Coding region paraoxonase polymorphisms dictate accentuated neuronal reactions in chronic, sub-threshold pesticide exposure

The FASEB Journal • FJ Express Full-Length Article

Coding region paraoxonase polymorphisms dictateaccentuated neuronal reactions in chronic,sub-threshold pesticide exposure

R. Orie Browne,* Liat Ben Moyal-Segal,† Dominik Zumsteg,‡ Yaron David,*Ora Kofman,§ Andrea Berger,§ Hermona Soreq,† and Alon Friedman*,1

*Departments of Physiology and Neurosurgery, Soroka University Medical Center, †The Departmentof Biological Chemistry, The Life Sciences Institute, The Hebrew University of Jerusalem, Jerusalem,Israel; ‡Krembil Neuroscience Centre, Toronto Western Hospital, University of Toronto, Toronto,Canada; and §Department of Behavioral Sciences, Zlotowski Center for Neurosciences, Ben-GurionUniversity of the Negev, Beersheva, Israel

ABSTRACT Organophosphate pesticides (OPs), knowninhibitors of acetylcholinesterase (AChE), are usedextensively throughout the world. Recent studies havefocused on the ACHE/PON1 locus as a determinant ofinherited susceptibility to environmental OP exposure.To explore the relationship of the corresponding gene-environment interactions with brain activity, we inte-grated neurophysiologic, neuropsychological, biochem-ical, and genetic methods. Importantly, we found thatsubthreshold OP exposure leads to discernible physio-logical consequences that are significantly influenced byinherited factors. Cortical EEG analyses by LORETArevealed significantly decreased theta activity in thehippocampus, parahippocampal regions, and the cin-gulate cortex, as well as increased beta activity in theprefrontal cortex of exposed individuals—areas knownto play a role in cholinergic-associated cognitive func-tions. Through neuropsychological testing, we identi-fied an appreciable deficit in the visual recall in ex-posed individuals. Other neuropsychological testsrevealed no significant differences between exposedand non-exposed individuals, attesting to the specificityof our findings. Biochemical analyses of blood samplesrevealed increases in paraoxonase and arylesteraseactivities and reduced serum acetylcholinesterase activ-ity in chronically exposed individuals. Notably, specificparaoxonase genotypes were found to be associatedwith these exposure-related changes in blood enzymeactivities and abnormal EEG patterns. Thus, gene-environment interactions involving the ACHE/PON1locus may be causally involved in determining thephysiological response to OP exposure.—Browne, R.O., Ben Moyal-Segal, L., Zumsteg, D., David, Y., Kof-man, O., Berger, A., Soreq, H., Friedman, A. Codingregion paraoxonase polymorphisms dictate accentu-ated neuronal reactions in chronic, sub-threshold pes-ticide exposure. FASEB J. 20, E1103–E1113 (2006)

Key Words: acetylcholinesterase � electroencephalography � LORETA� organophosphates � subthreshold exposure

Organophosphate compounds (ops) are commonlyused as agricultural pesticides and household insecti-cides throughout the world. Despite their common useand effectiveness in eradicating a wide range of harm-ful agricultural pests, their use poses a serious healthhazard for humans. Surprisingly, their toxic propertieshave been known since the 1930s (1), and extensiveresearch has verified that they act as inhibitors of theacetylcholine (ACh) hydrolyzing enzyme acetylcho-linesterase (AChE) (2). Notably, acute poisoning byOPs leads to accumulation of ACh at cholinergic syn-apses in the peripheral and central nervous systems,potentially causing nausea, excessive salivation, incon-tinence, bradycardia, headache, fatigue, seizure, coma,and death (3). Nevertheless, although the short-termeffects of acute OP poisoning are understood to a greatextent, the long-term consequences of acute poisoningand chronic, subthreshold exposure are still not clear.

Humans are capable of mounting a variety of re-sponses to OP exposure. AChE and butyrylcholinester-ase (BChE) in the circulation act as scavengers byirreversibly binding and consequentially inactivatingthe OP anticholinesterases (anti-AChEs) at their activesites. In experimental animals, the accumulation ofACh at cholinergic synapses in the brain and musclefollowing exposure was shown to initiate up-regulationof AChE mRNA, and protein (4). This protectiveresponse can be diminished as a consequence of adeletion in the AChE promoter, which induces consti-tutive overproduction of AChE, most likely exhaustingthe capacity for subsequent secondary overproductionin a reaction to exposure (5, 6). The paraoxonase gene(PON1), which maps close (5.5 Mb) to ACHE on thelong arm of chromosome 7, codes for paraoxonase(PON), a plasma enzyme that can hydrolyze OPs. PONalso possesses arylesterase and lactonase activity (7, 8)

1 Correspondence: Department of Physiology, Faculty forHealth Sciences, Ben-Gurion University, Beer-Sheva 84105,Israel. E-mail: [email protected]

doi: 10.1096/fj.05-5576fje

E11030892-6638/06/0020-1103 © FASEB

and is involved in inhibiting the oxidation of low-density lipoproteins (LDL) (9). PON1 is known toprotect against OP exposure, as affirmed by the sensi-tivity of PON1-null mice to OP intoxication (10). How-ever, the mechanism of this protection has not yet beenfully elucidated.

A variety of polymorphisms in the PON1 gene havebeen characterized (11). The PON1–108C/T promoterpolymorphism affects enzyme levels by modulatingexpression. Another polymorphism affecting expres-sion is the L55M substitution, which results in reducedPON enzyme levels (12). Substrate specificity is influ-enced by the Q192R substitution (13).

Several studies have reported restlessness, forgetful-ness, and other neuropsychiatric symptoms as commoncomplaints in exposed human populations (14–19).However, few significant changes in cognitive functionhave been detected by neuropsychological testing inpopulations exposed to low levels of OPs without anyhistory of acute poisoning (17, 19–23). Several factorsmay account for the apparent differences reported inthese studies. It is not clear whether the concentrationof exposure (which can be difficult to determine), theduration, type of exposure, or instead, individuallyinherited sensitivity is the most significant factor asso-ciated with exposure-related neuropsychological defi-cits. Variations in methods may also account for theapparent disparities in the literature. Furthermore, nospecific phenomenological or mechanistic data existregarding such changes. Despite the differing reports,accumulating data suggest that the function of thebrain may be appreciably affected following exposureto OPs. These changes are often subtle and may beassociated with genetic or environmental factors thathave not always been considered in previous studies.

Inherited, experiential, and environmental factorsare all presumed to influence higher brain functions inhumans. However, quantitative evaluation of each ofthese contributions is very difficult. We, therefore,combined biochemical, genetic, neurophysiologic, andneuropsychological methods in an effort to reveal therelation between environmental exposure and inher-ited sensitivity to OPs in an exposed population. Ourdata suggest the existence of a new underlying mecha-nism that explains the interactions between environ-mental and genetic factors in brain malfunction follow-ing low-concentration OP exposure.

MATERIALS AND METHODS

Participants

The local institutional review board approved the use ofhuman subjects for this study. The study was conducted usingvolunteers from a rural, agricultural community where or-ganophosphate pesticides (e.g., fenitrothion, chlorpyrifos,monocrotophos, ethion, and azinphos-methyl) are routinelyused. Living in the community for an average of 26 yr, theresidents are located 25–150 m from the fields. The exposedpopulation consisted of 291 individuals with the following

demographic profile: 149 males, 142 females, ages rangingfrom 4 to 90 yr (mean 36.5 yr, sd 21.7); 65 (22.3%) areagricultural workers. Twenty-eight individuals (9.6%) wereexcluded from the study because of a history of centralnervous system (CNS)-related diseases (as obtained in medi-cal questionnaire or medical records: meningioma, stroke,migraine, vertigo, anxiety, depression, attention deficit disor-der, and schizophrenia). Sixty individuals gave written con-sent to participate in the study after reading an extensiveconsent form. The study participants had no history of acutepesticide intoxication. All control subjects for this study camefrom urban areas with no history of exposure to anticholin-ergic agents and had no known neurological disease orpsychiatric problems. Control data for the biochemical anal-yses (n�91) were compiled in the same lab and with the sametechniques as those in the present study. Control subjects forthe EEG analysis (n�9) had an age and gender distributionsimilar to the exposed participants (exposed mean age: 46,control mean age: 42, exposed 58% male, 42% female,control 56% male, 44% female). Controls for the neuropsy-chological tests (n�24) were age (�5 yr), sex, and educationmatched.

Airborne spread of pesticides

The airborne spread of pesticides was evaluated using theAgri-Screen Ticket (Neogen, Lansing, MI, USA), whichscreens for the presence of cholinesterase inhibitors using alitmus-type disc impregnanted with cholinesterase enzyme.Inhibition of the enzyme results in a color change on testing.The test is sensitive to low parts per billion to �6 parts permillion. One test disc was placed in the field (positivecontrol), and 5 more were placed at 20-meter intervals fromthe field prior to the time of spraying by a tractor-drawnsprayer. A negative control was placed inside a closed auto-mobile at the same location. Test discs were collected after12 h of exposure and tested for the presence of cholinesteraseinhibitors. All discs except for the negative control testedpositive.

Blood samples

Blood samples were drawn to BD Vacutainer® blood collec-tion tubes (Becton-Dickinson, Franklin Lakes, NJ, USA) withcitrate as an anticoagulant, centrifuged (1300 rcf, 4°C, 15min) in a desktop centrifuge, and plasma was obtained.Whole blood and plasma were maintained at –70°C until use.

Enzyme activity measurements

Plasma paraoxonase activity was determined by adapting thespectrophotometric method (24) to a microtiter plate assay.Since several variations of the assay were published, the assaywas calibrated for plasma dilution and substrate concentra-tion. We found that a 1:10 dilution of plasma and a 1.2 mMparaoxon concentration were optimal, yielding high variabil-ity as reported for paraoxonase activity. Higher substrateconcentrations (up to 6 mM) enabled higher hydrolysis ratesbut yielded lower variability, thus obscuring the populationtrends. Briefly, 10 �l of plasma diluted 1:10 was placed inmicrotiter plate wells (Nunc, Roskilde, Denmark) in tripli-cate; the reaction was initiated by adding 190 �l of thesubstrate, 1.2 mM paraoxon (Sigma, Jerusalem, Israel), in0.26 mM Tris–HCl, pH 8.5, 25 mM CaCl2, and 0.5 M NaCl.Readings at 405 nm were repeated at minimal intervals for 10min. Nonenzymatic breakdown of paraoxon was subtractedfrom the total rate of hydrolysis. Enzyme activity was calcu-lated using the e405 for p-nitrophenol, 17 100 M/cm.

E1104 Vol. 20 August 2006 BROWNE ET AL.The FASEB Journal

Plasma arylesterase activity was measured in 10 �l of 1:40diluted plasma mixed with 190 �l of substrate (3.26 mMphenylacetate in 9 mM Tris–HCl, pH 8 and 0.9 mM CaCl2).Hydrolysis rates were determined at minimal intervals inUV-transparent 96-well plates (Greiner-Bio One GmbH,Frickenhausen, Germany), at 270 nm for 4 min. Enzymeactivity was calculated using the e270 for phenol, 1310 M/cm.

Plasma cholinesterase catalytic activity measurements in-volved adaptation of a spectrophotometric method (25) to amicrotiter plate assay. ATCh (Sigma, 1 mM) or butyrylthio-choline (BTCh, Sigma, 10 mM) hydrolysis rates were mea-sured following 20 min preincubation with 5 � 10�5 Miso-OMPA (tetraisopropylpyrophosphoramide) (Sigma), aspecific BChE inhibitor, or 10�5 M 1,5-bis(4-allyldimethylam-moniumphenyl)pentan-3-one dibromide (BW284C51, Sigma),a specific AChE inhibitor. Readings at 405 nm were repeatedat 2-min intervals for 20 min. Nonenzymatic breakdown ofsubstrate was subtracted from the total rate of hydrolysis.Enzyme activities were calculated using the e405 for 5-thio-2-nitrobenzoate, 13 600 M/cm.

Genotyping

Genomic DNA was prepared from blood cells using theGentra Whole Blood DNA Extraction Kit (Gentra, Minneap-olis, MN, USA). Genotyping involved polymerase chain reac-tion (PCR) amplification of the corresponding gene regions,using Taq polymerase (Sigma) followed by agarose gel elec-trophoresis and Exo-Sap enzymatic purification (USB, Cleve-land, OH, USA) of the PCR product. Standard automatedsequencing utilized the BigDye Terminator cycle sequencingchemistry, ABI 3700 DNA Analyser and Data collection andSequence Analysis software (Applied Biosystems, Foster City,CA, USA). PON1 Q192R and PON1 L55M polymorphismswere detected using the single nucleotide primer extensionmethod (SNaPshot ddNTP Primer Extension kit, AppliedBiosystems). Following PCR amplification and purification,the SNaPshot reaction was performed using primer 5�-GGCA-GAAACTGG CTCTGAAGAC-3� for the PON1 55 and 5�-GAT-CACTATTTTCTTGACCCCTACTTAC-3� for PON1 192. Fol-lowing extension and calf intestine phosphatase treatment(Amersham Biosciences, Freiburg, Germany), the productswere subjected to electrophoresis on a 3700 ABI analyser andthe results analyzed using Genescan software (ABI, Tel Aviv,Israel).

Electroencephalography

EEG was recorded from subjects using a clinical EEG unit(CEEGRAPH IV, Biological Systems Corp., Mundelein, IL,USA) with a sampling rate of 256 Hz (26,27). Twenty-fiveconventional Ag/AgCl surface electrodes were placed accord-ing to the international 10–20 system, with additional elec-trodes for the detection of eye-movements and electrocardio-gram activity. At least 25 suitable EEG segments (free ofmovement and eye artifacts, subjects with eyes closed) of 2 sduration were identified off-line by visual inspection for eachsubject. Power spectra (1–30 Hz, resolution of 0.5 Hz) werecalculated using Fourier transform. The spectral power fordiscrete frequency bands (delta: 1.5–6 Hz, theta: 6.5–8 Hz,alpha 1: 8.5–10 Hz, alpha 2: 10.5–12 Hz, beta 1: 12.5–18 Hz,beta 2:18.5–21 Hz, beta 3: 21.5–30 Hz) was calculated for eachsubject. We also calculated the population averaged power ofeach electrode. Spectral power distribution maps were cre-ated using MATLAB (version 7.0).

Source localization

Low resolution brain electromagnetic tomography (LORETA)was used for the localization of presumptive cortical sources

of EEG activity (28). LORETA is based on the neurophysio-logic assumption that neighboring cortical areas are likely tobe coactivated and permits three-dimensional tomography ofbrain electrical data while requiring only simple constraints(smoothness of the solution) and no predetermined knowl-edge about the putative number of discernible source regions(29–32). LORETA estimates the distribution of absolutecurrent density (numerically, and ultimately visually usingscaled color intensity) for brain electrical activity and displaysit on a dense grid of 2,394 voxels based on the digitizedTalairach Atlas, as provided by the Brain Imaging Centre,Montreal Neurological Institute (33). Calculation ofLORETA is limited to cortical gray matter and hippocampiand has a spatial resolution of �1–2 cm. The average currentsource densities for 23 distinct brain regions were calculatedby averaging the LORETA values for each voxel within ananatomically defined region (34). These cortical regions werecreated by allocating the raw LORETA values of individualvoxels to their corresponding Brodmann areas (BA) orcerebral gyri, based on the coordinates of the digitizedTalairach Atlas, and subsequent grouping of these BAs into23 arbitrary “functional” cortical regions (e.g., left mesialtemporal, right lateral temporal, etc.).

Neuropsychological tests

We used a battery of neuropsychological tests related toattention, memory and executive functions, which includedthe Tower of Hanoi, a Serial Reaction Time (SRT), a Contin-uous Performance Test (CPT) in which digits from 1 to 9were presented at a rate of 1/sec and the subject had to detectwhen the digit 9 followed the digit 1, the Digit Span (from theWechsler Adult Intelligence Scale), a visual reproductionsubtest of the Wechsler Memory Scale, and a long-term verbalmemory from a Hebrew version of the RALT (Rey AuditoryLearning Test). The two first tasks were computerized, andthe others were paper and pencil. Details of the methods canbe found in Lezak (35).

Data analysis and statistics

Student’s t test was used to calculate statistical significance forthe differences among AChE, BChE, arylesterase, and para-oxonase activities, as well as differences in the area under thecurve for each discrete frequency band of the EEG. Statisti-cally significant differences between LORETA values weretested by using a nonparametric t test, on a voxel-by-voxelbasis, with correction for repeated measure (36). We usedthree different thresholds for P values to determine statisticalsignificance (P�0.01, P�0.05, and P�0.1).

RESULTS

Neurophysiologic and neuropsychological assessment

To determine the effect of exposure on brain activity,we performed EEG in 19 exposed and 9 control sub-jects with a similar age and gender distribution. Calcu-lation of the averaged spectral power revealed a gener-ally lower power in the exposed group, which reachedstatistical significant only in the theta band (ANOVA,P�0.03). When individual power was normalized, sig-nificantly lower power was found in the theta band andincreased power in the beta 3 band (P�0.04 and

E1105PARAOXONASE POLYMORPHISMS AND PESTICIDE EXPOSURE

P�0.03, respectively) of exposed, compared with non-exposed groups (Fig. 1A–B). Topographically, the de-crease of theta power was most prominent over thecentral region, whereas the increase of beta 3 powershowed a more frontal distribution (Fig. 1C–D). Thesedifferences in surface spectral distribution are directlyrelated to region-specific changes of their underlyingcerebral sources, as evidenced by LORETA. Statisticalnonparametric mapping of LORETA revealed signifi-cant differences in the localization of activity in allseven frequency bands of the exposed subjects, com-pared with controls. Most importantly, the exposedsubjects exhibited markedly decreased theta activity inlimbic cortical areas of both hemispheres, includingmesial temporal structures, subcallosal areas, and pos-terior cingulate gyri. Cortical areas with significantlyincreased delta and beta3 activity were found in bilat-eral prefrontal areas, and alpha1 and alpha2 bandswere localized to the occipital region (Fig. 1E–F anddata not shown).

To reveal the possible neuropsychological signifi-cance of the neurophysiologic findings, we conductedneuropsychological tests and compared the results for23 of the exposed subjects and 23 controls, which werematched for age, sex, and education, using a t test fordependent measures. The only test that showed signif-icant differences between the exposed and non-ex-posed groups was the delayed memory (20 min) por-tion of the visual reproduction subtest of the Wechsler

Memory Scale (Fig. 2). An ANOVA for the effect of thegroup on the percentile score for immediate anddelayed memory revealed a marginally significant om-nibus effect for the group, F (1, 44) � 3.82, P � 0.056.

Figure 1. EEG analyses in OP-exposed groupindividuals show decreased power in thetheta and increased power in the beta bandsin specific brain regions: A) Average normal-ized power spectrum (1.5–30 Hz) of exposed(n�19) and non-exposed (n�9) groups. B)Area under the curve (AUC) of average nor-malized power by discrete frequency bands(delta: 1.5–6 Hz, theta: 6.5–8 Hz, alpha1:8.5–10 Hz, alpha2: 10.5–12 Hz, beta1:12.5–18 Hz, beta2: 18.5–21 Hz, beta3: 21.5–30Hz) in exposed and non-exposed groups. Inthe exposed group, theta activity was signifi-cantly decreased (P�0.04) and beta3 activitywas significantly increased (P�0.03). C–D)Percentage of increase in a normalized spec-tral power at the theta and beta3 bands isshown as a spectral power distribution map.Note the electrode-specific changes of spec-tral power. E–F) LORETA analyses showingcortical regions with statistically significantdifferences between exposed and non-ex-posed groups. E) Areas with significantly de-creased current intensity (blue, P�0.05) inthe theta band included bilaterally the amyg-dala, the hippocampus, the subcallosal area,the parahippocampal gyrus, and the anteriorand posterior cingulate gyrus (BrodmannAreas 25, 30, 31, 34, 36, 37). F) Areas withsignificantly decreased (blue, P�0.05) cur-rent density in the beta 3 band included thehippocampus and parahippocampal gyrus bi-laterally; increased sources (red, P�0.05) werefound in the inferior, middle, and superiorfrontal gyri (areas 10, 11, 27, 46).

Figure 2. Neuropsychological tests revealed significantly de-creased scores in delay memory: Exposed individuals showeddecreased scores in the immediate and delayed memoryportion of the visual reproduction subtest of the WechslerMemory Scale. Only the delayed (20 min) portion, however,showed a statistically significant difference between thegroups (see text for details).


Planned comparisons between groups for the immedi-ate and delayed tests indicated that the exposed groupwas significantly impaired in the delayed memory task,F (1, 44) � 4.86, P � 0.05, but not for the immediatememory task, F (1,44) � 1.78, n.s. The mean and sd ofthe percentile scores for the exposed group were 25.04 12.6 for delayed recall and 28.56 17.6 for imme-diate recall. The matched control group had percentilescores of 37.43 23.7 and 36.13 20.7 for the delayedand immediate recall, respectively.

To challenge our hypothesis that changes in currentsources in mesial temporal lobe structures is associatedwith poor memory performance we analyzed separatelymean current density in both right and left temporallobe (medial part) for exposed individuals with low andhigh performance. Exposed individuals who performedpoorly on memory tests (13.3 and 16% for immediateand delayed recall, respectively) also showed 10 timeslower beta activity and 50% increase in theta comparedto exposed individuals who showed better performance(32 and 38% for immediate and delayed recall, respec-tively). However due to the small sample size (n�6), thesechanges did not reach statistical significance (see Dis-cussion).

Exposure-related alterations in biochemical activity

Biochemical assays for AChE, BChE, PON, and aryles-terase activity were performed for 30 exposed subjects(see Table 1). Serum AChE activity was found to besignificantly lower in exposed individuals comparedwith controls, compatible with the hypothesis of irre-versible inhibition (36) (41% of control, P�0.001, Stu-dent’s t test). There was no significant difference inBChE activity. Both PON and arylesterase activity weresignificantly higher than predicted in the exposedsubjects (447 and 441% of control, respectively, P �0.001) (Fig 3A). Enzyme activities were not significantlydifferent for agricultural workers (n�9) when com-pared to the local population (n�20) (Table 1). Alter-ations in biochemical activity reflect, besides exposurelevels, also the individual’s reaction to the chemicalstress, as manifested by modified gene expression (37).This, in turn, represents a composite effect of inheri-tance and environmental status (i.e., exposure). There-fore, as a follow-up to the biochemical investigation, wesearched for genetic variation. Allele frequency in ourstudy population was similar to that of the generalIsraeli population for PON1 55, PON1 192, and ACHE-17130 HNF3�. There was a higher frequency of the

TABLE 1. Genetic and biochemical summary of exposed population

Subject AgeAgricultural

workerPON-108

PONL55M

PONQ192R

AChE SpecificActivity (nmol/

min*ml)

BChE- SpecificActivity (nmol/

min*ml)

PON-SpecificActivity (nmol/

min*ml)

Arylesterase- SpecificActivity

(�mol/min*ml)

1 44 TT LL QQ 101.97 2231.17 189.66 112.602 48 yes CT LM QQ 137.22 3983.90 108.89 74.493 51 CT LM QQ 160.87 3763.94 132.19 79.234 53 CT LL RR 168.87 4371.72 517.65 64.865 57 CC LL QQ 171.97 4528.69 150.83 94.036 47 yes CC LM QQ 172.70 4043.53 125.63 97.937 49 TT LL RR 177.18 4915.15 477.95 62.678 31 CC LL QQ 181.89 4342.50 224.60 117.469 26 CC LL QR 187.16 5077.09 466.97 107.75

10 51 TT LL QR 189.58 4410.14 425.09 -11 31 CC MM QQ 190.66 4665.69 105.19 69.1612 46 CC MM QQ 195.21 5815.05 90.00 66.9513 60 CC LL QQ 195.79 5901.26 188.39 101.9314 43 CC LL QR 195.93 4753.25 210.81 128.9815 53 CC LM QR 196.80 4338.76 229.04 58.0316 20 CC LL RR 196.87 4674.92 545.46 59.1817 50 yes CT LL QQ 198.33 5847.56 119.79 81.7018 16 CC LL QR 213.55 4984.36 478.24 96.7419 47 CC LL QR 214.58 6387.74 496.75 76.0420 51 yes CC MM QQ 216.88 4853.65 111.87 64.3521 70 yes CC MM QQ 217.79 4607.45 211.56 -22 52 yes CT LL QQ 221.26 6168.41 187.32 96.5823 42 TT MM QQ 226.53 2109.98 144.96 72.3124 46 CC LM QQ 230.46 3256.73 101.23 80.9825 54 CT LL QQ 232.28 6336.29 197.59 103.5326 40 yes CT LM QQ 238.62 6191.35 157.57 105.3027 60 yes CT MM QQ 252.54 6602.36 165.83 117.5328 62 CC LL QR 274.06 6980.74 452.71 111.2129 28 yes CC LM QR 279.39 5098.75 312.63 -30 53 yes TT LL QQ - - - -

Mean 46 Mean 201.27 4870.42 252.64 88.52SE 6.92924 220.66556 28.40421 4.1075


PON1 -108 C allele in our study than in the generalIsraeli population (0.70 and 0.40, respectively) (Ta-ble 2).

Biochemical analysis by PON1 55, 192, and –108genotypes revealed several significant trends. In gen-eral, AChE activity did not vary by PON1 genotypeexcept for a slight decrease in LLRR compared withMMQQ individuals (180.97�8.30041 and 216.6�9.16877nmole/min�ml, respectively, P�0.04) (Fig. 3B). Incontrast, different PON1 genotypes correlated withsignificant alterations in paraoxonase activity. The pres-ence of the PON1 L55M substitution (TTG to ATG) wasconsistent with a decrease in PON activity, presumablyrelated to reduced PON1 mRNA and protein levels(12). We controlled for the effects on PON activityassociated with the PON1 Q192R substitution by select-ing individuals without the R allele. Interestingly, inthe non-exposed population, MMQQ individuals(n�10) exhibited about one-third of the PON activityof LLQQ individuals (n�14) (MMQQ/LLQQ�0.36,P�0.0001). In contrast, in exposed subjects, MMQQ(n�6) and LLQQ (n�7) individuals displayed almost

equal PON activity (MMQQ/LLQQ�1.3), suggestingthat the M allele is associated with a greater increase inPON activity in exposed individuals. This is supportedby our data that show 8.1 times more PON activity inexposed MM individuals than in controls but only 3.8times more activity in exposed LL individuals than incontrols. PON enzyme levels can be assessed by mea-suring the arylesterase activity (38). Indeed, arylester-ase activity in individuals with the M allele was 6.0 timesgreater in exposed individuals than in non-exposedcontrols, whereas arylesterase activity in the exposedpopulation without the M allele was only 3.5 timesgreater than in non-exposed individuals, consistentwith the hypothesis of a more robust response toexposure in exposed individuals with the M allele.

The PON1 Q192R (CAA to CGA) substitution, whichaffects catalytic efficiency (13), was associated withincreased PON activity. In controls there was a gene-dependent increase in PON activity correlated with thepresence of the R allele (PON-specific activity (nmol/min*ml) QQ�29.6, QR�71.3, RR�99.9, slope�35.18,R2�0.9886) (Fig. 3B). Interestingly, changes in PONactivity in exposed subjects were characterized by amuch more robust change associated with the R allele(PON-specific activity (nmol/min*ml) QQ�150.7,QR�384.0, RR�513.7, slope�181.48, R2�0.9735). Tocontrol for the effects on PON activity contributed bythe PON L55M substitution, we analyzed PON activityin individuals who lacked the M allele. Indeed, The Rallele-associated changes in PON activity remained(Fig. 3B). In addition, we were able to identify animportant difference between exposed and non-ex-posed groups. Non-exposed LLRR individuals (n�13)showed �2 times more PON activity than non-exposedLLQQ individuals (n�10) (LLRR/LLQQ�2.09,P�0.0001), whereas in exposed subjects LLRR individ-uals (n�3) showed almost 3 times more PON activitythan LLQQ individuals (n�7) (LLRR/LLQQ�2.86,P�0.0001), suggesting an exposure-dependent re-sponse associated with the PON1 Q192R substitutionand a potential biosensor effect of the R allele inanti-AChE exposure.

TABLE 2. PON1 and AChE allele frequency in this study andthe general Israeli population

Position, allele This study Israeli populationa

PON1–108C 0.70 0.40T 0.30 0.60162 (55)T (L) 0.68 0.61A (M) 0.32 0.39575 (192)A (Q) 0.77 0.67G (R) 0.23 0.33ACHE-17130 HNF3� 0.017 0.019n 30 157

a Bryk et al., 2005

Figure 3. Biochemical analyses in OP-exposed individuals: A)Shown are the activities of AChE (nmol/min*ml) in exposedsubjects (n�29) compared to the Israeli non-exposed popu-lation (n�91) (P�0.001); BChE (mmol*10/min*ml) (notsignificantly different); PON (nmol/min*ml) and arylester-ase (�mol/min*ml) (P�0.001). B) AChE and PON activitiessegregated by the PON1 55/192 genotype combinations:LLRR (n�3), LLQR (n�6), LLQQ (n�8), LMQR (n�2), LMQQ(n�5), MMQQ (n�6). In all graphs: open bars representcontrol, non-exposed group; filled bars, exposed group.


Neurophysiologic alterations dependon PON genotype

Our biochemical data revealed significant differencesin the individual responses to exposure, associated withthe presence or absence of the M (L55M) and R(Q192R) alleles in the PON1 gene. Thus, we hypothe-sized that the increase in beta3 activity observed in theEEG of exposed individuals may also differ amongexposed individuals with different genetic profiles. Wefound significantly increased beta3 activity in the fron-tal cortical regions (decreased in temporal regions) ofexposed individuals with the R allele compared withcontrols and with exposed individuals without the Rallele. Interestingly, the M allele had no such effect(Fig. 4A). Comparing the LORETA values (represent-ing the current source densities) in these brain regionsrevealed that exposed individuals with the R alleleexhibited significantly increased frontal activity and

decreased temporal activity (Fig. 4B, P�0.03). Further-more, non-exposed and exposed individuals withoutthe R allele were more similar regarding the pattern ofbeta3 source activity and intensity.

DISCUSSION

Our results support the following conclusions regard-ing the results of chronic subthreshold exposure toOPs: 1) The exposure leads to significant changes inthe distribution of sources of brain activity in specificregions (mainly in limbic structures and frontal corti-ces); 2) These changes are associated with deficits invisual retention memory in exposed individuals; 3) Nosignificant alterations were found in the biochemicalactivities of serum AChE and PON; and 4) Both bio-chemical and neurophysiologic changes are dependenton the relevant genetic makeup of an individual.

Exposure to organophosphates leads to changesin brain activity

A previous study by Burchfiel and Duffy identifiedstatistically significant increases in beta activity in indi-viduals exposed to the AChE inhibitor, sarin (39). Ourresults, despite the small sample size, are consistentwith these findings and add the observation of signifi-cant decreased theta activity in exposed individuals.The relevance of these changes is difficult to determinewithout more detailed information about the sources ofthe brain activity. This is especially true since thedifferences in power were not equally distributed be-tween electrodes, which clearly suggest the involvementof specific brain regions (Fig. 1C–D). Using statisticalanalyses of LORETA current density distributions forthe determination of OP-exposure related changes incortical activity, we found markedly decreased thetaactivity in bilateral limbic structures (including theamygdala, the hippocampus, the parahippocampal re-gions, the subcallosal area and the cingulate cortex),and significantly increased beta3 activity in bilateralprefrontal cortical areas in the exposed group. Thesefindings are consistent with research detailing thefunctional roles of these brain regions and the poten-tial neuropsychological effects of OP exposure (seebelow). Notably, all of these regions are known toreceive significant cholinergic input originating fromthe magnocellular basal forebrain cholinergic system(40, 41). This cholinergic system can be subdivided intoa number of distinct cholinergic sites, including thenucleus basalis of Meynert, the medial septal nucleus,and the vertical and horizontal limb nuclei of thediagonal band of Broca. Efferents originating in thebasal forebrain project diffusely throughout the cortexas well as in the hippocampus, amygdala, and olfactorybulb. Presumably, areas of the brain with significantcholinergic input may be particularly susceptible to OPexposure. The hippocampus and parahippocampal re-gion were localized by LORETA as the sites of signifi-

Figure 4. Enhanced beta activity in carriers of the R allele: A)LORETA values (i.e., current density) were calculated foreach brain region in the beta3 frequency band and the %change compared to the control population is given. Robustchanges were found only for exposed individuals with the Rallele and in specific brain regions (Left temporal and Rightfrontal). B) Normalized current density values (arbitraryunits) of beta3 activity in selected brain regions for non-exposed controls (white bars), exposed without the R allele(gray bars, n�14), and exposed with the R allele present(dark gray, n�8). Note that significant differences werefound only for the exposed individuals carrying the R allele.


cant differences in theta activity between exposed andcontrol subjects. The theta frequency is extensivelystudied in the hippocampus and has been recognizedin other areas of the limbic region as well (42, 43).Numerous reports have elucidated the role of cholin-ergic input in this region from the medial septum anddiagonal band of Broca in generating the theta rhythm(42, 44–46), which is thought to be important inmnemonic function (43, 46, 47). Thus, the reducedtheta rhythms in mesial temporal regions of chronicallyexposed individuals suggest disturbed cholinergic func-tions at rest. The observed changes in the beta activityof the prefrontal cortex may be associated with in-creased demands on attention and other cognitivefunctions following OP exposure. Several studies de-scribed the role of the cholinergic system in attentionalprocessing (48–55) and reported increased neuronalactivity and ACh efflux in the prefrontal cortex of ratsperforming tasks that increase demands on attention(56–58).

Neuropsychological tests

Of the neuropsychological tests performed in thisstudy, the only one that showed a significant differencebetween the exposed and non-exposed groups was thedelayed memory portion of the visual reproductiontest. This outcome is consistent with the results ofneuropsychological testing in other studies, where few,if any, detectable deficits in cognitive function wereidentified in individuals exposed to low levels of or-ganophosphates (17, 19–23). Interestingly, some cor-relation was found between decreased hippocampaltheta and increased beta activity and poor performancein memory recall performance. While these strengthenthe potential of source localization of resting EEGactivity, future studies with increased sample size areawaited to confirm these results. No correlation wasfound between scores on the visual recall test andgenotype or enzyme activities. The small sample sizeand the relative infrequency of specific genotypes inthe present study make clear discrimination difficult.Future studies should seek to confirm genotypic varia-tions in the OP susceptibility alluded to in this study.

Alterations in biochemical activity

On OP exposure, PON acts in a protective role byhydrolyzing OPs, thus preventing inhibition of AChE.In addition, PON exhibits peroxidase activity that pro-tects AChE from oxidative damage (59–61). A reduc-tion in PON levels or a functional defect is expected toleave AChE particularly susceptible to anti-AChE expo-sure. In the brains of experimental animals, it wasshown that on AChE inhibition, there is a subsequentincrease in ACh available to ACh receptors, thus acti-vating a feedback response that results in overproduc-tion of AChE while suppressing choline acetyltrans-ferase (ChAT) and the vesicular ACh transporter (4).This feedback response may bring cholinergic activity

into balance for the short term but may also beassociated with cognitive deterioration in the long termas animal experiments suggest (62). Here we per-formed biochemical assays for AChE, BChE, PON, andarylesterase serum activity in 30 exposed subjects. Ourresults suggest a lasting reduction in serum AChEactivity (41% of control) in exposed individuals, with aconcomitant increase in PON (447%) and arylesterase(441%) activity.

Previous studies reported an age-dependent increasein AChE activity (63, 64). Given that the average age ofthe exposed subjects in the present study was higherthan our control population (46�0.95 vs. 34�2.28),this model would predict increased AChE activity in theexposed population. The fact that this was not the casesupports the notion of persistent inhibition. The de-crease of AChE activity and the increase of PON activityin exposed individuals could potentially reflect a dose-dependent effect of exposure. In such a case we wouldexpect a direct association between the extent of reduc-tion in AChE activity and an increase in PON activity inexposed subjects. But this hypothesis was not supportedby the data. An alternative hypothesis was that theincreased PON activity is protective and results inincreased AChE activity. This, however, was not sup-ported either. Rather, our results support the conclu-sion that enzyme levels are a result of a complexinteraction between inherited (genotype) and environ-mental (exposure) factors and that specific genotypesare associated with exposure-dependent changes in thebiochemical activity of PON in the serum. Interestingly,this appears to be the first reported incidence of PON1up-regulation in response to OP exposure, and it isconsistent with a previous report from Hernandez etal., showing higher levels of PON activity in exposedthan in control individuals during the period of lowerexposure (65).

Genetic influence on biochemistryand neurophysiology

The increase in PON activity is most likely the result ofmultiple processes possibly acting concomitantly. Theobserved increase in PON activity was associated withincreased arylesterase activity [arylesterase activity maybe used as a measure of protein levels (37)], suggestingthat higher PON1 protein levels contribute to in-creased PON activity. We have shown that the PON1L55M substitution, which tends to result in reducedmRNA and protein levels (12), is associated with amuch more robust up-regulated response in exposedindividuals, suggesting an autoregulated mode of pro-duction. Another factor related to increased PONactivity is the PON1 Q192R substitution, which signifi-cantly influences enzyme activity (Fig. 2B). This iscompatible with the hypothesis that R PON1 possessesan insecticide biosensor activity that leads to subse-quent increases in enzyme activity on exposure in amanner unrelated to enzyme levels (arylesterase activityin individuals with the R allele did not correlate with


increases in PON activity). Taken together, these find-ings suggest OP exposure-related changes in PON1transcriptional and post-transcriptional processes thatresult in alterations of both PON1 protein levels andenzymatic functions. It is presently not known whetherthis effect is primarily peripheral or whether it occurs inother tissue types as well (e.g., brain). Our EEG data,showing increased beta3 in exposed individuals anduncharacteristically increased activity in the frontalregions of the cerebral cortex in individual carriers ofthe R allele, suggest that this effect may not be limitedto the periphery. This notion is supported by theobservation of the possible role of PON1 polymor-phisms in the white matter lesions of the brain (66).Indeed, although PON is expressed in brain tissue (67),future research studies on animals and humans areneeded to explore the role of altered PON expressionon brain structure and function.

The exposure-related decrease in AChE activity mostlikely reflects continued exposure to irreversibly inhib-iting OPs in the exposed population. This finding issupported by a recent study showing decreased AChEactivity in OP-exposed Parkinsonian (PD) patients com-pared with non-exposed PD patients (37). Notably, thephysiological feedback response to anti-AChE exposurewas described in the CNS and cannot be generalized tothe periphery. Possibly, chronic exposure also leads toan “exhaustion” of the feedback pathway or that feed-back is “limited” in the case of chronic exposure.

In addition to the genetic influence on biochemistry,we found that markedly altered neurophysiologic func-tion was correlated with paraoxonase promoter poly-morphisms. The increased beta3 activity in frontalcortical regions of exposed individuals with the R allele,compared with controls and to exposed individualswithout the R allele, suggests that the R allele mayconfer less of a “protective” effect on its carriers, thusallowing greater changes in brain activity following OPexposure. Another finding that links the presence ofthe R allele to changes in brain function is the obser-vation that non-exposed individuals without the R allelewere more similar in the intensity of their beta3 activityto exposed noncarriers of the R allele than to exposedcarriers.

The observations that the presence of the R allele isassociated with both up-regulation of PON and alter-ations in brain activity following OP exposure may seemto contradict one another. Increased PON activityshould be a protective factor enabling the carrier tohydrolyze OPs more effectively. Although this may betrue in the short term, it is possible that this robustresponse either reflects or even plays a role in delete-rious effects over the long term. Such a mechanism wassuggested for the prolonged feedback response associ-ated with AChE, as transgenic mice over-expressingAChE show cognitive deterioration over time (62).PON1 may also be involved in this process, since the twogenes are so closely linked. We also mentioned thatPON is expressed in brain tissue (especially bloodvessels) and its over-expression may have a negative

influence on brain structure and function. Therefore,the increased PON1 activity that appears to be apotentially useful response to OP exposure may indeedbe associated with neurophysiologic deficits followingchronic exposure.

A putative model of chronic, subthresholdorganophosphate exposure

On exposure to OPs, PON protects AChE by enzymat-ically degrading the poison and reducing the threat ofoxidative damage (9, 61). OPs are stoichiometricallyscavenged in the blood by BChE (67). The surplus OPsthat remain active after these physiologically protectiveresponses inhibit AChE and cause an increase of AChavailable to muscarinic and nicotinic receptors. Thisrelative overabundance of ACh activates a feedbackresponse that overproduces AChE while suppressingcholine acetyltransferase (ChAT) and the vesicularACh transporter, in an attempt to balance cholinergicfunction in the short term (4). In the case of chronicexposure, this up-regulation fails to adequately meetthe demands of excessive cholinergic stimulation, lead-ing to up-regulation of PON and to consequent alter-ations in cholinergic neurotransmission, neuronal ac-tivity, and brain function. The entire biochemical andneurophysiological response to exposure depends onthe genetic profile of the individual, (i.e., on PON1polymorphism). Thus, although PON over-expressionmay protect against acute functional deficits associatedwith poisoning, it may lead to long-term alterations inhigher brain functions.

The authors are grateful to Ms. Carrie Zaga and Boris Brykfor assistance with experiments. This study was supported bythe European Union (LSHM-computed tomography-2002–503330), the German-Israeli-Foundation (Grant 673) and theRoestrees Foundation, UK, to H.S., the Israeli Ministry ofScience to A.F. and H.S., and National Institute for Psycho-biology in Israel grant 3–02 to A.B. and O.K. R.O. Browne isa Kreitman Doctoral Fellow. L.B.M.S. is an Eshkol DoctoralFellow.

REFERENCES

1. Beswick, F. W., and Maynard, R. L. (1987) Poisoning in conflict.In Oxford Textbook of Medicine (Weatherall, and colleagues, eds)pp. 6.59–56.65, Oxford University Press, Oxford, UK

2. DuBois, K. P., Doull, J., Salerno, P. R., and Coon, J. M. (1949)Studies on the toxicity and mechanisms of action of p-nitrophe-nyl diethyl thionosphosphate (parathion). J. Pharmaco. Exper.Therapeutics 95, 79–91

3. Taylor, P. (2001) Anticholinesterase agents. In Goodman andGilman’s The Pharmacological Basis of Therapeutics (Hard-man, G., Limbird, L., and Gilman, A., eds) pp. 131–149,Pergamon, New York

4. Kaufer, D., Friedman, A., Seidman, S., and Soreq, H. (1998)Acute stress facilitates long-lasting changes in cholinergic geneexpression. Nature 393, 373–377

5. Soreq, H. (2005) Gulf War Syndrome, Psychological, and Chem-ical Stressors. In Encyclopedia of Stress (Fink, G., ed), Elsevier, inpress

6. Shapira, M., Tur-Kaspa, I., Bosgraaf, L., Livni, N., Grant, A. D.,Grisaru, D., Korner, M., Ebstein, R. P., and Soreq, H. (2000) A


transcription-activating polymorphism in the ACHE promoterassociated with acute sensitivity to anti-acetylcholinesterases.Hum. Mol. Genet. 9, 1273–1281

7. Aharoni, A., Gaidukov, L., Yagur, S., Toker, L., Silman, I., andTawfik, D. S. (2004) Directed evolution of mammalian paraoxo-nases PON1 and PON3 for bacterial expression and catalyticspecialization. Proc. Natl. Acad. Sci. U. S. A. 101, 482–487

8. Aharoni, A., Gaidukov, L., Khersonsky, O., Mc, Q. G. S.,Roodveldt, C., and Tawfik, D. S. (2005) The “evolvability” ofpromiscuous protein functions. Nat. Genet. 37, 73–76

9. Aviram, M., Rosenblat, M., Bisgaier, C. L., Newton, R. S., Primo-Parmo, S. L., and La Du, B. N. (1998) Paraoxonase inhibitshigh-density lipoprotein oxidation and preserves its functions. Apossible peroxidative role for paraoxonase. J. Clin. Invest. 101,1581–1590

10. Shih, D. M., Gu, L., Xia, Y. R., Navab, M., Li, W. F., Hama, S.,Castellani, L. W., Furlong, C. E., Costa, L. G., Fogelman, A. M.,and Lusis, A. J. (1998) Mice lacking serum paraoxonase aresusceptible to organophosphate toxicity and atherosclerosis.Nature 394, 284–287

11. Costa, L. G., Cole, T. B., Jarvik, G. P., and Furlong, C. E. (2003)Functional genomic of the paraoxonase (PON1) polymor-phisms: effects on pesticide sensitivity, cardiovascular disease,and drug metabolism. Annu. Rev. Med. 54, 371–392

12. Garin, M. C., James, R. W., Dussoix, P., Blanche, H., Passa, P.,Froguel, P., and Ruiz, J. (1997) Paraoxonase polymorphismMet-Leu54 is associated with modified serum concentrations ofthe enzyme. A possible link between the paraoxonase gene andincreased risk of cardiovascular disease in diabetes. J. Clin.Invest. 99, 62–66

13. Davies, H. G., Richter, R. J., Keifer, M., Broomfield, C. A.,Sowalla, J., and Furlong, C. E. (1996) The effect of the humanserum paraoxonase polymorphism is reversed with diazoxon,soman, and sarin. Nat. Genet. 14, 334–336

14. Bowers, M. B., Jr., Goodman, E., and Sim, V. M. (1964) Somebehavioral changes in man following anticholinesterase admin-istration. J. Nerv. Ment. Dis. 138, 383–389

15. Metcalf, D. R., and Holmes, J. H. (1969) VII. Toxicology andphysiology. EEG, psychological, and neurological alterations inhumans with organophosphorus exposure. Ann. N. Y. Acad. Sci.160, 357–365

16. Kaufer, D., and Soreq, H. (1999) Tracking cholinergic pathwaysfrom psychological and chemical stressors to variable neurode-terioration paradigms. Curr. Opin. Neurol. 12, 739–743

17. Ray, D. E., and Richards, P. G. (2001) The potential for toxiceffects of chronic, low-dose exposure to organophosphates.Toxicol. Lett. 120, 343–351

18. Richter, E. D., Chuwers, P., Levy, Y., Gordon, M., Grauer, F.,Marzouk, J., Levy, S., Barron, S., and Gruener, N. (1992) Healtheffects from exposure to organophosphate pesticides in workersand residents in Israel. Isr. J. Med. Sci. 28, 584–598

19. Steenland, K., Dick, R. B., Howell, R. J., Chrislip, D. W., Hines,C. J., Reid, T. M., Lehman, E., Laber, P., Krieg, E. F., Jr., andKnott, C. (2000) Neurologic function among termiticide appli-cators exposed to chlorpyrifos. Environ. Health Perspect. 108,293–300

20. Brown, M. A., and Brix, K. A. (1998) Review of health conse-quences from high-, intermediate- and low-level exposure toorganophosphorus nerve agents. J. Appl. Toxicol. 18, 393–408

21. Fiedler, N., Kipen, H., Kelly-McNeil, K., and Fenske, R. (1997)Long-term use of organophosphates and neuropsychologicalperformance. Am. J. Ind. Med. 32, 487–496

22. Levin, H. S., Rodnitzky, R. L., and Mick, D. L. (1976) Anxietyassociated with exposure to organophosphate compounds. Arch.Gen. Psychiatry 33, 225–228

23. Rodnitzky, R. L., Levin, H. S., and Mick, D. L. (1975) Occupa-tional exposure to organophosphorous pesticides: a neurobe-havioral study. Arch. Environ. Health 30, 98–103

24. Furlong, C. E., Richter, R. J., Seidel, S. L., Costa, L. G., andMotulsky, A. G. (1989) Spectrophotometric assays for the enzy-matic hydrolysis of the active metabolites of chlorpyrifos andparathion by plasma paraoxonase/arylesterase. Anal. Biochem.180, 242–247

25. Ellman, G. L., Courtney, K. D., Andres, V., Jr., and Feather-Stone, R. M. (1961) A new and rapid colorimetric determina-tion of acetylcholinesterase activity. Biochem. Pharmacol. 7, 88–95

26. Korn, A., Golan, H., Melamed, I., Pascual-Marqui, R., andFriedman, A. (2005) Focal cortical dysfunction and blood-brainbarrier disruption in patients with Postconcussion syndrome.J. Clin. Neurophysiol. 22, 1–9

27. Pavlovsky, L., Seiffert, E., Heinemann, U., Korn, A., Golan, H.,and Friedman, A. (2005) Persistent BBB disruption may under-lie alpha interferon-induced seizures. J. Neurol. 252, 42–46

28. Pascual-Marqui, R. D., Michel, C. M., and Lehmann, D. (1994)Low resolution electromagnetic tomography: a new method forlocalizing electrical activity in the brain. Int. J. Psychophysiol. 18,49–65

29. Pascual-Marqui, R. D., Esslen, M., Kochi, K., and Lehmann, D.(2002) Functional imaging with low-resolution brain electro-magnetic tomography (LORETA): a review. Methods Find. Exp.Clin. Pharmacol. 24 Suppl C, 91–95

30. Sukov, W., and Barth, D. S. (1998) Three-dimensional analysisof spontaneous and thalamically evoked gamma oscillations inauditory cortex. J. Neurophysiol. 79, 2875–2884

31. Llinas, R. R. (1988) The intrinsic electrophysiological proper-ties of mammalian neurons: insights into central nervous systemfunction. Science 242, 1654–1664

32. Dale, A. M., Liu, A. K., Fischl, B. R., Buckner, R. L., Belliveau,J. W., Lewine, J. D., and Halgren, E. (2000) Dynamic statisticalparametric mapping: combining fMRI and MEG for high-resolution imaging of cortical activity. Neuron 26, 55–67

33. Pascual-Marqui, R. D. (2002) Standardized low-resolution brainelectromagnetic tomography (sLORETA): technical details.Methods Find Exp. Clin. Pharmacol. 24 Suppl D, 5–12

34. Talairach, J., and Tournoux, P., eds (1988) Co-Planar StereotaxicAtlas of the Human Brain: Three-Dimensional Proportional System,Thieme Medical Publishers, New York.

35. Lezak, M. D. (1995) Neuropsychological Assessment, Oxford Uni-versity Press, New York

36. Strik, W. K., Fallgatter, A. J., Brandeis, D., and Pascual-Marqui,R. D. (1998) Three-dimensional tomography of event-relatedpotentials during response inhibition: evidence for phasic fron-tal lobe activation. Electroencephalogr. Clin. Neurophysiol. 108,406–413

37. Benmoyal-Segal, L., Vander, T., Shifman, S., Bryk, B., Ebstein,R. P., Marcus, E. L., Stessman, J., Darvasi, A., Herishanu, Y.,Friedman, A., and Soreq, H. (2005) Acetylcholinesterase/para-oxonase interactions increase the risk of insecticide-inducedParkinson’s disease. FASEB J. 19, 452–454

38. Brophy, V. H., Jampsa, R. L., Clendenning, J. B., McKinstry,L. A., Jarvik, G. P., and Furlong, C. E. (2001) Effects of 5�regulatory-region polymorphisms on paraoxonase-gene (PON1)expression. Am. J. Hum. Genet. 68, 1428–1436

39. Burchfiel, J. L., and Duffy, F. H. (1982) Organophosphateneurotoxicity: chronic effects of sarin on the electroencephalo-gram of monkey and man. Neurobehav. Toxicol. Teratol. 4, 767–778

40. Mesulam, M. M., Mufson, E. J., Wainer, B. H., and Levey, A. I.(1983) Central cholinergic pathways in the rat: an overviewbased on an alternative nomenclature (Ch1-Ch6). Neuroscience10, 1185–1201

41. Mesulam, M. M., Mufson, E. J., Levey, A. I., and Wainer, B. H.(1983) Cholinergic innervation of cortex by the basal forebrain:cytochemistry and cortical connections of the septal area,diagonal band nuclei, nucleus basalis (substantia innominata),and hypothalamus in the rhesus monkey. J. Comp. Neurol. 214,170–197

42. Buzsaki, G. (2002) Theta oscillations in the hippocampus.Neuron 33, 325–340

43. Kocsis, B., Di Prisco, G. V., and Vertes, R. P. (2001) Thetasynchronization in the limbic system: the role of Gudden’stegmental nuclei. Eur. J. Neurosci. 13, 381–388

44. Stewart, M., and Fox, S. E. (1990) Do septal neurons pace thehippocampal theta rhythm? Trends Neurosci. 13, 163–168

45. Vinogradova, O. S. (1995) Expression, control, and probablefunctional significance of the neuronal theta-rhythm. Prog.Neurobiol. 45, 523–583

46. Vertes, R. P., and Kocsis, B. (1997) Brainstem-diencephalo-septohippocampal systems controlling the theta rhythm of thehippocampus. Neuroscience 81, 893–926

47. Vertes, R. P., Hoover, W. B., and Viana Di Prisco, G. (2004)Theta rhythm of the hippocampus: subcortical control andfunctional significance. Behav. Cogn. Neurosci. Rev. 3, 173–200


48. Bucci, D. J., Holland, P. C., and Gallagher, M. (1998) Removalof cholinergic input to rat posterior parietal cortex disruptsincremental processing of conditioned stimuli. J. Neurosci. 18,8038–8046

49. Chiba, A. A., Bushnell, P. J., Oshiro, W. M., and Gallagher, M.(1999) Selective removal of cholinergic neurons in the basalforebrain alters cued target detection. Neuroreport 10, 3119–3123

50. McGaughy, J., Everitt, B. J., Robbins, T. W., and Sarter, M.(2000) The role of cortical cholinergic afferent projections incognition: impact of new selective immunotoxins. Behav. BrainRes. 115, 251–263

51. McGaughy, J., Dalley, J. W., Morrison, C. H., Everitt, B. J., andRobbins, T. W. (2002) Selective behavioral and neurochemicaleffects of cholinergic lesions produced by intrabasalis infusionsof 192 IgG-saporin on attentional performance in a five-choiceserial reaction time task. J. Neurosci. 22, 1905–1913

52. Martinez, V., and Sarter, M. (2004) Lateralized attentionalfunctions of cortical cholinergic inputs. Behav. Neurosci. 118,984–991

53. Turchi, J., and Sarter, M. (2000) Cortical cholinergic inputsmediate processing capacity: effects of 192 IgG-saporin-inducedlesions on olfactory span performance. Eur. J. Neurosci. 12,4505–4514

54. Sarter, M., Bruno, J. P., and Turchi, J. (1999) Basal forebrainafferent projections modulating cortical acetylcholine, atten-tion, and implications for neuropsychiatric disorders. Ann. N. Y.Acad. Sci. 877, 368–382

55. Sarter, M., and Bruno, J. P. (1997) Cognitive functions ofcortical acetylcholine: toward a unifying hypothesis. Brain Res.Brain Res. Rev. 23, 28–46

56. Sarter, M., Hasselmo, M. E., Bruno, J. P., and Givens, B. (2005)Unraveling the attentional functions of cortical cholinergicinputs: interactions between signal-driven and cognitive modu-lation of signal detection. Brain Res. Brain Res. Rev. 48, 98–111

57. Kozak, R., Bruno, J. P., and Sarter, M. (2005) AugmentedPrefrontal Acetylcholine Release during Challenged AttentionalPerformance. Cereb. Cortex. 16, 9–17

58. Gill, T. M., Sarter, M., and Givens, B. (2000) Sustained visualattention performance-associated prefrontal neuronal activity:evidence for cholinergic modulation. J. Neurosci. 20, 4745–4757

59. Weiner, L., Kreimer, D., Roth, E., and Silman, I. (1994) Oxida-tive stress transforms acetylcholinesterase to a molten-globule-like state. Biochem. Biophys. Res. Commun. 198, 915–922

60. Aviram, M., Rosenblat, M., Bisgaier, C. L., Newton, R. S.,Primo-Parmo, S. L., and La Du, B. N. (1998) Paraoxonaseinhibits high-density lipoprotein oxidation and preserves itsfunctions. A possible peroxidative role for paraoxonase. J. Clin.Invest. 101, 1581–1590

61. Jenner, P. (2003) Oxidative stress in Parkinson’s disease. Ann.Neurol. 53 Suppl 3, S26–36; discussion S36–28

62. Beeri, R., Andres, C., Lev-Lehman, E., Timberg, R., Huberman,T., Shani, M., and Soreq, H. (1995) Transgenic expression ofhuman acetylcholinesterase induces progressive cognitive dete-rioration in mice. Curr. Biol. 5, 1063–1071

63. Sklan, E. H., Lowenthal, A., Korner, M., Ritov, Y., Landers,D. M., Rankinen, T., Bouchard, C., Leon, A. S., Rice, T., Rao,D. C., Wilmore, J. H., Skinner, J. S., and Soreq, H. (2004)Acetylcholinesterase/paraoxonase genotype and expressionpredict anxiety scores in Health, Risk Factors, Exercise Train-ing, and Genetics study. Proc. Natl. Acad. Sci. U. S. A. 101,5512–5517

64. Bryk, B., BenMoyal-Segal, L., Podoly, E., Livnah, O., Eisenkraft,A., Luria, S., Cohen, A., Yehezkelli, Y., Hourvitz, A., and Soreq,H. (2005) Inherited and acquired interactions between ACHEand PON1 polymorphisms modulate plasma acetylcholinester-ase and paraoxonase activities. J. Neurochem. 92, 1216–1227

65. Hernandez, A., Gomez, M. A., Pena, G., Gil, F., Rodrigo, L.,Villanueva, E., and Pla, A. (2004) Effect of long-term exposureto pesticides on plasma esterases from plastic greenhouseworkers. J. Toxicol. Environ. Health A. 67, 1095–1108

66. Schmidt, R., Schmidt, H., Kapeller, P., Lechner, A., and Fazekas,F. (2002) Evolution of white matter lesions. Cerebrovasc. Dis. 13Suppl. 2, 16–20

67. Rodrigo, L., Hernandez, A. F., Lopez-Caballero, J. J., Gil, F., andPla, A. (2001) Immunohistochemical evidence for the expres-sion and induction of paraoxonase in rat liver, kidney, lung andbrain tissue. Implications for its physiological role. Chem. Biol.Interact. 137, 123–137

68. Glick, D., Moyal, L. B., and Soreq, H. (2003) Genetic variationin butyrylcholinesterase and the physiological consequences foracetylcholinesterase function. In Butyrylcholinesterase: Functionand Inhibition (Giacobini, E., ed) pp. 55–67, Martin Dunitz,London

Received for publication January 29, 2006.Accepted for publication March 20, 2006.


The FASEB Journal • FJ Express Summary

Coding region paraoxonase polymorphisms dictateaccentuated neuronal reactions in chronic,sub-threshold pesticide exposure

R. Orie Browne,* Liat Ben Moyal-Segal,† Dominik Zumsteg,‡ Yaron David,*Ora Kofman,§ Andrea Berger,§ Hermona Soreq,† and Alon Friedman*,1

*Departments of Physiology and Neurosurgery, Soroka University Medical Center, †The Departmentof Biological Chemistry, The Life Sciences Institute, The Hebrew University of Jerusalem,Jerusalem, Israel; ‡Krembil Neuroscience Centre, Toronto Western Hospital, Universityof Toronto, Toronto, Canada; and §Department of Behavioral Sciences, Zlotowski Centerfor Neurosciences, Ben-Gurion University of the Negev, Beersheva, Israel

To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.05-5576fje

SPECIFIC AIMS

Organophosphate compounds (OPs) are commonly used asagricultural pesticides and household insecticides through-out the world. They act by inhibiting the acetylcholine(ACh) hydrolyzing enzyme acetylcholinesterase (AChE).Humans can mount a variety of responses to OP exposure.Several enzymes, including AChE, butyrylcholinesterase(BChE), and paraoxonase1 (PON1), are known to protectagainst OP exposure. Acute poisoning by OPs leads toaccumulation of ACh at cholinergic synapses in the periph-eral and central nervous systems, potentially causing nausea,excessive salivation, incontinence, bradycardia, headache,fatigue, seizures, coma, and death. Although the short-termeffects of acute OP poisoning are understood to a greatextent, the long-term consequences of acute poisoning andchronic, subthreshold exposure are still not clear. Severalstudies have reported restlessness, forgetfulness, and otherneuropsychiatric symptoms as common complaints inexposed human populations. However, few significantchanges in cognitive function have been detected byneuropsychological testing in populations exposed to lowlevels of OPs without any history of acute poisoning. Theaim of the study was to explore gene-environment inter-actions involving functional changes in brain activity fol-lowing chronic, subthreshold organophosphate expo-sure. Our candidate gene was the ACHE/PON1 locus onchromosome 7, and the physiological response was stud-ied by combining neurophysiologic and neuropsycholog-ical approaches with biochemical and genetic tests.

PRINCIPAL FINDINGS

1. Quantitative EEG analyses and source localizationmethods revealed significant exposure-inducedchanges in EEG activity in specific cortical regions

We recorded EEG from 19 exposed and 9 control subjectswith a similar age and gender distribution. Calculation of

the average power spectrum revealed significantly lowerpower in the theta band and increased power in the beta3 band (P�0.04 and P�0.03, respectively) of exposedcompared with non-exposed groups, (Fig. 1A–D). Suchdifferences suggest that changes in cortical activity maydiffer in different brain regions. We used LORETA todetermine the cortical regions responsible for generatingsite-specific differences in the power spectra. Visualizationof activity computed by LORETA shows significant differ-ences in the localization of activity in all seven frequencybands of exposed subjects compared with controls. Inter-estingly, exposed subjects showed significantly decreasedactivity in the limbic system and cingulate cortex com-pared with controls. Additionally, they displayed signifi-cantly increased prefrontal activity in the delta and beta3bands (Fig. 1E–F).

2. Neuropsychological testing identified a significantdeficit of visual recall in exposed individuals

We conducted neuropsychological tests and comparedthe results for 24 of the exposed subjects and 24 controlsmatched for age, sex, and education, using a t test fordependent measures. The delayed memory (20 min)portion of the visual reproduction subtest of the WechslerMemory Scale indicated significant differences.

3. Chronic, subthreshold doses of organophosphatepesticides were shown to cause significant reductionsin serum acetylcholinesterase activity and increase inserum paraoxonase and arylesterase activities

Biochemical assays for AChE, BChE, PON, and aryles-terase activity were performed for 30 exposed subjects.

1 Correspondence: Department of Physiology, Faculty forHealth Sciences, Ben-Gurion University, Beer-Sheva 84105,Israel. E-mail: [email protected]

doi: 10.1096/fj.05-5576fje

17330892-6638/06/0020-1733 © FASEB

Serum AChE activity was found to be significantly lowerin exposed individuals compared with controls. Therewas no significant difference in BChE activity. BothPON and arylesterase activity were significantly higherthan predicted in exposed subjects (447 and 441% ofthe control, respectively, P�0.001) (Fig. 2A). Alter-ations in biochemical activity reflect, besides exposurelevels, also the individual’s reaction to the chemicalstress manifested by modified gene expression. This, inturn, is a composite effect of inheritance and environ-mental status (i.e., exposure). Therefore, as a follow-upto the biochemical investigation, we searched for ge-netic variation.

4. Specific paraoxonase (PON1) genotypes(polymorphisms encoding the L55M and Q192Rsubstitutions) are associated with the exposure-relatedaccentuation of the exposure-induced biochemicaland EEG changes

Biochemical analysis in carriers of specific PON1 55,192, and –108 genotypes revealed several significanttrends. In general, AChE activity did not vary by PON1genotype except for a slight decrease in LLRR com-pared with MMQQ individuals (P�0.04) (Fig. 2B). Incontrast, different PON1 genotypes were correlatedwith significant alterations in paraoxonase activity. Thepresence of the PON1 L55M substitution (TTG to ATG)was consistent with a decrease in PON activity, presum-ably related to reduced PON1 mRNA and protein levels.We controlled for the effects on PON activity associatedwith the PON1 Q192R substitution by selecting individ-uals without the R allele. Interestingly, in the non-exposed population, MMQQ individuals (n�10) exhib-ited about one-third of the PON activity of LLQQindividuals (n�14) (MMQQ/LLQQ�0.36, P�0.0001).In contrast, in exposed subjects MMQQ (n�6) andLLQQ (n�7) individuals exhibited almost equal PONactivity (MMQQ/LLQQ�1.3), suggesting that the Mallele is associated with a greater increase in PONactivity in exposed individuals. This is supported by ourdata that show 8.1 times more PON activity in exposedMM individuals than in controls and only 3.8 timesmore activity in exposed LL individuals than in con-trols. PON enzyme levels can be assessed by measuringthe arylesterase activity. Indeed, arylesterase activity inindividuals with the M allele was 6.0 times greater inexposed individuals than in non-exposed controls,whereas arylesterase activity in the exposed populationwithout the M allele was only 3.5 times greater than innon-exposed individuals (Fig. 2).

The PON1 Q192R (CAA to CGA) substitution, whichaffects catalytic efficiency, was associated with increasedPON activity. In controls there was a gene-dependentincrease in PON activity correlated with the presence ofthe R allele (Fig. 2B). Interestingly, changes in PONactivity in exposed subjects were characterized by amuch more robust change associated with the R allele.In addition, we could identify an important differencebetween exposed and non-exposed groups. Non-ex-posed LLRR individuals (n�13) showed �2 times morePON activity than non-exposed LLQQ individuals(n�10) (LLRR/LLQQ�2.09, P�0.0001), whereas inexposed subjects LLRR individuals (n�3) showed al-most 3 times more PON activity than LLQQ individuals(n�7) (LLRR/LLQQ�2.86, P�0.0001), suggesting anexposure-dependent response associated with thePON1 Q192R substitution and a potential biosensoreffect of the R allele in anti-AChE exposure.

Our biochemical data showed significant differencesin the individual responses to exposure, associated withthe presence or absence of the M (L55M) and R(Q192R) alleles in the PON1 gene. Thus, we hypothe-

Figure 1. EEG analyses in OP-exposed group individuals showdecreased power in the theta and increased power in the betabands in specific brain regions: A) Average normalized powerspectrum (1.5–30 Hz) of exposed (n�19) and non-exposed(n�9) groups. B) Area under the curve (AUC) of averagenormalized power by discrete frequency bands (delta: 1.5–6 Hz,theta: 6.5–8 Hz, alpha1: 8.5–10 Hz, alpha2: 10.5–12 Hz, beta1:12.5–18 Hz, beta2: 18.5–21 Hz, beta3: 21.5–30 Hz) in exposedand non-exposed groups. In the exposed group, theta activitywas significantly decreased (P�0.04) and beta3 activity wassignificantly increased (P�0.03). C, D) Percentage of increase ina normalized spectral power at the theta and beta3 bands isshown as a spectral power distribution map. Note the electrode-specific changes of spectral power. E, F) LORETA analyses show-ing cortical regions with statistically significant differences be-tween exposed and non-exposed groups. E) Areas withsignificantly decreased current intensity (blue, P�0.05) in thetheta band included bilaterally the amygdala, the hippocampus,the subcallosal area, the parahippocampal gyrus, and the ante-rior and posterior cingulate gyrus (Brodmann Areas 25, 30, 31,34, 36, 37). F) Areas with significantly decreased (blue,P�0.05) current density in the beta 3 band included thehippocampus and parahippocampal gyrus bilaterally; increasedsources (red, P�0.05) were found in the inferior, middle, andsuperior frontal gyri (areas 10, 11, 27, 46).

1734 Vol. 20 August 2006 BROWNE ET AL.The FASEB Journal

sized that the increase in beta3 activity observed in theEEG of exposed individuals may also differ betweenexposed individuals with different genetic profiles.We found significantly increased beta3 activity in thefrontal cortical regions (decreased in the temporalregions) of exposed individuals with the R allelecompared with controls and with exposed individualswithout the R allele. The M allele had no such effect.Comparing the LORETA values (representing the cur-rent source densities) in these brain regions revealed thatexposed individuals with the R allele showed significantlyincreased frontal activity and decreased temporal activity(P�0.03).

CONCLUSIONS AND SIGNIFICANCE

These findings relate to the state of the field byaddressing four major questions relevant to genome-

environment interactions: 1) What, if any, neurologi-cal/neuropsychological effects result from chronic,subthreshold organophosphate exposure? 2) Do gene-environment interactions play a significant role indetermining increased susceptibility to organophos-phate exposure? 3) Is cholinergic signaling involved? 4)Can such increased risks be identified in EEG andblood test analyses?

Previous studies have suggested that chronic, sub-threshold organophosphate exposure in humans maycause cognitive deficits such as decreased attention andimpaired memory. Our study adds to the growing body ofknowledge by identifying specific brain regions affectedby such exposure. Additionally, we present findings thatlink genetic polymorphisms, previously identified as po-tential candidates for conferring susceptibility to organo-phosphates, to both functional alterations in brain activityand blood biochemistry measurements.

Figure 2. Biochemical analyses inOP-exposed individuals: A) Shownare the activities of AChE (nmol/min*ml) in exposed subjects(n�29) compared to the Israeli non-exposed population (n�91); BChE(mmol*10/min*ml); PON (nmol/min*ml) and arylesterase (�mol/min*ml) (*P�0.001). B) AChE andPON activities segregated by thePON1 55/192 genotype combina-tions: LLRR (n�3), LLQR(n�6), LLQQ (n�8), LMQR

(n�2), LMQQ (n�5), MMQQ (n�6). In all graphs: open bars represent non-exposed group; filled bars, exposed group.

Figure 3. Schematic summary and pu-tative model: On exposure to OPs,PON protects AChE by degrading thepoison and reducing the threat ofoxidative damage (left). The surplusOPs that remain active after the phys-iological protective response inhibitAChE and cause an increase in ACh,leading to a feedback response thatoverproduces AChE as well as to acti-vation (red areas) or suppression(blue) of brain activity (brain car-toon). The entire biochemical andneurophysiological response to expo-sure is dependent on the genetic pro-file of the individual (i.e., on PON1polymorphism and varies with thebrain region examined).

1735PARAOXONASE POLYMORPHISMS AND PESTICIDE EXPOSURE

Documents

Coding region paraoxonase polymorphisms dictate accentuated neuronal reactions in chronic, sub-threshold pesticide exposure