Anti-inflammatory and antispasmodic activity of Ipomoea imperati ( Vahl ) Griseb ( Convolvulaceae )

! " #$ ! % &' ( & $) " *) ! % &' + ( & * " ,! % !' ( , $# " & ! % !' ( , )" . / 0 0 * ! " + #*' ( & -" -*' + ( & ,& " !!! "1 )& % -' *2 , % )' + ! " ( 2! " , " + Correspondence J.C. Freitas Departamento de Fisiologia Instituto de Biociências, USP Rua do Matão, Travessa 14, 101 05508-900 São Paulo, SP Brasil Fax: +55-11-818-7416 E-mail: jfreitas@usp.br Research supported by FAPESP. Received July 6, 2001 Accepted October 2, 2002


Introduction
Natural products have served as a source of drugs for centuries, and about half of the pharmaceuticals in use today are derived from natural products (1).Dependence on plants as the source of medicines is prevalent in developing countries where traditional medicine plays a major role in health care (2).
Ipomoea littoralis (Convolvulaceae), cur-rently known as I. imperati (Vahl) Griseb, is a member of the Ipomoea section Batatas, and is the only species in the section that is native and endemic to the old world (3).I. imperati blooms from December to April and has white flowers with a yellow throat apex and a purple throat base as a particularly distinctive characteristic (4).This plant grows well on sandy seashores in tropical climates, especially in Atlantic coastal areas and it is used in traditional medicine for the treatment of inflammation (the leaves being employed to treat furunculosis), swelling and wounds, as well as a diuretic (5).The extract of I. imperati is usually obtained from its leaves, which are also used to treat pains after childbirth and for stomach problems.A clinical study has reported an analgesic effect of this extract (5).In addition, other species of the family Convolvulaceae such as I. pes-caprae also have been reported to have anti-inflammatory, antispasmodic and antihemolytic properties (6).
In the present study, we examined the anti-inflammatory and antispasmodic activity of an alcohol-water extract of I. imperati and of its aqueous and lipid subfractions following oral and topical administration in several animal models in vivo, and also in vitro using the enzyme phospholipase A 2 (PLA 2 ).

Animals
Fasted male Wistar rats (180 ± 10 g), male Swiss mice (25 ± 5 g) and male guinea pigs (200-350 g) obtained from the breeding colony of the Institute of Biology, University of São Paulo (IB/USP) were used.The animals had free access to water.All experiments were approved by the Ethics Committee of the Physiology Department, IB/USP, São Paulo.

Plant material
The leaves of I. imperati were collected by one of the authors (A.C.B.Paula) along the seashore at Boracéia, São Paulo State, Brazil, in January and February from 1995 to 1998.The specimens were identified by Dr. José Rubens Pirani (University of São Paulo, São Paulo, Brazil) and Dr. Rosângela Simão Bianchini (Botanical Institute, São Paulo, Brazil) and were deposited in the Botanical Institute under voucher number SP 351848.
Lyophilization was used for the aqueous fraction and evaporation in vacuo at 40ºC (lipid fraction) was used to obtain the two fractions used for the pharmacological tests.

Acute toxicity in animals
Acute toxicity was studied in groups (N = 10) of male albino mice (18-22 g).I. imperati (lipid and aqueous fraction) was injected intraperitoneally and mortality was recorded for 24 h (7).

Mouse ear edema induced by arachidonic acid and croton oil
The in vivo anti-inflammatory activity of I. imperati was assessed in the mouse ear edema model using arachidonic acid and croton oil to induce inflammation (8).Control mice received only the irritant agents, whereas experimental mice also received the extract (2.5, 5.0, 7.5 and 10.0 mg/ear of both fractions) applied topically together with the irritant agent.Arachidonic acid and croton oil were dissolved in acetone to concentrations of 100 and 10 mg/ml, respectively.Each mouse received 0.5 mg of arachidonic acid/ear or 200 µg of croton oil/ear in the left ear.
The drugs were applied topically to the inner surface of the ear with an automatic pipette in a volume of 5 µl (arachidonic acid) or 20 µl (croton oil).The right ear (control) received 20 µl of acetone alone (vehicle).
The mice were killed by cervical dislocation 1 and 6 h after arachidonic acid and after treatment with croton oil, respectively.Each ear was perforated with a metal punch (a 6-mm diameter disc) and edema was assessed by subtracting the weight of the disc from the right control ear from the weight of the disc from the left treated ear.Hydrocortisone (217.0 µg/ear) was used as the positive control.
Drug effects were calculated as percent inhibition using the following equation:

Cotton pellet granuloma
Dental cotton rolls (Johnson and Johnson, New Brunswick, NJ, USA) were cut into 5mm pieces and sterilized in groups of four pellets (160 mg).Rats were anesthetized and the pellets were implanted subcutaneously at four symmetrical positions in the abdomen (9,10).
Two hours after implantation of the cotton pellets, the animal groups (N = 10 each) were treated orally with carboxymethylcellulose (0.2 mg/kg), mineral water (1 ml), dexamethasone (0.2 mg/kg), or the lipid and aqueous fractions of the extract of I. imperati at doses of 10, 30, 100, 300 or 1000 mg/kg.Administration was then continued daily for 6 days.On the seventh day, the animals were killed by cervical dislocation, and the granulomas were removed, dried (60ºC), and weighed.The difference between the initial and final dry weight corresponded to the weight of the granulomatous tissue formed.The animals sacrificed after subchronic treatment had their stomachs excised and opened along the greater curvature to determine the number of lesions.

Phospholipase activity
The inhibition of PLA 2 activity (purified from Apis mellifera bee venom) by the I. imperati extract was assayed by measuring the decrease in the pH of the incubation mixture using a pH electrode in a closed stirring chamber.The assay medium contained 4 mM sodium taurocholate, 12 mM calcium chloride and 7 mM phosphatidylcholine dipalmitate (Sigma, St. Louis, MO, USA).According to Ref. 11, this technique is reliable down to pH 5.0.In the present study, the mean initial pH of the phospholipid mixture was 8.0.Positive controls were set up using purified PLA 2 (0.33 µg/ml) from bee venom.The total incubation volume was 2.5 ml.The four different concentrations used in these in vitro experiments were 2.5, 3.0, 3.5 and 5.0 mg/ml of both fractions.In the case of the lipid fraction we used propylene glycol as an apolar solvent and mineral water as a polar solvent.

Isolated ileum
Guinea pigs were sacrificed by cervical dislocation and exsanguinated.Ileum segments (2 cm each) were prepared and mounted to record isometric contractions using force transducers (Grass FTO3) connected to a Narco Bio Systems polygraph.A 30-min period for stabilization was followed by a 10-min period during which basal activity was recorded.The test substances (lipid and aqueous fractions of the extract dissolved in carboxymethylcellulose at doses of 1.0, 3.0, 5.0 and 10.0 mg/ml) were added to the bath in a volume of 50 µl; one concentration was tested per ileum segment.
In the same experiments, the preparations were electrically stimulated with supramaximal rectangular pulses (20 V, 1.0 ms, 0.1 Hz) and the effects of non-cumulative extract concentrations were evaluated as described above.
The influence of the extract on the contractions induced by acetylcholine (1 µM) and histamine (1 µM) was studied in preparations preincubated for 10 min with the fraction.The fractions of the I. imperati extract were tested at their corresponding IC 50 for the inhibition of electrically induced contractions.The effect of these fractions was assessed by comparing the contractile responses to acetylcholine and histamine obtained in the absence and in the presence of extract components (1.0, 3.0 and 10.0 mg/ ml) (12).

Statistical analysis
Results are reported as means ± SEM and were analyzed statistically by analysis of variance (ANOVA) followed by the Tukey test.P values of less than 0.05 were considered significant.

Acute toxicity study in animals
The methanol-water extract of I. imperati at doses up to 2000 mg/kg did not produce any mortality in male albino mice (N = 10) up to 24 h after intraperitoneal injection.

Effect of Ipomoea imperati on croton oil-and arachidonic acid-induced mouse ear edema
The mouse ear edema reached a maximum at 6 h after croton oil application and 1 h after arachidonic acid application.The lipid and aqueous fraction, 10 mg/ear of I. imperati extract, significantly inhibited swelling, considerably reducing the vascular permeability response to croton oil and arachidonic acid application.I. imperati inhibited the inflammation induced by croton oil in a concentration-dependent manner.The maximum inhibition induced by the lipid and aqueous fractions was 89.0 ± 1.3% with an IC 50 of 3.97 mg/ear and 57.0 ± 1.3% with an IC 50 of 3.5 mg/ear, respectively (P<0.05), with a confidence interval (CI) of 2.51-6.27and 2.98-6.09mg/ear, respectively.The positive control dexamethasone (217 µg/ear) induced a maximum inhibition of 52.0 ± 1.0% with an IC 50 of 28.3 µg/ear and a CI of 26.2-30.45µg/ear (Figure 1A).

Farmacological effects of Ipomoea imperati
3.43 mg/ml, respectively, and the CI was 2.47-3.98 and 2.83-4.04mg/ml, respectively, which are not proportional to the weight of the extract.The extract exhibited significant inhibitory activity on the formation of granulation tissue in the cotton pellet test.The lipid and aqueous fractions of I. imperati (at the dose of 1000 mg/kg) inhibited the inflammatory process by 73.2 ± 1.2% (IC 50 of 93.7 mg/ear and CI of 68.1-128.95mg/ear) and by 56.14 ± 2.7% (IC 50 of 102.5 mg/ear with a CI of 72.10-145.59mg/ear), respectively, 6 days after implantation of the cotton pellet (Table 2).The lipid fraction (1000 mg/kg) was as effective as dexamethasone (0.20 mg/kg) in inhibiting this inflammation by 73%.No gastric mucosal lesions were observed after 6 days in the rats used in these experiments.Other doses used in this experiment proved to be effective, but maximum inhibition was obtained with 1000 mg/kg (Table 2).
The methanol-water extract of I. imperati dose dependently inhibited the tone and amplitude of the contractions of guinea pig ileum in electrically stimulated preparations.The lipid fraction had an IC 50 of 1.60 mg/ml and caused a maximum inhibition of 85.0 ± 1.0% (CI of 0.72-3.58mg/ml), while the aqueous fractions had an IC 50 of 4.12 mg/ml and caused a maximum inhibition of 81.0 ± 1.53% (P<0.05; Figure 2) with a CI of 2.13-7.97mg/ml.

Discussion
A large number of herbal drugs are reputed to be of excellent medicinal value, and are used for the treatment of several ailments.In folk medicine, various indigenous drugs are used in single and/or in combined form to treat different types of inflammatory and arthritic conditions, with considerable success (13).
In the present study we showed that the lipid fraction of the methanol-water I. imperati leaf extract contains the active agent against topical inflammation induced by croton oil and arachidonic acid.However, the positive control dexamethasone was effective only in inhibiting croton oil-induced topical inflammation.It is known that the anti-inflammatory effect of dexamethasone depends on the model used.Lee et al. (14) reported that dexamethasone (0.05 mg/ear dissolved in acetone) applied topically 16 h before the induction of mouse ear edema inhibited by 92% the inflammation induced by 12-O tetradecanoyl phorbol acetate (TPA).It has been established that inflammation induced by TPA/croton oil is related to the activation of PLA 2 , which releases arachidonic acid from the cell membrane.Arachidonic acid, in turn, is metabolized to prostaglandins and leukotrienes.Substances able to inhibit edema could be inhibitors of cyclooxygenase (COX) and/or 5-lipoxygenase (15).However, other agents are involved in edema generation such as serotonin and bradykinin, which act directly through specific receptors (5HT 2 , B 2 ) present in the endothelial cells of postcapillary venules.On the other hand, pretreatment with antagonists of protein kinase C may suppress the inflammation induced in guinea pig illeum.However, the inflammation observed in this model is related to the activation of PLA 2 .Bresnick et al. (16) stated that PLA 2 catalyzes the sn-2 hydrolysis of phospholipids liberating free fatty acids, predominantly arachidonic acid, and lysophospho-lipids.These products can have biological actions or be further metabolized to form a variety of proinflammatory lipid mediators including prostaglandins, leukotrienes, or platelet-activating factor and thus the inhibition of PLA 2 by pharmacological agents should have led to an anti-inflammatory effect.
The mechanism of croton oil-induced inflammation involves an increase in PLA 2 activity (17,18), which in turn leads to the release of arachidonic acid and subsequent biosynthesis of leukotrienes and prostaglandins (19,20), thus also involving the lipoxygenase pathway.It has been firmly established that arachidonic acid metabolites act as mediators of the inflammatory response via COX and lipoxygenase activity, and have therefore been a target for the development of therapeutic agents (21).
Enzyme inhibition assays are important tools in the search for new drugs.Our experimental work also concerns the effects of I. imperati extracts on several in vivo and in vitro eicosanoid-releasing systems, to investigate whether they contribute to the antiinflammatory activity of these compounds.The anti-inflammatory action of I. imperati extracts on arachidonic acid-induced inflammation involved inhibition of the PLA 2 enzyme and probably of COX II (COX-2), the key enzyme involved in prostaglandin biosynthesis.So PLA 2 and probably the COX-2 enzyme of arachidonate metabolism can be inhibited by I. imperati, suggesting that this plant can provisionally be classified as a dual inhibitor, with greater PLA 2 than COX-2 activity.The step preceding arachidonic acid liberation, which is catalyzed by PLA 2 , can be a target of the drugs (22).
Since endothelial cell contraction increases vascular permeability, and since this permeability is inhibited by vasodilating agents, the antispasmodic activity of I. pescaprae can be assumed to contribute to the anti-inflammatory action of this plant (6).The cited investigators based their hypo-thesis on the tentative theory that endothelial cell contraction causes increased vascular permeability and on the fact that agents such as ß-adrenoceptor agonists and calcium antagonists inhibit this permeability with general relaxant properties.
During the process of acute inflammation vascular leakage is a cardinal effect believed to result from the actions of certain inflammatory mediators, such as histamine, 5-hydroxytryptamine, bradykinin, LTE 4 and platelet-activating factor, on the postcapillary venules.The mechanism of this leakage was proposed to be due to mediator-induced contraction of endothelial cells.The cells are thereby pulled apart from each other and create gaps, allowing blood components to leak into the interstitial compartment creating edema (23).Pongprayoon et al. (24) suggested that endothelial cells and smooth muscle cells share a similar contractile mechanism.Vasoconstriction has been implicated in the dermatitis caused by jellyfish stings, leading to local vascular insufficiency and gangrene.Agents with a direct vasodilatory action, e.g., papaverine (IC 50 values of 30 µg/ml) have been recommended for the treatment of such toxin-induced dermatitis because of their nonspecific antispasmodic action.
I. imperati extracts have local and systemic anti-inflammatory actions in mice and rats, respectively.The pharmacological mechanism involved in this anti-inflammatory effect may be related to the inhibition of PLA 2 and COX-2.Additional experiments are necessary to demonstrate the inhibition of COX-2 by I. imperati.The extract also presented a nonspecific antispasmodic activity on the isolated ileum, inhibiting histamine and acetylcholine.In the acute toxicity assay, 1 mg/kg of I. imperati methanol-water extract caused no mortality in mice after 24 h.

(
weight of left minus right control ears) -(weight of left minus right treated ears) x 100 (weight of left control ear)

Figure 2 .
Figure 2. Dose-dependent inhibition of contractions by Ipomoea imperati extracts in electrically stimulated isolated guinea pig ileum.Each point represents the mean ± SEM of 5 experiments.Lipid fractions (1.0, 3.0 and 5.0 mg/ml) were significantly different from the aqueous fraction (*P<0.05,Tukey test).

Figure 3 .
Figure 3. Dose-dependent inhibition of 1 µM acetylcholine-and 1 µM histamine-induced contractions in guinea pig ileum by Ipomoea imperati extracts.Data are reported as means ± SEM for 5 experiments.*P<0.05 compared to the aqueous fraction (Tukey test).

Table 1 .
Inhibitory action of the lipid and aqueous fractions of Ipomoea imperati leaf extract on phospholipase A 2 activity.

Table 2 .
Effect of the oral administration of Ipomoea imperati leaf lipid and aqueous extracts on rat granuloma tissue formation.
Data are reported as the means ± SEM for 10 rats.Lipid fractions 30, 100, 300 and 1000 mg/kg of both fractions and the positive control (dexamethasone) were significantly different (*P<0.05,Tukey test).