Food and Chemical Toxicology 45 (2007) 14871495 www.elsevier.

com/locate/foodchemtox

Eect of the extract of the tamarind (Tamarindus indica) fruit on the complement system: Studies in vitro and in hamsters submitted to a cholesterol-enriched diet
s Nader Chryso stomo a, Ana Elisa C.S. Azzolini a, Ana Paula Landi Librandi a, Ta a rgio Akira Uyemura b, Ana Isabel de Assis-Pandochi Carem Gledes Vargas Recchia , Se
a

a,*

ncias Farmace uticas de Ribeira sica e Qu mica, Faculdade de Cie , s/n, 14040-903, Departamento de F o Preto, Universidade de Sa o Paulo, Av. do Cafe Ribeira o Preto, SP, Brazil b ncias Farmace uticas de Ribeira lises Cl nicas, Toxicolo gicas e Bromatolo gicas, Faculdade de Cie Departamento de Ana o Preto, , s/n, 14040-903, Ribeira Universidade de Sa o Paulo, Av. do Cafe o Preto, SP, Brazil Received 2 August 2006; accepted 11 February 2007

Abstract This work evaluated a crude hydroalcoholic extract (ExT) from the pulp of the tamarind (Tamarindus indica) fruit as a source of compounds active on the complement system (CS) in vitro. ExT, previously characterized by other authors, had time and concentration dependent eects on the lytic activity of the CS. The activity of 0.8 mg/mL of the extract on the classical/lectin pathways (CP/LP) increased after 30 min of pre-incubation, while that of the alternative pathway (AP) decreased after 15 min at 1 mg/mL. Since the CS is a mediator of inammation, studies were also made in vivo, taking advantage of a model of hypercholesterolemia in hamsters to investigate the role of CS in the phase preceding the inammatory process of atherosclerosis. Hamsters submitted to a diet rich in cholesterol showed increased lytic activity of the CP/LP and AP after 45 days. The activity levels of C2 and factor B components on respectively, classical/lectin and alternative pathways of the CS also increased. Early events cooperating to trigger the process of atherosclerotic lesions are not completely understood, and these alterations of complement may participate in these events. When treatment with a diet rich in cholesterol was associated to the furnishing of ExT, evaluation of complement components and complement lytic activity showed values similar to those of the controls, showing that treatment with ExT blocked the increase of complement activity caused by the cholesterol-rich diet. By itself, ExT had no eect on the complement system in vivo. ExT activity on the CS may be of interest for therapy and research purposes. 2007 Elsevier Ltd. All rights reserved.
Keywords: Tamarind fruit extract; Complement system; Hyperlipidemia; Hamsters

1. Introduction In recent years there occurred a renewed interest in plants for the treatment of diseases (Dubey et al., 1994; Prince et al., 1998; Ladeji et al., 2003). Epidemiologic studies have shown that the consumption of diets rich in plantderived foods high in phenolic compounds, is associated with a decreased incidence of cardiovascular mortality
*

Corresponding author. Tel.: +55 16 36024179; fax: +55 16 36332960. E-mail address: anisa@usp.br (A.I. de Assis-Pandochi).

(Hertog et al., 1993, 1995; Keli et al., 1996; Geleijnse et al., 1999). Grape polyphenols decrease plasma triacylglycerols and of cholesterol accumulated in the aorta of ovariectomized guinea pigs (Zern et al., 2003). Components of Tamarindus indica, a tree indigenous to India and South East Asia, have been used as spices, food components and in snacks. In Thai traditional medicine, the fruit of T. indica is regarded as a digestive, carminative, laxative, expectorant and blood tonic (Komutarin et al., 2004). Recently, Pumthong (1999) demonstrated the antioxidant activity of the T. indica seeds coat extract. This extract is

0278-6915/$ - see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.fct.2007.02.008

1488

A.P. Landi Librandi et al. / Food and Chemical Toxicology 45 (2007) 14871495 0.0005 M of azide, 0.08 M of ethyleneglycol-bis-(b-aminoethyl ether) N,N,N 0 ,N 0 ,-tetraacetic acid and 0.002 M magnesium (TEA/EGTA/Mg2+), was used for the evaluation of AP activity. Phosphate buered saline (PBS) pH 7.4, maintained on ice, was used to stop hemolytic reactions.

composed of avonoids including tannins, polyphenols, anthocyanidins, and oligomeric proanthocyanidins. A polysaccharide isolated and puried from T. indica seeds showed immunomodulatory activities, like the enhancement of phagocytosis and inhibition of leucocyte migration and cell proliferation, suggesting possibly interesting biological applications of this plant (Sreelekha et al., 1993). A recent study has shown the benecial eects of the consumption of T. indica fruit extract in an experimental model of atherosclerosis in hamsters, including decreased levels of serum cholesterol and triglycerides (Martinello et al., 2006). Cardiovascular diseases (CVD) are the leading cause of morbidity and mortality in developed countries (Libby, 2002). Classic major risk factors for CVD include hyperlipidemia, elevated levels of LDL cholesterol, decreased HDL cholesterol and smoking. Atherosclerosis is an important pathologic manifestation underlying CVD, and has been suggested to be an inammatory disease (Ross, 1999). Data from the literature indicate it to be a consequence of a chronic inammatory process induced by the activation of macrophages, the complement system (CS) and T-lymphocytes. Several studies have implicated elements of the humoral-mediated immune response in atherogenesis (Geertinger and Sorensen, 1973; Hollander et al., 1979; Hansson et al., 1984; Vlaicu et al., 1985a,b). Among them, activation of CS has been associated with the pre-lesional stages, as well as with the progress of atherosclerosis (Torzewski et al., 1997). The CS is a complex cascade of enzymes and regulatory proteins that normally participate in host defenses against microorganisms through opsonization, chemoattraction of leucocytes, cell lysis and cell activation (Wallport, 2001). Several studies showed activated complement components in the atherosclerotic plaque, as well as membrane attack, the complex (MAC, C5b-9) that promotes cellular activation, upregulates adhesion molecules, stimulates chemokine secretion and can cause cell lysis (Yasojima et al., 2001). The potential activity of tamarind extracts on the CS had not been previously investigated; it may be of interest for application in research and therapy. The aim of the present study was to evaluate the eect of ExT on the activity of the CS in vitro, and on complement activity in an experimental model of hyperlipidemia in hamsters submitted to a cholesterol-enriched diet and treated or not with ExT. Components of the classical/lectin (CP/LP), or alternative pathways (AP) of the CS (Wallport, 2001) and complement lytic activity were investigated.
2. Materials and methods 2.1. Buers
Complement xation diluent (CFD) containing 0.1% gelatin, was used for hemolytic assays of CP/LP, according to Harrison and Lachamnn (1986). Triethanolamine (TEA) 0.02 M buer, pH 7.2, containing

2.2. Plant material

Ripe fruits of Tamarindus indica were collected in the region of Ribeira o Preto, SP, Brazil and peeled to obtain the pulp employed for the extract preparation.

2.3. Preparation of hydroethanolic extract of the pulp of Tamarindus indica (ExT)

Approximately 100 g of the fruit pulp were placed in a conical ask and soaked at room temperature for 5 days, in 400 mL of 70% ethanol in water. The resulting extract was ltered through a sieve and rotoevaporated twice, until complete alcohol evaporation had occurred. The composition of this ExT, previously determined by Martinello et al. (2006), is shown in Table 1. Total sugars were determined by the phenolsulfuric acid method according to Dubois et al. (1956); the reaction mixture contained, respectively, 500 lL of the sample (a solution at 0.1 mg/mL of crude ExT (w/v)), the blank or a glucose standard (Vetec mica Fina LTDa, Rio de Janeiro, Brazil) (20100 lg/mL solution), Qu respectively, 500 lL of an aqueous phenol 5% (w/v) solution and 2.5 mL of fuming sulfuric acid. After mixing and chilling, absorbance was measured at 490 nm in a Beckman DU-70 spectrophotometer to determine hexose content. Total soluble phenolic derivatives were determined according to Folin and Ciocalteau (1927), using a reaction medium containing 50 lL of the sample (a solution of 20 mg/mL of crude ExT (w/v)), the blank or a gallic acid standard (Sigma) (1001000 lM), respectively, 50 lL of 7% (v/v) acetic acid, 50 lL of the FolinCiocalteaus reagent, 50 lL of 35% (w/v) sodium carbonate and 800 lL of distilled water. After mixing, the reactants were incubated for 90 min at room temperature in the dark. Light absorbance was measured at 725 nm in a Beckman DU-70 spectrophotometer, and the total phenolic content expressed as gallic acid equivalents per mg of extract. Protein concentration was determined by Bradfords (1976), calorimetric method, using a commercially available kit (BIO-RAD Protein Assay) and a bovine serum albumin (BIO-RAD standard II). Fifty microliter of sample (undiluted ExT) blank or albumin standard (2.5 25 lg/mL), respectively, were added of 40 lL of BIO-RAD reagent. After mixing, the reaction mixture was incubated for 5 min, at room temperature. Its absorbance was measured at 600 nm in a Beckman DU-70 spectrophotometer. SomogyiNelsons method was employed to evaluate the concentration of reducing sugar (Kidby and Davidson, 1973). Briey, 1 mL of the sample (0.1 mg/mL of crude ExT (w/v)), of the blank or of the galactose standard (Sigma) (20100 lg/mL) were added to 1 mL of Somogyis reactant and incubated for 10 min at 100 C. After cooling, 1 mL of Nelsons reagent was added and the mixture made up to 10 mL with distilled water. Absorbance was measured at 540 nm in a Beckman DU-70 spectrophotometer. For the uronic acid determination (Blumenkratz and Assboe-Hansen, 1973), 200 lL of sample solution of 0.1 mg/mL of crude ExT (w/v), of the blank or the glucuronic acid standard (Sigma) (2.5100 lg/mL) were maintained in ice and then 1.2 mL of 0.0125 M sodium tetraborate were added. After mixing and incubation for 5 min incubation at 100 C, 20 lL

Table 1 Components of the ExT (%)a Total sugars 61.5

a

Reducing sugars 27.6

Phenolic compounds 0.25

Uronic acid 3.70

Protein 0.57

Determined as referred to in Section 2.

A.P. Landi Librandi et al. / Food and Chemical Toxicology 45 (2007) 14871495 of 0.15% of m-hydroxybiphenyl in 0.5% NaOH were added. After 5 min, absorbance was measured at 520 nm in a Beckman DU-70 spectrophotometer. In accordance with the results obtained by Martinello et al. (2006), the ExT prepared as described above, was diluted to 5% in water prior to use; this concentration is usually present in tamarind juice drinks consumed by the human population.

1489

2.8. Evaluation of factor B

Factor B (R(B))-depleted NHaS was prepared as described by Hamuro et al. (1978) and Joisel et al. (1983). After 3.3 min of incubation at 56 C, the serum showed no AP lytic activity. Restoration of this activity was obtained when puried human factor B was added at 30 lg/mL (not shown). This R(B) was used to evaluated the factor B of treated animals, using a modied Mayers method (1971). R(B) (45 lL) were added to 15 and 20 lL of hamster serum to a nal volume of 180 lL, in TEAEGTA Mg+2 and incubated with 120 lL of rabbit E suspension for 30 min at 37 C. After incubation, 300 lL of PBS were added and the samples were centrifuged at 2000 rpm for 10 min. The absorbance of supernatants was measured at 412 nm, and that of cells lysed in water (100% lysis) was employed for comparisons.

2.4. Preparation of a cholesterol-rich diet

The chow (Guabi or Nuvlab), furnished by the campus animal house, was ground and 1% cholesterol was added to obtain the lipidenriched diet. This chow or a control not containing this lipid was pressed, dried at 37 C and stored in a dry environment.

2.5. Selection of animals and experimental design

Male Gold Syrian hamsters (90100 g) were housed in colony cages (ve animals per cage) in a room at controlled temperature (25 2 C) and a 12 h light/12 h dark cycle. Animals were divided in four groups treated respectively as follows: (1) water and standard chow (control group, C); (2) a 5% ExT suspension in water and standard chow (extract group, ExT); (3) water and standard chow plus 1% cholesterol (cholesterol group, Col) and (4) a 5% ExT suspension in water and standard chow plus 1% cholesterol (cholesterol plus extract group, CE). Food and water or ExT were furnished ad libitum. After treatment during 1560 days, animals were sacriced by decapitation. Blood samples were collected, and serum separated and stored at 70 C until analysis. The guide-lines of our institutional ethics committee were strictly followed.

2.9. Analysis of biochemical parameters in serum

Biochemical parameters were investigated employing commercial kits (Labtest Diagnostica, Montes Claros, MG, Brazil). Hyperlipidemia was evaluated by determination of total cholesterol and triglycerides in serum of the experimental animals. Hepatic function was investigated by evaluation of albumin levels (Peters et al., 1982), total protein (Kingsley, 1942), aspartate (AST) and alanine (ALT) transaminases (Rej and Horder, 1983; Horder and Rej, 1983) in the sera of the animals following treatment.

2.10. Evaluation of in vitro ExT eect on hamster serum complement levels

NHaS was pre-incubated at 37 C with dierent concentrations of ExT for 15, 30, 45 and 60 min, respectively. The ExT concentrations tested were obtained by serial dilutions of a crude 2 mg/ml (w/v) ExT stock suspension in CFD (for CP/LP determination) or TEAEGTAMg2+ (for AP determination), plus the addition of diluted NHaS (1:140 for CP/LP and 1:10 for AP), to a nal volume of 180 lL. After 15 min (for CP/LP) and 60 min (for AP) of pre-incubation at 37 C, respectively, 120 lL of EA or E were added, followed after 30 min by 300 lL of cold PBS buer added to stop the reaction. The incubation tubes were centrifuged for 10 min at 2000 rpm, and absorbance measured at 412 nm; and the percentage of haemolysis was calculated, in terms of the absorbance of cells lysed in water (100% lysis controls). As an additional control, the eect of ExT alone (without hamster serum) on the lysis of EA or E, was investigated under the same conditions.

2.6. Evaluation of complement activity

Haemolytic assay measuring the kinetic of lysis (kinetic assay), according to Ewald et al. (1961) and Ferriani et al. (1990), was used to evaluated the lytic activity of CP/LP and of the AP of serum complement. This method is based on the determination of the time required to lyse 50% (t1/2), of an erythrocyte suspension under standard conditions. For the CP/LP assays, sheep erythrocytes were washed twice with CFD and sensitized with rabbit anti-sheep erythrocyte antibodies, produced in our laboratory. The cell suspension (EA) was standardized at 700 nm to show an absorbance of 0.70.8 (1.7 108 cells/mL). For the AP assays, rabbit erythrocytes were washed twice with TEA EGTAMg2+ and the cell suspension (E) was standardized as above described. EA or E (100 lL) were added to 200 lL of hamster serum diluted at 1:20 or 1:40 (CP/LP), and 1:2 or 1:2.5 (AP) in adequate buer, and assayed as described above.

2.11. Statistical analysis

Experimental data were analyzed using GraphPad Prism V.4 (GrapPad Software, Inc., San Diego, CA). Results are expressed as means SD. The statistical signicance of the dierences between means was determined by the KruskallWallis non-parametric test, followed when necessary, by post hoc testing by Dunns test. Dierences were considered signicant when p < 0.05.

2.7. Evaluation of C2
C2 (R(2))-depleted serum was prepared by heating pooled normal human serum (NHS) to 56 C, for 2.5 min, and aliquots were assayed after dierent times of incubation to investigate residual activity of CP/LP according to Mayer (1971). After 2.5 min of heating, CP/LP lytic activity was completely destroyed Selective inactivation of C2 was investigated by the addition of dierent concentrations of puried human C2 to the R(2) to restore CP/LP lytic activity (results not shown). This serum preparation was employed as R(2) reagent to evaluate C2 in the serum of animals. R(2) (15 lL) was added to 30 or 60 lL of hamster serum, taken to a nal volume of 200 lL with CFD, and incubated with 100 lL of EA suspension at 37 C to measure the kinetics of lysis. Normal hamster serum (NHaS) without R(2) was used as a control. Results were expressed as % of increase in lytic activity promoted by the hamster serum plus R(2) compared to that promoted by hamster serum without human R(2) (100%).

3. Results Fig. 1 shows that the ExT exhibited opposite in vitro pre-incubation time and ExT concentration-dependent eects on respectively, CP/LP (Fig. 1a) and AP lytic activity (Fig. 1b) in hamster serum. At 0.8 mg/mL it led to a 73% increase of the CP/LP lytic activity after 15 min of pre-incubation. In contrast, an anticomplementary action of 1 mg/mL of the extract, led to loss of lytic AP activity: causing after 60 min of pre-incubation, approximately 40% of inhibition of this activity.

1490

A.P. Landi Librandi et al. / Food and Chemical Toxicology 45 (2007) 14871495

a
Percentage of lysis

200 175 150 125 100 75 50 25 0 0 250 500 750 1000

Extract concentration (g/mL)

b
Percentage of inhibition of lysis

50 40 30 20 10 0 0.0

0.3

0.6

0.9

1.2

Extract concentration (mg/mL)

Fig. 1. Eect on the lytic activity of respectively CP/LP (a) and AP (b), of pre-incubation of the ExT with normal hamster serum for respectively, 15 and 60 min.

The administration of ExT to hamsters had no eect on CP/LP or AP lytic activities after 1560 days of treatment (results not shown). Results obtained following 30 and 45 days of treatments respectively, are from samples assayed

on the same day, and are presented on Figs. 2a and 3a for convenience. After 45 days of consumption of a hyperlipidemic diet (Col), increased CP/LP (Fig. 2a) and AP (Fig. 3a) activities, expressed by decreased values of t1/2 compared to controls, were observed. Similar results were obtained in other groups treated following the same protocol, with some variation in the periods of treatment between assays. Considering these results, we evaluated the activity of C2 and Factor B, components specic for respectively CP/LP and AP, by a hemolytic assay using reagents decient in these components (prepared in our laboratory, see Section 2 for details). When assayed using R(2), CP/LP lytic activity (expressed as percentage relative to controls) (Fig. 2b), was signicantly increased in the hyperlipidemic hamsters after 45 days of treatment. Similar results were obtained when Factor B was evaluated using R(B) (Fig. 3b). Interestingly, when ExT was administrated to animals given a normal diet, but also when given together with a diet enriched with cholesterol (CE), this eect did not occur. These results suggest that the increases of CP/LP and AP lytic activity observed in these assays could be due to alterations of complement at the level of components C2 and Factor B as a consequence of the administration of a diet rich in cholesterol, but that treatment with ExT was able to block this eect. Hyperlipidemia was evaluated by determination of total cholesterol and triacylglycerols in serum (Fig. 4). Levels of total serum cholesterol were signicantly higher in animals receiving the high lipid diet after all periods of treatment compared to controls (only periods of 15 and 45 days are shown for convenience). Increase of serum triacylglycerols

4.0

b
30 days

30

3.5

3.0

Percentage of increase of CP/LP lytic activity

20

10

2.5

T1/2 (minutes)

2.0 C 6 ExT COL CE

0 C 40 ExT COL CE

*
30

20

10

45 days
2 C ExT COL CE 0 C ExT COL CE

Fig. 2. Eects on the restoration of lytic serum activity CP/LP (a) and R(2) (b), by hamster serum from animal groups: c (controls, 814)#, ExT (extracttreated, 710), Col (hypercholesteremic 1118) and CE (hypercolesterolemic + extract-treated 810) treated respectively, for 30 and 45 days. #Number of animals tested.

A.P. Landi Librandi et al. / Food and Chemical Toxicology 45 (2007) 14871495

1491

7.5

b
30 days

160

***
140

6.0

4.5

120

3.0

100

T1/2 (minutes)

1.5 C 4 ExT COL CE

Percentage of lysis

80 C 130 120 110 ExT COL CE

*
3

45 days 100
90

1 C ExT COL CE

80 C ExT COL CE

Fig. 3. Eects on the restoration of lytic activity of serum AP (a) and of R(2) (b), by serum from hamsters from the groups: c (910)#; ExT (8), Col (719) and CE (916) (b) treated respectively, for 30 and 45 days. *p < 0.05; **p < 0.01; ***p < 0.001, #number of animals tested.

TOTAL CHOLESTEROL
300

TRIGLYCERIDES
300

**
Serum levels of total cholesterol (mg/dL)
200

*
15 days

100

Serum levels of total cholesterol (mg/dL)

200

100

0
Ex T C O L E C C

0
Ex T C L C C E O

300

300

**
200

*
200

**

45 days
100

100

Ex

0
C Ex T L C O CE

Fig. 4. Serum levels of total cholesterol, and triacylglycerides, from hamsters in-groups C, ExT, Col and CE. (10)# for all groups. Animals were treated for respectively, 15 and 45 days; **p < 0.01; ***p < 0.001, #number of animals tested.

was signicant after 45 days of cholesterol treatment; however, a signicant reduction of this level was observed when the ExT was associated with the hyperlipidemic colesterolrich diet. ExT by itself did not show a signicant eect on these serum parameters. Results in Fig. 5 show values of total protein, albumin levels, AST and ALT activities in the serum of animals submitted to the dierent treatments employed during 15 or 45 days. Considering the absence of alterations of these parameters, neither the ExT, nor the diet enriched in cholesterol, nor the association of these two treatments aected liver function.

4. Discussion The activity of the ExT on serum CS was investigated in vitro using a pool of normal hamster serum. Besides furnishing a background for the in vivo studies, the in vitro assays were performed to investigate whether the ExT contained compounds able to interfere with the CS; if so, they could be considered as potential tools for the analysis of the mechanisms involved in the activation and regulation of the complement cascade. Important studies investigating eects of natural substances on the CS as for example, the classic well known cobra venom factor

1492

A.P. Landi Librandi et al. / Food and Chemical Toxicology 45 (2007) 14871495
AST
150 100 75
2

TOTAL PROTEINS
10.0 7.5
3

ALBUMIN

ALT

Serum levels of albumin (g/dL)

100 50

Serum levels of total proteins (g/dL)

5.0 2.5 0.0

CO L Ex T C E C

50 25

AST (U/L)

0
L Ex T

ALT (U/L)

C O

C E

0
C O L CE C

0
CO L CE CE C

10.0 7.5 5.0 2.5 0.0

Ex T L C E C C O

4 3

150

100 75

100
2 1 0
Ex C E O

50 50 25 0
C O L CE C

0
C C O L

Fig. 5. Serum levels of total protein, albumin, aspartate (AST) and alanine (ALT) transaminases in hamsters in the groups C, ExT, Col and CE. n = 10 for all groups. Animals were treated for 15 and 45 days, respectively.

(CFV) (Oberholzer et al., 1999), were decisive to understand the activation and regulation of the AP. Our in vitro results showed that the ExT aects serum complement leading to a concentration dependent increase of CP/LP lytic activity. Since the ExT by itself did not lyse of EA or E (results not shown), we can conclude that increase in lysis through the CP/LP in vitro, is due to an eect on complement. This eect was opposite to that on AP, where a decrease in lytic activity was observed. These opposite results could be a consequence of the complex composition of the ExT, with dierent substances acting on dierent points of the complement cascade causing diverse eect on complement activity. Future studies are required to understand this discrepancy; they justify additional eorts to isolate the substances responsible for these actions and to investigate the mechanisms involved. Assays in vivo in hamsters given a normal diet (controls) or a of cholesterol-enriched diet were performed employing an experimental model widely used in atherosclerosis studies (Moghadasian, 2002; Kahlon et al., 1992; Dorfman et al., 2003; Nistor et al., 1987), in which diet cholesterol content varied between 0.05% and 3% (w/w). This model was also chosen based on previous results obtained by our group (Martinello et al., 2006). Animals treated for 10 weeks were hypercholesterolemic at the end of treatment and a well-developed and mature atherosclerotic lesion was demonstrated. Our aim was to investigate the eventual involvement of the complement system in the mechanism that triggers the inammatory process of atherosclerosis, as well as the eect of the ExT on this condition. It was therefore necessary to nd a condition where developing increases of serum total cholesterol and triacylglycerols would occur preceding the inammatory process of atherosclerosis. Thus, we would be able to investigate if eventual complement alterations, possibly associated with alterations in serum lipid levels, would favor this triggering process and consequently the inammation that leads to the atherosclerosis lesion. In animals treated with 1% cholesterol for periods between 15 and 60 days, total serum

cholesterol was raised by all periods of treatment. These periods were chosen by us to investigate the role of the CS in the initial phase (pre-lesion) of the disease, in which the primary event is the initiation of an inammatory process triggered by not completely understood mechanisms (Ross, 1999). Serum from animals submitted to a standard diet plus ExT (ExT group), showed lytic activity similar to that of the controls, indicating that such treatment did not aect the complement system. Two dierent protocols were used to evaluate the CS in animals receiving a diet with high levels of cholesterol. In the rst group, hamsters received water plus cholesterol (Col group), and in the second, received ExT plus the sterol (CE group). In the rst group, CP/LP and AP lytic activities triggered by respectively, rabbit anti-sheep erythrocytes (EA) or rabbit E, (see Section 2), were signicantly increased compared to controls after 45 days of treatment. It is important to reinforce the fact that a considerable number of additional assays were performed using the 45-day and other periods of treatment. Increases of CP/LP and AP lytic activity due to the cholesterol-rich diet were regularly replicated. A total of 40 animals were evaluated. However, considerable variability occurred related to the period of treatment in which the eect was statistically signicant. We attributed these observations to normal biologic variability, and were a measure taken to avoid the sacrice of a higher number of animals. This condition of hyperlipidemia possibly associated to an increased potential (increased concentration?) of complement to generate fragments such as C3a/C3b/C5a or sublytic doses of the membrane attack complex (MAC), could have important biological consequences. Pasqui et al. (2000) showed a signicant increase of sC5b-9 (but not of CH50, C3 or C4) in hypercholesterolemic subjects compared to controls, indicating complement activation. The plasma sC5b-9 level was inversely and signicantly related to HDL-cholesterol. An increase of circulating immune complexes containing cholesterol was also demon-

A.P. Landi Librandi et al. / Food and Chemical Toxicology 45 (2007) 14871495

1493

strated, suggesting that complement activation in dyslipidemia may be induced by this mechanism. Another important factor is the decrease of HDL in subjects under a hypercholesterolemic diet. Since complement regulatory proteins such as CD59 are transported in serum by HDL (Vakeva et al., 1994), this decrease could in some way aect serum complement activity. The evidences available in the literature about complement activation in established atherosclerosis, and our data showing that when activated, complement action is high in hyperlipidemia, may be of interest and as shown here, the eect of ExT under these conditions justies additional studies. In the CE group, lytic activity showed values similar to those of controls. The mechanism through which diets containing high levels of cholesterol aect complement activity, and through which ExT prevents this eect will be subjected to investigation. Similar results were obtained in assays using treated animals serum samples to restore the lytic activity of serum decient of C2 [R(2)] and of factor B [R(B)]. When serum from the Col (hypercholesteremic only) group was used, lysis was higher than in controls, but serum from the CE group restored lytic activity to values similar to those of the controls. These data point to alterations of CP/LPs at the level of C2 and of AP at the level of factor B, possibly involving the components themselves and/or regulators of complement activation at these points of the cascade. Such alterations were not observed when ExT was furnished to animals in association with a diet containing high levels of cholesterol, and to our knowledge, components of hamster complement or their respective antibodies are not commercially available. In order to enable us to presently further investigate the mechanism of these eects, the use of the R(2) and R(B) reagents for (indirect) evaluation of these components eects appears justied. We believe that a direct correlation between the results obtained in our in vivo and in vitro studies, is not to be expected. Factors such as intake, absorption and metabolism of our study material can aect its bioavailability and bioeciency (Ross and Kasum, 2002). The ExT is a mixture of many compounds whose bioavailability depends on their structure. The activity of compounds present in the crude tamarind extract could be modied or/and destroyed following absorption and metabolism. Nevertheless, besides its use in research, we consider the possibility that ExT is a potential source of compounds that if properly isolated, may eventually become of therapeutic use. The rationale to use drinking water containing 5% ExT was that this concentration is usually present in tamarind juice drinks consumed by the human population. In order to rule out an eventual damage to the liver by the hypercholestorlemic diet or by the ExT treatment, we investigated possible alterations of serum protein, albumin levels and transaminase activity following experimental treatments. As shown in Fig. 5, hamster liver function was not aected by them. This is of importance considering that the liver is the site of the synthesis of the majority of

the CS components (Paul, 1999), which could be altered by hepatic lesions. However, even if this is occurring at levels not detectable by our analyses, it is important to restate the protective action of the ExT in these conditions. Furthermore, besides its known role in the process of atherosclerosis, increased activity of CS as a consequence of hyperlipemic diets could eventually enhance the triggering of the inammatory process that precedes atherosclerosis. Serum levels of total cholesterol and triacylglycerols were increased in animals given a high-cholesterol diet. In the CE group, triacylglycerol levels were normal, but total cholesterol remained high, although it is worth considering that determinations of total cholesterol may mask variations due to the dierent types of components measured, like high (HDL) and low (LDL) density lipoproteins; this eect was already observed in recent studies on the eect of the ExT on cholesterol levels in our experimental animal model (Martinello et al., 2006). A decrease in serum triglycerides may be associated with the epicatechins present in the ExT (Sudjaroen et al., 2005); they promote high fecal excretion of total fatty acids, neutral and acidic sterols, suggesting that the hypolipidemic activity of this material may be mediated by its inuence on the absorption of dietary fat and cholesterol (Chan et al., 1999). Moreover, the ExT is rich in total sugars, and it has been described that consumption of b-glucan lowers total and nonHDL-cholesterol in plasma due to its soluble ber content (Wilson et al., 2004). Nevertheless, the mechanism by which high-level cholesterol diets aect complement activity and by which ExT prevents this eect, certainly deserves additional investigation. To conclude, the apparent discrepancy between in vivo and in vitro results produced by the ExT, deserves further study of factors involved in its in vivo actions dependent on the intake, absorption and metabolism of its components possibly aecting bioavailability and bioeciency. The nding that ExT appears to contain substances showing diering, even opposite eects on the complement system, deserves consideration, and justies additional eorts to isolate the substances responsible for such eects and to investigate their mechanisms of action.

Acknowledgements The authors are indebted to the Conselho Nacional de co e Tecnolo gico (CNPq) for Desenvolvimento Cient nancial support (Grant Nos. 475267/2003-6 and 306965/ ` Pesquisa do Estado 2003-8) and Fundac a o de Amparo a de Sa o Paulo (FAPESP) (Grant No. 05/00887-9). We thank Denise Pimenta da Silva Leita o and Dr. Adolfo Rothschild for manuscript revision, Ieda Maria Razaboni Prado for ExT components determinations, Alcides Pere nio Zanardo Filho and Joa ira for animal handling, Anto o Franco for biochemical determinations and the priJose sioners of the Ribeira o Preto, SP Penitentiary for helping in the peeling of the fruits of T. indica.

1494

A.P. Landi Librandi et al. / Food and Chemical Toxicology 45 (2007) 14871495 Kidby, D.K., Davidson, D.J., 1973. A convenient ferricyanide estimation of reducing sugars in the nanomole range. Anal. Biochem. 55, 3221 3325. Kingsley, G.R., 1942. The direct biuret method for the determination of serum proteins as applied to photoelectric and visual colometry. J. Lab. Clin. 27, 840845. Komutarin, T., Azadi, S., Butterworth, L., Keil, D., Chitsomboon, B., Suttajit, M., Meade, B.J., 2004. Extract of the seed coat of Tamarindus indica inhibits nitric oxide production by murine macrophages in vitro and in vivo. Food Chem. Toxicol. 42, 649658. Ladeji, O., Omekarah, L., Solomon, M., 2003. Hypoglicemic properties of aqueous bark extract of Ceiba pentandra in streptozotocin-induced diabetic rats. J. Ethnopharmacol. 84, 139142. Libby, P., 2002. Inammation in atherosclerosis. Nature 420, 868874. Martinello, F., Soare, S.M., Franco, J.J., Santos, A.C., Sugohara, A., Garcia, S.B., Curti, C., Uyemura, S.A., 2006. Hypolipemic and antioxidant activities from Tamarindus indica L. pulp fruit extract in hypercholesterolemic hamsters. Food Chem. Toxicol. 44, 810818. Mayer, M.M., 1971. Complement and complement xation. In: Kabat, E.A., Mayer, M.M. (Eds.), Experimental Immunochemistry. Charles C. Thomas Publisher, USA, pp. 133240. Moghadasian, H.M., 2002. Experimental atherosclerosis. A historical overview. Life Sci. 70, 855865. Nistor, A., Bulla, A., Filip, D.A., Radu, A., 1987. The hyperlipidemic hamster as a model of experimental atherosclerosis. Atherosclerosis 68, 159173. Oberholzer, J., Yu, D., Triponez, F., Cretin, N., Andereggen, E., Mentha, G., White, D., Buehler, L., Morel, P., Lou, J., 1999. Decomplementation with cobra venom factor prolongs survival of xenografted islets in a rat to mouse model. Immunology 97, 173180. Pasqui, A.L., Bova, G., Puccetti, L., Bruni, F., Pompella, G., Auteri, A., 2000. Complement activation in hypercholesterolemia. Nutr. Metab. Cardiovasc. Dis. 10, 137142. Paul, E.W., 1999. Complement. In: Paul, E.W. (Ed.), Fundamental Immunology. Lippincott-Raven Publishers, Philadelphia, p. 967. Peters, T., Biamont, G.T., Doumas, B.T., 1982. Albumin in serum. In: Faulkner, W.R., Meites, S. (Eds.), . In: Selected Methods of Clinical Chemistry, vol. 9. AACC Press, Washington, p. 319. Prince, P.S., Menon, V.P., Pari, L., 1998. Hypoglycemic activity of Syzigium cuminii seeds: eect on lipid peroxidation in alloxan diabetes rats. J. Ethnopharmacol. 61, 17. Pumthong, G., 1999. Antioxidant Activity of Polyphenolic Compounds Extracted from Seed Coat of Tamarindus indica Linn. Chiang Mai University, Thailand. Rej, R., Horder, M., 1983. Aspartate aminotransferase. In: Bergmeyer, H.U., Bergmeyer, J., Grasse, M. (Eds.), Methods of Enzymatic Analysis. Verlag Chemic, Germany, pp. 416443. Ross, R., 1999. Atherosclerosis: an inammatory disease. N. Engl. J. Med. 340, 115126. Ross, J.A., Kasum, C.M., 2002. Dietary avonoids: bioavailability, metabolic eects, and safety. Annu. Rev. Nutr. 22, 1934. Sreelekha, T.T., Vijayakumar, T., Ankanthil, R., Vijayan, K.K., Nair, M.K., 1993. Immunomodulatory eects of a polysaccharide from Tamarindus indica. Anticancer Drugs 4, 209212. Sudjaroen, Y., Haubner, R., Wu rtele, G., Hull, W.E., Erben, G., Spiegelhalder, B., Changbumrung, S., Bartsch, H., Owen, R.W., 2005. Isolation and structure elucidation of phenolics antioxidants from Tamarind (Tamarindus indica L.) seeds and pericarp. Food Chem. Toxicol. 43, 16731682. Torzewski, M., Torzewski, J., Bowyer, D.E., Waltenberger, J., Fitzsimmons, C., Hombach, V., Gabbert, H.E., 1997. colocalization of the terminal complex of human complement and smooth muscle cell alpha-actin in early atherosclerotic lesions. Arterioscler. Thromb. Vasc. Biol. 17, 24482452. Vakeva, A., Jauhiainen, M., Ehnholm, C., Lehto, T., Meri, S., 1994. Highdensity lipoproteins can act as carriers of glycophosphoinositol lipidanchored CD59 in human plasma. Immunology 82, 2833.

References
Blumenkratz, N., Assboe-Hansen, G., 1973. New method for quantitative determination of uronic acids. Anal. Biochem. 54, 484489. Bradford, M.M., 1976. A rapid and sensitive method for the quantication of microgram quantities of protein utilizing the principle of proteindye binding. Anal. Biochem. 72, 248254. Chan, P.T., Fong, W.P., Cheung, Y.L., Huang, Y., Ho, W.K.K., Chen, Z.Y., 1999. Jasmine green tea epicatechins are hypolipidemic in hamsters (Mesocricetus auratus) for a high fat diet. J. Nutr. 129, 1094 1101. Dorfman, S.E., Smith, D.E., Osgood, D.P., Lichtenstein, A.H., 2003. Study of diet-induced changes in lipoprotein metabolism in two strains of Golden-Syrian hamsters. J. Nutr. 133, 41834188. Dubey, G.P., Dixit, S.P., Alok, S., 1994. diabetes in rabbits and eect of herbal formulation D-400. Indian J. Pharmacol. 26, 225226. Dubois, M., Gilles, K.A., Hamilton, J.K., Rebers, P.A., Smith, F., 1956. Colorimetric method for determination of sugars and related substances. Anal. Chem. 28, 350356. Ewald, R.A., Williams, J.H., Bowden, D.H., 1961. Serum complement in the newborn. An investigation of complement activity in normal infants and in Rh and AB hemolytic disease. Vox Sang 6, 312319. Ferriani, V.P., Barbosa, J.E., de Carvalho, I.F., 1990. Serum haemolytic classical and alternative pathways of complement in infancy: agerelated changes. Acta Paediatr. Scand. 79, 322327. Folin, O., Ciocalteau, V., 1927. On tyrosine and tryptophane determinations in proteins. J. Biol. Chem. 73, 627650. Geertinger, P., Sorensen, H., 1973. Complement and atherosclerosis. Atherosclerosis 18, 6577. Geleijnse, J.M., Launer, L.J., Hofman, A., Pols, H.A., Witernan, J.C., 1999. Tea avonoids may protect against atherosclerosis: the Rotterdam Study. Arch. Intern. Med. 159, 21702174. Hamuro, J., Hadding, U., Bitter-Suermann, D., 1978. Fragments Ba and Bb derived from guinea pig factor B of the properdin system: purication, characterization, and biological activities. J. Immunol. 120, 438444. Hansson, G.K., Holm, J., Kral, J.G., 1984. Accumulation of IgG and complement factor C3 in human arterial endothelium and atherosclerotic lesions. Acta Pathol. Microbiol. Immunol. Scand. 92, 429435. Harrison, R.A., Lachamnn, P.J., 1986. Complement technology. In: Weir, D.M. (Ed.), Handbook of Experimental Immunology, vol. 1. Blackwell Scientic Publications, Oxford, pp. 39.139.4. Hertog, M.G., Feskens, E.J., Hollman, P.C., Katan, M.B., Kromhout, D., 1993. Dietary antioxidant avonoids and risk of coronary heart disease: the Zuptphen Elderly Study. Lancet 342, 10071011. Hertog, M.G., Kromhout, D., Aravanis, C., Blackburn, H., Buzina, R., Fidanza, F., Giampaoli, S., Jansen, A., Menotti, A., Nedeljkovic, S., Pekkarinen, M., Simic, B.S., Toshima, H., Feskens, E.J.M., Hollman, P.C.H., Katan, M.B., 1995. Flavonoid intake and long-term risk of coronary heart disease and cancer in the seven countries study. Arch. Intern. Med. 155, 381386. Hollander, W., Colombo, M.A., Krkpatrick, B., Paddock, J., 1979. Soluble proteins in the human atherosclerotic plaque. With spectral reference to immunoglobulins, C3-complement component, alpha 1antitrypsin and alpha 2-macroglobulin. Atherosclerosis 34, 391405. Horder, M., Rej, R., 1983. Alanine transaminase. In: Bergmeyer, H.U., Bergmeyer, J., Grasse, M. (Eds.), Methods of Enzymatic Analysis. Verlag Chemic, Germany, pp. 444456. Joisel, F., Leroux-Nicollet, I., Lebreton, J.P., Fontaine, M., 1983. A hemolytic assay for clinical investigation of human C2. J. Immunol. Methods 29, 229235. Kahlon, T.S., Chow, F.I., Sayre, R.N., Betschart, A.A., 1992. Cholesterollowering in hamsters fed rice bran at various levels, defatted rice bran and rice bran oil. J. Nutr. 122, 513519. Keli, S.O., Hertog, M.G., Feskens, E.J., Kromhout, D., 1996. Dietary avonoids, antoxidant vitamins, and incidence of stroke: the Zuptphen study. Arch. Intern. Med. 156, 637642.

A.P. Landi Librandi et al. / Food and Chemical Toxicology 45 (2007) 14871495 Vlaicu, R., Niculescu, F., Rus, H.G., Cristea, A., 1985a. Immunoglobulins and complement components in human aortic atherosclerotic intima. Atherosclerosis 55, 3550. Vlaicu, R., Niculescu, F., Rus, H.G., Cristea, A., 1985b. Immunohistochemical localization of the terminal C5b-9 complement complex in human aortic brous plaque. Atherosclerosis 57, 163177. Wallport, M.J., 2001. Complement. First of two parts. NE J. Med. 344, 10581066. Wilson, T.A., Nicolosi, R.J., Delaney, B., Chadwell, K., Moolchandani, V., Kotyla, T., Ponduru, S., Zheng, G.H., Hess, R., Knutson, N., Curry, L., Kolberg, L., Goulson, M., Ostergren, K., 2004. Reduced

1495

and high molecular weight barley b-glucans decrease plasma total and non-HDL-cholesterol in hypercholesterolemic Syrian Golden hamsters. J. Nutr. 134, 26172622. Yasojima, K., Schwab, C., McGeer, E.G., McGeer, P.L., 2001. Complement components, but not complement inhibitors, are upregulated in Atherosclerotic Plaques. Arterioscler. Thromb. Vasc. Biol. 21, 12141219. Zern, T.L., West, K.L., Fernandez, M.L., 2003. Grape polyphenols decrease plasma triglycerides and cholesterol accumulation in the aorta of ovariectomized guines pigs. J. Nutr. 133, 22682272.