Herbicidal Effects of Statin Pharmaceuticals in Lemna gibba

Jul 14, 2006 - and lovastatin for 7-days by measuring the concentrations of sterols and ubiquinone; products downstream in the MVA pathway. The effici...
0 downloads 0 Views 287KB Size
Environ. Sci. Technol. 2006, 40, 5116-5123

Herbicidal Effects of Statin Pharmaceuticals in Lemna gibba R I C H A R D A . B R A I N , * ,† TAMARA S. REITSMA,† LINDA I. LISSEMORE,‡ KETUT (JIM) BESTARI,† PAUL K. SIBLEY,† AND KEITH R. SOLOMON† Department of Environmental Biology, University of Guelph, Guelph, Ontario, Canada N1G 2W1, and Lab Services Division, University of Guelph, Guelph, Ontario, Canada N1G 2W1

Statin pharmaceuticals, heavily prescribed in the treatment of hypercholesterolemia, are competitive inhibitors of 3-hydroxy-3-methylglutaryl coenzyme-A reductase (HMGR). In plants, these compounds also inhibit HMGR, which regulates cytosolic isoprenoid biosynthesis in the mevalonic acid (MVA) pathway. Phytotoxicity was evaluated in the higher aquatic plant Lemna gibba exposed to atorvastatin and lovastatin for 7-days by measuring the concentrations of sterols and ubiquinone; products downstream in the MVA pathway. The efficiency of the parallel and unaffected methylerythritol phosphate pathway (MEP) was also evaluated by measuring the end product, plastoquinone. Statin treatment caused an accumulation of plastoquinone, and unexpectedly, ubiquinone, an artifact likely due to metabolite sharing from the plastidial MEP pathway. Statins were, however, highly phytotoxic to L. gibba and HPLC-UV analysis of plant extracts showed significantly decreased concentrations of both stigmasterol and β-sitosterol, which are critical components of plant membranes and regulate morphogenesis and development. EC10 values for atorvastatin and lovastatin were as small as 26.1 and 32.8 µg/L, respectively. However, hazard quotients indicated that statins present little risk to the model higher aquatic plant Lemna gibba at environmentally relevant concentrations, even though pathway-specific endpoints were 2-3 times more sensitive than traditional gross morphological endpoints typically used in risk assessment.

Introduction Statin pharmaceuticals are heavily prescribed in the treatment of hypercholesterolemia, a disorder that causes severe elevations in total cholesterol and low-density lipoprotein cholesterol (LDLc) (1). The first statin, mevinolin (lovastatin), isolated from Aspergillus terrus, was found to be a highly potent competitive inhibitor of 3-hydroxy-3-methylglutaryl coenzyme-A reductase (HMGR) (2), the rate-limiting enzyme in cholesterol biosynthesis. Several analogues of mevinolin have been developed, and among all classes of pharmaceuticals, Lipitor (atorvastatin) was the second most prescribed compound in the U.S. at nearly 67 million prescriptions in * Corresponding author: phone: (519) 821-4120 x58627; fax: (519) 837-3861; e-mail: [email protected]. † Department of Environmental Biology. ‡ Lab Services Division. 5116

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 16, 2006

2004 (3). The large number of statin prescriptions reflects a correspondingly large volume of material continually being released into the environment via sewage, raising concerns over potential nontarget effects, particularly in plants, though there is currently little information available with respect to the occurrence, fate, and properties of statins in the aquatic environment (4,5). As in humans, mevinolin is a highly specific inhibitor of HMGR in higher plants (6), the key enzyme regulating the mevalonic acid pathway (MVA) of isoprenoid biosynthesis. The presence of a corresponding receptor in plants, which is known to be inhibited by statins, suggests the question of whether there are significant ecological implications for phytotoxicity due to the discharge of these compounds through sewage. Seiler (7), maintains that, in plants, HMGR is not rate-limiting, making these organisms less, or not at all, susceptible to HMGR inhibitors. However, HMGR is considered the key enzyme in the MVA isoprenoid biosynthetic pathway (8) and statins have displayed phytotoxicity in radish (9) and Lemna (10), as well as Tobacco Bright Yellow-2 (TBY-2) cells (11). In plants, isoprenoids serve numerous key-role biochemical functions including electron transport chains (prenyl side chain of quinones), components of lipid membranes (sterols), subcellular targeting and regulation (prenylation of proteins), photosynthetic pigments (carotenoids and prenyl side chain of chlorophylls), hormones (gibberellins, brassinosteroids, abscisic acid, cytokinins), and defense compounds as well as pollinator attractants (monoterpenes, sequiterpenes, and diterpenes) (12, 13). Two discrete independent pathways for synthesizing isopentyl diphosphate (IPP) and its isomer dimethylallyl phosphate (DMAPP), the universal precursors to isoprenoid biosynthesis, exist in plants; a cytosolic (mevalonate (MVA)) pathway of eukaryotic origin and a chloroplastic (2-C-methyl-D-erythritol 4-phosphate (MEP)) pathway of prokaryotic origin (Figure 1) (14-16). Cytosolic formation of IPP proceeds through the intermediate mevalonate from acetyl-CoA, which under normal physiological conditions, is responsible for the formation of sterols and ubiquinone (8, 14). Chloroplastic formation of IPP involves a condensation of pyruvate and glyceraldehyde-3-phosphate via a 1-deoxy-D-xylulose 5-phosphate intermediate and is used to form isoprene, carotenoids, abscisic acid, and the side chains of chlorophylls and plastoquinone (8, 14, 1719). Although both pathways are compartmentalized, they operate in parallel (20) and cross-talk of intermediates does occur largely through the transfer of IPP, and to a lesser extent, DMAPP, geranylgeranyl diphosphate and farnesyl diphosphate (11, 12, 21). Although metabolite sharing appears to operate more readily from the chloroplast to the cytosol, complementary compensation has been shown for both pathways in the face of inhibitors (lovastatin) or mutants defective in the early steps of IPP biosynthesis (11, 12). The major objective of this study was to evaluate the phytotoxic effects of statin pharmaceuticals in Lemna gibba by analyzing the endogenous concentrations of sterols and quinones after treatment with statin pharmaceuticals. This investigation was conducted in a pathway-specific manner by measuring products downstream of the target enzyme (sterols and potentially ubiquinone) in the MVA pathway after treatment with known inhibitors (statins) of the key enzyme (HMGR). The efficiency of the uninhibited MEP pathway was also investigated by measuring a pathway end product (plastoquinone). Two different statins were used (atorvastatin and lovastatin) in order to evaluate and compare the consistency and severity of effects, and to validate the 10.1021/es0600274 CCC: $33.50

 2006 American Chemical Society Published on Web 07/14/2006

FIGURE 1. Overview of the pathways involved in isoprenoid biosynthesis, the plastidial methylerythritol phosphate (MEP) pathway and the cytosolic mevalonic acid (MVA) pathway modified from ref 19. the metabolism of sterols (β-sitosterol and stigmasterol) and quinones (plastoquinone and ubiquinone) (outlined in boxes) are emphasized, as well as the major pigments, hormones, and isoprenoid precursors. Direct reactions are represented by solid arrows and multiple step reactions are represented by dashed arrows. The inhibition of the rate limiting enzyme 3-hydroxy-3-methylglutaryl coenzyme-A reductase (HMGR) by statins is indicated by an X. The double headed arrow between isopentyl diphosphate and dimethylallayl phosphate indicates metabolite sharing of these intermediates, which occurs more readily from the chloroplast to the cytosol as indicated by the arrowhead weightings. herbicidal effects of statins as a class. The second objective was to evaluate the use and sensitivity of plant-based pathway-specific molecular/biochemical endpoints for use in risk assessment as compared to gross morphological endpoints. The final objective was to conduct a deterministic hazard assessment to evaluate the risk posed by statin pharmaceuticals to L. gibba at environmentally relevant concentrations using novel endpoints.

Experimental Section Chemicals and Reagents. All solvents used were HPLC grade and obtained from Caledon (Georgetown, ON, Canada). Atorvastatin (ATR, 98.0%) was supplied by Rugao Foreign Trade Corp. (Shanghai, China) and lovastatin (LOV, 99.0%) was supplied by Fermic (Mexico City, Mexico). Ubiquinone (UQ, coenzyme Q10) (98.0%), stigmasterol (95.0%), and β-sitosterol (97.0%) standards were obtained from SigmaAldrich (Oakville, ON, CA). Plastoquinone (PQ) was a kind gift from Dr. Jerzy Kruk (Faculty of Biotechnology, Jagiellonian University, Krako´w, Poland). For all analytical procedures, calibration standards were made daily with a minimum acceptable correlation coefficient of linearity of 99% for all calibration curves. Experimental Procedure. Tests were conducted according to a modified ASTM guideline (22) for seven-day static tests with L. gibba, which incorporate a daily static-renewal procedure as outlined in ref 23. Briefly, for each respective test compound, large trays (27 × 38 cm) were used, which

contained 18 experimental units (10 mL (60 × 15 mm) culture dishes) in a randomized complete block design (RCBD) with six treatment concentrations and three replicates. Two L. gibba plants, each with four fronds, were transferred to each culture dish from axenic test cultures. Two sets of identical trays and culture plates were employed for each test. During each static-renewal, respective ATR and LOV treatments were prepared daily in growth (Hunter’s) media and plants were carefully transferred between sets of corresponding trays. Each assay was conducted at 25°C under constant cool white fluorescent light (6756 ( 138 lux) in a growth chamber for 7 days. Extraction of ATR from Plant Growth Media. The ATR extraction procedure was modified from ref 4. All ATR standards were made up in 50:50 methanol:water. Plant growth media samples containing ATR were extracted in dim light using one of two methods available based upon treatment concentrations and detection limit. For treatment concentrations of >300µg/L, samples were directly injected by diluting with an equal volume of methanol. For treatment concentrations 1 indicate a potential for toxicity and values 50% (8, 11). Fresh weight reductions (50%) in L. gibba exposed to LOV were similar at 106-320 µg/L and at 135-157 µg/L for ATR. To our knowledge, these results are the first to show a definitive concentration response relationship between increasing statin treatment concentration and decreased sterol formation in higher plants. In plants, sterols reinforce membranes regulating acyl chain ordering and water permeability of the phospholipid bilayer (31). Alterations in plasma membrane sterol profiles can affect or modulate functions of membrane bound proteins such as enzymes, channels, receptors, or components of signal transduction pathways (31, 32). Sitosterol appears to generally regulate the integrity of the plasma membrane (32), whereas stigmasterol has been shown to modulate the activity of plasma membrane H+-ATPase (33). Sterols are also crucial to cellulose synthesis in the building of the plant cell wall and necessary for cellular division and

TABLE 2. Effective Concentrations Required to Cause a Change of 10, 25, and 50% (EC10’s, EC25’s and EC50’s) in Fresh Weight (FW), as Well as Total, Reduced and Oxidized Plastoquinone, Total Ubiquinone, Stigmasterol and β-sitosterol Concentrations in Lemna Gibba Exposed to Atorvastatin and Lovastatin Using a 7-day Static-renewal Test Procedure Atorvastatin endpoint

response type

FW (quinones) total PQ reduced PQ oxidized PQ total UQ FW (sterols) β-Sitosterol stigmasterol

Decrease Increase Increase Increase Increase Decrease Decrease decrease

endpoint

response type

EC10 (( SE)

EC25 (( SE)

EC50 (( SE)

model

r2b

FW (quinones) total PQ reduced PQ oxidized PQ total UQ FW (sterols) β-sitosterol stigmasterol

Decrease Increase Increase Increase Increase Decrease Decrease Decrease

40.9 ( 8.17 35.2 ( 4.80 19.7 ( 2.51 117 ( 35.3 137 ( 19.1 133 ( 25.8 32.9 ( 30.5 32.8 ( 43.9

64.8 ( 8.08 88.0 ( 11.9 49.1 ( 6.26 292 ( 87.4 343 ( 47.2 219 ( 24.1 34.7 ( 31.8 34.4 ( 46.1

106 ( 8.27 176 ( 23.9 98.3 ( 12.48 NCc 686 ( 95.0 320 ( 15.4 36.7 ( 32.3 36.4 ( 17.1

logistic 4-parameter linear linear linear linear logistic 4-parameter logistic 4-parameter logistic 4-parameter

0.97 0.86 0.92 0.42 0.78 0.97 0.98 0.96

EC10 (( SE)

EC25 (( SE)

EC50 (( SE)

84.1 ( 22.1 105 ( 23.4 135 ( 27.1 17.2 ( 4.12 42.6 ( 10.3 85.1 ( 21.6 8.48 ( 1.64 20.3 ( 12.5 37.7 ( 7.51 611 ( 294 1527 ( 702 3055 ( 1458 17.9 ( 20.1 173.4 ( 106.2 967 ( 332 111 ( 118 131 ( 100 157 ( 69.5 33.1 ( 7.87 45.8 ( 6.05 64.4 ( 2.41 26.1 ( 7.02 41.0 ( 6.45 71.6 ( 6.73

model logistic 4-parameter sigmoidal sigmoidal linear logistic 4-parameter logistic 4-parameter logistic 4-parameter logistic 4-parameter

r2b

LOEC (µg/L)

p-value

hazard quotiente

0.0031