cells by apoptosis
macrophages. Cancer Sci 102: 206–211. induced by corosolic acid isolated from Weigela subsessilis.
Fukushima M, Matsuyama F, Ueda N, et al. 2006. Effects of Phytother Res 24: 1857–1861.
corosolic acid on post-challenge plasma glucose levels. Diab Lee MS, Thuong PT. 2010. Stimulation of glucose uptake by triter-
Res Clin Pract 73: 174–177. penoids from Weigela subsessilis. Phytother Res 24: 49–53.
Garcia F. 1940. On the hypoglycemic effect of a decoction of Li Y, Kim J, Li J, et al. 2005. Natural anti-diabetic compound
Lagerstroemia speciosa leaves (banaba) administered orally. 1,2,3,4,6-penta-O-galloyl-D-glucopyranose binds to insulin re-
J Phil Med Assoc 20: 193–201. ceptor and activates insulin-mediated glucose transport sig-
Garcia F. 1941. Distribution and deterioration of insulin-like naling pathway. Biochem Biophys Res Comm 336: 430–437.
principle in Lagerstroemia speciosa (banaba). Acta Med Phil Lieberman S, Spahrs R, Stanton A, Martinez L, Grinder M. 2005.
3: 99–104. Weight loss, body measurements, and compliance: a 12-week
Hattori K, Sukenobu N, Sasaki T, et al. 2003. Activation of insulin total lifestyle intervention pilot study. Altern Compl Ther Dec:
receptors by lagerstroemin. J Pharmacol Sci 93: 69–73. 307–313.
Hayashi T, Maruyama H, Kasai R, et al. 2002. Ellagitannins from Liu F, Kim JK, Li Y, Liu XQ, Li J, Chen X. 2001. An extract of
Lagerstroemia speciosa as activators of glucose transport in Lagerstroemia speciosa L. has insulin-like glucose uptake-
fat cells. Planta Med 68: 173–175. stimulatory and adipocyte differentiation-inhibitory activities
Hong H, Maeng WJ. 2004. Effects of malted barley extract and in 3T3-L1 cells. J Nutr 131: 2242–2247.
banaba extract on blood glucose levels in genetically diabetic Liu P, Deng R, Duan H, Yin W. 2009. Chemical constituents from
mice. J Med Food 7: 487–490. roots of Phlomis umbrosa (In Chinese). Zhongguo Zhong Yao
Hosoyama H, Sugimoto A, Suzuki Y, Sakane I, Kakuda T. 2003. Za Zhi 34: 867–870.
Isolation and quantitative analysis of the a-amylase inhibitor Lu H, Chen J, Li WL, Zhang HQ. 2008. Studies on the triterpenes
in Lagerstroemia speciosa (L.) Pers. (banaba) (In Japanese). from ioquat leaf (Eriobotrya japonica) (In Chinese). Zhang
Yakugaku Zasshi 123: 599–605. Yao Cai 31: 1351–1354.
Hou W, Li Y, Zhang Q, et al. 2009. Triterpene acids isolated from Lu H, Xi C, Chen J, Li W. 2009. Determination of triterpenoid acids
Lagerstroemia speciosa leaves as alpha-glucosidase inhibitors. in leaves of Eriobotrya japonica collected at different seasons
Phytother Res 23: 614–618. (In Chinese). Zhongguo Zhong Yao Za Zhi 34: 2353–2355.
Hu C, Chen L, Xin Y, Cai Q. 2006. Determination of corosolic acid Matsuura T, Yoshikawa Y, Masui H, Sano M. 2004. Suppression of
in Eriobotrya japonica leaves by reversed-phase high perform- glucose absorption by various health teas in rats (In Japanese).
ance liquid chromatography (In Chinese). Se Pu 24: 492–494. Yakugaku Zasshi 124: 217–223.
Ichikawa H, Yagi H, Tanaka T, Cyong JC, Masaki T. 2010. Miura T, Itoh Y, Kaneko T, et al. 2004. Corosolic acid induces
Lagerstroemia speciosa extract inhibit TNF-induced activation GLUT4 translocation in genetically type 2 diabetic mice. Biol
of nuclear factor-kB in rat cardiomyocyte H9c2 cells. J Pharm Bull 27: 1103–1105.
Ethnopharmacol 128: 254–256. Miura T, Ueda N, Yamada K, et al. 2006. Antidiabetic effects of
Ikeda Y, Chen JT, Matsuda T. 1999. Effectiveness and safety of corosolic acid in KK-Ay diabetic mice. Biol Pharm Bull 29:
banabamin tablet containing extract from banaba in patients 585–587.
with mild type 2 diabetes (In Japanese). Jpn Pharmacol Ther Morshed A, Hossain MH, Shakil S, et al. 2010. Evaluation of the
27: 829–835. antinociceptive activity of two Bangladeshi medicinal plants,
Ikeda Y, Noguchi M, Kishi S, et al. 2002. Blood glucose controlling Kalanchoe pinnata (Lam.) Pers. and Lagerstroemia speciosa
effects and safety of single and long-term administration on (L.) Pers. Adv Nat Appl Sci 4: 193–197.
the extract of banaba leaves (In Japanese). J Nutr Food Musabayane CT, Tufts MA, Mapanga RF. 2010. Synergistic
5: 41–53. antihyperglycemic effects between plant-derived oleanolic
Judy WV, Hari SP, Stogsdill WW, Judy JS, Naguib YMA, acid and insulin in streptozotocin-induced diabetic rats. Ren
Passwater R. 2003. Antidiabetic activity of a standardized Fail 32: 832–839.
extract (Glucosol™) from Lagerstroemia speciosa leaves in Na M, Yang S, He L, et al. 2006. Inhibition of protein tyrosine
type II diabetics: A dose-dependence study. J Ethnopharmacol phosphatase 1B by ursane-type triterpenes isolated from
87: 115–117. Symplocos paniculata. Planta Med 72: 261–263.
Kakuda T, Sakane I, Takihara T, Ozaki Y, Takeuchi H, Kuroyangi M. Nagai M, Fukamachi T, Tsujimoto M, et al. 2009. Inhibitory effects
1996. Hypoglycemic effect of extracts of Lagerstroemia of herbal extracts on the activity of human sulfotransferase

Copyright © 2011 John Wiley & Sons, Ltd. Phytother. Res. 26: 317–324 (2012)
324 S. J. STOHS ET AL.

isoform sulfotransferase 1A3 (SULT1A3). Biol Pharm Bull 32: Tanquilut NC, Tanquilut MRC, Estacio MAC, Torres EB, Rosario JC,
105–109. Reyes BAS. 2009. Hypoglycemic effect of Lagerstroemia
Park C, Lee JS. 2011. Banaba: The natural remedy as antidiabetic speciosa L. Pers. on alloxan-induced diabetic mice. J Med
drug. Biomed Res 22: 127–131. Plant Res 3: 1066–1071.
Park MY, Lee KS, Sung MK. 2005. Effects of dietary mulberry, Thuong PT, Min BS, Jin W, et al. 2006. Anti-complementary
Korean red ginseng, and banaba on glucose homeostasis in re- activity of ursane-type triterpenoids from Weigela subsessilis.
lation to PPAR-a, PPAR-g, and LPL mRNA expressions. Life Sci Biol Pharm Bull 29: 830–833.
77: 3344–3354. Thuppia A, Rabintossaporn P, Saenthaweesuk S, Ingkaninan K,
Rollinger JM, Kratschmar DV, Schuster D, et al. 2010. 11-Beta- Sireeratawong S. 2009. The hypoglycemic effect of a water
hydroxysteroid dehydrogenase-1 inhibiting constituents extract from leaves of Lagerstroemia speciosa L. in streptozo-
from Eriobotrya japonica revealed by bioactivity-guided isola- tocin-induced diabetic rats. Songklanakarin J Sci Technol 31:
tion and computational approaches. Bioorg Med Chem 18: 133–137.
1507–1515. Tsuchibe S, Kataumi S, Mori M, Mori H. 2006. An inhibitory effect
Saha BK, Bhuiyan NH, Mazumder K, Haque KMF. 2009. on the increase in the postprandial glucose by banaba extract
Hypoglycemic activity of Lagerstroemia speciosa L. extract capsule enriched corosolic acid. J Integr Study Diet Habits
on streptozotocin-induced diabetic rat: underlying mechanism 17: 255–259.
of action. J Bangladesh Pharmacol Soc 4: 79–83. Ulbricht C, Dam C, Milkin T, Seamon E, Weissner W, Woods J.
Sato J, Goto K, Nanjo F, Kawai S, Murata K. 2000. Antifungal 2007. Banaba (Lagerstroemia speciosa L.): An evidence-
activity of plant extracts against Arthrinium sacchari and based systematic review by the natural Standard Research
Chaetomium funicola. J Biosci Bioeng 90: 442–446. Collaboration. J Herb Pharmacother 7: 99–113.
Saumya SM, Basha PM. 2011. Antioxidant effect of Lagerstroemia Unno T, Sakane I, Masumizu T, Kohno M, Kakuda T. 1997. Antioxi-
speciosa Pers (banaba) leaf extract in streptozotocin-induced dant activity of water extracts of Lagerstroemia speciosa
diabetic mice. Indian J Exp Biol 49: 125–131. leaves. Biosci Biotechnol Biochem 61: 1772–1774.
Shen Y, Wang QH, Lin HW, Shu W, Zhou JB, Li ZY. 2006. Study on Vijaykumar K, Murthy PB, Kannababu S, Syamasundar B,
chemical constituents of Potentilla chinensis Ser (In Chinese). Subbaraju GV. 2006. Quantitative determination of corosolic
Zhong Yao Cai 29: 237–239. acid in Lagerstroemia speciosa leaves, extracts and dosage
Shi L, Zhang W, Zhou YY, et al. 2008. Corosolic acid stimulates forms. Int J Appl Sci Eng 4: 103–114.
glucose uptake via enhancing insulin receptor phosphorylation. Xu Y, Ge R, Du J, et al. 2009. Corosolic acid induces apoptosis
Eur J Pharmacol 584: 21–29. through mitochondrial pathway and caspase activation in
Shim KS, Lee SU, Ryu SY, Min YK, Kim SH. 2009. Corosolic acid human cervix adenocarcinoma HeLa cells. Cancer Lett 284:
stimulates osteoblast differentiation by activating transcrip- 229–237.
tion factors and MAP kinases. Phytother Res 23: 1754–1758. Yamada K, Hogokowa M, Fujimoto S, et al. 2008b. Effect of
Suzuki Y, Hayashi K, Sukabe I, Kakuda T. 2001. Effects and mode corosolic acid on gluconeogenesis in rat liver. Diabetes Res
of action of banaba (Lagerstroemia speciosa L.) leaf extracts Clin Pract 80: 48–55.
on postprandial blood glucose in rats. J Jpn Soc Nutr Food Yamada K, Hosokawa M, Yamada C, et al. 2008a. Dietary corosolic
Sci 54: 131–137. acid ameliorates obesity and hepatic steatosis in KK-Ay mice.
Takagi S, Miura T, Ishibashi C, et al. 2008. Effect of corosolic acid Biol Pharm Bull 31: 651–655.
on the hydrolysis of disaccharides. J Nutr Sci Vitaminol 54: Yamaguchi Y, Yamada K, Yoshikawa N, Nakamura K, Haginaka J,
266–268. Kunitomo M. 2006. Corosolic acid prevents oxidative stress,
Takagi S, Miura T, Ishihara E, Ishida T, Chinzei Y. 2010. Effect of inflammation and hypertension in SHR/HDmcr-cp rats, a
corosolic acid on dietary hypercholesterolemia and hepatic model of metabolic syndrome. Life Sci 79: 2474–2479.
steatosis in KK-Ay diabetic rats. Biomed Res 31: 213–218. Yang NY, Duan JA, Li P, Qian SH. 2006. Chemical constituents
Tanaka T, Tong HH, Xu YM, Ishimaru K, Nonaka G, Nisioka I. 1992. of Glechoma longituba (In Chinese). Yao Xue Xue Bao 41:
Tannins and related compounds. CXVII. Isolation and 431–434.
characterization of three new ellagitannins, Lagerstannins A, Zheng JQ, Zheng CM, Lu KC. 2010. Corosolic acid-induced acute
B, and C, having a gluconic acid core, from Lagerstroemia kidney injury and lactic acidosis in a patient with impaired
speciosa (L.) Pers. Chem Pharm Bull 40: 2975–2980. kidney function. Am J Kidney Dis 56: 419–420.

Copyright © 2011 John Wiley & Sons, Ltd. Phytother. Res. 26: 317–324 (2012)