Supplementary Material
Measuring cytochrome P450 activity in aquatic invertebrates: A critical evaluation of in vitro and in vivo methods
Journal: Ecotoxicology
Authors: Michele Gottardi,* Andreas Kretschmann, Nina Cedergreen
*corresponding author, Department of Plant and Environmental Sciences, University of Copenhagen, Thorvaldsensvej 40, 1871 Frederiksberg, Denmark, email:
18 pages, 13 figures and 1 table
Contents
Section 1. In vivo cytochrome P450 activity assay - Optimization 2
Section 2. In vitro cytochrome P450 activity assay - Optimization 6
Section 3. Microsome preparation - Optimization 12
Section 4. 7-hydroxycoumarin standard curves 13
Section 5. The effect of microsomes on 7-hydroxycoumarin fluorescence 15
REFERENCES 18
Section 1. In vivo cytochrome P450 activity assay - Optimization
In vivo measurements of ECOD activity were based and adapted from the existing method described in Gagnaire et al. (2010). In order to optimize the number of organisms needed, one, two and three Chironomus riparius larvae (n = 2 biological replicates) or 10 and 20 Daphnia magna (n = 2 biological replicates) were incubated for 6 hours in 2 mL M7 medium containing the substrate 7-ethoxycoumarin (0.006 mM). The incubation was conducted in 4 mL amber glass vials. The organisms were transferred together with their own growth medium, subsequently the medium was removed and the incubation medium was added. A first blank was made with 7-ethoxycoumarin (0.006 mM) dissolved in M7 medium to correct for the substrate own fluorescence, a second blank was made with organisms in M7 medium to detect possible fluorescence due to molecules excreted by the organisms. Moreover, in order to verify whether product formation was cytochrome P450 dependent, D. magna were continuously pre-exposed to prochloraz (500 µg L-1) for 18 hours prior and during incubation. This was done in agreement with one of our previous investigations, which showed that prochloraz was found to be the strongest inhibitor of in vitro ECOD activity of rat liver microsomes as compared to other azoles fungicides (tebuconazole, epoxiconazole, propiconazole) and the cytochrome P450 inhibitor piperonyl butoxide (Fig. S1). The experiments were conducted at 22 ˚C. During incubation, 100 µL aliquots of medium were removed from the incubation vial and transferred directly into a 96-well plate every 30 min. The samples were stored at -20˚C until fluorescence measurement. Fluorescence due to the product formation (7-hydroxycoumarin; excitation: 380 nm, emission: 480 nm) was measured with a 96-well plate spectrofluorometer at room temperature (25 ˚C).
Fig. S1 ECOD activity of rat liver microsomes (0.01 mg protein mL-1) in the presence of azole fungicides and piperonyl butoxide, PBO (0.060 mM). All groups were statistically significantly different from acetonitrile (ACN, control) (One-way ANOVA, Tukey Test: p < 0.05). Prochloraz, epoxiconazole and propiconazole were not statistically significantly different from one another (One-way ANOVA, Tukey Test: p > 0.05). Data are means ± S.E. (3 analytical replicates)
As it is shown in Fig. S2 and Fig. S3, detectable fluorescence increased within 6 hours for both organisms at all densities. Linearity was observed within the first three hours for D. magna and for approximately five hours for C. riparius, depending on density. For C. riparius there was a lag-phase of approximately half an hour before the fluorescent product was released to the water. For D. magna no detectable lag-phase was observed. Based on these observations the number of organisms were set to ten D. magna and three C. riparious using an incubation time of three hours. Since the organisms in M7 medium did not show any increase in fluorescence over time (Fig. S2 and Fig. S3), this particular control was omitted in the subsequent studies. D. magna exposed to prochloraz prior and during incubation did not show any detectable product formation, hence, pre-exposure to prochloraz (500 µg L-1) was therefore used as a control for cytochrome P450 dependent activity.
Fig. S2 Daphnia magna in vivo ECOD activity. Raw data showing the fluorescence increase over time measured in samples of 0.100 mL M7 medium. The background fluorescence of the substrate (7-ethoxycoumarin) is shown with black circles, the background fluorescence of ten and 20 organisms incubated without substrate is shown with white circles and black triangles, respectively. The change in fluorescence in beakers with substrate and ten or 20 organisms are given with white triangles and black squares, respectively, while change in fluorescence in beakers with substrate and ten or 20 organisms pre-exposed to prochloraz (500 ug L-1) for 18 h prior to incubation and during incubation are given with white squares and black diamonds, respectively. Data are means ± S.E. (n = 2)
Fig. S3 Chironomus riparius larvae in vivo ECOD activity. Raw data showing the fluorescence increase over time measured in samples of 0.100 mL M7 medium. The background fluorescence of the substrate (7-ethoxycoumarin) is shown with black circles, background fluorescence of one, two or three organisms incubated without substrate are given with white circles, black triangles and white triangles, respectively, while changes in fluorescence in beakers incubated with substrate and one, two or three organisms are given with black squares, white squares and black diamonds, respectively. Data are means ± S.E. (n = 2)
Section 2. In vitro cytochrome P450 activity assay - Optimization
The starting point for measuring ECOD activity was the method adopted by Bach and Snegaroff (1989) for the liver, intestines and other organs of the rainbow trout (Salmo gairdneri) using phosphate buffer (KH2PO4/K2HPO4) (50 mM, pH = 7.5), MgCl2 (1.5 mM), NADP (0.25 mM), glucose 6-phosphate (2.5 mM), glucose 6-phosphate dehydrogenase (3 units) and 7-ethoxycoumarin (0.1 mM). The optimal concentrations of microsomes, substrate and of the two cofactors, NADPH and MgCl2 were then investigated using rat liver microsomes (Product nr: M9066, Sigma Aldrich), which provided a system known to possess high P450 activity. The aim was to get a very sensitive test with a low detection limit for ECOD activity that would be able to capture the low activities known to occur in aquatic invertebrates. The five parameters were varied one at a time to obtain the optimal concentration as shown in Table S1. All tests were run in phosphate buffer (50 mM, pH = 7.5).
Table S1. Summary of chemicals and their concentrations used during in vitro ECOD activity assay optimization. Underlined values represent the range tested during optimization, while the other factors were kept constant.
Microsomes / MgCl2 / NADPH / Substrate / SolventMicrosomes (mg protein mL-1) / 0.0006 – 0.08 / 0.03 / 0.03 / 0.03 / 0. 03
MgCl2 (mM) / 1. 5 / 0 - 26 / - / - / -
NADPH (mM) / 0.25 / 0. 25 / 0 – 0.5 / 0.0156 / 0.0156
7-ethoxycoumarin (mM) / 0.1 / 0.1 / 0.1 / 0.04 – 2.5 / 0.1
ACN (%) / 0.12 / 0.12 / 0.12 / 3 / 0.12 – 10.12
Geometrically distributed dilutions of microsomes were made and fluorescence was monitored over time (Fig. S4). A concentration of microsomes in the range of detectable product formation within 30 min equal to 0.03 mg protein mL-1 was used for further investigations.
Fig. S4 ECOD activity of rat liver microsomes – Fluorescence increase over time of geometrical dilutions of rat liver microsomes. Five of the eight concentrations are shown: 0.08 mg protein mL-1 by black circles, 0.04 mg protein mL-1 by white circles, 0.02 mg protein mL-1 by black triangles, 0.01 mg protein mL-1 by white triangles and 0.005 mg protein mL-1 by black squares. Data are means ± S.E. (3 analytical replicates)
Subsequently, ECOD activity was monitored over 30 min using different geometrically distributed MgCl2 concentrations (Fig. S5). MgCl2 did not seem to greatly enhance nor inhibit ECOD activity and was therefore excluded in all following experiments.
Fig. S5 ECOD activity of rat liver microsomes – Optimization of MgCl2 concentration. Data are means ± S.E. (3 analytical replicates)
Fluorescence increase over time was monitored in relation to geometrically distributed dilutions of NADPH (Fig. S6). In order to minimize background noise (cause by NADPH fluorescence at the working wavelengths) (Aitio 1978) the lowest concentration that showed no depletion (linear product formation) within 30 minutes was considered as optimal. For the chosen microsome concentration this was 0.0156 mM NADPH.
Fig. S6 ECOD activity of rat liver microsomes – Optimization of NADPH concentration. Fluorescence increase over time with geometric dilutions of NADPH. Four of the eight concentrations are shown: 0.0312 mM by black circles, 0.0156 mM by white circles, 0.0078 mM by black triangles and 0 mM by white triangles. Data are means ± S.E. (3 analytical replicates)
In order to find the optimal substrate concentration, ECOD activity was investigated in relation to ten geometrical dilutions of substrate and by estimating kinetic parameters such as Vmax and KM (Michaelis Menten kinetics: rate = Vmax [S]/([S]+KM). Curve fitted using SigmaPlot 12.5) (Fig. S7). Moreover, since the substrate was solubilized in a solvent: 30 % v/v acetonitrile (ACN) in MilliQ water, the possible inhibition of activity due to the solvent (Li et al. 2010) was tested by addition of known amounts of solvent to the test system (Fig. S8). A concentration of substrate within the range of Vmax that did not significantly decrease activity due to the solvent was considered optimal. We chose the concentration of 0.6 mM 7-ethoxycoumarin, giving a total acetonitrile concentration in the sample of 0.72 % v/v.
Fig. S7 ECOD activity of rat liver microsomes – Optimization of substrate (7-ethoxycoumarin) concentration. ECOD activity as a function of substrate concentration. Kinetic parameters: Vmax: 29.4 ± 0.6 flu min-1, KM: 0.19 ± 0.01 mM 7-ethoxycoumarin estimated with Michealis-Menten equation fit (black line). Data are means ± S.E. (3 analytical replicates)
Fig. S8 ECOD activity of rat liver microsomes – Optimization of solvent (Acetonitrile) concentration. Data are means ± S.E. (3 analytical replicates)
The lowest detectable activity of the method was investigated by measuring the activity of geometrical dilutions of rat liver microsomes (0.008 to 0.0001 mg protein mL-1) and Chironomus riparius larvae microsomes (0.04 to 0.002 mg protein mL-1) with the optimized amounts of reagents. The lowest fluorescence increase over time that was found to be statistically significant from the blank (no substrate) (ANCOVA, p < 0.01, run with the program "R") was considered to be the lowest detectable activity of the method developed in the present study. For rat liver microsomes, the lowest detectable activity was 0.21 ± 0.08 flu min-1 (± S.D.) corresponding to ~34 fmol min-1 and a concentration of microsomes of 0.001 mg protein mL-1. For Chironomus riparius larvae microsomes, the lowest detectable activity was 0.25 ± 0.07 flu min-1 (± S.D.) corresponding to ~37 fmol min-1 and a concentration of microsomes of 0.036 mg protein mL-1.
Section 3. Microsomes preparation - Optimization
As a proper extraction of microsomes from whole organisms with an exoskeleton requires severe homogenization to ensure cell breakage, we chose to use an ultrasonic stick (Digital Sonifier cell disruptor Model 450, Branson Ultrasonics, U.S.). Homogenization with ultrasonic stick can, however, potentially heat and thereby denature proteins resulting in a substantial decrease in detectable ECOD activity. Therefore two types of homogenization trials were conducted: First, studies using rat liver microsomes were conducted to identify optimal mode and processing times that would not halter P450 activity. Three settings: 3x3 sec, 6x3 sec and 12x3 sec with 10 sec pause in between on ice and the power set to 20 % were tested on rat liver microsomes. Foaming was observed for 6x3 sec and 12x3 sec, and subsequent measurements of ECOD activity showed that 12x3 sec decreased the activity of rat liver microsomes by 27 % as compared to the control. Secondly, homogenization trials were made with D. magna using either 20 or 40 five day old organisms in 0.9 mL of phosphate buffer (50 mM, pH = 7.5) and glycerol (10 % v/v). Two different power outputs, 10 % and 20 %, and various processing times starting with 1x3 sec to 14x3 sec with 10 sec pause on ice in between were tested. Results showed that the total number of organisms was not fully homogenized when low power output or processing time shorter than 12 sec were used. On the other hand, when high power or long processing time were used, the homogenate was foaming, which was previously shown to decrease P450 activity in rat-liver microsomes. To minimise the homogenization time using the ultrasonic stick while still ensuring full cell disruption, a preliminary cell disruption carried out by hand with a tissue grinder (Econo Grind, Radnoti, U.S.) was therefore adopted. After that, ultrasonic homogenization was carried out with the optimal settings found for rat liver microsomes: power output equal to 20 % and a time sequence of 3x3 sec with 10 sec pause in between.
Section 4. 7-hydroxycoumarin standard curves
Fig. S9 Standard curve of 7-hydroxycoumarin in MilliQ water and rat liver microsomes (0.03 mg protein mL-1) (Total volume 200 µL). Fit: y = 51.1 + 6.266x. Data are means ± S.E. (2 analytical replicates)
Fig. S10 Standard curve of 7-hydroxycoumarin in MilliQ water (Total volume 100 µL). Fit: y = 27.1 + 8.218x. Data are means ± S.E. (3 analytical replicates)
Fig. S11 Standard curve of 7-hydroxycoumarin in phosphate buffer (0.13 M) (Total volume 200 µL). Fit: y = 46.1 + 6.803x. Data are means ± S.E. (3 analytical replicates)
Section 5. The effect of microsomes on 7-hydroxycoumarin fluorescence
Different amounts of microsomes (final concentration in the well ranging from 0 to 0.05 mg protein mL-1) obtained from C. riparius larvae and D. magna were incubated with the ECOD product 7-hydroxycoumarin (60 pmol) in order to investigate possible decrease of fluorescence due to the conversion of the product to non-fluorescence molecules (Fig. S12) and possible scattering of fluorescence signal due to turbidity of the samples (Fig. S13).