Академический Документы
Профессиональный Документы
Культура Документы
www.elsevier.com/locate/chemosphere
Received 13 August 2003; received in revised form 14 June 2004; accepted 13 July 2004
Abstract
Postmetamorphic South African clawed frogs (Xenopus laevis) were exposed to a phytosterol mixture (ca. 80% b-
sitosterol and less sitostanol, campesterol, and campestanol) for 14 days at 30 lg l1 in a flow-through system. The
effects of phytosterols (PS) on the plasma thyroid hormone (T3 and T4), testosterone, leptin-immunoreactive peptide
and tissue glycogen concentrations were determined. The following enzyme activities were also analyzed from the liver
and muscle: glycogen phosphorylase and lipase, and from the liver only: glucose-6-phosphatase. The plasma T3 concen-
tration was lower in the PS-exposed female frogs. Both muscle lipase and glycogen phosphorylase activities were also
lower in the PS-exposed animals. These results could indicate that the basal metabolic rate and locomotion activity of
the frogs were decreased. The effects could not be attributed to the possible estrogenicity of the PS mixture. Further
studies will be needed to evaluate the possible significance of these effects.
2004 Elsevier Ltd. All rights reserved.
0045-6535/$ - see front matter 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.chemosphere.2004.07.029
1684 P.S. Koponen et al. / Chemosphere 57 (2004) 1683–1689
much interest has also been focused on PS due to their lowing day the temperature was raised to 20 ± 1 C. The
beneficial effects on the serum lipid profile. As a result control treatment contained pure tap water and the sol-
of this, they are consumed in various remedies and mar- vent control contained as much ethanol as the PS expo-
garines by the general population in developed countries sure water (0.03% (v/v)). The water volume in the glass
to lower elevated serum cholesterol concentrations test aquaria was adjusted to 10 l, and the inflow was ad-
(Moghadasian, 2000). justed to 7 ml min1 (approximately 10 l d1), with a re-
Amphibians may be exposed to xenobiotics via many newal time of 24 h. Fresh test and control solutions
routes: through food, their semipermeable skin and the were placed in the aquaria before the frogs were intro-
aquatic environment in which their eggs and larvae duced, and the flow-through was started. Previous exper-
develop. The ontogenetic shift from herbivorous tadpoles iments in our laboratory have showed that the nominal
to carnivorous adults makes them vulnerable to several water concentration of Ultrasitosterol remains constant
exposure routes (Duellman and Trueb, 1986). As a totally in the exposure waters after 48 h (Honkanen et al., unpub-
aquatic and carnivorous animal the postmetamorphic lished). The total number of aquaria was ten. Six aquaria
African clawed frog (Xenopus laevis) is an excellent model were used for the ultrasitosterol treatment and two aqua-
for studying the effects of PS. The aims of our experiment ria for both control and solvent control treatments. Each
were to study how a 14-d subacute flow-through PS expo- aquarium contained two individuals. Seven females and
sure (30 lg l1) affects particular parameters of reproduc- 5 males were randomly placed in the ultrasitosterol
tion and energy metabolism in postmetamorphic X. treatment aquaria. Two females and males were ran-
laevis. The plasma testosterone, leptin-immunoreactive domly placed in both control and solvent control treat-
peptide, thyroxine and triiodothyronine concentrations ments. The temperature of the aquaria was maintained
were measured at end of the exposure. Glycogen content, at 20 ± 1 C. The frogs were fed with commercial
and lipase and phosphorylase activities were determined frog food (Frog Brittle, Nasco) every other day, but not
from liver and muscle tissues, and glucose-6-phosphatase on the sampling day. The length of the exposure was 14
activity from liver tissue. days.
2.1. Sampling
2. Materials and methods
At the end of the exposure, the frogs were euthanized
Animals (n = 20, 9 males and 11 females, age >1.5 in a solution containing 1.5 g l1 of ethyl 3-aminobenzo-
years, sex verified at necropsy) were obtained from X. ate methanesulfonate salt (MS-222, Sigma–Aldrich
laevis stock of the University of Joensuu, Department Chemie GmbH, Steinheim, Germany) and sampled
of Biology. The temperature in the stock aquaria was immediately after the MS-222 treatment. Their body
maintained at 20 ± 1 C. Before the experiment the frogs mass was measured to the nearest 0.1 g. Blood samples
were fed ad libitum every other day with commercial were obtained by cardiac punctures with sterile needles
frog food (Frog Brittle, Nasco, Fort Atkinson, WI, and syringes containing EDTA to prevent clotting,
USA). The animals in the aquaria were maintained in transferred to Eppendorf vials and centrifuged at
high quality tap water (nonchlorinated natural ground 1000 · g to obtain plasma. The separated plasma was
water), and lighting was supplied by fluorescent lamps. frozen in liquid nitrogen, stored at 20 C, and used
The photoperiod was 16 h light, 8 h dark. for testosterone, leptin-immunoreactive peptide, triiodo-
A stock solution of PS mixture powder extracted thyronine (T3) and thyroxine (T4) analyses. The whole
from pulp mill effluents (Ultrasitosterol, UPM-Kym- liver, gonads and fat body, as well as a muscle tissue
mene Corporation, Chemical Mill, Lappeenranta, Fin- sample from the right hind leg were dissected and
land; 75.7% b-sitosterol, 13% sitostanol, 9% campe- weighed to the nearest 0.1 g. The liver and muscle sam-
sterol and campestanol, 0.9% artenols) was prepared ples were frozen in liquid nitrogen and stored at 20 C
by dissolving the powder in ethanol. The solution was for the enzyme analyses.
stored in the dark at 5 C. It is known that MS-222 may interact with biotrans-
A flow-through system was used to expose the animals formation enzymes (Fabacher, 1982). However, those
to PS and control and solvent control treatments were in- results were obtained by using relatively high MS-222
cluded. All solutions were prepared the day before the concentrations (112.5 g l1) compared to the concentra-
exposure in 11-l plastic vessels and mixed carefully. The tion used in this study (1.5 g l1). Even if the used con-
PS solution (30 lg l1) was prepared by diluting the centration had interfered with biotransformation
appropriate volume of stock solution with cold tap water. enzymes, it is obvious that the changes would not have
Earlier studies have shown that the used ultrasitosterol been immediately reflected in the circulating hormone
concentration is capable of inducing reproductive impacts concentrations. Furthermore, there should be no con-
on fish and is an environmentally relevant concentration founding factor since all the frogs were treated similarly
(Lehtinen et al., 1999; Mattson et al., 2001b). On the fol- with MS-222.
P.S. Koponen et al. / Chemosphere 57 (2004) 1683–1689 1685
2.2. Hormone analyses met. The tests were performed using SPSS 11.01 for
Windows (SPSS, Chicago, IL, USA). If there were sig-
The plasma testosterone and thyroid hormone con- nificant differences, a multiple comparison test was per-
centrations were measured with Spectria [125I] Coated formed according to Conover (1980). The normality of
Tube Radioimmunoassay (RIA) kits (T3: intraassay var- distribution was determined by the Kolmogorov–Smir-
iation 3.3–6.1 CV(%), interassay variation: 4.5–7.5 nov test and homogeneity of variances by Levene test.
CV(%), sensitivity: 0.2 nmol l1, specificity: T3: 100%, The correlations were calculated using SpearmanÕs cor-
T4: 0.15%, T2: 0.2%. T4: intraassay variation 3.3–6.8 relation coefficient. The differences were considered to
CV(%), interassay variation: 3.3–8.0 CV(%), sensitivity: be significant at the P < 0.05 level. The results are ex-
5 nmol l1, specificity T4: 100%, T3: 1.2%, T2: < 0.1%. pressed in the tables as the mean ± SD. As no differences
Testosterone: intraassay variation 3.8–7.5 CV(%), inter- were found between the control and solvent control
assay variation: 4.8–7.0% CV(%), sensitivity: 0.1 treatments, their results were pooled and presented
nmol l1, specificity: testosterone: 100%, 5-a-dihydrotes- together.
tosterone: 4.5%, others: <1%. Orion Diagnostica,
Espoo, Finland). Measurement of the plasma leptin-
immunoreactive peptide concentrations was carried out 3. Results
with the Multi-species Leptin RIA kit (intraassay varia-
tion: 2.8–3.4 CV(%), interassay variation: 6.5–8.7 The PS mixture caused a decrease in the plasma T3
CV(%), specificity: human leptin: 100%, insulin, gluca- concentrations in the exposed females (Table 1). The
gon, etc: <0.01%. Linco Research Inc., St. Charles, plasma T4 concentration was below the RIA kitÕs detec-
MO, USA). The leptin assay was validated by serial tion limit in all measured cases. The mean testosterone
dilutions of the X. laevis plasma, showing linear changes concentration was higher in the males than in the fe-
in BB1
0 values that were parallel to the standard curve males, as expected. However, no differences were found
produced with human standards (data not shown). For in the testosterone concentrations between the treat-
the quantification of radioactivity in the RIA, a gamma ments. There were no differences in the leptin-immuno-
counter, 1480 Wizard 300 , Wallac, Turku, Finland was reactive peptide concentrations between the sexes or
used. the treatments. Neither were there any correlations be-
tween the leptin-immunoreactive peptide concentrations
2.3. Enzyme analyses and any of the measured parameters.
Both exposed males and females showed a significant
The various enzyme activities were determined spec- decrease in muscle lipase activity due to the PS treat-
trophotometrically. The liver samples were weighed ment (Table 2). The liver lipase activity was slightly
and homogenized in cold citrate buffer at pH 6.5 for higher in the males than in the females, but the difference
the glucose-6-phosphatase (G6Pase), the liver and mus- was statistically significant only between the control
cle samples at pH 6.1 for the glycogen phosphorylase males and females. The muscle phosphorylase activity
measurements, and in cold 0.85% sodium chloride for was also lower in the PS-exposed animals, but signifi-
the lipase measurements. The activity of G6Pase was cantly lower only when compared to control females.
measured using glucose-6-phosphate as a substrate in The liver glycogen concentration was higher in the
the presence of EDTA after an incubation period of
30 min at 25 C (Hers and van Hoof, 1966). The glyco-
gen phosphorylase activity was measured in the presence Table 1
of glucose-1-phosphate, glycogen, sodium fluoride and Plasma hormone concentrations (mean ± SD) of Xenopus laevis
AMP according to the method of Hers and van Hoof Plasma hormone Control PS exposure
(1966). The lipase activity was measured according to
the method of Seligman and Nachlas (1962) using 2- Testosterone (nmol l1)
Males 4.94 ± 3.28b (4) 5.57 ± 4.04b (5)
naphthyl-laurate without taurocholate as a substrate.
Females 0.19 ± 0.17a (3) 5.48 ± 6.30b (4)
Glycogen concentrations in the liver and muscle were
measured spectrophotometrically according to the T3 (nmol l1)
method of Lo et al. (1970). Males 0.53 ± 0.20b (3) 0.41 ± 0.12b (5)
Females 0.57 ± 0.12b (3) 0.28 ± 0.08a (6)
2.4. Statistical analyses Leptin-immunoreactive peptide (ng ml1)
Males 0.90 ± 0.38 (3) 0.48 ± 0.13 (3)
In order to evaluate differences in the measured var- Females 0.99 ± 0.51 (3) 1.10 ± 0.70 (3)
iables between the exposed and control frogs, Kruskal- Values with no common letter differ at P < 0.05 level. Numbers
Wallis nonparametric analysis of variance (ANOVA) in brackets indicate number of successful analyses and n in
was used, since the requirements of ANOVA were not statistical tests.
1686 P.S. Koponen et al. / Chemosphere 57 (2004) 1683–1689
(1998). Probably reduced intestinal uptake or increased 2002). The decreased level of lipid metabolism may be
metabolism of cholesterol could contribute to reduced a direct effect of PS.
sex steroid synthesis (Tremblay and Van Der Kraak, Leptin, a protein produced primarily by adipocytes
1999). Several human and animal studies have also (Zhang et al., 1994; Reidy and Weber, 2000), increases
shown that PS significantly reduce plasma total and the levels of appetite-reducing factors such as a-melano-
low-density lipoprotein cholesterol concentrations cyte-stimulating hormone and lowers the levels of appe-
(Moghadasian, 2000). Another possible mechanism is tite-enhancing peptides, such as neuropeptide Y and
that b-sitosterol and other compounds present in pulp agouti-related protein (Di Marzo et al., 2001). Thus, lep-
mill effluents inhibit the activity of specific enzymes tin is considered to be a signal through which the hypo-
within the steroid biosynthetic pathway (McMaster thalamus senses the nutritional state of the animal.
et al., 1996). Previous studies on rainbow trout have Leptin is also an indicator of mammalian body adiposity
shown that PS may affect cholesterol metabolism by with a positive correlation between plasma leptin con-
increasing the levels of bound cholesterol in bile (Lehti- centrations and body mass index in humans and rodents
nen et al., 1999). In addition, animal studies have (Maffei et al., 1995). Leptin-like proteins have been dis-
shown that PS may accumulate in the liver, adrenal covered also in fish (Mustonen et al., 2002; Nieminen
glands, ovaries and testes (Moghadasian, 2000). In this et al., 2003b) and lizards (Niewiarowski et al., 2000).
regard, researchers have reported synthesis of cortisol To our knowledge, this was the first experiment where
and sex steroid hormones from PS in human and rat leptin-immunoreactive peptide could be measured from
endocrine tissues. X. laevis. The RIA kit used for the leptin measurements
seems to be functional with X. laevis, but it is, of course,
4.3. Carbohydrate metabolism not certain that the binding component is a ‘‘true’’ lep-
tin. However, the leptin-immunoreactive peptide con-
PS exposure did not affect the muscle or liver glyco- centrations were unchanged in the exposed animals,
gen concentrations. The exposed females showed a sig- and they were not an indicator of adiposity in X. laevis,
nificant decrease in muscle glycogen phosphorylase i.e. the correlations between the leptin-immunoreactive
activity, but liver G6Pase activity was not affected. Gly- peptide and the body mass or FSI were not significant.
cogen phosphorylase operates at the strategic point of This observation is parallel to previous findings on the
glycogen degradation, as the initial catalytic factor that common shrew (Sorex araneus) (Nieminen and Hyväri-
makes glycogen available for energy release (Lehninger nen, 2000) and the mink (Mustela vison) (Nieminen
et al., 1993). When glucose is consumed or created by et al., 2000).
gluconeogenesis, it is phosphorylated to glucose-6-phos-
phate in preparation for conversion into glycogen.
In the same way, glycogenolysis occurs by a reverse 5. Conclusions
mechanism, with the production of glucose-6-phos-
phate. This must be dephosphorylated in liver by In conclusion, the results of this study show that
G6Pase before it can be released as free glucose that exposure to an environmentally relevant concentration
can be used by the body. It is unclear why the muscle of PS induces physiological changes in frogs, but they
glycogen phosphorylase activity of the females was are not detrimental to adults. The effects of PS are quite
decreased. different in mammals, amphibians and fish. PS exposure
decreases plasma T3 concentration in frogs, and this
4.4. Lipid metabolism merits further experiments with larval stages, as the met-
amorphosis of frogs is directly stimulated by thyroid
The measured lipase activity indicates the hydrolyz- hormones (e.g. Just and Kraus-Just, 1996). Further
ing activity of lipases (Lehninger et al., 1993). Adipo- studies are needed to determine the biological signifi-
cytes contain lipases that break down stored cance of the observed changes in hormonal and enzyme
triacylglycerols (TG), releasing fatty acids for export to levels.
other tissues where they are required as fuel. The activity
of lipases was determined both from liver and muscle tis-
sue. In muscle, the energy content of TG store is higher Acknowledgments
than the energy content of the muscle glycogen pool.
The intramuscular TG constitutes a dynamic energy The authors thank Mr. Ulf Hotanen (UPM-Kym-
store, which can be mobilized by lipases. The PS- mene Oyj, Kaukas, Lappeenranta, Finland), who kindly
exposed frogs showed a significant decrease in the mus- provided the ultrasitosterol powder, and Mrs. Anita
cle lipase activity. This is similar to previous findings in Kervinen for technical assistance in the analyses of en-
mammals. The European polecat exposed to PS also zyme activities. This study was funded by the Maj and
showed decreased liver lipase activity (Nieminen et al., Tor Nessling Foundation (project 2003093).
1688 P.S. Koponen et al. / Chemosphere 57 (2004) 1683–1689
study the evolution of life histories and endothermy? J. Exp. Tremblay, L., Van Der Kraak, G., 1998. Use of a series of
Biol. 203, 295–300. homologous in vitro and in vivo assays to evaluate the
Rabelo, E.M., Baker, B.S., Tata, J.R., 1994. Interplay between endocrine modulating actions of b-sitosterol in rainbow
thyroid hormone and estrogen in modulating expression of trout. Aquat. Toxicol. 43, 149–162.
their receptor and vitellogenin genes during Xenopus met- Tremblay, L., Van Der Kraak, G., 1999. Comparison between
amorphosis. Mech. Dev. 45, 49–57. the effects of the phytosterol b-sitosterol and pulp and paper
Reidy, S.P., Weber, J.-M., 2000. Leptin: an essential regulator mill effluents on sexually immature rainbow trout. Environ.
of lipid metabolism. Comp. Biochem. Physiol. A 125, 285– Toxicol. Chem. 18, 329–336.
297. Zhang, Y., Proenca, R., Maffei, M., Barone, M., Leopold, L.,
Seligman, A.M., Nachlas, M.M., 1962. Lipase. In: Bergmeyer, Friedman, J.M., 1994. Positional cloning of the
H.U. (Ed.), Methoden der Entzymatischen Analyse. Verlag mouse obese gene and its human homologue. Nature 372,
Chemie GmbH, Weinheim, pp. 776–778. 425–432.