Journal Elsevier
of Ethnopharmacology, Scientific
Publishers
283
31 (1991) 283-289 Ireland Ltd.
POSSIBLE HEPATOTOXICITY DREGEA VOLUBILIS LEAVES
OF NIGELLA
KAMANI H. TENNEKOONa, S. JEEVATHAYAPARAN4 ARUNDATHIE P. KURUKULASOORIYA’ and ERIC
SATIVA
SEEDS
AND
H. KARUNANAYAKEb
“Department of Physiology, bDepartment of Biochemistry and “Department of Pathology, Faculty of Medicine, University of Colombo, PO Box 271, Colombo 8 (Sri Lanka) (Accepted
September
11, 1990)
Summary
Aqueous extracts of the seeds of Nigellu sativa and mature leaves of Dregea volubilis were administered orally under light ether anaesthesia to male Sprague-Dawley rats for 14 days. Key hepatic enzyme concentrations and histopathological changes in the liver in both treatment groups at the end of 14 days were compared with a control group which received distilled water under identical conditions for 30 days and with a group of normal animals. Serum gamma-glutamyl transferase concentrations were significantly increased in both extract groups while serum alkaline phosphatase concentrations were significantly increased following administration of only D. volubilis when compared with either the control or the normal group. Serum alanine aminotransferase concentrations were significantly increased in both extract groups when compared with the normal group but not with the control group. Degenerative changes in hepatocytes were seen following administration of D. volubilis while consistent significant histopathological changes were not evident following administration of N. satiwa.
Introduction
During investigations to evaluate the purported galactogogue activity of Nigella sativa L. (family, Umbelliferae; Sinhala: ‘Kaluduru’) and Dregea voluCorrespondence to: Kamani H. Tennekoon, Department versity of Colomho, PO Box 271, Colombo 8, Sri Lanka.
0378s8741lt03.50 Published
0 1991 Elsevier and Printed in Ireland
Scientific
Publishers
of Physiology,
Ireland
Ltd.
Faculty
of Medicine,
Uni-
284
bilis L. (family, Asclepiadaceae;
Sinhala: ‘Anguna’l a high incidence of mortality was observed among the experimental animals, l-2 weeks after completion of the experiment. In addition to its possible galactogogue properties, N. sativa is considered to have emetic and expectorant potential while D. volubilis is thought to have diuretic and anthelmintic properties (Jayaweera, 19821. Considering the extensive usage of these plants in the Ayurveda system of Medicine and our own observation of high mortality among experimental animals during our studies, it was thought necessary to investigate the possible hepatotoxicity of these two plants. Materials and Methods Preparation of plant extracts N. sativa seeds were bought
from an Ayurvedic drug store. These were washed to remove sand and other debris and air-dried. The seeds (250 gl were boiled in distilled water (1000 ml) for 90 min and filtered through muslin. The final volume of the filtrate obtained was 320 ml. The mature leaves (258 gl of D. volubilis were shredded and boiled in distilled water (1500 ml) for 90 min. The extract was squeezed and filtered through muslin. The final volume of the filtrate obtained was 690 ml.
Animal
experiments
Five-month-old healthy male Sprague - Dawley rats housed under standard conditions and liberally fed with rat pellets (Moosaji’s Ltd., No. 34, W.A.D. Ramanayake Mawatha, Colombo 21 and water were used for this study. Thirty-two rats were selected for the study and were randomly divided into 2 groups of 8 each and a group of 10. Aqueous extracts of N. sativa and D. volubilis were administered orally to rats in Group 1 Vv = 8) and Group 2 Vv = 81, respectively, at a daily dose of 10 ml/kg for 14 consecutive days under light ether anaesthesia. Group 3 (control group, N = 101 received an equal amount of distilled water for 30 days under identical conditions and served as a common control group for the two experiments. At the end of the experimental period (i.e. 14 days for both treatment groups and 30 days for the control group), animals were sacrificed and 5 ml of blood was collected by cardiac puncture for estimation of key hepatic enzymes. Serum was separated and stored at - 20°C till gammaalkaline phosphatase, alanine aminotransferase and glutamyl transferase, aspartate aminotransferase concentrations were measured using methods described by King et al. (19421, Reitman and Frankel (19571 and Rosalki and Rau (1972). Livers were harvested for histopathological examinations. Gamma-glutamyl transferase, alkaline phosphatase, alanine aminotransferase and aspartate aminotransferase concentrations in the two treatment groups and in the control group were compared using the Student t-test.
285
Further, these enzyme concentrations in the treatment groups and in the control group were also compared with the enzyme concentrations that have been established in our laboratory for 25 normal healthy Sprague-Dawley rats of the same age, again using the Student-t-test. Results and Discussion Serum gamma-glutamyl transferase and alkaline phosphatase concentrations (mean f S.E.M.) in the two treatment groups, control group and in normal healthy rats are shown in Fig. 1. Serum alanine aminotransferase and aspartate aminotransferase concentrations (mean + S.E.M.) for the same groups are shown in Fig. 2. Serum gamma glutamyl transferase concentrations (mean & S.E.M.) were 3.29 r 0.49, 3.99 f 0.54, 1.36 f 0.13 and 1.64 + 0.59 U/l in groups treated with N. sativa, D. volubilis and in the control and normal groups, respectively. These serum concentrations were significantly P < 0.001) increased following administration of N. sativa and D. volubilis relative to both the control group and the normal group. Serum alkaline phosphatase concentrations (mean f S.E.M.) were 43.70 + 3.88, 59.48 f 5.66, 45.7 + 0.85 and 42.24 f 1.66 K-A units1100 ml in groups treated with N. sativa. D. volubilis
Fig.
q NIGELLA SATIVA
H
CONTROL GROUP
q DREGEA VOLUBILIS
Cl
NORMAL GROUP
1. Serum
gamma-glutamyl
transferase
and alkaline
phosphatase
concentrations
(mean
+
S.E.M.) in male Sprague-Dawley rats following 10 ml/kg oral administration of boiling water extracts of the seeds of Nigello sativa W = 8) and mature leaves of Dregea volubilis W = 8) for 14 days and distilled water for 30 days (control group, N = 10) under light ether anaesthesia and in a group of normal animals W = 25).
286
q NIGELLA SATIVA
ggl
CONTROL GROUP
•3 DREGEA VOLUBILIS
Cl
NORMAL GROUP
Fig. 2. Serum aspartate aminotransferase and alanine aminotransferase concentrations (mean + S.E.M.) in male Sprague-Dawley rats following 10 ml/kg oral administration of boiiing water extracts of the seeds of Nigello sativa W = 8) and mature leaves of Dregea vo~ub~l~ Uv = 81 for 14 days and distilled water for 30 days (control group, N = 10) under light ether anaesthesia and in a group of normal animals W = 25).
and in the control and normal groups, respectively. Thus, the serum alkaline phosphatase concentration following administration of D. volubilis was significantly higher P < 0.0011 when compared with the control and normal groups while N. sativa administration did not have a significant effect on this enzyme. Further, there were no significant differences in regard to and alkaline phosphatase concentrations gamma glutamyl transferase between the control and the normal groups indicating that the transient but chronic ether administration experienced by the control group did not affect these enzymes. Serum gamma-glutamyl transferase and alkaline phosphatase are membrane-bound enzymes and are released unequally depending on the pathologiElevation cal phenomenon. of serum gamma-glutamyl transferase concentration is generally regarded as one of the most sensitive indices of hepatic damage (Szczeklik et al, 19611. In the present study, serum gammaglutamyl transferase concentrations were significantly increased following oral administration of both N. sativa and D. volubilis. Serum alkaline phosphatase concentrations are known to be markedly elevated in cholestasis and to be minimally increased in chronic hepatocellular disease (Wilkinson, 19761. The significantly increased serum alkaline phosphatase concentrations seen
287
following administration of D. volubilis indicate a difference in toxicity between this plant and N. sativa. Serum alanine aminotransferase concentrations (mean f S.E.M.1 were 18.04 + 2.40, 19.85 + 1.93, 15.16 + 1.45 and 12.7 rt 0.67 U/100 ml in the groups treated with N sativa, D. volubilis and in the control and normal groups, respectively, Serum alanine aminotransferase concentrations were significantly higher following administration of N. sativa P < 0.01) and D. voEubiZis P < 0.001) when compared with the normal group but not when compared with the control group. Alanine aminotransferase concentrations were higher in the control group when compared with the normal group but this difference was not statistically significant. Serum aspartate aminotransferase concentrations (mean -+ S.E.M.) were 13.13 1: 0.96, 13.13 & 1.77, 12.18 + 1.22 and 11.7 rt: 0.69 U/100 ml in the groups treated with N. sativa, D. volubilis and in the control and normal groups, respectively and hence were not significantly different from one another. Alanine aminotransferase is a cytoplasmic enzyme found in very high concentrations in the liver. Aspartate aminotransferase is present in the cytoplasm as well as in the mitochondria and is less specific than alanine aminotransferase as an indicator of hepatic damage since it is rapidly inactivated (Wilkinson, 1976). The oral administration of neither of the plant extracts used in this study nor chronic ether anaesthesia by itself appears to have a significant effect on the release of alanine aminotransferase. However, the significant increase in alanine aminotransferase concentrations seen following oral administration of N. sativa and D. volubilis when compared with normal animals may indicate that administration of these plant extracts when superimposed on chronic ether anaesthesia may cause an increased release of this enzyme. However, neither oral administration of the two plant extracts nor chronic ether anaesthesia appears to have a significant effect on aspartate aminotransferase concentrations. Histopathological examination of livers showed degenerative changes in hepatocytes in 6 out of 8 animals treated with D. volubilis. There was no evidence of degenerative changes in hepatocytes in animals treated with N sutivu, in control animals or in normal animals. Thus the degenerative changes in the hepatocytes in animals treated with D. volubilis appears to have been caused specifically by D. volubilis. The significant elevation of gamma-glutamyl transferase and alkaline phosphatase concentrations seen following oral administration of D. volub~l~ when compared with either the control or the normal group and the significant elevation of alanine aminotransferase concentrations when compared with the normal group may reflect the hepatic damage thus caused. Central veins and the sinusoids around them were dilated in 7 out of 8 animals treated either with N. sativa or D. volubilis and in 3 out of 10 control animals while the portal veins and the sinusoids around them were dilated in 5 out of 8 animals treated either with N. sativa or D. vo~u~~l~sand in 5 out of 8 control animals. Dilatation of central veins, portal veins or sinu-
288
soids was not seen in any of the normal animals. It may be that chronic ether anaesthesia led to dilatation of the central veins and surrounding sinusoids while oral administration of iV. sat&a and I). voZ~b~Z~ enhanced this effect. The prevalence of dilatation of portal veins and surrounding sinusoids was comparable in the two treatment groups and in the control group indicating that this may have been caused by chronic ether anaesthesia and that the oral administration of the plant extracts did not have a superimposing effect. Infiltration by mononuclear chronic in~ammatory cells was observed in the vicinity of the portal tracts in 6 out of 8 animals treated with N sat&a, in all control animals, in 5 out of 25 normal animals and in none of the animals treated with D. volubilis. This finding is rather interesting as infiltration by inflammatory cells was not seen in animals treated with D, volubdlis, yet they showed hepatic degeneration. Is it that the animals treated with N. saliva showed an in~ammatory response which prevented toxic material from causing degenerative changes in the hepatocytes? However, it is difficult to explain a possible defensive role of these inflammatory cells as these were also seen in all of the control animals. Kupffer cells laden with bile pigments were seen in 3 out of 8 animals treated with N sativa, in 1 out of 8 animals treated with D voZub~Z~ and in 4 out of 10 control animals. Arseculeratne et al. (19811 have previously reported hepatic lesions such as centrolobular sinusoidal congestion, segmental or total disruption of the central vein, hepatocellular necrosis, degeneration of hepatocytes, degeneration and oedema of the subendothelial connective tissue and periportal degeneration of hepatocytes with mononuclear infiltration in the portal tracts when they studied the toxicity of 3 plants which were positive for pyrrolizidine alkaloids. A later study by the same authors (Arseculeratne et al., 19851 demonstrated similar hepatic changes following administration of 5 other plants which were negative for pyrrolizidine alkaloids. These authors, therefore, suggested the occurrence of hepatotoxicity due to as yet unidentified hepatotoxins. Arseculeratne et al. (1981, 19851 have not studied the possible hepatotoxicity of N sat&a or D. votubilis. Further, their studies on toxicity were limited to histopathological examination and no attempt was made to measure key hepatic enzyme concentrations. This approach is inadequate as some plant toxins may damage the hepatic cells at the molecular level without causing overt histopathological changes. In our studies, a significant elevation of certain key hepatic enzymes and varying degrees of histopathological changes in the liver were documented following oral administration of boiling water extracts of seeds of A? sativa and D. volubilis leaves. An increase in gamma-glutamyl transferase, alkaline phosphatase and alanine aminotransferase concentrations in the presence of hepatocyte degeneration were observed following oral administration of ft. voZ~b~l~s extract and this suggests that these enzymes may have been released following damage to the hepatoeytes. An increase in gamma-gluta-
289
my1 transferase and alanine aminotransferase concentrations in the absence of hepatocyte degeneration were observed following oral administration of iV. sativu extract and this suggests that these enzymes may have been released due to hepato~ellular damage caused at the molecular level. However, an enzyme-inducing effect of N. sat& extract cannot be ruled out. In view of these preliminary findings, the repeated use of these medicinal plants as therapeutic agents should not be encouraged.
This work was supported by National Resources, Energy and Science Authority of Sri Lanka, Research Grant No: RG/88/M/3 and International Foundation for Science, Stockholm, Research Grant No: F/1296-1. We are extremely grateful to Prof. B.A. Abewickrama, Emeritus Professor of Botany, University of Colombo for authentication of all plant material. We thank Kamal Perera and D.C. Ranatunga for animal care. References Arseculeratna, S.N., GunatiIaka, A.A.L. and Panabokke, R.G. (1981) Studies on medicinal plants of Sri Lanka: Occurrence of pyrroli~dine alkaloids and hepatotoxie properties in some traditional medicinal herbs. JOIUX.U~of ~thnopha~ucoZog~ 4, 159- 177. Arseculeratna, S.N., Gunatilaka, A.A.L. and Panahokke, R.G. (1985) Studies on medicinal plants of Sri Lanka. Part 14: Toxicity of some traditional medicinal herbs. Journal of Ethnopharmacology 13, 323-335. Jayaweera, D.M.A. (1982) Medic&& Plants Used iu Ceylon. National Science Council of Sri Lanka, Colomho, Part 1, pp. 159 and Part 6 pp. 131. King, E.J., Harkwood, G.A.D., Derty, G.E. and Beally, D. (1942) Mi~obiochemical methods of blood analysis. Lancet 1, 208-210. Reitman, S. and Frankel, S. (1957) A calorimetric method for the determination of serum glutamate oxaloacetic acid and pyruvic acid transaminases. American Journal of Clinical Pathology 28, 56- 63. Rosalki, S.B. and Rau, D. (1972) Serum gamma glutamyl transpeptidase activity in alcoholism. Cbi?aicaChimica Acta 39, 41- 47. Szczeklik, E., Orlowski, M. and Szewesuk, A. (1961) Serum gamma glutamyl peptidase activity in liver disease. Gas troenterology 41,353 - 359. Wilkinson, J.H. (1976) The Principles and Practice of Diagnostic Enzymology. Edward Arnold (Publishers) Ltd., London, pp. 305-348.
of Ethnopharmacology, Scientific
Publishers
283
31 (1991) 283-289 Ireland Ltd.
POSSIBLE HEPATOTOXICITY DREGEA VOLUBILIS LEAVES
OF NIGELLA
KAMANI H. TENNEKOONa, S. JEEVATHAYAPARAN4 ARUNDATHIE P. KURUKULASOORIYA’ and ERIC
SATIVA
SEEDS
AND
H. KARUNANAYAKEb
“Department of Physiology, bDepartment of Biochemistry and “Department of Pathology, Faculty of Medicine, University of Colombo, PO Box 271, Colombo 8 (Sri Lanka) (Accepted
September
11, 1990)
Summary
Aqueous extracts of the seeds of Nigellu sativa and mature leaves of Dregea volubilis were administered orally under light ether anaesthesia to male Sprague-Dawley rats for 14 days. Key hepatic enzyme concentrations and histopathological changes in the liver in both treatment groups at the end of 14 days were compared with a control group which received distilled water under identical conditions for 30 days and with a group of normal animals. Serum gamma-glutamyl transferase concentrations were significantly increased in both extract groups while serum alkaline phosphatase concentrations were significantly increased following administration of only D. volubilis when compared with either the control or the normal group. Serum alanine aminotransferase concentrations were significantly increased in both extract groups when compared with the normal group but not with the control group. Degenerative changes in hepatocytes were seen following administration of D. volubilis while consistent significant histopathological changes were not evident following administration of N. satiwa.
Introduction
During investigations to evaluate the purported galactogogue activity of Nigella sativa L. (family, Umbelliferae; Sinhala: ‘Kaluduru’) and Dregea voluCorrespondence to: Kamani H. Tennekoon, Department versity of Colomho, PO Box 271, Colombo 8, Sri Lanka.
0378s8741lt03.50 Published
0 1991 Elsevier and Printed in Ireland
Scientific
Publishers
of Physiology,
Ireland
Ltd.
Faculty
of Medicine,
Uni-
284
bilis L. (family, Asclepiadaceae;
Sinhala: ‘Anguna’l a high incidence of mortality was observed among the experimental animals, l-2 weeks after completion of the experiment. In addition to its possible galactogogue properties, N. sativa is considered to have emetic and expectorant potential while D. volubilis is thought to have diuretic and anthelmintic properties (Jayaweera, 19821. Considering the extensive usage of these plants in the Ayurveda system of Medicine and our own observation of high mortality among experimental animals during our studies, it was thought necessary to investigate the possible hepatotoxicity of these two plants. Materials and Methods Preparation of plant extracts N. sativa seeds were bought
from an Ayurvedic drug store. These were washed to remove sand and other debris and air-dried. The seeds (250 gl were boiled in distilled water (1000 ml) for 90 min and filtered through muslin. The final volume of the filtrate obtained was 320 ml. The mature leaves (258 gl of D. volubilis were shredded and boiled in distilled water (1500 ml) for 90 min. The extract was squeezed and filtered through muslin. The final volume of the filtrate obtained was 690 ml.
Animal
experiments
Five-month-old healthy male Sprague - Dawley rats housed under standard conditions and liberally fed with rat pellets (Moosaji’s Ltd., No. 34, W.A.D. Ramanayake Mawatha, Colombo 21 and water were used for this study. Thirty-two rats were selected for the study and were randomly divided into 2 groups of 8 each and a group of 10. Aqueous extracts of N. sativa and D. volubilis were administered orally to rats in Group 1 Vv = 8) and Group 2 Vv = 81, respectively, at a daily dose of 10 ml/kg for 14 consecutive days under light ether anaesthesia. Group 3 (control group, N = 101 received an equal amount of distilled water for 30 days under identical conditions and served as a common control group for the two experiments. At the end of the experimental period (i.e. 14 days for both treatment groups and 30 days for the control group), animals were sacrificed and 5 ml of blood was collected by cardiac puncture for estimation of key hepatic enzymes. Serum was separated and stored at - 20°C till gammaalkaline phosphatase, alanine aminotransferase and glutamyl transferase, aspartate aminotransferase concentrations were measured using methods described by King et al. (19421, Reitman and Frankel (19571 and Rosalki and Rau (1972). Livers were harvested for histopathological examinations. Gamma-glutamyl transferase, alkaline phosphatase, alanine aminotransferase and aspartate aminotransferase concentrations in the two treatment groups and in the control group were compared using the Student t-test.
285
Further, these enzyme concentrations in the treatment groups and in the control group were also compared with the enzyme concentrations that have been established in our laboratory for 25 normal healthy Sprague-Dawley rats of the same age, again using the Student-t-test. Results and Discussion Serum gamma-glutamyl transferase and alkaline phosphatase concentrations (mean f S.E.M.) in the two treatment groups, control group and in normal healthy rats are shown in Fig. 1. Serum alanine aminotransferase and aspartate aminotransferase concentrations (mean + S.E.M.) for the same groups are shown in Fig. 2. Serum gamma glutamyl transferase concentrations (mean & S.E.M.) were 3.29 r 0.49, 3.99 f 0.54, 1.36 f 0.13 and 1.64 + 0.59 U/l in groups treated with N. sativa, D. volubilis and in the control and normal groups, respectively. These serum concentrations were significantly P < 0.001) increased following administration of N. sativa and D. volubilis relative to both the control group and the normal group. Serum alkaline phosphatase concentrations (mean f S.E.M.) were 43.70 + 3.88, 59.48 f 5.66, 45.7 + 0.85 and 42.24 f 1.66 K-A units1100 ml in groups treated with N. sativa. D. volubilis
Fig.
q NIGELLA SATIVA
H
CONTROL GROUP
q DREGEA VOLUBILIS
Cl
NORMAL GROUP
1. Serum
gamma-glutamyl
transferase
and alkaline
phosphatase
concentrations
(mean
+
S.E.M.) in male Sprague-Dawley rats following 10 ml/kg oral administration of boiling water extracts of the seeds of Nigello sativa W = 8) and mature leaves of Dregea volubilis W = 8) for 14 days and distilled water for 30 days (control group, N = 10) under light ether anaesthesia and in a group of normal animals W = 25).
286
q NIGELLA SATIVA
ggl
CONTROL GROUP
•3 DREGEA VOLUBILIS
Cl
NORMAL GROUP
Fig. 2. Serum aspartate aminotransferase and alanine aminotransferase concentrations (mean + S.E.M.) in male Sprague-Dawley rats following 10 ml/kg oral administration of boiiing water extracts of the seeds of Nigello sativa W = 8) and mature leaves of Dregea vo~ub~l~ Uv = 81 for 14 days and distilled water for 30 days (control group, N = 10) under light ether anaesthesia and in a group of normal animals W = 25).
and in the control and normal groups, respectively. Thus, the serum alkaline phosphatase concentration following administration of D. volubilis was significantly higher P < 0.0011 when compared with the control and normal groups while N. sativa administration did not have a significant effect on this enzyme. Further, there were no significant differences in regard to and alkaline phosphatase concentrations gamma glutamyl transferase between the control and the normal groups indicating that the transient but chronic ether administration experienced by the control group did not affect these enzymes. Serum gamma-glutamyl transferase and alkaline phosphatase are membrane-bound enzymes and are released unequally depending on the pathologiElevation cal phenomenon. of serum gamma-glutamyl transferase concentration is generally regarded as one of the most sensitive indices of hepatic damage (Szczeklik et al, 19611. In the present study, serum gammaglutamyl transferase concentrations were significantly increased following oral administration of both N. sativa and D. volubilis. Serum alkaline phosphatase concentrations are known to be markedly elevated in cholestasis and to be minimally increased in chronic hepatocellular disease (Wilkinson, 19761. The significantly increased serum alkaline phosphatase concentrations seen
287
following administration of D. volubilis indicate a difference in toxicity between this plant and N. sativa. Serum alanine aminotransferase concentrations (mean f S.E.M.1 were 18.04 + 2.40, 19.85 + 1.93, 15.16 + 1.45 and 12.7 rt 0.67 U/100 ml in the groups treated with N sativa, D. volubilis and in the control and normal groups, respectively, Serum alanine aminotransferase concentrations were significantly higher following administration of N. sativa P < 0.01) and D. voEubiZis P < 0.001) when compared with the normal group but not when compared with the control group. Alanine aminotransferase concentrations were higher in the control group when compared with the normal group but this difference was not statistically significant. Serum aspartate aminotransferase concentrations (mean -+ S.E.M.) were 13.13 1: 0.96, 13.13 & 1.77, 12.18 + 1.22 and 11.7 rt: 0.69 U/100 ml in the groups treated with N. sativa, D. volubilis and in the control and normal groups, respectively and hence were not significantly different from one another. Alanine aminotransferase is a cytoplasmic enzyme found in very high concentrations in the liver. Aspartate aminotransferase is present in the cytoplasm as well as in the mitochondria and is less specific than alanine aminotransferase as an indicator of hepatic damage since it is rapidly inactivated (Wilkinson, 1976). The oral administration of neither of the plant extracts used in this study nor chronic ether anaesthesia by itself appears to have a significant effect on the release of alanine aminotransferase. However, the significant increase in alanine aminotransferase concentrations seen following oral administration of N. sativa and D. volubilis when compared with normal animals may indicate that administration of these plant extracts when superimposed on chronic ether anaesthesia may cause an increased release of this enzyme. However, neither oral administration of the two plant extracts nor chronic ether anaesthesia appears to have a significant effect on aspartate aminotransferase concentrations. Histopathological examination of livers showed degenerative changes in hepatocytes in 6 out of 8 animals treated with D. volubilis. There was no evidence of degenerative changes in hepatocytes in animals treated with N sutivu, in control animals or in normal animals. Thus the degenerative changes in the hepatocytes in animals treated with D. volubilis appears to have been caused specifically by D. volubilis. The significant elevation of gamma-glutamyl transferase and alkaline phosphatase concentrations seen following oral administration of D. volub~l~ when compared with either the control or the normal group and the significant elevation of alanine aminotransferase concentrations when compared with the normal group may reflect the hepatic damage thus caused. Central veins and the sinusoids around them were dilated in 7 out of 8 animals treated either with N. sativa or D. volubilis and in 3 out of 10 control animals while the portal veins and the sinusoids around them were dilated in 5 out of 8 animals treated either with N. sativa or D. vo~u~~l~sand in 5 out of 8 control animals. Dilatation of central veins, portal veins or sinu-
288
soids was not seen in any of the normal animals. It may be that chronic ether anaesthesia led to dilatation of the central veins and surrounding sinusoids while oral administration of iV. sat&a and I). voZ~b~Z~ enhanced this effect. The prevalence of dilatation of portal veins and surrounding sinusoids was comparable in the two treatment groups and in the control group indicating that this may have been caused by chronic ether anaesthesia and that the oral administration of the plant extracts did not have a superimposing effect. Infiltration by mononuclear chronic in~ammatory cells was observed in the vicinity of the portal tracts in 6 out of 8 animals treated with N sat&a, in all control animals, in 5 out of 25 normal animals and in none of the animals treated with D. volubilis. This finding is rather interesting as infiltration by inflammatory cells was not seen in animals treated with D, volubdlis, yet they showed hepatic degeneration. Is it that the animals treated with N. saliva showed an in~ammatory response which prevented toxic material from causing degenerative changes in the hepatocytes? However, it is difficult to explain a possible defensive role of these inflammatory cells as these were also seen in all of the control animals. Kupffer cells laden with bile pigments were seen in 3 out of 8 animals treated with N sativa, in 1 out of 8 animals treated with D voZub~Z~ and in 4 out of 10 control animals. Arseculeratne et al. (19811 have previously reported hepatic lesions such as centrolobular sinusoidal congestion, segmental or total disruption of the central vein, hepatocellular necrosis, degeneration of hepatocytes, degeneration and oedema of the subendothelial connective tissue and periportal degeneration of hepatocytes with mononuclear infiltration in the portal tracts when they studied the toxicity of 3 plants which were positive for pyrrolizidine alkaloids. A later study by the same authors (Arseculeratne et al., 19851 demonstrated similar hepatic changes following administration of 5 other plants which were negative for pyrrolizidine alkaloids. These authors, therefore, suggested the occurrence of hepatotoxicity due to as yet unidentified hepatotoxins. Arseculeratne et al. (1981, 19851 have not studied the possible hepatotoxicity of N sat&a or D. votubilis. Further, their studies on toxicity were limited to histopathological examination and no attempt was made to measure key hepatic enzyme concentrations. This approach is inadequate as some plant toxins may damage the hepatic cells at the molecular level without causing overt histopathological changes. In our studies, a significant elevation of certain key hepatic enzymes and varying degrees of histopathological changes in the liver were documented following oral administration of boiling water extracts of seeds of A? sativa and D. volubilis leaves. An increase in gamma-glutamyl transferase, alkaline phosphatase and alanine aminotransferase concentrations in the presence of hepatocyte degeneration were observed following oral administration of ft. voZ~b~l~s extract and this suggests that these enzymes may have been released following damage to the hepatoeytes. An increase in gamma-gluta-
289
my1 transferase and alanine aminotransferase concentrations in the absence of hepatocyte degeneration were observed following oral administration of iV. sativu extract and this suggests that these enzymes may have been released due to hepato~ellular damage caused at the molecular level. However, an enzyme-inducing effect of N. sat& extract cannot be ruled out. In view of these preliminary findings, the repeated use of these medicinal plants as therapeutic agents should not be encouraged.
This work was supported by National Resources, Energy and Science Authority of Sri Lanka, Research Grant No: RG/88/M/3 and International Foundation for Science, Stockholm, Research Grant No: F/1296-1. We are extremely grateful to Prof. B.A. Abewickrama, Emeritus Professor of Botany, University of Colombo for authentication of all plant material. We thank Kamal Perera and D.C. Ranatunga for animal care. References Arseculeratna, S.N., GunatiIaka, A.A.L. and Panabokke, R.G. (1981) Studies on medicinal plants of Sri Lanka: Occurrence of pyrroli~dine alkaloids and hepatotoxie properties in some traditional medicinal herbs. JOIUX.U~of ~thnopha~ucoZog~ 4, 159- 177. Arseculeratna, S.N., Gunatilaka, A.A.L. and Panahokke, R.G. (1985) Studies on medicinal plants of Sri Lanka. Part 14: Toxicity of some traditional medicinal herbs. Journal of Ethnopharmacology 13, 323-335. Jayaweera, D.M.A. (1982) Medic&& Plants Used iu Ceylon. National Science Council of Sri Lanka, Colomho, Part 1, pp. 159 and Part 6 pp. 131. King, E.J., Harkwood, G.A.D., Derty, G.E. and Beally, D. (1942) Mi~obiochemical methods of blood analysis. Lancet 1, 208-210. Reitman, S. and Frankel, S. (1957) A calorimetric method for the determination of serum glutamate oxaloacetic acid and pyruvic acid transaminases. American Journal of Clinical Pathology 28, 56- 63. Rosalki, S.B. and Rau, D. (1972) Serum gamma glutamyl transpeptidase activity in alcoholism. Cbi?aicaChimica Acta 39, 41- 47. Szczeklik, E., Orlowski, M. and Szewesuk, A. (1961) Serum gamma glutamyl peptidase activity in liver disease. Gas troenterology 41,353 - 359. Wilkinson, J.H. (1976) The Principles and Practice of Diagnostic Enzymology. Edward Arnold (Publishers) Ltd., London, pp. 305-348.
