South Afncan Journal of Botany 2001, 67 349--353 Copyngh/ @NfSC pty Ltd Pnnled In Soutil Afnca - All ng/lts reserved ·SOUTH AFRICAN JOURNAL OF BOTANY ISSN 0254-6299 Pharmacological and phytochemical properties of Dombeya rotundifolia KA Reid, AK Jager and J van Staden' Research Centre for Plant Growth and Development, School of Botany and Zoology, Universily of Natal Pietermaritzburg, Privale Bag XOI, Scottsville 3209, South Africa *Corresponding author, e-mail: [email protected] Received 3 October 2000, accepted in revised form 17 November 2000 Dombeya rotundifolia (Hochst.) Planch, is used in tradi­ young shoot extracts. Material was phytochemically tional medicine to treat heart problems, nausea in preg­ screened for alkaloids, saponins, tannins, cardiac gly­ nant women, intestinal ulcers, headaches, stomach cos ides and cyanogenic glycosides. The results complaints, haemorroids, diarrhoea, dyspepsia and to obtained for material from both collection sites was hasten the onset of labour. D. rotundifolia material was very similar. No alkaloids and no cyanogenic glycosides collected from two sites and screened pharmacologi­ were detected. Tannins were present in the leaf and cally for anti-inflammatory and anti-bacterial activity young shoot material. Saponins were detected in the using the COX-1 and disc-diffusion assays. High anti­ bark material using the haemolysis test. Cardiac glyco­ inflammatory activity was detected in the ethanolic and sides were detected in the leaf, young shoot and bark dichloromethane leaf and young shoot extracts. Anti­ material. bacterial activity was highest in the ethanoHc leaf and Introduction The prime advantage of traditional medicine is that it is an in pregnant women. Decoctions of the bark are sometimes immediate, exis ting source of health care fo r people where used in delayed labour, to hasten the onset of the process. they live. It may not be as 'good' as what may be considered The preparation is thought also to be used for procuring ideal, but. in the absence of better alternatives use should abortion (Watt and Breyer-Brandwijk 1962, Hutchings et al. be made of what is available (Iwu 1993). Evaluation and 1996, Van Wyk et a/. 1997, Thomas and Grant 1998). screening of routinely used medicinal plants is becoming a Aqueous infusions of the bark or wood are used as enemas necessity to ensure the acceptance of the plants by the or are taken ora lly for tre atment of intestinal ulcers, medical doctors and health care authorities as potential aids headaches, stomach complaints, haemorroids and diar­ or cures for a number of ailments. rhoea (Watt and Breyer-Brandwijk 1962, Coates-Palgrave et Dombeya rotundifolia (Hochst.) Planch belongs to the al. 1985, Hutchings el al. 1996, Van Wyk et al. 1997 , Sterculiaceae which is found throughout the world (Palmer Thomas and Grant 1998). and Pitman 1961). This fam ily is often referred to as the The roots are made into a tonic and administered as ene­ chestnut family (Van Wyk 1974). It comprises approximately mas for dyspepsia and sharp pains in the stomach. In 1 200 species (50 genera). mainly trees and shrubs found in Tanzan ia the Zigula use the roots as a remedy for abdomi­ tropical and subtropical regions. Many of the African mem­ nal pains. It is also used as a colic remedy (Watt and Breyer­ bers of the fami ly serve as so urces of medicine, fibre, fire­ Brandwijk 1962, Coates-Palgrave el a/. 1985, Hutchings et wood. timber for furniture, and as decorative plants (Van al. 1996 , Venter and Venter 1996). In Zarnbia, the leaves are Wyk 1974). rubbed on abscesses as a counter irritant (Watt and Breyer­ D. rotundifolia, commonly known as the wi ld pear, is a sin­ Brandwijk 1962). gle-stemmed deciduous tree that grows to about 5-6m tall, No phytochemical and pharmacological studies of this with a moderate, irregular-shaped canopy (Irnrnelman el al. species have previously been undertaken, except for the 1973). It occurs in woodlands over a wide range of altitudes screening for alkaloids (Raffauf 1996). Th is study involved in Mpumalanga and KwaZulu-Nata l (Coates-Palgrave et af. the screening of D. rotundifolia for biologically active corn­ 1985) and is more abundant in the warmer, drier habita ts pounds. Plant material was obtained from two different sites (Immelman et al. 1973). to determine whether there is a difference in the chemical The inner bark is used to treat heart problems and nausea composition of the plant from widely different localities. 350 Reid. Jager and Van Sta den Materials and Methods Cardiac glycosides Plant material Extracts were screened for 2-deoxy -s ugars using the Keller­ Killiani test (Evans 1989, Jager and Van Staden 1995a). If a Plant material of D. rotundifolia was obtained from two sites : reddish-brawn-purple ring appears at the interphase, and 1) Umgeni Valley Nalure Reserve, Hawick, KwaZulu-Natal, the upper layer slowly becomes bluish-green , it is an ind ica­ and 2) Manato, a private nature reserve in the Oroogekloof tion that 2-deoxy-sugars are prese nt. suggesting the pres­ area, approximately 30km from Wa rmbaths in the Northern ence of cardiac glycosides. Province. Voucher specimens were deposited in the The extracts were screened for unsa turated lactone rings University of Na tal Herbarium, Pietermaritzburg, under the (Jager and Van Staden 1995a). A TLC plate was sp rayed numbers: REID1 UN and REI D2UN. The habitats varied with with Kedde's reagent (Wagner et a/. 1984) to test for the respect to climate, altitude and soil type (Ta ble 1). Leaf. presence of cardenolides . A second and th ird TLC pla te shoot and outer bark material was collected and dried at were sprayed with ch loramine T-trichloraacetic acid reagent SO ' C for 72h . The plant material was homogenously ground and antimony (III) ch loride reagent respectively (Wagner et and stored in the da rk at room temperatu re in airtight con ­ a/. 1984) to test for the presence of bufadienolides. tainers until further processing. Cyanogenic glycosides Extraction of plant material Pl ant materia l wa s screened for cyanog en ic glycos!des Ground, dried plant material (leaves, shoot and bark) was using the Grignard test (Jager el al. 1995b). A positive con­ extracted with water, ethyl acetate and ethanol respectively trol was prepared containing 0.1mg amygdalin. (100mg ml ') in an ultrasonic bath for 30m in . The extracts were fi ltered and ai r-d ri ed ove rn ight. They were fe-suspend ­ Saponins ed in the same solvents at SOmg resid ue mi l (anti-bacterial activity) and 2.Smg re sidue ml ' (anti-inflammatory activity) Plant material was tested for saponins using the haemolysis respectively. test (Fang el a/. 1974). Colu mbia Blood Agar Base (Oxoid CM 33 1) was prepa red . Fifteen ml of sterile 20% sodium cit­ Anti-bacterial activity rate was added to SOOml of who le blood. Aseptica lly, S0-70ml sterile citra ted catlle blood was added to the Columbia Blood The plant extracts were teste d for anti-bacterial activity Agar sol ution . Distilled water was used as a negative co ntrol using the disc-d iffusion assay (Ra soanaivo and and Saponaria officinalis extract as a positi ve control. Ratsimamanga-Urverg 1993) . Activity was tested against six strains of bacte ria: Bacillus sublilis (ATCC 60S1 ), Tannins Escherichia coli (ATCC 1177S), Klebsiella pneumoniae (ATCC 13883), Micrococcus luleus (ATGC 4698), Ta nnins were detected using the Gelatin-Salt Block test Staphylococcus aureus (ATCC 12600) and Staphylococcus (Duncan el al. 1999). epidermidis. Stock cultu res were stored at 4"C on nutri ent agar plates. Anti-bacterial activity was expressed as the ratio Alkaloids of the inhibition zone (mm) produced by the plant extract to the inhibition zone caused by the neomycin control (Rabe pH part ition ing for alka loids was performed according to and Van Staden 1997). Brimer el al. (1989). Dragendorff's reagent (Wagner et a/. The Minimal Inhibition Concentration (MIG) technique , 1984) was added to one part of the prepared extra cts , and developed by Eloff (1998), using 96-well microplates, was Mayer's reagent (Wagner el a/. 1984) to the other. used to determine the MIC's of extracts. Activity was tested Observations were made for the development of a red­ against 4 strain s of bacteria . K. pneumoniae and E. coli orange precipitate on th e addition of Dragendorff's reagent, (Gram-negative) and B. subtilis and S. aureus (Gram-posi­ and a white precipi tate on the addition of Mayer's reagent. tive). Bacterial growth in the wells was indicated by a red colour. Clear wells indicated inhibition by the test substance . Results and Discussion Anti-inflammatory activity The ethyl acetate , ethanol and water extracts, from plant material collected at both coll ection si tes in March 1999, Anti-inflammatory activity of leaf, shoot and bark extracts of showed varying degrees of anti-bacterial activity (Ta ble 2). O. rolundifolia was determined using the prostaglandin-syn­ Highest activity was recorded for the ethanolic leaf extracts thesis inhibitor assay (GOX-1) (Jager el a/. 1996). Inhibition from both collection sites, they were bacteriostatic against of the test solutions was obtain ed by analysing the amount M. luleus (0.83 and 0.71 respectively). Bacteriostatic of rad ioactivity present relative to the radioactivity in the sol­ extracts are those which prevent the mul tiplying of bacteria vent blank (McGaw el a/. 1997). without destroying them. A zone of inhibitio n forms which is not clear. Etha noJic shoot extracts were also bacteriostatic against S. aureus (1 .22).The ethanol extract fro m Mona to showed anti-bacteria l activi ty against M. luteus (0 .89). Bark material showed Jaw anti-bacterial activity.