EFFECT OF MEDICINAL PLANTS FROM PAKISTAN ON INSULIN SECRETION 73

Copyright © 2004 John Wiley & Sons, Ltd. Phytother. Res. 18, 73–77 (2004)

Copyright © 2004 John Wiley & Sons, Ltd.

Received 19 June 2001Accepted 4 April 2003

PHYTOTHERAPY RESEARCHPhytother. Res. 18, 73–77 (2004)Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/ptr.1372

The Effect of Medicinal Plants of Islamabadand Murree Region of Pakistan on InsulinSecretion from INS-1 Cells

Zakir Hussain1*, Abdul Waheed2, Rizwana Aleem Qureshi3, Dadu Khan Burdi1, Eugen J.Verspohl2, Naeema Khan1 and Mashooda Hasan1

1Department of Chemistry, Quaid-i-Azam University, Islamabad Pakistan2Institut für Pharmazeutische Chemie und Pharmakologie, Westfälische Wilhelms-Universität, Münster, Germany3Department of Biology, Quaid-i-Azam University, Islamabad Pakistan

In vitro testing of the extracts of medicinal plants collected from Islamabad and the Murree region on insulinsecretagogue activity was carried out. Dried ethanol extracts of all plants (ZH1-ZH19) were dissolved inethanol and DMSO, and tested at various concentrations (between 1 and 40 µµµµµg/mL) for insulin release fromINS-1 cells in the presence of 5.5 mM glucose. Glibenclamide was used as a control. Promising insulinsecretagogue activity in various plant extracts at 1, 10, 20 and 40 µµµµµg/mL was found, while in some cases adecrease in insulin secretion was also observed. Artemisia roxburghiana, Salvia coccinia and Monsteradeliciosa showed insulin secretagogue activity at 1 µµµµµg/mL (p <<<<< 0.05) while Abies pindrow, Centaurea ibericaand Euphorbia helioscopia were active at 10 µµµµµg/mL (p <<<<< 0.05). Extracts of Bauhinia variegata and Bergeniahimalacia showed effects at 20 µµµµµg/mL (p <<<<< 0.05), and Taraxacum officinale and Viburnum foetens at40 µµµµµg/mL (p <<<<< 0.05). Insulin secretagogue activity could not be detected in the extracts of Adhatoda vasica,Cassia fistula, Chrysanthemum leucanthemum, Morus alba, Plectranthus rugosus, Peganum harmala andOlea ferruginea. The results suggest that medicinal plants of Islamabad and the Murree region of Pakistanmay be potential natural resources for antidiabetic compounds. Copyright © 2004 John Wiley & Sons, Ltd.

Keywords: medicinal plants; insulinotropic activity; diabetes mellites; INS-1 cells.

* Correspondence to: Z. Hussain, Department of Chemistry, Quaid-i-Azam University, Islamabad 45320, Pakistan.E-mail: chem63@yahoo.comContract/grant sponsor: Quaid-i-Azam University.Contract/grant sponsor: DAAD Fellowship.Contract/grant sponsor: Alexander von Humboldt Stiftong; Contract/grantnumber: V-815/98074-PAK/1010276.

INTRODUCTION

The aim of the present study was to look into the effectof these medicinal plants on insulin secretion fromINS-1 cells. INS-1 cells secrete insulin without para-crine influence, i.e. by secreted glucagon or somatostatin;they respond to glucose and major insulinotropicpeptide (Asfari et al., 1992).

A review (Rahman and Zaman, 1989) summarizingthe literature on the anti-diabetic activity of 343 medi-cinal plants in the indigenous system of medicine wascarried out. Plants such as Adhatoda vasica (Claesonet al., 2000), Abies pindrow (Singh and Pandey, 1997;Singh et al., 2000), Artemisia roxburghiana (Bicchiet al., 1998), Bauhinia variegata (Sharma and Saxena,1996), Centaurea iberica (Shamyanov et al., 1998), Cassiafistula (Agarkar and Jadge, 1999), Euphorbia helioscopia(Park et al., 2001), Morus alba (Chen et al., 1995),Taraxacum officinale (Yasukawa et al., 1998) andPeganum harmala (Lamchouri et al., 1999) have exten-sively been investigated. Various chemical constituentshave been isolated from these plant species, so wedecided to select the only plant species or the part of

the plant, which have not been previously investigatedfor secretagogue activity.

Seventeen plant species were collected and depos-ited in the Herbarium, Department of Biology, Quaid-i-Azam University, Islamabad, where plants wereauthenticated, voucher numbers were given and speci-mens were stored (voucher numbers ZH1-103526,ZH2-123465, ZH3-115756, ZH4-21305, ZH6-113282,ZH8-44919, ZH9-5243, ZH10-14548, ZH11-44254,ZH12-22521, ZH13-5587, ZH14-17096, ZH15-57170,ZH16-123468, ZH17-66672, ZH18-115669 andZH19-8132).

MATERIALS AND METHODS

(Mono-1251-Tyr A14)-porcine insulin was from Hoechst(Frankfurt, Germany) rat insulin was from NovoNordisk (Bagsvaerd, Denmark) and anti-insulin anti-bodies were from Linco (St. Louis, USA).

Preparation of plant extracts

Air dried and coarsely milled plant material (30 g) wasexhaustively extracted by the method of percolation atroom temperature with 80–90% ethanol in an aspiratorbottle for seven days. The procedure was repeated threetimes. Evaporation of the combined extracts underreduced pressure yielded semisolid brownish material,the amount varied for different plants (Table 1).

74 Z. HUSSAIN ET AL.

Copyright © 2004 John Wiley & Sons, Ltd. Phytother. Res. 18, 73–77 (2004)

Extracts ZH1-ZH19 and Glibenclamide were dis-solved in DMSO and ethanol respectively. Final con-centrations up to 0.5% DMSO and 0.1% of ethanolhave been shown to be without any effect on insulinrelease and were well tolerated by the cells.

Cell Culture. INS-1 cells (insulinoma cell line) providedby Dr C. Wollheim, Geneva, Switzerland, were grownin microwells for 4–6 days (half confluence: 1–2 × 106

cells ml−1) in 300 µl RPMI medium supplemented with10% (v/v) fetal calf serum, 100 U of penicillin ml−1 and0.1 mg of streptomycin ml−1. Prior to the experiment,cells were washed two times and then incubated inKrebs-Ringer buffer (for 90 min) containing 10 mMHEPES and 0.5% bovine serum albumin (KRBH).

Insulin Release. To measure insulin secretionhalf-confluent cells in microwells were incubatedfor 90 min at 37 °C in the aforementioned KRBHbuffer in the presence of basic glucose concentra-tion (3.0 mM, substimulatory as control) and 5.6 mMglucose (stimulatory).

Insulin released into the medium was assayed with aradioimmunoassay using rat insulin as a standard,(mono-1251-Tyr A14)-porcine as the labeled compoundand anti-insulin antibodies. Each extract tested in thisinvestigation had been checked for non-interferencewith the insulin radioimmunoassay and results areshown as mean ± SEM Statistical significance wasdetermined using one way of analysis of variance(ANOVA) (RS/1 statistics pack, BBN Software Prod-ucts Corp.) followed by a post-hoc test (Newman Keuls).A p value > 0.05 was considered significant.

RESULTS

Effect of the extracts of 17 plants (Table 1) on insulinrelease from INS-1 cells in the presence of 5.6 mMglucose was determined. An increased insulin releasewas observed with 5–6 mM glucose compared to 3.0 mMglucose (a substimulatory glucose concentration).Glibenclamide (standard anti-diabetic compound) wasused in 10−3 M quantity as a control substance increasedinsulin release in the presence of 5.6 mM glucose.

Figure 1. Insulin stimulatory effect of ZH2, ZH3 and ZH4.

Ethanolic extracts (ZH1-ZH19) of the plantshave been investigated at various concentrations (1–40 µg/ml) for the insulin secretagogue activity andresults obtained are summarized in Table 1.

An insulin stimulatory effect was shown for the ex-tracts ZH2, ZH3, ZH4 (Fig. 1), ZH6, ZH8 (Fig. 2),ZH11 (Fig. 3) ZH14, ZH15, ZH16 (Fig. 4) and ZH18(Fig. 5) at various concentrations whereas we could notfind the insulin stimulatory effect in the extracts suchas ZH1, ZH9, ZH10, ZH12, ZH13, ZH17 and ZH19.Some extracts such as ZH9 and ZH10 showed a re-verse effect, i.e. decrease in insulin stimulatory effect.

DISCUSSION

The results obtained in the present work showed that itwas worthwhile to investigate some medicinal plantswhich have not been previously tested for insulinsecretagogue activity. Ethanolic extracts of the plantssuch as ZH3, ZH6, ZH8, ZH11, ZH14 and ZH16 whichhave not been investigated in previous studies haveshown insulinotropic activity in the preliminary in vitrotesting (Table 1).

Figure 3. Insulin stimulatory effect of ZH11.

Figure 2. Insulin stimulatory effect of ZH6, ZH8.

EF

FE

CT

OF

ME

DIC

INA

L P

LA

NT

S FR

OM

PA

KIST

AN

ON

INSU

LIN

SEC

RE

TIO

N75

Copyright ©

2004 John Wiley &

Sons, Ltd.

Phytother. R

es. 18, 73–77 (2004)

Table 1. Insulin secretagogue activity found in ethanolic extracts of different plant species

Plant No.

ZH1

ZH2

ZH3

ZH4

ZH6

ZH8

ZH9

ZH10

ZH11

ZH12

ZH13

ZH14

ZH15

ZH16ZH17

ZH18

ZH19

Plant/Family

Adhatoda vasica Nees/AcanthaceaeAbies pindrow Royle/pinaceaeArtemisia roxburghianaWall. ex Besser/AsteraceaeBauhinia variegata L./CaesalpiniaceaeBergenia himalacia Boiss./SaxifragaceaeCentaurea iberica Trev. exSpreng/AsteraceaeCassia fistula L./CaesalpiniaceaeChrysanthemumleucanthemum L./AsteraceaeEuphorbia helioscopia L./EuphorbiaceaeMorus alba L./MoraceaePlectranthus rugosus Wall.ex Bth/LamiaceaeSalvia coccinia Juss. exMurr/LamiaceaeTaraxacum officinaleWeber/AsteraceaeViburnum foetens Dcne/CaprifoliaceaePeganum harmala L.ZygophyllaceaeMonstera deliciosaAraceaeOlea ferruginea RoyleOleaceae

Place/date ofcollection

Uni. Campus/30.03.1998Murree/14.06.1998

Murree/14.06.1998

Uni. Campus/28.03.1998Murree/14.03.1998

Uni. Campus/30.03.1998Uni. Campus/20.03.1998Murree/14.03.1998

Uni. Campus/24.03.1998Uni. Campus/03.04.1998Murree/14.06.1998

Uni. Campus/March,1996Uni. Campus/20.03.1998Murree/14.06.1998

Chakwal/June, 1998

Uni. Campus/Dec.,1998Murree/March, 1999

Part of plantcollected

Leaves/flowers

Leaves

Leaves

Leaves

Roots

Leaves

Leaves

Leaves

Leaves andflowersLeaves

Leaves

Leaves

Whole plant(excluding roots)Leaves

Hull of seeds

Fruit pulp

Leaves

Amount ofextract obtd.

g, %

4, 13.3

6.38, 21.3

5, 16.6

4.6, 15.3

7, 23.3

3, 10

4.7, 15.6

5, 16.6

4.5, 15

5, 16.6

6, 20

7, 23.3

4, 13.3

6, 20

7, 23.3

10, 33.3

3, 10

Insulinsecretion(% basal)

No activity

62

74

140

38

62

No activity

No activity

26

No activity

No activity

50

112

125

No activity

40

No activity

Activity alreadyreported

Reported in leaves

Reported in wholeplant excluding rootsNot investigated

Reported in wholeplantNot investigated

Not investigated

Reported in stem bark

Activity investigatedin whole plant but noactive principleNot investigated

Reported in leavesand root barkNo activity

Not investigated

Reported in wholeplantNot investigated

Reported in seeds

Not investigated

Not investigated

Previous literature onhypoglycemic activity

(Ref.)

Rahman and Begum,1966Singh et al., 2000

–

Singh and Pandey,1997–

–

Shamyanov et al.,1998Akhtar et al., 1985

–

Sharma and Saxena,1996–

–

Yasokawa et al., 1998

–

Akhtar et al., 1985

–

–

Maximum Insulinstimulatory effect

shown (µµµµµg/ml)

–

10

1

20

20

10

–

–

10

–

–

1

40

40

–

1

–

76 Z. HUSSAIN ET AL.

Copyright © 2004 John Wiley & Sons, Ltd. Phytother. Res. 18, 73–77 (2004)

alba L. Another point worth noting is that the testingwas done in normal as well as in alloxan treated ani-mals. It is therefore possible that the extract pre-viously used had precursors of the original compoundand that precursors can only be transformed under thein vivo testing conditions into the active agent. Theother possibility is that the extract used in previousstudie may act as a stimulant of the actual insulinotropicagent (insulin present in the body) and may not act asa stimulant under in vitro conditions.

The hypoglycemic activity of some of the drugs iswell established. We now looked at one of the possiblemechanisms, i.e. insulin release. Some of the extractsshowed a clear insulinotropic activity, e.g. ZH2,ZH4 and ZH15. Several other extracts showed aninsulinotropic effect without having been tested ashypoglycemic agents: ZH3, ZH6, ZH16 and ZH18.It would be interesting to further evaluate these usingin vivo studies on hypoglycemic effect.

It is worth noting here that the ethanol extracts ofthe drugs do not contain any glucose; the cells werechallenged with glucose simultaneously with the addi-tion of the drug extracts, to observe the effects. Wemay consider that the extracts may merely be disrupt-ing the membrane of the INS-1 cells thereby causing arelease of the stored insulin. However, this possibilityis not likely since a disruption of cells would give morethan a 500 fold higher insulin level (leakage of cells)and insulin release in the cells is always controlled inthe absence of Ca2+ in the medium by which no insulinrelease would be seen. Therefore, all plant extractswhich showed an insulin secretagogue activity in thepresent in in vitro testing may be a good source ofactive hypoglycemic principle if investigated further.

The Effects of various extracts (ZH1-ZH19) oninsulin release from INS-1 cells in the presence of5.6 mM glucose are presented in Figs 1–5. The lowglucose concentration (3.0 mM) and glibenclamide(= glyburide; 10−3 M) served as negative and positivecontrols. The results are expressed as % secretioninsulin at 5.6 mM glucose alone. Each value representsthe mean ± SEM of three independent experiments(done on different days) run duplicates (each of thethree experiments done twice). In statistical analysis ap value > 0.05 was considered significant.

Acknowledgements

The research fellowship to Zakir Hussain (M Phil student) by Quaid-i-Azam University is gratefully acknowledged.

Dr Dadu Khan Burdi, Associate Professor, Department ofChemistry, University of Sindh, Jamshoro, Pakistan is grateful to theQuaid-i-Azam University for the award of a Post-Doctoral Fellow-ship, which enabled him to participate in this project.

Dr Abdul Waheed, Assistant Professor, Ayub Medical College,Abbottabad, Pakistan, expresses his thanks for the award of theDAAD fellowship for a study visit to the Institut für PharmazeutischeChemie und Pharmakologie, Westfälische Wilhelms-Universität,Münster, Germany.

Professer M. Hassan gratefully acknowledges the donation of equip-ment for the project by Alexander von Humboldt Stiftung, grant No.V-815/98074-PAK/1010276.

Figure 4. Insulin stimulatory effect of ZH14, ZH15 and ZH16.

Figure 5. Insulin stimulatory effect of ZH18.

It is evident from this study that all the results ob-tained are not in conformity to those reported in previ-ous studies using alternative methods had been used.Some of the plants like Adhatoda vasica Nees (Modakand Rao, 1966), Cassia fistula L. (Dhar et al., 1968) andMorus alba L. (Perry, 1980) for which activity had beenreported in the literature, did not show any activity inthe present in vitro testing, whereas others such as Abiespindrow Royle (Bhakuni et al., 1971), Bauhinia variegataL. (Rahman and Begum, 1966) and Taraxacum officinaleWeber (Akhtar et al., 1985) showed results which werein conformity with previous investigations. It is pos-sible that the hypoglycemic activity is dependent onthe part of the plant used. In our preliminary investiga-tions we used the ethanolic extract of the leaves andflower buds of Adhatoda vasica and leaves of Morusalba L. While previous studies have utilised the leavesof Adhatoda vasica and leaves and root bark of Morus

EFFECT OF MEDICINAL PLANTS FROM PAKISTAN ON INSULIN SECRETION 77

Copyright © 2004 John Wiley & Sons, Ltd. Phytother. Res. 18, 73–77 (2004)

Agarkar SV, Jadge DR. 1999. Phytochemical and pharmacologi-cal investigations of genus Cassia: A review. Asian J Chem11: 295–299.

Akhtar MS, Khan QM, Khaliq T. 1985. Effects of Portulaca-oleraceae and Taraxacum officinale in normoglycemic andalloxan-treated hyperglycemic rabbits. J Pak Med Assoc35(7): 207–210.

Asfari M, Janjic D, Meda PLiG, et al. 1992. Establishment of2-mercaptoethanol-dependent differential insulin-secretingcell lines. Endocrinology 130: 167–178.

Bhakuni DS, Dhar ML, Dhar MM, et al. 1971. Screening of In-dian plants for biological activity Part 3. Ind J Exper Biol9(1): 91–102.

Bicchi C, Rubiolo P, Marschall H, et al. 1998. Constituents ofArtemisia roxburghiana essential oil. Flavour Fragrance J13: 40–46.

Chen FJ, Nakashima N, Kimura I, et al. 1995. Potentiating ef-fects of pilocarpine-induced saliva secretion, by extracts andN-containing sugars derived from mulberry leaves, instreptozocin-diabetic mice. Biol Pharmaceut Bull 18:1676–1680.

Claeson UP, Malmfors T, Wikman G, Bruhn JG. 2000. Adhatodavasica: A critical review of ethnopharmacological andtoxicological data. J Ethnopharmacol 72: 1–20.

Dhar ML, Dhar MM, Dhawan BN, et al. 1968. Screening ofIndian plants for biological activity. Ind J Exper Biol 6:232–247.

Lamchouri F, Settaf A, Cherrah Y, et al. 1999. Antitumour prin-ciples from Peganum harmala seeds. Therapie 54: 753–758.

Modak AT, Rao MRR. 1966. Hypoglycaemic activity of a non-nitrogenous principle from the leaves of Adhatoda vasicaNees. Ind J Phar 28: 105.

Park H, Koh D, Lee S, et al. 2001. Anti-allergic and anti-asthmatic activity of Helioscopinin-A, a polyphenol com-pound isolated from Euphorbia helioscopia. J MicrobiolBiotech 11: 138–142.

Perry LM. 1980. Medicinal Plants of East and South East AsiaMIT Press: London.

Rahman A, Zaman K. 1989. Medicinal plants with hypoglycemicactivity. J Ethnopharmacol 26(1): 1–55.

Rahman W, Begum S. 1966. Flower pigments: flavonoids fromthe white flowers of Bauhinia variegata. Naturwissen 53:385.

Shamyanov ID, Akhmedov UA, Saidkhodzhaev AI. 1998.Sesquiterpene lactones and other components of Centaureaiberica. Khimiya Prirodnykh Soedinenii 3: 377–379.

Sharma RN, Saxena VK. 1996. In vitro antimicrobial efficacy ofleaves extracts of Bauhinia variegata. Asian J Chem 8:811–812.

Singh RK, Bhattacharya SK, Acharya SB. 2000. Pharmacolog-ical activity of Abies pindrow. J Ethnopharmacol 73:47–51.

Singh RK, Pandey BL. 1997. Further study of antiinflammatoryeffects of Abies pindrow Phytother Res 11: 535–537.

Yasukawa K, Akihisa T, Inoue Y, et al. 1998. Inhibitory effect ofthe methanol extracts from compositae plants on 12-O-tetradecanoylphorbol-13-acetate-induced ear oedema inmice. Phytother Res 12: 484–487.

REFERENCES