Journal of Ethnopharmacology 98 (2005) 231–239

Review tridentata ( bush), an abundant of Mexican and US-American deserts and its metabolite nordihydroguaiaretic acid Silvia Arteaga, Adolfo Andrade-Cetto, ReneC´ ardenas´ ∗

Departamento de Biolog´ıa Celular, Facultad de Ciencias, Universidad Nacional Aut´onoma de M´exico, 04510 M´exico, DF, M´exico Received 15 September 2004; accepted 8 February 2005

Abstract

Although controversial, Creosote bush, (Sesse and Moc. ex DC) Coville, is used to treat a variety of illnesses including infertility, rheumatism, arthritis, diabetes, gallbladder and kidney stones, pain and inflammation. Recently, it has been used as a nutritional supplement. The primary product extracted from this common plant of the arid regions of northern Mexico´ and Southwestern United States is the potent antioxidant nordihydroguaiaretic acid (NDGA). It was widely used during the 1950s as a food and to preserve naturals fibers. Later it was banned after reports of toxicity during the early 1960s. Renal and hepatotoxicity are also reported for chronic use of creosote bush and NDGA. This article reviews traditional and contemporary uses and pharmacology, including toxicology of this plant widely used in Mexican traditional medicine. © 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Larrea tridentata; Gobernadora; Creosote bush; Nordihydroguaiaretic acid; medicine; Traditional medicine; Hepatotoxicity; Renal toxicity

Contents

1. Introduction ...... 232 2. Botanical description and distribution ...... 232 3. Phytochemistry ...... 232 4. Ethnobotany and in vitro pharmacology related to traditional uses...... 233 4.1. Reported medicinal uses...... 233 4.2. Other reported uses ...... 234 5. In vivo pharmacology and pharmacodynamics ...... 235 5.1. Reports on Larrea with positive outcomes...... 235 5.2. Side effects and toxicity ...... 235 5.3. Experimental studies on possible additional uses of NDGA ...... 236 5.4. Studies on NDGA toxicity ...... 236 5.5. Hepatic metabolism of creosote bush and NDGA ...... 237 6. Conclusions ...... 237 Acknowledgements...... 237 References ...... 237

∗ Corresponding author. Tel.: +52 55 5622 4833; fax: +52 55 5622 4828. E-mail address: [email protected] (R. Cardenas).´

0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2005.02.002 232 S. Arteaga et al. / Journal of Ethnopharmacology 98 (2005) 231–239

1. Introduction temperatures within its range of distribution, without under- going irreparable damage. However, periodic extreme freezes Creosote bush, Larrea tridentata (Sesse and Moc. ex may contribute to limiting the present distribution of L. tri- DC, ) Coville is a common of North dentata since freezing induces xylem embolism and cavita- American warm deserts. Its dominance has increased within tion (Mart´ınez-Vilalta and Pockman, 2002). Creosote bush is 19 million ha of lands previously considered desert grass- unpalatable to livestock and most wildlife, is usually toxic, lands in response to disturbances such as grazing (Van Auken, sometimes causing (Gay and Dwyer, 1998). Sheep, es- 2000; Whitford et al., 2001). While often viewed as an in- pecially pregnant ewes, have been reported to die after eating dicator of desertified conditions and the focus of extensive the ( State University, 2005). Creosote bush is control efforts (Herbel and Gould, 1995) it is also an impor- abundant in the desert areas of the Mexican states of San tant plant with a long history of medicinal use (Timmermann, Luis Potos´ı, , , , , Zacate- 1977). Among the proposed medicinal properties of creosote cas, Baja Norte and Sur, and in the Southwest bush, the most prominent is its antioxidant effects (Sheikh of the United States in , California, , et al., 1997). The family Zygophyllaceae includes more than and (Rzedowsky and Huerta, 1994). Similar thirty genera and approximately 250 species (Jones, 1987). L. species are found in arid zones of South America, mostly in tridentata is used in a variety of forms. Traditionally leaves Argentina and Bolivia. It has been established that the plant and twigs are used to prepare a tea, but is also used in cap- is of South American origin, with a disjunct distribution (Lia sules and tablets, prepared for oral consumption. In Mexico´ et al., 2001). the tea is used traditionally as a treatment of kidney and gall- bladder stones (Diaz, 1976). Current use is limited by re- ports of toxic hepatitis (Brent, 1999), and a case of cystic 3. Phytochemistry renal disease (Smith, 1994) associated with its chronic use. Hepatic impairment resulting from the use of conventional Larrea tridentata is a notable source of natural products drugs is widely acknowledged, but there is less awareness of with approximately 50% of the leaves dry weight as ex- the potential hepatotoxicity of herbal preparations and other tractable matter. The that covers the leaves yielded 19 botanicals, many of which are believed to be harmless and flavonoid aglycones, as well as several lignans, notably in- are commonly used for self-medication without supervision cluding the antioxidant NDGA (Fig. 1; Konno et al., 1990). (Stickel et al., 2000). Creosote bush is also known as cha- Some glycosylated flavonoids, sapogenins, essential oils, parral and in the United States and gobernadora halogenic alkaloids (Argueta, 1994) and were isolated (governess) and hediondilla (little smelly one) in Mexico.´ from creosote bush (Romo de Vivar, 1985). Larrea tridentata This review initially presents a short botanical description contains about 0.1% of dry weight as volatile oils. Within the of the plant and an overall view of its phytochemical diver- volatile fraction 67 compounds have been identified constitut- sity. Then the traditional and contemporary uses reported are ing more than 90% of the know creosote bush oils; the remain- briefly summarised, together with in vitro studies dealing ing 10% is a mixture of more than 300 constituents, mainly with some medicinal uses. It should be mention that most of monoterpenoids and aromatic sesquiterpenoids (Mabry and the medicinal uses of Larrea tridentata are not supported by Bohnstedt, 1981; Xue et al., 1988). Products from the meval- experimental or clinical studies. In the section on pharmacol- onic, shikimic and fatty acid pathways are predominant. Be- ogy, in vivo studies on the beneficial, which propose possible sides a great number of substances, the vinyl and methyl therapeutic additional uses, and toxic properties of both the ketones contribute significantly to the creosote bush charac- plant and its main metabolite, nordihydroguaiaretic acid are teristic odor. Three common sterols have also been identified: reviewed, as well as their hepatic metabolism. campesterol, stigmasterol and sitosterol, as well as saponins of the C30-ursolic type that represent less than 1% of this species dry weight (Mabry and Bohnstedt, 1981). Alkaloids 2. Botanical description and distribution have been isolated from the bark and , but not from the leaves and flowers (Lara and Marquez,´ 1996). In terms of nat- Larrea tridentata is an evergreen shrub 1–3 m high, ural products chemistry creosote bush is best known by the branched and knotty. Leaves are opposite with two asym- large amount of the lignan NDGA, which is deposited on the metrical leaflets measuring ca. 1 cm in length. Leaves are glossy with a thick resinous coating secreted by a glandular epidermis of the stipules, located on the knots; the steam is woody, knotty and inerm. Flowers are complete and borne solitary in the axils, with five yellow clawed petals. The fruit is a roundish capsule, covered with a dense concentration of white hairs (De la Cerda, 1967; Nellesen, 1997). The plant discharges a penetrant odor and has a bitter flavor. It is peren- nial, retaining most of its leaves across the drought and low Fig. 1. Nordihydroguaiaretic acid structure. S. Arteaga et al. / Journal of Ethnopharmacology 98 (2005) 231–239 233 surface of the leaves (Mabry and Bohnstedt, 1981). It has been Externally Larrea tridentata has medicinal uses as tinc- found that between 5 and 10% of the leaves dry weight con- ture and salves, as and poultice to excoriations, sist of NDGA, 80% of all phenolics in the resin. A study on wounds (Timmermann, 1977; Mabry et al., 1979b; Argueta, the distribution of secondary phenolic compounds reported 1994; Kay, 1996; Lara and Marquez,´ 1996), acne, psoriasis that flowers, leaves, green stems and small woody stems and dandruff (Estudillo and Hinojosa, 1988; Brent, 1999). all contained NDGA, with highest concentrations in leaves It has also been used as salves against burns (Sheikh et (38.3 mg/g) and green stems (32.5 mg/g; Hyder et al., 2002). al., 1997; Brent, 1999), bruises and hemorrhoids, for cica- Due to numerous pharmacological activities of NDGA, a ver- trization (Argueta, 1994), chicken pox (Timmermann, 1977; satile method for its synthesis and that of a variety of deriva- Sheikh et al., 1997), snakebite pain, chronic cutaneous disor- tives was developed (Gezginci and Timmermann, 2001). Re- ders (Nellesen, 1997; Sheikh et al., 1997; Brent, 1999) and cently three new lignans, and six flavonoids and four lignans allergic problems (Brinker, 1993). known compounds of creosote bush were isolated from the Several antibiotic (Mabry et al., 1979a; Argueta, 1994), plant (Abou-Gazar et al., 2004). The epoxylignans with a antifungal (Mabry et al., 1979a; Barragan´ et al., 1994; Brent, tetrahydrofuran moiety showed strong antioxidant activity. 1999) and antiviral properties (Brent, 1999) have been at- NDGA and other of the surface deter herbivory. tributed to creosote bush. Dried is described as one On the other hand, the flavonoids function as antimicrobial of the best herbal antibiotics, being useful against bacteria, agents, and as protection against herbivores, UV radiation viruses and parasites, both internally and externally (Smith, and water loss, and so are potentially important in the species 1994). The alcoholic extract of creosote bush has antifungal success in desertic ambients (Mabry and Bohnstedt, 1981). activity against tested species of Aspergillus, Penicillium and Fusarium (Tequida et al., 2002). Similarly, the ethanolic ex- tract showed good antimicrobial activity against growth of 4. Ethnobotany and in vitro pharmacology related to yeast and some molds and bacteria (Verastegui et al., 1996). traditional uses Among the multiples medicinal properties of chaparral, it has been mentioned as antiamoeban at low doses (Brent, 1999). 4.1. Reported medicinal uses Ethanolic and chloroformic extracts of creosote bush and NDGA showed a marked growth inhibition of Entamoeba Creosote bush has been used in traditional medicine to hystolitica and Entamoeba invadens in culture (Segura, 1978; treat more than 50 illnesses (Table 1). Most common uses Mabry et al., 1979a; Calzado-Flores et al., 1995). are associated to diseases of renal and gynaecologic origins. It is said to possess analgesic and anti-inflammatory The plant is used as aqueous or alcoholic liquid extract of properties, when applied as a poultice of powdered leaves leaves and twigs; in addition, it is available in capsules and (Timmermann, 1977; Argueta, 1994; Kay, 1996; Tyler and tablets for oral use, while leaves and branches can be used Foster, 1999), is helpful in the treatment of neuritis and sci- for poultice and fomentation. Historically, aqueous extracts atica (Timmermann, 1977; Mabry et al., 1979b) and has of the creosote bush have been used by native healers of been employed as a tea to help with cramps (Brinker, 1993), the Southwest region of North America, and is commonly toothache (Brent, 1999) and headache (Argueta, 1994). Cha- referred to as chaparral tea. In Mexico´ it is reported that an parral and nordihydroguaiaretic acid (NDGA) have potent infusion of the leaves dissolves gallbladder and kidney stones anti-inflammatory activity, possibly due to their ability to in- when the tea is consumed throughout the day (“Agua de uso”; hibit the enzyme lipoxygenase in vitro (Bokoch and Reed, Mart´ınez, 1969; Diaz, 1976; Lara and Marquez,´ 1996). It is 1981; Salari et al., 1984; Safayhi et al., 1992). used in the treatment of diseases of the liver, and as a liver Other potential medicinal uses are against arthritis and tonic (Sheikh et al., 1997; Brent, 1999). Creosote bush has rheumatism. The branches are macerated in alcohol and also been employed for kidney pain and cystitis (Mart´ınez, rubbed onto the affected area (Mabry et al., 1979b; Brinker, 1969) dysuria (Mart´ınez, 1969; Lara and Marquez,´ 1996), 1993; Argueta, 1994; Lara and Marquez,´ 1996; Tyler and as a (Mabry et al., 1979b; Tyler and Foster, 1999), Foster, 1999). A tea of the branches is consumed for bowel to treat infections of the urinary tract and venereal diseases cramps and inflammation (Mabry et al., 1979b; Tyler and Fos- (Timmermann, 1977; Mabry et al., 1979b; Brent, 1999). Oral ter, 1999), stomach pain, diarrhea (Argueta, 1994; Sheikh et decoctions and extracts of leaves and twigs have been used al., 1997), ulcer and indigestion (Argueta, 1994), as an emetic by the Pima Indians in the Southwest of United States and in (Mabry et al., 1979b; Tyler and Foster, 1999), weight-loss Mexico for the treatment of diabetes (Winkelman, 1989). (Sheikh et al., 1997). Likewise, creosote bush is employed against sterility Chaparral tea has also been used in the treatment of res- through vaginal baths with infusions of the leaves and tea piratory diseases (Mabry et al., 1979b; Brent, 1999), such as (Estudillo and Hinojosa, 1988; Argueta, 1994), it is also cold, cough and influenza (Argueta, 1994; Brent, 1999; Tyler reported to be effective against menstrual pains and post- and Foster, 1999), bronchitis (Timmermann, 1977; Sheikh parturient inflammation (Estudillo and Hinojosa, 1988; Ar- et al., 1997) and tuberculosis (Timmermann, 1977; Tyler gueta, 1994; Brent, 1999). Whereas an infusion of the is and Foster, 1999). A large great variety of illnesses have also used as a contraceptive (Moser, 1970; Argueta, 1994). been treated with creosote bush, such as anemia (Argueta, 234 S. Arteaga et al. / Journal of Ethnopharmacology 98 (2005) 231–239

Table 1 Main ethnobotanical uses of the leaves and twigs of Creosote bush Uses Reference Acne, psoriasis and dandruff Estudillo and Hinojosa (1988) and Brent (1999) Allergic problems Brinker (1993) Altered blood pressure Argueta (1994) and Sheikh et al. (1997) Analgesic and anti-inflammatory Timmermann (1977), Argueta (1994), Kay (1996) and Tyler and Foster (1999) Anemia Argueta (1994) Antiamoebic Brent (1999) and Segura (1978) Antibiotic Mabry et al. (1979a) and Argueta (1994) Antifungic Mabry et al. (1979a), Barragan´ et al. (1994) and Brent (1999) Antineoplasic Sheikh et al. (1997) and Tyler and Foster (1999) Antiseptic Timmermann (1977), Mabry et al. (1979b), Argueta (1994), Kay (1996) and Lara and Marquez´ (1996) Antiviral Brent (1999) Arthritis and rheumatism Mabry et al. (1979b), Brinker (1993), Argueta (1994), Lara and Marquez´ (1996) and Tyler and Foster (1999) Blood purifier Sheikh et al. (1997) Bowel cramps and inflammation Mabry et al. (1979b) and Tyler and Foster (1999) Bronchitis Timmermann (1977) and Sheikh et al. (1997) Burns Sheikh et al. (1997) and Brent (1999) Chicken pox Timmermann (1977) and Sheikh et al. (1997) Cicatrization, bruises and hemorrhoids Argueta (1994) Cold, cough and influenza Argueta (1994), Brent (1999) and Tyler and Foster (1999) Contraceptive agent (roots of the plant) Moser (1970) and Argueta (1994) Cramping Brinker (1993) Diabetes Winkelman (1989) and Argueta (1994) Diseases of the liver, and as a liver tonic Sheikh et al. (1997) and Brent (1999) Dysuria Mart´ınez (1969) and Lara and Marquez´ (1996) Diuretic Mabry et al. (1979b) and Tyler and Foster (1999) Emetic Mabry et al. (1979b) and Tyler and Foster (1999) Headache Argueta (1994) Kidney and gallbladder stones Diaz (1976) Kidney pain and cystitis Mart´ınez (1969) Menstrual pains and inflammation after delivery Estudillo and Hinojosa (1988), Argueta (1994) and Brent (1999) Neuritis and sciatica Timmermann (1977) and Mabry et al. (1979b) Parasites Mabry et al. (1979a) and Brinker (1993) Snakebite pain, chronic cutaneous disorders Nellesen (1997), Sheikh et al. (1997) and Brent (1999) Sterility Estudillo and Hinojosa (1988) and Argueta (1994) Stomach pain and diarrhea Argueta (1994) and Sheikh et al. (1997) Toothache Brent (1999) Tuberculosis Timmermann (1977) and Tyler and Foster (1999) Ulcer and indigestion Argueta (1994) Urinary tract infections and venereal diseases Timmermann (1977), Mabry et al. (1979b) and Brent (1999) Weight-loss Sheikh et al. (1997)

1994), altered blood pressure (Argueta, 1994; Sheikh et al., 4.2. Other reported uses 1997) and diabetes (Argueta, 1994), as well as blood purifier (Sheikh et al., 1997) and antineoplasic (Sheikh et al., 1997; The essential oil has been employed in soaps and creams, Tyler and Foster, 1999). as well as a shoe polish. Due to its content it has The beneficial effect in most of these treatments has not been employed in tanning. The extraction, crystallization been demonstrated using appropriate in vivo models or clin- and use of NDGA from creosote bush as a food antioxi- ical studies. However, two possible properties of creosote dant was approved by the Meat Inspection Division of the bush (and/or NDGA) for which some in vitro evidence exists US War Food Administration in 1943. However, in 1970 this seem to be of relevance for many different diseases: antibi- use was banned by the US Food and Drug Administration otic and anti-inflammatory activities. In other cases this does (Timmermann, 1981). Currently, NDGA is employed as an not seem to be so, such as in diabetes and gallstones, for antioxidant in the storage of natural and synthetic rubber. The which, however, some experimental evidence exist (Luo et resin is used as a thermofixed polymer adhesive for wood and al., 1998; Arteaga, 1997). More recently, creosote bush has cardboard, due to its strong antimicrobial properties, which been introduced as a dietary supplement, mainly due to its prevent the rotting of natural fibers. It has also been used to antioxidant activity. This could explain its use and abuse in cover capsules, tablets and pills. The whole plant has been other situations, as well as the increase in toxicity reports. used for house roofing and firewood (Mabry and Bohnstedt, S. Arteaga et al. / Journal of Ethnopharmacology 98 (2005) 231–239 235

1981). Likewise, a boiling of creosote bush has been em- In several states and diseases, such as aging, heart dis- ployed to remove scale and to unplug boilers and pipelines, ease, cancer, Alzheimer’s disease, inflammatory-immune in- as well as to clean rifle barrels (Timmermann, 1977). Cel- juries/autoimmune diseases (rheumatoid arthritis, lupus, dia- lulose with a similar quality to that obtained from conifers betes), AIDS, adult respiratory distress syndrome, etc., reac- could be acquired from creosote bush. The later use is desir- tive species have a causal contribution (Frei, 1994). able because of the abundance of creosote bush and the need Creosote bush has been traditionally used in some of these to decrease creosote populations in order to increase the pro- disorders, and perhaps its antioxidant activity through NDGA ductive capacity of the ecological systems it inhabits. Effec- and several other lignans could be responsible for its thera- tive control would require that the whole plant be removed peutic benefits. since it is a prolific root sprouter. Based on proximate nutrient Despite the wide use of creosote bush in traditional profiles it could be a good food source for livestock, as it has medicine and several experimental studies that give support a profile similar to . Flour with 18–31% protein has to these claims, there is a lack of in vivo pharmacological or been obtained from leaves previously extracted with ethanol clinical trials that confirm these results. (Navarro et al., 1981). However, it is necessary to remove the resin and the odor before it is used by cattle (Timmermann, 5.2. Side effects and toxicity 1981). On the other hand, creosote bush exhibits an ability for rapid root uptake of copper(II) ions from solution, so it Chaparral products, mainly tablets and capsules of pow- may provide a useful and novel method of removing cop- dered leaves and twigs, have been marketed as dietary sup- per from contaminated soils (Gardea et al., 2001). Therefore plements, due to their antioxidant properties. However, non- it has been used as an indicator of soil contamination, to recommended uses of these products have led to hepatic identify the presence and distribution of tritium near radioac- damage (Stickel et al., 2000). Several cases of chaparral- tive disposal areas in the (Andraski et al., associated hepatitis have been reported to the U. S. Food 2003). and Drug Administration (FDA) between 1992 and 1994 (Obermeyer et al., 1995). In one report, there was evidence of hepatotoxicity in 13 of 18 patients; the predominant pat- 5. In vivo pharmacology and pharmacodynamics tern of liver injury was characterized as toxic or drug-induced cholestatic hepatitis, jaundice and marked increase in serum 5.1. Reports on Larrea with positive outcomes liver chemistry values; in four individuals progression to cirrhosis was observed and in two individuals there was A study examining the safety of low-dosage treatment with acute fulminant liver failure requiring liver transplant (Sheikh creosote bush has been reported for subjects using creosote et al., 1997). In another case, a patient developed similar bush prior to initiation of the study for traditional uses includ- symptoms after taking chaparral tablets, 160 mg/day, for 2 ing both oral and topical applications (Heron and Yarnell, months, documented severe cholestasis and hepatocellular 2001). In this study none of the subjects showed any history injury. The serum enzyme levels were markedly elevated of liver disease from the use of Larrea in a complex herbal and severely narrowed biliary ducts were observed, without formula containing less than 10% of tincture, or in an ex- sclerosing cholangitis, distal obstruction, tumor, or stenosis tract in ricinus oil for topical use. It may be preferable to (Alderman et al., 1994). Another patient developed hepatitis avoid the use of Larrea capsules because they have been as- 2–3 months after beginning daily consumption of creosote sociated with potentially dangerous overdosing (Heron and bush leaf (proven by biopsy). The patient recovered after Yarnell, 2001). Since extracts of the creosote bush have long ceasing creosote bush intake (Batchelor et al., 1995). Yet been used as a folk remedy for type II diabetes studies in this another study reported hepatic and renal failure attributed to area are of particular interest. Masoprocol (pure NDGA iso- prolonged consumption of creosote bush products. The au- lated from the plant) significantly reduces plasma glucose and thor of this report noted that when taken in capsule or tablet triglyceride (TG) concentrations in rats treated with strepto- form creosote bush can cause subacute hepatitis (Gordon et zotocine, by increasing glucose disposal and decreasing lipol- al., 1995). Six patients exhibited clinical, biochemical and ysis (Luo et al., 1998; Reed et al., 1999); it also decreases TG histological evidence of severe hepatitis after taking herbal secretion and liver TG content in rats with fructose-induced remedies, among them chaparral. The symptoms were jaun- hypertriglyceridemia, a nondiabetic model of hypertriglyc- dice, fatigue, pruritus and high liver enzymes, indicating hep- eridemia associated with insulin resistance and hyperinsu- atocellular damage, in all biopsies portal and lobular hepatitis linemia (Scribner et al., 2000). was found (Whiting et al., 2002). The addition of a hydroalcoholic extract of creosote bush There is also the risk of renal disease resulting from prevents the formation of pigment gallstones in hamsters chronic chaparral tea ingestion (Spencer and Jacobs, 1999). (Granados and Cardenas,´ 1994), and that the ethanolic extract In a case report, a 56-year-old woman was diagnosed as prevents cholesterol stones by reducing the biliary cholesterol having numerous microscopic and macroscopic cysts in the molar percent (Arteaga, 1997). This studies support the use kidney. The patient had been consuming three to four cups of creosote bush in gallbladder lithiasis. daily of chaparral tea during a 3-month period (Smith, 1994). 236 S. Arteaga et al. / Journal of Ethnopharmacology 98 (2005) 231–239

Creosote bush has been reported as an agent producing the anti-HIV activity, inhibited (IC50 = 43.5 ␮M) the replication biliary duct vanishing syndrome, observed in cholestasis in- of herpes simplex virus in Vero cells (Chen et al., 1998) and duced by drugs (Chitturi and Farrel, 2001). Herbal prepara- human papilloma virus (Craigo et al., 2000). tions are marketed as natural and safe alternatives to conven- NDGA has been used in research as an antioxidant to test tional medicines for the prevention and treatment of a variety the participation of lipid peroxidation in some processes. Cre- of ailments, however, consumers may not be fully aware of osote bush scavenges the superoxide anion radical O2− in a their potential side effects. dose-dependent manner (Zang et al., 1999). NDGA inhib- ited the alterations of airway epithelial barrier and active ion 5.3. Experimental studies on possible additional uses of transport properties of guinea pig tracheobronchial mono- NDGA layers induced by dioxide (Robison and Kwang-Jin, 1996), the apoptotic cell death of the trophoblast layer of This lignan posses several beneficial properties. It has chorion tissues during development (Ohyama et al., 2001), been used in the treatment of the Sjogren-Larsson syndrome, and protected cultured rat hippocampal neurons against the a severe neurocutaneous disorder due to fatty aldehyde de- toxicity of amyloid ␤-peptide, interrupting a neurodegenera- hydrogenase involved in the degradation of leukotriene B4 tive pathway relevant to the pathophysiology of Alzheimer’s (Willensen, 2000). This is its only clinical use reported. It disease (Goodman et al., 1994). modulates the expression of endothelial nitric oxide syn- thetase in vitro, which has implications in the treatment of 5.4. Studies on NDGA toxicity cardiopathies (Ramasamy et al., 1999) and it also reduces blood pressure in rats with hypertension induced by fructose Besides being a potent lipoxygenase and cyclooxygenase (Gowri et al., 1999). NDGA is converted by the gut microflora inhibitor (Safayhi et al., 1992), NDGA is also an inhibitor of to estrogenic compounds, which have estrogenic activity in intracellular vesicular transport at concentrations between 50 vitro as well as in vivo (Fujimoto et al., 2004). and 100 ␮M(Drecktrath et al., 1998; Ramoner et al., 1998): Several studies suggest that NDGA could have a role in it not only interrupts the secretory vesicular route, but also cancer therapy. NDGA and a leaf extract of a South American the endocytic pathway in human dendritic cells. Protein recy- subspecies of creosote bush were found to exert an antitumor cling between endoplasmic reticulum and Golgi is reversibly effect in rats (Birkenfeld et al., 1987). NDGA, at 2 mg/day blocked by NDGA at 30 ␮M, disrupting the Golgi apparatus. p.o. for 1 week, is a potent inhibitor of hepatic toxicity and However, at 100 ␮M it inhibits protein synthesis and alters renal tumor promotion mediated by ferric-nitrilotriacetate in the Golgi irreversibly (Fujiwara et al., 1998a). Likewise, it mice (Ansar et al., 1999). It was shown to be a possible has been found that the ethanolic extract of creosote bush chemoprotective agent in patients at risk or with lung cancer and NDGA have a reversible cholestatic effect in hepatocyte (Soriano et al., 1999). Moreover, non-small and small-cell couplets at concentrations between 2 and 12 ␮g/ml (Cardenas´ lung cancer cell lines are inhibited by NDGA with an IC50 of et al., 2000), which could be related to its inhibition of intra- 5–7 ␮M(Moody et al., 1998). It also reduced the frequency cellular movement of transporters. NDGA reduces cellular of micronuclei induces by methyl methanosulfonate in vivo ATP through inhibition of electron flux in the respiratory (Diaz et al., 1999). Similar to other lipoxygenase inhibitors, chain (Fujiwara et al., 1998b); it also inhibits the regula- NDGA induced a more differentiated state and apoptosis in tion of cellular volume by swelling, through an inhibition of several human pancreatic cancer cell lines (Ding et al., 1999), taurine channels at 50–150 ␮M(Ballatori and Wang, 1997). suppress breast cancer cell growth (Earashi, 1995), and also Increases in intracellular Ca2+ in a concentration-dependent shows an additive or synergistic effect with retinoic acid on manner are induced by NDGA between 10 and l00 ␮M. This the inhibition of mammary tumor cell transformation and action is modulated by phospholipase A2 (Jan and Tseng, proliferation (Kubow et al., 2000). NDGA significantly in- 2000). NDGA caused a dose-dependent reduction of the ovu- hibited UVB-induced signaling pathways in the human ker- latory rate in the isolated perfused rat ovary, with a reduc- atinocyte cell line HaCaT, which suggests it to be a potential tion of ovarian prostaglandin and leukotriene concentrations agent in the prevention of skin cancer (Gonzales and Bowden, (Mikuni et al., 1998). 2002). These reports suggest two different modes of action The US Federal Drugs Administration prohibited the use for NDGA in cancer. The first one as an antioxidant, prevent- of NDGA as food additive since it was shown to inhibit sev- ing the harmful effect of reactive oxygen species, the other as eral enzymes such as peroxidase, catalase and alcohol dehy- an agent that affects genetic expression and differentiation, drogenase, as well as NADH-dehydrogenase and succinate probably through its effect on leukotriene synthesis. dehydrogenase (Timmermann, 1981). NDGA also inhibits Several lignans derivatives from Larrea tridentata show phospholipase A2 (Jacobson and Schrier, 1993), cytochrome anti-HIV activities (Gnabre et al., 1996), and several methy- P-450 (Agarwal et al., 1991) and carboxylesterases (Satoh lated NDGA were produced in the laboratory which exhib- and Hosokawa, 1998). ited similar or even higher anti-HIV activities than the natural Reported acute NDGA LD50 for range between compounds (Hwu et al., 1998). Of these lignans the synthetic 800 and 5500 mg/kg b.w. orally (Oliveto, 1972). Rats fed derivative tetramethyl-O-NDGA, which shows the highest diets with 3% NDGA developed cortical and medullary cysts S. Arteaga et al. / Journal of Ethnopharmacology 98 (2005) 231–239 237 in the kidney (Timmermann, 1977), and one case of human has many beneficial applications. However, due to some re- cystic renal disease and cystic adenocarcinoma associated to ports of toxicity care must be taken when the plant is used in chaparral tea consumption has been reported (Smith, 1994). traditional preparation.

5.5. Hepatic metabolism of creosote bush and NDGA Acknowledgements Little is known on the metabolism of creosote bush com- ponents. However, it has been reported on the hepatic pro- This work was supported by Direccion´ General de Estu- cessing of NDGA. The intravenous injection of 50 mg/kg dios de Posgrado, UNAM and CONACYT, through a fellow- of NDGA to mice resulted in a peak plasma concentration ship to SA. of 14.7 mg/ml, with a half-life of 135.0 min, and a clear- ance of 201.9 ml/min/kg (Lambert et al., 2001). The high clearance indicates that NDGA may be cleared by non-renal References mechanisms. Mono and diglucuronides of NDGA have been identified in bile from mice injected intraperitoneally with Abou-Gazar, H., Bedir, E., Takamatsu, S., Ferreira, D., Khan, I.A., 120 mg/kg of NDGA (Lambert et al., 2002). In this latter 2004. Antioxidant lignans from Larrea tridentata. Phytochemistry 65, study, an LD50 of 75 mg/kg was established 5 days after a 2499–2505. single i.p. dose, and an increase in alanine amino transferase Agarwal, R., Wang, D., Bik, P., Mukhtar, H., 1991. Nordihydroguaiaretic acid, an inhibitor of lipoxygenase, also inhibits cytochrome P-450- was found with 50 mg/kg. mediated monooxygenase activity in rat epidermal and hepatic micro- Similarly, woodrats fed with alfalfa pellets containing somes. Drug Metabolism and Disposal 19, 620–624. increasing levels of the phenolic resin from creosote bush Alderman, S., Kailas, S., Goldfarb, S., Singaram, C., Malone, D., 1994. showed increased glucuronide and sulfide conjugates level Cholestatic hepatitis after ingestion of chaparral leaf: confirmation with increasing resin intake, and it seems that woodrats from by endoscopic retrograde cholangiopancreatography and liver biopsy. Journal of Clinical Gastroenterology 19, 242–247. the Mojave Desert tolerate more resin because they have a Andraski, B., Sandstrom, M., Michel, R., Radyk, J., Stonestrom, D., John- greater capacity for glucuronide excretion (Mangione et al., son, M., Mayers, C., 2003. Simplified method for detecting tritium 2001). contamination in and soil. Journal of Environmental Quality 32, 988–995. Ansar, S., Iqbal, M., Athar, M., 1999. Nordihydroguaiaretic acid is a po- tent inhibitor of ferric-nitroloacetate-mediated hepatic and renal toxic- 6. Conclusions ity and renal tumour promotion in mice. Carcinogenesis 20, 599–606. Argueta, V. (Ed.), 1994. Atlas of the Traditional Mexican Medicinal Creosote bush and its main metabolite NDGA have shown Plants, vol. II. National Indigenous Institute, Mexico, pp. 669–670 to be useful in traditional medicine, industry and research. (in Spanish). Although several medicinal properties of creosote bush have Arteaga, S., 1997. Effect of Gobernadora (Larrea tridentata) on choles- terol cholelithiasis in the golden hamster (Mesocricetus auratus). support in experimental studies, with some exceptions none Bachelor in Biology Thesis. Facultad de Ciencias, UNAM, Mexico´ has been tested at the clinical level. One of the main uses (in Spanish). reported in traditional medicine is the treatment of kidney Ballatori, N., Wang, W., 1997. Nordihydroguaiaretic acid depletes ATP stones. However, there has not been any experimental assay and inhibits a swelling-activated, ATP-sensitive taurine channel. dealing with this issue. American Journal of Physiology 272, C1429–C1436. Barragan,´ S., Alvarez, G., Zavalza, A., 1994. In vitro evaluation of the an- On the other hand, the toxicity of creosote bush has been tifungic activity of Larrea tridentata against human pathogenic fungi. demonstrated. However, as far as we can conclude from these In: Proceedings of the First Mexican Congress of Ethnobiology, Sum- data the reported toxic doses in humans and experimental maries, Cuernavaca, Mexico,´ p. 17 (in Spanish). always exceeded the traditional use of the plant and Batchelor, W., Heathcote, J., Wanless, I., 1995. Chaparral-induced hepatic are often confounded with use of other herbs and potentially injury. American Journal of Gastroenterology 90, 831–833. Birkenfeld, S., Zaltsman, Y., Krispin, M., 1987. Antitumor effects of with lifestyle choices. Overall, prescribed appropriately tradi- inhibitors of arachadonic acid cascade on experimentally induced in- tional uses of other herbal medications appear safe. Although testinal tumors. Diseases of the Colon and Rectum 30, 43–46. linkage to some adverse effects may not be discovered since Bokoch, G., Reed, P., 1981. Evidence for inhibition of leukotriene A4 syn- problems are likely not to be reported, it is reasonable to as- thesis by 5,8,11,14-eicosatetraynoic acid in guinea pig polymorphonu- sume that there is a wide margin of safety for many popular clear leukocytes. Journal of Biological Chemistry 256, 4156–4159. Brent, J., 1999. Three new herbal hepatotoxic syndromes. Journal of remedies (Elvin-Lewis, 2001). Toxicology and Clinical Toxicology 37, 715–719. In several in vitro assays the effective concentrations of Brinker, F., 1993. Larrea tridentata (DC) Coville (chaparral or creosote NDGA are a range of 10–100 ␮M, with negative effects being bush). British Journal of Phytotherapy 3, 10–31. reversible within this range although higher concentrations Calzado-Flores, C., Segura-Luna, J., Guajardo-Touche, E., 1995. Effects usually cause irreversible damage. of chaparrin, nordihydroguaiaretic acid and their structural analogues on Entamoeba histolytica cultures. Proceedings of the West Pharma- Further studies are needed to establish the therapeutic and cological Society 38, 105–106. toxic doses of creosote bush in humans and experimental ani- Cardenas,´ R., Mendez,´ N., Cardenas,´ T., Arteaga, S., 2000. Cholestasis mals. Creosote bush is readily available plant that potentially induced by nordihydroguaiaretic acid and Gobernadora (Larrea tri- 238 S. Arteaga et al. / Journal of Ethnopharmacology 98 (2005) 231–239

dentata) in rat hepatocyte couplets. Revista de Gastroenterolog´ıa de Goodman, Y., Steiner, M., Steiner, S., Mattson, M., 1994. Nordihydrogua- Mexico´ 65, 59 (in Spanish). iaretic acid protects hippocampal neurons against amyloid ␤-peptide Chen, H., Teng, L., Li, J.N., Park, R., Mold, D.E., Gnabre, J., Hwu, toxicity, and attenuates free radical and calcium accumulation. Brain J.R., Tseng, W.N., Huang, R.C., 1998. Antiviral activities of methy- Research 654, 171–176. lated nordihydroguaiaretic acids. 2. Targeting herpes simplex virus Gordon, D., Gayle, R., Hart, J., Sirota, R., Baker, A., 1995. Chaparral in- replication by the mutation insensitive transcription inhibitor tetra-O- gestion: the broadening spectrum of liver injury caused by herbal med- methyl-NDGA. Journal of Medical Chemistry 41, 3001–3007. ications. Journal of the American Medical Association 273, 489–490. Chitturi, S., Farrel, G.C., 2001. Drug-induced cholestasis. Seminars in Gowri, M., Reaven, G., Azhar, S., 1999. Masoprocol lowers blood pres- Gastrointestinal Disease 12, 113–124. sure in rats with fructose-induced hypertension. American Journal of Craigo, J., Callahan, M., Huang, R., DeLucia, L., 2000. Inhibition of hu- Hypertension 12, 744–746. man papillomavirus type 16 gene expression by nordihydroguaiaretic Granados, H., Cardenas,´ R., 1994. Biliary calculi in the golden hamster. acid plant lignan derivatives. Antiviral Research 47, 19–28. XXXVII. The prophylactic action of the creosote bush (Larrea tri- De la Cerda, A.J., 1967. The desertic Mexican lands. General Study on dentata) in the pigment cholelithiasis produced by Vitamin A. Revista Larrea Species, vol. II. Serie Aridocultura, Mexico,´ pp. 13–45 (in de Gastroenterolog´ıa de Mexico´ 59, 31–35 (in Spanish). Spanish). Herbel, C.H., Gould, W.L., 1995. Management of mesquite, creosote bush, Diaz, J., 1976. Uses of from Mexico. Scientific Mono- and tarbush with herbicides in the northern . Agri- graphies. Mexican Institute for the Study of Medicinal Plants, Mexico,´ cultural Experiment Station Bulletin No. 775. New Mexico State Uni- 329 pp. (in Spanish). versity, Las Cruces, 53 pp. Diaz, S.B., Madrigal-Bujaidar, E., Marquez, P., 1999. Inhibitory effect of Heron, S., Yarnell, E., 2001. The safety of low-dose Larrea triden- nordihydroguaiaretic acid on the frequency of micronuclei induced by tata (DC) Coville (creosote bush or chaparral): a retrospective clin- methyl methanesulfonate in vivo. Mutation Research 441, 53–58. ical study. Journal of Alternative and Complementary Medicine 7, Ding, X., Kuszynski, C., El-Metwally, T., Adrian, T., 1999. Lipoxygenase 175–185. inhibition induced apoptosis, morphological changes, and carbonic Hwu, J., Tseng, W., Gnabre, J., Giza, P., Huang, R., 1998. Antiviral activi- anhydrase expression in human pancreatic cancer cells. Biochemistry ties of methylated nordihydroguaiaretic acid: synthesis, structure iden- and Biophysics Research Communication 266, 392–399. tification, and inhibition of Tat-regulated HIV transactivation. Journal Drecktrath, P., De Figueiredo, R., Mason, M., Brown, W., 1998. Retro- of Medical Chemistry 41, 2994–3000. grade trafficking of both Golgi complex and TGN markers to the ER Hyder, P., Fredrickson, E.L., Rick, E., Estell, R., Tellez, M., Gibbens, R., induced by NDGA and cyclophenil diphenol. Journal of Cell Science 2002. Distribution and concentration of total phenolics, condensed tan- 111, 951–965. nins, and nordihydroguaiaretic acid (NDGA) in creosote bush (Larrea Earashi, M., 1995. Effects of eicosanoid synthesis inhibitors on the in tridentata). Biochemical Systematics and Ecology 30, 905–912. vitro growth and prostaglandin E and leukotriene B secretion of a Jacobson, P., Schrier, D., 1993. Regulation of CD11b/CD18 expres- human breast cancer cell line. Oncology 52, 150–155. sion in human neutrophils by phospholipase A2. Immunology 151, Elvin-Lewis, M., 2001. Should we be concerned about herbal remedies. 5639–5652. Journal of Ethnopharmacology 75, 141–164. Jan, C., Tseng, C., 2000. Mechanisms of nordihydroguaiaretic acid- Estudillo, R.L., Hinojosa, A.L., 1988. Catalog of Sonoran Medicinal induced Ca2+, increases in MDKC cells. Life Sciences 66, 1753–1762. Plants. University of Sonora, Hermosillo, 131 pp. (in Spanish). Jones, S., 1987. Sistematica´ Vegetal. McGraw-Hill, Mexico,´ pp. 405–406 Frei, B., 1994. Reactive oxygen species and antioxidant vitamins: mech- (in Spanish). anisms of action. The American Journal of Medicine 97, S5–S13. Kay, M., 1996. Healing with Plants in the American and Mexican West. Fujimoto, N., Kohtab, R., Kitamurab, S., Hondaa, H., 2004. Estrogenic University of Arizona Press, Tucson, pp. 178–181. activity of an antioxidant, nordihydroguaiaretic acid (NDGA). Life Konno, C., Lu, Z., Xue, H., Erdelmeier, C., Meksuriyen, D., Che, C., Sciences 74, 1417–1425. Cordell, G., Soejarto, D., Waller, D., Fong, H., 1990. Furanoid lignans Fujiwara, T., Misumi, Y., Ikehara, Y., 1998a. Dynamic recycling of ER- from Larrea tridentata. Journal of Natural Products 53, 396–406. GIC53 between the endoplasmic reticulum and the Golgi complex is Kubow, S., Woodward, T., Turner, J., Nicodemo, A., Long, E., Zhao, disrupted by nordihydroguaiaretic acid. Biochemical and Biophysical X., 2000. Lipid peroxidation is associated with the inhibitory action Research Communication 253, 869–876. of all-trans-retinoic acid on mammary cell transformation. Anticancer Fujiwara, T., Takami, N., Misumi, Y., Ikehara, Y., 1998b. Nordihydrogua- Research 20, 843–848. iaretic acid blocks protein transport in the secretory pathway causing Lambert, J., Meyers, R., Timmermann, B., Dorr, R., 2001. Pharmacoki- redistribution of Golgi proteins into the endoplasmic reticulum. Jour- netic analysis by high-performance liquid chromatography of intra- nal of Biological Chemistry 273, 3068–3075. venous nordihydroguaiaretic acid in the mouse. Journal of Chromatog- Gardea, J., Arteaga, S., Tiemann, K., Chianelli, R., Pingitore, N., Mackay, raphy B 754, 85–90. W., 2001. Absorption of copper (II) by creosote bush (Larrea triden- Lambert, J., Zhao, D., Meyers, R., Kuester, R., Timmermann, B., Dorr, tata): use of atomic and X-ray absorption spectroscopy. Environmental R., 2002. Nordihydroguaiaretic acid: hepatotoxicity and detoxification Toxicology and Chemistry 20, 2572–2579. in the mouse. Toxicon 40, 1701–1708. Gay, C., Dwyer, D., 1998. New Mexico range plants. In: Allison, C., Lara, F., Marquez,´ C., 1996. Medicinal Plants from Mexico: Composition, Hatch, S., Schickedanz, J. (Eds.), Cooperative Extension Service Cir- Uses and Biological Activity. UNAM, Mexico,´ pp. 59–61 (in Spanish). cular No. 374. New Mexico State University, Las Cruces, NM, p. Lia, V., Confalonieri, C., Comas, I., Hunziker, J., 2001. Molecular phy- 84. logeny of Larrea and its allies (Zygophyllaceae): reticulate evolution Gezginci, M., Timmermann, B., 2001. A short synthetic route to and the probable time of creosote bush arrival in North America. nordihydroguaiaretic acid (NDGA) and its stereoisomer using Ti- Molecular and Phylogenetic Evolution 21, 309–320. induced carbonyl-coupling reaction. Tetrahedron Letters 42, 6083– Luo, J., Chuang, T., Cheung, J., Quan, J., Tsai, J., Sullivan, C., Hector, R., 6085. Reed, M., Meszaros, K., King, S., Carlson, T., Reaven, G., 1998. Ma- Gnabre, J., Ito, Y., Ma, Y., Huang, R., 1996. Isolation of anti-HIV-1 soprocol (nordihydroguaiaretic acid): a new antihyperglycemic agent lignans from Larrea tridentata by counter-current chromatography. isolated from the creosote bush (Larrea tridentata). European Journal Journal of Chromatography A 719, 353–364. of Pharmacology 346, 77–79. Gonzales, M., Bowden, G., 2002. Nordihydroguaiaretic acid-mediated in- Mabry, T., DiFeo, D., Sakakibara, M., Bohnstedt, C., Sleiger, D., 1979a. hibition of ultraviolet B-induced activator protein-1 activation in hu- The natural products chemistry of Larrea. In: Mabry, J., Hunzinker, man keratinocytes. Molecular Carcinogenesis 34, 102–111. J., DiFeo, D. (Eds.), Creosote Bush. Biology and Chemistry of Larrea S. Arteaga et al. / Journal of Ethnopharmacology 98 (2005) 231–239 239

in the New World Desert. Dowden Hutchinson Ross Inc., USA, pp. Salari, H., Braquet, P., Borgeat, P., 1984. Comparative effects of in- 115–134. domethacin, acetylenic acids, 15-HETE, nordihydroguaiaretic acid and Mabry, J., Hunzinker, J., DiFeo, D., 1979b. Creosote Bush. Biology and BW755C on the metabolism of arachidonic acid in human leukocytes Chemistry of Larrea in the New World Deserts. Dowden Hutchinson and platelets. Prostaglandins and Leukotrienes in Medicine 13, 53–60. Ross Inc., USA, 2083 pp. Satoh, T., Hosokawa, M., 1998. The mammalian carboxylesterases: from Mabry, T., Bohnstedt, Ch., 1981. Larrea: a chemical resource. In: Cam- molecules to functions. Annual Review of Pharmacology and Toxi- pos, L.E., Mabry, J.J., Fernandez,´ T.S. (Eds.), Larrea. CONACYT, cology 38, 257–288. Mexico,´ p. 232. Scribner, K., Gadbois, T., Gowri, M., Azhar, S., Reaven, G., 2000. Mangione, A., Dearing, D., Karasov, W., 2001. Detoxification in relation Masoprocol decreases serum triglyceride concentrations in rats with to toxin tolerance in desert woodrats eating creosote bush. Journal of fructose-induced hypertriglyceridemia. Metabolism 49, 1106–1110. Chemical Ecology 27, 2559–2578. Segura, J., 1978. Effects of nordihydroguaiaretic acid and ethanol on the Mart´ınez, M., 1969. The Medicinal Plants from Mexico. Botas, Mexico,´ growth of Entamoeba invadens. Archivos de Investigacion´ Medica´ pp. 143–144 (in Spanish). (Mex) 9, 157–162 (in Spanish). Mart´ınez-Vilalta, J., Pockman, W.T., 2002. The vulnerability to freezing- Sheikh, N., Philen, R., Love, L., 1997. Chaparral-associated hepatotoxi- induced xylem cavitation of Larrea tridentata (Zygophyllaceae) in the city. Archives of International Medicine 157, 913–919. Chihuahuan desert. American Journal of 89, 1916–1924. Smith, A., 1994. Cystic renal cell carcinoma and acquired renal cystic Mikuni, M., Yoshida, M., Hellberg, P., Peterson, A., Edwin, S., disease associated with consumption of chaparral tea: a case report. Brannstrom,˚ M., Peterson, M., 1998. The lipooxygenase inhibitor, Journal of Urology 152, 2089–2091. nordihydroguaiaretic acid, inhibits ovulation and leukotriene and Soriano, A., Helfrich, B., Chan, D., Heasley, L., Bunn, P., Chou, T., 1999. prostaglandin levels in the rat ovary. Biological Reproduction 58, Synergist effects of new chemopreventive agents against human lung 1211–1216. cancer cell lines. Cancer Research 59, 6178–6184. Moody, T.W., Leyton, J., Martinez, A., Hong, S., Malkinson, A., Mul- Spencer, J., Jacobs, J., 1999. Complementary/: An shine, J.L., 1998. Lipoxygenase inhibitors prevent lung carcinogenesis Evidence Based Approach. Mosby, Toronto, pp. 147–148. and inhibit non-small cell lung cancer growth. Experimental Lung Re- Stickel, F., Egerer, G., Seitz, H., 2000. Hepatotoxicity of botanicals. Pub- search 24, 617–628. lic Health Nutrition 3, 113–124. Moser, M., 1970. Seri Indians ethnobotany. The Kiva 35, 201. Tequida, M., Cortez, R., Rosas, B., Lopez, S., Corrales, M., 2002. Effect Navarro, M., Garcia, C., Martinez, H., 1981. Advances in the extraction of alcoholic extracts of wild plants on the inhibition of growth of technology of Larrea. In: Campos, L., Mabry, J., Fernandez,´ T. (Eds.), Aspergillus flavus, Aspergillus niger, Penicillium chrysogenum, Peni- Larrea. CONACYT, Mexico,´ pp. 369–379. cillium expansum, Fusarium moniliforme and Fusarium poae moulds. Nellesen, J., 1997. Larrea tridentata. Oecologia 109, 19–27. Revista Iberoamericana de Micolog´ıa 19, 84–88 (in Spanish). Obermeyer, W., Musser, S., Betz, J., Casey, R., Pohland, A., Page, S., Timmermann, B., 1977. Practical uses of Larrea. In: Mabry, T., Hun- 1995. Chemical studies of phytoestrogens and related compounds in ziker, J., Difeo, D. (Eds.), Creosote Bush. Biology and Chemistry of dietary supplements: flax and chaparral. In: Proceedings of the Society Larrea in New World Deserts. Dowden Hutchinson Ross Inc., USA, of Experimental Biology and Medicine, vol. 208, pp. 6–12. pp. 252–257. Ohyama, K., Yuan, B., Bessho, T., Yamakawa, T., 2001. Progressive apop- Timmermann, B., 1981. Larrea: potential uses. In: Campos, L., Mabry, tosis in chorion leave trophoblast cells of human fetal membrane tis- J., Fernandez,´ T. (Eds.), Larrea. CONACYT, Mexico,´ pp. 240–241. sues during in vitro incubation is suppressed by antioxidative reagents. Tyler, V., Foster, S., 1999. Tyler’s Honest Herbal: A Sensible Guide to European Journal of Biochemistry 268, 6182–6189. the Use of Herbs and Related Remedies. Haworth Herbal Press, New Oliveto, E., 1972. Nordihydroguaiaretic acid, a naturally occurring an- York, pp. 109–111. tioxidant. Chemistry and Industry 2, 677–679. Utah State University, 2005. Accessed on January 2005. http://extension. Ramasamy, S., Drummond, G., Ahn, J., Storek, M., Pohl, J., usu.edu/rangeplants/Woody/creosotebush.htm. Parthasarathy, S., Harrison, D., 1999. Modulation of endothelial nitric Van Auken, O.W., 2000. Shrub invasions of North American semiarid oxide synthase by nordehydroguaiaretic acid, a phenolic antioxidant grasslands. Annual Review of Ecology and Systematics 31, 197–215. in cultured endothelial cells. Molecular Pharmacology 56, 116–123. Verastegui, M., Sanchez, C., Heredia, N., Garcia, J., 1996. Antimicrobial Ramoner, R., Rieser, C., Thurner, M., 1998. Nordihydroguaiaretic acid activity of extracts of three major plants from the Chihuahuan desert. blocks secretory and endocytic pathway in human dendritic cells. Journal of Ethnopharmacology 52, 175–177. Journal of Leukocyte Biology 64, 747–752. Whitford, W.G., Nielson, R., De Soyza, A., 2001. Establishment and Reed, M., Meszaros, K., Entes, L., Claypool, M., Pinkett, J., Brignetti, D., effects of creosote bush, Larrea tridentata, on a Chihuahuan Desert Luo, J., Khandwala, A., Reaven, G., 1999. Effect of masoprocol on watershed. Journal of Arid Environments 47, 1–10. carbohydrate and lipid metabolism in a rat model of Type II diabetes. Whiting, P., Coluston, A., Kerlin, P., 2002. Black cohosh and other herbal Diabetologia 42, 102–106. remedies associated with acute hepatitis. The Medical Journal of Aus- Robison, T., Kwang-Jin, K., 1996. Enhancement of epithelial Na+,K+- tralia 117, 440–443. ATPase activity by NO2 and protective role of nordihydroguaiaretic Willensen, M., 2000. 5-Lipoxygenase inhibition: a new treatment strategy acid. American Journal of Physiology 270, L266–L272. for Sjogren-Larsson syndrome. Neuropediatrics 31, 1–3. Romo de Vivar, A., 1985. Natural Products of the Mexican Flora. Limusa, Winkelman, M., 1989. Ethnobotanical treatments of diabetes in Baja Cal- Mexico,´ 56 pp. (in Spanish). ifornia Norte. Medical Anthropology 11, 255–268. Rzedowsky, J., Huerta, M., 1994. Xerophilous heath. In: Rzedowsky, J. Xue, H., Lu, Z., Konno, C., Soejarto, D., Cordell, G., Fong, H., Hodg- (Ed.), The Mexican Vegetation. Limusa, Mexico,´ pp. 237–261 (in son, W., 1988. 3␤-(3,4-Dihydrixycinnamoyl)-erythrodiol and 3␤-(4- Spanish). hydroxycinnamoyl)-erythrodiol from Larrea tridentata. Phytochem- Safayhi, H., Mack, J., Sabieraj, M., Anazodo, L., Subramanian, H., Am- istry 27, 233–235. mon, H., 1992. Boswellic acids: novel, specific, nonredox inhibitors of Zang, L., Cosma, G., Gardner, H., Starks, K., Shi, X., Vallyathan, V., 5-lipoxygenase. Journal of Pharmacology and Experimental Therapy 1999. Scavenging of superoxide anion radical by chaparral. Molecular 261, l143–l146. and Cellular Biochemistry 196, 157–161. Journal of Ethnopharmacology 98 (2005) 241–244

Anti-diarrhoeal activity of Butea monosperma in experimental animals A. Gunakkunru a, K. Padmanaban a, P. Thirumal a, J. Pritila a, G. Parimala a, N. Vengatesan a, N. Gnanasekar a, James B. Perianayagam b, S.K. Sharma b,∗, K.K. Pillai c

a Department of Pharmacology, K.P. College of Pharmacy, Thiruvannamalai, India b Phytochemistry and Pharmacognosy Division, Faculty of Pharmaceutical Sciences, Guru Jambheshwar University, Hisar 125001, India c Department of Pharmacology, Faculty of Pharmacy, Jamia Hamdard (Hamdard University), Hamdard Nagar, New Delhi 110062, India Received 20 October 2004; received in revised form 2 November 2004; accepted 7 December 2004

Abstract

The anti-diarrhoeal potential of the ethanolic extract of stem bark of Butea monosperma (Lam) Kuntz has been evaluated using several experimental models in Wistar albino rats. The extract inhibited castor oil induced diarrhoea and PGE2 induced enteropooling in rats; it also reduced gastrointestinal motility after charcoal meal administration. The results obtained establish the efficacy and substantiate the use of this herbal remedy as a non-specific treatment for diarrhoea in folk medicine. © 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Butea monosperma; Fabaceae; Anti-diarrhoeal; Intestinal secretion; Gastrointestinal motility

1. Introduction 2002), we wish to report therein on the anti-diarrhoeal effect of the ethanolic extract of the stem bark of the plant. In developing countries, a quarter of infant and child- hood mortality is related to diarrhoea (Jousilahti et al., 1977). Many plants conveniently available in India are used in tra- 2. Materials and methods ditional folklore medicine for the treatment of diarrhoea and dysentery (Chopra et al., 1956). The plant Butea monosperma 2.1. Plant material (Lam) Kuntz (Fabaceae), also known as Bastard Teak, is a medium sized tree native of the mountainous regions of In- Stem bark of Butea monosperma was collected from the dia and Burma (Anonymous, 1988). The stem bark of this Valasa Malai in the Dharmapuri district of Tamilnadu, India plant is used in indigenous medicine for the treatment of in November 2002. It was identified by Professor P. Jayara- dyspepsia, diarrhoea, dysentery, diabetes, ulcers, sore throat man, Taxonomist, Plant Anatomy Research Centre, Chen- and snake bites (Jayaweera, 1981; Varier, 1993). Although nai. A voucher specimen (GPT/19) has been deposited in the many chemical constituents such as, -tannic acid, gal- Herbarium Section of the Department of Pharmacognosy, lic acid and proanthocyanidins have been isolated from the K.P. College of Pharmacy, Thiruvanamalai, Tamilnadu, for stem bark and gum (Seshadri and Trikha, 1972; Nadkararni future reference. The stem bark (1.5 kg) was air dried and and Nadkarani, 1976), together with (−) medicarpin from the pulverized using a mechanical grinder. stem bark (Bandara et al., 1989), and pterocarpans, phenols and lipids from the stem (Mishra et al., 2000; Shukla et al., 2.2. Preparation of extract

The powdered plant material (750 g) was subjected to ∗ Corresponding author. Tel.: +91 1662 263162; fax: +91 1662 276240. maceration process with 95% ethanol at room temperature. E-mail address: drsks [email protected] (S.K. Sharma). After exhaustive extraction, the ethanolic extract (EBM) was

0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2004.12.021 242 A. Gunakkunru et al. / Journal of Ethnopharmacology 98 (2005) 241–244 concentrated under reduced pressure at 50–55 ◦C. A reddish- mal was sacrificed and the intestinal distance moved by the brown coloured residue was obtained (93 g; yield 12.4%, charcoal meal from the pylorus was cut, measured, and ex- w/w) and stored in desiccator. For pharmacological studies, a pressed as a percentage of the distance from the pylorus to weighed amount of the dried extract was suspended in a 2% caecum for each (Mukherjee et al., 1998). (w/v) aqueous acacia solution. 2.7. PGE2-induced enteropooling 2.3. Animals used In this method, rats were deprived of food and water for Swiss albino mice (30–40 g) and Albino (Wistar) rats 18 h and placed in five cages, with five animals per cage. The (210–235 g) of either sex were maintained at uniform lab- first three groups were treated with 200, 400 and 800 mg/kg oratory conditions in standard polypropylene cages and pro- doses of EBM. The fourth group was treated with 1 ml of vided with food and water ad libitum. The animals were ac- a 5% (v/v) ethanol in normal saline (i.p.) and then it was climatized for a period of 14 days prior to performing the treated with aqueous acacia suspension, which served as ve- experiments. hicle control. Immediately after the extract administration PGE2 (Astra Zeneca, India) was administered orally to each 2.4. Acute toxicity study rat (100 ␮g/kg) in the first three groups. The fifth group was treated with PGE2 (100 ␮g/kg) as well as with aqueous aca- Swiss albino mice were divided into eight groups of six in- cia suspension and served as the PGE2 control group. Af- dividuals. The extract was administered orally at doses rang- ter 30 min following administration of PGE2, each rat was ing from 0.1 to 5 g/kg following a standard method (Turner, sacrificed and the whole length of the intestine from the py- 1965). A group of animals treated with 2% (w/v) aqueous lorus to the caecum was dissected out, its content collected acacia suspension (vehicle control). The animals were contin- in a test tube, and the volume measured (Mukherjee et al., uously observed for 2 h to detect changes in the autonomic 1998). or behavioural responses. Mortality in each group was ob- served for 7 days. The doses of 200, 400 and 800 mg/kg 2.8. Statistical analysis were selected based on the results of preliminary toxicity testing. All data was expressed as the mean ± S.E.M. Statisti- cal significance testing was performed by Student’s t-test; 2.5. Castor oil induced diarrhoea in rats P < 0.05 imply significance.

The method reported by Awouters et al. (1978), with mod- ifications, has been used in the present study. Rats of either 3. Results sex (210–235 g) were fasted for 18 h; they were then divided into five groups of five individuals. The EBM extract was The presence of steroids, flavonoids, phenolic compounds, administered orally at doses of 200, 400 and 800 mg/kg by and glycosides were detected on preliminary phyto- gavage as suspension to the first three groups of animals. The chemical screening of the dried extract. The EBM extract, fourth group received loperamide (3 mg/kg) orally as suspen- when orally administered in the dose range 0.1–5.0 g/kg to sion (positive control). The fifth group, which served as the mice did not produce any significant changes in the auto- blank, was administered with aqueous acacia suspension. Af- nomic or behavioural response during the observation pe- ter 60 min of treatment, the animals of each group received riod. No mortality was observed up to 7 days of monitoring. 1 ml of castor oil orally, by gavage, and the consistency of Hence, the extract seem to be safe for administration up to faecal material and the frequency of defecation was noted 5.0 g/kg. up to 4 h in the transparent plastic dishes placed beneath the The EBM extract significantly inhibited the frequency individual rat cages (Gnansekar and Perianayagam, 2004). of defaecation, when compared to untreated control rats (Table 1); the activity was similar to that of loperamide, the 2.6. Gastrointestinal motility tests standard anti-diarrhoeal agent. The extract also reduced the wetness of faecal droppings. The EBM extract at doses of Rats were fasted for 18 h and then placed in five cages 400 and 800 mg/kg decreased the propulsion of charcoal meal containing five individuals in each cage. Each animal was through the gastrointestinal tract, as compared with the con- administered orally with 1 ml of charcoal meal (5% deacti- trol group. Atropine (0.1 mg/kg) reduced the motility of the vated charcoal in 10% aqueous tragacanth), followed by oral intestine to a greater extent (P < 0.001) (Table 2). The extract administration of EBM suspension to three groups of ani- (EBM) significantly inhibited PGE2 induced enteropooling mals in doses of 200, 400 and 800 mg/kg. The fourth group in rats in higher dose levels (Table 3). PGE2 induced a sig- received atropine (0.1 mg/kg, i.p.), the standard drug for com- nificant increase in the fluid volume of the rat intestine when parison and the fifth group was treated with aqueous acacia compared with control animals, received ethanol in normal suspension (vehicle control). Thirty minutes later, each ani- saline. A. Gunakkunru et al. / Journal of Ethnopharmacology 98 (2005) 241–244 243

Table 1 Effect of stem bark extract (EBM) on castor oil induced diarrhoea in rats Oral pre-treatment at 0 h + castor oil at 1 h Mean defecations in 4 h Mean number of wet faeces in 4 h Control (acacia suspension 5 ml/kg) 3.60 ± 0.50 3.00 ± 0.70 Standard (loporamide 3 mg/kg) 0.80 ± 0.58c 0.00c EBM (200 mg/kg) 2.4 ± 0.24b 1.20 ± 0.80a EBM (400 mg/kg) 1.8 ± 0.42c 0.80 ± 0.52b EBM (800 mg/kg) 1.00 ± 0.48c 0.40 ± 0.24b Results are mean ± S.E.M., n = 5. Statistical significance test with control was done by Student’s t-test. a P < 0.05. b P < 0.01. c P < 0.001.

Table 2 thesis delayed castor oil induced diarrhoea (Awouters et al., Effect of EBM extract on small intestinal transit 1978). Oral pre-treatment Movement of charcoal meal (%) The extract appears to act on all parts of the intestine. Thus, Control (acacia suspension 5 ml/kg) 87.88 ± 1.84 it reduced the intestinal propulsive movement in the charcoal Standard (atropine 0.1 mg/kg) 29.12 ± 0.98b meal treated model; at 800 mg/kg EBM showed activity sim- EBM (200 mg/kg) 86.70 ± 0.84 ilar to that of atropine. The extract (EBM) at different dose . ± a EBM (400 mg/kg) 75 32 1.42 levels 400 and 800 mg/kg significantly inhibited the PGE EBM (800 mg/kg) 30.18 ± 1.12b 2 induced intestinal fluid accumulation (enteropooling). These Results are mean ± S.E.M., n = 5. Statistical significance test with control was done by Student’s t-test. observations tend to suggest that the EBM extract at doses of a P < 0.01. 400 and 800 mg/kg reduced diarrhoea by inhibiting gastroin- b P < 0.001. testinal motility and PGE2 induced enteropooling. The present results indicate that the ethanolic extract of Table 3 Butea monosperma possesses significant anti-diarrhoeal ac- Anti-enteropooling effect of EBM extract in rats tivity due to its inhibitory effect both on gastrointestinal Treatment Volume of intestinal P-values propulsion and fluid secretion. The inhibitory effect of the fluid (ml) extract justifies the use of the plant as a non-specific anti- Ethanol in saline 0.98 ± 0.30 – diarrhoeal agent in folk medicine. a PGE2 in ethanol (100 ␮g/kg) 2.82 ± 0.26 0.001 EBM (200 mg/kg) 2.78 ± 0.14 – EBM (400 mg/kg) 2.12 ± 0.12 0.05b EBM (800 mg/kg) 1.72 ± 0.17 0.001b Acknowledgement Results are mean ± S.E.M., n = 5. Statistical significance test with control was done by Student’s t-test. The authors are grateful to M/s Astra Zeneca Limited, a With respect to ethanol in saline treatment. Bangalore, India, for providing free samples of PGE2. b With respect to PGE2 treatment.

References 4. Discussion Ammon, H.V., Thomas, P.J., Phillips, S., 1974. Effect of oleic and reci- In the present study, the extract of Butea monosperma ex- noleic acid on net jejunal water and electrolyte movement. Journal of hibited significant anti-diarrhoeal activity against castor oil Clinical Investigation 53, 374–379. induced diarrhoea in rats. The extract had a similar activity Anonymous, 1988. The Wealth of India (Raw Material). Publication and Information Directorate, CSIR, New Delhi, pp. 341–346. as loperamide, when tested at 400 and 800 mg/kg and signif- Awouters, F., Nimegeers, C.J.E., Lenaerts, F.M., Janssen, P.A.J., 1978. icantly inhibited the frequency of defecation and the wetness Delay of castor oil diarrhoea in rats; a new way to evaluate inhibitors of the faecal droppings when compared to control rats. Castor of prostaglandin biosynthesis. Journal of Pharmacy and Pharmacology oil releases ricinoleic acid which induces changes in mucosal 30, 41–45. fluid and electrolyte transport that results in a hypersecretory Bandara, B.M.R., Kumar, N.S., Samaranayake, K.M.S., 1989. An anti- fungal constituent from the stem bark of Butea monosperma. Journal response and diarrhoea (Ammon et al., 1974; Gaginella et of Ethanopharmacology 25, 73–75. al., 1975). The experimental studies in rats demonstrated a Beubler, E., Juan, H., 1979. Effect of ricinoleic acid and other significant increase in the portal venous PGE2 concentration on net water flux and prostaglandin E release by the rat colon. Journal following oral administration of castor oil (Luderer et al., of Pharmacy and Pharmacology 31, 681–685. Chopra, R.N., Nayar, S.L., Chopra, I.C., 1956. Glossary of Indian Medic- 1980). Ricinoleic acid markedly increased the PGE2 content inal Plants. Council of Scientific and Industrial Research, New Delhi. in the gut lumen and also caused on increase of the net se- Gaginella, T.S., Stewart, J.J., Olson, W.A., Bass, P., 1975. Action of cretion of the water and electrolytes into the small intestine recinoleic acid and structurally related fatty acid on the gastroin- (Beubler and Juan, 1979). Inhibitors of prostaglandin biosyn- testinal tract II. Effect on water a electrolyte absorption in-vitro. 244 A. Gunakkunru et al. / Journal of Ethnopharmacology 98 (2005) 241–244

Journal of Pharmacology and Experimental Therapeutics 195, 355– Mukherjee, P.K., Saha, K., Murugesan, T., Mandal, S.C., Pal, M., Saha, 361. B.P., 1998. Screening of anti-diarrhoeal profile of some plant extracts Gnansekar, N., Perianayagam, J.B., 2004. Influence of sodium curcumi- of a specific region of West Bengal, India. Journal of Ethanopharma- nate on castor oil-induced diarrhoea in rats. Indian Journal of Phar- cology 60, 85–89. macology 36, 177–178. Nadkararni, K.M., Nadkarani, A.K., 1976. Indian Materia Medica, vol. Jayaweera, D.M.A., 1981. Medicinal Plant Used in Ceylon. Part 3. Na- 1. Popular Prakashan, Bombay, p. 223. tional Science Council of Sri Lanka, Colombo, p. 161. Seshadri, T.R., Trikha, R.K., 1972. Proanthocyanidins from the bark Jousilahti, P., Madkour, S.M., Lambrechts, T., Sherwin, E., 1977. Di- and gum of Butea monosperma. Indian Journal of Chemistry 9, arrhoeal disease morbidity and home treatment practices in Egypt. 1201–1203. Public Health 111, 5–10. Shukla, Y.N., Mishra, M., Kumar, S., 2002. Pterocarpan, and lipid Luderer, J.R., Dermers, L.N., Hayn, A.T., 1980. Advances in constituents from Butea monosperma stem. Indian Journal of Chem- Prostaglandin and Thromboxane Research. Raven Press, New York, istry 41B, 1283–1285. pp. 1633–1638. Turner, R.A., 1965. Screening Methods in Pharmacology. Academic Press, Mishra, M., Shukla, Y.N., Kumar, S., 2000. Euphane triterpenoid and New York, p. 63. lipid constituents from Butea monosperma. Phytochemistry 54, 835– Varier, S.P.V., 1993. Indian Medicinal Plants, vol. 1, first ed. Orient Long- 838. man Limited, Madras, p. 284. Journal of Ethnopharmacology 98 (2005) 245–250

Effects of wild (Panax ginseng C.A. Meyer) leaves on lipid peroxidation levels and antioxidant enzyme activities in streptozotocin diabetic rats

Chang-Hwa Jung a,b, Ho-Moon Seog a,∗, In-Wook Choi a, Hee-Don Choi a, Hong-Yon Cho b

a Korea Food Research Institute, San 46-1, Baekhyun-dong, Bundang-ku, Songnam-si, Kyunggi-do 463-746, South Korea b Korea University, 1,5-ka, Anam-dong, Sungbuk-ku, Seoul 136-701, South Korea Received 17 May 2004; received in revised form 9 December 2004; accepted 13 December 2004

Abstract

The aim of this study was to examine the possible antioxidant activities of wild Panax ginseng leaf extract intake in streptozotocin (STZ)-induced diabetic rats (WGLE). Initial blood glucose levels increased abruptly after streptozotocin injection. After 4 weeks of WGLE supplementation, blood glucose levels were lower in animals fed 40 mg/kg (266 mg/dL) and 200 mg/kg (239 mg/dL) than those in no-WGLE fed diabetic rats (464 mg/dL). The concentration of blood TBARS, which are considered the main products of glucose oxidation in blood, was also lowered by WGLE supplementation. These results indicate that WGLE supplementation is involved in suppressing a sudden increase in blood glucose levels and a consequent decrease in TBARS levels in diabetic rats. TBARS levels in the liver, kidney and spleen of WGLE-fed diabetic groups were also significantly lower than in the control diabetic group indicating that oral administration of WGLE effectively suppresses lipid peroxidation that occurs in the organs of diabetic rats. Antioxidant activities of WGLE supplementation further extend in suppressing activities of antioxidant related enzymes, such as glutathione peroxidase (GSH-Px), catalase (CAT) and superoxide dismutase (SOD), in organs of diabetic rats. These results confirm the effectiveness of WGLE supplementation in detoxifying free radicals that are produced excessively in diabetic-induced complications. © 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Wild ginseng; Panax ginseng leaves; Streptozotocin-induced diabetic rats; Thiobarbituric acid reactive substances (TBARS); Antioxidant activities

1. Introduction on the role of oxidative stress. It has been suggested that oxidative stress may contribute to the pathogenesis of dif- Diabetes mellitus (DM), characterized by hyperglycemia ferent diabetic complications (Ceriello, 2000). Diabetic ex- and long-term complications affecting the eyes, kidneys, perimental animal models have shown that oxidative stress nerves and blood vessels, is the most common endocrine dis- causes persistent and chronic hyperglycemia, thereby deplet- order. Although the underlying mechanism of diabetic com- ing activities of the antioxidant defense system and otherwise plications remains unclear, much attention has been focused promoting free radicals generation (Kakkar et al., 1997; Bhor et al., 2004). Abbreviations: WGLE, wild Panax ginseng leaf extract; TBARS, thio- Panax ginseng C.A. Meyer (Araliaceae) is a valuable herb barbituric acid reactive substances; GSH-Px, glutathione peroxidase; CAT, in East Asia that has also gained popularity in the West be- catalase; SOD, superoxide dismutase; N, normal group; DM, streptozotocin- cause of its pharmacological properties (Zhang et al., 1996; diabetic group; N + W200, normal rats fed with WGLE (200 mg/kg day); Mahady et al., 2000). Panax ginseng is categorized as either DM + W40, diabetic animals fed with WGLE (40 mg/kg day); DM + W200, cultivated or wild according to different nurturing methods. diabetic animals fed with WGLE (200 mg/kg day) ∗ Corresponding author. Tel.: +82 31 780 9248; fax: +82 31 709 9876. Cultivated ginseng is systematically farmed on open land E-mail address: [email protected] (H.-M. Seog). and harvested after a 5–6 year cultivation period. On the

0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2004.12.030 246 C.-H. Jung et al. / Journal of Ethnopharmacology 98 (2005) 245–250 other hand, wild ginseng is planted through seeding in a deep 2.3. Preparation of ginseng leaf extract mountain at an altitude between 800 and 1500 m. Wild gin- seng is slower in growth and more sensitive to environmental The wild ginseng leaves were freeze-dried and ground changes than cultivated ginseng, showing a preference for into a fine powder. The powder was extracted twice with wa- areas with fluctuating daily temperatures and less exposure ter under reflux in a water bath at 90 ◦C for 1 h. The ex- to direct sunlight. These differences may result in a variation tract was then filtered and freeze-dried. The extract yield of active compounds between cultivated and wild ginseng. was 46% (w/w, dry basis). Finally, the extract was dis- It is widely accepted in both Korea and China that wild solved in distilled water for oral administration to the diabetic ginseng is more active than cultivated ginseng. However, only rats. a few studies have been conducted to compare the pharmaco- logical activities of wild and cultivated ginseng. Traditionally, Panax ginseng leaves have been consumed mostly in the form 2.4. Determination of ginsenosides, phenolics and of tea (KFDA, 2002). Ginseng leaves have long been used as flavonoids folk medicine in the treatment of diabetes in Korea. Because most studies on the use of ginseng as a treatment for diabetes The freeze-dried extract was dissolved in 20% (v/v) ace- focused on roots rather than leaves, little has been reported on tonitrile/water and filtrated through a 0.45 ␮m membrane the antidiabetic effectiveness and compounds found in gin- filter to analyze ginsenosides by HPLC (Li et al., 1996). seng leaves. Xie et al. (2004) have researched hypoglycemic Total phenolic contents in WGLE were determined us- activities in diabetic rat models and concluded that these ac- ing Folin–Ciocalteu’s reagent according to the method of tivities come primarily from ginsenosides (Xie et al., 2004). Singleton and Lamuela-Raventos (1999). Total flavonoid Ginseng leaves are known to have six major ginsenosides, contents in WGLE were determined by the method of Rb1,Rb2, Rc, Rd, Re and Rg1 (Li et al., 1996). It is generally Kumazawa et al. (2004). believed that dammaranne saponins in ginseng play major roles in antioxidant activities in animal models (Attele et al., 1999). In addition, the phenolic acids and flavonoids in gin- 2.5. Animals and treatment seng leaves exhibit very potent antioxidant activities (Park et al., 1990). Male Sprague–Dawley rats (250 ± 10 g) were housed in Streptozotocin (STZ) is widely used to induce diabetes steel cages. They were allowed free access to drinking water in experimental animals by causing the selective destruction and normal diet during the experiment. The rats were accli- of pancreatic ␤-cells that secret insulin (Aksoy et al., 2003). matized for 7 days and randomly allotted into five groups. Hyperglycemia in diabetes results in excessive protein gly- All animal experiments were performed under the guidelines cation and the production of reactive oxidants, which leads of the Laboratory Animal Experiment Committee of Korea to oxidative damage in organs (Brownlee, 2001). The main Food Research Institute. Body weight was recorded at 2-day objective of this study was to evaluate the effects of WGLE intervals in every group. Amount of food consumption was supplementation on suppressing such damage caused by hy- measured by subtracting leftovers from the diet provided to perglycemia in diabetic rats. the rats at 2-day intervals.

2. Materials and methods 2.6. Experimental design

2.1. Chemicals In the normal groups, rats were allotted into two sub- groups: normal rats (N) and normal rats fed with WGLE Standard ginsenoside Rg1, Rc, Rb1, Re and Rd were (200 mg/kg, N + W200). Diabetic animals were distributed obtained from Wako (Osaka, Japan). Ginsenoside Rb2 and into control diabetic (DM) and WGLE-fed diabetic group. Rf were purchased from Extrasynthese (Genay, France). WGLE-fed diabetic animals were allotted into DM + W40 Streptozotocin and all other chemicals of analytical grade (40 mg/kg) and DM + W200 (200 mg/kg) according to the were purchased from Sigma Chemical Co. (St. Louis, amounts of WGLE supplementation. Diabetes was induced MO). by intravenous injection of streptozotocin into the tail vein at a dose of 45 mg/kg body weight in a 0.1 M fresh cold citrate 2.2. Plant material buffer (pH 4.3). Six days after the STZ injection, glucose in the orbital vein plexus was measured in all test rats. STZ- The wild Panax ginseng C.A. Meyer leaves were collected treated rats having a blood glucose level of 280–380 mg/dL from ginseng plants that were grown for more than 12 years at were included in the study. After 4 weeks, all animals were a mountain in Inje-gun, Gangwon-do, Korea. Voucher speci- sacrificed and the liver, kidney and spleen were isolated from mens have been deposited in the laboratory of KT&G Center each rat, washed in 0.9% saline solution, blotted dry and Research Institute, in Gyunggi-do, Korea. weighed. C.-H. Jung et al. / Journal of Ethnopharmacology 98 (2005) 245–250 247

2.7. Determination of glucose and thiobarbituric acid of change of absorbance during the conversion of NADPH to reactive substances (TBARS) in blood NADP+ was recorded spectrophotometrically at 340 nm for 3 min. GSH-Px activity for tissues was expressed as ␮moles Blood was obtained from the orbital vein plexus of all of NADPH oxidized to NADP+ min−1 mg−1 protein. rats. Blood glucose concentrations were measured by a glu- Catalase (CAT) activity was measured according to the cometer kit (Abbott Laboratories, Illinois, USA). TBARS Aebi (1983) method. One unit of CAT activity was defined in the serum were estimated by Satoh (1978) method based as the amount of enzyme required to decompose 1 ␮mol of on thiobarbituric acid (TBA) reactivity. 1.0 mL of 20% H2O2 in 1 min. In a cuvette containing 1.95 mL of a 50 mM trichloroacetic acid (TCA) was added to the 0.5 mL of serum phosphate buffer (pH 7.0), 0.05 mL of tissue supernatant was and the tube was left to stand for 10 min at room temper- added. The reaction was started by the addition of 1.0 mL ature. After centrifugation at 4000 × g for 10 min, 0.5 mL of freshly prepared 30 mM H2O2. The rate of decomposition of supernatant and 0.8% TBA were added. The coupling of H2O2 was measured spectrophotometrically at 240 nm for of lipid peroxide with TBA was carried out by heating in 1 min. Using the reaction time (t) of the absorbance (A1 a boiling water bath for 30 min. After cooling in cold wa- and A2), the following equation was generated to calculate ter, the resulting chromogen was extracted with 4.0 mL of the rate constant (k): k = (2.3/t)(log A1/A2). The enzyme ac- n-butyl alcohol by vigorous shaking. Separating of the or- tivity was expressed as k mg−1 protein. ganic phase was facilitated by centrifugation at 4000 × g Superoxide dismutase (SOD) activity was measured by the for 10 min and its absorbance was determined at a wave- inhibition of pyrogallol autoxidation at 420 nm for 10 min length of 532 nm. The standard for TBARS in this study was according to Marklund and Marklund (1974) method. The malondialdehyde (MDA). MDA was prepared from 1,1,3,3- enzyme activity was expressed as U/mg protein, where 1 U tetramethoxypropane. The MDA concentration in tissue ho- is the amount of enzyme required to bring about 50% inhi- mogenates was normalized with protein concentration, which bition of the autoxidation of pyrogallol. The assay mixture was measured by the Lowry method (Lowy et al., 1951). consisted of 1.8 mL of 50 mM Tris–HCl buffer (containing 10 mM EDTA), 0.1 mL of 6.0 mM pyrogallol and the diluted 2.8. Tissue samples tissue supernatant to a final volume of 2.0 mL. The reaction was stopped by adding 0.05 mL of 1N HCl. Tissue was chopped into small pieces on ice. A 10% (w/v) homogenate was prepared in 10 mM phosphate buffer (pH 2.11. Statistical analysis 7.4) and centrifuged at 13,000 × g for 10 min at 4 ◦C. The supernatant was used for immediate thiobarbituric acid reac- All experimental data were analyzed for variance tive substances and enzyme activity evaluation. (ANOVA) and significant difference among the means from triplicate analysis was set at P < 0.05. This was determined 2.9. Determination of thiobarbituric acid reactive by Duncan’s multiple range test using the Statistical Analysis substances in tissues System (SAS).

TBARS in tissue was estimated by the method of Ohkawa et al. (1979). Some 0.2 mL of the tissue sample was added 3. Results to 0.2 mL of 8.1% sodium dodecyl sulfate (SDS), 1.5 mL of 20% acetic acid solution (pH 3.5) and 1.5 mL of 0.8% TBA. 3.1. Effects of WGLE supplementation on blood glucose The mixture was made up to 4.0 mL with distilled water and and TBARS in diabetic rats heated in a water bath at 90 ◦C for 60 min. After cooling with tap water, 1.0 mL of distilled water and 5.0 mL of n- As shown in Table 1, there were significant differences in butanol were added and shaken vigorously and centrifuged body weight gain between normal (N) and diabetic animals at 4000 × g for 10 min. The upper butanol layer was taken (DM). Unlike the normal group, which gained weight dur- and its absorbance at 532 nm was read. ing the experimental period, the diabetic group lost weight (−53.26 ± 10.38 g) after 4 weeks. Increased food consump- 2.10. Antioxidant enzyme activities tion and decreased body weight observed in diabetic rats in comparison to normal rats indicates polyphagic condition and Activity of glutathione peroxidase (GSH-Px) was deter- weight loss due to the excessive breakdown of tissue proteins mined according to the method of Lawrence and Burk (1976). (Chatterjea and Shinde, 2002). Weight loss of WGLE fed di- The assay mixture consisted of 2.0 mL of 75 mM phosphate abetic groups (DM + W40 and DM + W200), however, were buffer (pH 7.0), 50 ␮L of 60 mM glutathione, 0.1 mL of less severe than that of the DM group. 30 units/mL glutathione reductase, 0.1 mL of 15 mM EDTA, The result of WGLE effects on lowering blood glucose 0.1 mL of 3 mM NADPH and the appropriate amount of tis- and TBARS in diabetic rats is also presented in Table 1. sue supernatant to a final volume of 3.0 mL. The reaction was There was no significant difference in blood glucose levels started by the addition of 0.1 mL of 7.5 mM H2O2. The rate after 4 weeks of WGLE supplementation on normal groups 248 C.-H. Jung et al. / Journal of Ethnopharmacology 98 (2005) 245–250

Table 1 Effects of WGLE supplement on weight gain, food intake, feed efficiency ratios (FER), glucose and TBARS in blood of diabetic rats Group N(n = 8) N + W200 (n =4) DM(n =7) DM+W40(n = 6) DM + W200 (n =6) Weight gain (g) +78.29 ± 14.84 a +88.21 ± 20.12 a −53.26 ± 10.38 c −35.77 ± 15.86 bc −24.23 ± 15.21 b Food intake (g) 535.62 ± 85.09 a 629.35 ± 28.80 a 990.42 ± 70.76 b 1043.50 ± 46.50 b 1008.13 ± 8.49 b FER 0.15 ± 0.02 a 0.14 ± 0.06 a −0.05 ± 0.03 c −0.03 ± 0.02 bc −0.02 ± 0.02 b Glucose (mg/dL) Initial 100.38 ± 10.88 a 106.73 ± 7.33 a 338.67 ± 32.33 b 334.90 ± 19.00 b 343.33 ± 13.11 b Final 106.00 ± 4.50 a 101.33 ± 7.11 a 464.68 ± 72.78 c 266.00 ± 45.00 b 239.68 ± 71.78 b TBARS (nmol/mg protein) Initial 0.114 ± 0.003 a 0.114 ± 0.002 a 0.178 ± 0.024 b 0.174 ± 0.024 b 0.162 ± 0.024 b Final 0.114 ± 0.004 a 0.115 ± 0.005 a 0.234 ± 0.032 c 0.170 ± 0.032 b 0.174 ± 0.032 b N: normal group, no streptozotocin injected; N + W200: normal group orally fed with WGLE (200 mg/kg day); DM: diabetic control group; DM + W40: diabetic group orally fed with WGLE (40 mg/kg day); DM + W200: diabetic group orally fed with WGLE (200 mg/kg day). All values are mean ± S.E.M. Mean values with different letters are significantly different (P < 0.05) by Duncan’s multiple-range test.

indicating that feeding WGLE to normal rats does not cause 3.2. Effects of WGLE supplementation on antioxidant any changes in blood glucose levels. In diabetic groups, how- enzyme activities and TBARS in organs of diabetic rats ever, the effect of WGLE supplementation on lowering blood glucose was clearly observed. DM groups with no WGLE Table 2 shows antioxidant related enzyme activities in the supplementation showed an abrupt increase in blood glucose livers, kidneys and spleens of normal and diabetic groups. In from 338.6 to 464.6 mg/dL after 4 weeks. Although the ini- general, normal groups maintained higher enzyme activities tial blood glucose level in WGLE fed DM groups (DM + W40 than diabetic groups. Although daily administrations of 40 and DM + W200) was around 330 mg/dL, it was significantly and 200 mg/kg for 4 weeks were not effective in fully recov- reduced to 266.0 mg/dL in DM + W40 and 239.6 mg/dL in ering enzyme activities as detected in normal groups, WGLE DM + W200 after 4 weeks of WGLE feeding. As in the case supplementation seemed to play a certain role in recovering of blood glucose levels, TBARS levels in all DM groups were them in diabetic groups. By STZ-induction, TBARS in each initially higher than those in normal groups. All WGLE fed organ were increased by 25–40% (Table 2). In most of the diabetic rats maintained constant blood TBARS levels even organs in the diabetic groups, TBARS were effectively re- though control diabetic rats had increased TBARS levels after duced by the administration of WGLE in a dose dependent 4 weeks. manner.

Table 2 Activities of glutathione peroxidase, catalase, superoxide dismutase in liver, kidney and spleen from normal and diabetic rats treated with WGLE Group

N(n = 8) N + W200 (n =4) DM(n =7) DM+W40(n = 6) DM + W200 (n =6) Liver TBARS 1.07 ± 0.13 a 1.10 ± 0.16 a 1.71 ± 0.10 c 1.45 ± 0.05 b 1.41 ± 0.06 b GSH-Px 460.04 ± 23.39 a 455.80 ± 20.23 a 301.67 ± 12.94 c 321.67 ± 10.67 c 360.25 ± 17.31 b CAT 0.84 ± 0.04 a 0.87 ± 0.04 a 0.46 ± 0.03 d 0.70 ± 0.04 c 0.79 ± 0.02 b SOD 6.15 ± 0.58 a 6.56 ± 0.55 a 3.44 ± 0.34 c 4.46 ± 0.36 b 4.77 ± 0.31 b Kidney TBARS 0.77 ± 0.07 a 0.81 ± 0.03 a 1.05 ± 0.10 c 0.94 ± 0.06 bc 0.90 ± 0.05 b GSH-Px 146.75 ± 11.56 a 144.82 ± 11.52 a 97.93 ± 9.71 c 107.30 ± 10.96 c 129.55 ± 10.97 b CAT 0.61 ± 0.02 a 0.60 ± 0.05 a 0.25 ± 0.02 c 0.27 ± 0.01 bc 0.30 ± 0.03 b SOD 4.57 ± 0.34 a 4.58 ± 0.34 a 6.11 ± 0.46 c 5.38 ± 0.33 b 5.12 ± 0.29 b Spleen TBARS 0.27 ± 0.04 a 0.28 ± 0.03 a 0.38 ± 0.05 b 0.35 ± 0.03 ab 0.28 ± 0.04 a GSH-Px 177.55 ± 12.56 a 181.45 ± 10.10 a 138.20 ± 5.96 c 137.23 ± 11.41 c 165.85 ± 5.12 b CAT 0.21 ± 0.01 a 0.22 ± 0.02 a 0.14 ± 0.01 b 0.14 ± 0.01 b 0.14 ± 0.02 b SOD 4.43 ± 0.50 a 4.39 ± 0.22 a 2.63 ± 0.33 c 3.08 ± 0.34 bc 3.67 ± 0.20 b N: normal group, no streptozotocin injected; N + W200: normal group orally fed with WGLE (200 mg/kg day); DM: diabetic control group; DM + W40: diabetic group orally fed with WGLE (40 mg/kg day); DM + W200: diabetic group orally fed with WGLE (200 mg/kg day). All values are mean ± S.E.M. Mean values with different letters are significantly different (P < 0.05) by Duncan’s multiple-range test. C.-H. Jung et al. / Journal of Ethnopharmacology 98 (2005) 245–250 249

4. Discussion enging reactive oxygen species and (or) elevating the activity of antioxidant enzymes (Shirwaikar et al., 2004). STZ is a commonly employed compound for the induc- Another important component in ginseng leaves, ginseno- tion of type-1 diabetes in rats (Tomlinson et al., 1992). STZ sides, produces powerful antioxidant activities other than causes diabetes by the rapid depletion of ␤-cells, which leads radical scavenging activities by simulating gene express of to a reduction in the insulin release. An insufficient release antioxidant enzymes and enhancing their activities (Wee of insulin causes high blood glucose, namely hyperglycemia, et al., 1989; Kim et al., 1996a). These enhanced activities which results in oxidative damage by the generation of reac- of antioxidant enzymes contribute to further scavenging of tive oxygen species (ROS) (Mohamed et al., 1999) and the ROS or other free radicals formed through lipid peroxida- development of diabetic complications (Donnini et al., 1996). tion (Fan et al., 1995; Kim et al., 1996b). In our results, Furthermore, STZ diabetic animals may exhibit most other ginsenosides in WGLE composed 16.99% (data not shown). diabetic complications such as myocardial, cardiovascular, Conclusively, WGLE supplementation causes both reactiva- gastrointestinal, nervous and urinary bladder dysfunctions tion of antioxidant enzymes and scavenging of free radicals, (Ozturk et al., 1996). which, in turn, results in decreased TBARS synthesis and its In this study, we examined oxidative stress pathway mark- accumulation. ers in rats with STZ-induced diabetes. As shown in our re- sults, diabetic rats had much higher blood glucose levels than normal rats during the experimental period. By feed- Acknowledgments ing WGLE to diabetic rats, however, blood glucose levels were controlled to a certain degree. Although the mechanism This work was supported by a grant from Agricultural Re- involved with suppressing blood glucose levels by WGLE search Promotion Center, Ministry of Agriculture and Forest, supplement was not clearly demonstrated in this study, at Korea. least three possibilities can be suggested: (1) modulation of glucose transport (Yamasakiet al., 1993), (2) glucose disposal (Yokozawa et al., 1984), or (3) insulin secretion (Waki et al., 1982). Some components such as phenolics and flavonoids References that compose 1.67 and 0.95% of dry ginseng leaves (data not shown) are known to be responsible for hypoglycemic Aebi, H., 1983. Catalase in vitro. Methods in Enzymology 105, 121–126. activity (Ragunathan and Sulochana, 1994). Aksoy, N., Vural, H., Sabuncu, T., Aksoy, S., 2003. Effects of melatonin on oxidative–antioxidative status of tissues in streptozotocin-induced Since high blood glucose is susceptible to oxidation, diabetic rats. Cell Biochemistry and Function 21, 121–125. hyperglycemia causes high ROS production and, in turn, Attele, A.S., Wu, J.A., Yuan, C.S., 1999. Ginseng pharmacology. Bio- leads to high TBARS in tissues (Wolff and Dean, 1987; Das chemical Pharmacology 58, 1685–1693. et al., 2000). As in the case of hypoglycemic activity by Bhor, V.M., Raghuram, N., Sivakami, S., 2004. Oxidative damage WGLE feeding, it was assumed that TBARS levels in blood and altered antioxidant enzyme activities in the small intestine of streptozotocin-induced diabetic rats. The International Journal of Bio- were also lowered by WGLE feeding in diabetic groups. chemistry and Cell Biology 36, 89–97. In addition, direct radical scavenging activities by phenolic Brownlee, M., 2001. Biochemistry and molecular cell biology of diabetic compounds in ginseng leaves may result in lowered blood complications. Nature 414, 813–820. TBARS levels in WGLE supplemented groups (Park et al., Ceriello, A., 2000. Oxidative stress and glycemic regulation. Metabolism 1990). 49, 27–29. Chatterjea, M.N., Shinde, R., 2002. Text Book of Medical Biochemistry, As shown in the results (Table 2), activities of antioxi- fifth ed. Jaypee Brothers, Medical Publishers (P) Ltd., New Delhi, p. dant related enzymes were deteriorated by STZ-induction. 17. This may be due to a direct attack of reactive oxygen species Das, S., Vasisht, S., Snehalta, Das, N., Srivastava, M., 2000. Correlation on these enzymes. GSH-Px plays an important role in the between total antioxidant status and lipid peroxidation in hypercholes- reduction of H O in the presence of reduced glutathione terolemia. Current Science 78, 486. 2 2 Donnini, D., Zambito, A.M., Perella, G., Ambesi-Impiombat, O., Cur- (GSH). This detoxifying action of GSH-Px against H2O2 pro- cio, F., 1996. Glucose may induce cell death through a free radical- tects cell membrane against oxidative damage (Jacob, 1995). mediated mechanism. Biochemical and Biophysical Research Com- WGLE supplementation in diabetic rats protect, to a certain munications 219, 412–417. degree, further deterioration of GSH-Px and CAT in all or- Fan, Z.H., Isobe, K.I., Kiuchi, K., Nakashima, I., 1995. Enhancement gans. WGLE supplementation also increases SOD activity in of nitric oxide production from activated macrophages by a purified form of ginsenoside (Rg1). American Journal of Chinese Medicine the livers and spleens in diabetic groups whereas it signifi- 23, 279–287. cantly decreases in kidneys. Jacob, R.A., 1995. The integrated antioxidant system. Nutrition Research As in the case of blood analysis, it was found that STZ- 15, 755–766. induction caused a significant increase in TBARS in organs. Kakkar, R., Mantha, S.V., Radhi, J., Prasad, K., Kalra, J., 1997. Antiox- This increase in TBARS could be suppressed more or less idant defense system in diabetic kidney: a time course study. Life Science 60, 667–679. by WGLE feeding. The present study indicates that WGLE KFDA, 2002. Food Code. Korea Food and Drug Administration, Seoul, exerts a protective effect against lipid peroxidation by scav- Korea. 250 C.-H. Jung et al. / Journal of Ethnopharmacology 98 (2005) 245–250

Kim, J.S., Kim, K.W., Choi, K.J., Kwak, Y.K., Im, K.S., Lee, K.H., Satoh, K., 1978. Serum lipid peroxide in cerebrovascular disorders de- Chung, H.Y., 1996a. Screening of antioxidative components from termined by a new colorimetric method. Clinica Chimica Acta 90, red ginseng saponin. Korean Journal of Ginseng Science 20, 173– 37–43. 178. Shirwaikar, A., Rajendran, K., Kumar, D., Bodla, R., 2004. Antidiabetic Kim, Y.H., Park, K.H., Rho, H.M., 1996b. Transcriptional activation of the activity of aqueous leaf extract of Annona squamosa in streptozotocin- Cu,Zn-superoxide dismutase gene through the AP2 site by ginsenoside nicotinamide type 2 diabetic rats. Journal of Ethnopharmacology 91, Rb2 extracted from a medical plant, Panax ginseng. The Journal of 171–175. Biological Chemistry 271, 24539–24543. Singleton, V.L., Lamuela-Raventos, R.M., 1999. Analysis of total phe- Kumazawa, S., Hamasaka, T., Nakayama, T., 2004. Antioxidant activity of nols and other oxidation substrates and antioxidants by means propolis of various geographic origins. Food Chemistry 84, 329–339. of Folin–Ciocalteu reagent. Methods in Enzymology 299, 152– Lawrence, R.A., Burk, R.F., 1976. Glutathione peroxidase activity in 178. selenium-deficient rat liver. Biochemical Biophysical Research Com- Tomlinson, K.C., Gardiner, S.M., Hebden, R.A., Bennett, T., 1992. Func- munications 71, 952–958. tional consequences of streptozotocin-induced diabetes mellitus, with Li, T.S.C., Mazza, G., Cottrell, A.C., Gao, L., 1996. Ginsenosides in particular reference to the cardiovascular system. Pharmacological Re- roots and leaves of American ginseng. Journal of Agricultural and views 44, 103–150. Food Chemistry 44, 717–720. Waki, I., Kyo, H., Yasuda, M., Kimura, M., 1982. Effects of a hy- Lowy, O., Rosebrough, N.J., Farr, A.L., Randal, R.J., 1951. Protein mea- poglycemic component of ginseng radix on insulin biosynthesis in surement with the Folin phenol reagent. Journal of Biological Chem- normal and diabetic animals. Journal of Pharmacobio-dynamics 5, istry 193, 265–275. 547–554. Mahady, G.B., Gyllenhaal, C., Fong, H.H.S., Farnsworth, N.R., 2000. Wee, J.J., Park, J.D., Kim, M.W., Lee, H.J., 1989. Identification of pheno- Ginseng: a review of safety and efficacy. Nutrition in Clinical Care lic antioxidant components isolated from Panax ginseng. Agricultural 3, 90–101. Chemistry and Biotechnology 32, 50–56. Marklund, S., Marklund, G., 1974. Involvement of the superoxide anion Wolff, S.P., Dean, R.T., 1987. Glucose autoxidation and protein modifi- radical in the autoxidation of pyrogallol and a convenient assay for cation. The Biochemical Journal 245, 243–250. superoxide dismutase. European Journal of Biochemistry 47, 469–474. Xie, J.T., Mehendale, S.R., Wang, A., Han, A.H., Wu, J.A., Osinski, Mohamed, A.K., Bierhaus, A., Schiekofer, S., Tritschler, H., Ziegler, H., J., Yuan, C.S., 2004. American ginseng leaf: ginsenoside analy- Nawroth, P.P., 1999. The role of oxidative stress and NF (B) activation sis and hypoglycemic activity. Pharmacological Research 49, 113– in late diabetic complications. Biofactors 10, 175–179. 117. Ohkawa, H., Ohishi, N., Yagi, K., 1979. Assay for lipid peroxides in Yamasaki, K., Murakami, C., Ohtani, K., Kasai, R., Kurokawa, T., animal tissues by thiobarbituric acid reaction. Analytical Biochemistry Ishibashi, S., 1993. Effects of standardized Panax ginseng extract 95, 351–358. G115 on the d-glucose transport by Ehrlich ascites tumour cells. Phy- Ozturk, Y., Altan, V.M., Yildizoglu, A., 1996. Effects of experimental totherapy Research 7, 200–202. diabetes and insulin on smooth muscle functions. Pharmacological Yokozawa, T., Kobayashi, T., Kawai, A., Oura, H., Kawashima, Y., Reviews 48, 69–112. 1984. Stimulation of lipid and sugar metabolism in ginsenoside- Park, S.N., Choi, S.W., Boo, Y.C., Kim, C.K., Lee, T.Y., 1990. Effects Rb2 treated rats. Chemical and Pharmaceutical Bulletin 32, 2766– of flavonoids of ginseng leaves on erythrocyte membranes against 2772. singlet oxygen caused damage. Korean Journal of Ginseng Science Zhang, D., Yasuda, T., Yu, Y., Zheng, P., Kawabata, T., Ma, Y., Okada, 14, 191–199. S., 1996. Ginseng extract scavenges hydroxyl radical and protects Ragunathan, V., Sulochana, N., 1994. A new flavonol bioside from the unsaturated fatty acids from decomposition caused by iron-mediated flowers of Hibiscus vitifolius Linn. and its hypoglycemic activity. Jour- lipid peroxidation. Free Radical of Biological Medicine 20, 145– nal of Indian Chemical Society 71, 705–706. 150. Journal of Ethnopharmacology 98 (2005) 251–257

Reevaluation of risks with the use of Ficus insipida as a traditional anthelmintic remedy in the Amazon

Anders Hansson a, ∗, Julio C. Zelada b, Hugo P. Noriega c

a Ametra-Ucayali, Apartado 210, Pucallpa, Peru b Departamento de Pediatr´ıa, Hospital Regional de Pucallpa, Agust´ın Cauper 285, Pucallpa, Peru c Direcci´on, Hospital Regional de Pucallpa, Agust´ın Cauper 285, Pucallpa, Peru Received 24 May 2004; received in revised form 12 December 2004; accepted 13 December 2004

Abstract

The anthelmintic remedy oje,´ prepared latex of Ficus insipida, is still used by indigenous and local people in the Amazonian regions. However, overdosage leading to toxic reactions occurs despite the broadcasting of a clinically accepted dosage that is effective and safe. The intoxication of a 10-year-old girl in Pucallpa, who had received oje´ in a dose close to the recommended one, led us to study retrospectively the records of all hospitalized patients with toxic reactions to oje´ over a 12-year-period. The use of oje´ in and around Pucallpa was estimated. Most cases with toxic reactions, out of a total of 39 for the 12-year-period, were probably due to an overdose, defined as more than 1.5 cm3/kg; the recommended dose being 1 cm3/kg. In only five cases did toxic reactions occur at doses up to 1.5 cm3/kg, which were interpreted as idiosyncratic reactions; all of them occurred in children, and in two cases it was a severe reaction. One fatal outcome was noted among the 37 hospitalized patients. Two other fatal outcomes were observed in the 12-year-period but they occurred outside the hospital. The mortality rate is estimated to have been 0.01–0.015% among patients supposedly treated with oje´ in the area. Severe intoxication led to symptoms of cerebral edema. The main treatment was osmotic diuresis with which started in 1996. Although hypersensitivity reactions have been observed with other Ficus spp., there was apparently no such reaction in our cases. Recommendations are given so as to avoid toxic reactions from an expected continued use of oje.´ © 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Oje;´ Ficus insipida; Anthelmintic; Toxic reactions; Safety; Cerebral edema

1. Introduction moretenolactone could also be involved (Gaughran, 1976; Ayala, 1984; Hansson et al., 1986; Lopes et al., 1993; De The white latex of Ficus insipida Willdenow (Moraceae) Amorin et al., 1999). A clinical trial was carried out in 181 has been used for centuries among indigenous people and set- persons to find a dosage that was clinically effective for com- tlers in the neotropics, particularly in the Amazon region, for mon intestinal helminths in reducing the worm burden with- intestinal helminthiasis (Hansson et al., 1986; Phillips, 1990; out disturbing adverse effects (Hansson et al., 1986). The pop- Castner et al., 1998). In Peru this majestic fig tree is known as ular dosage had varied tremendously and occasional serious oje.´ Synonymous scientific names are Ficus glabrata H.B.K. adverse effects had been described. The overall prevalence and Ficus anthelmintica Martius. The pharmacologically ac- of intestinal helminthiasis has in many studies been reported tive component is thought to be the proteolytic enzyme ficin, to be up to more than 90% in Amazonian villages (Hansson in fact, a complex of sulfhydryl endopeptidases, although et al., 1986). With a dose of 1 cm3/kg (or a simplified dosage other components such as the terpenoids eloxanthine and with spoons according to age) of prepared latex taken for three consecutive mornings a reasonably effective and safe treat- ∗ Corresponding author. Tel.: +51 61 596223. ment was established and recommended for mass campaigns E-mail address: [email protected] (A. Hansson). or individual treatments (Hansson et al., 1986; Caldecott,

0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2004.12.029 252 A. Hansson et al. / Journal of Ethnopharmacology 98(2005) 251–257

1987). The regimen was optional when commercial an- As this is a retrospective study, informed consent could thelmintics such as mebendazol (previously) and albendazol not be given by the patients. The anonymity of the patients is were not available or were turned down by the patients. Peo- protected in the publication of this study. Furthermore, there ple continue using the latex in doses that are generally within was no experimental intervention as a result of this investi- the same range as the recommended dosage, although usu- gation. Consequently, there was no violation of international ally a single dose is given. However, it is apparent that some rules on bioethics. people still use very high doses. In March 2003, a 10-year-old girl was admitted to the 2.2. Plant material emergency department of the Regional Hospital in Pucallpa, an urban centre and capital of the Ucayali region in Eastern The oje´ tree is supposedly recognizable by all collectors of Peru currently with about 300,000 inhabitants. She had, to- the crude latex and also by a high proportion of the rural pop- gether with her brother and sister been given oje´ 5 h before by ulation in Ucayali. The tree has been botanically identified in her parents at a calculated dose of 1.4 cm3/kg. However, the published studies; it is Ficus insipida Willdenow, sometimes girl, in contrast to the other children, had developed symp- identified with a synonymous name (see, e.g. Tournon et al., toms of poisoning and she was unconscious with cerebral 1986). It is most improbable that collectors make a mistake edema on admittance. After 9 days she was discharged with- when choosing oje´ trees for latex tapping. There are other Fi- out any abiding disabilities. cus spp. in the area but they are generally very different with The serious character of this case made us evaluate pre- regard to visible signs, e.g. a strangler or an epiphyte instead vious cases of toxic reactions to oje´ at the regional hospital of a tree, small leaves instead of large ones, equipped with since the year 1992 and later on at the Hospital Amazonico,´ spines instead of having none, or with regard to the , an auxiliary hospital in Yarina Cocha, 10 km from Pucallpa. smell and colour of the latex. In none cases of toxic reac- We finally included other cases for the year 2003. We also tions were the preparations of oje´ consumed available. The decided to try to determine the annual number of patients in preparations of oje´ in the markets appeared to be authentic the Pucallpa area taking oje,´ as judged mainly by an estima- as judged by one of the authors who had much experience in tion of the sales of oje´ in the markets. Thus, a rough estimate preparations of oje´ from a previous investigation (Hansson of the incidence of oje´ intoxications and mortality could be et al., 1986). In that study it was also demonstrated that oje´ calculated. preparations from a market were all genuine. The white latex The aim of the study was to evaluate whether the is nearly always mixed with sugarcane brandy, “aguardiente”, ethnopharmacological use of oje´ and its later acceptance with an ethanol content of about 40%. This is done soon after by the local population in a standardized dosage could still tapping in the approximate proportion 4:1. With this prepa- be recommended in view of the increased concern about its ration the mixture can be stored for several months without safety. decomposition or fungal growth (Hansson et al., 1986). If the latex is not mixed with alcohol it easily ferments, changing in 2. Methods and materials colour and smell and is then regarded as unfit for consump- tion. Based on the above-mentioned facts, our assumption is 2.1. Hospital data that the preparations of oje´ taken by the hospitalized patients were actually based on latex from Ficus insipida although no All case records noted as “intoxicacion´ por oje”´ (oje´ poi- voucher specimen is available. soning) in the files of the Regional Hospital in Pucallpa were gone through from the year 1992, for a total period of 12 2.3. Statistical analysis years. Attention was also paid to possible records of post- mortem examinations; autopsies have regularly been per- Doses, times elapsed before toxic symptoms, and hospi- formed since the year 1994. It was judged that emergencies talization time are expressed with mean and median values caused by toxic reactions to oje´ in the Pucallpa urban area and are furthermore grouped together according to “normal” nearly always ended up at the Regional Hospital. A form dose, moderate overdose, and heavy overdose. Median val- was created to retrieve relevant information from the records ues were calculated to get a more stable measure as there including name, number of record, age, weight, dose of oje´ were 1–2 very extreme values for some parameters. Corre- in cm3 (calculated from the data in the records with, e.g. a lation coefficients have been calculated where possible and spoon corresponding to 10 cm3) and cm3/kg, preparation of their significance has been tested (Clarke, 1980). Toxic re- oje,´ other medication, time between administration and ad- actions and anatomical systems involved are expressed in verse or toxic reactions, type of reactions, treatment, number factual numbers and percentages. of days admitted and condition on discharge. An inquiry at the Social Security Hospital, EsSalud, in 2.4. Market data Pucallpa did not indicate any cases of toxic reactions to oje´ there. At the Hospital Amazonico´ an inquiry was carried out All six main markets in Pucallpa were visited once or twice for the same 12-year-period as for the Regional Hospital. in May 2003 to find where sales of oje´ took place. Apparently, A. Hansson et al. / Journal of Ethnopharmacology 98(2005) 251–257 253 no sales of oje´ took place in shops specializing in herbal Table 2 medicines at this time. Vendors selling oje´ were interviewed Systems involved in toxic reactions about the quantity sold, its preparation, recommended doses, System Children (n = 16) (%) Adults (n = 13) (%) and their knowledge of adverse effects. Central nervous system 15 (94) 11 (85) Gastrointestinal 10 (63) 9 (69) Musculoskeletal 4 (25) 1 (8) Cardiovascular 1 (6) 3 (23) 3. Results Respiratory 1 (6) 2 (15) Urinary 1 (6) 1 (8) 3.1. Hospital data

In the hospital files (from both the Regional Hospital and 8.2 h and median value was 6.0 h. The correlation coefficient the Hospital Amazonico)´ 37 patients with a diagnosis of “in- between dose and time was 0.58, p < 0.01. This positive cor- toxicacion´ por oje”´ (oje´ poisoning) were found for the 12 relation is surprising as it could be expected that a higher years from 1992 to 2003; 31 from the Regional Hospital and dose would lead to toxic reactions appearing more rapidly. 6 from the Hospital Amazonico.´ The cases were distributed However, when time was grouped according to dose interval throughout the years without any tendency to decrease or in- another picture could be seen (Table 1). At doses lower than crease; 17 cases in the first 6 years and 20 cases in the second. 1.5 cm3/kg median time was 5.0 h, at a moderate overdose There were 20 cases of children (less than 15 years old) and 7.0 h and at a heavy overdose 4.0 h. For the five non-overdose 17 of adults. The youngest child was only 1-year-old; - patients information on time was only available in three cases. thermore there was one 4-year-old child and three 5-year-old However, for these patients a relatively short time passed be- children. There were, in total, three cases with fatal outcomes, fore symptoms appeared, which might indicate some kind of two of which were not registered in the case records of the idiosyncratic reaction not directly related to the dose. hospital as they were dead on arrival. More or less complete The number of days the patients were hospitalized varied case records could be found for 29 patients: 17 children and between 1 and 12 days; records were available in 28 cases. 12 adults. Mean time was 3.9 days and median time 3.5 days. Admission It was stated to be an overdose when the single dose time was longer in cases of a heavy overdose (Table 1). The was 1.5 cm3/kg or larger, and a heavy overdose when it was five children with about the recommended doses were on 4cm3/kg or larger. This definition was made having in mind average admitted after 4.0 days, median time 2.0 days. The the safe dosage of 1 cm3/kg or less observed in many clin- correlation coefficient between dose and hospitalization time ical studies. In 19 out of 24 records, where the dosage was was 0.68, p < 0.001. This appears logical, and it concurs with stated, an overdose had been administered; in eight of these the results when the median values are compared. cases it was a heavy overdose (Table 1). The maximum dose Among the 29 patients (with available data) admitted for was 7.3 cm3/kg, noted for an adult. The highest dose for a oje´ poisoning CNS symptoms and gastrointestinal symp- child was 5.0 cm3/kg for a 12-year-old girl. Mean dose was toms dominated in both children and adults (Table 2). There 3.3 cm3/kg as was the median. In the five cases in which were slightly more musculoskeletal reactions among chil- the dose was close to the recommended one the doses were dren, and slightly more cardiovascular and respiratory re- 0.5 cm3/kg (9-year-old boy), 0.6 cm3/kg (11-year-old girl), actions among adults. The most common adverse or toxic 1.1 cm3/kg (6-year-old boy), 1.2 cm3/kg (1-year-old girl) and reactions were vomiting/nausea and irritability/psychomotor 1.4 cm3/kg (10-year-old girl). excitation (Table 3). Cerebral edema was apparent in many The time between ingestion and the appearance of ad- cases. Eleven patients were unconscious and four had fainting verse or toxic reactions varied between 20 min and 2 days in fits. There was no clear relation between dose and different 23 cases where information was available. Mean value was reactions and no obvious hypersensitivity reactions.

Table 1 Distribution of patients, time elapsed before toxic symptoms appeared and time in hospital vs. oje´ dose Oje´ dose (cm3/kg) <1.5 1.5–4.0 >4.0 Unknown Total Distribution of patients Children 5 6 2 7 20 Adults 0 5 6 6 17 Total 5 11 8 13 37 Time elapsed before symptoms appeared Median time (h) 5.0 (n =3) 7.0(n = 11) 4.0 (n =6) 4.0(n =3) 6.0(n = 23) Time in hospital Median time (days) 2.0 (n =5) 3.0(n = 11) 4.0 (n =8) 3.0(n =4) 3.5(n = 28) 254 A. Hansson et al. / Journal of Ethnopharmacology 98(2005) 251–257

Table 3 vendors were calculated to be 22 bottles (each bottle = 0.5 L) Most common toxic symptoms or signs a month. In all cases the latex had been mixed with “aguar- Symptom/sign Number of patients (n = 29) (%) diente” to prevent fermentation. There might be a few more Vomiting/nausea 14 (48) vendors and some people might get oje´ directly from oje´ col- Irritability/psychomotor excitation 14 (48) lectors or in some other way; we have supposed five bottles Unconsciousness 11 (38) a month. It was thus estimated that a total of 27 bottles were Convulsions 10 (35) administered in most months in the Pucallpa area. Accord- Diarrhea 8 (28) Headache 7 (24) ing to the vendors information sales were up 50% in January Abdominal discomfort 7 (24) to March before schools started after the holidays and when Somnolence 5 (17) more oje´ was available, as latex is more abundant in the trees Mydriasis 5 (17) in the rainy season. According to these estimates about 365 bottles (27 × 9 + 40.5 × 3) were sold per year around 2003. It is likely that the quantity has had a tendency to decrease Of the 11 patients who suffered unconsciousness 6 were because of greater access to commercial drugs. On the other children, 3 boys and 3 girls aged between 4 and 12 years. hand, the rapid population growth in Pucallpa has probably Doses for four of these children (with data available) were had an opposite effect. We therefore assume that this quan- 0.5, 1.4, 1.6 and 2.7 cm3/kg. Doses for four out of five adults 3 3 tity is representative for the whole period from 1992. Based were 2.1, 4.7, 6.0 and 7.3 cm /kg. The case with 0.5 cm /kg on the interviews and previous experience in the riverine vil- is the most alarming one. It involved a 9-year-old boy who lages we assume that the number of administrations of ojeis´ also showed symptoms of vomiting, hypotonia, mydriasis, roughly the same among children as adults. abdominal distension and psychomotor agitation. The vendors generally recommend a single dose of a quar- Treatment was according to symptoms and usually in- ter (of a bottle) for an adult and half the dose or less for a child cluded intravenous rehydration, registered in 17 of 29 cases. but rarely for children under the age of 5 years. This means When cerebral edema was diagnosed an infusion of mannitol a dose that is around 2 cm3/kg, but it is not clear whether was given, sometimes supported with furosemid as a diuretic. these recommendations are followed. The dose could also be However, the use of the mannitol as an osmotic diuretic for divided and given for 3 days. The dose should be taken in the cerebral edema was not prescribed in any cases before 1996. morning on an empty stomach and a traditional diet observed The adequate dose of mannitol was found to be 1–2 g/kg ev- during the day, avoiding, e.g. pork, chilies and salt. The dose ery sixth hour. Diazepam (or other benzodiazepine) was given should be repeated once or twice a year. For administration in 17 cases, mannitol in 15 cases, histamin-2-blockers (rani- the oje´ preparation can be mixed with orange or sugarcane tidine/famotidine) in 15 cases, antibiotics (for initial fear of juice, , or banana gruel. The vendors say they have sold sepsis) in 13 cases, dexamethasone or other steroid in 10 oje´ for many years and they know the collectors of the latex cases, anticonvulsants (phenobarbital/fenytoin) in 8 cases, well. Harvesting is mostly done from trees at the riversides antipyretics/analgesics in 8 cases, furosemid in 7 cases and or close to the river. antihypertensives in 4 cases. Other occasional treatments in- The vendors interviewed did not know of any toxic re- cluded neuroleptics, antacids, pyridoxin, oxygen, atropin and actions on the part of their respective clients. Asked about dimenhydrinate. Gastric lavage had been performed in two serious adverse reactions they, like many other people, as- cases. severate that it is due to (1) using latex from another tree, or The cases with fatal outcomes occurred in the years 1994 mixing one latex with another; (2) non-compliance with diet and 1995 (two cases). Unfortunately, no specific data are or other restrictions; (3) too high a dose. available in the last two cases when the persons were dead Based on the assumptions above ∼2000 persons ingest oje´ on arrival at the hospital; they were aged 14 and 8 years. In during a year. This figure is of course a rough estimate, but the third case, an 11-year-old boy died in the hospital of a the true figure should lie in a range between 1500 and 2500 ´ cerebral edema after 1 day; dose of oje unknown; symptoms patients per year. With these figures the incidence of severe appeared 4 h after intake. At this time, 1994, mannitol was adverse reactions and fatal reactions can be estimated. not given as a treatment for cerebral edema. 3.3. Estimate of incidence of toxic reactions 3.2. Market data A fatal outcome occurred in three cases over 12 years, Vendors of oje´ were found in two of the main markets in giving an incidence of between 0.01 and 0.015% among pa- Pucallpa, Mercado no. 2 and Bellavista. In four other markets tients supposedly treated with oje.´ Unconsciousness was ob- there was no sale of oje;´ despite questioning, it appeared as if served in at least 11 patients, giving an incidence of at least sales never occurred there and that sales of oje´ were concen- 0.06% among oje-ingesting´ persons. Toxic reactions leading trated in the two markets mentioned above. In a total of two to hospitalization occurred in 39 cases (two of which were visits to these markets 12 vendors were located and 7 of them dead on arrival), meaning an incidence of between 0.13 and were interviewed. The approximate sales of oje´ for these 12 0.2%. A. Hansson et al. / Journal of Ethnopharmacology 98(2005) 251–257 255

4. Discussion and conclusions

The therapeutic use of prepared latex of Ficus insipida, oje,´ still occurs to some extent in the city of Pucallpa includ- ing its surroundings. In distant, rural areas among indigenous ´ avez (1986) ıaz, 1992 peoples its use is more frequent, in some villages reaching ´ Landers (1984a) Giove, 1996 Landers (1984b) Ch Oporto et al., 1988 Matthews et al., 1986 D Hansson et al. (1986) up to 25% of the population, based on sporadic interviews. Cases of toxic reactions can occur as has previously been reported anecdotically (Hansson et al., 1986). In this study, the incidence of hospitalization due to intoxication was esti- mated at 0.13–0.2% of all oje´ consumers, between a half and third of these cases showing such serious reactions as uncon- sciousness. A fatal outcome is estimated to have an incidence rate of 0.01–0.015% during the 12-year-period. In most cases of toxic reactions an overdose of oje,´ more than 1.5 cm3/kg, had been administered, although in five children the dose was None 19 (5p) abdominal pain, four (1p) abdominal distension, two (4p) diarrhea, two (0p) leg edema None Six abdominal pain, two diarrhea, one allergic pruritus None One abdominal pain, one nausea, one miscarriage close to the recommended one; in these cases the reactions might be characterized as being atypical or idiosyncratic. ´ ´ e None e None ´ In the tropical regions of Peru many clinical trials of oje´ e, ´ ´ ´ e, e, were carried out in recent decades to establish an effective e, 80% oj ´ and safe anthelminthic dosage of this traditional remedy to e Adverse effects Reference ∼ be integrated into primary health programs in remote regions ´ as an alternative to commercial drugs. The study reports or e articles available to us were put together in a standardized ´ form (Table 4). Most patients belong to different indigenous e, sugar added sugar solution and orange juice added sugar added sugar added 30–70% oj sugar added peoples. The dose was generally about 1 cm3/kg, although oj in one minor study the dose was about 6 cm3/kg. In total, 395 patients received 1–7 doses of oje,´ generally in prepara-

tions with 70–85% of pure latex. There were 39 reports of 7 days Sugar solution to 70% oj ×

adverse reactions (Table 5). Gastrointestinal symptoms dom- 3 (7) days Aguardiente to 73–80% 2 × × inated. A miscarriage (at about 4 months of pregnancy and 5 1 day3 days Sugar solution to 50% oj Aguardiente to 80% oj 3 days1 day Aguardiente to 80% oj Aguardiente to ´ e Preparation of oj 3 days3 days Aguardiente to 85% oj Sugar cane juice to /kg 3 × × × × /kg 3

days after a dose of 0.25 cm /kg) occurred in an 18-year-old 3 × × /kg /kg /kg /kg girl who had not admitted being pregnant at her pretreatment 3 3 3 3 /kg /kg examination; oje´ is contraindicated in pregnant women. In 3 3 0.33 cm 1cm 6cm 3cm 1cm ∼ ∼ ∼ ∼ one of the studies placebo was used as a control (Chavez,´ ∼ 1986). Gastrointestinal reactions occurred in 10 cases with placebo (n = 70) versus 25 cases with oje(´ n = 78) (Table 4). Atypical reactions were noted in three cases; allergic itching in one case and edema of the lower extremities in two cases 2–21+ 0.08–1 cm 6–12 1 cm (Table 5). These atypical or idiosyncratic reactions, thus had 2–20 an incidence of 0.75%. In one of the studies it was clearly stated that there was no blood, microscopically or macroscopically, after treat- ment in the stool samples of the 181 patients examined for a ´ ´ e, 10 helminthic egg (larva) count (Hansson et al., 1986). In another e, 70 203, 181 completed 31control Children 1 cm of the studies there was an additional anecdotal report, involv- placebo (p) ing cases outside the anthelmintic trial, with unknown doses of oje´ where the persons had suffered generalized edema with renal insufficiency in three cases, and hematuria with severe ´ anemia in one case (Giove, 1996). e for treating intestinal helminthiasis in Peru with specific reference to dosage, preparation and reported adverse effects ın 6 Adults In this context a preclinical study in mice infected with ´ ın 20 Adults two oxiurid species and one cestod species should be men- ´ tioned (De Amorin et al., 1999). Prepared oje´ latex was ad- ministered intragastrically, 4 cm3/kg/day for 3 days. As an- ´ thelmintic activity was moderate and hemorrhagic enteritis en, Loreto N.Esperanza, Ucayali Runuya, Ucayali Dios ıo Pichis (alto), Pasco 20 oj ´ Rio Tambo, Junin 10 24–27 N. Ed Puerto Maldonado, Madre de R Puerto Arturo, Madre de Dios 27 1–15 Tarapoto, San Mart Iquitos/Nauta, Loreto 78 oj was observed at necropsy, continued clinical use of oje´ latex Table 4 Therapeutic trials with oj Place and departmentRio Tambo, Jun Patient number Age (years) Dosage of oj 256 A. Hansson et al. / Journal of Ethnopharmacology 98(2005) 251–257

Table 5 is not natural in our area. A significant cross-reactivity be- Frequency of adverse effects reported in eight therapeutic trials with ojein´ tween rubber latex, Hevea brasiliensis, and Ficus benjamina intestinal helminthiasis in a total of 395 patients and Ficus carica has also been demonstrated (Brehler et al., Adverse effect Number of patients (%) 1998; Erdmann et al., 2003). The rubber tree grows in the Abdominal pain 26 (6.6) area and prior sensitization to rubber latex could, perhaps, Abdominal distension 4 (1.0) be a possible explanation of the rare hypersensitivity reac- . Diarrhea 4 (1 0) tions observed with oje´ in other studies (Table 5). However, Edema of leg 2 (0.5) Nausea 1 (0.25) specific studies are needed to clarify such possible connec- Miscarriage 1 (0.25) tions. Ficin can be sensitizing but only after inhalation of Allergic pruritus 1 (0.25) dry powder. Sensitization by the oral and topical routes have not been noted (Gaughran, 1976). The importance of the tra- ditional diet generally followed together with an oje´ intake in traditional medicine was not recommended, apparently a must also be taken into account to explain idiosyncratic or conclusion without due regard to all the clinical trials per- hypersensitivity reactions. formed. With a continued therapeutic use of the latex of Ficus insip- In Pucallpa and its surroundings anthelmintic mass cam- ida the following precautions should be followed and broad- paigns with single-dosed albendazol are now carried out fre- cast through the health authorities in both urban and rural quently. This is a beneficial and welcome measure in an area areas and through indigenous organizations. High dosages of where intestinal helminthiasis is prevalent. The most com- oje´ should be avoided as the risk of toxic reactions increases. mon adverse effects with albendazol are non-serious and in- Oje´ should not be used with children under the age of 3 years clude: abdominal pain, nausea, vomiting, alopecia, increased and, as mentioned previously, not with pregnant women. Pre- serum aminotransferase and neutropenia (Liu and Weller, pared oje´ should only be purchased from known vendors or 1996). prepared by knowledgeable people so as not to use the poten- The use of oje´ as an anthelmintic remedy in the Ama- tially toxic latex of other tree species, which might be used zon area is conditioned by cultural, traditional and economic by unscrupulous people. Should symptoms of poisoning oc- factors as extensively discussed in a previous publication cur owing to an overdose or a rare idiosyncratic reaction, the (Hansson et al., 1986). The use will probably recede in the ur- patient should be transferred to a hospital or a health centre. ban areas with the availability of the commercial anthelmintic In severe cases with cerebral edema the use of mannitol is albendazol. However, we do not think that the treatment with recommended to save the patient. oje´ should be abandoned, particularly in rural and riverine ar- eas where registered pharmaceutical formulations are rarely available, although there is a small risk of idiosyncratic reac- tions at established therapeutic doses. In this study, toxic reac- Acknowledgements tions at therapeutical doses (< 1.5 cm3/kg), were interpreted as idiosyncratic reactions. This was observed in five cases, in We are grateful to Philip O’Brien, MA (Oxon), of the two of which there was a severe reaction with unconscious- British Centre, Pucallpa for proof-reading the manuscript for ness. However, in no case was an allergic reaction noted. Ac- any infelicities in our use of the English language. We are cording to the estimate of oje´ consumption, the occurrence also grateful to Beatriz Lujan´ and Christian Felipa for assis- of idiosyncratic reactions was in the range of 0.03–0.1% (5 tance in getting data from the Archive Departments of the or 18 cases, the latter figure with cases of unknown doses two hospitals involved. included, in 18,000 patients). The reason for the idiosyncratic reactions should be elu- cidated to discover whether they are a direct effect of the References proteolytic enzyme ficin or, possibly, the result of a hyper- sensitivity reaction to ficin or another component. Such reac- Axelsson, I.G.K., Johansson, S.G.O., Larsson, P.H., Zetterstrom,¨ O., 1990. tions have been reported after contact with another Ficus sp., Characterization of allergenic components in extract from the the ornamental indoor plant Ficus benjamina (Axelsson et weeping fig (Ficus benjamina). International Archives of Allergy and al., 1990; Bircher et al., 1995; Brehler and Theissen, 1996). Applied Immunology 91, 130–135. Axelsson, I.G.K., Johansson, S.G.O., Larsson, P.H., Zetterstrom,¨ O., 1991. However, Ficus benjamina allergens represent indoor aller- Serum reactivity to other indoor Ficus plants in patients with allergy gens. There seems to be cross-reactivity with other Ficus spp., to weeping fig (Ficus benjamina). Allergy 46, 92–98. especially those with small leaves (Axelsson et al., 1991). Ayala, F.F., 1984. Notes on some medicinal and poisonous plants of Allergic reactions to fresh or dried figs, Ficus carica, can Amazonian Peru. Advances in Economic Botany 1, 1–8. be present as a consequence of primary sensitization to air- Bircher, A.J., Langauer, S., Levy, F., Wahl, R., 1995. The allergen of Ficus benjamina in house dust. Clinical and Experimental Allergy borne Ficus benjamina allergens (Focke et al., 2003). Ficin 25, 228–233. is present in Ficus carica (Whitaker, 1958), but it is diffi- Brehler, R., Theissen, U., 1996. Ficus benjamina allergy. Hautarzt 47, cult to see a clear connection with our cases. The fig tree 780–782. A. Hansson et al. / Journal of Ethnopharmacology 98(2005) 251–257 257

Brehler, R., Abrams, E., Sedlmayr, S., 1998. Cross-reactivity between Giove, R.A., 1996. Medicina tradicional en el tratamiento de enteropara- Ficus benjamina (weeping fig) and latex. Allergy 53, sitosis. Revista Gastroenterologica´ del Peru´ 16, 197–202. 402–406. Hansson, A., Veliz, G., Naquira,´ C., Amren,´ M., Arroyo, M., Arevalo,´ G., Caldecott, J., 1987. Medicine and the fate of tropical forests. British 1986. Preclinical and clinical studies with latex from Ficus glabrata Medical Journal 295, 229–230. HBK, a traditional intestinal anthelmintic in the Amazonian area. Jour- Castner, J.L., Timme, S.L., Duke, J.A., 1998. A Field Guide to Medicinal nal of Ethnopharmacology 17, 105–138. and Useful Plants of the Upper Amazon. Feline Press, Gainesville, p. Landers, E., 1984a. Tratamiento y resultados de enteroparasitosis con 54. resina de Ficus anthelmintica (oje,´ potoc) en solucion´ al 50%. Chavez,´ R., 1986. Uso del latexdeoj´ e(´ Ficus glabrata HBK) en el Archives of Save the Children, Lima, pp. 1–2. tratamiento de helmintiasis intestinal. In: Cardenas,´ C. (Ed.), Semi- Landers, E., 1984b. Tratamiento y resultados de enteroparasitosis con nario taller sobre estudios, experiencias y aplicacion´ de la medicina resina de Ficus anthelmintica (oje,´ potoc) al 70%. Archives of Save tradicional en el ambito´ del grupo etnico´ Shipibo-Conibo del r´ıo Ucay- the Children, Lima, pp. 1–4. ali. Instituto Indigenista Peruano (Lima), FECONAU (Pucallpa) and Liu, L.X., Weller, P.F., 1996. Antiparasitic drugs. New England Journal Hospital Amazonico´ de Yarinacocha, pp. 112–122. of Medicine 334, 1178–1184. Clarke, G.M., 1980. Statistics and Experimental Design, second ed. Ed- Lopes, D., Villela, C.T., Kaplan, M.A.C., Carauta, J.P.P., 1993. Moreteno- ward Arnold, London, p. 83. lactone, a beta-lactone from Ficus insipida. Phytochemistry 34, De Amorin, A., Borba, H.R., Carauta, J.P.P., Lopes, D., Kaplan, M.A.C., 279–280. 1999. Anthelmintic activity of the latex of Ficus species. Journal of Matthews, C., Cruz, T., Revorado, G., Lacaze, D., Alexiades, M., 1986. Ethnopharmacology 64, 255–258. Uso de la resina de oje´ como antiparasitario en Madre de Dios. D´ıaz, E., 1992. Efecto de la resina de oje´ sobre la parasitosis intesti- Archives of Ametra 2001. Puerto Maldonado, pp. 1–6. nal. Tesis para optar el grado academico´ de bachiller en enfermer´ıa. Oporto, L.E., Cruz, T., Paredes, M., 1988. Informe del uso de resina de Universidad Nacional de Callao, Peru,´ pp. 1–34. oje´ en la comunidad de Pto, Arturo. Archives of Ametra 2001, Puerto Erdmann, S.M., Hipler, U.C., Merk, H.F., Raulf-Heimsoth, M., 2003. Maldonado, pp. 1–2. Sensitization to fig with cross-sensitization to weeping fig and natural Phillips, O., 1990. Ficus insipida (Moraceae): ethnobotany and ecology rubber latex. International Archives of Allergy and Applied Immunol- of an Amazonian anthelmintic. Economic Botany 44, 534–536. ogy 131, 138–142. Tournon, J., Serrano, L.G., Reategui,´ R.U., Alban,´ C.J., 1986. Plantas Focke, M., Hemmer, W., Wohrl, S., Gotz, M., Jansch, R., 2003. Cross- y arboles´ medicinales de los Conibo del alto Ucayali: concepciones reactivity between Ficus benjamina latex and fig fruit in patients with nativas y botanica.´ Revista Forestal del Peru´ 13, 107–130. clinical fig allergy. Clinical and Experimental Allergy 33, 971–977. Whitaker, J.R., 1958. The ficin content of the latex from different varieties Gaughran, E.R.L., 1976. Ficin: history and present status. Quarterly Jour- of Ficus carica and a comparison of several micro-methods of protein nal of Crude Drug Research 14, 1–21. determination. Food Research 23, 364–370. Journal of Ethnopharmacology 98 (2005) 259–266

Effects of Bambusae concretio Salicea (Chunchukhwang) on amyloid ␤-induced cell toxicity and antioxidative enzymes in cultured rat neuronal astrocytes

Ji-Cheon Jeong a, Cheol-Ho Yoon a, Woo-Hun Lee a, Kwan-Kyu Park b, Young-Chae Chang b, Yung Hyun Choi c, Cheorl-Ho Kim a,∗

a Department of Biochemistry, Molecular Biology and Internal Medicine, College of Oriental Medicine, Dongguk University and National Research Laboratory for Glycobiology, Kyungju City, Sukjang-Dong 707, Kyungbuk 780-714, Republic of Korea b Department of Pathology, College of Medicine, Catholic University of Daegu, Daegu, Republic of Korea c Department of Biochemistry, Dongeui University College of Oriental Medicine, Busan 614-052, Republic of Korea Received 6 January 2004; received in revised form 20 December 2004; accepted 20 December 2004

Abstract

Bambusae concretio Salicea (BCS; plant family name: Phyllostachys bambusoides Siebold et Zuccarinii) is a medicinal plant used in Korea for the treatment of various symptoms accompanying hypertension and cerebrovascular disorders. Previously, it was shown that BCS is an effective protectant against oxidative glutamate toxicity in the murine neuroblastoma cells and human neuroblastoma cells. Treatment with BCS increased the secretion of the non-amyloidogenic amyloid precursor protein fragment, and decreased the secretion of amyloid-␤ (A␤) peptides from neuronal cells [Jeong, J.C., Seo, Y.J., Kim, H.M., Lee, Y.C., Kim, C.H., 2003. Inhibitory effects of Bombusae concretio Salicea on neuronal secretion of Alzheimer’s ␤-amyloid peptides, a neuro-degenerative peptide. Neurochemical Research 28, 1785–1792.]. To further examine the pharmacological activity of BCS, we studied the protective effect of the water extracts on A␤25-35 peptide-induced neuronal death by microscopic observation and lactate dehydrogenase (LDH) assay, and action on antioxidative enzymes using cultured astrocyte cells. Ten ␮MA␤25-35-induced cell death was protected by the application of water extract of BCS in a dose-dependent manner, and concentrations of 1–10 ␮g/ml had a significant effect compared to exposure to A␤25-35 only. When antioxidative enzyme activities such as catalase, superoxide dismutase (SOD), glutathione peroxidase (GPx) and glutathione-S-transferase (GST) were assayed after A␤25-35 treatment, the enzymes were decreased in a similar fashion. However, those activities were enhanced by BCS treatment and this may have resulted from the potentiation of antioxidative ability by BCS. The ability of BCS to reduce cellular cytotoxicity induced by 10 ␮MA␤25-35 suggests that BCS may be a protective agent for free radical generating compounds such as A␤25-35, and that A␤25-35 is not only a potent lipid peroxide inducer, but also causes changes in antioxidative enzymes. From the results, it was concluded that BCS has a protective effect on A␤-induced neuronal death in cultured astrocyte cells through the inhibition of lipid peroxidation and protection of antioxidative enzymes. © 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Astrocyte; Alzheimer’s disease; Amyloid-␤; Bombusae concretio Salicea (Chunchukhwang); Antioxidative enzyme

Abbreviations: BCS, Bambusae concretio Salicea water extract; A␤, amyloid-␤; LDH, lactate dehydrohenase; SOD, superoxide dismutase; GPx, glutathione peroxidase; GST, glutathione-S-transferase; AD, Alzheimer’s disease; APP, amyloid precursor protein; FBS, fetal bovine serum; DMEM, Dul- becco’s modified eagle’s medium; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrasolium bromide; DMSO, dimethyl sulphoxide; TBA, 2-thiobabituric acid; TEP, 1,1,3,3-tetraethoxypropane; EDTA, ethylenediamine tetraacetic acid; SOD, superoxide dismutase; XOD, xanthine, xanthine oxidase; NBT, nitro blue tetrazolium; DETAPAC, diethylene triamine pentaacetic acid; GSH, reduced glutathione; GSSG-reductase, oxidized glutathione-reductase; NADPH, nicotinamide adenine dinucleotide phosphate; CDNB, 1-chloro-2,4-dinitrobenzene; NaN3, sodium azide; PBS, phosphate-buffered saline; MTT, 3-(4,5- dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide ∗ Corresponding author. Tel.: +82 54 770 2663; fax: +82 54 770 2281. E-mail address: [email protected] (C.-H. Kim).

0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2004.12.034 260 J.-C. Jeong et al. / Journal of Ethnopharmacology 98 (2005) 259–266

1. Introduction BCS had arisen because BCS showed comparable neuropro- tection (Jeong et al., 2003). Alzheimer’s disease (AD) is a neurodegenerative disor- This study was carried out to investigate the effect of BCS der characterized by a progressive cognitive decline resulting on cultured astrocytes, lipid peroxidation and antioxidative from selective neuronal dysfunction, synaptic loss, and neu- enzyme activities following A␤25-35 treatment. ronal cell death. The well-studied neuropathological features of AD show compacted deposits of amyloid-␤ (A␤) aggre- gates (Selkoe, 1991; Terry, 1996). A␤ is 39–43 amino acids 2. Materials and methods long and proteolytically derived from an integral membrane protein termed amyloid precursor protein (APP) (Kang et al., 2.1. Materials 1987; Kim and Suh, 1996), although the mechanism for APP processing is still unknown. There are many in vitro studies The A␤25-35 peptide was synthesized by Applied demonstrating that A␤ is directly neurotoxic and increases Biosystem’s Protein Synthesizer Model 470A (Peptron neuronal susceptibility to other toxic agents (Cotman et al., Co. Ltd., Taejon, Korea). Fetal bovine serum (FBS) and 1996; Kim and Suh, 1996; Yankner, 1996). The toxic effect penicillin–streptomycin were obtained from GIBCO-BRL of A␤ is correlated with its ability to form aggregates (Pike (Grand Island, New York, USA). Dulbecco’s Modified Ea- et al., 1995). Both oxygen species (Goodman et al., 1994) gle’s Medium (DMEM), glutamine, 3-(4,5-dimethylthiazol- and excessive Ca2+ influx (Mattson et al., 1993) are also im- 2-yl)-2,5-diphenyltetrasolium bromide (MTT), dimethyl plicated in the mechanism of A␤ neurotoxicity. In contrast, sulphoxide (DMSO), 2-thiobabituric acid (TBA), 1,1,3,3- it was also reported that A␤ promotes neurite outgrowth un- tetraethoxypropane (TEP), ethylenediamine tetraacetic acid der certain culture conditions instead of having a toxic action (EDTA), superoxide dismutase (SOD: from bovine liver), (Koo et al., 1990). xanthine, xanthine oxidase (XOD), nitro blue tetrazolium On the other hand, AD could be induced by surround- (NBT), catalase (from bovine liver), diethylene triamine pen- ing free radicals and also be protected by enzymes such as taacetic acid (DETAPAC), reduced glutathione (GSH), oxi- catalase, superoxide dismutase (SOD), glutathione peroxi- dized glutathione-reductase (GSSG-reductase), and nicoti- dase (GPx) and glutathione-S-transferase (GST) (Seifter et namide adenine dinucleotide phosphate (NADPH) were pur- al., 1988; Halliwell, 1994). Antioxidants scavenge and mini- chased from Sigma Chem. Co. (St. Louis, USA). 1-Chloro- mize the formation of oxygen-derived species and inhibit ox- 2,4-dinitrobenzene (CDNB) and sodium azide (NaN3) were idative damage induced by free radicals. They also recover obtain from Aldrich Chem. Co. (Milwaukee, WI). the level of intracellular antioxidants (vitamins, methionine, glutathione and glutathione-related minerals) (Ip et al., 1991; 2.2. Bombusae concretio Salicea and the extraction Schrauzer, 1992). Hence, these antioxidants may be particu- larly important in diminishing cumulative oxidative damage. Three hundred grams of BCS (Phyllostachys bambusoides Recently, several reports have suggested that natural dietary Siebold et Zuccarinii in plant family name) (OHC-B-3 in plants may play an antioxidative role in the prevention of herbarium record) was obtained from Oriental Herbal Cen- aging and carcinogenesis and may offer effective protection ter (OHC), Oriental Medical Hospital, Dongguk University from lipid peroxidative damage in vitro and in vivo (Ruch et College of Oriental Medicine, and extracted with 500 ml of al., 1989; Tsuda et al., 1994). Therefore, much attention has boiling water for 3 h. After the extract was centrifuged at been focused on natural antioxidants (Wang et al., 1988a,b). 7500 rpm for 30 min, the supernatant was lyophilized. For In particular, it was reported that the extract of Bam- direct use, the extract solution was stored at 4 ◦C in aliquots. busae concretio Salicea (Chunchukhwang in Korean; Phyl- lostachys bambusoides Siebold et Zuccarinii in plant fam- 2.3. Cell culture and treatment of BCS ily name; BCS) is specifically effective for cerebrovas- cular lesions and aphasia during the treatment of wind- Cortical astrocyte cultures were prepared from neonatal heat syndrome and heat-phlegm in oriental medicine (Lee, rat (1–2 day old) pups by the method of Levision and Mc- 1986). BCS has been long being used in Traditional Ko- Carthy (Levision and McCarthy, 1991). Cerebral cortex was rean Medicine for clinical treatment of degenerative neu- dissected from neonatal day 1–2 Sprague–Dawley rats and ronal disorders (Lee, 1986; Kosuwon et al., 1994). Previously, dissociated by gentle trituration. Cells were plated in six-well the pharmacological mechanism for BCS was attributed to culture plates (0.2 mg/ml in sodium borate buffer, pH 8.3) at anti-aging and sexual-reinforcing activities in experimental a density of 40,000 cells per well. After overnight incubation in vitro and in vivo systems (Chen et al., 1992; Lee and in DMEM supplemented with 20% fetal bovine serum, Chung, 1998). Recently, to delineate the mechanisms of ac- the medium was changed to serum-free defined medium tion of BCS in experimental systems of oxidative neuronal for neurons [DMEM supplemented with 2 mM glutamine, cell death, we have investigated the neuroprotective effects 1 mM pyruvate, penicillin–streptomycin–amphotericin exhibited by in different concentrations of BCS (Jeong et al., B mixture, 5 mM HEPES, 0.5% glucose, 10 ␮g/ml 2003). The idea of a novel and neuroprotective function of insulin, 30 nM sodium selenite, 20 nM progesterone, J.-C. Jeong et al. / Journal of Ethnopharmacology 98 (2005) 259–266 261

100 ␮M putrescine, and 20 ␮g/ml transferrin]. The cul- For all findings, each condition represents five separate tures were incubated at 37 ◦C in an atmosphere of 5% wells per experiment and is repeated in two or five indepen- CO2/95% room air, and the medium was replaced every dent experiments. other day. Experiments were performed on 6–7-day-old cultures. 2.5. Treatment of astrocytes with A␤ peptide Depending upon the experimental group, 1–10 ␮g/ml BCS was added (at 2 vol.% in culture medium) to or omitted from Confluent astrocytes were trypsinized and plated into T- flasks. After 16–18 h, cells were washed twice with warm 75 tissue culture flasks at a density of 5 × 106 cells/flask (for phosphate-buffered saline (PBS) and serum-free medium lipid peroxidation and antioxidative enzyme activity), or into added to the flask. Then the cells were treated with 10 ␮M 96-well plates at a density of 5 × 104 cells/well (for MTT A␤25-35 for 2 h and enzyme activities measured. A␤25- reduction assays). After 24 h, cells were washed with PBS to 35 was diluted in serum-free medium and added to the remove serum, and cultures were incubated in DMEM free cultures. FBS for an additional 12 h before addition of 0.1, 0.5, 1.0, 5.0, 10, 25, 50 and 100 ␮MA␤25-35 peptides or control 2.4. Determination of cell viability and toxicity assay buffer.

Cell viability was determined with 3-(4,5-dimethylth- 2.6. Examination of astrocyte morphology iazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) to the coloured formazan product by mitochondrial enzymes in Cell morphology was examined under a Nikon TMS (H- viable cells (Sladowski et al., 1993). Cells were cultured III) inverted microscope equipped with a Nikon FDX-35 in 24- or 96-well culture plates at a density of 10,000 photo camera. cells per well for lactate dehydrogenase (LDH) assay or 40,000 cells per well for MTT reduction assay. LDH activ- 2.7. Assay of catalase activity ities in the medium were measured by a Cytotox 96 non- radioactive cytotoxicity assay kit (Promega) according to Catalase activity was measured as the decrease in hy- the manufacturer’s instructions. The results were expressed drogen peroxide absorbance at 240 nm on a Gilford Re- as percentages of peak LDH release on complete cell lysis sponse spectrophotometer using 30 mM hydrogen peroxide (control). according to Aebi (1974). Briefly, 100 ␮l of cellular ex- The MTT reduction was measured essentially as described tract was placed on ice bath for 30 min and then for an- previously (Kim and Suh, 1996) with a slight modification. other 30 min at room temperature. Ten microliter Triton In brief, after incubating cells for 48 h with various samples, X-100 was added to the each tube. In a cuvette contain- A␤25-35, MTT (Sigma) solution in PBS was added to a final ing 200 ␮l phosphate buffer and 50 ␮l of cellular extract, concentration of 0.5 mg/ml, and the cells were further incu- was added 250 ␮l of 0.066 M H2O2 (in phosphate buffer) bated for 4 h at 37 ◦C. After incubation, the plates were cen- and decrease in optical density was measured at 240 nm trifuged at 900 × g for 10 min to obtain the resulting insolu- for 60 s. The molar extinction coefficient of 43.6 M cm−1 ble formazan precipitates. To dissolve the crystal precipitates, was used to determine catalase activity. One unit of ac- 150 or 600 ␮l of a 1:1 mix of ethanol and DMSO were added tivity is equal to the moles of H2O2 degraded/min/mg to each well. Each plate was gently shaken for approximately protein. 20 min before reading on the enzyme-linked immunosor- bent assay (ELISA) reader (measurement 570 nm, reference 2.8. Assay of superoxide dismutase activity 620 nm). Absorbance of converted dye was measured. Assay values obtained on addition of vehicle were taken as 100%, SOD was assayed by recording the inhibition of ferri- and complete inhibition of MTT reduction (0%) was defined cytochrome c reduction with xanthine and xanthine oxi- as the value obtained following addition of 0.9% Triton dase, EDTA being replaced by DETAPAC (1 mM) (Oberley X-100. and Spitz, 1984; Oyanagui, 1984). Briefly, 2.8 ml of reac- To examine whether BCS could attenuate the cytotoxicity tive mixture (xanthine 9.13 mg/200 ml distilled water, EDTA of A␤ peptide, cultures were pretreated with indicated con- 25 mg/100 ml, sodium carbonate, 2.54 g/60 ml, bovine al- centrations of 10 ␮g/ml of BCS for 4 h. Thereafter 10 ␮M bumin 30 mg/30 ml) is added to 0.1 ml sample and 50 ␮l A␤25-35 was added to cultures and incubated for 48 h. LDH xanthine oxidase (10 ␮l in 2 M ammonium sulphate), in- activity in the culture medium was determined as described cubated at 25 ◦C for 20 min and mixed with 0.1 ml copper above. To investigate the effect of pretreatment with A␤ pep- chloride (108 mg/100 ml). The colour reaction is detected at tide on the cytotoxicity induced by hydrogen peroxide or 560 nm. glutamate, cells were pretreated with 10 ␮MA␤25-35 pep- One unit of SOD activity was defined as the amount of tide for 48 h, and then 100 ␮M hydrogen peroxide or 100 ␮M enzyme needed to obtain 50% inhibition of cytochrome c glutamate was added to cultures and incubated further for 4 reduction at pH 7.8 (25 ◦C) in a 3.0 ml reaction volume at or 1 h, respectively. 560 nm using a Gilford Response spectrophotometer. 262 J.-C. Jeong et al. / Journal of Ethnopharmacology 98 (2005) 259–266

2.9. Assay of glutathione peroxidase activity

GPx activity was determined by the modified cou- pled assay developed by Paglia and Lawrence (Paglia and Valentine, 1967; Lawrence and Burk, 1976). The reaction mixture consisted of 500 ␮l phosphate buffer and 100 ␮l glu- tathione reductase (0.24 units). One hundred microliters of cellular extract was added to the reaction mixture and incu- bated at 37 ◦C for 10 min. Fifty microliters of 12 mM t-butyl hydroperoxide was added to 450 ␮l of the cellular reaction mixture and measured at 340 nm for 180 s. The molar extinc- tion coefficient of 6.22 × 103 cm−1 was used to determine the activity. The reaction was started by addition of 2.2 mM hydrogen peroxide as substrate. The change in absorbance at 340 nm was measured for 1 min on a Gilford Response spec- trophotometer. The activity was expressed as ␮M of NADPH oxidized/min/mg protein.

2.10. Assay of glutathione-S-transferase activity Fig. 1. Protective effect of BCS on neuronal cell damage induced by A␤25- 35. Rat cortical astrocyte cultures were treated for 48 h with control buffer GST activity was assayed with CDNB as substrate and (panel A), for 48 h with 1 ␮MA␤25-35 (panel B). Cells were treated with extract of BCS (10 ␮g/ml) and 1 ␮MA␤25-35 (panel C). The cells treated enzyme activity of GST towards the glutathione conjuga- with 1 ␮MA␤25-35 exhibited an activated morphology, stellate-shaped a tion of CDNB (Habig et al., 1974). Briefly, prepared TCA more spherical and phase-bright cell soma compared to the flat, polygonal supernatant was added to 1 ml of a solution containing control cells (100×). 0.1 M phosphate buffer (pH 6.5), 1 mM GSH and 1 mM CDNB. The formation of the CDNB-conjugate was followed ◦ ␤ at 340 nm (25 C) with a Gilford Response spectrophoto- toxic fragment of A had little effect on LDH release up ␮ meter. to 100 M. For examination of effects of BCS treatment, cells were treated with A␤25-35 for 2 h and further treated ␮ 2.11. Protein determination with the indicated concentration (10 g/ml) of BCS, and then LDH activities in the culture medium of cultured rat Protein was determined on each sample by the method of cortical astrocytes were assayed 48 h after treatment with Smith et al. (1985) (using bicichoninic acid), using bovine BCS. LDH releases were severely decreased in the cells, serum albumin as the standard. indicating that BCS treatment reduced the cell injury and protected the cells against A␤25-35-induced cytotoxicity 2.12. Statistical analysis (Table 1).

Standard procedures were use to calculate means and stan- 3.2. Aβ25-35 induces morphological activation of dard deviation of the mean. Mean values were compared astrocytes using Duncan’s Multiple Range Test with an SAS program (SAS Institute, Cary, NC); P < 0.05 was considered signifi- Cells, which were certificated to be approximately 98% cant. astrocytes, as assessed by positive staining for astrocyte in- termediate filament protein, GFAP (Cotman et al., 1996) were grown in serum-rich media for 24 h. They were flat 3. Results and polygonal-shaped and growing in a monolayer, a typ- ical morphology in culture (Fig. 1(A)). A␤25-35 expo- 3.1. Effect of Aβ25-35 peptide on cell cytotoxicity in sure induced neuronal degeneration appeared as swelling of cultured astrocyte cells the soma and bright pycnotic mass of nucleus (Fig. 1(B)). This indicates that cortical astrocyte cultures treated with The toxicity of A␤25-35 on astrocytes was assessed A␤25-35 exhibited a marked morphological changes, which by LDH assay. The results were expressed as a percent- show the activated morphology such as stellate-shaped age of peak LDH release obtained on complete lysis. The a more spherical and phase-bright cell soma compared A␤25-35 peptide increased LDH release by 52.6% of the to the flat, polygonal control cells. These morphologi- maximal value at 100 ␮M concentration at 4 h incubation. cal changes induced by A␤25-35 treatment were dose- A␤25-35 induced LDH release of 12.3% and 32.5%, re- dependent, with maximal activation reached at approxi- spectively, at 25 and 50 ␮M(Table 1). However, a non- mately 50 ␮M peptide. However, these signs of neuronal J.-C. Jeong et al. / Journal of Ethnopharmacology 98 (2005) 259–266 263

Table 1 LDH activity in the culture medium and MTT reduction of cultured rat cortical astrocytes at 48 h after treatment with indicated concentrations of A␤25-35 Concentration (␮M) LDH (% of maximal release) MTT reduction (% of control)

A␤25-35 A␤25-35 + BCS A␤25-35 A␤25-35 + BCS 0 1.2 ± 0.1 1.2 ± 0.1 100 100 0.1 2.6 ± 0.12 1.14 ± 0.05 58.3 ± 5.2 87.2 ± 4.7 0.5 2.6 ± 0.15 2.27 ± 0.21 55.6 ± 6.7 86.5 ± 5.4 1.0 3.2 ± 0.23 2.54 ± 0.32 50.2 ± 3.6 83.4 ± 6.8 5.0 4.0 ± 0.35 2.65 ± 0.29 49.4 ± 3.1 78.4 ± 4.6* 10 6.4 ± 0.54 3.54 ± 0.34* 45.7 ± 5.7 74.6 ± 7.6* 25 12.3 ± 1.5 5.42 ± 0.43** 45.2 ± 2.8 65.6 ± 3.6* 50 32.5 ± 2.6 13.22 ± 1.43*** 30.2 ± 2.3 51.5 ± 6.7* 100 52.6 ± 6.5 18.11 ± 2.23*** 19.4 ± 1.2 42.4 ± 2.5** The results are expressed as percentage of maximal LDH release that was obtained on complete cell lysis. For effects of BCS treatment (10 ␮g/ml at 2 vol.% in culture medium), cells were treated with A␤25-35 for 2 h and further treated with indicated concentration of BCS. MTT assay values obtained on addition of vehicle were taken as 100%, and complete inhibition of MTT reduction (0%) was defined as the value obtained following addition of 0.9% Triton X-100. Data are mean ± S.E.M. values obtained from five culture wells per experiment. ∗ P < 0.05. ∗∗ P < 0.01. ∗∗∗ P < 0.001. death were prevented by BCS (1–10 ␮g/ml). The represen- mal groups. However, BCS + A␤25-35 group was signifi- tative effect of BCS on the neuronal death at concentra- cantly increased compared to A␤25-35 treatment group, al- tion of 10 ␮g/ml was shown in Fig. 1(C), suggesting that though this increase was lower than that of BCS treatment BCS protected cells from this A␤25-35-induced neuronal group. damage. 3.4. Superoxide dismutase activity 3.3. Effect of BCS on antioxidative enzyme activities in Aβ25-35-treated astrocytes catalase activity Fig. 3 shows that A␤25-35 treatment for 2 h decreased about 2.0-fold of the normal group of the SOD activity in the The effect of BCS on catalase activity by A␤25-35 treat- cultured cell. BCS pretreatment at concentration of 10 ␮g/ml ment is shown in Fig. 2. The A␤25-35 treatment group had a slightly enhanced effect on SOD activity compared resulted in a decrease of catalase activity when compared to normal group. The SOD activity by BCS itself (10 ␮g/ml) with normal group. In contrast, BCS pretreatment (10 ␮g/ml) was significantly increased by 1.5-fold compared with normal markedly increased compared to those of untreated nor- group. Also, the enzyme activity of BCS + A␤25-35 group

Fig. 2. Effect of BCS on catalase activity in A␤25-35 treated astrocyte cells. Culture cell were prepared and incubated with BCS (10 ␮g/ml at 2 vol.% in Fig. 3. Effects of BCS on SOD activity in A␤25-35 treated astrocyte cells. culture medium). After 16–18 h, cells placed in DMEM medium without Culture cell were prepared and incubated with BCS (10 ␮g/ml at 2 vol.% serum, and then incubated with 1 ␮MA␤25-35 for 2 h. Catalase activity is in culture medium). After 16–18 h, cells placed in DMEM medium with- expressed as the moles of H2O2 degraded/min/mg protein in the axis. Results out serum, and then incubated with 1 ␮MA␤25-35 for 2 h. Results are are mean ± S.D. (N = 3). **P < 0.01, ***P < 0.001. mean ± S.D. (N = 3). *P < 0.05, **P < 0.01, ***P < 0.001. 264 J.-C. Jeong et al. / Journal of Ethnopharmacology 98 (2005) 259–266

␤ Fig. 4. Effects of BCS on GPx activity in A 25-35 treated astrocyte cells. Fig. 5. Effects of BCS on GST activity in A␤25-35 treated astrocyte cells. ␮ Culture cell were prepared and incubated with BCS (10 g/ml at 2 vol.% Culture cell were prepared and incubated with BCS (10 ␮g/ml at 2 vol.% in culture medium). After 16–18 h, cells placed in DMEM medium with- in culture medium). After 16–18 h, cells placed in DMEM medium with- ␮ ␤ out serum, and then incubated with 1 MA 25-35 for 2 h. The activity out serum, and then incubated with 1 ␮MA␤25-35 for 2 h. Results are ␮ was expressed as M of NADPH oxidized/min/mg protein. Results are mean ± S.D. (N = 3). *P < 0.05, **P < 0.01, ***P < 0.001. mean ± S.D. (N = 3). **P < 0.01, ***P < 0.001.

was significantly increased compared to A␤25-35 treatment vious paper (Lee and Chung, 1998), MDA level induced by group (P < 0.001). A␤25-35 treatment was significantly increased and the level was slightly reduced by BCS. These results of cellular cyto- 3.5. Glutathione peroxidase activity toxicity and MDA level by A␤25-35 treatment are in good agreement with that of Glascott et al. (1992, 1995). As shown in Fig. 4, GPx activity in A␤25-35 treatment Lipid peroxidation was prevented or greatly reduced by group was 1.7-fold higher than that of normal group, while the addition of antioxidants (Vit E, Vit C, DPPD or deferoxam- GPx activity was not changed by BCS itself (10 ␮g/ml). The ine) (Sharma and Buettner, 1993). For example, addition of GPx activity was significantly decreased in the BCS + A␤25- antioxidants in cell culture medium significantly reduced cell 35 group (P < 0.01) when compared with A␤25-35 treatment killing and content of intracellular antioxidants. Recently, it group. was reported that repeated oral administration of deer antler extract showed inhibitory effect on monoamine oxidase ac- 3.6. Glutathione-S-transferase activity tivity, one of the senescence-marker enzymes, reduced MDA level and increased SOD activity in the liver and brain tissues The effect of GST activity by A␤25-35 treatment and BCS of aged mice (Wang et al., 1988a,b). For plant resources, a pretreatment (10 ␮g/ml) is documented in Fig. 5. The GST large number of exogenous compounds such as quercetin, cat- activity by A␤25-35 itself was significantly increased by 4.0- echin, stilbens and resveratrol have also been shown to exhibit fold as compared to that of normal group. In addition, BCS antioxidative properties (Fauconneau et al., 1997; Murkies et pretreatment have significantly enhanced GST activity com- al., 1998; Bravo, 1998; Lien et al., 1999; Ashby et al., 2000). pared to the normal group (2.5-fold) (P < 0.001). In the case of They have been suggested to exert beneficial pharmacologi- the BCS + A␤25-35 group, the GST activity was also signif- cal effects on neurological disorders on the basis of in vitro icantly increased when compared to the A␤25-35 treatment observations (Oyama et al., 1994; Skaper et al., 1997). Sev- group (P < 0.01). eral novel antioxidant and prooxidant phenolic compounds such as 3-O-(3-methylcaffeoyl)quinic acid, 5-O-caffeoyl- 4-methylquinic acid and 3-O-caffeoyl-1-methylquinic acid 4. Discussion have been isolated from BCS (Hu et al., 2000; Kweon et al., 2001). Therefore, it was suggested that changes in MDA level The present study was carried out to investigate the effects are related to the alterations of antioxidative enzyme activity of BCS on cultured astrocyte cell system, A␤25-35-induced and these biochemical changes were recovered to normal lev- cytotoxicity and antioxidative enzyme activities in A␤25-35 els by the addition of antioxidants (Halliwell and Gutteridge, treated conditions. Cellular cytotoxicity was significantly en- 1990). hanced by addition of increasing concentrations of A␤25-35. In the present study, we have examined several enzymes to Pretreatment of BCS (10 ␮g/ml) attenuated in cell killing en- evaluate the effects of A␤25-35 treatment and pretreatment hanced by increasing concentrations of A␤25-35. In our pre- of BCS. Particular importance was placed on catalase, SOD, J.-C. Jeong et al. / Journal of Ethnopharmacology 98 (2005) 259–266 265

GPx and GST for well-known parameters as the antioxida- Cotman, C.W., Tenner, A.J., Cummings, B.J., 1996. Beta-amyloid con- tive enzymes. The results showed a decrease in the activity verts an acute phase injury response to chronic injury responses. Neu- of catalase in cultured cell by A␤25-35 treatment. However, robiological Aging 17, 284–723. Fauconneau, B., Waffo-Teguo, P., Huguet, F., Barrier, L., Decen- catalase activity by pretreatment of BCS was increased com- dit, A., Merillon, J.M., 1997. Comparative study of radical scav- pared to control groups. These results suggest that the reduced enger and antioxidant properties of phenolic compounds from Vitis catalase activity by A␤25-35 treatment possibly is due to a vinifera cell cultures using in vitro tests. Life Sciences 61, 2103– direct toxic effect of A␤25-35 or its metabolites. On the other 2110. hand, SOD activity by A␤25-35 treatment was significantly Glascott Jr., P.A., Gilfor, E., Farber, J.L., 1992. Effects of vitamin E on the killing of cultured hepatocytes by tert-butyl hydroperoxide. Molecular decreased when compared with the normal group, and the Pharmacology 41, 1155–1162. decreased activity was significantly increased by BCS pre- Glascott Jr., P.A., Gilfor, E., Farber, J.L., 1995. Relationship of the treatment (BCS + A␤25-35 group, P < 0.001). Furthermore, metabolism of vitamin C and E in cultured hepatocytes treated with the BCS pretreatment group showed a significantly increased tert-butyl hydroperoxide. Molecular Pharmacology 48, 80–88. effect on SOD activity when compared to the normal group Goodman, Y., Steiner, M.R., Steiner, M.S., Mattson, M.P., 1994. Nordi- hydroguaiaretic acid protects hippocampal neurons against amyloid (P < 0.05). In GPx, our results showed elevation in the activity beta-peptide toxicity, and attenuates free radical and calcium accumu- of GPx in A␤25-35 treated cultured cell. It has been shown lation. Brain Research 654, 171–176. that exposed to oxidant drug or materials resulted in the inac- Habig, W.H., Pabst, M.J., Jabby, W.B., 1974. Glutathione-S-transferase; tivation of GPx and these were recovered by addition of an- the first enzymatic step mercapturic acid formation. Journal of Bio- tioxidant. GST activity by A␤25-35 treatment was markedly chemistry 249, 7130–7139. Halliwell, B., 1994. Free radical, antioxidant and human disease: curiosity, increased when compared to the control group. Also, BCS cause, or consequence? Lancet 344, 721–724. pretreatment slightly enhanced GST activity above that of Halliwell, B., Gutteridge, J.M., 1990. The antioxidants of human extra- the normal group. cellular fluids. Archives of Biochemistry and Biophysics 280, 1–8. In the present study, antioxidative enzyme activities by Hu, C., Zhang, Y., Kitts, D.D., 2000. Evaluation of antioxidant and A␤25-35 treatment were decreased in a similar fashion. prooxidant activities of Phyllostachys nigra var. Henonis leaf extract in vitro. Journal of Agricultural and Food Chemistry 48, These results possibly due to adaptive phenomenon by the 3170–3176. direct toxic effect of A␤25-35 or its metabolites may cause Ip, C., Hayes, C., Budnick, R.M., Ganther, H.E., 1991. Chemical form of these effects. Furthermore, enhanced activities by BCS pre- selenium, critical metabolites, and cancer prevention. Cancer Research treatment may be resulted from the potentiation of antioxida- 51, 595–600. tive ability. In summary, the ability of BCS to reduce cellular Jeong, J.C., Seo, Y.J., Kim, H.M., Lee, Y.C., Kim, C.H., 2003. In- ␤ hibitory effects of Bombusae concretio Salicea on neuronal secre- cytotoxicity and MDA level induced by A 25-35 suggests tion of Alzheimer’s ␤-amyloid peptides, a neuro-degenerative peptide. that BCS may be a protective agent for free radical gener- Neurochemical Research 28, 1785–1792. ating compounds such as A␤25-35, although we cannot in Kang, J., Lemaire, H.G., Unterbeck, A., Salbaum, M.N., Masters, C.L., detail explain to effects of BCS at present. We are now in Grzeschik, K.H., Multhaup, G., Beyreuther, K., Muleller-Hill, B., progress to isolate the active molecules and investigate the 1987. The precursor of Alzheimer’s disease amyloid A4 protein re- sembles a cell surface receptor. Nature 325, 733–736. pharmacological mechanism. Kim, C.H., Suh, Y.H., 1996. Neurotoxicity of a carboxyl-terminal frag- ment of the Alzheimer’s amyloid precursor protein. Journal of Neu- rochemistry 67, 1172–1182. Acknowledgments Koo, E.H., Park, L., Selkoe, D.J., 1990. Amyloid beta-protein as a sub- strate interaction with extracellular matrix to promote neurite out- growth. Proceedings of the National Academy of Sciences USA 90, This study was supported by Dongguk University Re- 4748–4752. search Fund and National Research Laboratory Program Kosuwon, W., Laupattarakasem, W., Saengnipanthkul, S., Mahaisavariya, (2002–2007) from Korean Ministry of Science and Technol- B., Therapongpakdee, S., 1994. Charcoal bamboo as a bone substi- ogy (C-H. Kim). tute: an animal study. Journal of Medical Association of Thailand 77, 496–500. Kweon, M.H., Hwang, H.J., Sung, H.C., 2001. Identification and an- tioxidant activity of novel chlorogenic acid derivatives from bamboo References (Phyllostachys edulis). Journal of Agricultural and Food Chemistry 49, 4646–4655. Aebi, H., 1974. In: Begmeyer, H.U. (Ed.), Catalase: Methods of Enzy- Lawrence, R.A., Burk, R.F., 1976. Glutathione peroxidase activity in matic Analysis, 2nd ed. pp. 673–684. selenium-deficient rat liver. Biochemical and Biophysical Research Ashby, J., Odum, J., Paton, D., Lefevre, P.A., Beresford, N., Sumpter, Communication 71, 952–958. J.P., 2000. Re-evaluation of the first synthetic estrogen, 1-keto-1,2,3,4- Lee, S.I., 1986. Clinical Application of Oriental Medicinal Herbs. Sungbo tetrahydrophenanthrene, and bisphenol A, using both the ovariec- Publishing Co., Seoul, Korea, pp. 453–497. tomised rat model used in 1933 and additional assays. Toxicology Lee, W.H., Chung, J.C., 1998. Effects of Bombusae concretio Salicea on Letters 115, 231–238. Amyloid-␤-induced neuronal cell toxicity and lipid peroxidation in Bravo, L., 1998. : chemistry, dietary sources, metabolism, and cultured rat astrocytes. Journal of Korean Oriental Internal Medicine nutritional significance. Nutritional Reviews 56, 317–333. 19, 381–391. Chen, X., Jia, Y., Wang, B., 1992. Inhibitory effects of the extract of Levision, S.W., McCarthy, K.D., 1991. Astroglia in culture. In: Banker, Bombusae concretio Salicea on monoamine oxidase in aged mice. G., Goslin, K. (Eds.), Culturing Nerve Cells. MIT Press, Cambridge, Chung Kuo Chung Yao Tsa Chih 17, 107–110. MA, pp. 309–336. 266 J.-C. Jeong et al. / Journal of Ethnopharmacology 98 (2005) 259–266

Lien, E.J., Ren, S., Bui, H.H., Wang, R., 1999. Quantitative Selkoe, D.J., 1991. The molecular pathology of Alzheimer’s disease. Neu- structure–activity relationship analysis of phenolic antioxidants. Free ron 6, 487–498. Radical Biology and Medicine 26, 285–294. Sharma, M.K., Buettner, G.R., 1993. Interaction of vitamin C and vitamin Mattson, M.P., Tomaselli, K.J., Redel, R.E., 1993. Calcium destabilizing E during free radical stress in plasma: an ESR study. Free Radical and neurodegenerative effects of aggregated beta-amyolid peptide are Biology and Medicine 14, 649–653. attenuated by basic FGF. Brain Research 621, 35–49. Skaper, S.D., Fabris, M., Ferrari, V., Dalle Carbonare, M., Leon, A., Murkies, A.L., Wilcox, G., Davies, S.R., 1998. Clinical review 92: phy- 1997. Quercetin protects cutaneous tissue-associated cell types in- toestrogens. Journal of Clinical Endocrinology and Metabolism 83, cluding sensory neurons from oxidative stress induced by glutathione 297–303. depletion: cooperative effects of ascorbic acid. Free Radical Biology Oberley, L.W., Spitz, P.R., 1984. In: Colowick, S.P., Kaplan, N.O., Packer, and Medicine 22, 669–678. L. (Eds.), Methods in Enzymology, vol. 105. Academic Press, pp. Sladowski, D., Steer, S.J., Clothier, R.H., Balls, M., 1993. An im- 457–464. proved MTT assay. Journal of Immunological Methods 157, 203– Oyanagui, Y., 1984. Reevaluation of assay methods and establishment 207. of kit for superoxide dismutase activity. Analytical Biochemistry 42, Smith, P.K., Krohn, R.I., Hermanson, G.T., Mallia, A.K., Gartner, F.H., 290–296. Provenzano, M.D., Fujimoto, E.K., Goeke, B.J., Klenk, D.C., 1985. Oyama, Y., Fuchs, P.A., Katayama, N., Noda, K., 1994. Myricetin and Measurement of protein using bicichoninic acid. Analytical Biochem- quercetin, the flavonoid constituents of biloba extract, greatly istry 150, 76–85. reduce oxidative metabolism in both resting and Ca(2+)-loaded brain Terry, R.D., 1996. Neuropathological changes in Alzheimer’s disease. neurons. Brain Research 635, 125–129. Progress in Brain Research 101, 193–203. Paglia, D.E., Valentine, W.N., 1967. Studies on quantitative an qualita- Tsuda, T., Watanabe, M., Ohshima, K., Norinobu, S., Choi, S.W., Kawak- tive characterization of erythrocyte glutathione peroxidase. Journal of ishi, S., Osawa, T., 1994. Antioxidative activity of the anthocyanin Laboratory and Clinical Medicine 70, 158–169. pigments cyanidin 3-O-d-glucoside and cyanidin. Journal of Agricul- Pike, C.J., Walencewicz-Wasserman, A.J., Kosmoski, J., Cribbs, D.J., tural and Food Chemistry 42, 2407–2410. Glabe, C.G., Cotman, C.W., 1995. Structure–activity analyses of ␤- Wang, B.X., Zhao, X.H., Qi, S.B., Kaneko, S., Hattori, M., Namba, amyloid peptides: contributions of the ␤25-35 region to aggregation T., Nomura, Y., 1988a. Effects of repeated administration of and neurotoxicity. American Journal of Pathology 142, 17–24. deer antler extract on biochemical changes related to aging in Ruch, R.J., Cheng, S.J., Klaunig, J.E., 1989. Prevention of cytotoxicity senescence-accelerated mice. Chemical and Pharmaceutical Bulletin and inhibition of intracellular communication by antioxidant catechins 36, 2587–2592. isolated from Chinese green tea. Carcinogenesis 10, 1003–1008. Wang, B.X., Zhao, X.H., Qi, S.B., Yang, X.W., Kaneko, S., Hat- Schrauzer, G.N., 1992. Selenium: mechanistic aspects of anticarcinogenic tori, M., Namba, T., Nomura, Y., 1988b. Stimulating effect of action. Biology of Trace Element Research 33, 51–62. deer antler extract on protein synthesis in senescence accelerated Seifter, E., Mendecki, J., Holtzman, S., Kanofsky, J.D., Freidenthal, E., mice in vivo. Chemical and Pharmaceutical Bulletin 36, 2593– Davis, L., Weinzweig, J., 1988. Role of vitamin A and -carotene in ra- 2598. diation protection; relation to antioxidant properties. Pharmacological Yankner, B.A., 1996. Mechanisms of neuronal degeneration in Therapy 39, 357–365. Alzheimer’s disease. Neuron 16, 921–932. Journal of Ethnopharmacology 98 (2005) 267–273

Anti-inflammatory, analgesic and antipyretic effects of methanol extract from Bauhinia racemosa stem bark in animal models M. Gupta ∗, U.K. Mazumder, R. Sambath Kumar, P. Gomathi, Y. Rajeshwar, B.B. Kakoti, V. Tamil Selven

Department of Pharmaceutical Technology, Jadavpur University, Kolkata 700 032, India Received 5 March 2004; received in revised form 30 December 2004; accepted 5 January 2005

Abstract

In this study, the anti-inflammatory, analgesic, and antipyretic effects of 50, 100 and 200 mg/kg body weight of methanol extract obtained from Bauhinia racemosa stem bark, the so-called MEBR, were investigated. The effects of MEBR on the acute and chronic phases of inflammation were studied in carrageenan, dextran and mediators (histamine and serotonin)-induced paw oedema and cotton pellet-induced granuloma, respectively. Analgesic effect of MEBR was evaluated in acetic acid-induced writhing and hotplate tests. Antipyretic activity of MEBR was evaluated by yeast-induced hyperpyrexia in rats. The anti-oedema effect of MEBR was compared with 10 mg/kg of indomethacin orally. In acute phase of inflammation, a maximum inhibition of 44.9, 43.2, 44.8 and 45.9% (P < 0.001) was noted at the dose of 200 mg/kg b.w. after 3 h of treatment with MEBR in carrageenan, dextran, histamine and serotonin-induced paw oedema, respectively. Administration of MEBR (200 mg/kg b.w.) and indomethacin (10 mg/kg b.w.) significantly (P < 0.05) decreased the formation of granuloma tissue induced by cotton pellet method at a rate of 50.4 and 56.2%, respectively. The extract also inhibited peritoneal leukocyte migration in mice. The MEBR also produced significant (P < 0.01) analgesic activity in both models. Further, the MEBR potentiated the morphine- and aspirin-induced analgesic in mice. Treatment with MEBR showed a significant (P < 0.01) dose-dependent reduction in pyrexia in rats. The results suggest that MEBR possess potent anti-inflammatory, analgesic and antipyretic activity. © 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Bauhinia racemosa; Anti-inflammatory; Analgesic; Antipyretic

1. Introduction fresh flower of this plant were screened for antiulcer ac- tivity (Akhtar and Ahmad, 1995). Dried entire plant showed Bauhinia racemosa L. (Caesalpiniaceae), a small crooked antimicrobial activity (Ali et al., 1999). The cytotoxic, hy- tree with dark scabrous bark, is widely distributed through- potensive and hypothermic activities of seeds of Bauhinia out India, Ceylon, China and Timor. The bark and leaves racemosa have also been reported (Dhar et al., 1968). of this plant are reported to be medicinally important in Several chemical constituents of Bauhinia racemosa have the traditional system of medicine and are used extensively been identified mainly as flavonols, coumarins, triterpenoids, for the treatment of inflammation, headache, fever, tumors, stilbenes, steroids and tannins (El-Hossary et al., 2000; skin infection, disease of the blood, dysentery, and diarrhoea Prakash and Khosa, 1976; Anjaneyulu et al., 1984, 1986; (Kirtikar and Basu, 1975; Wealth of India, 1952). The ethanol Balasooriya et al., 1982). extract of leaves of this plant was evaluated for analgesic, anti- Previous results from this laboratory have also demon- inflammatory, antipyretic and antispasmodic activity and was strated the antioxidant and hepatoprotective effects in rats reported to be active (El-Khatiba and Khaleel, 1995). The (Gupta et al., 2004a), and antitumor and antioxidant activity of Bauhinia racemosa against Ehrlich ascites carcinoma in ∗ Corresponding author. Tel.: +91 33 24404123; fax: +91 33 24146967. Swiss albino mice (Gupta et al., 2004b). The present study E-mail address: [email protected] (M. Gupta). was focussed on anti-inflammatory, analgesic and antipyretic

0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2005.01.018 268 M. Gupta et al. / Journal of Ethnopharmacology 98 (2005) 267–273 effects of methanol extract of Bauhinia racemosa (MEBR) was calculated from the graph of percent mortality against stem bark in animal models. probit log dose of the extract.

2.5. Anti-inflammatory activity 2. Materials and methods 2.5.1. Carrageenan-induced rat paw oedema 2.1. Plant material The rats were divided into five groups (n = 6). The differ- ent groups were treated with MEBR (50, 100 and 200 mg/kg The stem bark of the plant Bauhinia racemosa L. b.w., p.o.), indomethacin (10 mg/kg, p.o.) and vehicle control (Family—Caesalpiniaceae) was collected from Kolli Hills (10% propylene) p.o. and the paw volume was measured at of TamilNadu, India. The plant material was taxonomically 0 h and 3 h after carrageenan injection using plethysmome- identified by the Botanical Survey of India, Kolkata. A ter (Winter and Porter, 1957). The animals were pretreated voucher specimen (No GMS-1) has been preserved in our with the extract 1 h before the administration of carrageenan. laboratory. The stem barks were dried under shade and then Acute inflammation was produced by the subplantar admin- powdered with a mechanical grinder and stored in airtight istration of 0.1 ml of 1% carrageenan in normal saline in the container. The dried powder material of the bark was ex- right paw of the rats. The anti-inflammatory effect of MEBR tracted with methanol (yield 9.25%), in a soxhlet appara- was calculated by the following equation: Anti-inflammatory tus. Phytochemical screening of the extracts revealed the activity (%) = (1 − D/C) × 100, where D represents the per- presence of flavonoids, triterpenoids, coumarins, tannins and centage difference in paw volume after MEBR was admin- steroids. istered to the rats and C represents the percentage difference of volume in the control groups (Suleyman et al., 1991). 2.2. Chemicals and drugs 2.5.2. Dextran-induced paw oedema Carrageenan (S.D. Fine Chemicals Limited, Bombay), The animals were treated in a manner similar to that of 5-hydroxytryptamine hydrochloride (serotonin), histamine carrageenan-induced paw oedema models; dextran (0.1 ml, (Sigma, USA) and dextran (Sigma, USA) were used in 1% w/v in normal saline) was used in the place of carrageenan the study, and indomethacin (Recon Bangalore), aspirin (Winter and Porter, 1957). (USV Bombay), paracetamol (IPCA, Bombay) and mor- phine (M.M. Pharma, New Delhi) were used as the standard 2.5.3. Histamine- and serotonin-induced inflammation drugs. The anti-inflammatory activity of the MEBR was mea- sured with phlogistic agents (viz. histamine, 5-HT) which act 2.3. Animals as mediator of inflammation. The paw oedema was induced in rats by subplantar injection of freshly prepared histamine Studies were carried out using male Wistar albino rats (1 mg/kg b.w.) and serotonin (1 mg/kg b.w.) solutions, re- weighing 180–200 g and male Swiss albino mice weighing spectively, and the paw oedema was measured as mentioned 18–22 g. They were obtained from the animal house, In- earlier (Winter et al., 1962). dian Institute of Chemical Biology (IICB), Kolkata, India. The animals were grouped and housed in polyacrylic cages 2.5.4. Cotton pellets-induced granuloma (38 cm × 23 cm × 10 cm) with not more than six animals per The rats were divided into five groups (n = 6). After shav- cage and maintained under standard laboratory conditions ing the fur, the rats were anaesthetized and 10 mg of ster- (temperature 25 ± 2 ◦C) with dark and light cycle (14/10 h). ile cotton pellets were inserted, one in each axilla. The They were allowed free access to standard dry pellet diet MEBR (50, 100 and 200 mg/kg b.w., p.o.) and indomethacin (Hindustan Lever, Kolkata, India) and water ad libitum. The (10 mg/kg b.w., p.o.) and control vehicle were administered rats were acclimatized to laboratory condition for 10 days be- orally for seven consecutive days from the day of cotton pellet fore commencement of experiment. All procedures described implantation. The animals were anaesthetized on the eighth were reviewed and approved by the University animal ethical day and cotton pellets were removed surgically and made committee. free from extraneous tissues. The moist pellets were weighed and then dried at 60 ◦C for 24 h, after that dried pellets were 2.4. Toxicity study weighed again. Increment in the dry weight of the pellets was taken as measure of granuloma formation. The antiprolifer- The LD50 was determined using the graphical method of ative effect of MEBR was compared with the control. Litchfield and Wilcoxon (1949), in mice. Briefly, geometric doses of the extract (100–1750 mg/kg) were administered i.p. 2.6. Mouse carrageenan peritonitis to 10 groups of mice. Control group received normal saline (5 ml/kg i.p.). Signs of toxicity and mortality within 24–72 h The mice were divided into five groups (n = 6). Inflam- were noted. Confirmatory test was carried out and the LD50 mation was induced by modification of the technique as M. Gupta et al. / Journal of Ethnopharmacology 98 (2005) 267–273 269 previously described (Griswold et al., 1987). The extract was brewer’s yeast in the back below the nape of the rat (Al- administered orally at doses of 50, 100 and 200 mg/kg and Ghamdi, 2001). The animals were then fasted for the indomethacin at a dose of 10 mg/kg p.o., and carrageenan duration of the experiment, water ad libitum. Control (0.25 ml, 0.75% in saline) was injected intraperitoneally 1 h temperatures were taken 24 h after the yeast injection to later and after 4 h the animals were sacrificed by cervical determine the pyretic response of yeast. Temperatures taken dislocation for further investigation. Ca2+ and Mg2+ free 1 h prior to drug administration in fevered animal served phosphate buffered saline was used during the collection of as the pre-drug control. Extract (50, 100 and 200 mg/kg peritoneal fluids. The total leukocyte count was determined b.w.) and paracetamol (150 mg/kg b.w.) served as the in a Neubauer chamber and the differential cell count was de- reference drug given orally 24 h after the yeast injection. termined (Wintrobe et al., 1961; D’Amour et al., 1965). The The temperatures were recorded at 1–4 h after the drug treat- percentage of the leukocyte inhibition = (1 − T/C) × 100, ment. where T represents the treated groups leukocyte counts and C represents the control groups leukocyte counts. 2.9. Statistical analysis Neutrophils changes were calculated by the follow- ing equation: Neutrophils changes = Neutrophils counts of Values were expressed as mean ± S.E.M. Statistical sig- treated groups/Neutrophils counts of control groups × 100. nificance was determined by ANOVA, followed by Student’s t-test; values with P < 0.05 were considered as statistically 2.7. Analgesic activity significant. This was investigated by monitoring test animals exposed to chemical and thermal stimuli. 3. Results 2.7.1. Acetic acid-induced writhing response in mice 3.1. Toxicity study Eight groups of mice were selected for the present study. Group one received acetic acid solution (300 mg/kg) The LD value of MEBR was estimated to be (control) and group two received standard drug aspirin 50 955.04 mg/kg (915.80–984.75 mg/kg) body weight i.p. in (100 mg/kg). Remaining six different groups of mice mice. received 50, 100 and 200 mg/kg b.w. of MEBR (i.p.) and the combination of these doses of extract with the standard drug (aspirin 100 mg/kg b.w.). Acetic acid solution (15 mg/ml) 3.2. Anti-inflammatory studies at the dose of 300 mg/kg b.w. was injected i.p. and the number of writhes during the following 30-min period The anti-inflammatory activity of MEBR was measured was observed (Turner, 1965). A significant reduction in the at the dose of 50, 100 and 200 mg/kg b.w. against acute paw number of writhes by drug treatments as compared to vehicle oedema induced by carrageenan, dextran and mediators (his- control animals was considered as a positive analgesic re- tamine and serotonin), and is summarized in Fig. 1. The sponse. The percentage inhibition of writhing was then cal- MEBR produced significant (P < 0.001) anti-inflammatory culated. activity and the results were comparable to that of in- domethacin as a standard anti-inflammatory drug. MEBR at 2.7.2. Hot plate reaction time in mice the doses 50, 100 and 200 mg/kg showed an inhibition (28.1, Eight groups of mice were selected for the present study. 39.3 and 44.9%), (26.1, 37.5 and 43.2%), (27.6, 39.1 and Group one received normal saline (5 ml/kg b.w.) (Control) 44.8%) and (28.2, 40.0, 45.9%) against acute paw oedema- and group two received standard drug morphine (5 mg/kg). induced by carrageenan, dextran, histamine and serotonin, Remaining six different groups of mice received 50, 100 and respectively. 200 mg/kg b.w. of MEBR (i.p.) and the combination of these The effects of MEBR and indomethacin on the pro- doses of extract with the standard drug morphine (5 mg/kg liferative phase of inflammation are summarized in b.w. i.p.). Mice were screened by placing them on a hot plate Table 1. It was seen that MEBR was responsible for maintained at 55 ± 1 ◦C and recorded the reaction time in anti-inflammatory effect, which would be calculated de- seconds for licking of hind paw or jumping (Turner, 1965). pending on the moist and dry weight of cotton pellets. The mice which reacted within 15 s and which did not show According to these results, the antiproliferative effects large variation when tested on four separated occasions were of MEBR (200 mg/kg b.w.) and indomethacin (10 mg/kg selected for studies. b.w.) were calculated as 44.8 and 51.5% (P < 0.05), respec- tively. After they were dried, the antiproliferative effects 2.8. Yeast-induced hyperpyrexia in mice were calculated on the basis of dry weight pellets; the inhibition of inflammation by MEBR and indomethacin Hyperpyrexia was induced in rat by subcutaneous were established as 50.4 and 56.2% (P < 0.05), respec- injection of 10 ml/kg b.w. of a 15% aqueous suspension of tively. 270 M. Gupta et al. / Journal of Ethnopharmacology 98 (2005) 267–273

Fig. 1. Effect of methanol extract of Bauhinia racemosa (MEBR) stem bark on carrageenan-, dextran-, histamine- and serotonin-induced pedal edema in rats. Difference of mean of edema volume (ml) between control and treatment values at different doses ±S.E.M. Variation compared to the control animals. ANOVA followed by Student’s t-test, *P < 0.001.

3.3. Mouse carrageenan peritonitis 100 mg/kg b.w. exhibited significant (P < 0.01) inhibition of the control writhes at the rate of 20.58, 36.71, 53.41 and The MEBR also inhibited peritoneal leukocyte migration 66.72%, respectively in the acetic acid-induced writhing. at the rate of 36.13, 58.14 and 78.51% at the doses of 50, In addition, MEBR at the different doses also potenti- 100 and 200 mg/kg, respectively, whereas the inhibition pro- ated (71.36, 77.04 and 84.02%) the aspirin-induced analge- duced by indomethacin (10 mg/kg) 60.81% was found to sia. be in carrageenan-induced peritonitis model as shown in Table 1. 3.5. Hot plate reaction time in mice 3.4. Acetic acid-induced writhing in mice As shown in Table 3, the MEBR produced significant Analgesic effects induced by different doses of MEBR (P < 0.01) analgesic activity at all the tested doses. Addi- on the writhing test in mice are shown in Table 2. MEBR tionally, MEBR at different doses potentiated the analgesic at the dose of 50, 100 and 200 mg/kg b.w. and aspirin activity of morphine (5 mg/kg b.w.).

Table 1 Effect of the methanol extract of Bauhinia racemosa stem bark on cotton pellets-induced granuloma in rats and on leukocytes migration in peritoneal exudation in carrageenan-induced mice Treatment Dose Weight of Percentage Weight of Percentage Leukocytes Leukocytes Neutrophils Neutrophils (mg/kg) cotton pellet of cotton pellet of (105 mL−1) inhibition 105 mL−1 changes (mg) (moist) inhibition (mg) (Dried) inhibition Control 203.51 ± 15.9 – 46.82 ± 1.9 4.18 ± 0.34 – 2.50 ± 0.41 – Indomethacin 10 98.42 ± 8.7* 51.46 20.54 ± 0.7* 56.23 2.70 ± 0.14* 60.81 0.84 ± 0.47* 33.63 MEBR 50 152.32 ± 12.1 25.15 33.43 ± 1.1* 28.64 2.67 ± 0.11* 36.13 0.92 ± 0.05** 36.82 MEBR 100 126.71 ± 10.9* 37.73 28.14 ± 0.9* 40.02 1.75 ± 0.14** 58.14 0.29 ± 0.07** 11.64 MEBR 200 112.33 ± 11.6* 44.81 23.21 ± 0.6* 50.41 0.90 ± 0.08** 78.51 0.21 ± 0.08* 8.43 Values are mean ± S.E.M. (n = 6). ∗ Experimental groups were compared with control (P < 0.01). ∗∗ Experimental groups were compared with control (P < 0.05). M. Gupta et al. / Journal of Ethnopharmacology 98 (2005) 267–273 271

Table 2 Effect of the methanol extract of Bauhinia racemosa (MEBR) stem bark on writhing induced by acetic acid in mice Treatment Dose (mg/kg) Number of writhes (per 30 min) Percentage of inhibition Acetic acid control 300 33.42 ± 2.75 – Aspirin 100 11.12 ± 1.19* 66.72 MEBR 50 26.54 ± 2.13* 20.58 MEBR 100 21.15 ± 1.12* 36.71 MEBR 200 15.57 ± 1.17 53.41 MEBR + Aspirin 50 + 100 9.57 ± 0.79* 71.36 MEBR + Aspirin 100 + 100 7.67 ± 0.71 77.04 MEBR + Aspirin 200 + 100 5.34 ± 0.49* 84.02 Values are mean ± SEM (n = 6). ∗ Experimental groups were compared with control (P < 0.01).

Table 3 Effect of the methanol extract of Bauhinia racemosa (MEBR) stem bark on hot plate reaction time in mice Mean latent time Treatment Dose (mg/kg) Initial After 30 min Normal (saline 5 ml/kg) 9.43 ± 0.83 9.36 ± 0.74 Morphine 5 9.03 ± 0.72 19.78 ± 1.12* MEBR 50 9.05 ± 0.74 14.13 ± 1.32* MEBR 100 9.37 ± 0.84 15.27 ± 1.21* MEBR 200 9.21 ± 0.65 16.57 ± 1.34 MEBR + Morphine 50 + 5 8.74 ± 0.65 22.15 ± 1.94* MEBR + Morphine 100 + 5 8.93 ± 0.71 24.74 ± 1.75* MEBR + Morphine 200 + 5 8.64 ± 0.69 27.96 ± 1.87* Values are mean ± SEM (n = 6). ∗ Experimental groups were compared with control (P < 0.01).

Fig. 2. Effect of methanol extract of Bauhinia reacemosa (MEBR) stem bark on yeast-induced pyrexia in rats. The results are given are mean ± S.E.M.; number of animals used (n = 6). *P < 0.01 compared to control by ANOVA. 272 M. Gupta et al. / Journal of Ethnopharmacology 98 (2005) 267–273

3.6. Yeast-induced hyperpyrexia in mice sites and accumulation of collagen and mucopolysaccha- ride. Subcutaneous injection of yeast suspension markedly el- evated the rectal temperature after 24 h of administration. 4.2. Mouse carrageenan peritonitis Treatment with the MEBR at the doses of 50, 100 and 200 mg/kg significantly (P < 0.01) decreased the rectal tem- Leukocyte aggregation at the site of inflammation is a perature of the rats in a dose-dependent manner. The an- fundamental event in the inflammatory process. Cell mi- tipyretic effect started as from the first hour and the effect gration occurs as a result of much different process in- was maintained for 4 h, after administration of the extract. cluding adhesion and cell mobility (Meade et al., 1986). The result obtained from both the standard and MEBR treated In the present investigation we compared the effect of the rats were compared with that of control and a significant re- extract at the doses of 50, 100 and 200 mg/kg b.w. and in- duction in the yeast-induced elevated rectal temperature was domethacin on the cell migration. The MEBR also inhibited observed (Fig. 2). the carrageenan-induced leukocyte migration in peritonitis model in mice. The MEBR was found to inhibit leukocyte migration more potent than indomethacin. The extract (in 4. Discussion peritonitis model) drastically reduced the migration of neu- trophils. 4.1. Anti-inflammatory activity

The MEBR showed a dose-dependent anti-oedematogenic 4.3. Analgesic activity effects on paw oedema induced by carrageenan at 3 h. Dextran-induced paw oedema is known to be mediated both In acetic acid-induced abdominal writhing which is the by histamine and serotonin. The MEBR also exhibited signif- visceral pain model, the processor releases arachidonic acid icant (P < 0.001) anti-inflammatory in dextran-induced paw via cyclooxygenase, and prostaglandin biosynthesis plays a oedema. role in the nociceptive mechanism (Franzotti et al., 2002). Histamine is one of the important inflammation media- Results of the present study show that all the doses of the tors and it is a potent vasodilator substance and increases MEBR produced significant analgesic effect and this effect the vascular permeability (Linardi et al., 2002; Cuman et al., may be due to inhibition of the synthesis of the arachidonic 2001). This study showed that all the doses of MEBR effec- acid metabolite. In addition, MEBR potentiates the analgesic tively suppressed the oedema produced by histamine, so it activity of aspirin. may be suggested that its anti-inflammatory activity is pos- The hot plate test has been found to be suitable for evalua- sibly backed by its antihistaminic activity. The MEBR also tion of centrally acting analgesics. The validity of this test has effectively suppressed the inflammation produced by sero- been shown even in the presence of substantial impairment of tonin induced by hind paw edema, which indicates that the motor performance (Plummer et al., 1996). The present study MEBR may exhibit its anti-inflammatory action by means of findings indicate that the MEBR may be centrally acting. either inhibiting the synthesis, release or action of inflamma- tory mediators viz. histamine, serotonin and prostaglandins 4.4. Antipyretic that might be involved in inflammation. From the above re- sults it is suggested that the anti-oedematogenic effects of Fever may be a result of infection or one of the sequelae MEBR on carrageenan, dextran and mediators-induced paw of tissue damage, inflammation, graft rejection, or other dis- oedema may be related to inhibition of inflammation media- ease states. Antipyretic are drugs, which reduce the elevated tor formation. body temperature. Regulation of body temperature requires a The cotton pellet granuloma method is widely used to delicate balance between production and loss of heat, and the evaluate the transudative and proliferative components of hypothalamus regulates the set point at which body tempera- the chronic inflammation. The moist weight of the cotton ture is maintained. In fever this set point elevates and a drug pellet correlates with the transuda; the dry weight of the like paracetamol does not influence body temperature when pellet correlates with the amount of the granulomatous tis- it is elevated by the factors such as exercise or increase in am- sue (Olajide et al., 1999, 2000). Administration of MEBR bient temperature (Goodman and Gilman, 1996). The MEBR (50, 100 and 200 mg/kg b.w.) and indomethacin (10 mg/kg possesses a significant antipyretic effect in yeast-induced el- b.w.) appear to be effective in inhibiting the moist weight evation of body temperature in rats and this may be due to of cotton pellet. On the other hand, the MEBR effect on anti-inflammatory effect. dry weight of the cotton pellet was almost near to that Based on the results of the present study it can be con- of indomethacin. These data support the hypothesis of the cluded that MEBR has potential anti-inflammatory activity greater effect of the MEBR on the inflammation media- against acute and chronic phases of inflammation. The MEBR tors in the immediate response of inflammation in rats. produce analgesic activity, which is both central and periph- This effect may be due to the cellular migration to injured eral analgesia, and also significant antipyretic action. M. Gupta et al. / Journal of Ethnopharmacology 98 (2005) 267–273 273

Acknowledgement inhibits lipoxygenase- and cyclooxygenase-mediated metabolism of arachidonic acid. Biochemical Pharmacology 36, 3463–3470. One of the authors R. Sambath Kumar (Senior Research Gupta, M., Mazumder, U.K., Sambath kumar, R., Siva Kumar, T., Go- mathi, P., 2004a. Antioxidant and hepatoprotective effect of Bauhinia Fellow), is grateful to AICTE, New Delhi, India, for provid- racemosa against paracetamol and carbon tetrachloride induced liver ing financial support to this work. damage in rata. Iranian Journal of Pharmacology and Therapeutics 3, 12–20. Gupta, M., Mazumder, U.K., Sambath Kumar, R., Siva Kumar, T., 2004b. References Antitumor effect of Bauhinia racemosa against Ehrlich ascites carci- noma with reference to lipid peroxidation and antioxidant system in Akhtar, A.H., Ahmad, K.U., 1995. Anti-ulcerogenic evaluation of the Swiss albino mice. Acta Pharmacologica Sinica 25, 1070–1076. methanol extracts of some indigenous medicinal plants of Pakistan in Kirtikar, K.R., Basu, B.D., 1975. Indian Medicinal Plants, vol. II. Bishen aspirin-ulcerated rats. Journal of Ethnopharmacology 46, 1–6. Mahendra Pal Singh, Dehradun, India, pp. 842–844. Al-Ghamdi, M.S., 2001. The anti-inflammmatory, analgesic and an- Linardi, A., Costa, S.K.P., Da Silva, G.R., Antunes, E., 2002. Involvement tipyretic activity of Nigella sativa. Journal of Ethanopharmacology of kinins, mast cells and sensory neurons in the plasma exudation and 76, 45–48. paw oedema induced by staphylococcal entrotoxin-B in the mouse. Ali, M.S., Azhar, I., Amtul, Z., Ahmad, V.U., Usmanghani, K., 1999. European Journal of Pharmacology 399, 235–242. Antimicrobial screening of some Caesalpiniaceae. Fitoterapia 70, Litchfield, J.T., Wilcoxon, F., 1949. A simplified method of evaluating 299–304. dose-effect experiments. Journal of Pharmacology and Experimental Anjaneyulu, A.S.R., Reddy, A.V.R., Reddy, D.S.K., Roe, S.P., Cameron, Therapeutics 96, 99–133. T.S., 1986. Racemosal: a novel tetracyclic phenol from Bauhinia race- Meade, C.J., Turner, G.A., Bateman, P.E., 1986. The role of polyphospho- mosa Lamk. Tetrahedron 42, 2417–2420. inositides and their breakdown products in A23187-induced release of Anjaneyulu, A.S.R., Reddy, A.V.R., Reddy, D.S.K., Ward, R.S., Adhike- arachidonic acid from rabbit polymorphonuclear leucocytes. Biologi- savalu, D., Cameron, T.S., 1984. Pacharin: a new dibenzo (2,3–6,7) cal Chemistry Journal 23, 425–436. oxepin derivative from Bauhinia racemosa Lamk. Tetrahedron 40, Olajide, O.A., Awe, S.O., Makinde, J.M., 1999. Effects of the aque- 4245–4252. ous extract of Bridelia ferruginea stem bark on carrageenan-induced Balasooriya, S.J., Sotheeswaran, S., Balasubramanium, S., 1982. Eco- edema and granuloma tissue formation in rats and mice. Journal of nomically useful plants of Sri Lanka part IV. Screening of Sri Lanka Ethnopharmacology 66, 113–117. plants for tannins. Journal of Natural Science Council Sri Lanka 10, Olajide, O.A., Awe, S.O., Makinde, J.M., Ekhelar, A.I., Olusola, A., 213–219. Morebise, O., Okpako, D.T., 2000. Studies on the anti-inflammatory, Cuman, R.K.N., Bersani-Amadio, C.A., Fortes, Z.B., 2001. Influence of antipyretic and analgesic properties of the Alstonia boonei stem bark. Type 2 diabetes on the inflammatory response in rats. Inflammation Journal of Ethnopharmacology 71, 179–186. Research 50, 460–465. Plummer, J.L., Cmielewski, P.L., Gourly, G.K., Owen, H., Cousins, M., D’Amour, F.E., Blood, F.R., Belden Jr., D.A., 1965. Manual for Labo- 1996. Assessment of antinociceptive drug effects in the presence of ratory Work in Mammalian Physiology, third ed. The University of impaired motor performance. Journal of Pharmacological Methods 26, Chicago Press, Chicago, pp. 4–6. 79–84. Dhar, M.L., Dhar, M.M., Dhawan, B.N., Mehrotra, B.N., Ray, C., 1968. Prakash, A., Khosa, R.L., 1976. Chemical studies on Bauhini racemosa. Screening of Indian plants for biological activity. Indian Journal of Current Sciences 45, 705–707. Experimental Biology 6, 232–247. Suleyman, H., Demirezer, L.O., Kuruuzum, A., Banoglu, Z.N., Gocer, F., El-Hossary, G.A., Selim, M.A., Sayed, A.E., Khaleel, A.E., 2000. Study Ozbakir, G., Gepdiremen, A., 1991. Anti-inflammatory effect of the of the flavonoid content of Bassia muricata L. and Bauhinia racemosa aqueous extract from Rumex patientia L. roots. Journal of Ethnophar- Lam. Bulletin of Faculty of Pharmacy, Cairo University 38, 93–97. macology 65, 141–148. El-Khatiba, A.S., Khaleel, A.E., 1995. Evaluation of some pharmacolog- Turner, R.A., 1965. In: Turner, R., Ebborn, P. (Eds.), Analgesics: Screen- ical properties of different extracts of Bauhinia racemosa leaf and ing Methods in Pharmacology. Academic Press, New York, p. 100. Bassia muricata whole plant. Bulletin of Faculty of Pharmacy, Cairo Wealth of India, Raw Materials, 1952. Council of Scientific and Industrial University 33, 59–65. Research, Publication and Information Directorate, New Delhi, 54–55. Franzotti, E.M., Santos, C.V.F., Rodrigues, H.M.S.L., Mourao, R.H.V., Winter, C.A., Porter, C.C., 1957. Effect of alteration in side chains upon Andrade, M.R., Antoniolli, A.R., 2002. Anti-inflammatory, analgesic anti-inflammatory and liver glycogen activities in hydrocortisone ester. and acute toxicity of Sida cardiafolia L. Journal of Ethnopharmacol- Journal of American Pharmacological Society 46, 515–519. ogy 72, 273–278. Winter, C.A., Risley, E.A., Nuss, G.W., 1962. Carregeenan-induced Goodman, L.S., Gilman, A.G., 1996. The Pharmacological Basis of Ther- oedema in hind paw of the rats as an assay for anti-inflammatory apeutics, nineteenth ed. McGraw-Hill, New York, pp. 959–975. drugs. Experimental Biology and Medicine 111, 544–547. Griswold, D.E., Mrarshall, P.J., Webb, E.F., Godfrey, R., Newton, J., Wintrobe, M.M., Lee, G.R., Boggs, D.R., Bithel, T.C., Athens, J.W., Diamartina, M.J., Sarau, H.M., Gleason, J.G., Poste, G., Hanna, N., Foerester, J., 1961. Clinical Hematology, fifth ed. Lea and Febiger, 1987. SK&F, 86002: a structurally novel anti-inflammatory agent that Philadelphia, p. 326. Journal of Ethnopharmacology 98 (2005) 275–279

CNS pharmacological effects of the hydroalcoholic extract of Sida cordifolia L. leaves C.I.F. Franco a, L.C.S.L. Morais a, L.J. Quintans-Junior´ a, R.N. Almeida a,∗, A.R. Antoniolli b

a Laborat´orio de Tecnologia Farmacˆeutica e Departamento de Fisiologia e Patologia, Universidade Federal da Para´ıba, Caixa Postal 5009, CEP: 58051-970 Jo˜ao Pessoa, PB, Brazil b Laborat´orio de Farmacologia/Bioqu´ımica e Departamento de Fisiologia-CCBS, Universidade Federal de Sergipe, S˜ao Crist´ov˜ao, SE, Brazil Received 26 September 2004; received in revised form 9 December 2004; accepted 5 January 2005

Abstract

Sida cordifolia L. (Malvaceae), known as “malva branca”, is a plant used in the popular medicine for the treatment stomatits, of asthma and nasal congestion. This work researched the acute toxicity of Sida cordifolia and its action on the central nervous system (CNS) because no data in the literature have been found about of pharmacological activity of this plant in the CNS. The hydroalcoholic extract of Sida cordifolia leaves (HESc) was used and the psychopharmacology approach began with the determination of LD50, where a low toxicity was observed in mice. Depressive activity on CNS was demonstrated by several alterations in mice’s behavior in the pharmacological screening. In the motility test, the HESc showed significant reduction of spontaneous activity at a dose of 1000 mg/kg (i.p.) at 30 and 60 min. The same form the HESc also decreased the ambulation and rearing in open-field test at 30, 60 and 120 min at a dose of 1000 mg/kg (i.p.). © 2005 Published by Elsevier Ireland Ltd.

Keywords: Sida cordifolia; Malvaceae; Medicinal plants; CNS depressant activity

1. Introduction Although preliminary pharmacological studies with Sida cordifolia have been undertaken, there are no data about the In Brazil Sida cordifolia is popularly known as “malva pharmacological effects of this species on behavior and CNS. branca” or “malva branca sedosa” and found throughout the The aim of this study was to carry out a pharmacological be- country with considerable distribution in the northeast region. havioral screening, determine the acute toxicity and to eval- It is used in the popular medicine for the treatment stomatits, uate the effects of HESc in psychopharmacological animal’s of asthma and nasal congestion (Balbach, 1978). models. This plant contains mainly alkaloids, oils, steroids, resin acids, mucin and potassium nitrate (Diwan and Kanth, 1999). Studies showed that the roots possess diuretic and tonic prop- 2. Material and methods erties and administered for nervous disorders such as hemi- plegia and facial paralysis (Rastogi and Malhotra, 1985). 2.1. Plant material and preparation of extract Pharmacological investigation carried out with an aque- ous extract of this plant’s leaves demonstrated an anti- Sida cordifolia was colleted in the botanical garden of Uni- inflammatory and analgesic activity (Antoniolli et al., 2000). versidade Federal de Sergipe (UFS) (Brazil) in January 1999. The plant was identified by Dr. C. Dias Silva Jr. and a voucher ∗ Corresponding author. Tel.: +55 83 216 7511; fax: +55 83 216 7375. specimen (no. 30171) is deposited in the Herbarium of the E-mail address: [email protected] (R.N. Almeida). Biology Department at the same institution. Sida cordifolia

0378-8741/$ – see front matter © 2005 Published by Elsevier Ireland Ltd. doi:10.1016/j.jep.2005.01.008 276 C.I.F. Franco et al. / Journal of Ethnopharmacology 98 (2005) 275–279 leaves were dried at 40 ± 1 ◦C and ground into a granulated After 30, 60 and 120 min of treatment, the number of squares powder. The extract was obtained using 494 g of this pow- invaded within a period of 3 min were counted (De Lima et al., der with EtOH 70% at 50 ◦C for 72 h in Soxhlet followed 1993; Almeida et al., 2001). The invasion criterion adopted by filtration. The filtrate was concentrated in rotaevaporator was the presence of all paws of the animal within the square at 50 ± 5 ◦C for 48 h, lyophilized for 8 h and stored at 5 ◦C, (Vasquez-Freire´ et al., 1994). yielding 88 g of lyophilized active material. The extract was freshly prepared with 0.9% saline and cremophor (vehicle) 2.7. Open-field for pharmacological experiments. This method is used to evaluate exploratory activity and 2.2. Animals emotionality of animals (Carlini et al., 1986). The open-field consisted of a white painted arena measuring 55 cm in diam- Male Swiss mice (weighing 25–35 g, 90 days old) were eter with a 100 W lamp. The floor of the arena was divided obtained from our research animal house and were main- ◦ into several units by black painted lines. Groups of 10 mice tained at controlled room temperature (21 ± 2 C) on a 12 h were treated with HESc at dose of 1000 mg/kg (i.p. or p.o.) light/dark cycle (lights on at 06:00–18:00 a.m.) with free ac- or vehicle. After 30, 60 and 120 min of administration, each cess to food and water. All experiments were conducted be- mouse was placed in the center of the arena and defecation, tween 8:00 and 13:00 h. Procedures were approved by the ambulation, rearing and grooming were recorded for 5 min Laboratorio´ de Tecnologia Farmaceuticaˆ Animal Care and (Arletti et al., 2000). Use Committee. 2.8. Rotarod test 2.3. Drugs This method was described by Dunham and Miya (1957). Sodium pentobarbital and cremophor were purchased Mice were placed on a rotating rod (2.5 cm diameter, rotat- from Sigma (USA). All drugs and the HESc were imme- ing at 7 rpm) for a pre-selection and those able to remain diately prepared before each assay and administered in a vol- on the rod for 3 or more minutes in two successive trials ume of 0.1 ml/10 g body weight (mice). At the time use, the were selected for testing (Morais et al., 1998). After 24 h of extract was suspended in vehicle (saline 0.9% and one drop pre-selection, groups of ten mice were treated with HESc at of cremophor) at the desired concentrations and the sodium dose of 1000 mg/kg (i.p. or p.o.) or vehicle. After 30, 60 and pentobarbital was diluted in a saline 0.9% solution. 120 min of treatments the animals were placed on a rotative bar of the rotarod apparatus for 5 min and the time spent by 2.4. Acute toxicity each animal on the rotarod was recorded (Carlini and Burgos, 1979; Morais et al., 1998). Different doses of HESc were administered in- traperitonally (i.p.) (500–3000 mg/kg) and orally (p.o.) (500–5000 mg/kg), while the control group received only the 2.9. Pentobarbital-induced sleep time vehicle. The groups were observed for 48 h and at the end of this period mortality was recorded for each group (Dietrich, This methodology evaluate the depressive action of a given 1983). drug in CNS that possess activity and characteris- tics of a hypnotic drug (Carlini et al., 1986). Groups of 10 2.5. Pharmacological behavioral screening mice were treated with HESc at a dose of 1000 mg/kg (i.p. or p.o.) or vehicle. After 60 min of the pre-treatment the ani- Groups of 10 mice were treated with HESc at the dose mals were treated with sodium pentobarbital (50 mg/kg, i.p.) of 1000 mg/kg, i.p. or p.o. (experimental) or vehicle (con- (Pal et al., 1996; Perez et al., 1998; Morais et al., 1998). The trol) while behavioral effects were observed and quantified time between loss and recovery of the righting reflex, taken as described by Almeida et al. (1999). After the treatment, as sleeping time, was recorded for the saline and the drug the mice were observed from 30 min up to 4 h for studying pre-treated animals (Speroni and Minghetti, 1988). behavioral changes. 2.10. Statistical analysis 2.6. Spontaneous locomotion test Calculation of the LD50 values with 95% confidence limits This experimental model was described by Carlini (1973) and comparisons of the results were performed using comput- to evaluate the interference of a substance in the motor activity erized linear regression analysis, in GraphPad Prism, version of the animals. Groups of 10 mice were treated with HESc 3.02, a registered trademark of GraphPad Software Inc. The of dose of 1000 mg/kg (i.p. or p.o.) or vehicle. The animals statistical analysis of data was made by analysis of variance were placed in the activity cage (with a square area of 48 cm, (ANOVA) followed by Bonfferoni test. In all cases differ- 30 cm in height and demarcation squares of 12 cm × 12 cm). ences were considered significant if p < 0.05. C.I.F. Franco et al. / Journal of Ethnopharmacology 98 (2005) 275–279 277

3. Results

3.1. Acute toxicity of HESc

The hydroalcoholic extract of S. cordifolia was toxic at high doses administered (i.p.). The LD50 values were 2639 mg/kg with 95% confidence limits of 2068–3367 mg/kg for i.p. administration. were not observed among orally treated animals.

3.2. Pharmacological behavioral screen of HESc Fig. 1. Effect of HESc on spontaneous locomotion in mice. The values rep- resent mean ± S.E.M. (n = 10); * p < 0.05, ** p < 0.001 significantly different from control. The HESc at a dose of 1000 mg/kg (i.p. and p.o.) produced sedation, decrease of the ambulation, reduction of answer to the touch, analgesia and decrease of urination. The effects of HESc were more pronounced in animals treated intraperiton- ally.

3.3. Effect of HESc on spontaneous motor activity in mice

The mice treated with HESc at a dose of 1000 mg/kg (i.p.) caused significant reduction (p < 0.001) of the spontaneous Fig. 2. Effect of HESc on the motor coordination. The values represent locomotor activity in comparison with the control group at mean ± S.E.M. (n = 10). 30 and 60 min. The animals treated p.o. showed a decrease (p < 0.05) of ambulation at 60 min (Fig. 1). 3.5. Effect of HESc on rotarod test

The HESc at a dose of 1000 mg/kg (i.p. or p.o.) did not 3.4. Effect of HESc in open-field cause a significant difference in the motor coordination of the treated animals in comparison with the control group (Fig. 2). The HESc at a dose of 1000 mg/kg (i.p.) significantly (p < 0.05) reduced the ambulation of mice at 30, 60 and 3.6. Effect of HESc on pentobarbital-induced sleep time 120 min. On the other hand, the animals treated p.o. pre- sented a decrease (p < 0.05) of ambulation at 60 and 120 min. The HESc at a dose of 1000 mg/kg (i.p. or p.o.) did not Another important data observed was the reduction of rear- produce a significant alteration of the latency and the time of ing in animals treated at 30, 60 and 120 min, both i.p. or p.o. sleep of the treated animals in comparison with those from (Table 1). the control group (Fig. 3a and b).

Table 1 Effect of HESc on the open-field in mice Time (min) Groups Dose/route (mg/kg) Ambulation Rearing Grooming Defecation 30 Control – 54.2 ± 2.8 23.4 ± 3.9 1.0 ± 0.2 0.2 ± 0.2 HESc 1000/i.p. 31.7 ± 7.5* 5.5 ± 2.8** 1.0 ± 0.3 0.1 ± 0.1 HESc 1000/p.o. 54.9 ± 4.9 8.6 ± 2.8** 0.6 ± 0.2 0.3 ± 0.1 60 Control – 41.6 ± 2.4 18.0 ± 2.7 1.8 ± 0.4 0.1 ± 0.1 HESc 1000/i.p. 28.4 ± 4.4* 1.8 ± 0.8*** 1.7 ± 0.5 0.1 ± 0.1 HESc 1000/p.o. 29.4 ± 5.2* 6.5 ± 2.6** 1.7 ± 0.5 0.1 ± 0.1 120 Control – 33.6 ± 3.8 10.1 ± 1.7 1.2 ± 0.4 0.1 ± 0.1 HESc 1000/i.p. 12.9 ± 3.9* 1.5 ± 1.0** 1.5 ± 0.4 0.0 ± 0.0 HESc 1000/p.o. 14.6 ± 3.7* 3.8 ± 2.1* 1.2 ± 0.4 0.2 ± 0.2 n = 10; values represent mean ± S.E.M. ∗ p < 0.05 significantly different from control. ∗∗ p < 0.01 significantly different from control. ∗∗∗ p < 0.001 significantly different from control. 278 C.I.F. Franco et al. / Journal of Ethnopharmacology 98 (2005) 275–279

The HESc decreased the ambulation and the rearing in the open-field, test confirming the depressive property of the extract. According to Masur et al. (1971) the rearing is a response to the levels of excitability of CNS. It was also ob- served that mice treated with HESc did not present significant alteration in the time of spent in the rotative bar. Vale et al. (1999) affirmed that the absence of interference with the mo- tor coordination in the rotative bar discards the possibility of a muscular relaxing effect. Another methodology carried out to assess the depressant effect of HESc in CNS was the pentobarbital-induced sleep test. The results showed that the animals treated with HESc did not present any alteration in latency and sleep time. Ac- cording to Lovell (1986), the animals’ response in this model can be affected by environmental (diet, temperature and bed- ding material) and genetic factors. However, the depressive activity of HESc in mice was better evidenced by the de- crease of ambulation in spontaneous locomotion and open- field tests. Therefore the absence of effects in motor coordi- nation performance and in pentobarbital-induced sleep time Fig. 3. Effect of HESc on: (a) the latency of pentobarbital-induced sleep suggests a possible absence of neurotoxicity. and (b) the time of sleep induced pentobarbital. The values represent On the basis of the present study, we may suggest that the ± mean S.E.M. (n = 10). HESc has depressant effect on CNS without interfering with motor coordination with a low toxicity, thus justifying its 4. Discussion and conclusion extensive use by the northeast Brazilian population. It is nec- essary to determine the major evidence indicating depressant In this work, the effects of the hydroalcoholic extract of activity as well as the possible action mechanisms. Sida cordifolia leaves was studied in several behavioral ani- mal models for the evaluation of central activity: pharmaco- logical behavioral screening, spontaneous locomotion, open- field, pentobarbital-induced sleeping time and rotarod test. Acknowledgements These are classical animal models of preliminary pharmaco- logical tests of activities on CNS, which provide information The authors would like to express their sincere thanks to about action upon psychomotor performance, motor behav- J.C. Duarte and R. Nonato for their technical assistance. This ior and neurotoxicity. Increasing doses of the hydroalcoholic work was supported by CNPq. extract of Sida cordifolia up to 5 g/kg administered to mice p.o. were not lethal, this is an indication of the low toxicity of the extract (Dietrich, 1983). References In pharmacological behavioral screening, the animals treated with HESc showed decrease of response to the touch Almeida, R.N., Barbosa-Filho, J.M., Antoniolli, A.R., et al., 1999. and reduction of motor activity. These data are indicative of Metodologia para avaliac¸ao˜ de plantas com atividade no Sistema depressive activity of the CNS according to Almeida et al. Nervoso Central e alguns dados experimentais. Revista Brasileira de (1999). It is important to note that the i.p. dose chosen make Farmacia´ 80, 72–76. Almeida, R.N., Assis, T.S., Medeiros, I.A., Barbosa-Filho, J.M., 2001. with basis in preliminary tests that showed more effective. CNS pharmacological effects of the total alkaloidal fraction from Al- The general depressive activity was confirmed in the spon- bizia inopinata leaves. Fitoterapia 72, 24–130. taneous locomotion test where the HESc significantly re- Antoniolli, A.R., Rodrigues, R.H.V., Mourao,˜ M.R., et al., 2000. Anti- duced spontaneous motor activity. The decrease in motor ac- inflammatory, analgesic activity and acute toxicity of Sida cordifolia tivity gives an indication of the level of excitability of the CNS L. (Malva-branca). Journal of Ethnopharmacology 72, 273–278. Arletti, R., Bandieri, E., Cicero, A.F.G., 2000. Orally administered (Masur et al., 1971) and this decrease may be related to seda- Panax notoginseng influence on rat spontaneous behaviour. Journal tion resulting from depression of CNS (Ozturk et al., 1996). of Ethnopharmacology 73, 387–391. This behavioral effect is similar to those obtained in previous Balbach, A., 1978. A Flora Medicinal na Medicina Domestica,´ vol. 2. studies of depressive drugs of CNS in agreement with Moore MVP, Itaquaquecetuba, p. 703. and Kenyon (1994, pp. 114–123), cited in Almeida et al. Carlini, E.A., 1973. Farmacologia Pratica´ sem Aparelhagem. Sarvier, Sao˜ Paulo. (2001); Radhakrishnan et al. (2001) affirmed that the reduc- Carlini, E.A., Burgos, V., 1979. Screening farmacologico´ de ansiol´ıticos: tion of motor activity can be due to the inhibitory effects of metodologia laboratorial e comparac¸ao˜ entre o diazepam e o cloroben- the extract in SNC or the muscular relaxing activity. zapam. Revista da Associac¸ao˜ Brasileira de Psiquiatria 1, 25–31. C.I.F. Franco et al. / Journal of Ethnopharmacology 98 (2005) 275–279 279

Carlini, E.A., Contar, J.D.P., Silva-Filho, A.R., Silveira-Filho, N.G., Ozturk, Y., Aydini, S., Beis, R., Baser, K.H.C., et al., 1996. Effect Frochtengarten, M.I., Bueno, O.F.A., 1986. Pharmacology of lemon- of Hypericum perforatum L. and Hypericum calycinum L. extracts grass (Cymbopogon citratus Stapf). I. Effects of teas prepared from on the central nervous system in mice. Phytomedicine 3, 139– the leaves on laboratory animals. Journal of Ethnopharmacology 17, 146. 37–64. Pal, S.K., Murkhejee, K., Saha, K., Pal, M., Saha, B.P., 1996. Stud- De Lima, T.C.M., Morato, G.S., Takahashi, R.N., 1993. Evaluation of ies on some psychopharmacological actions of Moringa oleifera the central properties of Artemisia verlotorum. Planta Medica 59, lam (Moringaceae) leaf extract. Phytotherapy Research 10, 402– 326–329. 405. Dietrich, L., 1983. A new approach to practical acute toxicity testing. Perez, G.R.M., Perez, L.J.A., Garcia, D.L.M., Sossa, M.H., 1998. Archives of Toxicology 54, 275–287. Neuropharmacological activity of Solanum nigrum fruit. Journal of Diwan, P.V., Kanth, V.R., 1999. Analgesic, antiinflammatory and hypo- Ethnopharmacology 62, 42–48. glycaemic activities of Sida cordifolia. Phytotherapy Research 13, Radhakrishnan, R., Zakaria, M.N.M., Islam, M.W., Chen, H.B., Kamil, 75–77. M., Chan, K., Al-Attas, A., 2001. Neuropharmacological actions of Dunham, N.M., Miya, T.S., 1957. A note on a simple apparatus for de- Portulaca oleraceae L. v. sativa (Hawk). Journal of Ethnopharmacol- tecting neurological deficit in rats and mice. Journal of the American ogy 76, 171–176. Pharmaceutical Association 46. Rastogi, R.P., Malhotra, B.N., 1985. Compendium of Indian Medical Lovell, D.P., 1986. Variation in barbiturate sleeping time in mice. Plants 4, 674. Strain × environment interaction. Laboratory Animals 20, 307–312. Speroni, E., Minghetti, A., 1988. Neuropharmacology activity of extracts Masur, J., Martz, R.M.W., Carlini, E.A., 1971. Effects of acute from Passiflora incarnata. Planta Medica 54, 488–491. and chronic administration of sativa and (−)9-trans- Vale, T.G., Matos, F.J.A., De Lima, T.C.M., Viana, G.S.B., 1999. Be- tetrahydrocannabinol on behavior of rats in open-field arena. Psy- havioral effects of essential oils from Lippia Alba (Mill.) N.E. Brow chopharmacology 19, 338–397. chemotypes. Journal of Ethnopharmacology 167, 127–133. Morais, L.C.S.L., Almeida, R.N., Barbosa-Filho, J.M., 1998. Central de- Vasquez-Freire,´ M.J., Lamela, M., Calleja, J.M., 1994. Laminaria pressant effects of reticuline extracted from Ocotea duckei in rats and ochroleuca: a preliminary study of its effects on the central nervous mice. Journal of Ethnopharmacology 62, 57–61. system. Phytotherapy Research 8, 422–425. Journal of Ethnopharmacology 98 (2005) 281–285

In vitro antiplasmodial activity and cytotoxicity of 33 West African plants used for treatment of malaria

Gued´ eNo´ el¨ Zirihi a, Lengo Mambu b,Fred´ eric´ Gued´ e-Guina´ c, Bernard Bodo b, Philippe Grellier d, ∗

a Laboratoire de Botanique, UFR Biosciences, Universit´e de Cocody, 22 BP 582 Abidjan 22, Ivory Coast b USM0502-UMR5154 CNRS Chimie et Biochimie des Substances Naturelles, D´epartement R´egulations, D´eveloppement, Diversit´e Mol´eculaire, Mus´eum National d’Histoire Naturelle, 63 Rue Buffon, 75231 Paris Cedex 05, France c Laboratoire de Pharmacodynamie Biochimique, UFR Biosciences, Universit´e de Cocody, 22 BP 582 Abidjan 22, Ivory Coast d USM0504 Biologie Fonctionnelle des Protozoaires, D´epartement R´egulations, D´eveloppement, Diversit´e Mol´eculaire, Mus´eum National d’Histoire Naturelle, 61 Rue Buffon, 75231 Paris Cedex 05, France Received 19 June 2004; received in revised form 26 October 2004; accepted 14 January 2005

Abstract

Thirty-three plants commonly used in West tropical Africa by traditional healers for the treatment of malaria were collected and ethanolic extracts were obtained by decoction. The antiplasmodial activity of extracts was evaluated in vitro against the chloroquine-resistant FcB1 strain of Plasmodium falciparum. Cytotoxicity was determined on the human MRC-5 and the rat L-6 cell lines. Of the 33 plant extracts, eight (24.5%) showed significant antimalarial activity (IC50 values ranging from 2.3 to 13.7 ␮g/ml), 14 (42.5%) weak activity (IC50 values ranging from 15 to 50 ␮g/ml) and 11 (33%) appeared inactive (IC50 values >50 ␮g/ml). Five plants were of particular interest, associating good antiplasmodial activity and weak cytotoxicity. These five included Nauclea latifolia with known antiplasmodial activity and four, Fagara macrophylla, Funtumia elastica, Phyllanthus muellerianus and Rauvolfia vomitoria, for which the description of antiplasmodial activity is entirely novel. © 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Plasmodium falciparum; Antiplasmodial activity; Medicinal plants; Fagara macrophylla; Rauvolfia vomitoria; Funtumia elastica

1. Introduction remedies. Indeed, indigenous plants play an important role in the treatment of many diseases (Phillipson and Wright, Malaria is the major tropical disease due to parasites, re- 1991) and 80% of people world wide are estimated to use sponsible for significant morbidity and mortality in the world. herbal remedies. However, few data are available on their A dramatic recrudescence of malaria is ongoing due to the efficiency and safety, despite the fact that validation of tradi- increasing resistance of vectors to insecticides and the pro- tional practices could lead to innovative strategies in malaria gressive resistance of the parasite, mainly Plasmodium fal- control. What is more, natural products from plants or other ciparum, to drugs. These developments and the difficulty of organisms, represent a virtually inexhaustible reservoir of creating efficient vaccines underline the urgent need for new molecules, most of which are hardly explored and can con- antimalarial drugs. stitute lead molecules for new antimalarial drugs, such as In endemic countries, accessible treatments against artemisinin, initially isolated from Artemisia annua (Kayser malaria are mainly based on the use of traditional herbal et al., 2003). During an ethnobotanical survey in the central west Ivory ∗ Corresponding author. Tel.: +33 1 40793510; fax: +33 1 40793499. Coast (Zirihi, 1991), 33 plants were selected on their com- E-mail address: [email protected] (P. Grellier). monly use by traditional healers for the treatment of malaria

0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2005.01.004 282 G.N. Zirihi et al. / Journal of Ethnopharmacology 98 (2005) 281–285

Table 1 Plants and parts used for analysis Family Plant species (specimen no.) Aspect Part used Amaranthaceae Alternanthera repens (Linn.) Link (L. Ake Assi 14199) Herb (0.5–1 m) Whole plant Anacardiaceae Mangifera indica Linn. (Adjanohou and Ake Assi 288) Tree (20–25 m) Stem bark Apocynaceae Alstonia boonei De Wild. (L. Ake Assi 13872) Tree (20–30 m) Stem bark Funtumia elastica (Preuss) Stapf. (L. Ake Assi 13376) Tree (10–20 m) Stem bark Rauvolfia vomitoria Afzel. (E. Ake Assi 155) Tree (4–8 m) Root bark Asclepiadaceae Parquetina nigrescens (Afzel.) Bullock. (L. Ake Assi 15031) Liana Leaf Asteraceae Acanthospermum hispidum (DC) Kuntze (Ake Assi 11965) Herb (0.5–1 m) Stem and leaf Erigeron floribundus (Kunth) Sch. Bip. (Ake Assi 14076) Herb (0.5–1 m) Stem and leaf Melanthera scandens (Schumach. and Thonn.) Roberty (Adjanohoun and Ake Assi 224) Herb (0.4–1 m) Whole plant Microglossa pyrifolia Lam. Kuntze (Ake Assi 15048) Herb (1–2 m) Stem and leaf Caesalpiniaceae Anthonotha macrophylla P. Beauv. (Adjanohoun and Ake Assi 22) Tree (10–15 m) Stem bark Cassia alata Linn. (L. Ake Assi 15915) Shrub (4–6 m) Leaf Cassia occidentalis Linn. (L. Ake Assi 239) Shrub (0.5–1 m) Leaf Euphorbiaceae Alchornea cordifolia (Schumach. and Thonn.) Mull.¨ Arg. (Adjanohoun and Ake Assi 234) Shrub (5–12 m) Leaf Linn. (E. Ake Assi 132) Herb (0.2–0.3 m) Whole plant Mareya micrantha (Benth.) Mull.¨ Arg. (Adjanohoun and Ake Assi 337) Tree (8–12 m) Stem bark Phyllanthus muellerianus (Kuntze) Exell. (E. Ake Assi 156) Shrub (3–4 m) Leaf Fabaceae Albizia ferruginea (Guill. and Perr.) Benth. (L. Ake Assi 12615) Tree (5–10 m) Leaf Millettia zechiana Harms (J.L. Guillaumet 1424) Tree (10–15 m) Stem bark Loganiaceae Anthocleista djalonensis A. Chevalier (L. Ake Assi 10513) Tree (20–25 m) Stem bark Strychnos spinosa Lam. (L. Ake Assi 101) Liana Bark Melianthaceae Bersama abyssinica Fresen. (Adjanohoun and Ake Assi 215) Tree (10–15 m) Leaf Menispermaceae Rhigiocarya racemifera Miers (J.L. Guillaumet 1430) Liana Leaf Moraceae Ficus capensis Thunb. (L. Ake Assi 12619) Tree (20–30 m) Leaf Myristicaceae Pycnanthus angolensis (Welw.) Warb. (Ake Assi 16771) Tree (20–25 m) Stem bark Pandaceae Microdesmis keayana J. Leonard´ (L. Ake Assi 13640) Shrub (4–6 m) Leaf Rubiaceae Morinda morindoides (Baker) Milne-Redh. (Ake Assi 17239) Liana Leaf Nauclea latifolia Sm. (L. Ake Assi 15927) Liana Bark Rutaceae Fagara macrophylla (Oliv.) Engl. (Adjanohoun and Ake Assi 73) Tree (20–25 m) Stem bark Simaroubaceae Irvingia gabonensis (Aubry-Lecomte ex O’Rorke) Baill. (L. Ake Assi 1175) Tree (20–25 m) Stem bark Solanaceae Physalis angulata Linn. (L. Ake Assi 15767) Herb (0.3–0.5 m) Whole plant Solanum indicum (Thonn.) Bitter (E. Ake Assi 101) Herb (0.3–0.5 m) Fruit Solanum nigrum Linn. (E. Ake Assi 100) Herb (0.3–0.5 m) Fruit and fever. The present study investigates the in vitro antiplas- Parts of plants tested were air-dried during 7–10 days at room modial activity of ethanolic extracts of these plants against temperature and powdered. For each plant, 100 g of material Plasmodium falciparum and their toxicity as assessed in two were extracted three times in ethanol at room temperature. mammalian cell lines. The combined extracts were filtered and concentrated to dry- ness using a rotary evaporator. Dry extracts were stored at 4 ◦C until analysis. 2. Materials and methods 2.2. In vitro antiplasmodial activity 2.1. Plant extracts Extracts were tested against the chloroquine-resistant Plants were selected during an ethnobotanical survey FcB1/Colombia strain of Plasmodium falciparum (IC50 value with the guidance of traditional healers (ACCT, 1986; for chloroquine of 0.1 ␮M, Frappier et al., 1996). The an- Zirihi, 1991) and collected in Issia (central west Ivory Coast). tiplasmodial activity was determined according to Desjardins Species were identified by Prof. Ake´ Assi of the “Centre Na- et al. (1979). Extracts were prepared in DMSO at a con- tional de Floristique”, University of Cocody, Abidjan, Ivory centration of 10 mg/ml and serially diluted with culture Coast, where voucher specimens were deposited (Table 1). medium before to be added to asynchronous parasite cul- G.N. Zirihi et al. / Journal of Ethnopharmacology 98 (2005) 281–285 283 tures (1% parasitemia and 1% final hematocrite) in 96-well cultures. IC50 values on cell growth were obtained from the microplates. Plates were maintained for 24 h at 37 ◦C. 0.5 ␮Ci drug concentration–response curves. of [3H]hypoxanthine was then added to each well and para- sites were maintained for further 24 h. The growth inhibition for each extract concentration was determined by compari- 3. Results and discussion son of the radioactivity incorporated into the treated culture with that in the control culture maintained on the same plate. Plant species and the parts of plants used for extract prepa- The concentrations causing 50% inhibition of parasite growth ration are shown in Table 1.IC50 values obtained on the FcB1 (IC50) were calculated from the drug concentration–response strain of Plasmodium falciparum and on the mammalian L-6 curves. and MRC-5 cell lines are summarized in Table 2. Selectiv- ity index (SI) of the most active extracts are also presented 2.3. Cytotoxicity test on mammalian cells in Table 2. SI was defined as the ratio of the IC50 value on the mammalian cells to the IC50 value on Plasmodium falci- The human diploid embryonic lung cells MRC-5 or the parum. Extracts that demonstrated high selectivity (high SI rat myoblast-derived cells L-6 were seeded into 96-well mi- value) should offer the potential of safer therapy. croplates at 5000 cells per well. After 24 h, the cells were Of the 33 plant extracts tested, 11 (33%) were considered washed and maintained with different concentrations of ex- inactive against Plasmodium falciparum in culture with IC50 ◦ tract for 5 days, at 37 C under a 5% CO2 atmosphere. Cy- values >50 ␮g/ml. Fourteen (42.5%) showed a weak antiplas- totoxicity was determined using the colorimetric MTT assay modial activity with IC50 values between 15 and 50 ␮g/ml (Mossman, 1983) and scored as percentage of absorbance re- and eight (24.5%) had good in vitro activity with IC50 val- duction at 540 nm of treated cultures versus untreated control ues ranging from 2.3 to 13.7 ␮g/ml. Certain plants included

Table 2 In vitro antiplasmodial activity of ethanol extracts of plants on Plasmodium falciparum and cytotoxicity on mammalian cell lines L-6 and MRC-5

Plant species Plasmodium falciparum,IC50 (␮g/ml) L-6, IC50 (␮g/ml) SI, L6/Pf MRC-5, IC50 (␮g/ml) SI, MRC-5/Pf Acanthospermum hispidum 13.7 ± 1.8 22.7 ± 2.4 1.6 10.4 ± 1.6 0.8 Albizia ferruginea >50 ND ND ND ND Alchornea cordifolia >50 ND ND ND ND Alstonia boonei >50 ND ND ND ND Alternanthera pungens >50 ND ND ND ND Anthocleista djalonensis >50 ND ND ND ND Anthonotha macrophylla 42.3 ± 8.7 >50 >1 49.7 ± 1.5 1 Bersama abyssinica 23.9 ± 5.7 ND ND 5.3 ± 0.8 0.5 Cassia alata >50 ND ND ND ND Cassia occidentalis 36.9 ± 5.6 >50 >1 32.8 ± 3.4 1 Erigeron floribundus 36.9 ± 5.8 15.2 ± 1.1 0.5 >50 >1 Euphorbia hirta 44.7 ± 14.7 >50 >1 >50 >1 Fagara macrophylla 2.3 ± 1.0 28.5 ± 0.5 12.4 20.8 ± 3.2 9 Ficus capensis 45.3 ± 5.1 >50 >1 >50 >1 Funtumia elastica 3.3 ± 0.9 ND ND >50 >15.2 Irvingia gabonensis 21.6 ± 7.4 37.4 ± 1.5 1.7 46.6 ± 2.0 2.2 Mangifera indica >50 ND ND ND ND Mareya micrantha 27.6 ± 1.9 14.6 ± 0.2 0.5 7.0 ± 0.2 0.25 Melanthera scandens >50 ND ND ND ND Microdesmis keayana >50 ND ND ND ND Microglossa pyrifolia 33.1 ± 4.1 >50 >1 >50 >1 Millettia zechiana 16.1 ± 2.6 32.3 ± 4.1 2 34.4 ± 3.0 2.1 Morinda morindoides 11.6 ± 3.7 ND ND 42.2 ± 3.2 3.6 Nauclea latifolia 8.9 ± 2.5 >50 >5.6 >50 >5.6 Parquetina nigrescens 21.2 ± 3.0 23.3 ± 0.8 1 27.5 ± 4.8 1 Phyllanthus muellerianus 9.4 ± 2.9 >50 >5.3 >50 >5.3 Physalis angulata 7.9 ± 0.7 27.4 ± 1.5 3.5 27.9 ± 1.4 3.5 Pycnanthus angolensis 18.2 ± 2.7 47.9 ± 2.6 2.6 >50 >2.7 Rauvolfia vomitoria 2.5 ± 1.0 22.5 ± 1.5 10 22.6 ± 1.4 9 Rhigiocarya racemifera >50 ND ND ND ND Solanum indicum 41.3 ± 7.0 >50 >1 >50 >1 Solanum nigrum >50 ND ND ND ND Strychnos spinosa 21.8 ± 6.7 42.7 ± 6.7 1.9 >50 >2.3

IC50 values are the mean ± the standard deviations from three independent experiments; SI: selectivity index, defined as the ratio of the IC50 value determined on the mammalian cell line on the IC50 value determined on Plasmodium falciparum; ND: not determined. 284 G.N. Zirihi et al. / Journal of Ethnopharmacology 98 (2005) 281–285 in the study have previously been described as having an- described potent antiplasmodial activity of Nauclea latifolia tiplasmodial activity. This was the case for Acanthospermum (Benoit-Vical et al., 1998), although the active components hispidum (Sanon et al., 2003), Morinda morindoides (Tona against Plasmodium have yet to be identified. Also, sev- et al., 2001), Pycnanthus angolensis (do Ceu´ de Madureira et eral Phyllanthus species showed antiplasmodial activity al., 2002), and Physalis angulata (Ankrah et al., 2003). With (Omulokoli et al., 1997), whereas, no activity has previously the exception of Mareya micrantha, Bersama abyssinica, been reported for Phyllanthus muellerianus. Erigeron floribundus and Acanthospermum hispidum, plant Investigations to identify the active antimalarial extracts presented no or a weak cytotoxicity on the mam- compounds of Fagara macrophylla, Funtumia elastica, malian cells (IC50 values >20 ␮g/ml). Phyllanthus muellerianus and Rauvolfia vomitoria by Several extracts are of particular interest, associating a bioassay-guided fractionation are now in progress. good antiplasmodial activity with a low cytotoxicity: the root bark extract of Rauvolfia vomitoria (IC50 = 2.5 ␮g/ml, SI = 9–10), the stem bark extract of Funtumia elastica Acknowledgements (IC50 = 3.3 ␮g/ml, SI > 15) and the stem bark extract of Fa- gara macrophylla (IC50 = 2.3 ␮g/ml, SI = 9–12). The authors wish to thank Professor Ake´ Assi for iden- Although Apocynaceae are used in traditional medicine tification of plant material, Dr. Elisabeth Mouray for cultur- in Africa (Omino and Kokwaro, 1993), the in vitro antiplas- ing Plasmodium falciparum and Prof. Barbara Demeneix for modial activity of Rauvolfia vomitoria and Funtumia elastica manuscript reading and correction. This work was supported have not previously been reported. Stem latex of Funtumia by the Centre National de la Recherche Scientifique (CNRS) elastica is used for the washing of wounds and the leaves and the French Ministry of Research and New Technolo- are widely used in traditional medicine for the treatment of gies. Dr. G.N. Zirihi was supported by the Museum´ National haemorrhoids. Rauvolfia vomitoria is an important medicinal d’Histoire Naturelle. plant used in many ills such venereal diseases, neuropsychi- atry disorders, jaundice, gastro-intestinal, sexual complaint, measles (Iwu and Court, 1977). Rauvolfia vomitoria also References belongs with Alstonia boonei and Elaies guineensis,toa Ghanaian herbal preparation with anti-inflammatory activity ACCT, 1986. Medecine´ traditionnelle et pharmacopee,´ contribution aux known to be used in the management of rheumatoid arthritis etudes´ ethnobotaniques et floristiques au Togo. Agence de Cooperation´ Culturelle et Technique (ACCT), 671 pp. (Kweifio-Okai, 1991). Ankrah, N.A., Nyarko, A.K., Addo, P.G., Ofosuhene, M., Dzokoto, C., Fagara macrophylla possesses a potent antifeedant Marley, E., Addae, M.M., Ekuban, F.A., 2003. Evaluation of efficacy activity against larvae of both Spodoptera frugiperda and and safety of a used for the treatment of malaria. Spodoptera littoralis (Tringali et al., 2001). Acridone Phytotherapy Research 17, 697–701. alkaloids that contribute to this antifeeding activity have Basco, L.K., Mitaku, S., Skaltsounis, A.L., Ravelomanantsoa, N., Tille- quin, F., Koch, M., Le Bras, J., 1994. In vitro activities of furoquino- been isolated from this plant (Tringali et al., 2001) and these line and acridone alkaloids against Plasmodium falciparum. Antimi- substances might also be involved in the antiplasmodial crobial Agents and Chemotherapy 38, 1169–1171. activity. This hypothesis is supported by the fact that several Benoit-Vical, F., Valentin, A., Peletier, Y., Cournac, V., Kone-Bamba, acridone derivatives have been described as having potent D., Mallie, M., Bastide, J.M., 1998. Differential antimalarial activity activity against Plasmodium falciparum (Basco et al., 1994). in vitro of root and bark extracts of Nauclea latifolia. Journal of Ethnopharmacology 61, 173–178. Benzo[c]phenanthridine alkaloids with antitumor activity Desjardins, R.E., Canfield, C.J., Haynes, J.D., Chulay, J.D., 1979. Quan- have also been isolated from Fagara macrophylla (Wall et al., titative assessment of antimalarial activity by a semi-automated mi- 1987). They are DNA ligands and active as topoisomerase I- crodilution technique. Antimicrobial Agents and Chemotherapy 16, targeting agents (Li et al., 2003), the nitidine alkaloid present 710–718. in Fagara macrophylla being the best known. Nitidine has do Ceu´ de Madureira, M., Paula Martins, A., Gomes, M., Paiva, J., Proenca da Cunha, A., do Rosario, V., 2002. Antimalarial activity of also been isolated from Toddalia asiatica, a plant used by medicinal plants used in traditional medicine in S. Tome and Principe the Pokot tribe of Kenya to treat fevers and showed a potent islands. Journal of Ethnopharmacology 81, 23–29. antiplasmodial activity against Plasmodium falciparum Frappier, F., Jossang, A., Soudon, J., Calvo, F., Rasoanaivo, P., with IC50 values in the low nanomolar range, and a lack of Rastimamanga-Urveg, S., Saez, J., Schrevel, J., Grellier, P., 1996. Bis- cross-resistance between chloroquine and nitidine (Gakunju benzylisoquinoline as modulators of chloroquine resistance in Plas- modium falciparum and multidrug resistance in tumor cells. Antimi- et al., 1995). Benzo[c] phenanthridine alkaloids might thus crobial Agents and Chemotherapy 6, 1476–1481. be involved in the antiplasmodial activity of the Fagara Gakunju, D.M., Mberu, E.K., Dossaji, S.F., Gray, A.I., Waigh, R.D., Wa- macrophylla extract by inhibiting parasite DNA replication. terman, P.G., Watkins, W.M., 1995. Potent antimalarial activity of the The liana bark extract of Nauclea latifolia and the alkaloid nitidine, isolated from a Kenyan herbal remedy. Antimicro- leaf extract of Phyllanthus muellerianus showed a weaker bial Agents and Chemotherapy 39, 2606–2609. ␮ Iwu, M.M., Court, M.M., 1977. Root alkaloids of Rauwolfia vomitoria antiplasmodial activity (IC50 values of 8.9 and 9.4 g/ml, afz. Planta Medica 32, 88–99. respectively) but they also have significant parasitic selectiv- Kayser, O., Kiderlen, A.F., Croft, S.L., 2003. Natural products as antipar- ity (SI > 5). These observations also confirm the previously asitic drugs. Parasitology Research 90, S55–S62. G.N. Zirihi et al. / Journal of Ethnopharmacology 98 (2005) 281–285 285

Kweifio-Okai, G., 1991. Antiinflammatory activity of a Ghanaian an- David, P., 2003. Antiplasmodial activity of alkaloid extracts from tiarthritic herbal preparation, I. Journal of Ethnopharmacology 33, Pavetta crassipes (K. Schum) and Acanthospermum hispidum (DC), 263–267. two plants used in traditional medicine in Burkina Faso. Parasitology Li, D., Zhao, B., Sim, S.P., Li, T.K., Liu, A., Liu, L.F., La Voie, E.J., 2003. Research 90, 314–317. 2,3-Dimethoxybenzo[i]phenanthridines: topoisomerase I-targeting an- Tona, L., Mesia, K., Ngimbi, N.P., Chrimwami, B., Okond’ahoka, ticancer agents. Bioorganic Medicinal Chemistry 20, 521–528. Cimanga, K., de Bruyne, T., Apers, S., Hermans, N., Totte, J., Pieters, Mossman, T., 1983. Rapid colorimetric assay for cellular growth and L., Vlietinck, A.J., 2001. In vivo antimalarial activity of Cassia oc- survival: application to proliferation and cytotoxicity assay. Journal cidentalis, Morinda morindoides and Phyllanthus niruri. Annals of of Immunological Methods 65, 55–63. Tropical Medicine and Parasitology 95, 47–57. Omino, E.A., Kokwaro, J.O., 1993. Ethnobotany of Apocynaceae species Tringali, C., Spatafora, C., Cali, V., Simmonds, M.S., 2001. Antifeeding in Kenya. Journal of Ethnopharmacology 40, 167–180. constituents from Fagara macrophylla. Fitoterapia 75, 538–543. Omulokoli, E., Khan, B., Chhabra, S.C., 1997. Antiplasmodial activity Wall, M.E., Wani, M.C., Taylor, H., 1987. Plant antitumor agents. 27. of four Kenyan medicinal plants. Journal of Ethnopharmacology 56, Isolation, structure, and structure activity relationships of alkaloids 133–137. from Fagara macrophylla. Journal of Natural Products 50, 1095–1099. Phillipson, J.D., Wright, C.W., 1991. Can ethnopharmacology contribute Zirihi, G.N., 1991. Contribution au recensement, a` l’identification et a` to the development of antimalarial agents? Journal of Ethnopharma- la connaissance de quelques especes` veg´ etales´ utilisees´ en medecine´ cology 32, 155–165. traditionnelle chez les Bet´ es´ du departement´ d’Issia, Coteˆ d’Ivoire. Sanon, S., Azas, N., Gasquet, M., Ollivier, E., Mahiou, V., Barro, N., These` de Doctorat 3eme` cycle. Universite´ Nationale de Coteˆ d’Ivoire, Cuzin-Ouattara, N., Traore, A.S., Esposito, F., Balansard, G., Timon- p. 179. Journal of Ethnopharmacology 98 (2005) 287–294

Effect of Commiphora opobalsamum (L.) Engl. (Balessan) on experimental gastric ulcers and secretion in rats

Tawfeq Al-Howiriny a, Mohammed Al-Sohaibani b, Mansour Al-Said a, Mohammed Al-Yahya a, Kamal El-Tahir a, Syed Rafatullah a,∗

a Medicinal, Aromatic and Poisonous Plants Research Center (MAPPRC), College of Pharmacy, King Saud University, P.O. Box 2457, Riyadh 11451, Saudi Arabia b Department of Pathology, College of Medicine, King Khalid University Hospital, King Saud University, P.O. Box 2925, Riyadh 11461, Saudi Arabia Received 18 November 2004; received in revised form 14 January 2005; accepted 14 January 2005

Abstract

The ulcer protective potential of an ethanol extract of Commiphora opobalsamum (L.) Engl. (Burseraceae) ‘Balessan’ was assessed against different acute gastric ulcer models in rats induced by necrotizing agents (80% ethanol, 0.2 M NaOH and 25% NaCl), hypothermic restraint stress, pyloric ligation (Shay) and indomethacin. Balessan, 250 and 500 mg/kg administered orally (intraperitoneally in Shay rat model) showed a dose-dependent ulcer protective effects in all the above ulcer models. Besides, the extract offered protection against ethanol- induced depletion of stomach wall mucus and reduction in nonprotein sulfhydryl (NP-SH) concentration. Ethanol treatment also caused histopathological lesions of the stomach wall. Pretreatment with Balessan extract provided a complete protection of gastric mucosa through supporting both the offensive and defensive factors. Balessan extract was also showed a large margin of safety without any apparent adverse effects in rats. © 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Commiphora opobalosamum; Burseraceae; Gastric anti-ulcer; Biochemical and histopathological changes in rats

1. Introduction (1982) reported its potential use in stomach problems and cancer. In an earlier investigation, Abdul and Amin (1997) During field surveys in various parts of the Kingdom of have reported hypotensive effect of an aqueous extract of Saudi Arabia, we found that the aerial parts of Commiphora Commiphora opobalsamum in rats. Since, no studies of anti- opobalsamum (L.) Engl. (Burseraceae) are used by local peo- ulcer activity with this plant are available, the present study ple for the treatment of various diseases including dyspepsia, was carried to investigate the influence of this plant on dif- colic, joint pain and to promote urine output or to expel renal ferent experimental models of gastric ulcers, which operate calculi. This information was gathered from the local inhab- via distinct mechanisms of ulcerogenesis (Desai and Parmar, itants of Farasan Island (an island in the Red Sea, Southwest 1994). of Saudi Arabia). In Unani or Greeco–Arab medicine, this plant is one of the important ingredients of a polyherbal for- mulation and in the form of decoction is used to improve 2. Materials and methods digestion and to treat gastrointestinal disorders (Dymock et al., 1890; Anonymous, 1967; Chopra et al., 1956). Hartwell 2.1. Plant material and preparation of crude extract

∗ Corresponding author. Fax: +966 1 467 6383. The aerial parts of Balessan used in this study were col- E-mail address: [email protected] (S. Rafatullah). lected from Farasan Island of Red Sea (Saudi Arabia) in the

0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2005.01.034 288 T.Al-Howiriny et al./ Journal of Ethnopharmacology 98 (2005) 287–294 month of March 2002, and identified by Dr. M.A. Rahman, according to the following arbitrary scale: 0, no blood de- a voucher specimen (#14312) was deposited in the herbar- tectable; 1, thin blood follows the rugae; 2, thick blood fol- ium of Medicinal, Aromatic and Poisonous Plants Research lows the rugae; 3, thick blood follows the rugae with blood Center, College of Pharmacy, King Saud University, Riyadh, clots in certain areas; and 4, thick blood (Chiu et al., 1984). Saudi Arabia for future reference. Each 500 g coarse pow- After wiping the blood off, the total area of lesions in each der of aerial parts of the plant was macerated in 3 L of 96% stomach was scored as described above. ethanol for 72 h using percolation method. The solvent was then removed at 40 ◦C under reduced pressure in a rotary 2.6. Pylorus ligated (Shay) rats evaporator. The yield was 11.02% of (w/w) in terms of the dried starting material. The extract was kept in a refrigerator The animals were fasted for 36 h with access to water ad for biological testing. The extract was suspended in distilled libitum before the pylorus was ligated under anesthesia, water using sodium carboxymethylcellulose (0.25%) just be- care being taken not to cause bleeding or to occlude blood ves- fore administration to the animals. sels (Shay et al., 1945). Balessan extract (250 and 500 mg/kg body weight) was administered immediately after pylorus 2.2. Animal stock ligation by intraperitoneal injection. The animals were sacri- ficed 6 h after the pylorus ligation, stomachs were removed, Wistar albino rats of either sex (home bred) aged 7–8 and contents were collected, measured, centrifuged, and sub- weeks and weighing 150–200 g, were obtained from the jected to analysis for titratable acidity against 0.01N NaOH Experimental Animal Care Centre, King Saud University, to pH 7. Each stomach was examined for lesions as described Riyadh, Saudi Arabia. The animals were fed on Purina chow above. diet and water ad libitum and were maintained under standard conditions of humidity (55 ± 5%), temperature (22 ± 2 ◦C) 2.7. Gastric wall mucus determination and light (12-h light/12-h dark cycle). The rats were randomly assigned to different control and treatment groups. The modified procedure of Corne et al. (1974) was used to determine gastric-wall mucus. The glandular segments 2.3. Chemicals from the stomachs of control and treated rats were removed and weighed. Each segment was transferred immediately to Ethanol (BDH, England), indomethacin (Sigma), sodium 1% Alcian blue solution (in sucrose solution, buffered with hydroxide and sodium chloride (Merck) were used. sodium acetate, pH 5), and the excess dye was removed by rinsing with sucrose solution. The dye complexed with the 2.4. Gastric lesions induced by necrotizing agents gastric wall mucus was extracted with magnesium chloride. A 4-mL sample of blue extract was then shaken with an equal The animals in the test groups were given 1 mL of necrotic volume of diethyl ether. The resulting emulsion was cen- agents, either 80% ethanol, 0.2 M NaOH or 25% NaCl, which trifuged and the absorbance of the aqueous layer was mea- are known to produce gastric lesions (Robert et al., 1979). sured at 580 nm. The quantity of Alcian blue extracted per Hypertonic saline and NaOH (0.2 M) were used only in cy- gram (net) of glandular tissue was then calculated. toprotection studies. Balessan extract was given 30 min be- fore the necrotizing agents. The animals were killed under 2.8. Indomethacin-induced gastric ulcers anesthesia, using diethyl ether 1 h after treatment with the necrotic agents. The stomach of each of the animals was ex- Indomethacin was suspended in 1% carboxymethylcellu- cised and opened along the greater curvature. After washing lose in water (6 mg/mL) and administered to the fasted rats with normal saline the gastric lesions were quantified using a in a dose of 30 mg/kg (0.5 mL/100 g). Rats were treated with binocular magnifier. The ulcers were scored according to the Balessan extract (250 and 500 mg/kg, orally) 30 min before method of Valcavi et al. (1982). Control animals were treated indomethacin. Control rats were treated similarly with an with vehicle only. equivalent amount of vehicle (Bhargava et al., 1973). The stomachs of the animals were removed, rinsed with normal 2.5. Hypothermic restraint stress-induced ulcers saline and studied according to the standard procedure (Szabo et al., 1985). The method of Levine (1971) was followed with slight modification. The animals were fasted for 36 h with access 2.9. Estimation of nonprotein sulfhydryl groups (NP-SH) to water ad libitum. One hour after receiving oral Balessan extract (250 and 500 mg/kg body weight), the rats were im- Gastric mucosal NP-SH was measured according to the mobilized in restraint cages and placed inside a ventilated method reported earlier (Sedlak and Lindsay, 1968). The refrigerator maintained at a temperature of 2–4 ◦C. After 3 h, glandular stomachs of control and treated rats were removed they were taken out and sacrified. The stomachs were ex- and homogenized in ice-cold 0.02 M ethylenediaminete- cised and examined for the severity of intraluminal bleeding traacetic acid. The homogenate was mixed with distilled T.Al-Howiriny et al./ Journal of Ethnopharmacology 98 (2005) 287–294 289

Table 1 Effect of Balessan extract on the induction of gastric ulcers by various necrotic agents in rats No. Treatment and dose (mg/kg body weight) Ulcer index (mean ± S.E.) 80% EtOH 0.2 M NaOH 25% NaCl 1. Control 6.66 ± 0.72 7.33 ± 0.42 5.50 ± 0.42 2. Balessan extract (250) 4.33 ± 0.55** (34.98) 4.83 ± 0.70* (34.10) 2.83 ± 0.40** (48.54) 3. Balessan extract (500) 1.66 ± 0.33*** (75.05) 2.33 ± 0.33*** (68.20) 2.16 ± 0.60*** (60.72) Figures in parenthesis indicate percent inhibition. Groups 2 and 3 were statistically compared with group1. Six animals were used in each group. ∗ P < 0.05 (Student’s t-test). ∗∗ P < 0.01 (Student’s t-test). ∗∗∗ P < 0.001 (Student’s t-test). water and 50% TCA, and centrifuged; the supernatants were 2.13. Histopathological studies mixed with Tris buffer, 5,5-dithio-bis(2-nitrobenzoic acid) (DTNB) was added and the sample was shaken. The ab- The gastric tissue was fixed in 10% ethanol buffered for- sorbance was measured, within 5 min of addition of DTNB, malin and processed though graded ethanol, and im- at 412 nm, against a reagent blank with no homogenate. pregnated with paraffin ; sections were made by mi- crotome. After staining with haemotoxylin and eosin stain (Culling, 1974), the sections were examined under a research 2.10. Determination of LD50 in rats microscope by a person who was not aware of experimental Wistar albino rats were divided into various groups and protocols. The different histopathological indices screened each group was orally treated with Balessan extract in the were: congestion, hemorrhage, edema, , inflamma- dose range up to 10 g/kg. tory and dysplastic changes erosions and ulcerations. Following treatments, the animals were observed for six continuous hours and thereafter at intervals of 2–12 h for up to 72 h. All behavioral changes and death during the observation 3. Results and discussion period were recorded. The percentage of death in each group was then calculated. The LD50 was then determined using The results of the present study are summarized in the methods outlined by Ghosh (1984). Tables 1–6. Balessan treatment was found to offer the gas- tric mucosa, a statistically significant and dose-dependent protection against ulcerations caused by 80% ethanol, 0.2 M 2.11. Phytochemical screening NaOH and 25% NaCl (Table 1). The results were substan- tiated by histopathological findings. The histopathological A preliminary phytochemical screening of aerial parts of studies showed that ethanol treatment caused haemorrhagic Balessan was conducted to determine the presence or ab- necrosis to the gastric mucosa in rats (Figs. 1 and 2). Pre- sence of alkaloids, cardiac glycosides, flavonoids, tannins, treatment with Balessan in doses of (250 and 500 mg/kg) coumarins, anthaquinones, saponins, volatile oil, volatile was found to preserve the functional cytoarchitecture of the bases, cyanogenic glycosides, glucosinolates, sugars, sterols entire gastric mucosa (Figs. 3 and 4). These findings con- and/or triterpenes according to the methods described by firms the cytoprotective nature of Balessan (Rafatullah et al., Farnsworth (1966). 1994). Ethanol-induced gastric lesions are thought to arise as a result of direct damage of gastric mucosal cells, resulting in 2.12. Statistical analysis the development of free radicals and hyperoxidation of lipid (Puurunen et al., 1980). The cytoprotective action of some The differences between control and treated groups were anti-ulcer drugs are mediated by the action of endogenous compared using the ANOVA and Student’s t-test as appro- prostaglandins known to play an important role in maintain- priate and were considered significant if P was < 0.05. ing mucosal integrity (Miller, 1983) and to protect the gas-

Table 2 Effect of Balessan extract on hypothermic restraint stress-induced intraluminal bleeding and gastric lesions in rats No. Treatment and dose (mg/kg body weight) Intraluminal bleeding Gastric lesion Percent inhibition Score (mean ± S.E.) Ulcer index (mean ± S.E.) 1. Control 1.50 ± 0.34 26.16 ± 2.30 2. Balessan extract (250) 1.00 ± 0.36 11.00 ± 2.35** 57.95 3. Balessan extract (500) 0.50 ± 0.22* 4.00 ± 2.53** 84.70 Six animals were used in each group. ∗ P < 0.05 (Student’s t-test). ∗∗ P < 0.001 (Student’s t-test). 290 T.Al-Howiriny et al./ Journal of Ethnopharmacology 98 (2005) 287–294

Table 3 Balessan extract on the basal gastric secretion, acidity and lesion in 6 h pylorus ligated Shay rats No. Treatment and dose Volume of basal gastric Titratable acidity Ulcer index (mg/kg body weight) secretion (mean ± S.E.) (mEq/L) (mean ± S.E.) (mean ± S.E.) 1. Control 8.66 ± 0.66 163.27 ± 3.94 1.33 ± 0.33 2. Balessan extract (250) 6.00 ± 0.85* 140.55 ± 9.16* 0.00** 3. Balessan extract (500) 6.16 ± 0.98 81.10 ± 6.18** 0.00** Group 2 and 3 were statistically compared with group 1. Six animals were used in each group. ∗ P < 0.05 (Student’s t-test). ∗∗ P < 0.001 (Student’s t-test).

Table 4 Effect of Balessan extract on the induction of changes in gastric wall mucus by 80% ethanol No. Treatment and dose (mg/kg body weight) Number of rats Gastric-wall mucus (Alcian blue ␮g/g wet glandular tissue) 1. Control (vehicle 1 mL/rat) 6 338.57 ± 8.06 80% Ethanol (1 mL/rat) 6 184.22 ± 6.16* 2. Balessan extract (250) + 80% ethanol (1 mL/rat) 6 209.31 ± 12.00 3. Balessan extract (500) + 80% ethanol (1 mL/rat) 6 270.76 ± 5.71** Tabular values represent the mean ± S.E. Group 2 was statistically compared with group 1. Group 3 and 4 were statistically compared with group 2. ∗ P < 0.01 (Student’s t-test). ∗∗ P < 0.001 (Student’s t-test).

Table 5 Effect of Balessan extract on indomethacin-induced gastric mucosal damage in rats No. Treatment and dose (mg/kg body weight) Lesions score (mean ± S.E.) Percent inhibition 1. Indomethacin (30) + (distilled water) 33.50 ± 3.24 2. Indomethacin (30) + Balessan extract (250) 14.60 ± 2.92** 56.41 3. Indomethacin (30) + Balessan extract (500) 9.66 ± 2.25** 71.16 Group 2 and 3 were statistically compared with group 1. Six animals were used in each group. ∗∗ P < 0.01 and <0.001 (Student’s t-test). tric mucosa against various damaging agents (Chaudhury and The significant reduction in basal gastric secretion and Robert, 1980; Robert et al., 1983). complete inhibition of ulcers by Balessan extract (Table 3) Hypothermic restraint stress model has been widely used after pylorus ligation suggest that the cytoprotective mecha- in rats because of data reproducibility (Murakami et al., nism of action of the extract on gastric mucosa may involve 1985). Disturbances of gastric mucosal microcirculation direct reduction of gastric secretion though one or more of (Guth and Hall, 1966), alteration of gastric secretion and the possible mechanisms discussed by Parmar (1983); Martin abnormal motility (Garrick et al., 1986) have been consid- et al. (1993) and Njar et al. (1995). Moreover, gastric acid is ered to be the pathogenic mechanisms responsible for stress- an important factor for the genesis of ulceration of pylorus- induced gastric mucosal lesions, which develop as a result ligation ulcer in rats (Shay et al., 1945). Gastric acid secretion of stimulation of the vagal nerve which increases gastric se- is regulated by many factors including anxietic effect in the cretion (Kitagawa et al., 1979), this often termed the aggres- CNS, vagal activity, cholinergic, histaminergic and gastriner- sive factor (Goa and Monk, 1987), and gastric mucus de- gic neurotransmissions, the activities of various post-synaptic pletion (Koo et al., 1986). In the current study, the Balessan receptors and the proton pump (Isenberg et al., 1991). It is extract inhibited the production of stress-induced gastric le- therefore, difficult to elucidate the relationship between the sions (Table 2). This finding further indicated that the Ba- mechanism of inhibition of gastric acid and Balessan extract. lessan extract may enhance gastric mucosal defensive fac- The current data clearly demonstrated that, at the very least, tor. Balessan extract dose-dependently inhibited the aggressive

Table 6 Effect of Balessan extract on NP-SH concentrations in gastric ulcers induced by 80% ethanol No. Treatment and dose (mg/kg body weight) Number of animals NP-SH concentration (␮mol/g tissue, mean ± S.E.) 1. Control (vehicle) 6 1.66 ± 0.09 80% Ethanol (1 mL/rat) 6 0.96 ± 0.04** 2. Balessan extract (250) + 80% ethanol (1 mL/rat) 6 1.11 ± 0.04* 3. Balessan extract (500) + 80% ethanol (1 mL/rat) 6 1.37 ± 0.05** Group 2 was statistically compared with group 1. Groups 3 and 4 were statistically compared with group 2. ∗ P < 0.05 (Student’s t-test). ∗∗ P < 0.001 (Student’s t-test). T.Al-Howiriny et al./ Journal of Ethnopharmacology 98 (2005) 287–294 291

Fig. 2. Gastric mucosa of rat after treatment with 1 mL of 80% by gavage. Superficial mucus secreting cell layer is disarranged as well lining of deep Fig. 1. Normal gastric mucosa: note mucus-secreting cells on the surface glands is undergoing hemorrhagic necrosis. Haemotoxylin and eosin 200×. and chief and parietal cells in deep glands. Haemotoxylin and eosin 200×. factor, gastric acid secretion. The anti-ulcerogenic effect of increase the amount of gastric mucus secretion in gastric the extract may be related to its anti-secretory action since mucosa (Bolton et al., 1976; Robert et al., 1984). This mu- acid is a major factor in the development of peptic ulcer cus consists of mucin-type glycoproteins, which can be de- (Glavin and Szabo, 1992). However, certain anti-ulcer drugs tected by amounts of alcian blue binding (Bolton et al., 1978).

Fig. 3. Gastric mucosa of rat after treatment with Balessan extract (250 mg/kg body weight) and 1 mL of 80% ethanol, by gavage. Note that apart from lamina propria edema, there is no evidence of hemorrhagic necrosis as well mucin secreting superficial epithelium is relatively well preserved. Haemotoxylin and eosin 200×. 292 T.Al-Howiriny et al./ Journal of Ethnopharmacology 98 (2005) 287–294

agents (Wallace and Whittle, 1985). In this assay, the Ba- lessan extract was also able to produce a significant reduc- tion of the gastric mucosal damage induced by indomethacin (Table 5), indicating the probable local increase in prostaglandin synthesis. Our results also showed (Table 6)a significant reduction in nonprotein sulfhydryls content of gas- tric mucosa after 80% ethanol administration. Pretreatment of rats with Balessan extract significantly prevented NP-SH de- pletion. Decreased NP-SH level by ethanol is in agreement with other reports which have demonstrated the important role of NP-SH in gastric mucosal damage by ethanol (Miller et al., 1985). An increased NP-SH level is thought to protect gastric damage against various noxious chemicals (Szabo et al., 1981). These findings clearly showed the possible in- volvement of NP-SH in the cytoprotective and antioxidant activities of Balessan extract. Al-Harbi et al. (1994, 1997) have reported similar finding for another Commiphora spec. (Commiphora molmol)ofCommiphora opobalsamum. The preliminary phytochemical screening of Balessan re- vealed the presence of flavonoids, saponins, volatile oil, sterol and/or triterpenes. Though, we have not studied the active principles responsible for the anti-ulcer activity of Balessan, it is likely that the presence of flavonoids in this plant may be involved in the ulcer preventing action as flavonoids possess significant anti-ulcer activity in various experimental models of gastric and duodenal ulceration (Parmar and Parmar, 1998; Fig. 4. Gastric mucosa of rat after treatment with Balessan extract Deshpande et al., 2003). Additionally, saponins and volatile (500 mg/kg body weight) and 1 mL of 80% ethanol, by gavage. No evi- oil of some plants are known to affect the integrity of mucus dence of necrosis is noted as well functional cytoarchitecture of the entire membranes and shown their ability to prevent the formation mucosa is well preserved. Haemotoxylin and eosin 200×. of ulceration of the gastric mucosa in different models of ex- perimentally induced gastric ulcers in rats and mice (Hiruma- Balessan extract increased the alcian blue binding to mucosa. Lima et al., 1999; Guaraldo et al., 2001; Al-Rehaily et al., Alcian blue dye is able to bind negatively charged materials. 2002). LD50 studies on acute toxicity determination showed The increase in bound alcian blue suggested protective ef- no deleterious symptoms and a large margin of safety in rats fect of orally administered Balessan extract. This may be via was recorded. The LD50 value was found to be 4.75 g. the formation of protecting complexes between Balessan ex- In the results of histopathalogical investigation tract and mucus, which can act as a barrier against several (Figs. 1–4), the gastric mucosa of rats revealed that agents introduced in the stomach (Clamp et al., 1978; Sun et the pretreatment of Balessan extract absolutely inhibited the al., 1991). In the present study, the extract of Balessan sig- ethanol-induced haemorrhagic necrosis of rat stomach. Our nificantly replenished the depleted level of gastric mucus by results are in corroboration with the antigastric ulcer activity ethanol challenge (Table 4). Thus, the possible mechanism of of the extract observed under the studies on pharmacological gastric mucosal protection by orally administered extract may and biochemical evaluation. be partly due to reinforcement of resistance of the mucosal In conclusion, the oral administration of an ethanol extract barrier by a protective coating. In addition, its anti-secretory of Commiphora opobalsamum displayed a significant antiul- activity and direct cytoprotective action cannot be excluded. cer activity without any apparent toxicological effects, which Ani-inflammatory agents such as indomethacin reduce supports the use of Balessan in herbal medicine of Saudi Ara- gastric cyclooxygenase activity and decrease endogenous bia. Further experiments and detailed phytochemical analy- prostaglandin levels (Konturek et al., 1984) and increase ses are underway to determine the phytoconstituent(s) re- acid secretion (Sairam et al., 2002). These agents break sponsible for as well as the anti-ulcer mechanisms involved. the mucosal barrier, provoke an increase in gastric mu- cosal permeability to H+ and Na+ ions and a drop in the transmucosal potential difference, and induce the for- Acknowledgements mation of erosions and ulcers (Droy-Lefaix, 1988). There is mounting evidence that an increase of certain endoge- The authors wish to thank Research Center, College of nous prostaglandins can enhance gastric mucosal resistance Pharmacy, KSU, for partial funding (C.P.R.C. #123) and to against ulcerogenic substances such as anti-inflammatory Mr. Malik Sawood Ahmed for his technical assistance. T.Al-Howiriny et al./ Journal of Ethnopharmacology 98 (2005) 287–294 293

References Goa, K.L., Monk, J.P., 1987. Enprostil: a preliminary review of its phar- macodynamics and pharmacokinetic properties and therapeutic effi- Abdul-Ghani, A.S., Amin, R., 1997. Effect of aqueous extract of Com- cacy in the treatment of peptic ulcer disease. Drugs 3, 539–559. miphora opobalsamum on blood pressure and heart rate in rats. Jour- Guaraldo, L., Sertie, J.A.A., Bacchi, E.M., 2001. Antiulcer action of the nal of Ethnopharmacology 57, 219. hydroalcoholic extract and fractionsof Darilla rugosa Poiret in the Al-Harbi, M.M., Qureshi, S., Ahmed, M.M., Rafatullah, S., Shah, A.H., rat. Journal of Ethnopharmacology 76, 191–195. 1994. Effect of Commiphora molmol (oleo-gum-resin) on the cytolog- Guth, P.H., Hall, P., 1966. Microcirculatory and mast cell changes in ical and biochemical changes induced by cyclophosphamide in mice. restrain induced gastric ulcer. Gastroenterology 50, 562–570. American Journal of Chinese Medicine 22, 77–82. Hartwell, J.L., 1982. Plants Used Against Cancer, A Survey, vol. II. Al-Harbi, M.M., Qureshi, S., Raza, M., Ahmed, M.M., Afzal, M., Shah, Quaterman Publication, Lawrence, MA, pp. 89–93. A.H., 1997. Gastric antiulcer and cytoprotective effect of Commiphora Hiruma-Lima, C.A., Gracioso, J.S., Nunes, D.S., Souza Brito, A.R.M., molmol in rats. Journal of Ethnopharmacology 55, 141–150. 1999. Effects of an essential oil from the bark of Croton cajucara Al-Rehaily, A.J., Al-Howiriny, T.A., Al-Sohaibani, M.O., Rafatullah, S., Benth. On experimental gastric ulcer models in rats and mice. Journal 2002. Gastroprotective effects of ‘Amla’ Emblica officinalis on in vivo of Pharmacy and Pharmacology 51, 341–346. test models in rats. Phytomedicine 9, 515–522. Isenberg, J.I., McQuid, K.R., Laine, L., Rubin, W., 1991. Acid-peptic Anonymous, 1967. Unani Pharmacopoeia, vol. II. Indian Medicine De- disorders. In: Yamada, T., Alpers, D.H., Ozyang, C., Powell, D.W., partment, Government of Andha Pradesh, Hyderabad, India, p. 21 (in Silverstein, F.E. (Eds.), Textbook of Gasteroenterology. J.B. Lippin- Urdu). cott, Philadelphia, pp. 1241–1339. Bhargava, K.P., Gupta, M.B., Tangri, K.K., 1973. Mechanism of ulcero- Kitagawa, H., Fujiwara, M., Osami, Y., 1979. Effect of water immersion genic activity of indomethacin and oxyphenbutazone. European Jour- stress on gastric secretion and mucosal blood flow in rats. Gastroen- nal of Pharmacology 22, 191–195. terology 77, 298–302. Bolton, J.P., Palmer, D., Cohen, M.M., 1976. Effect of the E2 Konturek, S.J., Kuriecien, N., Obtulowiecz, W., Oleksy, J., Sito, E., Kopp, prostaglandins on gastric mucus production in rats. Surgery Forum B., 1984. Prostaglandins in peptic ulcer disease: effect of non-steroidal 27, 402–403. anti-inflammatory compounds. Scandinavian Journal of Gastroenterol- Bolton, J.P., Palmer, D., Cohen, M.M., 1978. Stimulation of mucus and ogy 92 (Suppl.), 250–254. nonparietal cell secretion by the E2 prostaglandins. Digestive Diseases Koo, M.W.L., Ogle, C.W., Cho, C.H., 1986. Effect of verapamil, car- Science 23, 359–364. benoxolone and N-acetyl-cystein on gastric wall mucus and ulceration Chaudhury, T.K., Robert, A., 1980. Prevention by mild irritants of gastric in stressed rats. Pharmacology 32, 326–334. necrosis produced in rats by sodium taurocholate. Digestive Diseases Levine, R.J., 1971. A Method for Rapid Production of Stress Ulcers in Science 25, 830–836. Rats. Munksgaard, Copenhagen, pp. 92–97. Chiu, P.J.S., Gerhart, C., Brown, A.D., Barnett, A., 1984. Effects of gastric Martin, M.J., Motilva, V., Alarcon de la Lastra, C., 1993. Quercetin and antisecretory-cytoprotectant 2-methyl-8-(phenylmethoxy)imidazo (1,2- naringenin: effects on ulcer formation and gastric secretion in rats. a)-pyridine-3-acetonitrile (Sch 28080) on cysteamine, reserpine and Phytotherapy Research 7, 150–153. stress ulcers in rats. Arzneimittelforschung 34, 783–786. Miller, T.A., 1983. Protective effects of prostaglandins against gastric mu- Chopra, R.N., Nayer, S.L., Chopra, K., 1956. Glossary of Indian Medic- cosal damage: current knowledge and proposed mechanisms. Ameri- inal Plants. CSIR, New Delhi, India, p. 75. can Journal of Physiology 245, 610–662. Clamp, J.R., Allen, A., Gibbons, R.A., Roberts, G.P., 1978. Chemical Miller, T.A., Kuo, Y.J., Schmidt, K.L., Shambour, L.L., 1985. Non protein aspects of mucus. British Medical Bulletin 34, 25–41. sulfhydryl compounds in canine gastric mucosa; effect of PGE2 and Corne, S.J., Morrissey, S.M., Woods, R.J., 1974. A method for the quanti- ethanol. American Journal of Physiology 249, G137–G142. tative estimation of gastric barrier mucus. Journal of Physiology 242, Murakami, M., Lam, S.K., Inada, M., Miyake, T., 1985. Pathophysiology 116–117. and pathogenesis of acute gastric mucosal lesions after hypothermic Culling, C.F.A., 1974. Handbook of Histopathological and Histochemical restraint stress in rats. Gastroenterology 88, 660–665. Techniques, 73, third ed. Butterworth and Co., London, pp. 73, 126, Njar, V.C.O., Adesanwo, J.K., Raji, Y., 1995. Methyl angolensate: the 159. antiulcer agent of the stem bark of Etandrophgma angolensse. Planta Desai, J.K., Parmar, N.S., 1994. Gastric and duodenal anti-ulcer activity of Medica 61, 91–92. sulpiride a dopamine D2 receptor antagonist, in rats. Agents Actions Parmar, N.S., 1983. Gastric antiulcer activity of naringenin, a special histi- 42, 149–152. dine decarboxylase inhibitor. International Journal of Tissue Research Deshpande, S.S., Shah, G.B., Parmar, N.S., 2003. Antiulcer activity 4, 415–420. of Tephosia purpurea in rats. Indian Journal of Pharmacology 35, Parmar, N.S., Parmar, S., 1998. Antiulcer potential of flavonoids. Indian 168–172. Journal of Physiology and Pharmacology 42, 343–351. Droy-Lefaix, M.T., 1988. In: Curtis Prior, P.B. (Ed.), Prostaglandins: Puurunen, J., Huttunen, P., Hirronen, J., 1980. Is ethanol-induced damage Biology and Chemistry of Prostaglandins and Related Eicosanoids, of the gastric mucosa a hyperosomotic effect? Comparative studies Churchill Livingstone, New York, pp. 345–360. on the effects of ethanol, some other hyperosmotic solutions and Dymock, W., Warden, C.J.H., Hopper, D., 1890. Pharmacographia indica. acetylsalicylic acid on rat gastric mucosa. Acta Pharmacology and In: Kegan, P. (Ed.), A History of the Principal Drugs of Vegetable Toxicology 47, 321–327. Origin. Trench, Trubner and Co. Ltd, London, p. 330. Rafatullah, S., Al-Yahya, M.A., Al-Said, M.S., Abdul Hameed Taragan, Farnsworth, N.R., 1966. Biological and phytochemical screening of K.U., Mossa, J.S., 1994. Gastric anti-ulcer and cytoprotective effects plants. Journal of Pharmaceutical Science 55, 225–272. of Cyamopsis tetragonoloba (Guar) in rats. International Journal of Garrick, T., Leung, F.W., Buack, S., Hirabayashi, K., Guth, P.H., Pharmacognosy 32, 163–170. 1986. Gastric motility in stimulated but overall blood flow in un- Robert, A., Nezamis, J.E., Lancaster, C., Hanchar, A.J., 1979. Cytoprotec- affected during cold restraint in the rat. Gastroenterology, 141– tion by prostaglandins in rats. Prevention of gastric necrosis produced 148. by alcohol, HCl, NaOH, hypertonic NaCl, and thermal injury. Gas- Ghosh, M.N., 1984. Fundamentals of Experimental Pharmacology. Sci- troenterology 77, 433–443. entific Book Agency, Calcutta, pp. 153–158, 187–189. Robert, A., Nezamis, J.E., Lancaster, C., Davis, J.P., Field, S.O., Han- Glavin, G.B., Szabo, S., 1992. Experimental gastric mucosal injury: labo- char, A.J., 1983. Mild irritants prevent gastric necrosis though “adap- ratory models reveal mechanisms of pathogenesis and new therapeutic tive cytoprotection” mediated by prostaglandins. American Journal of strategies. FASEB Journal 6, 825–830. Physiology 245, G113–G121. 294 T.Al-Howiriny et al./ Journal of Ethnopharmacology 98 (2005) 287–294

Robert, A., Bottcher, W., Golanska, E., Kauffman, G.L., 1984. Lack of gastric ulcer models in rats and mice. Journal of Pharmacy and Phar- correlation between mucus gel thickness and gastric cytoprotection in macology 43, 699–704. rats. Gastroenterology 89, 670–674. Szabo, S., Trier, J.S., Frank, P.W., 1981. Sulfhydryl compounds may Sairam, K., Rao, Ch.V., Dora Babu, M., Vijay Kumar, K., Agarwal, V.K., mediate gastric protection. Science 214, 200–202. Goel, R.K., 2002. Antiulcerogenic effect of methanolic extract of Em- Szabo, S., Trier, J.S., Brown, A., Schnoor, J., Homan, H.D., Bradford, blica officinalis: an experimental study. Journal of Ethanopharmacol- J.C., 1985. A quantification method for assessing the extent of exper- ogy 82, 1–9. imental erosions and ulcers. Journal of Pharmacological Methods 13, Sedlak, J., Lindsay, R.H., 1968. Estimation of total, protein bound and 59–66. nonprotein sulfhydryl groups in tissue with Ellman’s reagent. Analyt- Valcavi, U., Caponl, K., Brambilla, A., Palmira, F., Monoja, F., Bernini, F., ical Biochemistry 25, 192. Mustani, R., Fumagall, R., 1982. Gastric antisecretory, antiulcer and Shay, M., Kamarov, S.A., Fels, D., Meranze, D., Gruenstein, H., Siplet, cytoprotective properties of 9-hydroxy-19,20-bis-nor-prostanic acid in H., 1945. A simple method for the uniform production of gastric experimental animals. Arzneimittelforschung 32, 657–663. ulceration in the rats. Gastroenterology 5, 43–61. Wallace, J.L., Whittle, B.J., 1985. Role of prostanoids in the protective Sun, X.B., Matsumoto, T., Yamada, H., 1991. Effects of polysaccharide action of BW755c on the gastric mucosa. European Journal of Phar- fraction from the roots of Bupleurum falcatum L. on experimental macology 115, 45–51. Journal of Ethnopharmacology 98 (2005) 295–300

In vitro estrogenic activities of Chinese medicinal plants traditionally used for the management of menopausal symptoms

C.Z. Zhang a,b, S.X. Wang b, Y. Zhang a, J.P. Chen a, X.M. Liang a,∗

a Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Zhongshan Road No. 161, Dalian 116011, PR China b College of Bio and Food Technology, Dalian Institute of Light Industry, Qinggong-yuan No. 1, Ganjingzi-qu, Dalian 116034, PR China Received 14 December 2004; received in revised form 15 December 2004; accepted 15 January 2005

Abstract

The estrogenic activity of 70% EtOH extracts of 32 traditional Chinese medicinal plants, selected according to their reported efficacy for the treatment of menopausal symptoms, was assessed using a recombinant yeast system with both a human estrogen receptor expression plasmid and a reporter plasmid. Among them, 11 (34%) species proved to be active. Polygonum cuspidatum had the highest estrogenic relative potency (RP) (3.28 × 10−3), followed by Rheum palmatum (3.85 × 10−4), Cassia obtusifolia (3.49 × 10−4), Polygonum multiflorum (2.87 × 10−4), Epimedium brevicornum (2.30 × 10−4), Psoralea corylifolia (1.90 × 10−4), Cynomorium songaricum (1.78 × 10−4), Belamcanda chinensis −4 −5 −5 −5 (1.26 × 10 ), Scutellaria baicalensis (8.77 × 10 ), Astragalus membranaceus (8.47 × 10 ) and Pueraria lobata (6.17 × 10 ). The EC50 value of 17␤-estradiol used as the positive control was 0.205 ± 0.025 ng/ml (RP = 100). This study gave support to the reported efficacy of Chinese medicines used for hormone replacement therapy. © 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Phytoestrogens; Hormone replacement therapy; Chinese medicine; Saccharomyces cerevisiae

1. Introduction symptoms 20 years ago (Nichols et al., 1984), which quickly took effect but increased the risk of breast cancer (Beral et Hormones such as estrogen and progesterone play a very al., 1999). It was found that natural compounds from cer- important role in human growth. It is responsible in regulating tain plants called phytoestrogens could be used for manage- the complex cellular events associated with differentiation, ment of menopausal symptoms and have few side effects function and growth of female reproductive tissues. Women (Thompson, 1993; Glazier and Bowman, 2001). in the menopause have always had to suffer bone density Traditional Chinese medicine has been used to heal many reduction, sweating and anxieties because of a lack of hor- diseases for thousands of years and is now well known mones (Harlow and Signorello, 2000). Hormone replacement as natural medicine throughout the world. Many herbal therapy (HRT) was introduced to improve the menopausal medicines such as Angelica sinensis (Oliv.) Diels, Panax notoginseng (Burk.) F. H. Chen and so on are effective for improving female function according to the oldest Abbreviations: DMSO, dimethyl sulfoxide; E2,17␤-estradiol; EC50, sample concentration at half-maximum ␤-galactosidase activity; ER, es- traditional Chinese medical book, Sheng-nong Ben-cao Jing. trogen receptor; ERE, estrogen response elements; hER, human estrogen It has been proven that some plant extracts have estrogenic receptor; HRT, hormone replacement therapy; lacZEscherichia, coli ␤- components possessing a potential human use in dietary sup- galactosidase gene; OD420, absorbance at 420nm; OD600, absorbance at plements and treatment of menopausal symptoms (Liu et al., 600nm; oNPG, o-nitrophenol-␤-d-pyrogalactoside; RIE, estrogenic relative inductive efficiency; RP, estrogenic relative potency 2001). ∗ Corresponding author. Tel.: +86 411 83681012; fax: +86 411 83698905. In vivo and in vitro assays have been developed to test es- E-mail address: [email protected] (X.M. Liang). trogenic substances. Although in vivo assays are widely used,

0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2005.01.033 296 C.Z. Zhang et al. / Journal of Ethnopharmacology 98 (2005) 295–300 they are unsuitable for large-scale screening and their utility was kindly provided by Prof. W.Z. Wu, Institute of Hydrobi- is further limited due to the cost and relatively poor sensi- ology, Chinese Academy of Sciences. This strain carried the tivity. In vitro assays, however, are based on well-elucidated hER expression plasmid YEPE10 and the estrogen respon- mechanisms of action and utilize more definitive end points sive reporter plasmid YRPE2 (Santiso-Mere et al., 1991). than in vivo assays (Zacharewski, 1997). Among them, the The reporter gene (␤-galactosidase) was controlled by ERE. assay system based on the binding of a ligand to estrogen re- The activity of ␤-galactosidase resulted in a color reaction, ceptor is the simplest. Yeastcells carrying the human estrogen which was measured absorbance at 420 nm. The absorbance receptor (hER) gene, estrogen response elements (ERE) and at 600 nm was selected to measure cell density and viability. Escherichia coli ␤-galactosidase gene (lacZ) are very suitable The yeast strain was grown at 30 ◦C, 180 rpm, in selective for large-scale screening and sensitive analysis of estrogenic medium with 50 ␮M CuSO4 but without tryptophane and compounds. It is useful for the assay and discovery of novel uracil (Wu et al., 2002a). estrogenic substances in natural specimens (Breithofer et al., 1998; Routledge and Sumpter, 1996). 2.4. Preparation of test samples In this study, a recombinant yeast with both a human estro- gen receptor expression plasmid and a reporter plasmid was The plant extracts were dissolved in dimethylsulfoxide employed to search for phytoestrogens in selected Chinese (DMSO) and used as samples for screening tests. 17␤- medicinal plants, which have been used for hormone replace- Estradiol was dissolved in DMSO and used as positive con- ment therapy. A total of 32 Chinese medicinal plants (Table 1) trol. were selected according to their reported efficacy for the treat- ment of menopausal symptoms and the estrogenic activities of their crude 70% EtOH extracts was assessed in order to 2.5. Design of the experiments give support to their reported activity and find crude drugs containing phytoestrogens in high concentration or highly The experiments were designed according to Wang active. et al. (2003) with some modification. For all exper- iments, overnight cultures were diluted to OD600 = 1.0 prior to the induction of hER expression and addition of 2. Materials and methods 17␤-estradiol (positive control), test samples or DMSO (negative control). All the samples at concentrations of 2.1. Chemicals 0.1–1000 ␮g/ml (dried extract/ml) and 17␤-estradiol at con- centrations of 0.001–10 ng/ml were prepared in DMSO. 17␤-Estradiol (E ) and o-nitrophenol-␤-d-pyrogalacto- 2 The final concentration of DMSO in the assays was less side (oNPG) were purchased from Sigma. Yeast nitro- than 1.0%. In this test, 5 ␮l of samples, 17␤-estradiol or gen base without amino acids was purchased from Fluka. DMSO were added to 995 ␮l of yeast culture containing All other reagents used in the study were of analytical 50 ␮МCuSO4, which induced the expression of estrogen re- grade. ◦ ceptor (ER). After incubation at 30 C for 2 h with shaking (150–180 rpm), the yeast cells were used for ␤-galactosidase 2.2. Plant materials and extraction assays. A total of 32 Chinese medicinal plants were purchased from Darentang drugstore in Dalian, China, originating from 2.6. β-Galactosidase assays different regions in China. The plants were identified by Dr. H. Sun, College of Pharmacy, Hei Longjiang University. For the ␤-galactosidase assays, 100 ␮l cell suspensions Voucher specimens were preserved in College of Bio and were added to the wells of a 96-well microplate. The Food Technology, Dalian Institute of Light Industry, China. cells were permeabilized by addition of 100 ␮l assay buffer The voucher numbers are shown in Table 1. (60 mM Na2HPO4, 40 mM NaH2PO4, 10 mM KCl, 1 mM The minced plants (100 g) were extracted with 70% EtOH MgSO4, 2 mg/ml oNPG, 38 mM ␤-mercaptoethanol, 0.01% (800 ml) at room temperature for 24 h with shaking (Chen Triton X-100, 15 U/ml lyticase). The microplate was incu- ◦ et al., 1987). The extracts were taken to dryness under re- bated at 30 C until the color became yellow, which resulted duced pressure at a temperature of 40–50 ◦C to yield gummy from ␤-galactosidase cleavage of oNPG. Then, 100 ␮lof solids. The extraction yields (w/w) ranged from 0.6 to 12.1% 1M Na2CO3 was added to stop the reaction. In our test (Table 1). The extracts were kept in a refrigerator for further system, for 17␤-estradiol, the incubation time was 50 min, activity assay. and for the samples, it was less than 90 min. The result- ing absorption was measured at 420 nm with a plate reader 2.3. Yeast strain and growth (TECAN, Austria). Each test sample and E2 was assayed in triplicate. E2 as a positive control and DMSO as a For all transactivation assays Saccharomyces cerevisiae negative control for activity were performed in each test strain BJ3505 originally developed by Glaxo was used, which run. C.Z. Zhang et al. / Journal of Ethnopharmacology 98 (2005) 295–300 297

Table 1 Selected Chinese medicinal plants, extraction yields and estrogenic activity of 70% EtOH extracts assessed by the recombinant yeast bioassay

Plant family and species Chinese name Plant part Collection place and Extraction EC50 RIE RP voucher number yield (w/w) (␮g/ml) Amaranthaceae Achyranthes bidentata Bl. Niuxi Root Sichuan, DL-A AB318 6.2––– Amaryllidaceae Curculigo orchioides Gaertn. Xianmao Rhizome Sichuan, DL-ACO412 4.3––– Araliaceae Panax ginseng C. A. Mey. Renshen Root Jilin, DL-APG423 6.7––– Panax notoginseng (Burk.) F. H. Chen Sanqi Root Yunnan, DL-APN216 4.4––– Asteraceae Atractylodes macrocephala Koidz. Baishu Rhizome Jiangsu, DL-AAM617 4.1––– Berberidaceae Epimedium brevicornum Maxim. Yinyanghuo Leaf Shanxi, DL-BEB414 4.3 100.0 31.5 2.30 × 10−4 Convolvulaceae Cuscuta chinensis Lam. Tusizi Seed Guangdong, DL-CCC527 0.6––– Cornaceae Cornus officinalis Sieb.et Zucc. Shanzhuyu Fruit Henan, DL-CCO601 4.6 – – – Cynomoriaceae Cynomorium songaricum Rupr. Suoyang Stem Gansu, DL-CCS620 5.4 112.237.1 1.78 × 10−4 Dipsacaceae Dipsacus asperoides C. Y.Cheng et Xuduan Root Yunnan, DL-DDA519 2.5––– T. M. Ai Eucommiaceae Eucommia ulmoides Oliv. Duzhong Bark Jiangsu, DL-EEU323 8.0––– Iridaceae Belamcanda chinensis (L.) DC. Shegan Rhizome Jiangsu, DL-IBC329 6.5 142.883.7 1.26 × 10−4 Labiatae Scutellaria baicalensis Georgi huangqin Root Neimenggu, DL-LSB325 12.1 262.38.9 8.77 × 10−5 Leguminosae Astragalus membranaceus (Fisch.) Bge. Huangqi Root Gansu, DL-LAM518 9.5 236.117.2 8.47 × 10−5 Cassia obtusifolia L. Juemingzi Seed Liaoning, DL-LAO617 11.660.223.5 3.49 × 10−4 Psoralea corylifolia L. Buguzhi Fruit Henan, DL-LPO423 5.3 120.666.5 1.90 × 10−4 Pueraria lobata (Willd.) Ohwi Gegen Root Sichuan, DL-LPL515 7.8 373.032.6 6.17 × 10−5 Liliaceae Polygonatum odoratum (Mill.) Druce Yuzhu Rhizome Heilongjiang, DL-LPO624 3.2––– Polygonatum sibiricum Red. Huangjing Rhizome Jiangsu, DL-LPS811 8.9––– Magnoliaceae Schisandra chinensis (Turcz.) Baill. Wuweizi Fruit Liaoning, DL-MSC617 9.4––– Oleaceae Ligustrum lucidum Ait. Nuzhenzi Fruit Jiangsu, DL-OLL119 8.7––– Orobanchaceae Cistanche deserticola Y. C. Ma Roucongrong Stem Neimenggu, DL-OCD320 15.6––– Polygonaceae Polygonum cuspidatum Sieb.et Zucc. Huzhang Root and Jiangsu, DL-PPC421 8.46.430.1 3.28 × 10-3 rhizome Polygonum multiflorum Thunb. Heshouwu Root Guizhou, DL-PPM203 4.180.252.1 2.87 × 10−4 Rheum palmatum L. Dahuang Root Jiangsu, DL-PRP421 9.746.739.2 3.85 × 10−4 Ranunculaceae Aconitum carmichaeli Debx. Fuzi Root Helongjiang, DL-RAC124 0.8 – – – Clematis armandii Franch. Chuanmutong Stem Sichuan, DL-RCA515 4.7 – – – Rubiaceae Morinda officinalis How Bajitian Root Guangdong, DL-RMO527 10.6 – – – Scrophulariaceae Rehmannia glutinosa Libosch. Dihuang Root Henan, DL-SRG403 8.9 – – – 298 C.Z. Zhang et al. / Journal of Ethnopharmacology 98 (2005) 295–300

Table 1 ( continued )

Plant family and species Chinese name Plant part Collection place and Extraction EC50 RIE RP voucher number yield (w/w) (␮g/ml) Umbelliferae Angelica sinensis (Oliv.) Diels Danggui Root Gansu, DL-UAS508 6.1 – – – Ligusticum chuanxiong Hort. Chuanxiong Rhizome Sichuan, DL-ULC722 2.1 – – – Cnidium monnieri (L.) Cuss. Shechuangzi Rruit Hebei, DL-UCM123 5.9 – – –

Positive control: the EC50 values of 17 ␤-estradiol was 0.205 ng/ml; the maximal ␤-galactosidase activity of 17 ␤-estradiol was 3.80 u; RP and RIE of 17 ␤-estradiol were both 100 by definition. RP: Estrogenic relative potency; RIE: estrogenic relative inductive efficiency. Values represent the mean of three experiments (standard deviation not shown). (–): Inactive.

2.7. Calculation of β-galactosidase activity 3. Results

The ␤-galactosidase activity is dependant on the binding 3.1. Standard dose–response in yeast of the ligand to the estrogen receptor and was measured ac- cording to Miller (1972), using the following formula (Wang To the induced culture 17␤-estradiol was added to reach et al., 2003): a final hormone concentration between 0.001 and 10 ng/ml   and incubated for 2 h, then the ␤-galactosidase activity A A 50 u = 100% 420 (sample) − 420 (blank) was assayed. The limit of detection was 0.04, the maxi- A600 A6000 t mum ␤-galactosidase activity was 3.80 u and the EC50 was where u is the ␤-galactosidase activity, t the incubation time 0.205 ± 0.025 ng/ml. (min) of enzyme reaction, A420 the absorbance of enzyme reaction at 420 nm and A the absorbance of the cells of the 600 3.2. Estrogenic activities of the selected Chinese sample at 600 nm. medicinal plants 2.8. Curve fitting and EC 50 The EtOH extracts of 32 Chinese medicinal plants used to treat menopausal symptoms were assayed for estrogenic Data derived from transactivation assays were fitted activities by a recombinant yeast system. The test samples in using a four parameters logistic model based on the DMSO were added to the culture reaching final concentra- Marquardt–Levenberg algorithm (Sigmaplot 4.0, SPSS Inc., tions between 0.1 and 1000 ␮g/ml and incubated for 2 h, and Chicago, IL, USA) (Rehmann et al., 1999): then the ␤-galactosidase activity was assayed. [A − D] Table 1 shows that 11 extracts activated the transcription Y =   + D C B 1 + of lacZ. Polygonum cuspidatum, Rheum palmatum, Cassia X obtusifolia and Polygonum multiflorum had a higher estro- where Y is the response value (␤-galactosidase activity), X genic relative potency with RP ranging from 3.28 × 10−3 to the sample concentration, A the maximum induction of ␤- 2.87 × 10−4. Among them, Polygonum cuspidatum had the galactosidase activity, B the relative slope of the middle re- highest estrogenic potency and was about 100,000 times less gion, C the sample concentration at half-maximal response estrogenic than 17␤-estradiol (RP of E2 was 100). On the and D the detection limit. EC50 value is the value of C in the other hand, Belamcanda chinensis, Psoralea corylifolia and equation. Polygonum multiflorum had a higher estrogenic relative in- ductive efficiency with RIE ranging from 83.7 to 52.1 (RIE 2.9. Definition of estrogenic relative potency and of E2 was 100). The results indicated their potential efficacy relative inductive efficiency (RIE) for the treatment of menopausal symptoms.

In order to compare each assay directly, the relative po- tency and the relative inductive efficiency are employed. The 4. Discussion estrogenic relative potency of samples are computed by di- viding the EC50 of E2 by the EC50 of the test samples and then The recombinant yeast cells, MCF-7 human breast can- multiplying these values by 100 (the RP value of E2 is 100 cer cells and a prepubertal mouse uterotrophic bioassay have in the definition). But it is not enough to evaluate estrogenic been used for phytoestrogen screening and environmental es- activity using RP alone (Coldham et al., 1997). The relative trogen assays. The recombinant yeast cell bioassay is approx- inductive efficiency of ␤-galactosidase activity could be used imately two and five orders of magnitude more sensitive to for further evaluating estrogenic activity. The RIE is the ratio E2 than MCF-7 cells and the uterotrophic assay, respectively of maximal ␤-galactosidase activity of the samples to that of (Coldham et al., 1997). So it is thought to be a good method E2 and then multiplying these values by 100 (the RIE value for screening potential estrogens because of its exquisite sen- of E2 is 100 in the definition). sitivity, absence of test compound biotransformation, ease of C.Z. Zhang et al. / Journal of Ethnopharmacology 98 (2005) 295–300 299 use and the possibility of measuring antiestrogenic activity. Acknowledgements But the yeast system has a drawback that is the thick cell wall (Zysk et al., 1995). On the one hand, the limited permeability The authors want to thank Prof. Robert E. Levin, Food Mi- of small molecules might cause the error of assaying estro- crobiology Department of Food Science, University of Mas- genic activity. On the other hand, it blocked the permeation sachusetts, USA and Prof. Le F. Zhang, Dalian Institute of of ␤-galactosidase, which was tested outside the cells. There- Chemical Physics, Chinese Academy of Sciences, China for fore, cell disintegration was the key step to ␤-galactosidase reading through the manuscript and Prof. Wen Z. Wu for tech- assays. It was shown that the sensitivity of chemical dis- nical aid. This study was supported by the National Science integration is lower than that of mechanical disintegration Foundation of China. (Jungbauer and Beck, 2002), but chemical disintegration is still used widely because of its simplicity (Wang et al., 2003; Wu et al., 2002b; Tran et al., 1997; Arnold et al., 1996). Chem- References ical disintegration was used in this study and the experimental ␤ results were very stable and the EC50 of 17 -estradiol was Arnold, S.F., Collins, B.M., Robinson, M.K., Guillette Jr., L.J., Mclachlan, 0.205 ± 0.025 ng/ml. J.A., 1996. Differential interaction of natural and synthetic estrogens Normally, estrogenic activity is judged by the RP value with extracellular binding proteins in a yeast estrogen screen. Steroids (Wu et al., 2002a). According to this, Polygonum cuspidatum 61, 642–646. is the highest estrogenic, followed by Rheum palmatum, Andersen, H.R., Andersson, A.M., Arnold, S.F., Autrup, H., Barfoed, M., Bereford, N.A., Bjerregaard, P., Christiansen, L.B., Gissel, B., Hum- Cassia obtusifolia, Polygonum multiflorum, Epimedium mel, R., Jorgensen, E.B., Korsgaard, B., Le Guevel, R., Leffers, H., brevicornum, Psoralea corylifolia, Cynomorium songar- McLachlan, J., Moller, A., Nielsen, J.B., Olea, N., Oles-Karasko, A., icum, Belamcanda chinensis, Scutellaria baicalensis and Pakdel, F., Pedersen, K.L., Perez, P., Skakkeboek, N.E., Sonnenschein, Astragalus membranaceus. Estrogenic activity is also judged C., Soto, A.M., 1999. Comparison of short-term estrogenicity tests for as fully estrogenic if the RIE is >75%, partially estrogenic if identification of hormone-disrupting chemicals. Environmental Health Perspectives 107, 89–108. the RIE is 25–75%, weakly estrogenic if the RIE is 10–25% Beral, V., Banks, E., Reeves, G., Appleby, P., 1999. Use of HRT and the and negative if the RIE is below 10% (Andersen et al., subsequent risk of cancer. Journal of Epidemiology and Biostatistics 1999). According to this, Belamcanda chinensis is fully 4, 191–210. estrogenic, Cynomorium songaricum, Epimedium brevi- Breithofer, A., Graumann, K., Scicchitano, M.S., Karathanasis, S.K., Butt, cornum, Polygonum cuspidatum, Polygonum multiflorum, T.R., Jungbauer, A., 1998. Regulation of human estrogen receptor by phytoestrogens in yeast and human cells. Journal of Steroid Biochem- Psoralea corylifolia, Pueraria lobata and Rheum palmatum istry and Molecular Biology 67, 421–429. are partially estrogenic, Cassia obtusifolia and Astragalus Chen, Y.J., Xu, S.X., Ma, Q.F., Pei, Y.P., Xie, H., Yao, X.S., 1987. Studies membranaceus are weakly estrogenic, and other species on new minor saponins isolated from leaves of Panax ginseng C. A. are negative. These data suggest a potential use of certain Meyer. Acta Pharmaceutica Sinica 22, 685–689. Chinese crude drugs against hormone-dependent breast Coldham, N.G., Dave, M., Sivapathasundaram, S., McDonnell, D.P., Con- nor, C., Sauer, M.J., 1997. Evaluation of a recombinant yeast cell and prostate cancers, and cardiovascular diseases, and as estrogen screening assay. Environmental Health Perspectives 105, estrogen replacement therapy for postmenopausal women. 734–742. The physiological role of bioactive compounds present in Cos, P., De Bruyne, T., Apers, S., Vanden Berghe, D., Pieters, L., Vliet- plants has attracted more attention over the last decade, es- inck, A.J., 2003. Phytoestrogens: recent developments. Planta Medica pecially the phytoestrogens. Traditional Chinese medicines 69, 589–599. Glazier, M.G., Bowman, M.A., 2001. A review of the evidence for the have been used in the treatment of menopausal symptoms use of phytoestrogen for traditional estrogen replacement therapy. for several thousands of years in Asian countries, but many Archives of Internal Medicine 16, 1161–1171. mysteries remain about the relationship between the active Harlow, B.L., Signorello, L.B., 2000. Factors associated with early component and their function owing to the complexity of the menopause. Maturitas 35, 3–9. mixtures. It has been proven that phytoestrogens of genistein Jungbauer, A., Beck, V., 2002. Yeast reporter system for rapid determi- nation of estrogenic activity. Journal of Chromatography B-Analytical and daidzein, the main components of soy, make a contribute Technologies in the Biomedical and Life Sciences 777, 167–168. to the inhibition of the growth of breast cell lines and re- Liu, J., Burdette, J.E., Xu, H., Gu, C., Breemen, R.B., Bhat, K.P., Booth, duce bone turnover (Vissac-Sabatier et al., 2003; Setchell N., Constantinou, A.I., Pezzuto, J.M., Fong, H.H., Farnsworth, N.R., and Lydeking-Olsen, 2003). Bolton, J.L., 2001. Evaluation of estrogenic activity of plant extracts Based on their chemical structure, phytoestrogens can for the potential treatment of menopausal symptoms. Journal of Agri- cultural and Food Chemistry 49, 2472–2479. be classified into four main groups, i. e., isoflavonoids, Miller, J.M., 1972. Assays for ␤-glactosidase. In: Miller, J.M. (Ed.), Ex- flavonoids, stilbenes and lignans (Cos et al., 2003). It is periments in Molecular Genetics. Cold Spring Harbor, New York, pp. known that isoflavonoids, flavonoids, anthraquinones, triter- 352–355. penes and saponins are the main components in the 11 active Nichols, K.C., Schenkel, L., Benson, H., 1984. 17 Beta-estradiol for post- herbs, but it is unknown whether or not a single component menopausal estrogen replacement therapy. Obstetrical and Gynaeco- logical Survey 39, 230–245. or the additive effect of certain components contributes to Rehmann, K., Schramm, K.-W., Kettrup, A., 1999. Applicability of a estrogenic activity. Further studies will focus on the isolation yeast oestrogen screen for the detection of oestrogen-like activities in and identification of active compounds in the active species. environmental samples. Chemosphere 38, 3303–3312. 300 C.Z. Zhang et al. / Journal of Ethnopharmacology 98 (2005) 295–300

Routledge, E.J., Sumpter, J.P., 1996. Estrogenic activity of surfactants Vissac-Sabatier, C., Bignon, Y.J., Bernard-Gallon, D.J., 2003. Effects of and some of their degradation products assessed using a recombinant the phytoestrogens genistein and daidzein on BRCA2 tumor suppres- yeast assay. Environmental Toxicology and Chemistry 15, 241–248. sor gene expression in breast cell lines. Nutrition and Cancer—An Santiso-Mere, D., Nawaz, Z., McDonnell, D.P., O’Malley, B.W., 1991. International Journal 45, 247–255. The development of S. cerevisiae as a host system for steroid hor- Wang, J., Wu, W., Henkelmann, B., You, L., Kettrup, A., Schramm, K.- mone action studies. In: Huges, M.R., Schrader, W.T., O’Malley, B.W. W., 2003. Presence of estrogenic activity from emission of fossil fuel (Eds.), Hormone Action and Molecular Endocrinology. Houston Bio- combustion as detected by a recombinant yeast bioassay. Atmospheric logical Associates Inc., p. 23-1-24. Environment 37, 3225–3235. Setchell, K.D., Lydeking-Olsen, E., 2003. Dietary phytoestrogens and Wu, W.Z., Chen, J., Rehmann, K., Schramm, K.-W., Kettrup, A., 2002a. their effect on bone: evidence from in vitro and in vivo, human obser- Estrogenic effects from household stoves. Ecotoxicology and Envi- vational, and dietary intervention studies. American Journal of Clinical ronmental Safety 53, 65–69. Nutrition 78, 593S–609S. Wu, W.Z., Wang, J.X., Xu, Y., Huang, M.H., 2002b. Recombinant gene Thompson, L.U., 1993. Potential health benefits and problems associated yeast for assaying environmental estrogen pollutants. China Environ- with antinutrients in foods. Food Research International 26, 131–149. mental Science 22, 60–63. Tran, D.Q., Jin, L.Z., Chen, J., McLachlan, J.A., Arnold, S.F., 1997. Zacharewski, T., 1997. In vitro bioassays for assessing estrogenic sub- Evaluation of clinical and environmental anti-estrogens with human stances. Environmental Science and Technology 31, 613–623. estrogen receptor expressed in Saccharomyces cerevisiae: a novel role Zysk, J.R., Johnson, B., Ozenberger, B.A., Bingham, B., Gorski, J., 1995. for ABC-cassette transporters in mediating anti-estrogenic activity. Selective uptake of estrogenic compounds by Saccharomyces cere- Biochemical and Biophysical Research Communications 235, 669– visiae: a mechanism for antiestrogen resistance in yeast expressing 674. the mammalian estrogen receptor. Endocrinology 13, 1323–1326. Journal of Ethnopharmacology 98 (2005) 301–305

Effect of propolis, some isolated compounds and its source plant on antibody production

J.M. Sforcin a,∗, R.O. Orsi a, V. Bankova b

a Department of Microbiology and Immunology, Biosciences Institute, UNESP, 18618-000 Botucatu, S.P., Brazil b Institute of Organic Chemistry with Centre of Phytochemistry, Bulgarian Academy of Sciences, 1113 Sofia, Bulgaria Received 1 May 2003; received in revised form 1 January 2005; accepted 17 January 2005

Abstract

Propolis is a beehive product with a very complex chemical composition, widely used in folk medicine because of its several therapeutic activities. Its biological properties and chemical composition may vary according to the geographic location and to the different plant sources. The possible mechanism of action of propolis as well as of its active compounds has been the subject of researchers in recent years. In this work, first we reported the results of our study on the seasonal effect of the immunomodulatory action of propolis on antibody production in bovine serum albumin (BSA)-immunized rats. Then, we compared the effect of Brazilian and Bulgarian propolis, some isolated compounds and Baccharis extract on anti-BSA antibody levels. Based on the results, we conclude that propolis stimulates antibody production, independently of the season and geographic origin. Caffeic acid, quercetin and Baccharis extract had no effect on antibody production, although the importance of isolated compounds is well reported in other biological assays. Propolis action is a consequence of plant-derived products with synergic effects, while isolated compounds or extracts from its plant sources had no effect in this assay. © 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Propolis; Antibody; Caffeic acid; Quercetin; Baccharis

1. Introduction no poplars in tropical regions and the main propolis source in Botucatu, Sao˜ Paulo State, Brazil, is Baccharis dracun- Propolis is a resinous hive product, built by honeybees culifolia DC. (Compositae), followed by Eucalyptus citri- from various plant sources, consisting of a complex mixture odora Hook. (Myrtaceae) and Araucaria angustifolia (Bert.) of constituents (Burdock, 1998; Bankova et al., 2000). This O. Kuntze (Araucariaceae) (Bankova et al., 1999). Moreover, natural product has been used in folk medicine and its var- in the temperate zone of the Northern Hemisphere, col- ious biological activities have been reported, including an- lect propolis only in the summer (including late spring and tibacterial (Sforcin et al., 2000), antiviral (Vynograd et al., early autumn), while in Brazil, propolis collection proceeds 2000), antiinflammatory (Khayyal et al., 1993), anticarcino- throughout the year, but seasonal variations are possible. genic (Bazo et al., 2002) and immunomodulatory (Sforcin With respect to the immune system, in recent studies in et al., 2002a; Sa-Nunes´ et al., 2003) activities. our laboratory, we have shown that propolis increases the nat- Its biological properties and chemical composition may ural killer activity against tumor cells (Sforcin et al., 2002a), vary according to the geographic location and to the differ- modulates both in vitro and in vivo nitric oxide and hydro- ent plant sources (Markham et al., 1996). In temperate zone, gen peroxide production by peritoneal macrophages (Orsi propolis is collected from the exsudate of poplar trees, et al., 2000), and increases the fungicidal activity of these such as Populus species (Bankova et al., 1998b). There are cells (Murad et al., 2002). The possible mechanism of action of propolis, as well ∗ Corresponding author. as its active compounds, has been the subject of research in E-mail address: [email protected] (J.M. Sforcin). recent years (Banskota et al., 2001).

0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2005.01.042 302 J.M. Sforcin et al. / Journal of Ethnopharmacology 98 (2005) 301–305

In this work, we wish to report the results of our study on ethanol effect as propolis solvent; while G6, control, received the effect of seasonal variations over propolis immunomod- physiological salt solution (0.9% NaCl) only. ulatory action, specifically on antibody production, by BSA- Animals received 0.4 ml of the respective solutions by gav- immunized rats. We compared the effect of propolis from age, twice a day, for 3 days (Sforcin et al., 2002a). After 24 h Brazil against propolis from Bulgaria, and some active com- of the last treatment, rats were immunized through sc in- pounds on anti-BSA antibody levels. The effect of Baccharis jection with bovine serum albumin (Sigma: 4 mg/ml) with extract, one of the source plants in Brazil, was also analyzed, complete Freund adjuvant, in 0.5 ml saline. After 15 and 30 in order to compare the plant source with propolis activity. days, rats were similarly immunized (sc) with BSA (4 mg/ml) with incomplete Freund adjuvant and only BSA (2 mg/ml), respectively. 2. Methodology 2.4.2. Second assay 2.1. Propolis samples The effect of Brazilian and Bulgarian propolis, as well as the effect of caffeic acid, quercetin and Baccharis extract, Propolis was produced by Africanized honeybees (Apis were analysed for antibody production in order to observe mellifera L.) in the Beekeeping Section of the Lageado Farm a possible involvement of active compounds as well as the (UNESP, Botucatu). Samples were collected throughout the main vegetal source of propolis in this assay. year from plastic nets. At the end of each month, nets were Animals were divided into seven groups of seven rats each. taken and frozen to facilitate propolis removal. Samples were G1 and G2 were treated with Brazilian and Bulgarian 10% pooled by season and were ground and extracted (30 g of propolis extracts, respectively. G3 and G4 received caffeic propolis, completing the volume to 100 ml with 70% ethanol), acid and quercetin (100 mg/kg), respectively. G5 received protecting it from bright light, with moderate shaking, at room 10% Baccharis extract. G6 received 10% hydroalcoholic so- temperature. After a week, extracts were filtered and final lution (propolis vehicle) and G7 was considered as control concentrations were calculated, thus obtaining the dry weight (0.9% NaCl). of the extracts (120 mg/ml) (Sforcin et al., 1995). Animals were treated and immunized as previously de- Bulgarian propolis was kindly given by Dr. Bankova and scribed. was prepared in the same way (dry weight: 170 mg/ml). 2.5. Serum 2.2. Caffeic acid and quercetin administration Blood samples were collected 15 and 30 days after the last Caffeic acid and quercetin were isolated at the Institute immunization by retro-orbital puncture, in order to determine of Organic Chemistry in Bulgaria by Dr. Bankova and were the anti-BSA antibodies. Blood samples were centrifuged, dissolved in hot distilled water before administration to rats serum was taken and aliquots were frozen at −20 ◦C. (100 mg/kg body weight), giving 0.4 ml p.o. 2.6. Humoral immune response 2.3. Baccharis extract Enzyme-linked immunosorbent assay (ELISA) was used Extracts of Baccharis dracunculifolia were prepared in to detect antibodies (anti-BSA IgG) (Ivanovska et al., 1997). the same way, obtaining 10% Baccharis extracts (1 ml ex- Plates were coated by overnight incubation with BSA tract+9ml distilled water) (Lopes et al., 2003). The plant (20 ␮g/ml) in carbonate–bicarbonate buffer 0.2 M, pH 9.6, was identified in the Herbario´ Botu, Department of Botany then washed with phosphate-buffered saline (PBS)/0.05% of Biosciences Institute, UNESP, where a voucher specimen Tween 20 (0.1 ml) and quenched with PBS/5% low-fat milk BOTU (09867–18.03.98) has been deposited. (0.12 ml). After 30 min, appropriately diluted test serum in PBS/5% low-fat milk was added for 1 h at 37 ◦C. 2.4. Animals, treatments and immunization procedure Plates were washed and 80 ␮l goat peroxidase-labelled antirat IgG was added. Within 45 min, 80 ␮lofo- Male rats (Rattus norvegicus) weighing 200 g were used. phenylenediamine (0.4 mg/ml in 0.1 M phosphate citrate buffer, pH 5.5), containing 0.8 ␮l/ml hydrogen peroxide, was 2.4.1. First assay added to each well. After 30 min, 0.05 ml of 4 M H2SO4 was In order to investigate the seasonal effect on Brazilian added and the absorbance (OD 492 nm) was measured using propolis activity, 42 rats were divided into 6 groups (G1, G2, ELISA reader (Labsystems). G3, G4, G5 and G6) of 7 rats each. G1, G2, G3 and G4 received 10% propolis extracts (1 ml of 2.7. Statistical analysis propolis+9mldistilled water) from spring, summer, autumn and winter samples, respectively. G5 received 10% hydroal- Antibody titers were determined as −log2 of the coholic solution (Labsynth), in order to observe a possible last dilution to statistical analysis. Non-parametric J.M. Sforcin et al. / Journal of Ethnopharmacology 98 (2005) 301–305 303

Fig. 1. Effect of seasons on propolis action on anti-BSA antibody titers after 15 (A) and 30 (B) days of immunization. Geographic origin, isolated compounds and Baccharis effect on antibody production after 15 (C) and 30 (D) days of immunization. Results are expressed in −log2 X (n = 7). (A) Control significantly different from four seasons (p < 0.05). (C) Control significantly different from Brazilian and Bulgarian propolis (p < 0.05). (D) Brazilian propolis significantly different from Baccharis, caffeic acid and quercetin (p < 0.05).

Kruskal–Wallis test was used to independent samples (Zar, tion (Fig. 1C; Table 2)(p < 0.05). Although the effect of 1984). Brazilian propolis was a little higher than the control and the Bulgarian propolis after 30 days, no significant differences 3. Results were detected (Fig. 1D; Table 2). Hydroalcoholic solution-treated groups showed a pattern Brazilian propolis administration to rats increased anti- of antibody production similar to control, in all assays. body production after 15 days of immunization (Fig. 1A; Caffeic acid and quercetin had no effects on antibody pro- Table 1)(p < 0.05). No significant differences were seen af- duction. ter 30 days of the last immunization (Fig. 1B; Table 1). Baccharis dracunculifolia extract increased antibody pro- No differences were seen between the samples from each duction not significantly after 15 days of immunization season. when compared to control, but efficiently when compared Brazilian and Bulgarian propolis stimulated antibody pro- to propolis-treated rats. No effects were seen after 30 days duction, in the same magnitude, after 15 days of immuniza- (Fig. 1C and D).

Table 1 Effect of season on Brazilian propolis action on antibody production Control Alcohol Spring Summer Autumn Winter 15 daysa 1:8.192 1:16.384 1:65.536 1:131.072 1:65.536 1:65.536 30 days 1:65.536 1:131.072 1:131.072 1:131.072 1:65.536 1:131.072 Anti-BSA antibody titer of control groups and groups treated with propolis from each season, 15 and 30 days after immunization. a Control significantly different from four seasons (p < 0.05). 304 J.M. Sforcin et al. / Journal of Ethnopharmacology 98 (2005) 301–305

Table 2 Geographic origin, isolated compounds of Brazilian and Bulgarian propolis and Baccharis effect on antibody production Control Alcohol Brazil Bulgaria Baccharis Caffeic acid Quercetin 15 daysa 1:16.384 1:16.384 1:262.144 1:524.288 1:131.072 1:16.384 1:8.192 30 daysb 1: 16.384 1: 16.384 1:131.072 1:65.536 1:4.096 1:4.096 1:4.096 Anti-BSA antibody titer of control groups and groups treated with Brazilian and Bulgarian propolis, Baccharis extract, caffeic acid and quercetin, 15 and 30 days after immunization. a Control significantly different from Brazilian and Bulgarian propolis (p < 0.05). b Brazilian propolis sifgnificantly different from Baccharis, caffeic acid and quercetin (p < 0.05).

4. Discussion and conclusion garian propolis had a modulatory action on macrophages, without significant differences between them (Murad et al., Many scientific articles have been published in the last 2002). years related to the biological activities of propolis. Banskota From our data, it is clear that caffeic acid and quercetin et al. (2001) compiled important findings on pharmacological have no effects on antibody production. Caffeic acid properties of propolis and its possible mechanism of action, phenethyl ester and quercetin are propolis constituents re- as well as its active constituents. sponsible for several biological properties, such as antimi- Propolis administration to rats increased antibody produc- crobial effect (Mirzoeva et al., 1997). Caffeic acid esters pos- tion. Propolis ability to modulate antibody synthesis is a part sess significant cytotoxicity towards various tumor cells and of its adjuvant activity, since it has been shown recently that antitumor activity (Lee et al., 2000), although other pheno- propolis has a potent effect on different cells of innate im- lic compounds and diterpenoids isolated from propolis also mune response (Orsi et al., 2000; Murad et al., 2002; Sforcin have antitumor properties (Banskota et al., 2001). Besides the et al., 2002a). effect of individual constituents, there may be synergistic ef- Scheller et al. (1988) observed that propolis stimulates fects of several compounds, thus conferring propolis different antibody production in mice immunized by sheep red blood pharmacological activities. cells (SRBC) using a different methodology (plaque-forming Kujumgiev et al. (1999) suggested that general pharma- cells). These authors concluded that propolis acts on a cological properties of propolis are due to a natural mix- short-term basis on the immune system. Ivanovska et al. ture of its components and a single propolis constituent (1993) observed that SRBC, administered simultaneously does not have an activity greater than that of the total with cinnamic acid lysine derivative (CN.Ly), strongly el- extract. evated the antibody titers of mice, through haemagglutina- Baccharis dracunculifolia extract, the main source of tion determination. Increasing concentrations of CN.Ly low- propolis in our region, did not increase antibody production ered the antibody levels. Some propolis constituents aug- significantly after 15 days of immunization when compared ment host-defence response influencing lymphocyte prolif- to control, but efficiently when compared to propolis-treated eration and induction of cytokines release (Ivanovska et al., rats. No effects were seen after 30 days. 1995). We evaluated the action of three vegetal sources of propo- With respect to the effects of seasons on propolis activity, lis (Araucaria angustifolia, Baccharis dracunculifolia and no differences were seen between samples from each season. Eucalyptus citriodora) on macrophages activation, through These data are in accordance with those previously obtained oxygen (H2O2) and nitrogen (NO) metabolites determina- in our laboratory (Sforcin et al., 2000, 2001, 2002a, 2002b). tion, comparing these results with those previously obtained The chemical analysis of our propolis samples by gas chro- with propolis, in the same assays, in our laboratory (Orsi matography (GC), gas chromatography–mass spectrometry et al., 2000). Propolis and plant secretions were investigated (GC–MS) and thin-layer chromatography (TLC) revealed using TLC and GC–MS analysis (Bankova et al., 1999). The that seasonal variations in its composition are not significant data suggest that no effects are attributed to such extracts on and are predominantly quantitative (Boudourova-Krasteva metabolite production, while propolis induces an elevation et al., 1997; Bankova et al., 1998a,b). in H2O2 release and inhibits NO generation, depending on In the second part of this investigation, we compared the concentration (Lopes et al., 2003). Propolis action is a Brazilian and Bulgarian propolis, since their biological activ- consequence of plant-derived products and isolated extracts ities and chemical compositions may show some differences of its vegetal sources, which did not have the same effect in because of the local flora. One may see that both samples this assay. stimulated antibody production, in the same magnitude, af- Raised antibody titers are important for microbial killing ter 15 days of immunization. Although Brazilian propolis by antibody-mediated mechanisms. Based on the results of showed a little higher activity than control and the Bulgar- the present work, we may conclude that propolis stimulates ian propolis after 30 days, no significant differences were antibody production, independently of the season and geo- detected. These results are in conformity with our recent graphic origin, while caffeic acid, quercetin and Baccharis work, in which we observed that both Brazilian and Bul- extract had no effect on antibody production. J.M. Sforcin et al. / Journal of Ethnopharmacology 98 (2005) 301–305 305

New assays are being carried out in our laboratory, in order Lee, Y.J., Liao, P.H., Chen, W.K., Yang, C.Y., 2000. Preferential cytotox- to investigate propolis effect on the immune system and to icity of caffeic acid phenethyl ester analogues on oral cancer cells. permit a better understanding of its mechanism of action. Cancer Letters 153, 51–56. Lopes, F.C., Bankova, V., Sforcin, J.M., 2003. Effect of three vegetal sources of propolis on macrophages activation. Phytomedicine 10, 343. References Markham, K.E., Mitchel, K.A., Wilkins, A.L., Daldy, J.A., Lu, Y., 1996. HPLC and GC–MS identification of the major organic constituents in Bankova, V., Boudourova-Krasteva, G., Popov, S., Sforcin, J.M., Funari, New Zealand propolis. Phytochemistry 42, 205–211. S.R.C., 1998a. Seasonal variations in essential oil from Brazilian Mirzoeva, O.K., Grishanin, R.N., Calder, P.C., 1997. Antimicrobial ac- propolis. Journal of Essential Oil Research 10, 693–696. tion of propolis and its components, the effects on growth mem- Bankova, V., Boudourova-Krasteva, G., Popov, S., Sforcin, J.M., Funari, brane potential, and motility of bacteria. Microbiological Research S.R.C., 1998b. Seasonal variations of the chemical composition of 152, 239–246. Brazilian propolis. Apidologie 29, 361–367. Murad, J.M., Calvi, S.A., Soares, A.M.V.C., Bankova, V., Sforcin, J.M., Bankova, V., Boudourova-Krasteva, G., Sforcin, J.M., Frete, X., Ku- 2002. Effects of propolis from Brazil and Bulgaria on fungicidal ac- jumgiev, A., Maimoni-Rodella, R., Popov, S., 1999. Phytochemical tivity of macrophages against Paracoccidioides brasiliensis. Journal evidence for the plant origin of Brazilian propolis from Sao˜ Paulo of Ethnopharmacology 79, 331–334. State. Zeitschrift fur¨ Naturforschung 54c, 401–405. Orsi, R.O., Funari, S.R.C., Soares, A.M.V.C., Calvi, S.A., Oliveira, S.L., Bankova, V., Castro, S.L., Marcucci, M.C., 2000. Propolis: recent ad- Sforcin, J.M., Bankova, V., 2000. Immunomodulatory action of propo- vances in chemistry and plant origin. Apidologie 31, 3–15. lis on macrophage activation. Journal of Venomous Animals and Tox- Banskota, A.H., Tezuka, Y., Kadota, S., 2001. Recent progress in phar- ins 6, 205–219. macological research of propolis. Phytotherapy Research 15, 561– Sa-Nunes,´ A., Faccioli, L.H., Sforcin, J.M., 2003. Propolis: lymphocyte 571. proliferation and IFN-␥ production. Journal of Ethnopharmacology Bazo, A.P., Rodrigues, M.A.M., Sforcin, J.M., Camargo, J.L.V., Ribeiro, 87, 93–97. L.R., Salvadori, D.M.F., 2002. Protective action of propolis on the rat Scheller, S., Gazda, G., Pietsz, G., Gabrys, J., Szumlas, J., Eckert, L., colon carcinogenesis. Teratogenesis, Carcinogenesis and Mutagenesis Shani, J., 1988. The ability of ethanol extract of propolis to stimulate 22, 183–194. plaque formation in immunized mouse spleen cells. Pharmacological Boudourova-Krasteva, G., Bankova, V., Sforcin, J.M., Nikolova, N., Research Communications 20, 323–328. Popov, S., 1997. Phenolics from Brazilian propolis. Zeitschrift fur¨ Sforcin, J.M., Fernandes Jr., A., Lopes, C.A.M., Bankova, V., Funari, Naturforschung 52c, 676–679. S.R.C., 2000. Seasonal effect on Brazilian propolis antibacterial ac- Burdock, G.A., 1998. Review of the biological properties and toxicity of tivity. Journal of Ethnopharmacology 73, 243–249. propolis (propolis). Food and Chemical Toxicology 36, 347–363. Sforcin, J.M., Fernandes Jr., A., Lopes, C.A.M., Funari, S.R.C., Bankova, Ivanovska, N., Neychev, H., Stefanova, Z., Bankova, V., Popov, S., 1995. V., 2001. Seasonal effect of Brazilian propolis on Candida albicans Influence of cinnamic acid on lymphocyte proliferation, cytokine re- and Candida tropicalis. Journal of Venomous Animals and Toxins 7, lease and Klebsiella infection in mice. Apidologie 26, 73–81. 139–144. Ivanovska, N., Philipov, S., Georgieva, P., 1997. Immunopharmacologi- Sforcin, J.M., Funari, S.R.C., Novelli, E.L.B., 1995. Serum biochemical cal activity of aporphinoid alkaloid oxoglaucine. Pharmacological Re- determinations of propolis-treated rats. Journal of Venomous Animals search 35, 267–272. and Toxins 1, 31–37. Ivanovska, N., Stefanova, Z., Valeva, V., Neychev, H., 1993. Immunomod- Sforcin, J.M., Kaneno, R., Funari, S.R.C., 2002a. Absence of seasonal ef- ulatory action of propolis. VII. A comparative study on cinnamic and fect on the immunomodulatory action of Brazilian propolis on natural caffeic lysine derivatives. Comptes Rendus de lAcademie Bulgare des killer activity. Journal of Venomous Animals and Toxins 8, 19–29. Sciences 46, 115–117. Sforcin, J.M., Novelli, E.L.B., Funari, S.R.C., 2002b. Seasonal effect of Khayyal, M.T., el-Ghazaly, M.A., el-Khatib, A.S., 1993. Mechanisms in- Brazilian propolis on seric biochemical variables. Journal of Ven- volved in the antiinflammatory effect of propolis extract. Drugs Ex- omous Animals and Toxins 8, 244–254. perimental and Clinical Research 19, 197–203. Vynograd, N., Vynograd, I., Sosnowski, Z., 2000. A comparative multi- Kujumgiev, A., Tsvetkova, I., Serkedjieva, Y., Bankova, V., Christov, R., centre study of the efficacy of propolis, acyclovir and placebo in the Popov, S., 1999. Antibacterial, antifungal and antiviral activity of treatment of genital herpes (HSV). Phytomedicine 7, 1–6. propolis of different geographic origin. Journal of Ethnopharmacology Zar, J.H., 1984. Biostatistical Analysis, second ed. Prentice-Hall, Engle- 64, 235–240. wood Cliffs, 718 pp. Journal of Ethnopharmacology 98 (2005) 307–312

Medicinal plant wealth of local communities in some villages in Shimoga District of Karnataka, India Parinitha Mahishi, B.H. Srinivasa, M.B. Shivanna ∗

Department of Studies and Research in Applied Botany, Kuvempu University Jnana Sahyadri, Shankaraghatta 577 451, Shimoga District, Karnataka, India Received 1 November 2003; received in revised form 1 January 2005; accepted 17 January 2005

Abstract

An ethnomedicinal survey (1998–2000) was conducted in three villages of Shimoga district of Karnataka, India, using a questionnaire designed by Sinha (1996) [Sinha, R.K., 1996. Ethnobotany—The Renaissance of Traditional Herbal Medicine. Ina Shree Publishers, Jaipur, India, 242 pp.]. The herbal practitioners in the study area were interviewed and information on medicinal plants, their local names, habitat and their seasonal availability was collected. The survey revealed the utilization of 47 species of plants belonging to 46 genera in 28 families used to treat 9 infectious and 16 non-infectious diseases. Twelve new claims on ethnomedical knowledge were reported and there were formulations that were similar to that described already in the literature. © 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Ethnomedicine; Local community; Infectious diseases; Non-infectious diseases; Western Ghats; Karnataka

1. Introduction in three villages are still practicing herbal medicine exten- sively in their primary health care. These villages are located Traditional knowledge of herbal remedy to treat human next to Bhadra wild life sanctuary. There are no previous diseases is fast declining in many parts of the world, includ- records on ethnomedical knowledge from the study area. ing India. Even today, tribals and certain local communities Hence, an attempt has been made to document plant species, in India still practice herbal medicine to cure a variety of medicinal formulations and treatment of particular diseases diseases and disorders. They collect and preserve locally by various communities residing in this area. available, wild and cultivated plant species. Compared to the large number of floristic surveys in southern peninsular India (Saldanha and Nicolson, 1976; Yoganarasimhan et al., 1981; 2. Methodology Gamble, 1995), there are few surveys that reveal the practice of herbal medicine by either tribals or indigenous communi- The study covered three villages: Nellisara, Malenahalli ties (Bhandary et al., 1995, 1996; Harsha et al., 2002, 2003; and Shankaraghatta located about 30 km away from Shimoga ◦ Parinitha et al., 2004). It is apparent from these surveys that town, at an elevation of 620 m above sea level, 13 43 latitude ◦ tribals and communities residing in remote places followed and 75 38 longitude (Fig. 1). The study area is situated in the different practices. However, even certain local communities Western Ghats of Karnataka, which is one of the biodiversity residing near towns and cities do follow traditional healing hotspots in India. The population of the study area is about systems. A preliminary survey of villages around Shimoga 2500, comprising various communities and castes whose ma- town of Karnataka, revealed that local communities residing jor occupation is agriculture, while some are labourers. About 60% of households in these villages use locally ∗ Corresponding author. Fax: +91 8282 256262. available, wild and cultivated medicinal plants to treat com- E-mail address: [email protected] (M.B. Shivanna). mon diseases. In each village, medicinal plants are used by

0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2005.01.035 308 P. Mahishi et al. / Journal of Ethnopharmacology 98 (2005) 307–312

Fig. 1. Map of the study area. healers in different formulations to treat ailments, including years, who practiced herbal medicine in the villages of the skin diseases, stomach and kidney ailments, asthma, cough, study area, were interviewed. They had knowledge of 47 diabetes, leprosy, jaundice and wounds. species of medicinal plants used in therapy, belonging to 46 Extensive surveys were undertaken for the period of genera in 28 families, to treat 25 diseases and disorders. Of 1998–1999 and 1999–2000 in the study area for the purpose the 47 species, 28 were collected from the wild, 11 cultivated, of documenting plants used by the local communities. They and 8 both collected from the wild and cultivated (25 trees, 10 were convinced of the academic significance and bona fide , 9 herbs and 3 climbers). For the sake of convenience, intention of the study through repeated contacts, explana- 25 diseases and disorders were grouped into infectious and tions and interviews. They consented orally to document and non-infectious diseases (Tables 1 and 2 ). The information publish the results of the study in the interest of the soci- gathered is arranged alphabetically by disease together with ety. A previously prepared questionnaire designed by Sinha the botanical name of the plants, their families, local and (1996) was used to collect ethnobotanical information from common names and information on part used, method of the herbal practitioners and knowledgeable elders in the study preparation, dosage, duration, ingredients and other recom- area, and information on the plant species and their parts used mendations (Tables 1 and 2). for the formulation of medicine. Information on the habitat of The present study reveals that the local medicine men of the plants, their local names and seasonal availability was also the study area have good knowledge of the medicinal property collected. Plants were identified with the help of published re- of a variety of plant species that grow around their locality. gional flora (Saldanha and Nicolson, 1976; Yoganarasimhan They use 20 plant species to treat nine infectious diseases and et al., 1981; Gamble, 1995) and by comparing voucher speci- 30 species to treat 16 non-infectious diseases. Twelve species mens with identified herbarium collections. A set of voucher of plants belonging to 19 genera and 14 families have not been specimens has been deposited at the Department of Applied previously cited in the literature for the treatment of human Botany, Kuvempu University. The information recorded was diseases. Among the species of plants listed in Tables 1 and 2, further ascertained or cross-checked by consulting the bene- five are endangered in the wild, three are vulnerable and two ficiaries, villagers and other medicine men. The conservation are of lower risk category based on the Red Book category status of each medicinal plant species collected was assessed (Nayar and Sastri, 1990; Gowda et al., 1997; Seetharam et al., using the IUCN Red list and its criteria (Nayar and Sastri, 1998; Ravikumar and Ved, 2000). Most of the species used in 1990; Gowda et al., 1997; Ravikumar and Ved, 2000). the preparation of herbal medicine are collected fresh; very rarely, stored materials are used. Among the various plant parts used for the herbal formulations, leaves, followed by 3. Results and discussion stem bark and root, were preferred over other plant parts. Upon interviews with the beneficiaries, elders and Twenty informants of Nayaka, Chelvadi, Lambani, Tamil- residents of the study area and neighbouring villages, they ian and Muslim communities, in the age group of 30–80 unanimously agree to the efficacy of the herbal formulation P. Mahishi et al. / Journal of Ethnopharmacology 98 (2005) 307–312 309

Table 1 Plant species used to treat infectious diseases Disease Botanical name and family Local (common) name Part and method of use Oldenlandia auricularia K. Schum.a Nelae nekkare (Indian Paste of whole plant mixed with cow’s milk or water (Rubiaceae) (KU/SG/V 055) madder) and taken orally with sugar, for 10–12 days. Terminalia paniculata Rothm.a (Combretaceae) Hunalu (Flowering Flowers mixed with leaves of Cocculus villosus DC.a (KU/SG/V 056) murdah) (Vasantikta) (KU/SG/MH 322) plant, made into a paste and taken orally. Chronic dysentery Phyllanthus amarus Schum. et Thonn.a Nela nelli Leaves ground with Acacia Senegal Willd.a (Euphorbiaceae) (KU/SG/V 057) (Bhumyamalaki) (Snetakhadira) (KU/SG/N 323) leaves, add sugar and give orally, or tender leaves ground with cow’s milk curd given orally, for 2–5 days, before food. Mangifera indica L.b (Anacardiaceae) Mavu (Mango) Bark powdered, sieved and taken with cow’s milk, (KU/SG/V 058) two to three times a day. Jatropha curcas L.c (Euphorbiaceae) (KU/SG/V Bettada haralu (Angular Fruit and seeds ground and mixed with cow’s milk, 059) leaved physic ) taken orally for 2 days. Syzygium cumini Lam.c (Myrtaceae) (KU/SG/V Naeralae (Black plum) Stem bark macerated with cow’s milk curd taken 060) orally, for 2–3 days before food. Holarrhena antidysenterica (L.) Wall,a [EW] Kodasalu (Kurchi) Paste made with flower and cow’s milk taken orally, (Apocynaceae) (KU/BS/KG 032) for 4 days. Cough Ocimum sanctum L.b (Lamiaceae) (KU/SG/V Tulsi (Holy basil) Leaf paste made with cow’s milk or water taken 061) orally, two to three times a day. Butea frondosa Koenig ex Roxb.a Muttuga (Bastard teak) Gum of tree with cow’s milk taken orally, for 2–3 (Papilionaceae) (KU/SG/V 062) days. Areca catechu L.b (Arecaceae) (KU/SG/MH Adikae (Arecanut) Nut paste mixed with three to four drops of honey 063) taken orally, for 4 days. Daemia extensa R. Br.a (Asclepiadaceae) Yugaphala (Uttara varuni) Leaf decoction taken orally, for 5 days. (KU/BS/TH 045) Herpes simplex Tamarindus indica L.b (Leguminosae) Hunasae ( tree) Fruit paste with oil and Argemone mexicana (KU/SG/MH 064) L.a (Prickly poppy) (KU/BS/KH 043) leaves ground and applied locally throughout the affected part. Holoptelea integrifolia Planch,a [Vu] Tapasi (Indian elm) Bark paste applied over the affected part until it (Ulmaceae) (KU/BS/GG 032) disappears. Patient advised not to use cow’s milk curd and sour foodstuffs. Santalum album L.c [EW] (Santalaceae) Srigandha (White sandal Wood paste with lemon juice applied on the affected (KU/SG/V 065) wood tree) area. Itching Pongamia glabra Vent,a (Papilionaceae) Hongae (Indian ) Oil from seeds applied over the itching body parts, (KU/SG/V 066) for 5 days. Plumbago zeylanica L.a [EW] (Plumbaginaceae) Chitramula (White lead Root paste applied over the skin during the night, for (KU/BS/MA 023) wort) 3–5 days. Leprosy A. Juss.c (Meliaceae) Baevu (Neem) Gum exudate from the plant mixed with cow’s milk (KU/SG/MH 067) given orally, for five to six times a day. Xylia xylocarpa (Roxb.) Taub.a (Mimosaceae) Jambae (Scimsapa) Bark decoction with cow’s milk, taken for a week. (KU/SG/V 068) Malaria Holoptelea integrifolia Planch. (Ulmaceae) Tapasi (Indian elm) Bark cut in the shape of a coin tied on left arm (KU/BS/GG 032) below the shoulder. Jatropha curcas L. (Euphorbiaceae) (KU/SG/V Bettadaharalu (Angular Latex mixed with white jaggery and taken orally. 059) leaved physic nut) Clerodendron inerme (L.) Gaertn.a Vishamarae (Kundali) Leaf decoction of this plant and of Azadirachta indica (Verbenaceae) (KU/SG/NS 069) (Baevu) (KU/SG/MH 067) taken orally. Skin diseases, dhobis Leucas aspera Spreng.a (Lamiaceae) Tumbae (Chota halkusa) Leaf paste applied over the skin, for 8 days. itch and ringworm (KU/BS/MU 013) Momordica charantia L.b (Cucurbitaceae) Haagala (Bitter gourd) Leaf paste with lime applied over skin, for 5 days. (KU/SG/JS 136) Butea frondosa Koenig ex Roxb. (Leguminosae) Muttuga (Bastard tree) Flower paste applied over the infected parts. (KU/SG/V 062) Sore-wounds Plumbago zeylanica L. (Plumbaginaceae) Chitramula (White lead Bark juice applied using cotton and bandaged for (KU/BS/MA 023) wort) 2–3 days. [EW]: Endangered in wild; [Vu]: Vulnerable. a Wild. b Cultivated. c Wild as well as cultivated. 310 P. Mahishi et al. / Journal of Ethnopharmacology 98 (2005) 307–312

Table 2 Plant species used to treat non-infectious diseases Disease Botanical name and family Local (common) name Part and method of use Asthma Terminalia bellerica Roxb.a [Vu] Tara (Beleric Macerated fruit taken orally with honey, for 2–3 (Combretaceae) (KU/SG/V 070) myrobalans) days. Tylophora asthamatica Wight et Arn.a Pitta mari (Indian Leaf decoction mixed with two to three drops of (Verbenaceae) (KU/BS/KH 042) ipecacuanha) honey taken orally, for 5 days in the morning. Withania somnifera Dunal.a [EW] (Solanaceae) Ashwagandha (Winter Root decoction is taken orally with garlic and cow’s (KU/SG/NS 071) cherry) milk, two to four times a day. Ficus religiosa L.a (Moraceae) (KU/SG/V 072) Arali (Sacred fig) Dried fruits pulverized and taken with water. Boils, bums and sores Diospyros montana Roxb.a [LR] (Ebenaceae) Jagalaganti (Mountain Fruit made into a powder and applied on burnt parts. (KU/SG/V 073) persimmon) Bombax malabaricum DC.a (Bombacaceae) Booruga (Silk cotton tree) Flowers macerated and applied on boils and sores. (KU/SG/V 074) Breast cancer Plumbago zeylanica L. (Plumbaginaceae) Chitramula (White lead Roots ground with lime juice and applied over the (KU/BS/MA 023) wort) part with symptom, three to five times a day. Cataract Breynia rhamnoides Muell.- Arg.a Hullkadi (Tikhar) Root exudate collected in the morning and dropped (Euphorbiaceae) (KU/BS/LV 035) carefully into eyes, two to three times a day. Medicine men need to take care not to touch the exudate. Dog bite Acalypa indica L.a (Euphorbiaceae) Kuppi (Indian acalypha) Leaf paste with a little lime applied to bitten area (KU/SG/MH 075) two times a day, for 3–4 days. Ricinus communis L.b (Euphorbiaceae) Haralu (Castor) Leaf paste applied over bitten area for 5 days and a (KU/SG/NS 076) small quantity of paste taken orally with food. Azadirachta indica A. Juss. (Meliaceae) Baevu (Neem) Leaf paste of this and of Butea frondosa Koenig ex (KU/SG/MH 067) Roxb. (Muttuga KU/SG/V 062) applied over the bit- ten area, for 3–4 days. Advised not to use cow’s milk or curd. Fever Ruta graveolens L.b (Rutaceae) (KU/BS/MA Naagadali (Garden rue) Leaf paste with honey or cow’s milk taken orally, 018) two times a day for 5 days. Adhatoda vasica Nees. c (Acanthaceae) Adusoge (Malabar nut) Leaf paste mixed with powder (Piper (KU/SG/NS 077) nigrum L) (KU/BS/SM 053) made into pills taken orally, two to three times daily. Hair fall (severe) Vitex trifolia L.a (Verbenaceae) (KU/SG/MH Nira lakki (Indian wild Leaf paste blended with coconut oil and applied to 078) pepper) hair and scalp. Infertility Careya arborea Roxb.a (Lecythidaceae) Kavalu (Kumbi) Flower paste prepared with fruits of Emblica offici- (KU/BS/LV 036) nalis Gaertn. J (Indian goose ) (KU/SG/JS 223), Terminalia chebula Retz.c (Chebulic myrobalan) (KU/SG/JS 218) and macerated with ghee, taken orally in empty stomach for 4 days. Advised not to use salt. Jaundice Tylophora asthmatica Wight et Arn. Pitta maari (Indian Roots with black pepper (Piper nigrum L.) (Verbenaceae) (KU/BS/KH 042) ipecacuanha) (KU/BS/SM 053), garlic and fruits of Syzygium cumini Lam. (KU/SG/V 060) made into paste, taken orally for 2–3 days. Phyllanthus amarus Schum. et Thonn. Nela nelli Leaf paste with taken internally, two tea (Euphorbiaceae) (KU/SG/V 057) (Bhumyamalaki) spoons daily. Plumbago zeylanica L. (Plumbaginaceae) Chitramula (White lead Roots, gum of Acacia concinna DC.a (Saptala) (KU/BS/MA 023) wort) (KU/SG/JS 235), cardamom seeds, dates, coconut fruit pulp and sugar mixed well and taken orally for 9 days. Root paste prepared with lime juice applied onto the body for 3–5 days. Migraine and headache Euphorbia tirucalli L.a (Euphorbiaceae) Sanna kalli (Milk hedge) Milky latex applied on the forehead and leaves of (KU/SG/MH 079) Moringa oleifera Lam.b (Drum stick) (KU/BS/MU 012) stuck on it for 6–8 days during morning hours before sunrise. Asteracantha longifolia Neesa (Acanthaceae) Kulyanka (Kolistha) Root juice is dropped into the ear, which is on the (KU/SG/MH 080) opposite side of the headache. Eucalyptus globulus Labill.b (Myrtaceae) Neelagiri (Blue gum tree) Oil from leaves applied over forehead, for 5 days. (KU/SG/V 081) P. Mahishi et al. / Journal of Ethnopharmacology 98 (2005) 307–312 311

Table 2 (Continued)

Paralysis Mentha arvensis L.b (Lamiaceae) (KU/SG/NS 082) Pudina () Plant parts of M. arvensis and seeds of Trachyspermum ammi (L.) Sprague.b (Carum) in equal proportions along with rock salt taken with coffee, three to four times a day. Piles Careya arborea Roxb. (Lecythidaceae) (KU/BS/LV Kavalu (Kumbi) Bark powder with honey taken orally, early in the 036) morning. Snakebite Acalypa indica L. (Euphorbiaceae) (KU/SG/MH 075) Kuppi (Indian acalypha) Leaf paste applied over the bitten part for relief from poisoning. Rauvolfia serpentina Benth. ex Kurza [EW] Sarpagandha (Serpentine) Root paste taken orally. For quick relief, root paste (Apocynaceae) (KU/BS/HB 025) with cow’s milk curd in a copper vessel taken orally. Elaeodendron glaucum Jacq.f.a (Celastraceae) Mukarthi (Bhutphal) Roots and of plant made into paste taken orally with (KU/SG/NS 083) cow’s milk. Tylophora asthmatica Wight et Arn. (Asclepiadaceae) Pitta maari (Indian Leaf juice taken orally (KU/BS/KH 042) ipecacuanha) Canthium parviflorum Lam.a (Rubiaceae) (KU/SG/NS Kaarae (Kirni) Root powder with cow’s milk curd or cow’s milk 084) taken orally. Calotropis procera R. Br.a (Asclepiadaceae) Ekka (Safed Ak) Root bark paste given orally to patient to induce vom- (KU/BS/GG 031) iting and latex applied over the bitten area. Scorpion sting Careya arborea Roxb. (Lecythidaceae) (KU/BS/LV Kavalu (Kumbi) Fresh bark paste applied over affected part and infu- 036) sion of the fruit taken orally for quick relief. Stomach ache Azadirachta indica A. Juss. (Meliaceae) (KU/SG/MH Baevu (Neem) Leaves ground with castor oil and sugar, taken 067) orally for 3 days. Hemidesmus indicus R. Br.a [LR] (Asclepiadaceae) Sogadae beru (Indian Whole plant paste with cold unboiled cow’s milk, (KU/BS/MG 001) sarasaparilla) taken orally for 2–3 days. Mouth ulcer Aegle marmelos Correa ex Roxb.c [Vu] (Rutaceae) Bilva patre (Bael tree) Gum of the plant applied over the affected area for (KU/SG/V 085) 3–4 days. Wrightia tinctoria R. Br.a (Apocynaceae) (KU/SG/V Bepalle (Sweet indrajo) Fruit homogenised and applied over the affected 086) part for 5 days. [EW]: Endangered in wild, [Vu]: Vulnerable, [LR]: Lower risk. a Wild. b Cultivated. c Wild as well as cultivated. suggested by the local herbalists. They also point out that understanding of the pharmacological effect of herbal drugs allopathic medicines, which are available in the nearby is necessary for effective therapy of diseases. towns, are expensive and have side effects in comparison Several cases of indirect evidence on the pharmacological to the herbal medicine. This might indicate the reason for effects of certain medicinal herbs (Aristolochia trilobata, the dependence of local residents on herbal medicine rather Artemisia absinthium, Centella asiatica, Leucas aspera and than allopathic medicine. Plumbago zeylanica) have been documented in the literature The findings of the present study are in conformity with (Samy and Ignacimuthu, 2000; Quinlan et al., 2002; Somchit study published by Nadakarni (1976) in the treatment of et al., 2003 and Nessa et al., 2004). Rao (2000) provides certain diseases with specific medicinal plants. For example, a list of herbs that have various medicinal properties in his plant species recommended for the treatment of asthma, book on database of medicinal plants. Certain herbal drugs cholera, cough, dysentery, jaundice, leprosy and snakebite listed in the and other traditional medicine systems are essentially the same species, although the plant parts are not only time-tested but have also been screened for their differed. However, there are certain examples of other plant pharmacological properties (Dev, 1997, 1999). For example, species, which are used exclusively for the treatment of guggul, the gum resin from Commiphora wightti Engl. (Burs- specific diseases in the study area and represent the first eraceae) has been used to treat rheumatoid arthritis and lipid report of such uses. For example, Breynia rhamnoides is disorders in addition to other diseases. It has been shown that used to treat cataract, leaf paste of Ricinus communis to treat two antihyperlipoproteinemic compounds, Z-guggulsterone dog bite, flower of Careya arborea to treat infertility, and the and E-guggulsterone extracted from the gum resin, have bark of Xylia xylocarpa for leprosy. It is interesting to note hypolipidemic activity similar to that of the synthetic drug, that decoction of various parts of Holoptelea integrifolia, clofibrate (Dev, 1999). The presence of resveratrol and Jatropha curcas, Clerodendron inerme and Azadirachta pterostilbene in darakchasava, an ayurvedic medicine whose indica are used to treat malaria. principal component being grapes, has been shown to reduce The present investigation points out that often, more than cardiac disease rate and carcinogenesis (Paul et al., 1999). one plant belonging to different taxonomic groups are being Certain Chinese traditional herbal drugs used to retard used to treat a specific disease or disorder. A proper scientific ageing and to treat several other diseases have been found to 312 P. Mahishi et al. / Journal of Ethnopharmacology 98 (2005) 307–312 contain considerably high amounts of melatonin, which is Chen, G., Huo, Y., Tan, D., Liang, Z., Zhang, W., Zhang, Y., 2003. related to the scavenging of free radicals (Chen et al., 2003). Melatonin in Chinese medicinal herbs. Life Sciences 73, 19–26. The present results also suggest that different groups of Dev, S., 1997. Ethnotherapeutics and modern drug development: the po- tential of Ayurveda. Current Science 73, 909–928. people practice herbal medicine differently, and their formu- Dev, S., 1999. Ancient-modern concordance in ayurvedic plants: some lations might not resemble that of herbal medicine of people examples. Environmental Health Perspectives 107, 783–789. residing either in the far-off places or nearby places. For ex- Gamble, J.S., 1995. Flora of the Presidency of Madras, vols. 1–3. Bishen ample, ‘Siddis’ and ‘Gowlis’ of Uttar Kannada in Kamataka Singh Mahendra Pal Singh Publications, Dehradun, India, 2017 pp. used entirely different types of plant species for the treat- Gowda, B., Nissar, M.M., Seetharam, Y.N., Ramesh, S.R., 1997. Threat- ened plants of peninsular India. My Forest 33, 327–334. ment of a variety of human diseases. Bhandary et al. (1995) Harsha, V.H., Hebbar, S.S., Hegde, G.R., Shripathi, V., 2002. Ethnomed- reported that ‘Siddis’ used bark of Careya arborea to treat ical knowledge of plants used by Kunabi Tribe of Karnataka in India. dysentery and ear pain. On the other hand, local commu- Fitoterapia 73, 281–287. nities of the study area used flowers of the above plant for Harsha, V.H., Hebbar, S.S., Shripathi, V., Hegde, G.R., 2003. Ethnomedi- treating infertility, piles and scorpion sting. ‘Gowlis’ of Uttar cobotany of Uttara Kannada district in Karnataka, India—plants in treatment of skin diseases. Journal of Ethnopharmacology 84, 37– Kannada used Rauvolfia serpentina to treat herpes infection 40. (Bhandary et al., 1996), while people of the study area used Nadakarni, A.K., 1976. Indian Materia Medica, vol. I. Popular Prakashan, this plant for treating snakebite. Bombay, 1319 pp. Finally, this study also points out that certain species of Nayar, M.P., Sastri, A.R.K., 1990. Red Data Plants of India. CSIR Pub- medicinal plants that are endangered are being exploited by lication, New Delhi, 271 pp. Nessa, F., Mohamed, N., Ismail, Z., 2004. Antibacterial activity of leaf the local residents who are unaware of their importance in the extracts of Blumea balsamifera. Journal of Tropical Medicinal Plants ecosystem. In view of this, there is a great necessity to educate 5, 1–3. the local population and healers to adopt conservation mea- Parinitha, M., Harish, G.U., Vivek, N.C., Mahesh, T., Shivanna, M.B., sures as necessary, so that over-collection of such species will 2004. Ethnobotanical wealth of Bhadra Wild Life Sanctuary in Kar- not lead to their extinction in their territory, which signifies nataka. Indian Journal of Traditional Knowledge 3, 37–50. Paul, B., Masih, I., Deopujari, J., Charpentier, C., 1999. Occurrence of the loss of their source medicinal material. resveratrol and pterostilbene in age-old darakchasava, an ayurvedic medicine from India. Journal of Ethnopharmacology 68, 71–76. Quinlan, M.B., Quinlan, R.J., Nolan, J.M., 2002. Ethnophysiology and Acknowledgements herbal treatments of intestinal worms in Dominica, West Indies. Jour- nal of Ethnopharmacology 83, 75–83. Rao, C.K., 2000. Data Base of Medicinal Plants. Karnataka State Council The authors express thanks to the herbal doctors in the for Science and Technology, Karnataka Government, 458 pp. study area for revealing their traditional medico-botanical Ravikumar, K., Ved, D.K., 2000. Illustrated Field Guide of 100 Red- knowledge and for their permission to communicate their listed Medicinal Plants of Conservation Concern in Southern India. knowledge to a wider audience for the benefit of every one. Foundation for Revitalisation of Local Health Traditions (FRLHT) Thanks are also expressed to Prof. M. Krishnappa, Depart- Publication, Bangalore, 463 pp. Saldanha, C.J., Nicolson, D.H., 1976. Flora of Hassan District, Karnataka, ment of Post Graduate Studies and Research in Applied vol. I. Oxford and IBH Publication, New Delhi, 923 pp. Botany, Kuvempu University for the co-operation extended Samy, R.P., Ignacimuthu, S., 2000. Antibacterial activity of some folklore during the study. medicinal plants used by tribals in Western Ghats of India. Journal of Ethnopharmacology 69, 63–71. Seetharam, Y.N., Haleshi, C., Vijay, 1998. Medicinal plants of North Eastern Karnataka and their status. My Forest 33, 767–772. References Sinha, R.K., 1996. Ethnobotany—The Renaissance of Traditional Herbal Medicine. Ina Shree Publishers, Jaipur, India, 242 pp. Bhandary, M.J., Chandrashekar, K.R., Kaveriappa, K.M., 1995. Medical Somchit, N., Reezal, I., Elyshanur, I., Mutalib, A.R., 2003. In vitro antimi- ethnobotany of Siddis of Uttara Kannada District, Karnataka, India. crobial activity of ethanol and water extract of Cassia alata. Journal Journal of Ethnopharmacology 47, 149–158. of Ethnopharmacology 84, 1–4. Bhandary, M.J., Chandrashekar, K.R., Kaveriappa, K.M., 1996. Ethnob- Yoganarasimhan, S.N., Subramanyam, K., Razi, B.A., 1981. Flora of otany of Gowlis of Uttara Kannada District, Karnataka. Journal of Chikmagalur District, Karnataka, India. Book Distributors, Dehradun, Economic and Taxonomical Botany, Additional Series 12, 244–249. 407 pp. Journal of Ethnopharmacology 98 (2005) 313–317

The ameliorative effect of dates (Phoenix dactylifera L.) on ethanol-induced gastric ulcer in rats A.A. Al-Qarawi, H. Abdel-Rahman, B.H. Ali ∗, H.M. Mousa, S.A. El-Mougy

Department of Veterinary Medicine, Faculty of Agriculture and Veterinary Medicine, King Saud University, Al-Gaseem Branch, P.O. 10158, Buraydah, Al-Gaseem 81999, Saudi Arabia Received 18 November 2004; received in revised form 17 January 2005; accepted 17 January 2005

Abstract

The present work aimed at testing, in a rat model of ethanol-induced gastric ulceration, a local folk medicinal claim that dates are beneficial in gastric ulcers in humans. Aqueous and ethanolic undialyzed and dialyzed extracts from date fruit and pits were given orally to rats at a dose of 4 ml/kg for 14 consecutive days. On the last day of treatment, rats were fasted for 24 h, and were then given ethanol, 80% (1 ml/rat) by gastric intubation to induce gastric ulcer. Rats were killed after 1 h of ethanol exposure, and the incidence and severity of the ulceration were estimated, as well as the concentrations of gastrin in plasma, and histamine and mucus in the gastric mucosa. A single group of rats that were fasted for 24 h, was administered orally with lansoprazole (30 mg/kg), and was given 80% ethanol as above, 8 h thereafter, served as a positive control. The results indicated that the aqueous and ethanolic extracts of the date fruit and, to a lesser extent, date pits, were effective in ameliorating the severity of gastric ulceration and mitigating the ethanol-induced increase in histamine and gastrin concentrations, and the decrease in mucin gastric levels. The ethanolic undialyzed extract was more effective than the rest of the other extracts used. It is postulated that the basis of the gastroprotective action of date extracts may be multi-factorial, and may include an anti-oxidant action. © 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Dates; Ethanol-induced ulcer; Gastric histamine; Gastrin; Lansoprazole; Mucin; Phoenix dactylifera L.; Palmae

1. Introduction The pollen grains of date palm have been used in Egyptian local practices to improve fertility in women, and in some Date palms (Phoenix dactylifera L., Palmae) have been locations in Arabia date pits are roasted and used in lieu of cultivated in the Middle East over at least 6000 years ago coffee as a hot beverage. (Copley et al., 2001). For the natives in this region, dates are Relatively few pharmacological studies have been considered a staple carbohydrate food (Al-Shahib and Mar- conducted on dates. For example, it has been shown that, shall, 2003). Date fruits are also used in the production of depending on the type of extract used, date fruit and pit local beverages and spirits. In local medicinal practices dates extracts significantly increase or decrease gastrointestinal are considered a “tonic” and “aphrodisiac”, and in some com- transit (GIT) in mice (Al-Qarawi et al., 2003), and that munities they are thought to be useful against ulcer (Rasheed, date fruit extract has strong antioxidant and antimutagenic personal communication). In fact, Muslims believe that “He properties (Vayalil, 2002). Date palm kernels have been who eats seven dates every morning will not be affected by shown to exhibit antiaging properties and significant reduc- poison or magic on the day he eats them” (cited by Miller et tion in skin wrinkles in women (Bauza, 2002), and natural al., 2003). fats from date palm has been reported to prevent irritant contact dermatitis (Schliemann-Willers et al., 2002). In ∗ Corresponding author. Tel.: +966 6 3813372; fax: +966 6 3813372. animals, the pits have been included in the diet of chickens, E-mail address: [email protected] (B.H. Ali). sheep, fish and rats, and have been shown to enhance

0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2005.01.023 314 A.A. Al-Qarawi et al. / Journal of Ethnopharmacology 98 (2005) 313–317 growth in these species (see Ali et al., 1999 and references and the residue dissolved in distilled water to the appropriate therein). doses just prior to use. In view of the wide consumption of dates in our region, Dialysis to remove sugars was performed using cellu- the fact that dates are anecdotally reputed to be useful against lose tubing (Spectra/Por, width 32 mm, diameter 20.4 mm, peptic ulcers, and the fact that Muslims customarily consume volume/length 303 ml, from Spectrum Medical Instruments, more of the dates during the fasting month of Ramadan, pos- Inc., USA). sibly to protect the gastric mucosa from the damaging effect of gastric acid, and because of the scarcity of information 2.4. Experimental design on the pharmacological properties of date fruits and pits, we considered undertaking this study to assess the influence of Rats were randomly assigned to the following experimen- date extracts on the incidence and severity of ethanol-induced tal groups: gastric ulceration. In addition, the effect of date extracts on the gastric concentrations of histamine and mucin, and the Group 1: given distilled water (4 ml/kg) orally for 14 con- plasma concentration of the hormone gastrin has also been secutive days. On the last day rats were given normal saline investigated. (1 ml) 1 h before killing. Group 2: given distilled water (4 ml/kg (orally for 14 con- secutive days. On the last day rats were given 80% ethanol 2. Materials and methods (1 ml) 1 h before killing. Group 3: Rats were given the aqueous date fruit extract 2.1. Animals (4 ml/kg) for 14 consecutive days. On the last day rats were given 80% ethanol (1 ml) 1 h before killing. Fifty-four adult male Wistar rats weighing between 200 Group 4: Rats were given the dialyzed aqueous date fruit and 250 g were used in this work. They were obtained for extract (4 ml/kg) for 14 consecutive days. On the last day the Animal House of King Saud University in Riyadh, and rats were given 80% ethanol (1 ml) 1 h before killing. were divided into eight equal groups. The animals were kept Group 5: Rats were given the ethanolic date fruit extract at a controlled temperature of 23 ± 2 ◦C, relative humidity of (4 ml/kg) for 14 consecutive days. On the last day rats were 65–80% and a light regime of 12 h light:12 h dark (lights on given 80% ethanol (1 ml) 1 h before killing. at 6:00). Except otherwise mentioned, pelleted Purina chow Group 6: Rats were given the aqueous date pit extract and water were provided to the rats ad libitum. (4 ml/kg) for 14 consecutive days. On the last day rats were given 80% ethanol (1 ml) 1 h before killing. 2.2. Plant material Group 7: Rats were given the dialyzed date pit extract (4 ml/kg) for 14 consecutive days. On the last day rats were Fresh fruits of Sukari dates (P. dactylifera L.) were ob- given 80% ethanol (1 ml) 1 h before killing. tained from a local date manufacturing factory. Samples of Group 8: Rats were given the ethanolic date pit extract these dates were kept frozen for future reference. (4 ml/kg) for 14 consecutive days. Group 9: Rats were given a single oral dose of lansoprazole 2.3. Plant preparation and administration (30 mg/kg), and 8 h later was given 80% ethanol as above, 1 h before killing. The date fruits were manually separated from the pits and the latter were washed clear of any fruit, dried at room 2.5. Ethanol-induced gastric lesion temperature and ground into powder using a stainless-steel blender. The water extract of the date fruit was made by Rats were deprived from food (but not water) on day adding distilled water to coarsely pounded date fruit (3:1), 14 of the experiment. On the last day of experiment (day and leaving for 48 h in a refrigerator (4 ◦C) with continu- 15) rats were given 80% ethanol (1 ml) by gastric intuba- ous stirring. The aqueous extract was then used daily for tion 1 h before killing, except for rats in group 1 which 14 consecutive days. To remove sugars from the extract, was given normal saline (1 ml/rat), and group 9 (positive the aqueous extract was dialyzed. Dialysis was carried out control) which consisted of six rats that were fasted for under running tap water for 24 h. The dialyzed water ex- 24 h, administered orally with lansoprazole (Sigma, MO, tract was kept refrigerated and used daily for 14 consecutive USA) (30 mg/kg), and was given 80% ethanol as above, 8 h days. thereafter. A soxhlet apparatus was used to obtain an ethanol extract. The animals were anesthetized with ether, and rapidly de- The ethanol was then evaporated and the residue diluted with capitated 1 h after ethanol treatment. Blood was collected water to give the required concentration. The pounded date in heparinized tubes and centrifuged at 900 × g for 15 min fruit or pits were added (1:3) to either ethanol or distilled wa- at 5 ◦C. The plasma obtained was stored at −20 ◦C pend- ter. Extraction was carried out at 4 ◦C with continuous stir- ing gastrin assay. The stomach of each animal was excised ring. The ethanol extracts were then concentrated to dryness and opened along the greater curvature. After washing with A.A. Al-Qarawi et al. / Journal of Ethnopharmacology 98 (2005) 313–317 315 normal saline and removal of blood, the gastric lesion was 2.10. Statistical analysis quantified using a stereo-microscope. Gross mucosal dam- age was assessed in a “blinded manner by calculation of Values reported are expressed as mean ± S.E.M. The sig- a lesion index based on the number and severity factor of nificance of differences between the control and ethanol lesions as described previously (Agrawal et al., 2000). The and/or date-treated groups was tested by least squares anal- stomachs were inflated with 10 ml of 1% formalin for 10 min ysis of variance using the general linear models (GLM) pro- to fix the inner walls. The average number of ulcers per cedures of the statistical analysis system (SAS, 1996). stomach was recorded. The lesion index was calculated as the total number of lesions added to their respective severity factor. 3. Results

The results of this work are shown in Tables 1–3. 2.6. Measurement of the mucin content in the gastric Table 1 shows the effect of treatment with various date ex- wall tracts on the number, severity and ulcer index. The ethanolic and, to a lesser extent, the aqueous date fruit extract amelio- Gastric mucus was quantitatively measured as described rated the severity of gastric ulceration. The effect of the date by Corne et al. (1974). The stomachs were removed and were pit extract was less evident. The positive control lansoprazole soaked in 0.1% Alcian blue solution for 2 h. The uncomplexed significantly reduced the lesion index. dye was removed by two successive washes at 15 and 45 min Table 2 summarizes the histological changes seen af- in 0.25 M aqueous sucrose solution. Dye complexes with gas- ter treatment with ethanol and date extracts. Ethanol treat- tric wall mucous were extracted by immersion in 10 ml of ment induced severe necrosis, haemorrhage, congestion and 0.5 M MgCl for 2 h. The resulting blue solution was shaken 2 oedema in stomach sections. These actions were markedly with equal volumes of diethyl ether and the optical density of ameliorated by lansoprazole, and to a lesser extent by pre- the aqueous phase was measured at 605 mm by a UV–visible treatment with date fruits and pits. The ethanolic undialyzed spectrophotometer (B&L 2000). The quantity of mucin was date fruit extract was more active in this regard than the other expressed as ␮g of Alcian blue extracted per weight (g) of extracts tested. stomach. The concentrations of gastrin in the plasma, and histamine and mucin in the gastric mucosae in control and treated rats 2.7. Gastrin measurement with date extracts and lansoprazole are shown in Table 3. Ethanol treatment induced a significant increase in the The gastrin hormone was assayed in the collected plasma concentrations of gastrin and histamine, and significantly samples using a radioimmunoassay (RIA) technique accord- decreased that of mucin (P < 0.05). These effects were ing to the procedure described by the manufacturer (IBL- significantly antagonized by the pre-treatment of rats with Hamburg, Germany, cat. No: MI 131 01). lansoprazole and by date fruit aqueous and ethanolic extracts (P < 0.05). Date pits extracts were not significantly effective 2.8. Histamine estimation in antagonizing these effects.

Histamine in the gastric mucosa was separated by thin layer chromatography according to the modified method of 4. Discussion Shalaby (1994). Histamine dihydrochloride (Sigma, MO, USA) was used as standard for identification of the spots in Rat gastric mucosal damage induced by high concentra- the plates and for spectrophotometeric determination (Spec- tions of ethanol has widely been used to investigate gastro- tronic 2000, Bausch and Lomb at 570 nm) of samples after protective effect of medicinal plants (e.g. Zhu et al., 1997). elution from the plates. Values were expressed as mg/g of The present results suggest that pretreatment with date fruit stomach weight. (and to lesser extent pit) ethanolic and aqueous extracts for 14 days markedly ameliorated the ulcer index, and some his- tological and biochemical indices of ethanol-induced gastric 2.9. Histopathological examination ulceration in rats. This lends support to the local folk medic- inal claim that dates may be useful to humans with ulcers. Gastric tissue samples were fixed in 10% neutral formalin In this work we selected lansoprazole as a reference anti- and processed for routine paraffin blocking and H&E stain- ulcer drug (rather than a histamine antagonist) because it ing. These were “blindly” examined under the microscope has been shown that prostaglandins provide a much better for histopathological change such as congestion, oedema, anti-ulcer effect on ethanol-induced gastric damage (Cho and erosions, ulcerations and necrosis. The severity of histopatho- Ogle, 1992). Treatment with ethanol-induced the expected logical changes was expressed according to an arbitrary scale actions in the stomach of rats, and those included severe (between − to++++). histological damages (necrosis, haemorrhages), and a signi- 316 A.A. Al-Qarawi et al. / Journal of Ethnopharmacology 98 (2005) 313–317

Table 1 Effect of date fruit and pits extracts on ethanol-induced gastric ulcers in rats Treatment Number of ulcers/stomach (a) Severity per stomach (b) Lesion index (a + b) Group 1: distilled water + saline 0 0 No ulcers seen Group 2: distilled water + ethanol 8.1 ± 0.9a 8.0 ± 0.9a 16.1 ± 1.8a Group 3: date fruit aqueous extract + ethanol 4.2 ± 0.5b 5.4 ± 0.6b 9.6 ± 1.1b Group 4: dialyzed date fruit aqueous extract + ethanol 6.1 ± 0.6c 6.1 ± 0.5c 12.2 ± 1.1c Group 5: date fruit ethanolic extract + ethanol 3.1 ± 0.3d 4.2 ± 0.4b 7.3 ± 0.7d Group 6: date pits aqueous extract + ethanol 7.0 ± 0.6ac 6.1 ± 0.6b 13.1 ± 1.2c Group 7: dialyzed date pits aqueous extract + ethanol 6.6 ± 0.6ac 5.4 ± 0.5b 12.0 ± 1.1c Group 8: date pits ethanolic extract + ethanol 6.0 ± 0.5ac 4.9 ± 0.5b 10.9 ± 1.0b Group 9: lansoprazole + ethanol 1.6 ± 0.3d 0.9 ± 0.1d 2.5 ± 0.4d Values (means ± S.E.M. from six rats), with different superscripts are significantly different (P < 0.05) from the values in the same column. Distilled water (4 ml/kg) or various date extracts were given daily at an oral dose of 4 ml/kg orally for 14 consecutive days; 24 h after the last day of treatment, rats weregiven 80% ethanol (1 ml) per os 1 h before killing. A single group of rats that were fasted for 24 h, was administered orally with lansoprazole (30 mg/kg), and was given 80% ethanol as above, 8 h thereafter, served as a positive control.

Table 2 Histopathological evaluations of the effects of date aqueous and ethanolic extracts on the induction of gastric lesions in the rats Treatment Necrosis Haemorrhage Congestion Oedema Group 1: distilled water + saline −− − − Group 2: distilled water + ethanol +++ ++++ ++++ ++++ Group 3: date fruit aqueous extract + ethanol − +++ +++ +++ Group 4: dialyzed date fruit aqueous extract + ethanol −− ++ Group 5: date fruit ethanolic extract + ethanol − +++++ Group 6: date pits aqueous extract + ethanol + ++ ++ ++ Group 7: dialyzed Date pits aqueous extract + ethanol + ++ ++ + Group 8: date pits ethanolic extract + ethanol + + + + Group 9: lansoprazole + ethanol −− ++ −, normal; +, little effect; ++, appreciable effect; +++, severe effect; ++++, intensively severe effect, distilled water (4 ml/kg) or various date extracts were given daily at an oral dose of 4 ml/kg orally for 14 consecutive days; 24 h after the last day of treatment, rats (six in each group) were given 80% ethanol (1 ml) per os 1 h before killing. A single group of rats that were fasted for 24 h, was administered orally with lansoprazole (30 mg/kg), and was given 80% ethanolas above, 8 h thereafter, served as a positive control.

Table 3 Effect of treatment of rats with date fruit and pits extracts on plasma gastrin, gastric histamine and gastric juice mucin activity Treatments Gastrin (pg/ml plasma) Histamine mg/g stomach weight Mucin ␮g/g stomach weight Group 1: distilled water + saline 15.2 ± 1.0a 22.0 ± 1.5a 2.4 ± 0.1a Group 2: distilled water + ethanol 125.2 ± 2.6b 298.3 ± 18.2b 0.4 ± 0.3b Group 3: date fruit aqueous extract + ethanol 107.7 ± 12.3c 217.3 ± 19.6c 1.9 ± 0.2c Group 4: dialyzed date fruit aqueous extract + ethanol 118.5 ± 13.2c 230.2 ± 24.7c 2.0 ± 0.2c Group 5: date fruit ethanolic extract + ethanol 98.5 ± 10.1c 211.2 ± 3. 2c 2.1 ± 0.2c Group 6: date pits aqueous extract + ethanol 116.5 ± 12.3c 228.4 ± 19.8c 1.6 ± 0.2c Group 7: dialyzed Date pits aqueous extract + ethanol 113.8 ± 13.1c 225.8 ± 17.4c 1.4 ± 0.2c Group 8: date pits ethanolic extract + ethanol 107.7 ± 11.9c 205.5 ± 20.2c 1.2 ± 0.2c Group 9: lansoprazole + ethanol 34.3 ± 4.3d 205.5 ± 20.2d 1.2 ± 0.2d Values (means ± S.E.M. from six rats), with different superscripts are significantly different (P < 0.05) from the values in the same column. Distilled water (4 ml/kg) or various date extracts were given daily at an oral dose of 4 ml/kg orally for 14 consecutive days; 24 h after the last day of treatment, rats weregiven 80% ethanol (1 ml) 1 h before killing. A single group of rats that were fasted for 24 h, was administered orally with lansoprazole (30 mg/kg), and was given 80% ethanol as above, 8 h thereafter, served as a positive control.

ficant increase in plasma concentrations of the gastric events leading to the production of arteriolar vasodilatation hormone gastrin, reduction in mucin and an increase in the in injured tissues (Black, 1993). Gastric mucus (mucin) is an histamine concentrations in the gastric mucosa. These pa- important protective factor for the gastric mucosa and con- rameters were selected for study because of their relevance sists of a viscous, elastic, adherent and transparent gel formed to the pathogenesis of gastric ulceration. For example, gastrin by 95% water and 5% glycoproteins that cover the entire gas- is a gastrointestinal hormone that, among other various func- trointestinal mucosa. Moreover, mucus is capable of acting tions, regulates gastric acid secretion, releases histamine, and as an antioxidant, and thus can reduce mucosal damage me- regulates gastric endocrine cell proliferation (Walsh, 1993). diated by oxygen free radicals (Repetto and Llesuy, 2002). Stimulation of the oxyntic cells by histamine is the final com- In this work we used dialyzed and non-dialyzed extracts. mon pathway by which neural and endocrine mechanisms act Dialysis was used to remove the sugars from the extracts. Pre- in inducing acid secretion. Histamine is involved in a cycle of viously dialyzed and non-dialyzed extracts caused opposing A.A. Al-Qarawi et al. / Journal of Ethnopharmacology 98 (2005) 313–317 317 effects on GIT transit (Al-Qarawi et al., 2003). In the present Ali, B.H., Bashir, A.K., Alhadrami, G., 1999. Reproductive hormonal work, non-dialyzed extracts seemed to be more active as gas- status of rats treated with date pits. Food Chemistry 66, 437–441. troprotectants than dialyzed extracts. Al-Qarawi, A.A., Ali, B.H., Mougy, S., Mousa, H.M., 2003. Gastroin- testinal transit in mice treated with various extracts of date (Phoenix Ethanol-induced gastric ulceration is known to be related dactylifera L.). Food and Chemical Toxicology 41, 37–39. to an anti-oxidant action, increased lipid peroxidation and Al-Shahib, W., Marshall, R.J., 2003. The fruit of the date palm: its pos- generation of free-radicals (Terano et al., 1989). Recently sible use as the best food for the future. International Journal of Food Vayalil (2002) discovered that date extracts possess signif- Science and Nutrition 54, 247–259. icant antioxidant action in vitro. This may, at least partially, Bandyopadhyan, D., Biswas, K., Bhattacharyya, M., Reiter, R.J., Baner- jee, R.K., 2001. Gastric toxicity and mucosal ulceration induced be one of the possible mechanisms by which date extracts by oxygen-derived reactive species: protection by melatonin. Current have ameliorated the ethanol-induced gastric ulceration. Molecular Medicine 1, 501–513. Recently we have found that dates contain relatively high Bauza, E., 2002. Date palm kernel extract exhibits antiaging properties concentrations of the anti-oxidants melatonin and vitamin E and significantly reduces skin wrinkles. International Journal of Tissue (Al-Qarawi et al., unpublished data). It has been reported that Reactions 24, 131–136. Black, J., 1993. Reflections on the analytical pharmacology of histamine treatment with melatonin prevents gastric ulcerogenesis and H2-receptor antagonists. Gastroenterology 105, 963–968. decreases ulcer index (Bandyopadhyan et al., 2001; Bubenik, Bubenik, K.G.A., 2002. Gastrointestinal melatonin: localization, function 2002). Vitamin E in has also been shown to reduce and Clinical relevance. Digestive Disease Sciences 47, 2336–2348. ethanol-induced gastric ulcer (Jarrin et al., 1999). Taken Cho, C.H., Ogle, C.W., 1992. The pharmacological differences and simi- together, these results corroborate our present finding of larities between stress- and ethanol-induced gastric mucosal damage. Life Sciences 51, 1833–1842. an ameliorative action of dates on ethanol-induced gastric Copley, M.S., Rose, P.J., Clampham, A., Edwards, D.N., Horton, M.C., ulceration, possibly due to its relatively high content of Evershed, R.P., 2001. Detection of palm fruit lipids in archaeological antioxidant substances. pottery from Qasr Ibrim, Egyptian Nubia. Proceedings of the Royal In conclusion, the present work has suggested that date Society, London 268, 593–597. extracts can ameliorate ethanol-induced gastric ulcers, and Corne, S.J., Morrissey, S.M., Woods, R.J., 1974. A method for the quanti- tative estimation of gastric barrier mucus. Journal of Physiology 242, that, in view of the fact that ethanol induces ulceration by 116–117. an antioxidant action, and that dates contain relatively high Jarrin, K., Renuvathani, M., Nafeeza, M.I., Gapor, M.T., 1999. Compara- amounts of anti-oxidant substances, it is possible that the tive effect of palm vitamin E and ranitidine on the healing of ethanol- mechanism of the gastroprotective action is via an oxidant induced gastric lesions in rats. International Journal of Experimental action, however, other mechanisms cannot be excluded. Pathology 80, 259–263. Miller, C.J., Dunn, E.V., Hashim, I.B., 2003. The glycaemic index of Pending further pharmacological and toxicological studies dates and date/yoghurt mixed meals. Are dates “the candy that to delineate the mechanism(s) of action of dates as gastro- grows on trees”? European Journal of Clinical Nutrition 57, 427– protective agent, and their toxic effects, compounds from 430. dates may potentially be useful in ulcers. Repetto, M.G., Llesuy, S.F., 2002. Antioxidant properties of natural com- pounds used in popular medicine for gastric ulcers. Brazilian Journal of Medical and Biological Research 35, 523–534. Acknowledgements SAS, 1996. SAS User’s Guide. SAS institute, Cary, NY, USA. Schliemann-Willers, S., Wigger-Alberti, W., Kleesz, P., Grieshaber, R., Elsner, P., 2002. Natural vegetable fats in the prevention of irritant This work was supported by funds from the College Re- contact dermatitis. Contact Dermatitis 4, 6–12. search Committee. We thank the Dean of the College of Agri- Shalaby, A.R., 1994. Separation, identification and estimation of biogenic culture and Veterinary Medicine, King Saud University for amines in foods by thin layer chromatography. Food Chemistry 49, his interest in this project, and Al Gaseem Date Factory for 305–310. providing free samples of dates for this study. Terano, A., Hiraishi, H., Ota, S., Shiga, J., Sugimoto, T., 1989. Role of superoxide and hydroxyl radicals in rat gastric mucosal injury induced by ethanol. Gastroenterologia Japonica 24, 488–493. Vayalil, P.K., 2002. Antioxidant and antimutagenic properties of aque- References ous extract of date fruit. Journal of Agricultural Food Chemistry 50, 610–617. Agrawal, A.K., Rao, C.H., Sairam, K., Joshil, V.K., Goel, R.K., 2000. Ef- Walsh, J.H., 1993. In: Walsh, J.H. (Ed.), Gastrin. Raven Press, New York, fect of Piper longum L., Zingiber officinalis L. and Ferula species on pp. 273–318. gastric ulceration and secretion in rats. Indian Journal of Experimental Zhu, M., Lew, T.H., Luk, C.T., 1997. Gastric protective effect of Lentinus Biology 38, 994–998. edodes against ethanol-induced ulceration. Fitoterapia 68, 537–542. Journal of Ethnopharmacology 98 (2005) 319–322

Tanshinone inhibits intimal hyperplasia in the ligated carotid artery in mice

Jun-rong Du a,b,∗,1, Xin Li a,1, Rong Zhang a, Zhong-ming Qian b

a Department of Pharmacology, West China School of Pharmacy, Sichuan University, Chengdu 610041, China b ABCT, The Hong Kong Polytechnic University, Hong Kong Received 26 November 2004; received in revised form 17 January 2005; accepted 17 January 2005

Abstract

Vascular smooth muscle cell (VSMC) proliferation is considered to play a central role in the development of intimal hyperplasia with pathological artery healing. Danshen, the Salvia miltiorrhiza Bge., has long been regarded as an effective traditional Chinese medicine for cardiovascular diseases. In this paper, the effects of tanshinone (TA), the lipid-soluble pharmacological constituents of danshen, on the intima hyperplasia and proliferating state of VSMC were described in a mouse carotid artery injured by complete cessation of blood flow. This study showed that oral administration of TA could significantly decrease the intimal thickening of injured vessels and proliferating cell nuclear antigen (PCNA)-positive VSMC in intimal area. These results suggested that the suppressive effects of TA on intimal hyperplasia might partly result from its inhibitory effect against VSMC proliferation. © 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Tanshinone; Restenosis; Proliferation of smooth muscle cell; Intimal hyperplasia

Percutaneous transluminal coronary angioplasty (PTCA) 1995). Intimal hyperplasia is a process of vascular smooth has revolutionized the treatment of atherosclerosis coronary muscle cell (VSMC) proliferation, migration, and transition arterial stenosis, but the long-term success of angioplasty to a secretory phenotype that is induced by multiple growth remains limited by the high occurrence of restenosis. From stimuli released during the arterial injury (Cassells et al., a number of clinical and experimental studies that used 1994). Therapeutic interventions that inhibit the proliferation intracoronary ultrasound and/or histological evaluation of of VSMC may therefore be beneficial in the prevention tissues, it is clear that there are three major mechanisms of restenosis. responsible for restenosis: intimal hyperplasia, elastic recoil, Danshen is a well-known traditional Chinese medicine and vessel wall remodeling (Safian et al., 1990; Post et used for the treatment of cardiovascular diseases. Recent al., 1994; Mintz et al., 1996). Vessel wall remodeling and studies have showed that danshen significantly inhibited inti- elastic recoil can be effectively counteracted by permanently mal hyperplasia and attenuated restenosis (Zhang et al., 1998; fixing the vessel size by the placement of an intracoronary Chen et al., 2001). Tanshinone (TA) is the major lipid-soluble stent (Gordon et al., 1993; Painter et al., 1995). However, pharmacological constituent of danshen. Our previous study restenosis due to intimal hyperplasia and production of demonstrated that TA-IIA, one of the main components of extracellular matrix material within the neointima still occurs TA, had dose-dependent inhibition on the basic fibroblast in a significant percentage of patients, more importantly after growth factor (bFGF)-induced human Smooth muscle cell intracoronary stenting (Gordon et al., 1993; Painter et al., (SMC) proliferation in vitro (Du, 1999). The purpose of this paper is to investigate the effect of TA on intimal hyperpla- ∗ Corresponding author. Tel.: +86 28 85501381; fax: +86 28 85503938. sia in vivo. Therefore, we used a mouse model with carotid E-mail address: dujr [email protected] (J.-r. Du). artery ligation, in which the lumen narrowing occurred by 1 Both are the first authors. the formation of an extensive SMC-rich neointima and a re-

0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2005.01.038 320 J.-r. Du et al. / Journal of Ethnopharmacology 98 (2005) 319–322 duction in vessel diameter (Kumar and Lindner, 1997). The 1.3. Immunohistochemical staining inhibition of TA on intimal hyperplasia and proliferating state of VSMC was observed. PCNA-positive VSMC was identified by staining with a mouse monoclonal antibody (Mab, DAKO, Carpinteria, CA). The avidin–biotin–peroxidase complex method was 1. Materials and methods used for the immunostaining of PCNA as previously re- ported (Lohr et al., 1995). In brief, after dewaxing, inacti- 1.1. Mouse intimal hyperplasia model vating endogenous peroxidase activity and blocking cross- reactivity with normal serum (Vectastain Elite Kit; Vector, All animal studies were approved by the Institutional Ani- Burlingame, CA), the sections were incubated overnight at mal Care and Use Committee in China. Thirty-six female KM 4 ◦C with a diluted solution of the primary antibodies (1:200 mice (7–8 weeks old, West China animal center of Sichuan for PCNA). Location of the primary antibodies was achieved University) weighing 20–22 g were used in all experiments. by subsequent application of a biotinylated anti-primary an- The animals were randomly divided into three groups: model tibody, an avidin–biotin complex conjugated to horseradish control, TA 0.3 and 0.6 g/kg BW/d groups. Each group had peroxidase, and diaminobenzidine (Vectastain Elite Kit, Vec- 12 mice. The controlateral carotid artery of ligated vessel of tor, Burlingame, CA). The slides were counter-stained by model control group was regarded as normal control group. hematoxylin. Negative controls were established by replac- The mouse intimal hyperplasia model was set up as described ing the primary antibody with PBS. Known immunostaining- (Kumar and Lindner, 1997). In brief, mice were anesthetized positive slides were used as positive controls. The numbers by intraperitoneal injection of a solution of xylazine (5 mg/kg of total and stained nuclei were counted separately for the in- body weight) and ketamine (80 mg/kg body weight). The left tima with image analysis software (Nikon & Spot, USA) and carotid arteries of the mice were dissected and ligated near PCNA-positive indices [(stained nuclei/total nuclei) × 100] the carotid bifurcation, leading to intimal hyperplasia and were calculated. thus establishing restenosis model. All animals recovered and showed no symptoms of a stroke. 90.5% purity of TA (w/w) 1.4. Statistical analysis (Hang Zhou Tian Nong Material Extract Company, Hang Zhou, China) was dispersed in 0.5% sodium carboxymethyl Data were expressed as mean ± S.E.M. Unpaired Stu- cellulose, and orally administered 2 days before ligation of dent’s t-tests were used to compare the values between nor- vessel and continued until the mice were killed 4 weeks after mal vessels and ligated vessels. The effect of TA treatment operation. The control animals were administrated the equal was compared by repeated-measures ANOVA. Means were volume of the solvent. considered significantly different if P < 0.05.

1.2. Histopathological staining and morphometry 2. Results After a 4-week treatment, the mice were anesthetized again and fixed for 3 min by perfusion with 4% p- 2.1. Histopathological changes formaldehyde in 0.1 mol/L sodium phosphate buffer (pH 7.3). After excision of the left and right carotid arteries, all spec- The entire length of the vessels was obtained from the mice imens were immersion fixed in 10% formalin and then em- 4 weeks after ligation of the common carotid artery. Sections bedded in paraffin. Serial sections (5 ␮m thick) were cut for from the middle portion of specimens were analyzed. Among hematoxylin–eosin staining. The morphormetric analysis of the normal control vessel, the luminal surface is smooth, en- vessels was carried out by the observation under light micro- dothelial cell tightly lined with IEL, vascular wall only com- scope, and the digitized images were analyzed with graphic posed of few layers of cell and lumen area enough for the analysis software (Nikon & Spot, USA). The circumferences blood flow (Fig. 1a). As for the model control vessel, intimal (lengths) of the lumen, internal elastic lamina (IEL), and ex- hyperplasia is significant. Extensive VSMC was present in ternal elastic lamina (EEL) were determined by tracing along subendothelial space, lumen area was reduced and vascular the luminal surface, IEL, and EEL. Under the assumption that wall increased in thickness (Fig. 1b). Treatment with TA 0.3 the structures were circular, these measurements were used or 0.6 g/kg/d could efficiently improve the above histopatho- to calculate intimal area, medial area, intima-to-media ratio logical changes and prevent the extent of neointima (Fig. 1c and relative lumen area. The medial area was calculated by and d). subtracting the area defined by the IEL from the area defined by the EEL, and intimal area determined by subtracting the 2.2. Morphometric analysis lumen area from the area defined by the IEL. Intima-to-media ratio was the ratio of vessel intimal area to medial area. Rel- Morphometric analysis of the changes in vascular wall ge- ative lumen area is the percentage of lumen area to the sum ometry was carried out by the software. All histological pa- of lumen and vessel wall area. rameters of neointima formation in ligated vessels were sig- J.-r. Du et al. / Journal of Ethnopharmacology 98 (2005) 319–322 321

Fig. 2. Effect of TA on the PCNA expression in neointima of ligated carotid arteries in mice. **P < 0.01, compared with model group.

3. Discussion

The main pathological changes of restenosis after PTCA are the abnormal hyperplasia of vascular intima and narrowed Fig. 1. TA inhibits intimal thickening in ligated mouse carotid arteries. HE lumen. Proliferation and immigration of VSMC from media staining (original magnification ×200). (a) cross-section of a normal vessel; to intima have long been regarded as the primary mecha- (b) cross-section of a 4-week ligated vessel showing an intimal hyperplasia nism of restenosis. Among which models of animal resteno- and narrowed lumen; (c) and (d) cross-sections of 4-week treated ligated sis (Kumar and Lindner, 1997; Leidenfrost et al., 2003; vessels with TA 0.6 or 0.3 g/kg BW/d showing the inhibition of intimal Sirsjo et al., 2003; Uwatoku et al., 2003), the carotid-ligated hyperplasia and improvement of narrowed lumen. mouse is regarded as a reproducible one since complete ces- sation of blood flow can induce rapid proliferation of me- dial VSMC, thus leading to extensive neointima formation nificantly different from those in normal vessels (P < 0.01). and also the other conspicuous advantages such as easy ma- As compared with those of the normal group, the intimal nipulation and low cost. In the present experiment, the lig- area, medial area and intima-to-media ratio of ligated artery ated vessels showed marked intimal hyperplasia as compared of the model control group increased and the relative lumen with the normal ones. TA could efficiently prevent the neoin- area decreased obviously (P < 0.01). TA could significantly tima formation. By orally administrated 0.3 or 0.6 g/kg/d, TA improve all the changes in these morphological parameters could significantly improve the histopathological changes in (P < 0.01). No significant difference was observed between neointima: reducing the intimal area, medial area and the TA 0.3 and 0.6 g/kg/d groups (Table 1). ratio of intima to media, and increasing the relative lumen area (P < 0.01). TA also inhibited PCNA expression in in- tima. PCNA is a special marker for cellular proliferation and 2.3. PCNA expression found most cell phases during proliferation(Yue et al., 2000). The number of PCNA-positive cells was used as an index The dark brown labeled PCNA-positive nuclei were ob- of VSMC proliferation in injured arteries. In this paper, a served within the vascular neointima. PCNA is significantly significant difference in PCNA-positive indices in neointima less expressed in the TA-treated groups as compared to the was observed between model control and TA-treated groups model control group. The PCNA-positive indice in model (P < 0.01). Thus, the suppressive effect of TA on intimal hy- control, TA 0.3 and 0.6 g/kg BW/d groups were 32.52 ± 2.46, perplasia may, at least in part, result from the inhibition of 5.67 ± 1.42 and 5.21 ± 0.95%, respectively (Fig. 2). Thus, the proliferation of VSMC. The inhibitory effects of TA on TA significantly reduced the PCNA-positive indice in neoin- VSMC proliferation in the present in vivo experiment were tima, consistent with histopathological changes in neointima comparable with those found previously in in vitro experi- (P < 0.01). ment (Du, 1999).

Table 1 Effect of TA on intimal hyperplasia of ligated carotid arteries in mice (¯x ± s, n = 12) Groups Intimal area (␮m2) Medial area (␮m2) Relative lumen area (%) Ratio of intima/media (%) Normal control 410 ± 141 56637 ± 33224 329.7 ± 1.7 1.2 ± 1.1 Model control 37778 ± 22534## 125311 ± 62403## 32.2 ± 14.9## 31.8 ± 11.2## TA 0.3 g/kg/d × 4 weeks 5334 ± 2706** 58746 ± 30753** 43.4 ± 6.9* 10.0 ± 4.9** TA 0.6g/kg/d × 4 weeks 4787 ± 2679** 44984 ± 23420** 46.8 ± 9.0* 10.9 ± 5.0** ## P < 0.01, compared with normal control group. ∗ P < 0.05, compared with model control group. ∗∗ P < 0.01, compared with model control group. 322 J.-r. Du et al. / Journal of Ethnopharmacology 98 (2005) 319–322

It is known that restenosis is a complex vascular repair Kumar, A., Lindner, V., 1997. Remodeling with neointima formation in process in response to luminal injury that is induced by the mouse carotid artery after cessation of blood flow. Arteriosclerosis, multiple growth stimuli released during the arterial injury. Thrombosis and Vascular Biology 17, 2238–2244. Leidenfrost, J.E., Khan, M.F., Boc, K.P., Villa, B.R., Collins, E.T., Parks, These growth stimuli include local factors and blood-borne W.C., Abendschein, D.R., Choi, E.T., 2003. A model of primary growth factors, especially associated with lipoprotein oxida- atherosclerosis and post-angioplasty restenosis in mice. American tion (Steinberg et al., 1989) and activation of platelets (Liu et Journal of Pathology 163, 773–778. al., 1989). On the basis of the ‘oxidation theory,’ the oxida- Mintz, G.S., Popma, J.J., Pichard, A.D., Kent, K.M., Satler, L.F., Chiu, tion of LDL in the artery wall is an early initiating event and W.S., Hong, M.K., Kovach, J.A., Leon, M.B., 1996. Arterial remodel- ing after coronary angioplasty: a serial intravascular ultrasound study. contributes to atherogenesis. Previous studies have shown Circulation 94, 35–43. that TA was an effective antioxidant against LDL oxidation Painter, J.A., Mintz, G.S., Wong, C., Popma, J.J., Pichard, A.D., Kent, (Zhou et al., 1999; Niu et al., 2000). TA cannot only inhibit K.M., Satler, L.F., Leon, M.B., 1995. Serial intravascular ultrasound lipid peroxidation but also increase some endogenous antiox- studies fail to show evidence of chronic Palmaz-Schatz stent recoil. idant enzymes, such as superoxide dismutase, glutathione American Journal of Cardiology 75, 398–400. Post, M.J., Borst, C., Kuntz, R.E., 1994. The relative importance of ar- peroxidase and catalase. Therefore, TA may interfere with terial remodeling compared with intimal hyperplasia in lumen renar- restenosis process via its antioxidant activity. rowing after balloon angioplasty: a study in the normal rabbit and Taken together, intimal hyperplasia remains a significant the hypercholesterolemic Yucatan micropig. Circulation 89, 2816– problem of restenosis because effective clinical therapy to re- 2821. duce intimal hyperplasia is still unavailable. Here, our novel Liu, M.W., Roubin, G.S., King III, S.B., 1989. Restenosis after coro- nary angioplasty: potential biologic determinants and role of intimal in vivo finding shows that TA inhibits the development of hyperplasia. Circulation 79, 1374–1387. intimal thickening in an animal model of arterial ligation, Niu, X.L., Ichimori, K., Yang, X., Hirota, Y., Hoshiai, K., Li, M., which, as expected, is associated with the inhibition of VSMC Nakazawa, H., 2000. Tanshinone II-A inhibits low density lipopro- proliferation. Since TA may affect many vascular processes tein oxidation in vitro. Free Radical Research 33, 305–312. responsible for restenosis, more experiments are needed to Safian, R.D., Gelbfish, J.S., Erny, R.E., Schnitt, S.J., Schmidt, D.A., Baim, D.S., 1990. Coronary atherectomy: clinical, angiographic, and his- clarify the exact mechanisms underlying its protective effect tological findings and observations regarding potential mechanisms. on restenosis. The efficacy of TA in preventing neointimal Circulation 82, 69–79. proliferation in injured vessels and the other vascular protec- Sirsjo, A., Lofving, A., Hansson, G.K., Wagsater, D., Tokuno, S., Valen, tive effects of TA support a potential clinical use of TA to G., 2003. Deficiency of nitric oxide synthase 2 results in increased reduce restenosis in humans. neointima formation in a mouse model of vascular injury. Journal of Cardiovascular Pharmacology 41, 897–902. Steinberg, D., Parthasarathy, S., Carew, T.E., Khoo, J.C., Witztum, J.L., 1989. Beyond cholesterol: modifications of low-density lipoprotein References that increase its atherogenicity. New England Journal of Medicine 320, 915–924. Cassells, W., Engler, D., Willerson, J., et al., 1994. Mechanism of resteno- Uwatoku, T., Shimokawa, H., Abe, K., Matsumoto, Y., Hattori, T., Oi, sis. Texas Heart Institute Journal 21, 68–77. K., Matsuda, T., Kataoka, K., Takeshita, A., 2003. Application of Chen, Y.L., Yang, S.P., Shiao, M.S., Chen, J.W., Lin, S.J., 2001. Salvia nanoparticle technology for the prevention of restenosis after balloon miltiorrhiza inhibits intimal hyperplasia and monocyte chemotactic injury in rats. Circulation Research 92, e62–e69. protein-1 expression after balloon injury in cholesterol-fed rabbits. Yue, T.L., Vickery-Clark, L., Louden, C.S., Gu, J.L., Ma, X.L., Journal of Cellular Biochemistry 83, 484–493. Narayanan, P.K., Li, X., Chen, J., Storer, B., Willette, R., Gossett, Du, J.R., 1999. Inhibition of proliferation of human smooth muscle cells K.A., Ohlstein, E.H., 2000. Selective estrogen receptor modulator by tanshinone II A. West China Journal of Pharmaceutical Sciences idoxifene inhibits smooth muscle cell proliferation, enhances reen- 14, 1–3. dothelialization, and inhibits neointimal formation in vivo after vas- Gordon, P.C., Gibson, M., Cohen, D.J., Carrozza, J.P., Kuntz, R.E., Baim, cular injury. Circulation 102, III281–III288. D.S., 1993. Mechanisms of restenosis and redilation within coronary Zhang, T., Du, L., Cai, P., et al., 1998. Experimental study of Hirado and stents: quantitative angiographic assessment. Journal of The American Salvia Miltiorrhiza on smooth muscle cell hyperplasia after balloon College of Cardiology 21, 1166–1174. angioplasty. Zhongguo Zhong Xi Yi Jie He Za Zhi 18, 159–161. Lohr, F., Wenz, F., Haas, S., Flentje, M., 1995. Comparison of proliferat- Zhou, G.Y., Zhao, B.L., Hou, J.W., Ma, G.E., Xin, W.J., 1999. Protec- ing cell nuclear antigen (PCNA) staining and BrdUrd-labelling index tive effects of sodium tanshinone IIA sulphonate against adriamycin- under different proliferative conditions in vitro by flow cytometry. induced lipid peroxidation in mice hearts in vivo and in vitro. Phar- Cell Proliferation 28, 93–104. macological Research 40, 487–491. Journal of Ethnopharmacology 98 (2005) 323–327

Antiviral activity in vitro of Urtica dioica L., Parietaria diffusa M. et K. and Sambucus nigra L.

R.E. Uncini Manganelli a,∗, L. Zaccaro b, P.E. Tomei a

a Department of Agronomia e Gestione dell’Agroecosistema, University of Pisa, Via S. Michele degli Scalzi 2, Pisa 56100, Italy b Department of Experimental Pathology and Retrovirus Centre, University of Pisa, Via S. zeno 35 Pisa, Italy Received 10 November 2004; received in revised form 18 January 2005; accepted 18 January 2005

Abstract

Parietaria diffusa M. et K., Urtica dioica L. (Urticaceae) and Sambucus nigra L. (Caprifoliaceae) are plants usually used in popular medicine of central Italy for treating numerous diseases, first of all Herpes zoster. Several plant products have been described as potential antiviral agents, with special attention being devoted to those having retroviruses as etiological agents, including acquired immunodeficiency syndrome (AIDS), in which a retrovirus, the designated human immunodeficiency virus HIV, has been clearly identified as the primary cause of this disease. The present study proposes a preliminary screening of the antiviral activity of Parietaria diffusa, Sambucus nigra and Urtica dioica preparation against the feline immunodeficiency virus (FIV) infection. The feline immunodeficiency virus is a widespread lentivirus of domestic cats sharing numerous biological and pathogenic features with the human immunodeficiency virus (HIV). FIV infection in cats has therefore been proposed as an animal model for AIDS studies with respect to pathogenesis, chemotherapy, and vaccine development [Pedersen, N.C., 1993. Feline immunodeficiency virus infection. In: Levy, J.A. (Ed.), The Retroviridae. Plenum Press, New York; Bendinelli, M., Pistello, M., Lombardi, S., Poli, A., Garzelli, C., Matteucci, D., Ceccherini-Nelli, L., Malvaldi, G., Tozzini, F., 1995. Feline immunodeficiency virus: an interesting model for AIDS studies and an important cat pathogen. Clinical Microbiology Revue 8, 87–112; North, T.W., LaCasse, R.A., 1995. Testing anti-HIV drugs in the FIV model. Nature Medicine 1, 410–411; Matteucci, D., Pistello, M., Mazzetti, P., Giannechini, S., Isola, P., Merico, A., Zaccaro, L., Rizzati, A., Bendinelli, M., 2000. AIDS vaccination studies using feline immunodeficiency virus as a model: immunisation with inactivated whole virus suppresses viraemia levels following intravaginal challenge with infected cells but non-following intravenous challenge with cell-free virus. Vaccine 18, 119–130]. Early studies showed that some of them presented antiviral activity against infection of FIV as assayed by syncytia formation using feline kidney Crandell cells (CrFK). © 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Ethnopharmacobotany; Tuscany; Italy

1. Introduction Manganelli et al., 2002), some belong to popular medicine (Uncini Manganelli and Tomei, 1999a; Camangi et al., 2001; Also in fully-developed countries – such as central Italy – Camangi and Uncini Manganelli, 2002). numerous plants are used in popular medicine and very inter- An anthropological and botanical in loco research was esting traditions are still maintained (Martini, 1983; Leporatti done first, which was then followed by pharmacological and et al., 1985; Leporatti and Pavesi, 1990; Uncini Manganelli phytochemical research, to verify the value of some original and Tomei, 1999a). In Tuscany ethno-pharmacobotanical uses (Calderone et al., 1998; Uncini Manganelli et al., 2000; studies have been in progress for over ten years with the Testai et al., 2002; Chericoni et al., 2003). presence of more than 500 plant species; some of these are Among the plants used as antivirals we began the used in handicrafts, in liqueurs, as ornamentals, etc. (Uncini analysis of Parietaria diffusa M. et K. and Urtica dioica L. (Urticaceae) and of Sambucus nigra L. (Caprifoliaceae) ∗ Corresponding author. Parietaria diffusa M. et K., is also used for treating Herpes E-mail address: [email protected] (R.E. Uncini Manganelli). zoster and Herpes labialis; for this the freshly crushed plant

0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2005.01.021 324 R.E. Uncini Manganelli et al. / Journal of Ethnopharmacology 98 (2005) 323–327 is applied as a poultice which is changed every 5–6 h (Uncini and the number of syncytia was counted under the micro- Manganelli and Tomei, 1999b). scope. Only the syncytia composed of at least seven nuclei In inland areas – for example, the Appennino – Sambucus were counted. The cytotoxicity of the extracts of the three nigra L. is used; the bark is cooked in olive oil and bees-wax plants in CrFK cells was determined by microscopic exam- is added; this cream is applied often during the day (Uncini ination of cell death and integrity after an incubation time Manganelli and Tomei, 1996). of 72 h (data not shown). The positive check/test was carried The Urtica dioica decoction is applied with a compress out using a compound whose antiviral activity and action and is changed every 5–6 h. mechanism in vitro are well known, that is to say destrane (Tanabe-Tochikura et al., 1992)(Fig. 1).

2. Methodology 2.5. Statistical analysis 2.1. Plant material and extraction Data are expressed as means ± S.D. The significance of All the plant material was collected in their respective differences between groups was determined by single-factor areas of usage, and a voucher specimen was deposited at the ANOVA, when compared to control cultures in the absence Botanic Garden of Lucca (LU) (Parietaria diffusa M. et K.: of plant extract. A value of p lower than 0.05 was considered LUCCA 5 229; Sambucus nigra L.: LUCCA 73 3611; Urtica as significant. dioica L.: LUCCA 5 223). The fresh bark of Sambucus nigra was boiled for 20 min about (1 g/10 ml); the extract was filtered and then was 3. Results lyophilized. The lyophilized extract was melted in bidistilled water at protocol percentage. The same process was done Acute infection and cell-to-cell transmission of FIV were with the root of Urtica dioica. assayed by the syncytia formation assay. This assay, based Parietaria officinalis was instead centrifuged, filtered and on the interaction between fusigenic FIV-infected cells, has then lyophilized. been used with other natural products (Calderone et al., 1998; Nicoletti et al., 1999). After an incubation of six days all 2.2. Cells extracts showed an antiviral activity, with a pattern similar to that of destrane (Fig. 1). Crandell feline kidney cells (CrFK, ATCC clone CCL94) The aqueous extract of Urtica dioica indicated a good in- were grown in Eagle’s minimal essential medium containing hibition on the development of syncytia (Fig. 2) with low 0.5% fetal calf serum (FBS), 2 nM l-glutamine, penicillin doses (0.5–1 ␮g/ml) and increased when the concentration (50 IU/ml) and streptomycin (50 ␮g/ml), a variety of growth rose until it reached an inhibition level of 84% which, how- factors and other supplements (Tozziniand Bandecchi, 1985). ever, began to show cytotoxic effects. The inhibition trend on the syncytia formation based on 2.3. Virus Parietaria diffusa is reported in Fig. 3. The inhibiting activity of the syncytia reached good levels in a short time, but then it Feline immunodeficiency viruses, the California isolate remained stable at insufficient values and obtained complete Petaluma strain (FIV-Pet) kindly provided by Yamamoto et inhibition reaching the maximum value of 88.5% regarding al. (1988, 1991) has been used. This virus induces formation the sub-toxic concentration of 32 ␮g/ml. of syncytia in the CrFk cells. The titration of the viral stocks was performed as described by Tozzini et al. (1992). The viral titres were expressed as syncytium forming units virus (SFU/ml).

2.4. Experimental protocol

The extract plants were screened for the capability to in- hibit FIV replication using an assay based on FIV-induced syncytia formation (Tozzini et al., 1992). Briefly, different concentrations of plant extracts were added to CrFK cells grown in 24-well plates (2 × 104 cells per well) in 0.8 ml of culture medium containing 0.5% FBS. After 1 h at 37 ◦C approximately 50–100 syncytia-forming units (SFU) of FIV- Pet were added in 0.2 ml volume. Six days later, the cul- tures were stained by crystal violet (0.5%) in methanol 30% Fig. 1. Inhibition of FIV-induced syncytium formation by destrane. R.E. Uncini Manganelli et al. / Journal of Ethnopharmacology 98 (2005) 323–327 325

Fig. 2. Inhibition of FIV-induced syncytium formation by Urtica dioica. Fig. 4. Inhibition of FIV-induced syncytium formation by Sambucus nigra. Each point represents the mean ± S.D. of three independent experiments; Each point represents the mean ± S.D. of three independent experiments; * p < 0.05. * p < 0.05.

tecting substances that block FIV infection, such as natural extracts (Nicoletti et al., 1999; Schmitt et al., 2001). The results of our research have shown that all three plants have a good antiviral activity; the cytotoxicity stopped the attainment of total inhibition for Urtica dioica and Parietaria diffusa. Parietaria diffusa is a well-known and studied plant due to its allergenic properties, its pollen being mainly responsible for numerous spring allergies. From a therapeutic point of view, in popular medicine the infusion of its aerial components, taken orally, is considered to be helpful in treating the urinary tracts and to increase diuresis; the fresh plant is applied locally to soothe skin bitten by insects or touched by stinging nettles. Fig. 3. Inhibition of FIV-inducedsyncytium formation by Parietaria diffusa. Each point represents the mean ± S.D. of three independent experiments; Not many studies on the therapeutical uses of this plant * p < 0.05. can be found in literature, however, its diuretic and uricosuric properties are well known (Giachetti et al., 1986). No information is available concerning its possible antivi- The assay of syncytia inhibition for Sambucus nigra ral use. showed a noticeable anti-FIV activity, even if the inhibition The nettle, in popular medicine, is used very much in var- curve is not regular (Fig. 4). ious minor pathologies, besides the above-mentioned use for At the beginning, the decrease in number of syncytia hap- treating Herpes zoster the infusion of its aerial components pens progressively and reaches values which are over 43% can be drunk to favor diuresis, to soothe rheumatism, and to for concentrations of 1 ␮g/ml and to values over 85% for con- lower high blood pressure; externally it can be rubbed on the centrations of 31 ␮g/ml; then there is a digression of about scalp to stimulate hair growth. Recent studies have confirmed 60–70% for concentrations between 62 and 125 ␮g/ml con- diuretic and hypertensive properties regarding Urtica dioica tinuing up to 100% inhibition at 500 ␮g/ml. (Tahri et al., 2000); interesting results have been obtained from the discovery that the aqueous extract has antihyper- glycemic properties (Bnouham et al., 2003), and it is also a 4. Discussion and conclusion good antioxidant (Pieroni et al., 2002). The data we have obtained using the entire extract of Ur- The therapeutic properties of plants have been attributed tica dioica and the syncytium formation test are similar to to crude extracts or isolated compounds, often reflecting their those obtained using the active principles derived from the role in traditional medicine. same plant. The N-acetylglucosamine-specific lectin from In the study of natural products that have antiviral activ- Urtica dioica was a strong inhibitor of syncytium formation ity, previous studies have shown that inhibition of syncytium between HUT-78 cells and CD4 + Molt/4 cells permanently formation in CrFK cells is a useful screening method for de- infected by HIV-1 and HIV-2 (Balzarini et al., 1992). 326 R.E. Uncini Manganelli et al. / Journal of Ethnopharmacology 98 (2005) 323–327

The most important differences regard the effective dosage lectins from Cymbidium hybrid and Epipactis helleborine and the (lectin dosage is lower than the raw extract) and the amount of (N-acetylglucosamine)n-specific plant lectin from Urtica dioica are inhibition, which towards HIV is total, towards FIV is partial, potent selective inhibitors of human immunodeficiency virus and cytomegalovirus replication in vitro. Antiviral Research 18, 191– even if high. 207. The reason for this is likely to be found in the different Bnouham, M., Merhfour, F.Z., Ziyyat, A., Mekhfi, H., Aziz, M., Legssyer, composition of the substances which have been tested: the A., 2003. Antihyperglycemic activity of aqueous extract of Urtica extract used in the inhibition test is a raw extract, which is, dioica. Fitoterapia 74, 677–681. qualitatively and quantitatively, not well defined, while the Calderone, V., Nicoletti, E., Bandecchi, P., Pistello, M., Mazzetti, P., Mar- tinotti, E., Morelli, I., 1998. In vitro antiviral effects of an aqueous results reported in literature were obtained assaying a pure extract of Artemisia verlotorum Lamotte (Asteraceae). Phytotherapy compound having a well-defined biological activity. Research 12, 595–597. It is necessary to continue these studies to determine the Camangi, F., Uncini Manganelli, R.E., Tomei, P.E., 2001. L’uso delle chemical identification of all constituents. piante nella medicina popolare del territorio di Collodi. Acta Phy- Sambucus nigra is a very well-known plant in official phy- totharapeutica 2, 58–65. Camangi, F., Uncini Manganelli, R.E., 2002. Alla scoperta delle piante totherapy, whose flowers stimulate bronchial secretion and medicinali nella tradizione popolare della Media Valle del Serchio. have a diaphoretic activity (Longo, 1996); recent studies have Comune di Barga (LU), Felici Editore, Pisa. confirmed the presence of non-toxic ribosome-inactivating Chericoni, S., Testai, L., Calderone, V., Flamini, G., Nieri, P., Morelli, I., protein (RIP) in the bark, leading to the inhibition of protein Martinotti, E., 2003. The xantones Gentiacaulein and Gentiakochianin synthesis. Conjugation of RIP to monoclonal antibodies is a are responsible for the vasodilator action of the roots of Gentiana kochiana. Planta Medica 69, 770–772. promising tool for cancer therapy (Girbes et al., 2003). Giachetti, D., Taddei, E., Taddei, I., 1986. Diuretic and uricosuric activity In folk medicine many parts of this tree are used: the fruit of Parietaria judaica L. della Societa` Italiana di Biologia Sperimen- as a ; the leaves are applied to wounds and bruises tale. Bollettino 62, 197–202. to help soothe and heal, the infusion of the flowers is ap- Girbes, T., Ferreras, J.M., Arias, F.J., Munoz, R., Iglesias, R., Jimenez, plied to sore red eyes or is drunk for gastro-intestinal prob- P., Rojo, M.A., Arias, Y., Perez, Y., Benitez, J., Sanchez, D., Gayoso, M.J., 2003. Non-toxic type 2 ribosome-inactivating pro- lems. Concerning the antiviral activity of elder, the analysis teins (RIPs) from Sambucus: occurrence, cellular and molecular ac- of the bark infusion, based on popular medicine, is not in tivities and potential uses. Cellular Molecular Biology 49, 537– agreement with the data, which are few in literature. Data 545. can be found regarding two lectins isolated from elder, one Leporatti, M.L., Pavesi, A., 1990. New or uncommon uses of several specific for N-acetylgalactosamine/galactose the other for 5- medicinal plants in some areas of central Italy. Journal of Ethnophar- macology 29, 213–223. acetylneuroaminic acid, both resulting inefficacious in hin- Leporatti, M.L., Pavesi, A., Posocco, E., 1985. Phytotherapy in the dering the development of regressive cytopathology induced Valnerina Marche (central Italy). Journal of Ethnopharmacology 14, by HIV-1 and HIV-2 on the MT-4 cells and the appearance 53–63. of induced syncytia form the cultivation of Molt-4 cells with Longo, R., 1996. Le Monografie Tedesche. Studio Edizioni, Milano, p. HUT-78 permanently infected (Balzarini et al., 1992). 50. Martini, E., 1983. La fitoterapia popolare dell’alta Valle d’Orba (Appen- The assay of syncytia inhibition instead shows that with nino Ligure). Atti della Accademia Ligure di Scienze e Lettere 39, non-toxic concentrations the extract of elder has a high anti- 3–25. FIV activity. It could be hypothesized that substances having Nicoletti, E., Della Pieta,` F., Calderone, V., Bandecchi, P., Pistello, M., anti-FIV activity are present in the extract which are different Morelli, I., Cinelli, F., 1999. Antiviral properties of a crude extract from the two lectins studied with HIV. from a green alga Caulerpa taxifolia (Vahl) C. Agardh Phytotherapy Research 13, 245–247. This fact reinforces the value of ethno-pharmacological Pieroni, A., Janiak, V., Durr, C.M., Ludeke, S., Trachsel, E., Heinrich, data in the search for bioactive substances. Thus, more de- M., 2002. In vitro antioxidant activity of non-cultivated vegetables of tailed studies should be undertaken to isolate and identify ethnic Albanians in southern Italy. Phytotherapy Research 16, 467– antiviral components of specific parts of this plant and most 473. active extracts contributing to find antiviral agents against Schmitt, A.C., Ravazzolo, A.P., von Poser, G.L., 2001. Investigation of some Hypericum species native to southern Brazil for antiviral activity. lentiviruses. Journal of Ethnopharmacology 77, 239–245. Further studies to make a sub-division of the extract and Tahri, A., Yamani, S., Legssyer, A., Aziz, M., Mekhfi, H., Bnouham, M., the analysis of its main components are now in progress. Ziyyat, A., 2000. Acute diuretic, natriuretic and hypotensive effects of a continuous perfusion of aqueous extract of Urtica dioica in the rat. Journal of Ethnopharmacology 73, 95–100. Acknowledgement Tanabe-Tochikura, A., Tochikura, T.S., Blakeslee Jr., J.R., Olsen, R.G., Mathes, L.E., 1992. Anti-human immunodeficiency virus (HIV) agents The authors thank Jane Errkila for the English review. are also potent and selective inhibitors of feline immunodeficiency virus (FIV)-induced cytopathic effect: development of a new method for screening of anti-FIV substances in vitro. Antiviral Research 19, 161–172. References Testai, L., Chericoni, S., Calderone, V., Nencioni, G., Nieri, P., Morelli, I., Martinotti, E., 2002. Cardiovascular effects of Urtica dioica L. (Ur- Balzarini, J., Neyts, J., Schols, D., Hosoya, M., Van Damme, E., ticaceae) roots extracts: in vitro and in vivo pharmacological studies. Pneumans, W., De Clercq, E., 1992. The mannose-specific plant Journal of Ethnopharmacology 81, 105–109. R.E. Uncini Manganelli et al. / Journal of Ethnopharmacology 98 (2005) 323–327 327

Tozzini, F., Bandecchi, P., 1985. Terreno di coltura per le cellule animali Uncini Manganelli, R.E., Chericoni, S., Baragatti, B., 2000. Ethnophar- contenente fattori di crescita e sostanze nutritive, in alternativa al siero macobotany in Tuscany: plants used as antihypertensives. Fitoterapia fetale bovino. Archivi Veterinari Italiani 36, 151–158. 71, 95–100. Tozzini, F., Matteucci, D., Bandecchi, P., Baldinotti, F., Poli, A., Pistello, Uncini Manganelli, R.E., Camangi, F., Tomei, P.E., Oggiano, N., 2002. M., Siebelink, K.H., Ceccherini-Nelli, L., Bendinelli, M., 1992. Sim- Riscoprire pratiche antiche per creare opportunita’ d’impresa: l’uso ple in vitro methods for titrating feline immunodeficiency virus (FIV) delle erbe nella tradizione rurale Toscana, vols. I–II. Edizioni ARSIA, and FIV neutralizing antibodies. Journal of Virological Methods 37, Firenze. 241–252. Yamamoto, J.K., Sparger, E., Ho, E.W., Andersen, P.R., O’Connor, T.P., Uncini Manganelli, R.E., Tomei, P.E., 1996. Indagini farmaco-botaniche Mandell, C.P., Lowenstine, L., Munn, R., Pedersen, N.C., 1988. in Garfagnana (Lucca): il versante appenninico. Atti Societa` Toscana Pathogenesis of experimentally induced feline immunodeficiency virus di Scienze Naturali, Memorie, Serie B 102, 3–18. infection in cats. American Journal of Veterinary Research 49, Uncini Manganelli, R.E., Tomei, P.E., 1999. Documenti per la conoscenza 1246–1258. delle tradizioni etno-farmacobotaniche in Toscana. San Marco Yamamoto, J.K., Ackley, C.D., Zochlinski, H., Louie, H., Pembroke, E., Litotipo, Lucca. Torten, M., Hansen, H., Munn, R., Okuda, T., 1991. Development of Uncini Manganelli, R.E., Tomei, P.E., 1999b. Ethno-pharmacobotanical IL-2 independent feline lymphoid cell lines chronically infected with studies of the Tuscan Archipelago. Journal of Ethnopharmacology feline immunodeficiency virus: importance for diagnostic reagents and 65, 181–202. vaccines. Intervirology 32, 361–375. Journal of Ethnopharmacology 98 (2005) 329–333

In vitro anti-Helicobacter pylori action of 30 Chinese herbal medicines used to treat ulcer diseases Yang Li, Chen Xu, Qiang Zhang, Jun Yan Liu, Ren Xiang Tan ∗

Institute of Functional Biomolecules, State Key Laboratory of Pharmaceutical Biotechnology, Nanjing University, Nanjing 210093, PR China Received 9 December 2004; received in revised form 18 January 2005; accepted 18 January 2005

Abstract

Infection by Helicobacter pylori has been ascertained to be an important etiologic impetus leading usually to chronic active gastritis and gastric ulcer with growing incidences worldwide. Utilizing as the test pathogen a standard and five clinic strains of Helicobacter pylori, the antibacterial action was assessed in vitro with ethanol extracts of 30 Chinese herbal medicines which have been frequently prescribed since ancient times for treating gastritis-like disorders. Among the 30 tested materials, the ethanol extracts of Abrus cantoniensis (Fabaceae), Saussurea lappa (Asteraceae) and Eugenia caryophyllata (Myrtaceae) were strongly inhibitory to all test strains (MICs: ∼40 ␮g/ml), and Hippophae rhamnoides (Elaeagnaceae), Fritillaria thunbergii (Liliaceae), Magnolia officinalis and Schisandra chinensis (Magnoliaceae), (Papaveraceae), Citrus reticulata (Rutaceae), Bupleurum chinense and Ligusticum chuanxiong (Apiaceae) substantially active with MICs close to 60.0 ␮g/ml. As to antibacterial actions of the aqueous extracts of the same drugs, those derived from Cassia obtusifolia (Fabaceae), Fritillaria thunbergii and Eugenia caryophyllata were remarkably inhibitory against all the six Helicobacter pylori strains (MICs: ∼60 ␮g/ml). The work compared almost quantitatively the magnitude of the anti-Helicobacter pylori actions of the 30 most prescribed gastritis-treating Chinese herbal drugs, and located as well some source plants where potent anti-Helicobacter pylori phytochemicals could be characterized. © 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Traditional Chinese medicines; Helicobacter pylori; Gastric ulcer; Antibacterial

1. Introduction al., 1994; Bell et al., 1995; Bayerdorffer¨ et al., 1995). This situation could be attributed to a couple of reasons: (1) Some The human bacterial pathogen Helicobacter pylori has strains of Helicobacter pylori harboring deeply inside gut been recognized as a major causative factor in peptic ulcer tissues could not be entirely killed even after a long-time ad- diseases (Graham, 1989; Blaser, 1992; Axon, 1993). As to ministration of pertinent antibiotics (Korman et al., 1997); (2) the therapy of those illnesses, high cure-rates are sometime The drug resistance of the bacterium could be developed dur- afforded by sophisticated therapeutic approaches based on an ing repeated applications of anti-Helicobacter pylori agents on-demand empirical combination of histamine-2- (or H2-) (Adamek et al., 1998; Marshall, 1991); and (3) as a result of receptor antagonist(s), proton pump inhibitor(s) and bismuth patient compliance and/or ulcerogenic side-effects of some compound(s) (De Korwin and Lozniewski, 1996). However, drugs co-administered imperatively for the management of the infection of Helicobacter pylori is still as a whole under other accompanying diseases. Accordingly, there is a grow- a poor management since the eradication failure rate remains ing need for finding new anti-Helicobacter pylori agents that as high as 5–20% along with frequent relapses of gastric ul- can hopefully eradicate the invasion and presence of the sur- cers even after the discerned ‘complete healings’ (Bazzoli et vived strain of Helicobacter pylori to avoid relapses of the gastric ulcer it causes. ∗ Corresponding author. Tel.: +86 25 8359 2945; fax: +86 25 8359 3201. In China, most traditional medicines have experienced for E-mail address: [email protected] (R.X. Tan). dozens of centuries continuous clinical practices with opti-

0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2005.01.020 330 Y. Li et al. / Journal of Ethnopharmacology 98 (2005) 329–333 mized combinations. Among those well-proven herbal drugs, (Megraud et al., 1999). Briefly, 1 ml of each stock solu- some have been documented to have significant efficacy for tions at given concentrations of every plant extract in ster- treating gastric ulcers. Thus, a thorough literature search was ile water (possibly less amounts of DMSO were used for an conducted to select a total of 30 traditional Chinese herbal acceptable intermiscibility) was separately added into petri medicines (belonging to 19 families), which have been most dishes containing 8 ml of unsolidified Columbia agar base frequently prescribed for that therapeutic purpose. A compar- supplemented with 1 ml of horse serum. Final concentrations ative investigation was subsequently performed concerning of each plant extract in the medium were set to be 200.0, the anti-Helicobacter pylori actions of the ethanol and aque- 100.0, 60.0, 40.0, 20.0, 10.0 and 5.0 ␮g/ml with DMSO ous extracts of the selected herbs. concentration lower than 1%. Different Helicobacter pylori strains pre-activated in seed cultures were immediately di- luted with Brucella broth with bacterial cells at approximately 2. Materials and methods 5 × 107 CFU/ml in resultant liquors. Subsequently, 0.1 ml of each thus-prepared liquor was inoculated onto the surface of 2.1. General the sample-supplemented agar plates, followed by being kept ◦ at 37 C for 72 h in anaerobic jar (containing: 85% N2, 10% Columbia agar was purchased from Biomerieux´ sa CO2 and 5% O2). The MICs were defined as the lowest con- (France). Horse serum was obtained from Yuanheng Shenma centration at which no microbial growth could be detectable. (Beijing, China). Ampicillin was from the National Institute Ampicillin was co-assayed as a positive reference at concen- for the Control of Pharmaceutical and Biological Products trations of 0.125, 0.25, 0.5, 1.0, 2.0, 4.0 and 8.0 ␮g/ml. (China). All other chemicals used in the study are of chemi- cal grade. 3. Results and discussion 2.2. Plant material Ethanol and aqueous extracts of 30 herbal medicines used The plant materials were collected or purchased from dif- traditionally for the treatments of gastric ulcers were screened ferent parts of China (Table 1). Each voucher specimen under in vitro for the anti-Helicobacter pylori actions. As sum- a given registration number, authenticated by Associate Prof. marized in Table 1, the strongest growth inhibitions (MICs: L.X. Zhang, was preserved in the Herbarium of Nanjing Uni- ∼40.0 ␮g/ml) against all the six test strains of Helicobacter versity, Nanjing, People’s Republic of China. pylori were discerned with the ethanol extracts from the aerial parts of Abrus cantoniensis, the stems of Saussurea lappa and 2.3. Preparation of extracts the flowers of Eugenia caryophyllata. Also significant anti- Helicobacter pylori actions with MICs around 60.0 ␮g/ml Two sets (40 g each) of roughly ground air-dried plant were exhibited by 11 additional ethanol extracts of Hip- materials of every selected species were extracted twice sep- pophae rhamnoides, Fritillaria thunbergii, Magnolia offic- arately with 95% ethanol and water (400 ml for each extrac- inalis and Schisandra chinensis, Corydalis yanhusuo, Citrus tion) by refluxing for 4 h on a water bath at 90 ◦C. In vacuo reticulata, Bupleurum chinense and Ligusticum chuanxiong. evaporation of solvents from the filtered ethanol and water As to the anti-Helicobacter pylori action of aqueous extracts, extracts gave residues, which were subjected to an immedi- only three derived from the leaves of Cassia obtusifolia, the ate lyophilization into dry powders with the yields given in stems of Fritillaria thunbergii and the flowers of Eugenia Table 1. caryophyllata exhibited strong inhibitions against all the test strains with the MICs of ∼60 ␮g/ml. As far as the ethanol and 2.4. Helicobacter pylori strains aqueous extracts of a single herb were compared, the former seemed more active than the latter (Table 1). Furthermore, the Used in this study were a standard strain (ATCC 43504) MICs of those active extracts being usually complex mixtures and five clinical isolates of Helicobacter pylori obtained from consisting of numerous phytochemicals are quite acceptable antral biopsies of childish and adult patients hospitalized at although they are higher than that (2.0 ␮g/ml) of ampicillin Jiangsu People’s Hospital in Nanjing. They were inoculated co-assayed as a positive reference in the study. into Columbia agar base plates supplemented with 7% horse A reference survey on the chemical composition of the serum and cultured for 3 days at 37 ◦C under microaerophilic three most anti-Helicobacter pylori medicines have been per- conditions with high humidity as detailed elsewhere (Fabry formed. The previous phytochemical attention to Eugenia et al., 1996). caryophyllata had led to the characterization of volatile oil (mainly contains eugenol and caryophyllen), a chromone C- 2.5. MIC measurements glucoside, elagitennins, phytosterols (Yu and Hung, 1981; Xiao et al., 2002). Among these plant constituents, eugenol The minimum inhibitory concentration (MIC) was de- was disclosed to have prominently antibacterial properties tected with the agar dilution method described earlier (Yu and Hung, 1981; Pattnaik et al., 1997). Furthermore, the Y. Li et al. / Journal of Ethnopharmacology 98 (2005) 329–333 331

Table 1 Minimum inhibitory concentrations (MICs: ␮g/ml) of ethanol (EE) and aqueous (AE) extracts of 30 medicinal plants against Helicobacter pylori strains Species Partsa MIC (␮g/ml) Collection Yield (%) Voucher number

EE AE Siteb Period EE AE Apiaceae Bupleurum chinense DC. AP 60 0 JS April 2001 14.814.6 NU-10532 Foeniculum vulgare Mill. FT >100 0 HLJ October 2000 3.514.8 NU-10545 Ligusticum chuanxiong Hort. RT 60 100 SC July 1999 6.832.0 NU-10587 Aristolochiaceae Aristolochia debilis Sieb. et Zucc RT >100 0 HN May 1999 10.212.5 NU-65442 Aristolochia mallissima Hance LF >100 >100 JS October 1998 10.17.8 NU-65430 Asteraceae Saussurea lappa Clarke. RT 40 >100 YN December 2001 21.855.0 NU-23410 Xanthium sibiricum Patr. FT 0 >100 JL September 1999 2.515.7 NU-23541 Elaeagnaceae Hippophae rhamnoides L. LF 60 >100 JS November 2000 5.318.0 NU-86531 Fabaceae Abrus cantoniensis Bge. AP 40 0 GD September 1999 3.97.6 NU-34156 Astragalus membranaceus Bge. RT >100 0 GS September 2001 30.6 NU-34123 Cassia obtusifolia Vahl. LF >100 60 YN August 2000 10.928.0 NU-34112 Glycyrrhiza uralensis Fisch. RH >100 100 HB October 2001 20.125.0 NU-34123 Sophora flavescens Ait. RT >100 >100 JS April 2001 22.623.3 NU-34198 Trigonella foenum-graecum L. SD 100 >100 SC October 1999 3.428.2 NU-34158 Lauraceae Lindera strychnifolia Vill. RT >100 0 HB March 2001 13.313.0 NU-71630 Lamiaceae Mentha haplocalyx Briq. AP >100 0 JS October 2001 9.818.0 NU-15632 Scutellaria barbata D.Don WP >100 100 SX August 2000 7.416.2 NU-15644 Liliaceae Fritillaria thunbergii Miq. ST 60 60 ZJ July 1998 1.712.3 NU-35210 Magnoliaceae Magnolia officinalis Rehd. et Wils. BK 60 >100 HB March 2000 7.67.5 NU-71563 Schisandra chinensis Baill. FT 60 0 LN October 1999 34.315.7 NU-71542 Meliaceae Melia toosendan Sieb. et Zucc. FT 100 >100 GS December 2001 13.422.3 NU-45633 Menispermaceae Tinospora sagittata Gagnep. RT >100 100 HN September 1999 9.116.8 NU-71652 Myrtaceae Eugenia caryophyllata Thunb FL 40 60 HN October 2000 27.928.8 NU-96210 Orchidaceae Bletilla striata Reichb.F. ST >100 0 AH October 2001 1.121.6 NU-52631 Papaveraceae Corydalis yanhusuo W.T. Wang ST 60 100 SD June 2001 2.58.9 NU-48210 Piperaceae Piper longum L. SP 100 >100 YN November 1998 4.87.3 NJ-45623 Polyporaceae Poria cocos Wolf RT 0 0 AH November 2000 9.540.9 NU-22145 Primulaceae Lysimachia christinae Hance WP 100 0 FJ April 2000 12.820.9 NU-52312 Rutaceae Citrus reticulata Blanco FS 60 100 ZJ October 2000 32.734.3 NU-21345 Zingiberaceae Amomum villosum Lour. FT 100 0 GD November 2000 6.45.3 NU-78015 a AP, aerial part; BK, bark; FL, flower; FT, fruit; FS, fruit shell; LF, leaf; RH, rhizome; RT, root; SD, seed; SP, spike; ST, stem; WP, whole plant. b AH, Anhui; FJ, Fujian; GD, Guangdong; GS, Gansu; HB, Hubei; HLJ, Heilongjiang; HN, Henan; JL, Jilin; JS, Jiangsu; LN, Liaoning; SC, Sichuan; SD, Shandong; SX, Shanxi; YN, Yunnan; ZJ, Zhejiang. 332 Y. Li et al. / Journal of Ethnopharmacology 98 (2005) 329–333 preliminary study also showed that this plant has an anti- metronidazole, and amoxicillin-influence on treatment outcome. The Helicobacter pylori activity, which was consistent with our American Journal of Gastroenterology 93, 386–389. results (Bae et al., 1998). Accordingly, anti-Helicobacter py- Axon, A.T.R., 1993. Helicobacter pylori infection. Journal of Antimicro- bial: Chemotherapy 32, 61–68. lori action of the ethanol extract of Eugenia caryophyllata Bae, E.A., Han, M.J., Kim, D.H., 1999. In vitro anti-Helicobacter pylori in this study might be related to the presence of eugenol al- activity of some flavonoids and their metabolites. Planta Medica 65, though none of the reported principles has been shown to be 442–443. anti-Helicobacter pylori. Bae, E.-A., Han, M.J., Kim, N.J., Kim, D.H., 1998. Anti-Helicobacter The ethanol extract of Abrus cantoniensis displayed as pylori activity of herbal medicines. Biological and Pharmaceutical Bulletin 21, 990–992. well the most significant inhibition to the growth of He- Bayerdorffer,¨ E., Miehlke, S., Mannes, G.A., Sommer, W., Hochter, licobacter pylori. Phytochemically, this plant has been as- J., Weingart, W., Helewein, H., Klann, T., Simon, W., Schmitt, E., certained to be a common source of triterpenic saponins, Bastlein, A., Eimiller, R., Hatz, N., Lehn, P., Dirschedl, P., Stolte, anthraquinones, alkaloids, flavonoids, and some other con- M., 1995. Double-blind trial of omeprazole and amoxicillin to cure stituents such as abrin, abrussic acid, uric acid, glabrolide Helicobacter pylori infection in patients with duodenal ulcer. Gas- troenterology 108, 1412–1417. (Xiao et al., 2002). Flavonoids and isoflavonoids present Bazzoli, F., Zagari, R.M., Fossi, S., Pozzato, P., Alampi, G., Simoni, in the plant could be the anti-Helicobacter pylori sub- P., Sottili, S., Roda, A., Roda, E., 1994. Short-term low-dose triple stances since some structurally related analogs such as therapy for the eradication of Helicobacter pylori infection. European flavonoids ponciretin and hesperetin as well as isoflavonoids Journal of Gastroenterology Hepatology 6, 773–777. i.e. cabreuvin, irisolidone, genistein and licoisoflavone were Bell, G.D., Powell, K.U., Burridge, S.M., Bowden, A.F., Atoyebi, W., Bolton, G.H., 1995. Rapid eradication of Helicobacter pylori infec- revealed to possess potent anti-Helicobacter pylori activities tion. Alimentary Pharmacology Therapeutics 9, 41–46. (Bae et al., 1999; Ohsaki et al., 1999; Fukai et al., 2002). Blaser, M.J., 1992. Helicobacter pylori: its role in disease. Clinical In- The third most anti-Helicobacter pylori traditional medic- fectious Diseases 15, 386–391. inal plant is Saussurea lappa, which has been demonstrated Cowan, M.M., 1999. Plant products as antimicrobial agents. Clinical Mi- to contain sesquiterpenes, monoterpenes, triterpenes, aro- crobiology Review 12, 564–582. De Korwin, J.D., Lozniewski, A., 1996. Le traitement de I’infection a` matic compounds, sterols, alkaloid (Yang et al., 1997; Mat- Helicobacter pylori. La Presse Medicale´ 25, 1917–1922. suda et al., 2003, 2000). Concerning the antibacterial action, Fabry, W., Okemo, P., Mwatha, W.E., Chauder, S., Ansorg, R., 1996. volatile oils from the plant were demonstrated to be antibacte- Susceptibility of Helicobacter pylori and Candida spp. To the East rial to Staphylococcus aureus and Streptococcus pneumoniae African Plant Terminalia spinosa. Arznermittel-Forschung/Drug Re- (Pattnaik et al., 1997; Wang, 1997). Therefore it is suggested search 46, 539–541. Fukai, T., Marumo, A., Kaitou, K., Kanda, T., Terada, S., Nomur, T., that the volatile oils maybe the constituents responsible for 2002. Anti-Helicobacter pylori flavonoids from licorice extract. Life the anti-Helicobacter pylori activity. Sciences 71, 1449–1463. Some anti-Helicobacter pylori natural products have been Gadhi, C.A., Mory, F., Benharref, A., Lion, C., Jana, M., Weber, M., reviewed earlier (Cowan, 1999). Also reported elsewhere to Lozniewski, A., 1999. Antibacterial activity of Aristolochia paucin- be anti-Helicobacter pylori were oil-macerated garlic con- ervis Pomel. Journal of Ethnopharmacology 67, 87–92. Graham, D.Y., 1989. Campylobacter pylori and peptic ulcer disease. Gas- stituents (Ohta et al., 1999), fatty acids and terpenes in Aris- troenterology 96, 615–625. tolochia paucinervis (Gadhi et al., 1999), costunolides in Korman, M.G., Bolin, T.D., Engelmann, J.I., Pianko, S., 1997. Sucralfate Magnolia sieboldii (Park et al., 1997), and some acidic and as an alternative to bismuth in quadruple therapy for Helicobacter alkaline constituents in Coptis japonica, Eugenia caryophyl- pylori eradication. Helicobacter 2, 140–143. lata, Magnolia officinalis and Rhus javanica (Bae et al., Marshall, B.J., 1991. In: McCallum, R.W., Ruerrant, R.T. (Eds.), Heli- cobacter pylori in Peptic Ulceration and Gastritis. Oxford Press, pp. 1998). The present results, along with the observation, sug- 160–186. gested that a wider range of phytochemicals could be anti- Matsuda, H., Kageura, T., Inoue, Y., Morikawa, T., Yoshikawa, M., 2000. Helicobacter pylori, rationalizing the utility and efficacy of Absolute stereostructures and syntheses of saussureamines A, B, C, the traditional Chinese medicines in the treatment of gastric D and E, amino acid-sesquiterpene conjugates with gastroprotective ulcers. effect, from the roots of Saussurea lappa. Tetrahedron 56, 7763–7777. Matsuda, H., Toguchida, I., Ninomiya, K., Kageura, T., Morikawa, T., Yoshikawa, M., 2003. Effects of sesquiterpenes and amino acid-sesquiterpene conjugates from the roots of Saussurea lappa Acknowledgements on inducible nitric oxide synthase and heat shock protein in lipopolysaccharide-activated macrophages. Bioorganic and Medicinal Chemistry 11, 709–715. The work was co-financed by grants from National Natural Megraud, F., Lehn, N., Lind, T., Bayerdorffer,¨ E., O’Morain, C., Spiller, Science Foundation of China (No. 30171104) and from the R., Unge, P., Zanten, S.V.V., Wrangstadh, M., Burman, C.F., 1999. Ministry of Education (No. 104195). Antimicrobial susceptibility testing of Helicobacter pylori in a large multicenter trial: the MACH 2 study. Antimicrobial Agents and Chemotherapy 43, 2747–2752. Ohsaki, A., Takashima, J., Chiba, N., Kawamura, M., 1999. Micro- References analysis of a selective potent anti-Helicobacter pylori compound in a Brazilian medicinal plant, Myroxylon peruiferum and the activity Adamek, R.J., Suerbaum, S., Pfaffenbach, B., Opferkuch, W., 1998. Pri- of analogues. Bioorganic and Medicinal Chemistry Letters 9, 1109– mary and acquired Helicobacter pylori resistance to clarithromycin, 1112. Y. Li et al. / Journal of Ethnopharmacology 98 (2005) 329–333 333

Ohta, R., Yamada, N., Kaneko, H., Ishikawa, K., Fukuda, H., Fujino, T., Wang, B.Y., 1997. Modern Traditional Medicine Pharmacology. Tianjin Suzuki, A., 1999. In vitro inhibition of the growth of Helicobacter Scientific and Technological Press, Tianjin, pp. 655. pylori by oil-macerated garlic constituents. Antimicrobial Agents and Xiao, P.G., Li, D.P., Yang, S.L., 2002. Modern Chinese Materia Medica. Chemotherapy 43, 1811–1812. Press, Beijing, pp. 184–188. Park, B.J., Lee, C.K., Park, H.J., 1997. Anti-Helicobacter pylori effect Yang, H., Xie, J.L., Sun, H.D., 1997. Research progress on the medicinal of costunolide isolated from the stem bark of Magnolia sieboldii. plant—Saussurea lappa. Natural Product Research and Development Archives of Pharmacal Research 20, 275–279. 10, 90–98. Pattnaik, S., Subramanyam, V.R., Bapaji, M., Kole, C.R., 1997. Antibac- Yu, J.G., Hung, F., 1981. Studies on the essential oils of buds and terial and antifungal activity of aromatic constituents of essential oils. clove leaves. Zhong Caoyao 12, 339–342. Microbios 89, 39–46. Journal of Ethnopharmacology 98 (2005) 335–338

Effect of an avocado oil-rich diet over an angiotensin II-induced blood pressure response

M.J. Salazar a, M. El Hafidi b, G. Pastelin a, M.C. Ram´ırez-Ortega a, M.A. Sanchez-Mendoza´ a, ∗

a Instituto Nacional de Cardiolog´ıa “Ignacio Ch´avez”, Department of Pharmacology, Juan Badiano No. 1, Col. Secci´on XVI, Deleg. Tlalpan, Distrito Federal, C.P. 14080, M´exico b Instituto Nacional de Cardiolog´ıa “Ignacio Ch´avez”, Department of Biochemistry, Juan Badiano No. 1, Col. Secci´on XVI, Deleg. Tlalpan, Distrito Federal, C.P. 14080, M´exico Received 1 November 2003; received in revised form 2 January 2005; accepted 19 January 2005

Abstract

We studied the effect of an avocado oil-rich diet on (1) the blood pressure response to angiotensin II (AngII) and (2) the fatty acid composition of cardiac and renal membranes on male Wistar rats. The avocado oil-rich diet induced a slightly higher AngII-induced blood pressure response in the rats as compared to the control rats. In cardiac microsomes, avocado oil induced an increase in oleic acid content (13.18 ± 0.33% versus 15.46 ± 0.59%), while in renal microsomes, the oil decreased ␣-linolenic acid content (0.34 ± 0.02% versus 0.16 ± 0.12%), but increased the arachidonic acid proportion (24.02 ± 0.54% versus 26.25 ± 0.54%), compared to control. In conclusion, avocado oil-rich diet modifies the fatty acid content in cardiac and renal membranes in a tissue-specific manner. The rise in renal arachidonic acid suggests that diet content can be a key factor in vascular responses. © 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Avocado oil; Blood pressure; Membrane fatty acid composition; Arachidonic acid; Angiotensin II

1. Introduction people in countries like Brazil, Indonesia, Jamaica, Nige- ria and Panama also use it for the same therapeutic benefits Persea americana Miller (Lauraceae) or avocado, native (Ross, 1999; Adeboye et al., 1999). to Mexico, is valued due to its nutritional and therapeutic Recent studies carried out among Mexican populations qualities. Avocado fruit and leaves have been used in Mex- that consume avocados have shown that the avocado de- ican folk medicine to treat a wide variety of diseases. Fran- creases serum total cholesterol, LDL-cholesterol and triglyc- cisco Hernandez´ reported as early as the XVI century, that oil erides, and increases HDL-cholesterol levels compared to the obtained from pressing the seed was useful in the treatment control diet (Alvizouri-Munoz et al., 1992; Lopez-Ledesma of rashes and scars, had an astringent effect, and could also et al., 1996). be used to treat dysentery (Argueta-Villamar et al., 1994). The consumption of oil from different sources exerts dif- Hot water infusion from the leaves can be taken as an em- ferent effects over the lipid composition of the cellular mem- menagogue, diuretic, to treat coughs and colds, and diarrhea. branes and their function. Therefore, we studied the effect An important use of avocado leaves is to treat hypertension, of a diet rich in avocado oil and a control diet, on the which is not used exclusively by Mexican populations, but blood pressure response to angiotensin II (AngII). In ad- dition, we evaluated the effect of an avocado oil diet on ∗ Corresponding author. Tel.: +52 55 5573 2911x1317; fatty acid composition of cardiac and renal membranes in fax: +52 55 5573 0926. order to correlate biochemical changes and physiological E-mail address: [email protected] (M.A. Sanchez-Mendoza).´ responses.

0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2005.01.044 336 M.J. Salazar et al. / Journal of Ethnopharmacology 98 (2005) 335–338

2. Materials and methods 2.5. Statistical analysis

2.1. Experimental animals Data were expressed as mean ± S.E.M. Statistical eval- uation of the data was performed using Student’s t-test for Male Wistar rats (230–250 g), bred and raised in our fa- unpaired comparisons; p < 0.05 was considered statistically cilities, were placed into metabolic cages 3 days prior to the significant. beginning of the protocol, offered tap water and lab chow ad libitum, and maintained on a 12-h light:12-h dark cycle in a temperature-controlled room. Animals were randomly di- 3. Results vided into two experimental groups of five rats each. The con- trol group received lab chow, while the treated group received 3.1. Effect on body weight a 10% (w/w) avocado oil-rich diet for a 2-week period. At the end of the treatment, the rats were anaesthetized with sodium Two weeks of an avocado oil-rich diet had no significant pentobarbital (50 mg/kg, i.p.) and were either prepared for influence on the rat’s body weight. Body weight of control blood pressure measurement, or the heart and kidneys were rats was 247.5 ± 1.4 g, while in the avocado oil-rich diet was removed. All procedures were conducted in accordance with 249.1 ± 1.7 g. Institutional ethical guidelines. 3.2. Effect on AngII-induced change in blood 2.2. Diet preparation pressure

The avocado-enriched diet was prepared in bulk in our lab- Basal systolic blood pressure for control and avocado oil- oratory by mixing ground lab chow with avocado oil obtained rich diet rats was: 95 ± 3.1 and 97 ± 2.6 mmHg, respectively. from fresh fruit (10%, w/w). Both the control diet (ground lab The administration of AngII (100, 300, and 1000 ng/kg, i.v.) chow) and avocado oil-rich diet were partitioned into daily induced an increase in blood pressure in both control rats and rations packaged in plastic bags, and flushed with nitrogen to ◦ rats fed with avocado oil-rich diet; however, this increment minimize oxidation and stored at 4 C. was slightly higher in the rats fed with avocado oil-rich diet, as compared to the control rats (Fig. 1). 2.3. Fatty acid methyl esters analysis 3.3. Effect on fatty acid composition Cardiac and renal microsomes were obtained as described by Garg and co-workers (1988) in a homogenizing buffer con- Avocado oil-rich diet induced an increase in oleic acid pro- taining: 250 mM sucrose, 0.1 mM etilenediamino tetraacetic portion in cardiac microsomes (15.46 ± 0.5% of total) com- acid (EDTA), 62 mM potassium phosphate, 150 mM potas- pared to those from control rats (13.18 ± 0.3%, p < 0.05). In sium chloride, 5 mM magnesium chloride, and 1 mM dithio- renal microsomes, avocado oil-rich diet elicited a decrease in threitol (DTT), at pH 7.4. The microsomes, containing buty- ␣-linolenic acid (0.34 ± 0.02 and 0.16 ± 0.12% of total, for lated hydroxy (BHT, 0.02%), were stored at −70 ◦C control and avocado oil-treated rats, respectively, p < 0.05); as until processed. The lipids were extracted as described by well as an increase in arachidonic acid proportion: for control, Folch et al. (1957). The lipid extracts were trans-esterified to their fatty acid methyl esters as described by Christie (1989), separated and identified by gas–liquid chromatog- raphy in a Carlo Erba Fratovap 2300 chromatograph, fit- ted with a 25 m × 0.25 mm i.d. fussed-silica capillary col- umn, coated with CP-Sil 88 (film thickness, 0.25 ␮m). The analysis was carried out at an isotherm temperature of 195 ◦C, using helium gas as a carrier at a flow rate of 1 ml/min.

2.4. Angiotensin II-induced increase of blood pressure

At the end of the 2-week treatment, both control rats and rats fed with avocado oil-rich diet were anaesthetized and we performed intra-artery measurement of the blood pressure as described by Adeboye et al. (1999). AngII (100, 300, and Fig. 1. Effect of angiotensin II-stimulation on blood pressure response in rats fed a control and avocado oil-rich diet. AngII (100, 300, and 1000 ng/kg, i.v.) 1000 ng/kg, i.v.) was administered and the basal- and AngII- was administered to rats fed control diet (), and avocado oil-rich diet (10%, induced blood pressure changes were measured with a Blood w/w, ) for 2 weeks. The bars represent the mean ± S.E. of five different Pressure monitor (BP Monitor, WPI, USA). experiments. M.J. Salazar et al. / Journal of Ethnopharmacology 98 (2005) 335–338 337

Table 1 Effect of an avocado oil-rich diet over the fatty acid composition of cardiac and renal microsomes Fatty acid Heart Kidney

Control (% of total) Avocado oil (% of total) Control (% of total) Avocado oil (% of total) Palmitic 19.36 ± 0.11 18.54 ± 0.44 26.55 ± 0.48 25.27 ± 0.44 Palmitoleic 1.41 ± 0.38 1.83 ± 0.61 1.40 ± 0.20 1.03 ± 0.16 Margaric 0.95 ± 0.29 0.46 ± 0.04 0.29 ± 0.10 0.51 ± 0.05 Stearic 23.90 ± 1.09 23.62 ± 0.62 18.25 ± 0.58 19.53 ± 0.23 Oleic 13.18 ± 0.33 15.46 ± 0.59* 13.61 ± 0.68 11.77 ± 0.80 Linoleic 14.56 ± 0.28 12.79 ± 0.71 10.42 ± 0.23 10.92 ± 0.36 Gamma-linolenic 2.48 ± 1.44 0.86 ± 0.14 1.52 ± 0.18 0.86 ± 0.26 Alfa-linolenic 0.43 ± 0.01 0.83 ± 0.32 0.34 ± 0.02 0.16 ± 0.12* Eicosanoic 1.26 ± 0.45 1.05 ± 0.26 0.95 ± 0.02 0.92 ± 0.06 Di-homo-gamma-linolenic 0.45 ± 0.04 0.41 ± 0.03 0.94 ± 0.07 0.93 ± 0.03 Arachidonic 19.08 ± 1.08 20.98 ± 0.95 24.02 ± 0.54 26.25 ± 0.54* Eicosapentaenoic 0.33 ± 0.01 0.29 ± 0.03 0.81 ± 0.04 0.86 ± 0.25 Docosa-pentaenoic 0.29 ± 0.1 0.31 ± 0.1 0.13 ± 0.03 0.17 ± 0.06 Docosa-hexaenoic 2.11 ± 0.25 2.48 ± 0.13 0.47 ± 0.04 0.53 ± 0.11 The data represent the mean ± S.E. of five different experiments. ∗ p < 0.05.

24.02 ± 0.5%, and for avocado oil-treated rats, 26.25 ± 0.5%; Upon AngII-stimulation, AA is released from the mem- p < 0.05 (Table 1). brane phospholipids and metabolized in a cell/tissue-specific manner (Croft et al., 2000). Since the avocado oil-rich diet increased the renal content of AA, and we observed a higher 4. Discussion and conclusions increase in AngII-induced blood pressure, it is possible to postulate that the production of metabolites with vasocon- In this present study, we found that an avocado oil- strictor properties is increased. rich diet administered for 2 weeks to Wistar rat, in- The use of avocado (fresh fruit and leaves) in folk medicine duced a higher AngII-induced blood pressure response, and has shown to be effective. However, our results show that the modified the fatty acid composition of cardiac and renal favorable effects for the cardiovascular system cannot be at- microsomes. tributed to the components of the oil and suggests that other The fatty acid composition of cardiac membranes is sen- active substances present in fresh fruit and leaves may be re- sitive to modification through dietary factors (Pepe and sponsible. Further studies are required to establish the ratio of McLennan, 2002). It has been reported that the level and hypocholesterolemic/cardiovascular benefits, when avocado nature of polyunsaturated fatty acids (PUFA) incorporated oil is consumed as well as the component(s) responsible. In in cardiac membranes is related to the dietary n-3 PUFA. In conclusion, avocado oil-rich diet modifies the fatty acid con- their data, Sergiel et al. (1998) reported that rats given a diet tent in cardiac and renal membranes in a tissue-specific man- rich in docosahexaenoic acid (DHA) had a decrease in the ner. The rise in renal AA content suggests that diet is a key arachidonic acid content of the cardiac membranes, and that factor in vascular responses. DHA constituted almost the only membrane n-3 PUFA, sug- gesting a very low level of retroconversion. Our data show that avocado oil induces an increase in the proportion of oleic Acknowledgment acid in the heart microsomes and thus can modify the struc- ture and function of the membrane, including the biosynthesis The study was supported by grant (SEP-2003-C02-44173) of prostaglandins involved in the regulation of the vascular of Consejo Nacional de Ciencia y Tecnolog´ıa (CONACyT), function. Indeed, oleic acid has been shown to exert a pressor Mexico.´ effect in rats, when administered intravenously (Grekin et al., 1995). However, this pattern is different in renal microsomes, where we found not only a lack of differences in oleic acid References content, but a decrease in ␣-linolenic acid and an increase in arachidonic acid content. The tendency to a higher increase in Adeboye, J.O., Fajanyomi, M.O., Makinde, J.M., Taiwo, O.B., 1999. A Ang II-induced blood pressure may be partially explained by preliminary study on the hypotensive activity of Persea americana leaf these renal modifications. It has been reported that enhanced extracts in anaesthetized normotensive rats. Fitoterapia 70, 15–20. ␣ Alvizouri-Munoz, M., Carranza-Madrigal, J., Herrera-Abarca, J.E., dietary intake of -linolenic acid decreased blood pressure in Chavez-Carbajal, F., Amezcua-Gastelum, J.L., 1992. Effects of av- spontaneously hypertensive rats and increased prostaglandin ocado as a source of monounsaturated fatty acids on plasma lipid formation (Rupp et al., 1996). levels. Archives of Medical Research 23, 163–167. 338 M.J. Salazar et al. / Journal of Ethnopharmacology 98 (2005) 335–338

Argueta-Villamar, A., Cano, L., Rodarte, M., 1994. Atlas de las Plantas de Lopez-Ledesma, R., Frati-Munari, A.C., Hernandez-Dominguez,´ B.C., la Medicina Tradicional Mexicana. Ed. Instituto Nacional Indigenista, Cervantes-Montalvo, S., Hernandez-Luna,´ M.H., Juarez,´ C., Moran- Mexico,´ p. 55. Lira, S., 1996. Monounsaturated fatty acid (avocado) rich diet for Christie, W.W., 1989. The preparation of derivatives of fatty acids. In: mild hypercholesterolemia. Archives of Medical Research 27, 519– Christie, W.W. (Ed.), Gas Chromatography and Lipids. The Oily Press, 523. Ayr, UK, pp. 64–84. Pepe, S., McLennan, P.L., 2002. Cardiac membrane fatty acid composition Croft, K.D., McGiff, J.C., Sanchez-Mendoza, A., Carroll, M.A., 2000. modulates myocardial oxygen consumption and postischemic recovery Angiotensin II releases 20-HETE from rat renal microvessels. Amer- of contractile function. Circulation 105, 2303–2308. ican Journal of Physiology 279, F544–F551. Ross, I.A., 1999. Medicinal Plants of the World. Humana Press Inc., Folch, L., Lees, M., Sloane-Stanley, C.H., 1957. A simple method for the Totowa, NJ, p. 241. isolation and purification of total lipid from animal tissues. Journal Rupp, H., Turcani, M., Ohkubo, T., Maisch, B., Brilla, C.G., 1996. Dietary of Biological Chemistry 226, 497–509. linolenic acid-mediated increase in vascular prostacyclin formation. Garg, M.L., Sebokova, E., Thomson, A.B.R., Clandin, M.T., 1988. Delta Molecular Cell Biochemistry 162, 59–64. 6-desaturase activity in liver microsomes of rats fed diets enriched Sergiel, J.P., Martinez, L., Raederstorff, D., Grynberg, A., Demaison, L., with cholesterol and/or w-3 fatty acids. Biochemical Journal 249, 1998. Individual effects of dietary EPA and DHA on the functioning 351–356. of the isolated working rat heart. Canadian Journal of Physiology and Grekin, R.J., Vollmer, A.P., Sider, R.S., 1995. Pressor effects of portal ve- Pharmacology 76, 728–736. nous oleate infusion. A proposed mechanism for obesity hypertension. Hypertension 26, 193–198. Journal of Ethnopharmacology 98 (2005) 339–343

Inhibitory effect of jaceosidin isolated from Artemisia argyi on the function of E6 and E7 oncoproteins of HPV 16

Hee-Gu Lee a,1, Kyung-Ae Yu a,1, Won-Keun Oh b, Tae-Woong Baeg a, Hyun-Cheol Oh b, Jong-Seog Ahn b, Won-Cheoul Jang c, Jong-Wan Kim a,d, Jong-Seok Lim a, Yong-Kyung Choe a, Do-Young Yoon a,∗

a Laboratory of Cell Biology, Korea Research Institute of Bioscience and Biotechnology, Yuseong P.O. Box 115, Daejeon 305-600, Korea b Laboratory of Cellular Signaling Modulator, Korea Research Institute of Bioscience and Biotechnology, Yuseong P.O. Box 115, Daejeon 305-600, Korea c Department of Chemistry, DanKook University, 16-5 AnSeo Dong, CheonAn 330-714, Korea d Department of Laboratory Medicine, DanKook University, 16-5 AnSeo Dong, CheonAn 330-714, Korea Received 18 March 2004; received in revised form 5 January 2005; accepted 26 January 2005

Abstract

Jaceosidin (4,5,7-trihydroxy-3,6-dimethoxyflavone) was isolated from Artemisia argyi as a putative oncogene inhibitor. Jaceosidin inhibited binding between oncoprotein E6 of the human papillomavirus and the p53 tumor suppressor protein. In addition, jaceosidin inhibited binding between the E7 oncoprotein and the Rb tumor suppressor protein, and also inhibited the function of HPV-16 harboring cervical cancer cells, including SiHa and CaSki. Collectively, jaceosidin inhibited the functions of the E6 and E7 oncoproteins of the human papillomavirus, suggesting that this compound might be used as a potential drug for the treatment of cervical cancers associated with the human papillomavirus. © 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Cervical cancer; HPV; E6; E7; Artermisia argyi; Jaceosidin

1. Introduction protein) gene are uniformly retained and regularly expressed in tumors, underlining their importance in the development The Artemisia consists of more than 350 species, or maintenance of neoplastic phenotypes (Woodworth et al., and the aerial portions of the plant have been used in tradi- 1992). The E6 and E7 proteins have been shown to interact tional medicine for the treatment of prickly heat and jaun- specifically with the p53 and pRb tumor suppressor proteins, dice, as well as for treating uterine metrorrhagia and metritis respectively (Scheffner et al., 1991). Both pRb and p53 nega- in Korea. Human papillomaviruses (HPVs) are small epithe- tively regulate the cell cycle, and also appear to inhibit G0–G1 liotropic DNA viruses, which are involved, in several human and G1–S phase transitions. These interactions apparently malignancies. HPVs are major risk factors associated with the play important roles in the induction of cell immortality. The development of cervical carcinomas (Wright and Sun, 1996). importance of p53-mediated apoptosis has been recognized The HPV E6 and E7 oncogenes and the E6AP (E6-associated in terms of the maintenance of homeostasis, as well as the prevention of neoplastic transformation. Above all, p53 is a target of several oncoproteins, which are encoded by DNA tu- Abbreviations: HPV, human papillomavirus; E6AP, E6-associated pro- mor viruses (Kessis et al., 1993). E6 forms a ternary complex tein; GST, glutathione S-transferase; ELISA, enzyme-linked immunosorbent with p53 and E6AP, resulting in the degradation of p53 via assay; His, histidine ∗ Corresponding author. Tel.: +82 42 860 4218; fax: +82 42 860 4593. the ubiquitination pathway (Hubbert et al., 1992). pRb is nor- E-mail address: [email protected] (D.-Y. Yoon). mally activated when it is hypophosphorylated, and thus, neg- 1 These authors contributed equally to this work. atively regulates G1–S cell cycle phase transitions, resulting

0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2005.01.054 340 H.-G. Lee et al. / Journal of Ethnopharmacology 98 (2005) 339–343 in the prevention of the cell cycle from entering S phase. subjected to silica gel chromatography (Silica gel, Kiesegel Therefore, the loss of pRb function should remove the pro- 60, 0.063–0.2 mm, MERCK Co.) via elution with a mixture liferation block, thus indirectly stimulating cell proliferation. of hexane, ethylacetate, chloroform, and MeOH. Fractions of The E7 protein binds to pRb, and facilitates its dissociation similar composition, as determined by TLC analysis, were from the E2F transcriptional activator, which is associated pooled. The active fraction was further chromatographed on with cell proliferation and immortalization (Pagano et al., ODS-18 resin (ODS-A 120A, S-150 ␮m, YMC Inc.), eluting 1992). Both E6 and E7 induce DNA synthesis in growth- with a stepwise gradient from 40 to 100% MeOH, in H2O. arrested cells, thus extending cellular lifespans. Collectively, The fraction eluted with 70% MeOH exhibited inhibitory ac- the E6 protein promotes p53 degradation via complex forma- tivity on ELISA systems, based on binding between oncopro- tion, and E7 releases E2F from pRb, thus promoting mitotic teins and tumor suppressors, as described below. Finally, the cell division. Approximately half a million women die annu- active fraction was purified by preparative HPLC (J’sphere ally from cervical cancer, and the possibility of preventing ODS-H80, 150 mm × 20 mm i.d., S-4 ␮m, 80 A,˚ detection at HPV infection by vaccination is now being explored, in an 254 nm) using 75% MeOH to yield jaceosidin (40 mg). effort to reduce cervical cancer mortality rates (Koutsky et al., 2002). In this study, we attempted to isolate natural prod- 2.4. Overexpression and purification of E6, E6AP, E7 ucts, which could inhibit individual HPV infections via the and pRb disruption of the functions of viral gene products. This work describes the isolation of jaceosidin from the MeOH extract Fresh E6, E6AP, and E7 were prepared with pGEX vector of Artemisia argyi Levi. et Vant., as guided by observed in- (Amersham Pharmacia Biotech AB) and pET vector (No- hibitory effects on binding between oncoproteins and tumor vagen, Madison, USA), as described in our previous reports suppressors. In addition, we evaluated the inhibitory effects (Cho et al., 2000, 2001). pRb tumor suppressor protein was of jaceosidin on the cell viability of HPV positive cervical expressed in pET3 vector, and was constructed to improve cancer cell lines (Beerheide et al., 1999). solubility in a supernatant fraction, resulting in the produc- tion of a truncated protein composed of an N-terminal and a pocket domain (aa 373–928), as described in previous re- 2. Materials and methods ports (Cho et al., 2001). To express recombinant proteins, pET or pGEX recombinant vectors were transformed into 2.1. General DH5␣. The other vector, pET28a, was expressed in BL21 (DE3). These recombinant proteins were induced and puri- The following were purchased: organic solvents (Ducksan fied for use, as described in previous reports (Cho et al., 2000, Chemical Co., Daejeon, Korea); cisplatin (Sigma, Deisen- 2001). hofen, Germany); the analytical and preparative HPLC col- umn (J’sphere ODSH80, 150 mm × 20 mm, S-4 ␮m, 80A,˚ 2.5. Effects of jaceosidin on binding between YMC Inc., Kyoto, Japan); HPLC solvents (Fisher Scientific, oncoproteins and tumor suppressors Pittsburgh, PA); TLC: pre-coated silica gel plate (Silica gel 60F254, 0.5 mm) and pre-coated RP plate (RP-18F254, 25DC- The purified E6, E6AP, and E7 proteins were each coated platten, Merck Co., Darmstadt, Germany); Maxisorb 96-well onto Maxisorb 96-well plates (Nunc, Netherlands), at a final plates (Nunc, Netherlands); Ni2+-NTA column (Peptron Inc., concentration of 4 ␮g/ml. These plates were blocked with Daejeon, Korea) and glutathione sepharose 4B (Amersham PBS containing 3% skimmed milk for 2 h at room tempera- Pharmacia Biotech AB, Uppsala, Sweden). ture, in order to preclude non-specific binding. Jaceosidin and p53 lysate were added to plates, which had been pre-coated 2.2. Plant materials with E6 or E6AP. Jaceosidin and pRb lysate were added to E7 pre-coated plates as described (Cho et al., 2001). After Artemisia argyi leaves were purchased from the herbal 1 h of incubation, the respective plates were washed three medicine cooperative association of Daejeon, Korea, in Oc- times with PBS containing 0.05% Tween-20 (PBST). E6 or tober 2001. A voucher specimen (No. 143) was deposited in E6AP coated plates were treated with p53 antibody (Onco- our laboratory. The authenticity of the plant was confirmed gene, Cambridge, MA), and E7-coated plates were treated by Dr. Hyeong-Kyu Lee, a manager at the Plant Extract Bank, with pRb antibody (Ab-6) (Oncogene) for 1 h. Horseradish Plant Diversity Research Center, KRIBB. peroxidase-conjugated secondary antibody was then added to the plates, after an additional washing with PBST. Af- 2.3. Isolation of jaceosidin from Artemisia argyi ter 30 min of further incubation, 100 ␮l of substrates (4 mg, o-phenylenediamine 5 ␮l, 37% H2O2/10 ml of 0.1 M citrate Dried Artemisia argyi (2 kg) was extracted twice with buffer, pH 5.1) were added. The enzyme reaction was stopped MeOH at room temperature. The evaporated MeOH extract with 50 ␮lof2NH2SO4, and the plate was measured at (250 g) was then dissolved in water, and partitioned with 490 nm with an ELISA reader (Molecular Devices, Sunny- EtOAc. After concentration, the EtOAc-soluble extract was vale, CA). H.-G. Lee et al. / Journal of Ethnopharmacology 98 (2005) 339–343 341

2.6. Cytotoxic effect of jaceosidin on HPV (+) cervical carcinomas and HPV (−) control cell lines

Based on the cleavage of tetrazolium salt WST-1 (Boehringer Mannheim, Germany) by mitochondrial dehy- drogenases in viable cells, a colorimetric assay for cell via- bility was employed. CaSki, SiHa, HeLa, C33A, HaCaT, and C3 were used as cell lines in the viability inhibition test. HPV 16-positive cervical carcinoma cell lines, SiHa, and CaSki, were purchased from the American Type Culture Collection (Rockville, MD). C33A is an HPV (−) cervical carcinoma cell line. It has been established that the SiHa and CaSki cell lines contain 1,2 copies and 60–600 copies of the HPV 16 genome, respectively (Baker et al., 1987; Meissner, 1999). Fig. 1. The structure of jaceosidin. These cell lines were grown in DMEM containing 100 U/ml penicillin–streptomycin, 25 ng/ml amphotericin B, and 10% FBS. A humidified incubator was maintained at 37 ◦C and E7 should be inhibited in order to prevent HPV infec- tion. 5% CO2. These cell lines were then seeded in 96-well plates, at a concentration of 1 × 105/100 ␮l/well. After an overnight In order to isolate the HPV oncogene inhibitor from incubation, the culture medium on the plates was exchanged Artemisia argyi, the MeOH extract, which exhibited in- for fresh medium. Various concentrations of jaceosidin, or hibitory activity on ELISA systems based on binding be- the known anti-cancer agent, cisplatin, were subsequently tween oncoproteins and tumor suppressors, was further inves- added. After another 20 h of incubation, 10 ␮l of cell prolif- tigated. The ELISA assay guided fractionation, accomplished eration reagent WST-1 was added to the plates. After more by a variety of chromatographic techniques, led to the iso- incubation for 0.5 to 4 h, the plates were shaken thoroughly lation of the active component. This active component was for 1 min with a shaker. The absorbance was measured with identified as jaceosidin (Fig. 1), by comparing NMR and MS an ELISA reader (Molecular Devices) at 490 nm. data results with those in the literature (Tomas-Barberan et al., 1987; Yadava and Saurabh, 1998; Nakasugi et al., 2000). In order to evaluate the inhibitory effects of jaceosidin on 2.7. Statistical analysis the function of the E6 and E7 oncoproteins of HPV,we evalu- ated the effects of jaceosidin on the binding between viral on- All data were expressed as the means ± S.E.M. These data coproteins (E6 and E7) and tumor suppressor proteins (p53, were analyzed by one-way ANOVA using SPSS/PC(+). Post Rb). In the ELISA-based binding assay, jaceosidin exhibited hoc comparisons were assessed using Tukey’s method. inhibitory effects on the binding between E6 and p53, as well the binding between E6AP and p53 (Fig. 2). Jaceosidin also 3. Results and discussion inhibited binding between E7 and Rb. Cisplatin, a potent in- ducer of growth arrest and/or apoptosis in most cell types, is HPV E6 and E7 oncogenes play critical roles in car- among the most effective and widely used chemotherapeutic cinogenesis of the cervix. Thus, the functions of E6 and agents employed in the treatment of human cancers (Wang et

Fig. 2. The effects of jaceosidin on ELISAs, based on binding between oncoproteins and tumor suppressors (p53, and Rb). E6, E6AP, and E7 oncoproteins were coated onto a Maxisorb 96-well plate. Various concentrations of jaceosidin and lysate (p53 or pRb) were added to the blocked plates. E6 or E6AP coated plates were treated with p53 antibody, and E7 coated plates were coated with anti-pRb antibody for 1 h, respectively. Horseradish peroxidase-conjugated secondary antibody was then added to the plate, after a wash with PBST. These plates were treated with p53 antibody or pRb antibody. Horseradish peroxidase-conjugated anti-mouse IgG secondary antibody was also added to the plates. After the addition of substrate and developing, the plate was examined at 490 nm by an ELISA reader. Data are representative of three independent experiments performed in duplicate. 342 H.-G. Lee et al. / Journal of Ethnopharmacology 98 (2005) 339–343

Fig. 3. The cytotoxic effects of jaceosidin on HPV-containing or non-HPV-containing cervical carcinoma cell lines, and HPV-negative keratinocytes. Cells were seeded in 96-well plates at a concentration of 1 × 105/100 ␮l/well. After 1 day incubation, jaceosidin or the anti-cancer agent, cisplatin, were added, followed by 20 h of incubation, at which time, 10 ␮l of WST-1 was dispensed into these plates, and the absorbance was measured with an ELISA reader. Vehicle 70% MeOH was used as a control solvent. Cytotoxicity is represented as the mean value (mean ± S.E.M) of at least three independent experiments. Data were represented from three independent experiments performed at least in duplicate; *p < 0.05 vs. untreated control (ANOVA followed by Tukey test). al., 2000). Cisplatin effectively inhibited binding between E7 immortalized cell lines containing HPV 16 (Fig. 3). The in- and Rb, and also inhibited binding between E6 (or E6AP) and hibition of these interactions between oncogenes and tumor the p53 tumor suppressor (Fig. 2). Collectively, jaceosidin suppressors appears to result in the recovery of the functions proved able to attenuate the binding activity between onco- of the p53 and pRb tumor suppressors. proteins and tumor suppressors in a dose-dependent manner. Some previous studies have reported the existence of com- Therefore, jaceosidin may be considered a putative oncogene pounds, which interfere with the binding of zinc to E6 and E7 inhibitor. (Beerheide et al., 1999). Such research has been undertaken The HPV E6 and E7 oncogenes are required for the contin- in order to develop potential drugs against cervical cancer uous growth of tumor cell lines containing HPV. Jaceosidin, arising from HPV infection. If both jaceosidin and inhibitory as a putative inhibitor of the E6 and E7 oncoproteins, may not compounds for zinc binding were applied during the early only interfere with the HPV life cycle, but also arrest HPV stages of HPV infection, HPV might be effectively inhibited. 16-associated malignancies. As jaceosidin had clearly ex- Methods of effective HPV prevention, involving vaccination hibited inhibitory effects on the interactions between E6 and and immune therapy, are currently being developed. Most p53, and between E7 and pRb, we attempted to determine cervical cancer patients have, thus far, been treated with sur- whether it would also have an effect on cervical cancer cell gical techniques. As an alternative approach, cervical cancer lines. Therefore, we administered jaceosidin to the HPV 16- may be treatable with HPV oncogene inhibitors, including containing cervical type cell lines, SiHa and CaSki. HaCaT jaceosidin. It would first, however, be necessary to assess cells were also used as an immortalized human skin epithelial the inhibitory effects and side effects of jaceosidin on HPV- cell line, which did not contain HPV. In addition, we used the containing mouse models, and to adequately delineate the HPV 18 type cell line, HeLa. C33A is a cervical carcinoma role and functions of jaceosidin as a putative anti-oncogenic cell, which contain no HPV. Jaceosidin’s inhibitory effects or cancer chemopreventive agent. on cell viability were evaluated by its addition to the HPV- containing cervical carcinomas and non-HPV-containing cell lines (Fig. 3). In this set of assays, jaceosidin was found to Acknowledgements specifically inhibit HPV 16-positive cervical carcinoma cell lines SiHa and CaSki, in a dose-responsive manner (Fig. 3). This work was supported by the Plant Diversity Re- Little or no inhibition, however, was observed in the HPV- search Center of the 21st Century Frontier Research Pro- negative HaCaT and HPV 18-positive HeLa cells, suggesting gram (M103KD010034-03K0401-03410) from the Ministry that jaceosidin was specific to the E6 and E7 oncogenes of of Science and Technology of Korea and partially by the HPV type 16. Cisplatin was also added into cervical cancer Biotechnology Program of Kyung-Nam Province. cells, as a control anti-cancer agent. The viability of cervical cancer cells was also inhibited (Fig. 3). Jaceosidin effectively inhibited the viability of CaSki cells, which harbor a larger References amount of HPV 16 genomes (60–600 copies) than do other cervical cells. Our results suggest that jaceosidin abrogated Baker, C.C., Phelps, W.C., Lindgren, V., Braun, M.J., Gonda, M.A., How- the function of HPV 16 types, via the functional inhibition of ley, P.M., 1987. Structural and transcriptional analysis of human papil- lomavirus type 16 sequences in cervical carcinoma cell lines. Journal the E6 and E7 oncoproteins. In summary, jaceosidin inhib- of Virology 61, 962–971. ited interactions between oncoproteins (E6 and E7) and tumor Beerheide, W., Bernard, H.U., Tan, Y.J., Ganesan, A., Rice, W.G., Ting, suppressors (p53, pRb) (Fig. 2), and decreased the growth of A.E., 1999. Potential drugs against cervical cancer; zinc-ejecting in- H.-G. Lee et al. / Journal of Ethnopharmacology 98 (2005) 339–343 343

hibitors of the human papillomavirus type 16 E6 oncoprotein. Journal Nakasugi, T., Nakashima, M., Komai, K., 2000. Antimutagens in gaiyou of the National Cancer Institute 91, 1211–1220. (Artemisia argyi Levi. et Vant.). Journal of Agricultural and Food Cho, Y.S., Cho, C.W., Ok, J., Lee, K.A., Park, S.N., Yoon, D.Y., Chemistry 48, 3256–3266. 2000. Development of screening system for drugs against human Pagano, M., Durst, M., Joswig, S., Draetta, G., Jansen-Durr, P., 1992. papillomavirus-associated cervical cancer: based on E6-E6AP bind- Binding of the human E2F transcription factor to the retinoblastoma ing. Antiviral Research 47, 199–206. protein but not to cyclin A is abolished in HPV-16-immortalized cells. Cho, Y.S., Cho, C.W., Ok, J., Lee, K.A., Shim, J.H., Park, S.N., Yoon, Oncogene 7, 1681–1686. D.Y., 2001. Development of screening system for drugs against Scheffner, M., Munger, K., Byrne, J.C., Howley, P.M., 1991. The state of human papillomvirus-associated cervical cancer: based on E7-Rb the p53 and retinoblastoma genes in human cervical carcinoma cell binding. Journal of Biochemistry and Molecular Biology 34, 80– lines. In: Proceedings of the National Academy of Sciences of the 86. United States of America, vol. 88, pp. 5523–5527. Hubbert, N.L., Sedman, S.A., Schiller, J.T., 1992. Human papillomavirus Tomas-Barberan, F.A., Harborne, J.B., Self, R., 1987. Twelve 6- type 16 E6 increases the degradation rate of p53 in human ker- oxygenated flavone sulphates from Lippia nodiflora and L. canescens. atinocytes. Journal of Virology 66, 6237–6241. Phytochemistry 26, 2281–2284. Kessis, T.D., Slebos, R.J., Nelson, W.G., Kastan, M.B., Plunkett, B.S., Wang, X., Martindale, J.M., Holbrook, N.J., 2000. Requirement for ERK Han, S.M., Lorincz, A.T., Hedrick, L., Cho, K.R., 1993. Human pa- activation in cisplatin-induced apoptosis. Journal of Biological Chem- pillomavirus 16 E6 expression disrupts the p53-mediated cellular re- istry 275, 39435–39443. sponse to DNA damage. In: Proceedings of the National Academy Woodworth, C.D., Lichti, U., Simpson, S., Evans, C.H., DiPaolo, J.A., of Sciences of the United States of America, vol. 90, pp. 3988– 1992. Recombinant retroviruses encoding human papillomavirus type 3992. 18 E6 and E7 genes stimulate proliferation and delay differentiation Koutsky, L.A., Ault, K.A., Wheeler, C.M., Brown, D.R., Barr, E., Alvarez, of human keratinocytes early after infection. Oncogene 7, 619–626. F.B., Chiacchierini, L.M., Jansen, K.U., 2002. A controlled trial of Wright, T.C., Sun, X.W., 1996. Anogenital papillomavirus infection and a human papillomavirus type 16 vaccine. New England Journal of neoplasia in immuno-deficient women. Obstetrics and Gynecology Medicine 347, 1645–1651. Clinics of North America 23, 861–893. Meissner, J.D., 1999. Nucleotide sequences and further characterization Yadava, R.N., Saurabh, K., 1998. A new flavone glycoside: 5,7,4- of human papillomavirus DNA present in the CaSki, SiHa and HeLa trihydroxy-6,3-dimethoxy-flavone 5-o-alpha-l-rhamnopyranoside cervical carcinoma cell lines. Journal of General Virology 8, 1725– from the leaves tridax procumbens Linn. Journal of Asian Natural 1733. Products Research 1, 147–152. Journal of Ethnopharmacology 98 (2005) 345–350

Hypolipidemic activity of aqueous extract of Capparis spinosa L. in normal and diabetic rats

M. Eddouks a,∗, A. Lemhadri a, J.-B. Michel b,1

a UFR PNPE, B.P. 21, Boutalamine, Errachidia 52000, Morocco b Inserm U460, CHU X. Bichat, 16 rue Henri Huchard, 75018 Paris, France Received 24 May 2004; received in revised form 13 December 2004; accepted 26 January 2005

Abstract

The purpose of this study was to examine the effect of single and repeated oral administrations of the aqueous extract of Capparis spinosa L. (CS) at a dose of 20 mg/kg on lipid metabolism in normal and streptozotocin-induced diabetic rats. In normal rats, the aqueous extract of CS induced a significant decrease on plasma triglycerides concentrations 1 week (p < 0.05) and 2 weeks (p < 0.01) after once daily repeated oral administration. A significant decrease of plasma cholesterol levels was also observed 4 days (p < 0.05) and 1 week (p < 0.05) after repeated oral administration. In diabetic rats, CS treatment caused a significant decrease of plasma triglycerides levels after repeated oral administration. Four days after repeated oral administration of aqueous CS extract, the plasma cholesterol levels were significantly decreased (p < 0.05) and still dropped after 2 weeks (p < 0.01). On the other hand, the repeated oral administration of CS aqueous extract caused a significant decrease of body weight 4 days after repeated oral treatment in diabetic rats (p < 0.05). We conclude that the aqueous extract of CS (20 mg/kg) exhibits a potent lipid lowering activity in both normal and severe hyperglycemic rats after repeated oral administration of CS aqueous extract. © 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Capparis spinosa L.; Hypolipidemic; Cholesterol; Triglycerides; Body weight; Aqueous extract; Oral administration

1. Introduction Despite the remarkable progress in the management of diabetes mellitus by synthetic drugs, there has been a re- Diabetes mellitus is a chronic disease caused by inher- newed interest in medicinal plants attributed with therapeu- ited and/or acquired deficiency in production of insulin by tic virtues. Morocco has a rich heritage of medicinal plants the pancreas and/or by the ineffectiveness of the insulin pro- of wide diversity, which are used by the local population and duced. Such a deficiency results in increased plasma glucose, the traditional healers for the treatment of several diseases in- which in turn damages many of the body’s systems in partic- cluding diabetes (Ziyyat et al., 1997; Eddouks et al., 2002). ular the blood vessels and nerves. Besides hyperglycaemia, For cultural and economic reasons, the local pharmacopoeia the levels of plasma lipids are usually raised in diabetes mel- continues to be considered as an important source of remedies litus causing a risk factor for coronary heart disease (Kannel for primary healthcare. Capparis spinosa L. (CS) (Cappari- and Mc Gee, 1979). Hypertriglyceridemia is also related to daceae), locally known as “Kebbar” is a native shrub widely insulin resistance and glucose intolerance (Gingsberg, 1994). distributed throughout the south-eastern region of Morocco (Tafilalet). This plant is traditionally used in diabetes con- trol and treatment according to our previous ethnopharma- cological surveys in two great areas of Morocco, Tafilalet ∗ Corresponding author. Tel.: +212 55 57 44 97; fax: +212 55 57 44 85. E-mail addresses: [email protected] (M. Eddouks), and Fez-Boulemane regions (Jouad et al., 2001; Eddouks et [email protected] (J.-B. Michel). al., 2002). CS is used in phytomedicine around the world 1 Tel.: +33 1 44 85 61 60; fax: +33 1 44 85 61 57. as anti-oxidative (Germano et al., 2002), antifungal (Ali-

0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2005.01.053 346 M. Eddouks et al. / Journal of Ethnopharmacology 98 (2005) 345–350

Shtayeh and Abu Ghdeib, 1999), antihepatotoxic (Gadgoli 2001, and air-dried at 40 ◦C. The plant was previously iden- and Mishra, 1999), anti-inflammatory (Al-Said et al., 1988) tified and authenticated by Pr. M. Rejdali (Agronomy and and anti-diabetic (Yaniv et al., 1987; Ziyyat et al., 1997). Sev- Veterinary Institute, Rabat) and a voucher specimen (ME 60) eral medicinal plants have been reported in the scientific liter- (Eddouks et al., 2002) was deposited at the herbarium of the ature to possess significant hypolipidemic activities in animal Faculty of Sciences and Techniques Errachidia. models of diabetes (Ahmed et al., 2001; Sharma et al., 2003). In diabetes phytotherapy, the effect of aqueous CS extract on 2.2. Preparation of the aqueous extract lipid metabolism has never been demonstrated experimen- tally. In the south-eastern region of Morocco (Tafilalet), CS Plant material was prepared according to the traditional fruits are recognized as potent hypoglycaemic agents by sev- method used in Morocco (decoction): 1 g of powdered fruits eral traditional healers (Eddouks et al., 2002). mixed with 100 ml distilled water were boiled for 10 min and The present study was conducted in order to evaluate the then cooled for 15 min. Thereafter, the aqueous extract was beneficial effects of the oral administration of the aqueous filtered using a Millipore filter (Millipore 0.2 mm, St Quentin extract of CS fruits, which are the most commonly found in en Yvelines, France) to remove particulate matter. The filtrate the Moroccan kitchen and used as medicinal plant for the was then freeze-dried and the desired dose (mg of lyophilized treatment of diabetes mellitus, on plasma lipid parameters in aqueous extract of CS fruits per kg body weight) was then normal and streptozotocin (STZ) diabetic rats. prepared and reconstituted in 1.5 ml of distilled water. The aqueous extracts were prepared daily, just before administra- tion. The extracts obtained were then given orally to different 2. Material and methods groups of rats at a dose of 20 mg/kg body weight. The dose of 20 mg/kg was used according to the Moroccan traditional 2.1. Plant material phytotherapy. The extract was green coloured with a percent yield of 13%, its average osmolarity was 34 mOsm, pH 6.5, Specimens of CS (Capparidaceae) were collected from and with a very low viscosity. the Tafilalet region (semi-arid area) of Morocco in May–June

Fig. 1. Plasma triglycerides levels over 6 h after single oral administration of Fig. 2. Plasma cholesterol levels over 6 h after single oral administration of an aqueous CS extract (20 mg/kg) in normal (panel a) and diabetic rats (panel an aqueous CS extract (20 mg/kg) in normal (panel a) and diabetic rats (panel b). Data are expressed as means ± S.E.M., n = 6 rats per group. **p < 0.001, b). Data are expressed as means ± S.E.M., n = 6 rats per group. **p < 0.01, when compared to baseline values. () Control; ( ) CS; ( ) Vanadate. when compared to baseline values. () Control; ( ) CS; ( ) Vanadate. M. Eddouks et al. / Journal of Ethnopharmacology 98 (2005) 345–350 347

2.3. Animals used lyophilized aqueous extract of CS per kg body weight) and the + − third group received a reference drug (Vanadate (Na VO3 ) Experiments were performed in adult male Wistar rats at a dose of 0.8 mg/kg). For single oral administration, dis- weighing from 200 to 230 g. The animals were housed under tilled water (control), Vanadate (0.8 mg/kg) or the aqueous standard environmental conditions (23 ± 1 ◦C, with 55 ± 5% extract (20 mg/kg) were administered and plasma cholesterol humidity and a 12 h light/dark cycle) and maintained with and triglycerides levels were measured before and 6 h after free access to water and ad libitum standard laboratory diet CS treatment. (70% carbohydrates, 25% proteins, 5% lipids). 2.6. Repeated oral administration 2.4. Induction of diabetes For repeated oral administration, rats were treated once Streptozotocin (Sigma, St Louis, MO, USA) was dis- daily at a dose of 20 mg/kg for 2 weeks and plasma choles- solved in 0.1 M fresh cold citrate buffer at pH 4.5 before terol and triglycerides levels were followed during this pe- use, and injected intravenously into the tail vein at a dose riod. The rats (n = 6 in each group) were treated once daily. of 65 mg/kg. After 18 h, the rats with stable fasting blood Blood samples were collected from the tail vein and plasma glucose levels greater than 16 mmol/l were considered as di- triglycerides and cholesterol levels are determined enzymat- abetic and used in the present study. The percentage of re- ically by colorimetric specific kits (Randox, UK), respec- sponse to streptozotocin injection was 90%. tively. The kits used in this study for substrates analysis were specified for both human and rat blood samples at the same 2.5. Single oral administration percentage.

Normal and diabetic rats were randomly assigned to three 2.7. Statistical analysis different groups containing six rats each. One control group received distilled water, a second treated group received the Data were expressed as mean ± S.E.M. The statistical aqueous extract of CS fruits at a dose of 20 mg/kg (20 mg of analysis was performed by the Student’s t-test. The values were considered significantly different when the P-value was

Fig. 3. Plasma triglycerides levels after once daily repeated oral adminis- Fig. 4. Plasma cholesterol levels after once daily repeated oral administra- tration of an aqueous CS extract (20 mg/kg) for 15 days in normal (panel a) tion of an aqueous CS extract (20 mg/kg) for 15 days in normal (panel a) and diabetic rats (panel b). Data are expressed as means ± S.E.M., n = 6 rats and diabetic rats (panel b). Data are expressed as means ± S.E.M., n = 6 rats per group. *p < 0.05; **p < 0.01; ***p < 0.001, when compared to baseline per group. *p < 0.05; **p < 0.01; ***p < 0.001, when compared to baseline values. () Control; ( ) CS; ( ) Vanadate. values. () Control; ( ) CS; ( ) Vanadate. 348 M. Eddouks et al. / Journal of Ethnopharmacology 98 (2005) 345–350 less than 0.05 in comparison to baseline values (starting val- CS administration. Vanadate caused a significant decrease of ues). plasma triglycerides (p < 0.05) (Fig. 3a) and cholesterol lev- els (p < 0.001) after a single administration of CS (20 mg/kg) (Fig. 4a). 3. Results In STZ diabetic rats, CS extract decreased significantly the plasma triglycerides levels from the first week (p < 0.01) 3.1. Single oral administration to the 2nd (p < 0.01) (Fig. 3b). The plasma cholesterol levels were decreased from the 4th day (p < 0.05) to the 50th day In normal and STZ diabetic rats, no significant changes (p < 0.01) after CS treatment (Fig. 4b). Daily vanadate admin- of both plasma cholesterol and triglycerides concentrations istration (0.8 mg/kg) for 2 weeks produced a statistically sig- 6 h were noted after a single administration of CS (20 mg/kg) nificant decrease in both plasma cholesterol (p < 0.001) and (Figs. 1 and 2). In normal rats, Vanadate (0.8 mg/kg) did not triglycerides (p < 0.001) concentrations (Figs. 3b and 4b). affect significantly the plasma cholesterol and triglycerides concentrations after a single oral dose (Figs. 1a and 2a). How- 3.3. Body weight loss ever, in STZ rats, vanadate reduced both the plasma choles- terol (p < 0.01) and triglycerides levels (p < 0.01) 6 h after CS A significant decrease on body weight was observed in administration (Figs. 1b and 2b). distilled water treated group from the second day (p < 0.05) of STZ injection. CS aqueous extract (20 mg/kg) caused a 3.2. Repeated oral administration significant weight loss 2 weeks after once daily repeated oral treatment only in STZ rats (p < 0.01) (Fig. 5b) but not in nor- In normal rats, a significant reduction in plasma triglyc- mal rats (Fig. 5a). Vanadatecaused also a significant decrease erides was observed in CS-treated group from the 7th day on body weight in both normal (p < 0.01) and STZ (p < 0.001) (p < 0.05) to the 50th day of CS treatment (p < 0.01) (Fig. 3a). rats after repeated oral administration (Fig. 5). Plasma cholesterol levels dropped significantly, in normal rats, from the 4th day (p < 0.05) to the 7th day (p < 0.01) after 4. Discussion

This study was carried out in order to elucidate the influ- ence of daily oral administration for 15 days of the aqueous extract of CS fruits on plasma triglycerides, plasma choles- terol and body weight in normal and STZ diabetic rats. Vana- date, a potent inhibitor of tyrosine phosphatases (Tsiani et al., 1998), was used as a drug reference. Vanadate mimics sev- eral insulin actions in vivo: the stimulation of hexose uptake, the stimulation of lipogenesis and the inhibition of lipoly- sis. Administration of this compound to STZ diabetic rats normalises plasma levels of glucose, lipids, creatinine and thyroid hormones (Gupta et al., 1999). The levels of plasma lipids are usually raised in diabetes mellitus. Such elevation represents a risk factor for coronary heart diseases (Kannel and Mc Gee, 1979). The abnormally high concentration of plasma lipids is mainly due to the in- crease in the mobilization of free fatty acids from the periph- eral depots (Ahmed et al., 2001). The results demonstrated that the aqueous extracts of CS fruits produced a significant decrease in plasma triglycerides and cholesterol levels for the repeated oral administration (long-term treatment) in both normal and STZ diabetic rats. However, CS treatment did not affect lipid profile after single oral administration (short- term treatment) in both normal and STZ diabetic rats. The underlying mechanism by which CS exerts its cholesterol lowering effect seems to be a decrease in cholesterol absorp- tion from the intestine, by binding with bile acids within the Fig. 5. Effect of CS aqueous extract treatment (20 mg/kg) on body weight (g) in normal and diabetic rats over 2 weeks. Data are expressed as intestine and increasing bile acids excretion (Kritchevsky, means ± S.E.M., n = 6 rats in each group. *p < 0.05; **p < 0.01; ***p < 0.001, 1978; Kelly and Tsai, 1978). CS can also act by decreas- when compared to baseline values. () Control; ( ) CS; ( ) Vanadate. ing the cholesterol biosynthesis especially by decreasing the M. Eddouks et al. / Journal of Ethnopharmacology 98 (2005) 345–350 349

3-hydroxy-3-methyl-glutaryl coenzyme A reductase (HMG- In conclusion, our study demonstrated that repeated oral CoA reductase) activity, a key enzyme of cholesterol biosyn- administration of CS fruits for 15 days evokes a beneficial thesis (Kedar and Chakrabarti, 1982; Sharma et al., 2003) effect on the hyperlipidemia associated with hyperglycaemia and/or by reducing the NADPH required for fatty acids and in. This finding supports its use by the Moroccan population cholesterol synthesis (Chi, 1982). for the treatment and management of diabetes mellitus and In addition, CS fruits may improve hypercholesterolemia cardiovascular diseases. This implies that CS fruits consump- by modifying lipoprotein metabolism: enhanced uptake of tion can prevent or be helpful in reducing the complications LDL by increasing LDL receptors (Slater et al., 1980) of dyslipidemia associated with diabetes. Finally, the precise and/or by increasing the lecethin:cholesterol acyl transferase mechanism(s) and site(s) of this activity and the active con- (LCAT) activity (Khanna et al., 2002) which may contribute stituent(s) of CS fruits involved are still to be determined in to the regulation of blood lipids. LCAT plays a key role addition to toxicological studies. in incorporating free cholesterol into HDL and transferring back to VLDL or IDL, which is taken back by the liver cells (Rajlakshmi and Sharma, 2004). Caper may facilitate rapid Acknowledgements catabolism of LDL. On other hand, repeated oral administra- tion of aqueous extract of CS fruits for 15 days produced a sig- We would like to extend our thanks to the “Comite´ In- ◦ nificant decrease in plasma triglycerides in both normal and ter Universitaire Maroco-Franc¸ais, Action integree´ N MA/ STZ diabetic rats. The observed hypotriglyceridemic effect 03/83” and Moroccan Government for supporting this work. may be due to a decrease of fatty acids synthesis (Bopanna et al., 1997), enhanced catabolism of LDL, activation of LCAT and tissues lipases (Khanna et al., 2002) and/or inhibition of References acetyl-CoA carboxylase (Mc Carty, 2001) and production of triglycerides precursors such acetyl-CoA and glycerol phos- Ahmed, I., Lakhani, M.S., Gillet, M., John, A., Raza, H., 2001. Hy- potriglyceridemic and hypocholesterolemic effects of antidiabetic Mo- phate. mordica charantia (Karela) fruit extract in streptozotocin-induced di- The marked hyperlipidemia that characterizes the STZ di- abetic rats. Diabetes Research and Clinical Practice 51, 155–161. abetic rats (Riyad et al., 1988; Choi et al., 1991) seems to be Ali-Shtayeh, M.S., Abu Ghdeib, S.L., 1999. Antifungal activity of plant a consequence of uninhibited action of lipolytic hormones on extracts against dermatophytes. Mycoses 42, 665–672. adipose tissue. Since insulin inhibits adipose tissue hormone- Al-Said, M.S., Abdelsattar, E.A., Khalifa, S.I., El-feraly, F.S., 1988. Iso- lation and identification of an anti-inflammatory principle from Cap- sensitive lipase and reduces lipolysis, CS may mimic insulin paris spinosa. Pharmazie 43, 640–641. action. Anila, L., Vijayalakshmi, N.R., 2002. Flavonoids from Embilica offici- No significant change in plasma insulin concentrations nalis and Mangifera indica, effectiveness for dyslipidemia. Journal of was noted in both normal and STZ diabetic rats after CS Ethnopharmacology 79, 81–87. treatment (Eddouks et al., 2004). It seems that CS reduced Bopanna, K.N., Kannan, J., Gadgil, S., Balaraman, E.R., Rathore, S.P., 1997. Antidiabetic and antihyperglycaemic effects of neem seed kernel plasma cholesterol and triglycerides levels without stimu- powder on alloxan diabetic rabbits. Indian Journal of Pharmacology lating insulin secretion. It is well known that the level of 29, 162–167. glycemic control is the major determinant of plasma VLDL Brevard, H., Brambilla, M., Chaintreau, A., Marison, J.P., 1992. Oc- and triglycerides levels. Previously, we have reported that currence of elements sulphur in capers (Capparis spinosa L.) and CS produced a potent antihyperglycaemic activity in STZ di- first investigation of the flavour profile. Flavour Fragrance Journal 7, 313–321. abetic rats (Eddouks et al., 2004), so the strong hypolipidemic Calis, I., Kuruuzum-Uz, A., Lorenzetto, P.A., Ruedi, P., 2002. (6S)- effect of CS fruits could also be mediated by the improvement Hydroxy-3-oxalpha-ionol glucosides from Capparis spinosa fruits. of glycemia. Phytochemistry 59, 451–457. Chemical studies on CS have reported the presence of al- Chi, M.S., 1982. Effects of garlic products on lipids metabolism in choles- kaloids, flavonoids and glucosonates as main chemicals con- terol fed rats. In: Proceedings of Society of Experimental Biology and Medicine, vol. 171, pp. 174–178. stituents (Brevard and Brambilla, 1992; Sharaf et al., 2000; Choi, J.S., Yokozawa, T., Oura, H., 1991. Improvement of hyperglycemia Calis et al., 2002). One or more of these chemical compounds and hyperlipidemia in streptozotocin-diabetic rats by a methanolic of the plant is/are also likely to have contributed to the ob- extract of Prunus davidiana stems and its main component, pruning. served hypolipidemic activity of aqueous extract of CS fruit. Planta Medica 57, 208–211. Flavonoids are plant polyphenols found frequently in fruits. Eddouks, M., Maghrani, M., Lemhadri, A., Ouahidi, M.-L., Jouad, H., 2002. Ethnopharmacological survey of medicinal plants used for the A myriad of nutritional benefits has been attributed to these treatment of diabetes mellitus, hypertension and cardiac diseases in phytochemicals. It has been reported that administration of the south-east region of Morocco (Tafilalet). Journal of Ethnopharma- these compounds to hypercholesterolemic and hypertriglyc- cology 82, 97–103. eridemic rats evokes a significant lipid lowering activity Eddouks, M., Lemhadri, A., Michel, J.-B., 2004. and caper: and improves dyslipidemia (Koshy et al., 2001; Anila and a potential anti-hyperglycaemic plants in diabetic rats. Journal of Ethnopharmacology 94, 143–148. Vijayalakshmi, 2002). The intervention of other phytochem- Gadgoli, C., Mishra, S.H., 1999. Antihepatotoxic activity of p-methoxy ical constituents as bioactive hypolipidemic agents is not benzoic acid from Capparis spinosa. Journal of Ethnopharmacology excluded. 66, 187–192. 350 M. Eddouks et al. / Journal of Ethnopharmacology 98 (2005) 345–350

Germano, M.P., De Pasquale, R., D’Angelo, V., Catania, S., Silvari, V., Mc Carty, M.F., 2001. Inhibition of acetyl-CoA carboxylase by cystamine Costa, C., 2002. Evaluation of extracts and isolated fraction from Cap- may mediate the hypotriglyceridemic activity of pantetheine. Medical paris spinosa L. buds as an antioxidant source. Journal of Agriculture Hypotheses 56, 314–317. and Food Chemistry 27, 1168–1171. Rajlakshmi, D., Sharma, D.K., 2004. Hypolipidemic effect of dif- Gingsberg, H.N., 1994. Lipoprotein metabolism and its relationship to ferent extracts of Clerodendron colebrookranum in normal and atherosclerosis. Medicinal and Clinical North America 78, 1–20. high-fat diet fed rats. Journal of Ethnopharmacology 90, 63– Gupta, D., Raju, J., Prakash, J., Baquer, N.-Z., 1999. Change in the lipid 68. profile, lipogenic and related enzymes in the livers of experimental Riyad, A., Abdul-Ghani Abdul-Salam, S., Suleiman, S.M., 1988. Effect diabetic rats: effect of insulin and vanadate. Diabetes Research and of and lupine seeds on the development of experimental Clinical Practice 46, 1–7. diabetes in rats. Planta Medica 54, 286–290. Jouad, H., Haloui, M., Rhiouani, H., El Hilaly, J., Eddouks, M., 2001. Sharaf, M., El-Ansari, M.A., Saleh, N.A., 2000. Quercetin triglycoside Ethnopharmacological survey of medicinal plants used or the treat- from Capparis spinosa. Fitoterapia 71, 46–49. ment of diabetes, cardiac and renal diseases in the North centre re- Sharma, S.B., Nasir, A., Prabhu, K.M., Murthy, P.S., Dev, G., 2003. gion of Morocco (Fez-Boulemane). Journal of Ethnopharmacology 77, Hypoglycaemic and hypolipidemic effect of ethanolic extract of seeds 175–182. of Eugenia jambolana in alloxan-induced diabetic rabbits. Journal of Kannel, W.B., Mc Gee, D.L., 1979. Diabetes and cardiovascular risk Ethnopharmacology 85, 201–206. factors: the Framingham study. Circulation 59, 8–13. Slater, H.R., Packard, C.J., Bicker, S., Shephered, 1980. Effects of Kedar, P., Chakrabarti, C.H., 1982. Effects of bittergourd (Momordica cholestyramine on receptor mediated plasma clearance and tissue up- charantia) seed and glibenclamide in streptozotocin induced diabetes take of humain low dnsity lipoprotein in the rabbit. Journal of Bio- mellitus. Indian Journal of Experimental Biology 20, 232–235. logical Chemistry 255, 10210–10213. Kelly, J.J., Tsai, A.C., 1978. Effect of pectin, and agar on Tsiani, E., Bogdanovic, E., Sorisky, A., Nagy, L., Fantus, I.-G., 1998. cholesterol absorption, synthesis and turnover in rats. Journal of Nu- Tyrosine phosphatase inhibitors, vanadate and pervanadate, stimulate trition 108, 630–639. glucose transport and GLUT translocation in muscle cells by a mech- Khanna, K., Rizvi, F., Chander, R., 2002. Lipid lowering activity of Phyl- anism independent of phosphatidylinositol 3-kinase and protein kinase lanthus niruri in hyperlipemic rats. Journal of Ethnopharmacology 82, C. Diabetes 47, 1676–1686. 19–22. Yaniv, Z., Dafni, A., Friedman, J., Palevitch, D., 1987. Plants used for Koshy, A.S., Anila, L., Vijayalakshmi, N.R., 2001. Flavonoids from the treatment of diabetes in Israel. Journal of Ethnopharmacology 19, Garcinia cambogia lower lipid levels in hypocholesterolemic rats. 145–151. Food Chemistry 72, 289–294. Ziyyat, A., Legssyer, A., Mekhfi, H., Dassouli, A., Serhrouchni, M., Ben- Kritchevsky, D., 1978. Fiber, lipids and atherosclerosis. American Journal jelloun, W., 1997. Phytotherapy of hypertension and diabetes in ori- of Clinical Nutrition 31S, 65–74. ental Morocco. Journal of Ethnopharmacology 58, 45–54. Journal of Ethnopharmacology 98 (2005) 351–360

Antimalarial remedies in French Guiana: A knowledge attitudes and practices study

M. Vigneron a, X. Deparis b, E. Deharo a, G. Bourdy a,∗

a Laboratoire de Pharmacochimie des Substances Naturelles et Pharmacophores Redox, UMR-152 IRD –Universit´e Paul Sabatier, Centre IRD de Cayenne, BP 165, 97323 Cayenne, Guyane, France b Laboratoire d’Epid´emiologie, Institut Pasteur de Guyane, BP 6010, 97306 Cayenne, Guyane, France Received 10 June 2004; received in revised form 20 December 2004; accepted 26 January 2005

Abstract

A “knowledge attitudes and practices” study about malaria treatments was undertaken in French Guiana, along with an ethnopharmacological study. One hundred and seventeen people from five different groups and nationalities (Creole,´ Palikur, Galibi, Brazilian, and European) answered the questionnaire. The results were analysed using univariate and multivariate statistical analysis. First, we evaluated the overall knowledge about malaria from the interviewed people. According to bio-medical concepts, we noticed that they have a good knowledge of this illness. Secondly, we studied the treatment used by sick people during their last malaria attack. We demonstrated that, although bio-medical treatment is available in this area, people use both modern drugs and traditional remedies. Finally, preventive attitudes have been examined. One-third of the interviewed people drink regularly some herbal remedy to prevent febrile illnesses and malaria, thus displaying a strong concern about this disease. The ethnopharmacological study highlighted the frequent use of traditional remedies, along with their mode of preparation and administration. A total of 34 different species (both from flora and fauna) have been registered as antimalarial. Twenty-seven are used for curative purposes, 20 as preventive and 13 of them are used for both purposes. Quassia amara (Simaroubaceae) whose antimalarial activity has already been demonstrated was the species most frequently used as antimalarial for curative and preventive purposes. © 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: French Guiana; Malaria; Treatment; Medicinal plant; Traditional medicine; KAP

1. Introduction wet season. Mean annual precipitation on the coastline is 3000 mm (Groussin, 2001). Floristically, French Guiana be- French Guiana is a 91,000 km2 overseas department of longs to the Guianas floristic province, delimited east by the France, located northeast of the South American continent River Amazon and west by the River Orenoque. More than between 2◦ and 6◦ North and 52◦–54◦ West. Its frontiers 90% of French Guiana territory is covered by equatorial for- with Brazil are formed by the River Oiapoque (east) and est, and other types of vegetation include savannahs, swamps, the Tumuc-Humac mountains (south), while the River Ma- and mangroves. Actually, approximately 5200 phanerogamic roni delimits the western border with Suriname. The average species have been registered (De Granville, 2001). temperature ranges from 26 to 28 ◦C all year round, with Due to its history and geographic location, this French two marked seasons: the dry one occurring from July to department is a mosaic of ethnic groups. The overall pop- November and the rainy one from December to June. Hu- ulation is estimated 170,000 (Grenand and Grenand, 2001). midity varies from 50 (dry season) up to 100% during the Amerindians constitute the first inhabitants and belong to three distinct linguistic groups: the Tupi-Guarani (Wayampi ∗ Corresponding author. Fax: +33 594 31 98 55. and Emerillon), the Arawak (Palikur and Arawak) and the E-mail address: [email protected] (G. Bourdy). Karib (Galibi and Wayana). During the 17th Century, an-

0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2005.01.049 352 M. Vigneron et al. / Journal of Ethnopharmacology 98 (2005) 351–360

Fig. 1. Place of study. other community the “noirs marrons” (including the Nd- specific vectors control program together with educational juka, Aluku, Sarmaka, Paramaka ethnic groups) settled down program strongly help in reducing the impact on the popula- in French Guiana and today they number around 15,000 tion (Anonymous, 2002). people (Price and Price, 2003). Asian people and Euro- Within our research program aiming to detect new nat- pean people are also important communities. The Creoles´ ural substances with antimalarial activities (Bourdy et al., (from French Guiana or originating from other part of the 2004; Deharo et al., 2001, 2002), we became interested in the Caribbean zone) represent the highest part of the population. biological evaluation of indigenous phytotherapeutic prepa- Finally, Haitian, Brazilian, and people from Suriname rep- rations, rather than solvent-based plant extracts. This re- resent almost one-third of the overall population (Charrier, orientation of our work is in accordance with new trends 2002). in this field, as displayed by recent publications (Willcox et This department is severely affected by endemic malaria. al., 2004) and WHO directives (WHO, 1998, 2002). The fact It is estimated to be one of the South American countries that malaria affects the world’s poorest, and that many treat where the highest endemic prevalence exists, together with themselves with herbal medicine because it is more available Guiana and Suriname. Every year, between 3500 and 5000 and affordable, proves the urgency of antimalarial programs malaria cases are recorded (Carme and Venturin, 1999). Case that would detect local remedies with the greatest efficacy. repartition depends upon the place, most of them occurring These remedies should be investigated thoroughly, and could along the Oiapoque (higher prevalence of Plasmodium vivax) be promoted in other areas where the relevant plant exist and Maroni (higher prevalence of Plasmodium falciparum) or could be cultivated (Willcox and Bodeker, 2004; WHO, boarding river (Huguet and Clausse, 2003) (see Fig. 1). Al- 1997). In a long-term program, questions to be addressed though epidemiological factors are unfavourable (in-land dif- here are: which plant species are used, how widely they are ficult access, constant and unpredictable migration of the pop- used, and which one should be priorized for further research ulation, high prevalence of Plasmodium falciparum, high- (Willcox et al., 2004)? In order to answer these questions, chemoresistance level), the malaria mortality rate is very low we started a KAP (knowledge attitudes and practices) survey (no more than five cases a year). It is assumed that good in French Guiana, applied to 117 people, mainly directed to health coverage, free access to diagnostic and treatment, and assess disease perception and knowledge, treatment seeking M. Vigneron et al. / Journal of Ethnopharmacology 98 (2005) 351–360 353 attitude, use of therapeutic and prophylactic treatments. This cidence. Other preventive attitudes (use of bed nets, etc.) survey was analysed using correspondence analysis proce- were not recorded. dure, a specific statistical analysis routinely used in epidemi- The questionnaire was usually given at the house of the in- ological studies (Murgue et al., 1999). The following reveals terviewed people, sometimes outside, in an open place. Dur- the results of this survey with emphasis on the treatment fol- ing interviews with people who did not speak French, an lowed and the traditional remedies used, both preventives and interpreter was used. curatives. Other issues, such as validation of activity, will be dealt in following papers. 2.2. Ethnopharmacological study

2. Methodology Preventive and curative recipes were recorded with as many details as possible. Four parts of herbarium vouchers The study encompassed two different sections: the first for each species mentioned were collected. Voucherswere de- one was strictly a KAP questionnaire; the second part was posited in French Guiana Herbarium (CAY) and determined based on the compilation of traditional remedies used by in- with the help of specialists. formants. 2.3. Choice of the place of study 2.1. The KAP questionnaire Places of study were selected according to their facility of access, and also their epidemiological status in order to have a The KAP questionnaire was previously tested in a pre- representation of different transmission zones, sporadic and survey including 20 people. This pre-survey allowed us to medium-high, as classified in French Guiana (Huguet and check if the questions were well understood, and all together Clausse, 2003). The villages of Saul,¨ Ouanary, Saint-Georges pertinent to our project. Minor corrections were made ac- and Kamuyene were selected as places of study (see Fig. 1). cordingly, so that the questionnaire is in a definitive form. The last version included open and closed questions, and 2.4. Choice of the interviewed population was divided in seven parts (available from GB). They were as follows: In Saul¨ and Ouanary, the whole adult population was in- (1) Survey data. terviewed. In other places, the choice of the interviewed peo- (2) Sociological data. ple was made according to the availability and willingness (3) Knowledge of malaria. This part aimed to define what the of people. No selection was made upon sex, ethnic group, word malaria meant to the interviewed people. Questions or age. Interviewed people were adults, men or women who were oriented towards numerous symptoms encountered had previously had or had not suffered a malaria crisis, who when a person was suffering from this disease. People knew and used or did not use medicinal plants. were also invited to speak freely on this topic. This pre- liminary approach ensured us to reduce possible misun- 2.5. Statistical analysis derstanding between local conception of the disease and its bio-medical definition (Willcox and Bodeker, 2004; Data have been recorded using Excel® software. Espino et al., 1997; Gessler et al., 1995). We made univariate and multivariate statistical analysis. (4) Characteristics of the last malaria crisis (if any) and di- The goal of the univariate analysis is to analyse all answers agnosis/treatment seeking attitudes. of each question in the questionnaire. Mean, median and (5) Treatment used during the last malarial crisis. variance are calculated for assessing the distribution of each (6) Estimation and comparison of the different treatments quantitative variable. Proportions are calculated for express- available, definition of the ideal treatment (everyone was ing the results of each category of answer of the questions in questioned). the questionnaire. (7) Context of use of the preventives remedies. In this part, The multivariate statistical analysis is a correspondence we focused on self-medication practices used in order to analysis procedure (CAP), which allows assessing the partic- avoid malaria or fever. Generally epidemiological stud- ular link of several variables to each other. Correspondence ies try to quantify some risk-related attitudes (Quinones analysis procedure is a multivariate technique that converts et al., 1998; Snow et al., 1998) but preventives measures, tables of qualitative data into graphical displays in which rows such as regular ingestion of some plant remedies, are (individual cases) and columns (variables categories) are de- rarely documented nor taken into account except, maybe picted as labelled points on a graph (Deparis et al., 1999; from an ethnopharmacological point of view (Brandao et Murgue et al., 1999). Instead of trying to compare them using al., 1992). Within this survey, it seemed pertinent to focus proportions in several analysis, we broke down the measures on these practices, keeping in mind that these prophylac- of association between variables into a number of compo- tic preparations might have an impact on the malaria in- nents, or “factors” represented by axes. The coordinates of 354 M. Vigneron et al. / Journal of Ethnopharmacology 98 (2005) 351–360 each individual’s case and each variable’s category on these The mode of transmission is also well known, 96% of axes are calculated and computed, so that each successive the people quoting the mosquito as the agent responsible for axis or factor accounts for a decreasing portion of the to- malaria. The perceived severity of this disease is underlined tal association between variables, as represented by Pearson by the fact that 115 people said that malaria could be fatal, X statistics or, in other words, a decreasing portion of the both for children and adults. total. Due to the mechanism of this analysis, the first calculated 3.1.3. Characteristics of the last malaria crisis, and factor (F1) contains more information than the second (F2), diagnosis/treatment seeking attitudes the second more than the third (F3), and so on. Therefore, the Seventy-two people stated that they did suffer a malaria first two first factors are the most informative because they crisis, at least once in their life. The Galibis, Brazilian and Pa- contain the largest portion of the information resulting from likurs, along with some non-native people of the area where the analysis of the questionnaires. This permits most of the the interview took place, appeared to have a higher risk of information included in the data to be graphically presented catching malaria. They represent the mobile population as at the same time. In the plots, (i) the greater the distance opposed to the rest (Creoles,´ Europeans) who travel few in- from the origin indicates the greater importance of the cat- land, if not at all. Risk of catching malaria is the same for egories of the variable in the analysis, (ii) the proximity of both sexes, and sick people can suffer many crises, between points representing categories from two distinct variables is two and five. interpreted as an association between these categories; e.g. Almost everyone suspecting a malaria crisis went directly the proximity of the points “take preventive remedies” and to the nearest health centre for a diagnosis set. Everyone, but “age > 60 years” suggests an association which can be inter- two informants, mentioned that the diagnostic method used preted as people >60 years old often drink more preventive to confirm a malaria crisis was a smear test, which is routinely remedies than other people, (iii) the proximity of the category made in French Guiana. Only four people did not go to the points from the same variable is interpreted as similarity in the health centre because they quickly made their own prompt di- mean presentation between the groups of people expressed agnostic based upon the occurring symptoms. More than half in these categories; e.g. the proximity of the points “male” (51%) of the sick people remembered what type of parasite and “female” indicates similarity of the mean presentation they got infected with during their last crisis, Plasmodium fal- independently of genders. ciparum or Plasmodium vivax. The compilation of both data confirms that informants have a good knowledge of malaria, not only because they live in endemic area, but also because 3. Results and discussion of educational and preventive measures undertaken in French Guiana. 3.1. Epidemiological study: analysis of the KAP questionnaire 3.1.4. Treatment used during the last malaria crisis Seventy-two informants said that they had been sick at 3.1.1. Sociological data least once. Twenty-seven declared that they had treated them- A total of 117 people were interviewed. The sex ratio of selves with modern drugs alone and three had used traditional the interviewed population was nearly 1:1 (61 men and 56 remedy alone. The other 42 had used a combination of medic- women). Median and average ages were both close to 40 inal plants and modern drugs. When traditional remedies years of age. The interviewed population included Galibi (14 were used in combination with modern drugs, we recorded people), Creole´ (46), Palikur (35), Hmong (1), European (7) different patterns of association: (i) traditional remedies are and Brazilian (14). administered at the same time as modern drugs. Sometimes, the administration of traditional remedies overpasses the time of drugs, and the length of this type of treatment is of 15 3.1.2. Knowledge of malaria days approximately, (ii) treatment starts with modern drugs A great homogeneity was recorded in the answers to ques- alone. Then, traditional remedies are administered, for ap- tions dealing with symptoms, transmission mode, and esti- proximately 7 days after the last tablet, (iii) same as above mation of malaria severity. A large majority of informants re- but there are a few days (from 1 week up to one month) sponded in accordance with bio-medical definitions demon- of break between both treatments, (iv) treatment starts with strating strong knowledge of malaria. For the majority, the traditional remedies alone (two cases). In both cases of tra- fever symptom is always associated with malaria, but chills, ditional remedy treatment, it was followed by a modern drug strong headache; diarrhoea, vomits and loss of appetite are treatment, because in one case the Geissospermum sp. bark also identified as possible accompanying symptoms. decoction was too bitter to be drunk, and in the other case, the A very interesting fact is that 80% of the interviewed peo- wealthening effect of Phyllanthus amarus did not last. After ple declared that malaria was different from fever; malaria is one week of the Phyllanthus treatment, fever rose up again, recognized as a disease of its own and consequently, people and prompted the patient to change his treatment for modern said that malaria treatment is different from fever treatment. drugs. M. Vigneron et al. / Journal of Ethnopharmacology 98 (2005) 351–360 355

Throughout the correspondence analysis procedure, we ity and personal taste. Generally, the bitter is prepared and tried to better define who were the people who relied on tra- drunk at home because it is different from its aromatic ver- ditional remedies to cure their last malaria crisis. We found sion, which is drunk at the bar, called d´ecollage, richer in out that the choice of traditional remedies was not depen- fragrance and less bitter species. For people who do not wish dent upon sociological factors (i.e. ethnic group, place of to drink alcohol, a bitter can be prepared simply by boiling residence, sex, age, education level, marital status) but was plants parts. Finally a bitter, among the Creoles´ is also a three- dependent upon the fact of having suffered a previous episode step sequence of administration of different plants prepara- of malaria before the last crisis; in other words, the occurrence tions (namely rafraˆıchi, lok, purge), especially to children in of a first malaria crisis shapes the next treatment and orients order to get rid of worms and parasites. In that very case, patients towards the use of traditional remedies. plants are totally different from the previous ones (Grenand et al., 1987, and our observations). Underlying this practice is 3.1.5. Estimation and comparison of the different the idea that a bitter body (skin and blood, outside and inside) treatments available, definition of the ideal treatment is much stronger and is well protected against any attack; this For a high majority of informants, modern drugs are per- idea has been expressed many times by our informants, what- ceived as very efficient, acting very fast. The efficiency of ever their ethnic origin. modern drugs is beyond doubts. The fact that in most cases During our interviews, we discovered that repeated admin- the fever diminishes quickly, like within 2 days of a treatment, istration of preventive remedies was also performed within is also very much appreciated and is a conclusive proof of ac- other ethnic groups, especially among Palikur and Galibi. In tivity for everyone. Nevertheless, modern drugs are perceived the Palikur group, the word tiviyine (which is used to name all as potentially toxic, and are perceived to have strong damag- things considered as unpalatable or bitter) was used for these ing incidence on hepatic functions, thus enhancing the toxic preparations, perceived as a special class of remedies, charac- effect of malaria itself on the liver. terised also by their bitterness. In fact, in French Guiana and On the other hand, traditional remedies are said to be effi- among Galibi, Palikur, and Creole´ people, bitter seems to be cient, but for the majority of the interviewed people they are a generic term, passed over people, used for any plant-based perceived more as “liver and body purifier” and “cleansing preparation administered preventively. body impurities”. From our questionnaire, the main reason alleged to drink It was mentioned that both, plants and modern drugs are bitters were in order of importance: stay healthy, prevent easy to find, free, and impair relapses. In fact, the ideal treat- worms attack, and prevent malaria and fever. One-third of ment as defined by all our informants, is a combination of the interviewed people (38) declared expressly that they did modern drugs along with plants and it was often quoted that drink bitters regularly for that last purpose. “Modern drugs cure, but plants heal”. The role of modern For those who drank preventive traditional remedies drugs is, therefore, considered essential, but complementary against malaria and fever, the frequency of administration to plants, warrant of true healing. of these preparations could vary strongly, but on average, a traditional preventive remedy is taken every 28 days. Any- 3.1.6. Context of use of preventive remedies how, for some people, these prophylactic preparations are In French Guiana and in the French West Indies, especially considered efficient only if they drank some every day, this within the Creole´ ethnic group exists the tradition of the bit- being especially necessary if there is a malaria epidemic run- ter (“amer” in French), initially an alcoholic drink (Grenand ning around. Their action was compared to be like “a vaccine et al., 1987; Vilayleck, 2002). A bitter is basically a prepara- against fever and malaria”. Unlikely, others consider it is not tion made out of ingredients selected upon their organoleptic good to take these preparations daily because when a strong characteristic of strong bitterness. Many therapeutic virtues crisis arises, “their body will not respond” because it is too are attributed to these ingredients, i.e., stomachic, digestive, accustomed. They say that the “strong”, the “bitter” must be cholagogue, choleretic, liver protective, appetite enhancer, or used sparingly otherwise it loses its power. vermifuge properties. This way of selecting medicinal plants Also, though many informants mentioned the fact that upon organoleptic characteristics has also been recorded in drugs delivered in the pharmacy/health centre can also be many places, and theoretically demonstrated by Leonti in used for prophylaxis, none of them use them for such pur- Mexico (Leonti et al., 2002). Apparently, the “bitter” crite- poses; preventives remedies are always local recipes. rion is quite common for selecting medicinal plants used to Using the correspondence analysis procedure, we discov- cure malaria, thought it might have proved inefficient (Krettli ered that bitter self-administration against fever and malaria et al., 2001). was not dependent upon sociological factors, but was depen- In French Guiana, a traditional bitter is made by alco- dent upon the occurrence of a malaria crisis in the informant’s holic maceration (in rum, locally called taffia, or in another life, and also, was correlated with the administration of bitter alcoholic beverage) of selected ingredients for a few weeks. during childhood as a family habit. It’s as if the informants When ready, a small liquor glass is drunk every now and that already got malaria, had a protection that was recognized then and sometimes, many times a week. Species used to necessary and actively looked for. This protection-seeking at- prepare bitters are selected depending upon their availabil- titude is part of a tradition and perpetuates in adult life. 356 M. Vigneron et al. / Journal of Ethnopharmacology 98 (2005) 351–360

Table 1 Species used in antimalarial curative and preventive remedies Species Part used In curative treatment, used alone or In preventive treatment, used alone or in combinationa in combinationa Aristolochia stahelii O.C. Schmidt Stem – Used alone Aristolochiaceae Aristolochia trilobata L. Aristolochiaceae Leaf Always in combination and Always in combination and preferably with Geissospermum spp. preferably with Geissospermum spp., Quassia amara, Tinospora crispa, Eryngium foetidum, Artemisia spp., Curcuma longa Artemisia spp. Asteraceae Leaf – Always in combination and preferably with: Geissospermum spp., Quassia amara, Tinospora crispa, Eryngium foetidum, Aristolochia trilobata, Curcuma longa Ayapana triplinervis (Vahl.) R. King and H. Aerial part Always in combination – Robinson Asteraceae Banara guianensis Aubl. Flacourtiaceae Leaf Always in combination – Campomanesia aromatica (Aubl.) Griseb. or Leaf Always in combination and – Campomanesia grandiflora (Aubl.) Sagot. preferably with Siparuna spp. Myrtaceae Carica papaya L. Caricaceae Root, leaf Always in combination and – preferably with Quassia amara, Citrus sp. (lemon), Euterpe oleracea Citrus sp. (lemon) Rutaceae Root, leaf Always in combination and – preferably with Quassia amara, Carica papaya, Euterpe oleracea Citrus sp. (orange) Rutaceae Leaf Always in combination Alone Coutoubea spicata Aubl. Gentianaceae Whole plant Alone, or preferably in combination Alone or in combination with Geissospermum spp. and Quassia amara Curcuma longa L. Ziniberaceae Rhizome – Always in combination and preferably with Geissospermum spp., Quassia amara, Tinospora crispa, Aristolochia trilobata, Eryngium foetidum, Artemisia spp. Cymbopogon citratus Stapf. Poaceae Leaf Always in combination Always in combination and preferably with Eryngium foetidum Eryngium foetidum L. Apiaceae Whole plant – Always in combination and preferably with Geissospermum spp., Quassia amara, Tinospora crispa, Aristolochia trilobata, Artemisia sp., Curcuma longa Euterpe oleracea Mart. Arecaceae Root Always in combination and – preferably with Carica papaya, Citrus sp. (lemon), Quassia amara Geissospermum leave (Vell.) Miers or Inner bark Alone, or preferably in combination Alone, or preferably with Quassia Geissospermum sericeum Benth. and with Quassia amara, Aristolochia amara, Tinospora crispa, Hook. f. ex Miers Apocynaceae trilobata, Coutoubea spicata Aristolochia trilobata, Picrolemma pseudocoffea, Curcuma longa, Artemisia sp., Eryngium foetidum Gymnanthemum amygdalinum A. Chev. Leaf Always in combination Alone or in combination Asteraceae Hyptis pectinata (L.) Poit. Lamiaceae – Always in combination Lantana camara L. Verbenaceae Leaf Always in combination – Leonotis nepetifolia (L.) R. Br. Lamiaceae – Always in combination, preferably with Quassia amara Lippia alba (Mill.) N. E. Brown Verbenaceae Leaf Always in combination – Mikania guaco Humb. and Bompl. or Leaf Alone, or preferably in combination Alone, or preferably in combination Mikania micrantha Kunth. Asteraceae with Quassia amara, with Quassia amara Gymnanthemum amygdalinum Ocimum campechianum P. Mill. Lamiaceae Leaf Always in combination Always in combination Parides sp. Papilionidae Whole animal – Always in combination Petiveria alliacea L. Phytolaccaceae Leaf Always in combination Alone M. Vigneron et al. / Journal of Ethnopharmacology 98 (2005) 351–360 357

Table 1 (Continued ) Phyllanthus amarus Schumacher and Whole plant Alone, or preferably in combination Always in combination Thonning or Phyllanthus niruri L. with Geissospermum spp., Euphorbiaceae Aristolochia trilobata, Coutoubea spicata Picrolemma pseudocoffea Ducke Stem, leaf, root Always in combination and Always in combination, and Simaroubaceae preferably with Quassia amara preferably with Geissospermum spp. and Quassia amara Piper marginatum Jacq. Piperaceae Leaf Always in combination preferably – with Quassia amara Plectranthus barbatus Andrews Lamiaceae Leaf Alone – Quassia amara L. Simaroubaceae Leaf Alone, or preferably in combination Alone, or preferably with with Carica papaya, Piper Geissospermum spp., Tinospora marginatum, Geissospermum spp., crispa, Picrolemma pseudocoffea, Euterpe oleracea, Picrolemma Aristolochia triloba, Erygium pseudocoffea, Citrus sp. (lemon) foetidum, Artemisia spp., Curcuma longa Senna alata (L.) Roxb. or Senna reticulata Root Alone – (Willd.) HS Irwin and Barneby Fabaceae Siparuna guianensis Aubl. or Siparuna Leaf Always in combination and – poeppigii (Tul.) A.DC. Monnimiaceae preferably with Campomanesia spp. Tinospora crispaL. Hook f. and Thomson Stem Always in combination Always in combination, and Menispermaceae preferably with Quassia amara, Geissospermum spp., Aristolochia triloba, Eryngium foetidum, Artemisia spp., Curcuma longa Solanum leucocarpon Dunal Solanaceae Leaf Alone, or in combination – Zanthoxylum rhoifolium Lam. Rutaceae Inner bark Always in combination – (–): Not used for this purpose. a Used alone or in combination: listed species are the ones that have been recorded at least twice in association with the first column species. If no associated species are indicated, it means that no matching species have been associated at least twice with the first column species.

Unfortunately, the correspondence analysis procedure re- species have been mentioned to be used alone; they are vealed that drinking bitters did not have any influence on Coutoubea spicata, Geissospermum spp., Mikania spp., the occurrence of malaria crisis. The study reveals that peo- Phyllanthus spp., Quassia amara, Senna spp. and Solanum ple who declared that they drank bitter regularly in order to leucocarpon. avoid malaria are just as prone as the others to get it. In multi-ingredients recipes, we noticed that they were species more likely to be found together in combination; this 3.2. Ethnopharmacological study is the case for Cymbopogon citratus, Lippia alba, Ocimum campechianum, and Citrus sp. (orange). Another combina- 3.2.1. Curatives remedies tion frequently encountered is from Carica papaya, Euterpe Forty-five informants declared that they used a tradi- oleracea and lemon roots. tional recipe to cure their last malarial crisis. Thirty-nine Most frequently used species, such as Coutoubea spicata different recipes were registered made out of 27 species and Quassia amara, are used alone or they appear in combi- listed in Table 1. We counted as single species very close- nations with almost any listed species; they can be considered by species that were used indiscriminately by our infor- as the most versatile species, being the stable element of the mants and considered as a unique folk-taxa (Berlin, 1992). recipes. This is maybe due to their effective antimalarial ac- Such species were Geissospermum laeve and Geissospermum tivity (Moretti, 1986) though this still needs to be confirmed sericeum, Mikania guaco and Mikania micrantha, Phyllan- for Coutoubea spicata. thus amarus and Phyllanthus niruri, Siparuna poeppigiana All recipes (but one, a concentrated juice of Mikania and Siparuna guianensis, Campomanesia grandiflora and sp. soft leaves) are long-lasting decoction; generally, all in- Campomanesia aromatica, Senna alata and Senna reticulata. gredients are put in a large aluminium pot, cold water is Plants are the unique ingredients of curative remedies. The poured over and is left to boil for a while. A dose is then most used species are Quassia amara, Coutoubea spicata, taken from the aluminium pot, generally without remov- Carica papaya, Piper marginatum, Geissospermum spp., ing plant material. The dose to be drunk varies from one Euterpe oleracea, Picrolemma pseudocoffea and Citrus sp. to four cups (150 ml) daily, all day round, as water sub- (lemon). stitute. For preparations containing Picrolemma pseudocof- Most of the recipes (82%) are multi-components recipes, fea, recognised by all informants as very toxic, and Plec- ranking from two to five different ingredients. Only seven tranthus barbatus (designated as extremely bitter), the dose 358 M. Vigneron et al. / Journal of Ethnopharmacology 98 (2005) 351–360 is a small coffee cup (50 ml) three times a day for a few be administered warm, preventive remedies are drunk cold, days. even at fridge temperature. The warm decoction is also poured all over the patient As far as we know, none of the species used in this study body, and this is called a “bath”. Some people, enhancing the as preventive have been evaluated for such activity. This def- symbolic aspect of this cure, explained that the preparation initely should be done, at least the most used ones, such as had to be passed all over the skin with the help of an old pa- Geissospermum spp., due to the serious lack of appropriate tient’s cloth previously soaked in the decoction, then thrown antimalarial prophylactic drugs. away. Warm “baths” must be administered at least twice a day. 3.2.3. Preventive remedies versus curative remedies Reasons behind administering warm remedies is that they Even though, it has been expressed frequently by our infor- cause profuse sweating which “makes the fever go out the mants that “the plants which cure malaria are the same ones body”, “helps eliminate malaria completely”, “drives malaria that can prevent it”, in fact, less than one-third (13 over 34) of out of the body”, and “eliminates all impurities”. the species are used for both purposes. This might be due to Some of the species used as curative in this KAP study the fact that those were retained for analysis “malaria and/or have already been retained for biological and chemical anal- fever preventive remedies”, thus increasing and widening the ysis, and a few of them displayed in vitro or in vivo activities. range of preventive species. This is the case for Quassia amara (Ajaiyeoba et al., 1999, Digging deeper, we discovered that, among the 13 plants Moretti, 1986), Geissospermum spp. (Munoz˜ et al., 2000), species indiscriminately as preventive and curative, their ci- Tinospora crispa (Pachaly and Adnan, 1992), Picrolemma tation frequency in remedies varies greatly according to their pseudocoffea (Moretti, 1986), Gymnanthemum amygdalinum use. In other words, even if a species is quoted for both uses, (Huffman et al., 1997), Siparuna spp. (Jenett Siems et al., it has a strong tendency to be much more used for one or 1999), Zanthoxylum rhoifolium (Georges, 2002; DeFilipps the other purpose. For example it was quite a surprise to et al., 2004). In regards to the other species, antimalarial ac- see that one of the uttermost renowned antimalarial Amazo- tivity still needs to be proved. nian species, Geissospermum spp. (Milliken, 1997; Grenand et al., 1987; Brandao et al., 1992; Munoz˜ et al., 2000; and 3.2.2. Remedies used for prevention many others) was so poorly employed as curative, but on We retained for analysis only the last preventive reme- the contrary, used much more in favour as preventive. Same dies that have been taken by informants against malaria thing for Mikania spp.,Tinospora crispa, all of them being and/or fever. Thirty-eight different preventives recipes were species having a strong antimalarial reputation, but in fact, recorded based on the use of 20 different species (listed in in the frame of our study, poorly used as curative. From our Table 1). point of view, this reflects the difference between theoretical Eight species have been mentioned to be used as pro- knowledge and real practice. phylactics alone, but preventive remedies are mainly multi- components recipes, made out of plants and one animal (a caterpillar). Two modes of preparation were recorded: al- 4. Conclusion coholic maceration or decoction. When prepared in alco- holic maceration, bitters have a significantly higher number Previous ethnopharmacological studies have been under- of ingredients than when they are made out of water. We taken in French Guiana in order to document related tra- also revealed through the CAP that the ingredients combi- ditional knowledge, and also to detect plants with promis- nation within a recipe is related to the ethnic group. This ing biological activity (Andreu-Verin, 1998; Berton, 1997; confirms the previous statement: bitters are part of tradition, Grenand et al., 1987; Hay, 1998; Lecat, 2002; Fleury, 1991; so bitters recipes are passed over generation within the fam- Leduc, 2002), and especially plants with antimalarial poten- ily. tial (Fandeur et al., 1985; Moretti, 1986; Sauvain, 1989). Most species used in alcoholic bitters are: Geissosper- From some already published ethnopharmacological data, mum spp. bark, Tinospora crispa stem, Aristolochia trilobata it appears that there are broadly around 150 species used leaves, Quassia amara (bark or leaves), Picrolemma pseudo- against malaria and/or fever in French Guiana (Grenand et coffea (stem, leaves or roots), and Parides sp., a butterfly- al., 1987). In Latin America, another study (Milliken, 1997) caterpillar feeding on Aristolochia trilobata leaves. Some- reported 956 species with antimalarial or antipyretic reputa- times, in these alcoholic bitters macerations other flavoured tion. Oliveira et al. (2002) reviewed the Brazilian ethnomed- species might also be used, an aromatic additive such as Cur- ical literature and reported that, in this country, 197 species cuma longa (rhizom) or Artemisia spp. (leaves). The latter are used for the treatment of malaria and fever. species are never used in water. In our study, only thirty-four species have been used as Preventive remedies are administered orally and also, are antimalarial, and this number is more in accordance with the rubbed all over the body (as “bath” in case of watery prepa- survey undertaken in northern Brazil by Brandao et al. (1992), rations and in friction for the alcoholic ones). An interesting in which 4689 people declared they used 40 plants as malaria fact is that, in opposition with curative remedies that have to curative (and one as preventive). This discrepancy between M. Vigneron et al. / Journal of Ethnopharmacology 98 (2005) 351–360 359 the numbers of plants displayed in different studies might be Environnement, Temps, Espace, Societ´ es´ (ETES). Universite´ de Paris due to the fact that the literature compiled plants used against VII. p. 59 + Annexes. malaria and plants used against fever, thus increasing greatly Bourdy, G., Oporto, A., Gimenez, A., Deharo, E., 2004. A search for natural bioactive compounds in Bolivia through a multidisciplinary the number of species. In our study, it was clear that because approach. Part VI. Evaluation of the antimalarial activity of plants malaria is perceived differently than fever, treatments used to used by Isoceno-Guarani˜ Indians. Journal of Ethnopharmacology 93, treat malaria are very specific and more likely to prove some 269–277. efficiency. This is of importance when it comes to selecting Brandao, M.G., Grandi, T.S., Rocha, E.M., Sawyer, D.R., Krettli, A.U., accurately plants for future antimalarial study. 1992. Survey of medicinal plants used as antimalarial in the Amazon. Journal of Ethnopharmacology 36, 175–182. As far as we know, no study was fully dedicated to the Berlin, B., 1992. Ethnobiological classification. Principles of organization analysis of antimalarial treatment really used by the popu- of plants and animals in traditional societies. Princeton University lation, nor in Guiana, nor elsewhere. Also, very few studies Press, New Jersey, p. 335. are centred on the preventive plants-based methods used by Carme, B., Venturin, C., 1999. Le paludisme dans les Ameriques.´ the population, though some studies try to correlate the vari- Medecine´ Tropicale 59, 298–302. Charrier, R., 2002. Guyane, des peuples et des histoires. Antiane 54, ations in the diet with the incidence of malaria cases (Etkin 14–17. and Ross, 1983). We are clearly aware that this study should DeFilipps, R., Maina, S., Crepin J., 2004. Medicinal plants of the Guianas be considered as preliminary, and should be extent to a much (Guyana, Suriname, French Guiana). Department of Botany, National larger part of the population, including more people from dif- Museum of Natural History, Smithsonian Institution, p. 477. ferent places, especially people living along the Maroni river De Granville, J., 2001. Veg´ etation.´ In: Barret, J. (Ed.), Atlas illustrede´ la Guyane. IRD Editions, Paris, France, pp. 52–55. considered as a zone of permanent high transmission with a Deharo, E., Bourdy, G., Quenevo, C., Munoz,˜ V., Ruiz, G., Sauvain, M., predominance of Plasmodium falciparum cases (Huguet and 2001. A search for natural bioactive compounds in Bolivia through Clausse, 2003). a multidisciplinary approach. Part V. Evaluation of the antimalarial Further studies should be oriented to the biological evalua- activity of plants used by the Tacana Indians. Journal of Ethnophar- tion of some remedies in the form they are prepared locally by macology 77, 91–98. Deharo, E., Garcia, R., Oporto, P., Sauvain, M., Gautret, Ph., Ginsburg, the population (mainly water extract). The evaluation should H., 2002. A non-radiolabeled ferriprotoporphyrin IX biomineralization be done also on some multi-components remedies, in order to inhibition test (FBIT) for the high throughput screening of antimalarial evaluate the possible synergism of action of the constituents compounds. Experimental Parasitology 100, 252–256. of the preparation. Preventive activities of remedies should Deparis, X., Migliani, R., Merlin, M., 1999. Mise en evidence´ de facteurs be looked for. The study of the interaction between remedy favorisant la rupture de preservatifs´ chez des militaires franc¸ais outre mer. Medecine´ Tropicale 59, 1–5. and modern antimalarial drugs should also be undertaken. Espino, F., Manderson, L., Aciun, C., Domingo, F., Ventura, C., 1997. This last analysis would be a good indicator in regard to the Perception of malaria in the Philippines; transmission and prevention estimation of possible benefits gained by the local population of disease. Acta Tropica 63, 221–239. when using both treatments in combination. Etkin, N.L., Ross, P.J., 1983. Malaria medicine and meals: plant use among the Hausa and its impact on disease. In: Romanucci- Ross, L., Moerman, D., Tancredi, L. (Eds.), The Anthropology of Medicine: from Culture to Method. Prager, New York, pp. 231– Acknowledgements 259. Fandeur, T., Moretti, C., Polonsky, J., 1985. In vitro and in vivo as- We would like to express our thanks to people who were sessment of the antimalarial activity of sergeolide. Planta Medica 51, willing to share their knowledge about plants, especially An- 20–23. Fleury, M., 1991. Busi-nenge: les hommes-foret,ˆ essai d’ethnobotanique tonio and Ines` Narciso and their family. Special thanks are chez les Aluku (Boni) en Guyane Franc¸aise. These` de doctorat de due to Marie-Franc¸oise Prevost´ and Franc¸oise Crozier for l’Universite´ Paris 6. Specialit´ e´ de biologie veg´ etale´ tropicale. Univer- their help in voucher determinations, and to Diane Bergeron site´ de Paris 6, Paris, France, p. 357. for the English revision of the text. Gessler, M.C., Msuya, D.E., Nkunya, M.H.H., Schar,¨ A., Heinrich, M., Tanner, M., 1995. Traditional healer in Tanzania: the perception of malaria and its causes. Journal of Ethnopharmacology 48, 119–130. Georges, S., 2002. Etude chimique et biologique d’une plante guyanaise References a` forte reputation´ antimalarique. DEA Sciences des Agroressources, Faculte´ des Sciences Pharmaceutiques, Universite´ Paul Sabatier, Ajaiyeoba, E., Abalogu, U., Krebs, H., Oduola, A., 1999. In vivo an- Toulouse, France, p. 29 + Annexes. timalarial activities of Quassia amara and Quassia undulata plant Grenand, P., Grenand, F., 2001. Les groupes humains. In: Barret, J. (Ed.), extracts in mice. Journal of Ethnopharmacology 67, 321–325. Atlas illustre´ de la Guyane. IRD Editions, Paris, France, pp. 46–49. Andreu-Verin, S., 1998. Therapeutique´ occidentale et therapeutique´ tradi- Grenand, P., Moretti, C., Jacquemin, H., 1987. Pharmacopees´ tradi- tionnelle par les plantes chez les amerindiens´ Palikurs de Macouria. tionnelles en Guyane, (Creoles,´ Palikours, Wayampis). Editions de These` d’etat´ de Docteur en Medecine.´ Universite´ de Bordeaux II. p. l’ORSTOM, Collection Memoires,´ Numero´ 108. Paris, France, p. 569. 165. Groussin, J., 2001. Le climat guyanais. In: Barret, J. (Ed.), Atlas illustre´ Anonymous, 2002. Conclusions de la Troisieme` Conference´ de Consensus de la Guyane. IRD Editions, Paris, France, pp. 46–49. sur le Paludisme en Guyane, 4 et 5 octobre 2002. Cayenne, Guyane, Hay, Y., 1998. Etude´ de l’automedication´ par les plantes medicinales´ France. a` St. Georges de l’Oyapock (Guyane franc¸aise). DESS Gestion des Berton, M., 1997. Les plantes medicinales´ chez les ameridiens´ Pailkurs systemes` agro-pastoraux en zones tropicales. Universite´ Paris XII, Val de St Georges de l’Oyapock et de Macouria (Guyane franc¸aise). DEA de Marne. France, p. 56. 360 M. Vigneron et al. / Journal of Ethnopharmacology 98 (2005) 351–360

Huguet, P., Clausse, J., 2003. Profil epid´ emiologique´ du paludisme en Oliveira, F.Q., Brandao, M.G.L., Junqueira, R.G., Stehmann, J.R., Krettli, Guyane 2000–2002. Cellule de Veille sanitaire, DSDS, Cayenne, A.U., 2002. Plants used to treat fevers and malaria in Brazil. Poster Guyane, France. B192 of the 50th annual congress of the Society for Medicinal Plant Huffman, M., Gotoh, S., Turner, A., Hamai, M., Yoshida, K., 1997. Fur- Research. Fitoterapia 2, 310. ther observations on the use of Vernonia amygdalina, a medicinal Pachaly, P., Adnan, A., 1992. Tinocrisposide, a new bitter furanoditerpene plant used by a wild chimpanzee, its possible effect on parasite load glycoside from Tinospora crispa Miers. Archiv der Pharmazie 325, and its phytochemistry. African Study Monography 14, 227–240. 705–708. Jenett Siems, K., Mockenhaupt, F., Bienzle, U., Gupta, M., Eich, E., Price, R., Price, S., 2003. Les Marrons. Vents d’ailleurs, Chateauneuf-le-ˆ 1999. In vitro antiplasmodial activity of Central American medic- Rouge. France, p. 127. inal plants. Tropical Medicine and International Health 4, 611– Quinones, M., Lines, J., Thomson, M., Jawara, M., Greenwood, B., 1998. 615. Permethrin-treated bed nets do not have a mass-killing effect on vil- Krettli, A., Andrade, Neto, V.F., Bradao, M., Ferrari, W.M., 2001. The lage populations of Anopheles gambiae in the Gambia. Transactions search for antimalarial drugs from plants used to treat fever and of the Royal Society of Tropical Medicine and Hygiene 92, 373–378. malaria or plants randomly selected: a review. Memorias do Insti- Sauvain, M., 1989. Etude des plantes antiparasitaires du plateau des tuto Oswaldo Cruz 96, 1033–1042. Guyanes en Amazonie: antipaludiques et anti-leishmaniens. Doctorat Lecat, J., 2002. Le paludisme a` Camopi: situation, incidence, connais- de l’Universite´ Paris Sud, Unite´ de Formation et de Recherche de la sances attitudes et pratiques de la population amerindienne:´ facteurs Faculte´ de Pharmacie de Chatenay-Malabry, France, p. 180. de risque de nature comportementale. DEA de Sante´ Publique des Snow, R., Peshu, N., Forster, D., Bomu, G., Mitsanze, E., Ngumbao, E., Pays en voie de Developpement.´ Universite´ Pierre et Marie Curie, Chisengwa, R., Armstrong, A., Schellenberg, J., Hayes, R., Newbold, Paris VI, Institut SanteetD´ eveloppement,´ p. 86. C., Marsh, K., 1998. Environmental and entomological risk factors for Leduc, C., 2002. Enquetesˆ ethnobotaniques semi-quantitatives dans les the development of clinical malaria among children on the Kenyan populations Hmong de Guyane. Rapport de Maˆıtrise BPE, Universite´ coast. Transactions of the Royal Society of Tropical Medicine and de Rennes I, Rennes, France, p. 24. Hygiene 92, 381–385. Leonti, M., Sticher, O., Heinrich, M., 2002. Medicinal plants of the Popu- Vilayleck, E., 2002. Ethnobotanique et medecine´ traditionnelle creoles.´ laca, Mexico: organoleptic properties as indigenous selection criteria. Ibis Rouge Editions, Martinique, France, p. 142. Journal of Ethnopharmacology 81, 307–315. WHO, 1997. International Conferences held on Malaria, Recommenda- Milliken, W., 1997. Traditional antimalarial medicine in Roaramia, Brazil. tions. Dakar, Sen´ egal.´ NIH, OMS. Economic Botany 51, 212–237. WHO, 1998. Guidelines for the appropriate use of herbal medicines. Moretti, C., 1986. Contribution al’` etude´ de quatre Simaroubacees´ WHO Regional Publications. Western Pacific Series No. 23. Manila. medicinales´ de la Guyane. Quassinoides de Picrolemma pseudocof- WHO Regional Office for the Western Pacific. fea Ducke, Simaba cuspidata Spruce, Simaba multiflora Juss., Simaba WHO, 2002. Traditional Medicine Strategy, 2002–2005. morettii C. Feuillet. These` Universite´ de Paris-Sud, Centre d’Orsay. WHO/EDM/TRM. France, p. 120. Willcox, M., Bodeker, G., 2004. Frequency of use of traditional herbal Munoz,˜ V., Sauvain, M., Bourdy, G., Callapa, J., Bergeron, S., Rojas, I., medicines for the treatment and prevention of malaria, an overview Bravo, J., Balderrama, L., Ortiz, B., Gimenez, A., Deharo, E., 2000. of the literature. In: Willcox, M., Bodeker, F., Rasoanaivo, P. (Eds.), A search for natural bioactive compounds in Bolivia through a multi- Traditional Herbal Medicines for Modern times. Traditional Medici- disciplinary approach. Part I. Evaluation of the antimalarial activity of nal Plants and Malaria. CRC Press, Boca Raton, London/New York, plants used by the Chacobo Indians. Journal of Ethnopharmacology Washington DC, pp. 161–185. 69, 127–137. Willcox, M., Burford, G., Bodeker, G., 2004. An overview of ethnobotan- Murgue, B., Deparis, X., Chungue, E., Cassar, O., Roche, C., 1999. ical studies on plants used for the treatment of malaria. In: Willcox, Dengue: a re-evaluation of disease severity based on an analysis of M., Bodeker, G., Rasoanaivo, P. (Eds.), Traditional Herbal Medicines 403 laboratory-confirmed cases. Tropical Medicine and International for Modern times. Traditional Medicinal Plants And Malaria. CRC Health 4, 765–773. Press, Boca Raton, London/New York, Washington DC, pp. 187–197. Journal of Ethnopharmacology 98 (2005) 361–365

Ethnopharmacological communication Inhibitory effects of Okbyungpoong-Gamhmi on anaphylactic responses

Seung-Heon Hong a, Jin-Mu Yi a, b, Hongjun Kim b, Ho-Young Choi b, Yun-Kyung Kim a, Han-Jung Chae c, Hyung-Ryong Kim d, Cheorl-Ho Kim e, Hyung-Min Kim b, ∗

a Department of Oriental Pharmacy, College of Pharmacy, Wonkwang University, Iksan, Jeonbuk 570-749, Republic of Korea b College of Oriental Medicine, Kyung Hee University, 1 Hoegi-Dong, Dongdaemun-Gu, Seoul 130-701, Republic of Korea c Department of Pharmacology and CardioVascular Research Institute, Chonbuk National University School of Medicine, Jeonju, Jeonbuk 565-701, Republic of Korea d Department of Dental Pharmacology and Nano Science and Technology Research Institute, School of Dentistry, Wonkwang University, Iksan, Jeonbuk, Republic of Korea e Department of Biochemistry and Molecular Biology and Gynecology, College of Oriental Medicine, Dongguk University, Sukjang-Dong 707, Kyungju City, Kyungbuk 780-714, Republic of Korea Received 28 January 2005; received in revised form 4 February 2005; accepted 4 February 2005

Abstract

We investigated the effect of a herbal formulation Okbyungpoong-Gamhmi (OG) on mast cell-dependent anaphylactic reactions by intra- rectal administration. OG concentration dependently inhibited compound 48/80-induced anaphylaxis-like response and ear swelling response with doses of 0.01–1 g/kg. OG also inhibited the passive cutaneous anaphylaxis at the same concentrations. The histamine release induced by compound 48/80 or IgE from the rat peritoneal mast cells was reduced by 64.2 and 63.6%, respectively, at 1 g/l. These results provide evidence that intra-rectal therapy of OG may be beneficial in the treatment of anaphylactic response. © 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Intra-rectal administration; Ear swelling response; Passive cutaneous anaphylaxis; Systemic allergic reaction; Atractylodis japonica

1. Introduction and Magnoliae Flos (Magnolia denudata Desrousseaux, Magnoliaceae). Each of the dried individual herbs were pur- 1.1. Plant chased from Keum-dang yakupsa, a herbal drug company, and authenticated by Professor H. J. Song, Wonkwang Uni- The traditional herbal formulation Okbyungpoong- versity. Gamhmi (OG) consists of Atractylodis Rhizoma Alba (Atractylodis japonica Koidzumi, Compositae), Astragali 1.2. Uses in traditional medicine Radix (Astragalus membranaceus Bunge, Leguminosae), Ledebouriellae Radix (Ledebouriella seseloides Wolff, Um- OG has been used for the treatment of various types of belliferae), Cinnamomi Ramulus (Cinnamomum cassia allergic diseases in far eastern countries including Korea. Es- Blume, Lauraceae), Sutellariae Radix (Sutellaria baicalen- pecially this formulation has conventionally been applied to sis Georgi, Labiatae), Perillae herba (Perilla frutescens Brit- infant allergic patients by intra-rectal administration. But the ton var. acuta Kudo, Labiatae), Moutan Cortex Radicis pharmacological activity of this agent has not been investi- (Paeonoia moutan Sims, Paeoniaceae), Adenophorae Radix gated scientifically. (Adenophora tryphylla var. japonica Hara, Campanulaceae), Albizziae Cortex (Albizzia julibrissin Durazz, Leguminosae), 1.3. Previously isolated classes of constituents

∗ Corresponding author. Tel.: +82 2 961 9448; fax: +82 2 968 1085. Some classes of chemical constituents of the herbs con- E-mail address: [email protected] (H.-M. Kim). sisting this formula have been found. Essential oils from

0378-8741/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jep.2005.02.006 362 S.-H. Hong et al. / Journal of Ethnopharmacology 98 (2005) 361–365

Atractylodis Rhizoma Alba, Cinnamomi Ramulus, Perillae and applied to HPLC. Well known major chemical con- Herba, and Magnoliae Flos (Studennikova and Khaletskii, stituents of individual herbs include atractylon (Atractylodis 1966; Xu et al., 2001; Wilson et al., 1977; Nagasawa et al., Rhizoma Alba), astragaloside (Astragali Radix), psoralen 1969), saponins from Astragali Radix (Yu, 1986), coumarines (Ledebouriellae Radix), cinnamaldehyde (Cinnamomi Ra- from Ledebouriellae Radix (Jin et al., 1992), flavonoids from mulus), baicalin (Scutellariae Radix), perillaldehyde (Peril- Scutellariae Radix (Takido et al., 1975), phenolic compounds lae Herba), paeonol (Moutan Cortex Radicis), and cineole from Moutan Cortex Radicis (Harada and Yamashita, 1969), (Magnoliae Flos). triterpenoids from Albizziae Cortex (Kinjo et al., 1992), and alkaloids from Adenophorae Radix (Asano et al., 2000)have 2.4. Compound 48/80-induced systemic anaphylactic been previously reported. reaction

Mice were given an intraperitoneal (i.p.) injection of 2. Material and methods 7 mg/kg compound 48/80 to evoke systemic anaphylactic re- action. The OG was dissolved in saline and administered rec- 2.1. Reagents tally 1 h before the injection of compound 48/80. Mortality was monitored for 1 h after induction of the fatal response. Compound 48/80, anti-DNP IgE, DNP-human serum al- bumin (HSA), o-phthaldialdehyde (OPA), Evans blue, fetal 2.5. Ear swelling response bovine serum, ␣-minimum essential medium, and metriza- mide (density = 1.120 g/ml) were purchased from Sigma Ear swelling response was investigated by the method Chemical Co. (St. Louis, MO). previously used (Hong et al., 2001). Compound 48/80 was freshly dissolved in saline (5 g/l) and injected intrader- 2.2. Animals mally into the dorsal aspect of a mouse ear (100 ␮g/site, 20 ␮l) using a microsyringe with a 28-gauge hypoder- The original stock of male ICR mice (4-week age) and mic needle. Ear thickness was measured with a digital male Sprague-Dawley rats (7-week age) were purchased from micrometer (Mitutoyo, Japan) under mild anesthesia in- the Dae-Han Experimental Animal Center (Taejon, Korea) duced by i.p. injection of 1:1 mixture (50 ␮l) of ketamin and the animals were maintained at the College of Pharmacy, (1 mg/ml) and xylazine hydrochloride (23.32 mg/ml). Mice Wonkwang University. The animals were housed from 5 to 10 were kept in immobility state during the measurement. per cage in a laminar air-flow room, maintained at a tempera- Ear swelling response represented an increment in thick- ture of 22 ± 2 ◦C and relative humidity of 55 ± 10% through- ness above baseline control values. Ear swelling response out the study. was determined 40 min after compound 48/80 or vehicle injection. 2.3. Preparation of OG OG was administered rectally 40 min before the com- pound 48/80-injection (100 ␮g/site). The values obtained Extract of OG was prepared by decocting the dried pre- would appear to represent the effect of compound 48/80 scription of herbs with boiling distilled water. The extraction rather than the effect of the vehicle injection, since the ear decocted for approximately 3 h has been filtered, lyophilized, swelling response evoked by physiologic saline returned to and kept at 4 ◦C. Yield of lyophilized powder was 16% almost baseline thickness within 40 min. (w/w). OG was administered to mice in liquid state dissolved in saline. OG formula contains 11.8% of Atractylodis Rhi- 2.6. Preparation of rat peritoneal mast cells (RPMC) zoma Alba, 5.9% of Astragali Radix, 5.9% of Ledebouriellae Radix, 7.1% of Cinnamomi Ramulus, 7.1% of Scutellariae RPMC were isolated as previously described (Jippo et al., Radix, 10.5% of Perillae Herba, 10.5% of Moutan Cortex 2000). Mast cells were separated from the major components Radicis, 17.6% of Adenophorae Radix, 11.8% of Albizziae of rat peritoneal cells (i.e., macrophages and small lympho- Cortex, and 11.8% of Magnoliae Flos. cytes) according to the method described earlier (Yurt et al., Components of OG were analyzed by HPLC. The chro- 1977). More than 97% of the cells were viable as judged by matographic system consisted of a pump (Waters 2690XE Trypan blue uptake. HPLC pump), and a UV-detector (Waters M996 PDA detec- tor). A Nova-pak C18 (3.9 × 150) column (waters, U.S.) was 2.7. Histamine release used. Acetonitrile–H2O–Acetic acid (100:900:10) was used as the mobile phase. Detection of the peaks at 254 nm and Histamine release was assayed as previously described the sensitivity was set of 0.50 AUFS. The injection volume (Kim et al., 2003). The histamine content was measured by was 30 ␮l and flow rate was 1.0 ml/min. Standard solution a reported method of the OPA spectrofluorometric procedure was prepared by dissolving in distilled water (10 mg/100 ml). (Shore et al., 1959). The fluorescence intensity was measured The solution was filtered through 0.45 ␮l membrane filter at 438 nm (excitation at 353 nm) in a spectrofluorometer. The S.-H. Hong et al. / Journal of Ethnopharmacology 98 (2005) 361–365 363 inhibition percentage of histamine release can be calculated using the following equation: %Inhibition = {(histamine release without OG − histam- ine release with OG)/histamine release without OG} × 100

2.8. PCA reaction

The effect of PCA was investigated as another way to eval- uate anti-anaphylactic property of OG (Hong et al., 2001). OG dissolved in saline was administered intra-rectally 1 h prior to the challenge with antigen. The DNP-HSA was diluted in phosphate-buffered saline (PBS). The mice was injected in- tradermally with 250 ng/site of anti-DNP IgE into each of the three dorsal skin sites that had been shaved 48 h earlier. DNP- HSA was applied to the mice together with Evans blue as a Fig. 1. Effects of OG on histamine release from rat peritoneal mast cells. mixture. The pigment sites on the dorsal skin were punched Each value is presented as the mean ± S.E.M. of three independent exper- out. Dye was extracted from the skin with 1 ml of 1.0 N KOH iments. The cells (2 × 105 cells/ml) were pre-incubated with drug at 37 ◦C and 9 ml of a mixture of acetone and phosphoric acid (with for 15 min prior to incubation with DNP-HSA (hollow) or compound 48/80 ␮ * the ratio of 5:13) (Katayama et al., 1978). The cutaneous (6 g/ml) (hatched). P < 0.05; significantly different from IgE stimulation. **P < 0.05; significantly different from compound 48/80 stimulation. reaction can be quantified by measuring the concentration of leaked dye. Quantified dye amount measured from the OG-administered mouse was compared with that from the in each group was 18 (6/group × 3 independent experiment). vehicle-administered mouse. The absorbance at 620 nm of the PCA reaction was inhibited dose-dependently at the OG con- extract was measured in spectrophotometer and the amount centration range of 0.01–1 g/kg. of dye was calculated with the standard line of Evans blue. An in vitro study was performed to test the anti-allergic re- action of OG. The inhibitory effect of OG on compound 48/80 2.9. Statistical analysis and anti-DNP IgE-induced histamine release from RPMC was investigated. Fig. 1 shows OG dose-dependently inhib- The results were expressed as mean ± S.E. for the num- ited the compound 48/80 and anti-DNP IgE-induced his- ber of experiments. Statistical significance was compared be- tamine release at the concentration range of 0.01–1 mg/ml. tween each treated and control group by the Student’s t-test. At 0.1 and 1 mg/ml, the inhibition rate reached almost up to Results with P < 0.05 were considered statistically signifi- 25.0% and 64.2% for the compound 48/80-induced histamine cant. release and 24.7% and 63.6% for the anti-DNP IgE-induced histamine release. OG alone did not affect spontaneous his- tamine release. 3. Results The components of OG were analyzed by HPLC. Peaks of the principal components have not yet been identified in The in vivo model of systemic anaphylactic shock reaction this study. was designed to assess the contribution of OG in anaphylac- tic reactions. As shown in Table 1, an oral administration of 200 ␮l saline as a control (i.e. compound 48/80 injection 4. Discussion and conclusion without using OG) proved a 100% fatal shock. When the OG was administered through rectum (at doses ranging from 0.01 Stimulation of mast cell with compound 48/80 initiates to 1 g/kg) 1 h before compound 48/80 injection, the mortal- the activation of signal-transduction pathway that leads to ity was reduced dose-dependently. OG-inhibited compound histamine release. Some studies have shown that compound 48/80 induced mortality by 42.7 and 75% at doses of 1 and 48/80 and other polybasic compounds are able to activate G 0.1 g/kg, respectively. It has been reported that the compound proteins (Mousli et al., 1990a, 1990b). The membrane per- 48/80 significantly induced an ear-oedema response at an meability increase may be an essential factor for the release amount of 50–200 ␮g/site (Na et al., 2002). An amount of of the mediators from mast cells. We assume that OG might 100 ␮g/site was chosen for compound 48/80 which induced act on the lipid bilayer membrane affecting the prevention optimal ear-oedema response in this experiment. When mice of the perturbation being induced by compound 48/80 and were pre-treated with OG for 40 min through rectum, the ear- regulate the degranulation of the mast cells in mouse skin by oedema response derived by compound 48/80 was reduced stabilizing membrane fluidity. in a dose-dependent manner. OG-inhibited compound 48/80- The OG-administered mice were protected from IgE- induced ear-oedema response by 26 and 32.3% at the doses of mediated local allergic reaction. This was supported by the 0.1 and 1 g/kg, respectively. The number of mice employed effect of OG on PCA. OG also showed protection activity of 364 S.-H. Hong et al. / Journal of Ethnopharmacology 98 (2005) 361–365

Table 1 Effects of OG on allergic reaction in murine in vivo models Dose of OG (mg/kg) Systemic allergic reactiona Ear swelling responseb PCAc

Compound Mortality Compound 48/80 Ear swelling Inhibition Inhibition 48/80 (7 mg/kg) (%) (100 ␮g/site) (× 100 ␮m) (%) (%) None (D.W.) + 100 + 2.093 ± 0.134 – 0 ± 3.5 10 + 100 + 2.012 ± 0.104 4 5.7 ± 3.5 100 + 75.0 + 1.550 ± 0.118 26* 19.9 ± 3.8* 1000 + 42.7 + 1.417 ± 0.134 32.3** 32.8 ± 3.6** Each table shows effect of OG on compound 48/80-induced systemic allergic reaction on compound 48/80-induced ear-swelling response, and on the PCA in mice, respectively. a 200 ␮l distilled water (D.W.) or OG was intra-rectally given at various doses 1 h before (n = 8/group) the compound 48/80 injections. The compound 48/80 solutions were intraperitoneally given to mice. Mortality (%) within 1 h following the compound 48/80 injections is presented as the number of dead mice × 100/total number of experimental mice. b 50 ␮l D.W. or OG was intra-rectally given at various doses 40 min before (n = 6/group) the compound 48/80 injection. 20 ␮l of compound 48/80 (100 ␮g/site) were applied topically to the ears of mice. Each value represents the mean ± S.E.M. of three independent experiments. *P < 0.05. **P < 0.01; significantly different from the D.W. value. c OG was administered intra-rectally 1 h (n = 8/group) prior to the challenge with antigen. Each value is presented as the mean ± S.E.M. of four independent experiments. *P < 0.05. **P < 0.01; significantly different from the D.W. value.

IgE-induced histamine release from RPMCs. Antigen stimu- Jippo, T., Nomura, S., Kitamura, Y., 2000. Cichorium Intybus inhibits lation of mast cells via IgE receptor elicits release of numer- mast cell-mediated immediate-type allergic reaction. Internal Journal ous mediators including histamine in minutes and cytokines of Oriental Medicine 1, 82–88. Katayama, S., Shionoya, H., Ohtake, S., 1978. A new method for extrac- in hours. Consequently, these mediators induce immediate or tion of extravasated dye in the skin and the influence of fasting stress late hypersensitive reaction. The mechanism of histamine re- on passive cutaneous anaphylaxis in guinea pigs and rats. Microbiol- lease inhibition of this prescription against compound 48/80 ogy and Immunology 22, 89–101. or anti-DNP IgE-induced degranulation in mast cell is re- Kim, M.S., Na, H.J., Han, S.W., Jin, J.S., Song, U.Y., Lee, E.J., Song, mained for further studies. B.K., Hong, S.H., Kim, H.M., 2003. Forsythia fructus inhibits the mast cell-mediated allergic inflammatory reactions. Inflammation 27, In conclusion, our results obtained in this study proved that 129–135. OG-inhibited both compound 48/80-induced anaphylaxis- Kinjo, J., Araki, K., Fukui, K., Higuchi, H., Ikeda, T., Nohara, T., like response and IgE-mediated anaphylactic response in vivo Ida, Y., Takemoto, N., Miyakoshi, M., Shoji, J., 1992. Six new and in vitro in murine model. To our knowledge, this is the triterpenoidal glycosides including two new sapogenols from Al- first report of mast cell-mediated immediate-type allergic re- bizziae Cortex. Chemical and Pharmaceutical Bulletin 40, 3269– 3273. action by OG. Mousli, M.C., Bronner, C., Bockaert, J., Rouot, B., Landry, Y., 1990b. Interaction of substance P, compound 48/80, and mastoparan with ␣-sub unit C-terminal of G protein. Immunology Letters 25, 355– Acknowledgements 358. Mousli, M.C., Bronner, C., Landry, Y., Bockaert, J., Rouot, B., 1990a. Direct activation of GTP-binding regulatory proteins (G proteins) This study was supported by a grant of the Korea Health by substance P and compound 48/80. FEBS Letters 25, 260– 21 R&D Project, Ministry of Health & Welfare, Republic of 262. Korea (01-PJ2-PG4-J201PT01-0006). Na, H.J., Jeong, H.J., Bae, H., Kim, Y.B., Park, S.T., Yun, Y.G., Kim, H.M., 2002. Tongkyutang inhibits mast cell-dependent allergic reac- tions and inflammatory cytokines secretion. Clinica Chimica Acta 319, 35–41. References Nagasawa, M., Murakami, T., Ikeda, K., Hisada, Y., 1969. The geograph- ical variation of essential oils of flos Magnoliae. Yakugaku Zasshi 89, Asano, N., Nishida, M., Miyauchi, M., Ikeda, K., Yamamoto, M., Kizu, 454–459. H., Kameda, Y., Watson, A.A., Nash, R.J., Fleet, G.W., 2000. Poly- Shore, P.A., Burkhalter, A., Cohn, V.H., 1959. A method for fluorometric hydroxylated pyrrolidine and piperidine alkaloids from Adenophora assay of histamine in tissues. Journal of Pharmacology and Experi- triphylla var. japonica (Campanulaceae). Phytochemistry 53, 379– mental Therapeutics 127, 182–186. 382. Studennikova, L.D., Khaletskii, A.M., 1966. Data on the Harada, M., Yamashita, A., 1969. Pharmacological studies on the root study of the chemical composition of Atractylodis ovata bark of paeonia moutan. I. Central effects of paeonol. Yakugaku (Atractylodis ovata (Thunb.) DC). Aptechn Delo 15, 27– Zasshi 89, 1205–1211. 32. Hong, S.H., Kim, M.S., Lee, J.Y., Hwang, C.Y., Baek, S.H., Han, Takido, M., Aimi, M., Takahashi, S., Yamanouchi, S., Torii, H., 1975. D.S., Jung, W.Y., Seo, S.B., Kajiuchi, T., Kim, H.M., 2001. Novel Studies on the constituents in the water extracts of crude drugs. I. findings in inhibition of mast cell-dependent immediate-type cuta- On the roots of Scutellaria baicalensis Georgi. Yakugaku Zasshi 95, neous reactions by Gahmi-Shini-San. Clinica Chimica Acta 309, 85– 108–113. 90. Wilson, B.J., Garst, J.E., Linnabary, R.D., Channell, R.B., 1977. Perilla Jin, G., Li, J., Piao, H., 1992. Chemical constituents of Ledebouriella ketone: a potent lung toxin from the mint plant, Perilla frutescens seseloides Wolff. Zhongguo Zhong Yao Za Zhi 17, 38–40. Britton. Science 197, 573–574. S.-H. Hong et al. / Journal of Ethnopharmacology 98 (2005) 361–365 365

Xu, C.J., Liang, Y.Z., Song, Y.Q., Li, J.S., 2001. Resolution of the essen- Yu, J.H., 1986. Quantitative determination of astragaloside A in Astra- tial constituents of Ramulus cinnamomi by an evolving chemomet- galus membranaceus. Zhong Yao Tong Bao 11, 38–40. ric approach. Fresenius Journal of Analytical Chemistry 371, 331– Yurt, R.W., Leid, R.W., Austen, K.F., 1977. Native heparin from rat peri- 336. toneal mast cells. Journal of Biological Chemistry 252, 518–521. Journal of Ethnopharmacology 98 (2005) 367–370

Ethnopharmacological communication Alkaloids from Boophane disticha with affinity to the serotonin transporter in rat brain

Mikkel Sandager a,b, Nicolaj D. Nielsen a,b, Gary I. Stafford a, Johannes van Staden a, Anna K. Jager¨ b,∗

a Research Centre for Plant Growth and Development, School of Botany and Zoology, University of KwaNatal Pietermaritzburg, P/Bag X01, Scottsville 3209, South Africa b Department of Medicinal Chemistry, The Danish University of Pharmaceutical Sciences, Universitetsparken 2, DK-2100 Copenhagen Ø, Denmark Received 17 August 2004; received in revised form 21 January 2005; accepted 28 January 2005

Abstract

Bulbs and leaves of Boophane disticha are used in South African traditional medicine in the treatment of anxiety. Crude extracts of the leaves have shown affinity to the SSRI site on the serotonin transporter in a radioligand binding assay. In this study, two compounds, buphanadrine and buphanamine, were isolated by bioassay-guided fractionation on VLC and preparative TLC. The structures of the compounds were determined by 1H and 13C NMR. Fractions were tested for affinity to the serotonin transporter in a binding assay using [3H]-citalopram as ligand. The IC50 values of buphanidrine and buphanamine were 274 ␮M(Ki = 132 ␮M) and 1799 ␮M(Ki = 868 ␮M), respectively. The two alkaloids were also tested for affinity to the 5HT1A receptor, but only showed slight affinity. © 2005 Published by Elsevier Ireland Ltd.

3 Keywords: Boophane disticha; Buphanamine; Buphanidrine; [ H]-Citalopram; Depression; Radioligand binding assay; Serotonin transporter; 5HT1A

Plant leaves of Boophane disticha (L.f.) Herb (Amarylli- Previously isolated classes of constituents 11 alkaloids daceae) were collected in the Botanical Gardens, Pietermar- have been isolated from the bulb, with a total yield of 0.31%. itzburg, South Africa in March. A voucher specimen (Stafford Buphanidrine constituted 19.4% and buphanamine 14.1% of 53 NU) is deposited in the Bews Herbarium, University of the total alkaloid (Hautch and Stauffacher, 1961). KwaZulu-Natal, Pietermaritzburg. Uses in traditional medicine scales from the bulb of Boo- phane disticha is traditionally used for numerous of purposes, 1. Materials and methods e.g., treatment of hysteria in young women (Van Wyk et al., 1997). The bulb has narcotic properties and leads to visual 1.1. Isolation hallucinations in toxic doses (Du Plooy et al., 2001). In a screening of plants used for anxiety and depression for affin- Hundred g fresh leaves of Boophane disticha were ex- ity to the serotonin transporter in rat brain, leaf extracts of tracted with 70% ethanol (3 × 500 ml). The extract was frac- Boophane disticha had high affinity for the SSRI site (Nielsen tionated on 150 g of Merck Silica gel 60 in a vacuum column. et al., 2004). Five hundred millilitres of each of following solvents were used as eluents:hexane; hexane:ethyl acetate 50:50; 25:75; ethyl acetate; ethyl acetate:methanol 90:10; 80:20; 70:30; Abbreviations: DPAT, dipropylaminotetralin; SERT, serotonin trans- × porter; SSRI, selective serotonin reuptake inhibitor 60:40; 50:50; 40:60; 30:70; 20:80; 10:90; 2 methanol; wa- ∗ Corresponding author. Tel.: +45 3530 6339; fax: +45 3530 6041. ter. Active fractions from VLC (ethyl acetate:methanol 90:10 E-mail address: [email protected] (A.K. Jager).¨ and 80:20) where dissolved in 100 ml 70% ethanol, adjusted

0378-8741/$ – see front matter © 2005 Published by Elsevier Ireland Ltd. doi:10.1016/j.jep.2005.01.037 368 M. Sandager et al. / Journal of Ethnopharmacology 98 (2005) 367–370

Fig. 1. Structures of (1) buphanidrine and (2) buphanamine. to pH 3 with 4% acetic acid and partitioned against di- 1.2. Tissue preparation ethylether (3 × 100 ml). The pH was adjusted to 10 with NaOH and the aqueous phase was partitioned against di- All procedures were carried out at 0–4 ◦C. Whole rat ethylether (3 × 100 ml), ethylacetate (3 × 100 ml) and bu- brains, except cerebellum, were homogenised with an Ul- tanol (4 × 70 ml). The active fraction (diethylether from basic tra Turrax homogenizer in 1:10 (w/v) buffer (5 mM TRIS partition) was separated on preparative TLC using ethyl ac- base, 150 mM NaCl and 20 mM EDTA, pH 7.5). The ho- etate:methanol:water (90:20:10) as mobile phase. The TLC mogenate was centrifuged at 16,000 × g for 10 min and the plate was detected under UV 254/365 nm and a small part of tissue membranes washed with 120 ml of the same buffer. the TLC plate was sprayed with Dragendorrf reagent to detect The supernatant was discarded and the pellet was suspended alkaloids. Five bands were scraped off the TLC, eluted with in buffer (5 mM TRIS base and 5 mM EDTA, pH 7.5), left methanol and tested for activity. The isolated compounds to react for 20 min and then centrifuged at 16,000 × g for were dissolved in CDCl3 and the structures elucidated by 10 min. The supernatant was discarded and the pellet was 1H NMR and 13C H NMR. washed with 120 ml buffer (50 mM TRIS base, 120 mM NaCl

Table 1 13 1 NMR data for buphanamine (100 MHz for C, 600 MHz for H, CDCl3) Position 1H 13C COSY (H → H) NOESY (H → H) HMBC (C → H) 3 1 4.70 d ( J1,2 = 5.6) 63.6 2 2, 10, 11endo, 11exo 3 3 3 3 2 6.05 dddd ( J2,3 = 10.0, J2,1 = 5.6, J2,4␤ = 2.8, J2,4␣ = 1.9) 125.4 1,3,4␤ 1, 3 1 3 3 3 3 5.86 ddd ( J3,2 = 10.0, J3,4␣ = 4.6, J3,4␤ = 2.9) 127.02,4␣,4␤ 2, 4␣,4␤ 1 2 3 3 4 ␣: 2.83 dddd ( J4␣,4␤ = 19.6, J4␣,4a = 5.6, J4␣,3 = 4.6, 26.13,4␤,4a 3,4␤,4a 4 J4␣,2 = 1.9) 2 3 3 4 ␤: 2.21 ddt ( J4␤,4␣ = 19.6, J4␤,4a = 8.2, J4␤,3 = J4␤,2 = 2.8) 2, 3, 4␣,4a 3,4␣, 11exo, 12 exo 3 3 4a 3.79 t ( J4a,4␣ = J4a,4␤ = 8.2) 60.34␣,4␤ 4␣,6␣ 1, 3, 6␤ 2 6 ␣:4.43d( J6␣,6␤ = 16.7) 55.46␤ 6␤, 4a 12endo 2 ␤:4.08d( J6␤,6␣ = 16.7) 6␣ 6␣, 12 endo 6a 113.310 7 140.76␤, 10, 14 8 134.0 10, 13 9 149.7 10, 13 10 6.61 s 98.3 1, 11endo 10a 134.8 11exo 10b 49.01,4␣,4␤, 4a, 6␣,6␤,10 2 3 11 endo: 1.99 dddd ( J11endo,11exo = 12.3, J11endo,12endo = 8.7, 36.9 11exo, 12endo, 12exo 1, 10, 11exo, 12endo 1 3 4 J11endo,12exo = 3.0, J11endo,4a = 1.3) 2 3 exo: 2.12 ddd ( J11exo,11endo = 12.3, J11exo,12exo = 10.9, 11endo, 12endo, 12exo 1,4␤, 11endo, 12exo 3 J11exo,12endo = 8.2) 2 3 12 endo: 2.99 ddd ( J12endo,12exo = 13.2, J12endo,1endo = 8.7, 51.2 11exo, 11 endo, 12 exo 6␤, 11endo, 12exo 6␣,6␤ 3 J12endo,11exo = 8.2) exo: 3.91 m 11exo, 11 endo, 12 endo 9 4␤, 11exo, 12endo 2 13 5.93 AB-system ( JA,B = 1.5) 101.2 14 4.01 s 59.3 M. Sandager et al. / Journal of Ethnopharmacology 98 (2005) 367–370 369

and 5 mM KCl, pH 7.5) and then centrifuged at 16,000 × g for 10 min. The supernatant was discarded and the pellet fi- nally suspended in 120 ml buffer (50 mM TRIS base, 120 mM NaCl and 5 mM KCl, pH 7.5). This homogenate was kept at −70 ◦C until use.

1.3. [3H]-Citalopram binding assay

The method described by Plenge et al. (1990) was used with modifications (Nielsen et al., 2004). Two hundred mi- crolitres of test solution were mixed with 50 ␮lof[3H]- citalopram (4 nM) and 50 ␮l of tissue suspension, respec- tively. Paroxetin (1.5 ␮M) was used for the determination of unspecific binding. The total binding of [3H]-citalopram was determined with a buffer blank. All samples were incu- bated for 2 h at 25 ◦C and then filtered under vacuum using glass fibre filters. After 24 h the radioactivity on the filters was determined by liquid scintillation. Specific binding was calculated as total binding minus unspecific binding. All ex- periments were done in duplicate.

1.4. 8-Hydroxy[3H]DPAT binding assay

Two hundred microlitres of test solution were mixed with 50 ␮l of 8-hydroxy[3H]DPAT and 50 ␮l of tissue suspension. Buspirone (1 ␮M) was used to determine unspecific binding. Subsequent procedures were as for the [3H]-citalopram bind-

ing assay. Fig. 2. IC50 determinations of buphanidrine and buphanamine in the SERT and 5HT1A assays. 1.5. Estimation of IC50 values and Ki values

The IC50 values were calculated using GraFit 5 from The Ki values of the two compounds were calculated to be Erithacus Softwarethe utilising a full four-parameter equa- 132 ␮M for buphanidrine and 868 ␮M for buphanamine. For tion. The Ki values, which is independent of ligand con- citalopram the IC50 value was 1.3 nM, giving a Ki value of centration and thus more useful for inter-lab comparison, 0.6 nM. Both buphanidrine and buphanamine bound to the was calculated by Ki =IC50/(1+[L]/Kd), where [L] is the 5HT1A receptor with low affinity, the IC50 values were 1203 ␮ ligand concentration (here 0.75 nM) and Kd is 0.7 nM for and 2975 M, respectively (Fig. 2.). citalopram. For the SSRIs, and most other types of antidepressant drugs, it takes 2–4 weeks for the therapeutic action to de- velop. A leading hypothesis of this delayed pharmacologi- 2. Results cal action is desensitation of somatodendritic serotonin 1A autoreceptors (5HT1A) in the midbrain raphe, which act Two compounds, buphanidrine (7.6 mg) and buphanamine by reducing neuronal firing (Stahl, 1998; Holmes et al., (5.6 mg) (Fig. 1), with activity on the SERT were isolated. 2002). A major goal of antidepressant development is to im- Buphanidrine was identified by comparison with NMR data prove preceding drug classes with more rapid onset of an- previously reported (Viladomat et al., 1995). NMR data for tidepressant action and fewer unwanted side effects. Ther- buphanamine are given in Table 1. apeutic agents acting both by inhibition of serotonin reup- Structurally, buphanamine and buphanadrine have the take and by inhibiting the action of 5HT1A autoreceptors benzo-1,3-dioxole moiety in common with the clinically used might result in a more rapid onset of antidepressant action. SSRI paroxetin, which could explain their affinity to the Buphanidrine and buphanamine had only slight affinity to SERT. The hallucinogenic effects obtained after accidental or 5HT1A, eliminating the possibility of a dual-action system. purposeful overdosing with Boophane disticha extracts (De Although the activities of buphanidrine and buphanamine Smet, 1996; van Wyk et al., 2002) indicate that the alkaloids on the SERT were lower than the activity of the clinically reach the CNS. used SSRI citalopram, the activity supports the traditional In the SERT assay the IC50 values of buphanidrine and use of Boophane disticha as a remedy for depression and buphanamine were 274 and 1799 ␮M, respectively (Fig. 2.). anxiety. 370 M. Sandager et al. / Journal of Ethnopharmacology 98 (2005) 367–370

Acknowledgements Holmes, A., Yang, R.J., Murphy, D.L., Crawley, J.N., 2002. Evaluation of antidepressant-related behavioral responses in mice lacking the sero- The Danish Medical Research Council and The National tonin transporter. Neuropsychopharmacology 27, 914–923. Nielsen, N.D., Sandager, M., Stafford, G.I., Jager,¨ A.K., van Staden, J., Research Foundation, Pretoria are thanked for financial sup- 2004. Screening of indigenous plants from South Africa for affinity port. NMR equipment used in this work was purchased with to the serotonin reuptake transport protein. Journal of Ethnopharma- grants from Apotekerfonden af 1991, Copenhagen, and The cology 94, 159–163. Danish University of Pharmaceutical Sciences. Dan Stærk is Plenge, P., Mellerup, E.T., Nielsen, M., 1990. Inhibitory and regulatory thanked for performing the NMR experiments. binding sites on the rat brain serotonin transporter: molecular weight of the [3H]paroxetine and [3H]citalopram binding proteins. European Journal of Pharmacy: Molecular Pharmacology 189, 129–134. Stahl, S.M., 1998. Mechanism of action of serotonin selective reuptake in- References hibitors: serotonin receptors and pathways mediate therapeutic effects and side effects. Journal of Affective Disorders 51, 215–235. De Smet, P.A.G.M., 1996. Some ethnopharmacological notes on African Van Wyk, B., van Oudtshoorn, B., Gericke, N., 1997. Medicinal plants hallucinogens. Journal of Ethnopharmacology 50, 141–146. of South Africa. Briza Publications, Pretoria. Du Plooy, W.J., Swart, L., van Huysteen, G.W., 2001. Poisoning with van Wyk, B.-E., van Heerden, F.R., van Oudtshoorn, B., 2002. Poisonous Boophane disticha: a forensic case. Human and Experimental Toxi- Plants of South Africa. Briza Publications, Pretoria. cology 20, 277–278. Viladomat, F., Codina, C., Bastida, J., Shaheed, M., Cambell, W.E., Hautch, H., Stauffacher, D., 1961. Die Alkaloide von Buphane disticha 1995. Further alkaloids from Brunsvigia Josphinae. Phytochemistry (L.f.) Herb. Helvetica Chimica Acta 44, 491–502. 40, 961–965.

The author has requested enhancement of the downloaded file. All in-text references underlined in blue are linked to publications on ResearchGate.