Persistent behavioral abnormalities have been commonly associated with acute organophosphate (OP) pesticide poisoning; however, relatively little is known about the consequences of chronic OP exposures that are not associated with acute cholinergic symptoms. In this study, the behavioral and neurochemical effects of chronic, intermittent, and subthreshold exposures to the OP pesticide, chlorpyrifos (CPF), were investigated. Rats were injected with CPF s.c. (dose range, 2.5–18.0 mg/kg) every other day over the course of 30 days and then were given a 2-week CPF-free washout period. In behavioral experiments conducted during the washout period, dose-dependent decrements in a water-maze hidden platform task and a prepulse inhibition procedure were observed, without significant effects on open-field activity, Rotorod performance, grip strength, or a spontaneous novel object recognition task. After washout, levels of CPF and its metabolite 3,5,6-trichloro-2-pyridinol were minimal in plasma and brain; however, cholinesterase inhibition was still detectable. Furthermore, the 18.0 mg/kg dose of CPF was associated with (brain region-dependent) decreases in nerve growth factor receptors and cholinergic proteins including the vesicular acetylcholine transporter, the high-affinity choline transporter, and the α7-nicotinic acetylcholine receptor. These deficits were accompanied by decreases in anterograde and retrograde axonal transport measured in sciatic nerves ex vivo. Thus, low-level (intermittent) exposure to CPF has persistent effects on neurotrophin receptors and cholinergic proteins, possibly through inhibition of fast axonal transport. Such neurochemical changes may lead to deficits in information processing and cognitive function.
Neurobehavioral sequelae of acute and chronic organophosphate (OP) exposure have been described in the literature for decades (Gershon and Shaw, 1961; Tabershaw and Cooper, 1966). Such effects have been reported in cases of exposure to military OP nerve agents, including sarin, soman, and VX (Brown and Brix, 1998); however, most of the more recent human data come from studies of occupational poisonings from OP insecticides (for review, see Roldán-Tapia et al., 2005). Persistent behavioral consequences of OP pesticide poisoning include the deterioration of intellectual functioning, reasoning, and academic ability as well as impaired motor skills (Savage et al., 1988). It should be noted, however, that although a large number of human and animal studies have focused on the long-term consequences of acute OP exposure, relatively little attention has been given to the subject of chronic, “low-level” OP exposures that are not associated with acute cholinergic symptoms (Ray and Richards, 2001). This type of chronic exposure may be a particular concern, given the widespread use of OP insecticides (and consequent human exposure) in household, agricultural, and commercial environments worldwide.
One OP that has been used extensively as an agricultural and commercial pesticide since its introduction in 1965 (Hayes and Laws, 1990) is chlorpyrifos (O,O-diethyl O-[3,5,6,-trichloro-2-pyridyl] phosphorothionate, CPF). This broad spectrum OP insecticide has been described as “non-neurotoxic,” “non-neuropathic,” or “moderately toxic,” on the basis of published evidence that it exhibits only moderate acute toxicity in mammalian species and a greater inhibitory potency for acetylcholinesterase than for neurotoxic esterase (for review, see Richardson, 1995). However, because of concerns over published evidence of developmental and neurobehavioral anomalies in young animals exposed to CPF, its use (particularly for residential applications) has been restricted in the United States by the Environmental Protection Agency (U.S. Environmental Agency, 2002). Despite such restrictions, the major metabolite of CPF, 3,5,6-trichloro-2-pyridinol (TCP), was recently detected in 96% of approximately 2000 urine samples collected from individuals (ages 2–59 years) living in the United States (Barr et al., 2005). Furthermore, CPF continues to be used for residential applications (and for other pest control-related purposes) in the countries of the European Union and elsewhere in the world.
Whereas the insecticidal actions of CPF and its acute toxicity in nontarget organisms have been attributed to inhibition of acetylcholinesterase by the reactive CPF-oxon metabolite (Amitai et al., 1998), as in the case of other OPs, the consequences of chronic, low-level exposures to CPF are poorly understood (a focus of our laboratories over the last several years). Previously we observed spatial learning deficits in rats after 14 days of daily exposure to CPF (18.0 and 25.0 mg/kg) when testing was initiated 24 h after the last injection but not after a 14-day washout. We also observed that axonal transport was impaired in sciatic nerves that were isolated from these rats (Terry et al., 2003), a potentially important finding given the fundamental role of axonal transport in the function of neurons. Here we specifically focused on persistent effects (i.e., during and after an extended CPF-free washout) of repeated, intermittent, and subthreshold exposures to CPF on behavior, neurotrophin and cholinergic marker proteins, and axonal transport. We have operationally defined “subthreshold exposures” as doses that do not produce overt signs of cholinergic toxicity such as muscle fasciculations, respiratory muscle paralysis, seizures, diarrhea, urination, meiosis, salivation, and lacrimation (for reviews, see Rusyniak and Nanagas, 2004; Sungurtekin et al., 2006). The intermittent dosing regimen was used to provide a model for the types of exposure that might be experienced by agricultural or industrial workers. Subsequent neurochemical studies focused on central cholinergic pathways (i.e., those originating from the basal forebrain), which are known to play important roles in many aspects of cognition (for review, see Bartus, 2000). The expression and/or the intrinsic activities of protein markers residing in these cholinergic pathways (e.g., choline acetyltransferase, the high-affinity choline transporter, and the vesicular acetylcholine transporter) have been used for decades to ascertain the consequence of disease or injury in these cells. We also assessed the effects of CPF on homo-oligomeric α7 nicotinic acetylcholine receptors (nAChRs), which are important to cognitive function and highly expressed in the cortical and hippocampal neurons that receive innervation from the basal forebrain (Buccafusco, 2004; Dani and Bertrand, 2007). Finally, we evaluated the effects of CPF on the receptors for the neurotrophin, nerve growth factor, which, through its interactions with high-affinity TrkA and p75 neurotrophin receptors (p75NTR), is important for the maintenance and survival of basal forebrain cholinergic neurons (Li et al., 1995; Auld et al., 2001).
Materials and Methods
Male albino Wistar rats (Harlan, Indianapolis, IN), 2 to 3 months old, were doubly housed in a temperature-controlled room (25°C) and were maintained on a reversed 12-h light/dark cycle with free access to food (Teklad Rodent Diet 8604 pellets; Harlan, Madison, WI). Table 1 provides the details for all study cohorts, the numbers of animals tested per group, and the experiments conducted in each group. All procedures used during this study were reviewed and approved by the Medical College of Georgia Institutional Animal Care and Use Committee and are consistent with Association for Assessment and Accreditation of Laboratory Animal Care guidelines. Measures were taken to minimize pain or discomfort in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (Publication 80-23, revised 1996). Significant efforts were also made to minimize the total number of animals used while maintaining statistically valid group numbers.
Drug Administration and Observational Studies
Each experimental group received s.c. injections of vehicle (3% dimethyl sulfoxide and 97% v/v peanut oil) or CPF (Chem Service, Inc., West Chester, PA) dissolved in vehicle in a volume of 0.7 ml/kg b.wt. every other day over a 30-day treatment period. Individual rats were weighed and monitored (in their home cages for a period of approximately 5 min) for visible cholinergic symptoms (e.g., diarrhea, excessive salivation or lacrimation, respiratory difficulties, or muscle fasciculations) or other signs of distress throughout the study.
All behavioral experiments were conducted in rooms equipped with white noise generators (San Diego Instruments, San Diego, CA) set to provide a constant background level of 70 dB and ambient lighting of approximately 25 to 30 lux (lumens per square meter). Animals were transferred (in their home cages) to the behavioral testing rooms each morning approximately 30 min before the beginning of experiments.
Motor Function Tests
Open-Field Activity. Rat open-field activity monitors (43.2 × 43.2 cm; MED Associates, St. Albans, VT) were used for these experiments. The following parameters were recorded for each 5-min test session: horizontal activity (horizontal photobeam breaks or counts), number of stereotypy movements, and vertical activity (vertical photobeam breaks). Thus, spontaneous locomotor activity, olfactory activity (rearing and sniffing movements), and stereotypical movements were assessed. We also recorded the time spent in the central and peripheral zones of the apparatus (defined areas represented approximately 75 and 25% of the total floor area, respectively) as an anxiety-related behavioral assessment.
Accelerating Rotarod. Motor coordination, balance, and motor learning were evaluated with an accelerating Rotarod (Rotor-Rod System; San Diego Instruments). Individual rats were assessed for their ability to maintain balance on a rotating bar that accelerated from 4 to 40 rpm over a 5-min period. The amount of time elapsed before each subject fell from the rod was recorded. Each test subject was given four trials per day for 2 consecutive days with an intertrial interval of 30 min.
Grip Strength. Forelimb grip strength was measured with a digital grip strength meter (Animal Grip Strength System; San Diego Instruments) by holding the rat by the nape of the neck and by the base of the tail. The forelimbs were placed on the tension bar, and the rat was pulled back gently until it released the bar. Each animal was assessed three times, and mean grip strength (measured in kilograms of resistance) ± S.E.M. was calculated.
Water-Maze. Water-maze experiments were conducted as described in detail previously (Terry et al., 2006). In brief, the hidden platform test, rats were given two trials per day for 6 consecutive days to locate and climb on to the hidden platform. Probe trials were conducted 24 h after the last hidden platform trial to measure spatial bias for the previous platform location. Visible platform tests were subsequently conducted after probe trials (as a gross estimate of visual acuity) using a highly visible (white) cover fitted with a small white flag attached to the platform.
Spontaneous Novel OR Test. OR tests were conducted as described in detail previously (Terry et al., 2007). In brief, habituation to the test apparatus consisted of two daily 10-min sessions in which the animals were allowed to freely explore the open-field box. Video-recorded OR testing began on the 3rd day and ended on day 5. Each test day began with a 3-min information session (i.e., the A/A session with identical objects) followed by a 1.0-, 15.0-, or 60.0-min delay period (administered in a pseudorandom order) and a subsequent 3-min dissimilar stimuli (A/B) session. The objects discriminated were made of glass, ceramic, clay, or plastic. The proportion of the total exploration time that the animal spent investigating the novel object was the index of recognition memory. A recognition index calculated for each animal was expressed as the ratio TB/(TA + TB), where TA indicates the time spent exploring object A (familiar object) and TB indicates the time spent exploring object B (novel object).
PPI. To assess the effects of CPF exposure on sensorimotor gating, a PPI procedure was conducted as described in detail previously (Hohnadel et al., 2007). In brief, four startle chambers (San Diego Instruments) were used, the background white noise was set at 70 dB, and the PPI trials consisted of a prepulse (20-ms burst of white noise with intensities of 75, 80, or 85 dB) followed 100 ms later by a startle stimulus (120 dB, 20 ms of white noise). The PPI was calculated according to the formula: [100 – (startle amplitude on prepulse-pulse trials/startle amplitude on pulse alone trials) × 100]. The mean level of PPI (i.e., averaged across the three prepulse intensities) was also analyzed.
Blood Collection for Plasma Assays
Blood sampling occurred weekly throughout the CPF treatment regimen and 2-week washout. Rats were anesthetized by i.p. injection (1 ml/kg b.wt.) of a cocktail containing ketamine (40 mg/ml) and xylazine (8 mg/ml). Blood was collected from the jugular vein using a 1.0-ml syringe fitted with a 25-gauge needle; 0.7 ml of blood was immediately divided into two separate Microtainer tubes as follows. Blood (300 μl) was added to an EDTA Microtainer Tube (catalog no. 365973; BD Biosciences, Franklin Lakes, NJ) that contained 25 μlof trichloroacetic acid (22.5 mg) in ultrapure water. The tube was vortexed, snap-frozen in liquid nitrogen, and stored at –70°C until analyzed for CPF and its metabolites. The remaining blood (400 μl) was added to a Microtainer plasma separator tube containing lithium heparin (catalog no. 365958; BD Biosciences). This tube was inverted eight times and then centrifuged according to the BD Biosciences protocol. The resulting plasma was aliquoted into 0.5-ml tubes, snap-frozen in liquid nitrogen, and stored at –70°C until analyzed for cholinesterase activity.
At the end of the washout period, rats used in the blood sampling study were anesthetized with isoflurane; brains were harvested and snap-frozen in dry ice-chilled isopentane and then were cut in half (sagittally) before storage at –70°C. One-half of each brain was used in enzyme activity assays (choline acetyltransferase and cholinesterase), and the other half of the brain was analyzed for levels of CPF and its metabolites.
Plasma and Brain CPF and TCP Levels
CPF and TCP were measured in samples of brain tissue using coupled-column liquid chromatography/electrospray ionization tandem mass spectrometry as we described previously (Williamson et al., 2006). Acidified blood samples (previously collected in EDTA Microtainer tubes) were extracted by liquid-liquid extraction. Ethyl acetate-acetonitrile (200 μl, 60:40, v/v) was added to 50 μl of acidified rat blood. The samples were then vortexed for 5 min and centrifuged at 16,000g for 10 min. The upper organic layer was dried under a gentle stream of nitrogen, and the resulting residue was reconstituted in acetonitrile. The samples were briefly sonicated, centrifuged for 10 min at 16,000g, and then the supernatant (10 μl) was analyzed by liquid chromatography (LC)/electrospray ionization tandem mass spectrometry. The sample extraction and LC method used for blood analysis was different from the LC method used for brain (Williamson et al., 2006). A C8 column (2.0 × 150 mm, 5 μm; Agilent, Palo Alto, CA) equipped with a 4.0 × 2.0 mm Security Guard C8 guard column (Phenomenex, Torrance, CA) was used. The compounds were separated by gradient elution using 0.0025% formic acid in acetonitrile (mobile phase A) and 0.0025% formic acid deionized water (mobile phase B). The flow rate was 0.3 ml/min, and the column temperature was 25°C. Mobile phase A was linearly increased from 60 to 80% over 2 min and held at 80% for 3 min. The column was returned to the initial mobile phase conditions and re-equilibrated for 4 min. The total run time for each injection was 10 min. TCP eluted first at 3.5 min, followed by CPF-oxon and CPF at 4.2 and 8.0 min, respectively. The column effluent was analyzed by electrospray ionization tandem mass spectrometry (Williamson et al., 2007).
Homogenization of Brain for Enzyme Assays
The dissection protocol described by Gearhart et al. (2007) was used to isolate six brain regions—basal forebrain, hippocampus (anterior and posterior), striatum, prefrontal cortex, and cortex—from frozen half brains (collected above). Brain regions were manually homogenized using a polypropylene pestle in a 1.5-ml microcentrifuge tube (Scienceware catalog no. 19923-0000; Bel-Art Products, Pequannock, NJ). Homogenization buffer contained 0.25 M sucrose, 10 mM EDTA, 0.5% Triton X-100 (v/v), and 0.01 M sodium phosphate (pH 7.4). After manual homogenization in 5 volumes (5 μl of buffer/mg of tissue) of ice-cold homogenization buffer, the crude homogenate was briefly sonicated (on ice) using a Sonic Dismemberator (model 100, set at level 1; Fisher Scientific Co., Waltham, MA). Sonicated homogenates were aliquoted into 0.5-ml tubes (20 μl/tube), and stored at –20°C. Within 2 weeks of freezing, the homogenates from all six brain regions were analyzed for total protein (Coomassie Plus Assay, catalog no. 23236; Pierce Biotechnology, Rockford, IL), cholinesterase activity, and choline acetyltransferase activity.
Plasma and Brain Cholinesterase Activities
Cholinesterase activity in plasma samples and brain homogenates was measured according to Ellman et al. (1961) in a 96-well plate format at room temperature. Five microliters of plasma (100–130 μg of protein/μl) or brain homogenate (20–50 μg of protein/μl) was dispensed into the bottom of the wells of the 96-well plate (catalog no. 12-565-501; Fisher Scientific Co.). An 8- or 12-channel pipetter was used to quickly add 310 μl of reaction mixture to the wells. The reaction mixture contained acetylthiocholine (0.48 mM, D-8130; Sigma-Aldrich, St. Louis, MO) and dithiobisnitrobenzoic acid (0.52 mM, 102710050; Acros Organics, Geel, Belgium) in 0.1 M sodium phosphate buffer (pH 8.0). The microplate was shaken for ∼30 s using a Jitterbug plate shaker (Boekel Scientific, Feasterville, PA), before it was placed in a μQuant Microplate Spectrophotometer (BioTek Instruments Inc., Winooski, VT). The formation of reaction product (yellow color) was monitored by measuring absorbance at 412 nm every 2 min for 16 min. The cholinesterase-mediated reaction rate (moles per liter per minute) was calculated by dividing the change in absorbance per minute by 13,600 (for details, see Ellman et al., 1961). Each plasma sample or brain homogenate was assayed in triplicate.
Choline Acetyltransferase Activity
Brain homogenates (prepared as above) were analyzed for choline acetyltransferase activity using a modification of the method described by Fonnum (1969). Typically a set of 24 to 30 samples were processed at once, e.g., cortex homogenates from the four treatment groups (n = 5–6 samples per group). Each assay (20 μl of total volume) was prepared in a 0.5-ml tube. For each set of assays, brain homogenate (8 μl; 20–50 μg of protein/μl) was added to every tube, and then 12 μl of “reagent master mix” was quickly added to start the reaction. The reagent master mix contained (per assay tube) 0.2 μCi of tritium-labeled acetyl coenzyme A (NET290; PerkinElmer Life and Analytical Sciences, Boston, MA), 0.4 mM nontritiated acetyl coenzyme A, 10 mM choline chloride, 0.2 mM eserine, 10 mM EDTA, 0.3 mM sodium chloride, and 0.5% Triton X-100 v/v in 50 mM sodium phosphate, pH 7.4. Tubes were maintained at 37°C for 30 min. The reaction was quenched by placing the assay tubes in a prechilled (–20°C) microcentrifuge rack and adding 100 μl of ice-cold ultrapure water to each tube. Liquid-liquid extraction was used to isolate the acetylcholine formed during the reaction: 300 μl of an organic solution containing 20 mg/ml sodium tetraphenylborate in 3-heptanone was added to the quenched reaction. The mixture was vortexed for 30 s and centrifuged for 5 min at 7000g, and 200 μl of the upper organic layer was carefully pipetted out of the assay tube and then transferred to a 20-ml glass scintillation vial that contained 10 ml of Scintiverse BD (Fisher Scientific Co.) liquid scintillation cocktail. A Beckman LS6000TA (Beckman Coulter, Fullerton, CA) was used for liquid scintillation counting.
ELISA Methods for Cholinergic Proteins
At the end of the 14-day washout, relative levels of six specific brain proteins (p75NTR, TrkA, phosphorylated TrkA, VAChT, CHT, and α7-nAChR) were measured in tissue lysates from four brain regions (basal forebrain, hippocampus, prefrontal cortex, and cortex) prepared from vehicle- and CPF-treated rats. We previously published detailed methods for the dissection of brain regions, preparation of brain lysates, and ELISAs (Gearhart et al., 2006).
The effects of the CPF treatment regimen on fast anterograde and retrograde axonal transport were evaluated in single axons of sciatic nerves (ex vivo) by direct visualization of vesicle movement using video-enhanced differential interference contrast microscopy. This procedure has been described in detail previously (Terry et al., 2003). In brief, rats were anesthetized with 4% chloral hydrate (10 ml/kg i.p.), the midthigh sciatic nerve was exposed, and 6-0 silk ligatures were tied at the proximal and distal ends. Exceptional care was taken to prevent stretching and trauma to the nerve during excision. The proximal-to-distal orientation of the nerve was maintained throughout the experiment. The sample was placed between two coverslips in a custom-designed aluminum chamber that was sealed in place with 1:1:1 Vaseline petroleum jelly-lanolin-paraffin. Before the chamber was sealed, the nerve was extended to its original length, and the ligatures were attached to the bottom of the chamber using 1:1:1 Vaseline petroleum jelly-lanolin-paraffin. All procedures were accomplished with the nerve continually submersed in oxygenated physiologic buffer (94 mM NaCl, 5 mM KCl, 1.5 mM CaCl2, 1.0 mM MgSO4, 2.0 mM Na2HPO4, 24 mM NaHCO3, and 11 mM glucose, pH 7.4). Axons were viewed through a Zeiss Axiovert 10 microscope with differential interference contrast optics (Lehman Scientific, Red Lion, PA) equipped with a Hamamatsu C2400-07 camera (Hamamatsu Corporation, Bridgewater, NJ), an Argus-20 image processor (Spectra Services, Inc., Webster, NY), and a Hamamatsu high-resolution monitor) in an observation chamber on a 37°C heated stage (Zeiss TRZ model 3700). Video enhancement of the axons was achieved with analog contrast enhancement (camera controller) and digital contrast enhancement (video computer) with background subtraction. The number of vesicles moving in the anterograde and retrograde directions that completely traversed a 2-cm square window (drawn directly on the screen of the video monitor) was counted for a 10-min interval.
Comparisons between treatment groups were made using analysis of variance (with repeated measures when indicated) followed by the Student-Newman-Keuls method for post hoc analysis. Statistical significance was assessed at an α level of 0.05. In all studies, the investigator performing the experiments was blind to the treatment group.
Body Weight and Observational Studies
The effect of repeated (i.e., every other day) exposures to CPF on body weight over a 30-day period are illustrated in Fig. 1. As indicated, the rats in all treatment groups progressively gained weight over the 30-day period [day effect, F(11,33) = 184.2, p < 0.001]. There were no significant differences between any of the CPF-treated groups and vehicle controls, i.e., the dose effect and the dose × day interaction were not significant (p > 0.05). There were no observations of acute cholinergic side effects associated with any of these CPF doses at any point in the study.
Assessments of Exploratory Activity and Motor Function
In these experiments, we were interested in determining whether CPF exposure had significant effects on exploratory activity or motor function (i.e., effects that might have influenced performance in the memory-related tests, particularly the water-maze experiments). The results of these experiments are provided in Fig. 2, A to C.
Open-Field Activity.Figure 2A illustrates the effects of CPF on open-field locomotor activity at various points during exposure and during the drug-free washout period. Horizontal and vertical locomotor activity and stereotypical movements are depicted. There were no significant treatment-related effects (p > 0.05 for all differences) on any of these measures. The time spent in the peripheral versus central zones of the test apparatus was also assessed. Again, no significant CPF-related effects were observed (data not shown).
Grip Strength. The effects of CPF exposure on forelimb grip strength at various points during exposure and during the drug-free washout period are illustrated in Fig. 2B. There was a statistically significant trial effect (p < 0.05), indicating that grip strength increased in all groups over the course of repeated testing; however, no significant treatment-related differences (p > 0.05 for all comparisons) were observed.
Rotarod Performance. The effects of CPF exposure on the performance of the Rotarod task at various points during exposure and during the drug-free washout period are illustrated in Fig. 2C. As in the case of grip strength testing, there was a statistically significant trial effect (p < 0.001), indicating that Rotarod performance improved in all groups over the course of repeated testing. There was no significant overall dose effect, but there was a nearly significant CPF dose × trial interaction (p < 0.07). Again, no significant drug effects were observed in post hoc analysis, although there were some cases in which trends toward inferior Rotarod performance in CPF-treated rats were evident, particularly later in the study during the drug-free washout period.
Hidden Platform Test. The latencies required to locate a hidden platform in the water maze beginning on day 7 of a drug-free washout (i.e., after the 30-day regimen of CPF exposure) are illustrated in Fig. 3A. Statistical comparisons of latencies across the four groups revealed the following results: dose effect, F(3,55) = 2.9, p < 0.05; day effect, F(5,15) = 46.8, p < 0.001; and dose × day interaction, F(274,352) = 1.2, p = 0.3. Similar results were evident when swim distances were analyzed. Thus, after exposure to vehicle or CPF, the rats learned to locate the hidden platform with progressively shorter latencies (and swim distances) across the 6 days of training. Post hoc analyses indicated that the 18.0 mg/kg dose of CPF was associated with modest, but significant (p < 0.05), impairment in performance of the task (i.e., indicated by higher mean latencies to locate the hidden platform) on days 1 and 4 of testing.
Swim Speeds. Swim speeds (Fig. 3B), i.e., the distance swam (centimeters) divided by the latency to find the platform (seconds), were compared daily across the treatment groups for all 6 days of water-maze testing. Statistical comparisons revealed the following results: dose effect, F(3,55) = 0.5, p = 0.7; day effect, F(5,15) = 8.8, p < 0.001; and dose × day interaction, F(274,352) = 2.6, p < 0.001. Interestingly, post hoc analyses indicated that the animals previously exposed to CPF at 10.0 mg/kg swam significantly (p < 0.05) faster than did controls on day 6 of testing.
Probe Trials.Figure 3C illustrates the performance of probe trials by the various treatment groups conducted on day 7 of water-maze testing. There was a clear trend toward inferior performance in all of the CPF groups (as indicated by a reduced number of crossings over the previous 10 cm × 10 cm target area); however, the effects did not reach the required level of statistical significance (main effect, p = 0.1).
Visible Platform Test. The average times required to reach a highly visible (reflective) platform (data not shown) ranged between 8.4 and 13.6 s across all groups in the study and were not significantly different (i.e., all p values were > 0.05). This finding indicated that differences in performance of the previous hidden platform tests or probe trials were unlikely to be a result of gross impairments in visual acuity associated with CPF.
Spontaneous Novel OR Test
Figure 4 illustrates the effects of CPF exposure on performance in the OR task (only the A/B sessions are illustrated). There was a clear preference noted for all treatment groups for the novel object at each delay interval in the A/B sessions (p < 0.01). All other factors including dose, delay, and the interactions between dose, delay, and trial type were not statistically significant.
The effects of prior exposure to CPF on PPI testing assessed on day 12 of the drug-free washout are presented in Fig. 5, A to C. As indicated in Fig. 5A, there were significant dose-related differences [F(3,51) = 3.9, p =<0.02]; there was also a highly significant difference in response to the different prepulse levels [F(2,6) = 66.3, p = <0.001], but the dose × prepulse level interaction was not significant (p > 0.05). Post hoc analyses indicated that prior exposure to CPF at 10.0 and 18.0 mg/kg significantly (p < 0.05) diminished PPI at the 80-dB prepulse level (compared with vehicle controls) and that CPF at 18.0 mg/kg significantly (p < 0.05) diminished PPI at the 85-dB prepulse level. The PPI-impairing effects of the 10.0 and 18.0 mg/kg doses of CPF were also apparent when the data were averaged across the prepulse levels (Fig. 5C). There were no significant effects of any of the doses of CPF on startle amplitude (Fig. 5B).
Alternate day administration of CPF over 30 days produced a concentration-dependent increase in the plasma levels of CPF (Fig. 6A). Maximal levels were attained by experimental day 21. After the final injection on experimental day 30, plasma levels dropped precipitously such that minimally detectable levels were present after the 2-week washout. For the nontoxic metabolite TCP, plasma levels rose more rapidly than those for the parent compound, peaking on experimental day 7 (Fig. 6B). Despite the chronic regimen, TCP levels gradually decreased through experimental day 30, becoming minimally detectable during the discontinuation period. Whole brain levels of CPF and TCP were only measured on experimental day 44 because the primary objective was to determine whether significant amounts of CPF or TCP remained after washout. The data are presented in Table 2. CPF was detectable in the brains of all of the animals treated with the 10.0 and 18.0 mg/kg doses, whereas TCP was only present in 1 of 12 animals.
Alternate day administration of CPF produced a concentration-dependent decrease in plasma cholinesterase activity, which was maximal at experimental day 7 (Fig. 7A). Enzyme activity remained at approximately 10% of control during the 18 mg/kg CPF regimen until experimental day 30. During the 2-week washout period, cholinesterase activity began to increase; however, 14 days after the final 18 mg/kg dose, plasma cholinesterase activity was still decreased to 44% of control. Figure 7B shows the data for CPF (18 mg/kg) and TCP levels and for cholinesterase activity over time normalized with respect to the respective percent maximal level in each parameter, respectively. Thus, cholinesterase activity was maximally inhibited at a time when CPF levels were at approximately 50% of maximal attained levels. During washout, cholinesterase activity continued to be inhibited when CPF and TCP levels were only minimally detectable.
Cholinesterase activity was measured in six brain regions 14 days after the final administration of CPF (Fig. 8). For five of the six brain regions cholinesterase activity was significantly decreased for 14 days after the 10 and 18 mg/kg regimens [F(3,20) = 5.0–24.0, p < 0.01]. For the striatum, cholinesterase inhibition was significant 14 days after all three dose regimens were completed [F(3,20) = 10.0, p < 0.001]. Across the six brain regions, 14 days after the final 18 mg/kg dose of CPF, cholinesterase activity was still inhibited by 55%. After discontinuation of the 18 mg/kg CPF regimen, the brain/plasma ratio for cholinesterase activity varied from a low of 0.67 in the striatum to a high of 1.04 in the anterior hippocampus. Across all three dose regimens and across all six brain regions the ratio averaged 0.82 ± 0.033.
Cholinergic Marker Proteins
Six proteins that are important for cholinergic function were measured by ELISA in four brain regions derived from tissues harvested on experimental day 44 of the vehicle or 18 mg/kg CPF treatment regimen (Fig. 9). In the prefrontal cortex there were significant decreases in TrkA, phosphorylated TrkA, and α7 nAChRs. In the remaining cortical tissues, phosphorylated TrkA was the only marker protein to show a significant decrease relative to control. In the hippocampus, the high-affinity choline transporter and the vesicular acetylcholine transporter were decreased as was the expression of α7 nAChRs. The basal forebrain was the only region to show a decrease in the expression of p75 neurotropin receptor. This was accompanied by a decrease in the expression of the high-affinity choline transporter and the α7 nAChR. Thus, the most consistent finding was the decrease in the expression of α7 nAChRs.
Choline acetyltransferase (ChAT) activities were measured in the same brain regions as were the cholinergic protein markers. ChAT activity was measured 14 days after the last administration of 18 mg/kg CPF. None of the brain regions showed significant changes relative to control (p > 0.21; data not shown).
Anterograde and retrograde axonal transports were assessed ex vivo in sciatic nerve fibers obtained from rats on experimental days 30 and 43 of the 18 mg/kg CPF regimen (Fig. 10, A and B). In vehicle-treated rats, the number of vesicles moving in the anterograde direction was about twice that of those moving in the retrograde direction. On experimental day 30 (the day after the last CPF injection) anterograde and retrograde transports were significantly decreased by approximately 20% [anterograde, F(3,16) = 33.3, p < 0.0001; and retrograde, F(3,16 = 10.9, p < 0.001)]. Two weeks after the regimen was discontinued anterograde transport was still depressed by 20%, and retrograde transport was depressed by approximately 30%. Fast anterograde and retrograde axonal transports were also assessed ex vivo in sciatic nerves obtained from rats after a single injection of 18 mg/kg CPF (Fig. 10, C and D). In general, the number of vesicles moving in the anterograde direction was about twice that of those moving in the retrograde direction. In both cases, treatment with CPF resulted in a significant time-dependent decrease in the number of vesicles moving of up to approximately 20% in the anterograde direction [F(4,13) = 25.9, p < 0.0001] and up to approximately 33% for the retrograde direction [F(4,13) = 22.8, p < 0.0001]. The time point after injection for maximal inhibition of transport was 24 h. There was some recovery in axonal transport by 48 h, but it was still significantly decreased relative to control means.
In this study, there was no evidence of overt cholinergic side effects or CPF-related effects on weight gain, confirming our premise that the doses evaluated were “subthreshold” for acute toxicity. Furthermore, CPF treatment had no effect in the open field, Rotarod, or grip strength analyses or on swim speeds or visible platform tests in the water maze. These observations argue that deficits in memory-related task performance were not due to residual OP effects on locomotor activity, anxiety levels, or visual acuity. The highest dose of CPF evaluated (18.0 mg/kg) was associated with a 90% inhibition of cholinesterase by day 6 of treatment, a level of inhibition that would normally be expected to produce overt cholinergic symptoms. The lack of symptoms is suggestive of some type of adaptive mechanism or tolerance that protected the animals against the more severe toxic effects of CPF after repeated exposure.
The water-maze procedure was used because it is a visuospatial learning task that is sensitive to cholinergic dysfunction (McNamara and Skelton, 1993), and, importantly, deficits in visuospatial processing have been identified as some of the negative outcomes in patients previously exposed to OPs for chronic periods (Roldán-Tapia et al., 2005). In the hidden platform task, rats exposed to the 18.0 mg/kg dose of CPF were impaired on days 1 and 4; however, all treatment groups reached an asymptotic level of performance by day 5, indicating a modest impairment of task acquisition that could be overcome with repeated testing. Furthermore, there was some (albeit not statistically significant) evidence of CPF-related impairment of task retention in subsequent probe trials. In our earlier study (Terry et al., 2003), daily treatment with CPF (dose range of 2.5–25.0 mg/kg) for 2 weeks resulted in a dose-dependent decrement in the water maze that was evident from 1 to 5 days after CPF discontinuation. However, within 2 weeks after discontinuation, performance was similar to that of vehicle-treated controls. The results reported here support the findings of the prior study; however, in the present study, exposure was intermittent, and performance deficits were more prolonged (i.e., evident for up to 2 weeks after CPF discontinuation). The protracted effect of CPF in this study might be expected in view of the longer treatment period (30 days versus 14 days), although the total numbers of CPF injections were similar in the two studies (15 versus 14).
The CPF-related impairments in PPI were interesting in light of reports of sensorimotor abnormalities (Kamel and Hoppin, 2004) and attentional impairments in humans previously exposed to OPs. PPI, defined as the reduction in startle response produced by a low-intensity stimulus presented before a high-intensity, startle-producing stimulus has been widely used as a neurophysiological measure of the early preattentive stages of information processing (Braff and Geyer, 1990), Further, it has been shown in animals to be sensitive to cholinergic manipulations (Hohnadel et al., 2007). PPI impairments are evident in a variety of conditions in which deficits in attention and information processing are observed, such as Asperger's syndrome (McAlonan et al., 2002), Huntington's disease (Swerdlow et al., 1995), and schizophrenia (Braff et al., 2001), leading to the logical question of whether OP exposure might be a risk factor for such illnesses. There have been several reports in which pesticide exposures were linked to a higher incidence of Parkinson's disease (and other movement disorders); however, most of these studies have not implicated specific compounds (for review, see Kamel and Hoppin, 2004).
Given the water maze and PPI results, we were surprised that there were no significant effects of CPF in the novel object recognition task. This test for rodents was selected because it is a nonspatial procedure that models some components of episodic and recognition memory (Ennaceur and Delacour, 1988) and because it is sensitive to cholinergic alterations (Bartolini et al., 1996). The lack of a CPF effect may simply indicate that residual effects of the OP are task (i.e., cognitive domain)-specific. Because we did not evaluate delays longer than 60 min, it is certainly feasible that negative effects of CPF might emerge as the cognitive demands of the procedure increase.
In neurochemical studies, we sought to determine whether the protracted behavioral deficits could be related to alterations in proteins that have important roles in cholinergic function. Protein levels (or the activity) of molecules selective for cholinergic neurons (VAChT, CHT, and ChAT) were measured along with molecules related to but not specific for cholinergic cells (p75NTR, TrkA, phosphorylated TrkA, and α7 nAChRs). Decrements in one or more of these proteins were evident for each of four brain regions analyzed. For example, α7 nAChRs were decreased in basal forebrain, hippocampus, and prefrontal cortex, whereas VAChT and CHT were decreased in the hippocampus. The decreases in α7 nAChRs and the CHT in the basal forebrain were interesting because cholinergic neurons arising from the basal forebrain nucleus basalis innervate sensory regions of the cerebral cortex and play a role in attentional processes (Sarter et al., 2006). Furthermore, the decrease in TrkA and phosphorylated TrkA (i.e., the activated form of the receptor) in the prefrontal cortex was interesting given that nerve growth factor-mediated TrkA activation is required for the survival and maintenance of basal forebrain cholinergic neurons (Li et al., 1995; Auld et al., 2001). The PPI impairments and α7 nAChR decrements in the hippocampus and prefrontal cortex were interesting because α7 nAChR deficits in these same brain regions are thought to contribute to impairments in sensory gating and attention in schizophrenics (for review, see Martin et al., 2004). Finally, the notion that the VAChT and CHT deficits observed in the hippocampus contributed to the water-maze learning impairments is consistent with the well documented role of cholinergic neurons and hippocampal function to place learning (for review, see McNamara and Skelton, 1993).
The decreased protein markers (described above) could reflect neuronal loss, altered protein turnover, or perhaps changes in axonal transport. We reported previously that after rats completed a treatment regimen of 14 daily (25 mg/kg) injections of CPF, bidirectional fast axonal transport measured in peripheral nerves was significantly decreased up to 20 days after discontinuation of CPF (Terry et al., 2003). The results of the present study support these earlier findings and indicate that CPF can decrease axonal transport for up to at least 48 h after a single injection. This decrease was still evident 2 weeks after the intermittent 30-day regimen, coinciding with the decreased cholinergic markers. Although we have no direct evidence that axonal transport was inhibited in brain neurons, it has been reported that the development of delayed neuropathies after OP exposure in susceptible species often results in alterations in axonal transport in peripheral nerves (Gupta et al., 1997; Damodaran et al., 2001). Furthermore, elements of axonal transport such as tubulin and tau proteins are excessively phosphorylated after animals are exposed to OP agents (Gupta and Abou-Donia, 1998; Damodaran et al., 2001). More recently, we demonstrated that the OPs CPF, CPF-oxon, and diisopropylfluorophosphate directly affect the motor protein kinesin, thereby disrupting kinesin-dependent transport on microtubules (Gearhart et al., 2007).
The contribution of residual levels of CPF or its metabolite TCP to the protracted effects on neurotrophin receptors and cholinergic proteins appears minimal because the levels of CPF and TCP were low or nearly undetectable at the time of neurochemical assessments. Conversely, plasma and brain cholinesterase continued to be inhibited by at least 50% 2 weeks after the final CPF injection. Although it is not yet possible to rule out the effects of residual cholinesterase inhibition, it is unlikely that the differential effects on specific neuronal markers in different brain regions would be explained by cholinesterase inhibition. Alternately, axonal transport deficits might be expected to have differential effects in different brain regions, depending on the length of neuronal projections and the specific motor proteins present in those projections.
As noted previously, the CPF dosing regimen was selected to provide a model for the types of exposure that might be experienced by agricultural or industrial workers. From an environmental toxicology perspective, the exposure levels would be considerably higher than those expected to occur in the general population. Because even the lower dose (2.5 mg/kg/day) caused cholinesterase inhibition (and no behavioral effects), this represents the lowest observed effect level of the study. By standard correction for safety factors (10 × 10 × 10), this provides a reference dose (RfD) of 2.5 μg/kg/day, in close agreement with the reported RfD of 3.0 μg/kg/day derived from a no observed effect level of 0.3 mg/kg/day (U.S. Environmental Protection Agency, 1991). On the basis of the TCP urinary data, Barr et al. (2005) estimated that the median exposure of adults to CPF is 0.008 μg/kg/day, i.e., more than 300-fold less than the RfD, 300,000-fold less than the lowest observed effect level in this study, and more than 2 × 106-fold less than the dose level that caused behavioral and biochemical effects reported here. However, occupational exposures to CPF such as that found in pesticide applicators is considerably higher than exposure in the general population. In a study of termiticide applicators in North Carolina, TCP levels were found to exceed that of the general population by an average factor of 70 (Hines and Deddens, 2001). Even more germane to our results is a study of termiticide applicators in Australia where, in the absence of acute cholinergic symptoms, serum cholinesterase activity was found to be only 52% of control levels (Dyer et al., 2001).
In conclusion, the results of this study are consistent with the possibility that low-level intermittent exposures to commercial OP pesticides such as CPF have the potential to induce protracted deficits in information processing and cognitive function. Such deficits could be subsequent to functional changes in brain cholinergic pathways resulting from alterations in bidirectional fast axonal transport.
This work was supported by the National Institute of Environmental Health Sciences (ES012241 to A.V.T.). Salary support (to J.J.B.) also was contributed through a Merit Review Award from the Veterans Administration.
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
ABBREVIATIONS: OP, organophosphate; CPF, chlorpyrifos, O,O-diethyl O-[3,5,6,-trichloro-2-pyridyl] phosphorothionate; TCP, 3,5,6-trichloro-2-pyridinol; nAChR, nicotinic acetylcholine receptor; TrkA, tropomyosin-receptor kinase A; p75NTR, p75 neurotrophin receptor; OR, object recognition; PPI, prepulse inhibition; LC, liquid chromatography; CHT, high-affinity choline transporter; VAChT, vesicular acetylcholine transporter; ELISA, enzyme-linked immunosorbent assay; ChAT, choline acetyltransferase; RfD, reference dose.
- Received May 10, 2007.
- Accepted June 1, 2007.
- The American Society for Pharmacology and Experimental Therapeutics