Grantee Research Project Results

Specific Aim 1 was to develop and validate immunomagnetic bead (IMB) isolation protocols for the biomarker proteins of interest: butyrylcholinesterase (BChE), acyl peptide hydrolase (APH), and acetylcholinesterase (AChE), and to use a proteomic approach to validate the use of these proteins as biomarkers of OP exposure.

We had demonstrated that a single step IMB protocol will purify BChE from plasma and APH from red cell extracts. The OP adducts to these proteins have been identified. The serine adduct from the active metabolite of tri-cresyl phosphates (TCP) was found to undergo a unique aging reaction. In vitro inhibition of BChE with the active metabolite of TCP, cyclic saligenin cresyl phosphate (CBDP) generated both a cresyl phosphoserine and an aged phosphoserine. However, in samples from exposed individuals, only the fully aged phosphoserine modification was observed. More recent experiments have shown that AChE does not age in the same way and retains a cresyl group. We had anti-AChE antibodies generated in chicken (Aves Labs, Tigard, OR) based on the sequence differences between human and chicken upstream from the active site serine. We have used these antibodies to develop a protocol for isolation of the human active site tryptic peptide using the chicken antibodies similar to the protocols that we have developed for the rapid isolation of BChE and APH. We currently are using this protocol to examine AChE adduction by OP insecticides as well as the neurotoxins to which children are exposed during flights in commercial aircraft, the triaryl phosphates.

A second biomarker protein, acylpeptide hydrolase (APH) has proven in our mouse in vivo experiments to be very sensitive to chlorpyrifos oxon (CPO) exposure and is useful because it has a half-life three-times as long as BChE (33 d vs. 11 d). We have prepared and sent to the Centers for Disease Control and Prevention (CDC) OP modified and unmodified APH as well as a small quantity of ¹⁵N-labeled APH produced in our E. coli expression system to serve as an internal standard when quantifying OP exposures.

The immunomagnetic bead/mass spec protocols are sufficiently sensitive to determine exposures from dried blood spots, which are much easier to collect, ship, and archive.

We also added a fourth potentially useful biomarker to our targeted proteins that are modified by OP exposures. While it has been known for many years, there is no carboxylesterase 1 (CES1) in human serum/plasma, nonetheless, we can pull sufficient CES1 out of white cells to analyze for OP adduction; in fact, CES1 also is known as monocyte carboxylesterase. It has the unique property of being modified by non-bioactivated triaryl phosphates.

Specific Aims 2 and 3 made use of knockout and humanized mouse models to evaluate interactions among biomarkers of OP sensitivity, exposure, and response during critical stages of development. The experiments in Aim 2 involved chronic exposure of pregnant mice to chlorpyrifos (CP) and CPO. The experiments in Aim 3 involve neonatal exposures.

For the experiments characterizing PON1 status as a biomarker of sensitivity to CP and CPO during gestation, the initial chronic dose-response studies in pregnant and nonpregnant mice demonstrated steep dose-response curves for inhibition of BChE and AChE, with higher doses leading to weight loss and fetal abnormalities. PON1^-/- mice were more sensitive than PON1^+/+mice to the effects of chronic CPO exposure. Fetuses were more resistant than dams to CPO inhibition of BChE, and tgHuPON1_R192 females were more resistant than females expressing tgHuPON1_Q192. In our earlier studies, dose-response experiments were completed for CPO that involved gestational exposure of mice of all four genotypes (n = 10 pregnant dams per treatment group). After timed matings were performed (n = 20 females per treatment group), mice with copulatory plugs were exposed transdermally to CPO daily from GD6 to GD17. PON1^+/+, tgHuPON1_Q192, and tgHuPON1_R192 mice were exposed to vehicle (acetone) or to CPO at 0.75 mg/kg/d or 0.85 mg/kg/d. PON1^-/- mice were exposed to vehicle alone, or to CPO at 0.50 mg/kg/d or 0.75 mg/kg/d. Dams were sacrificed at GD18, and fetal and maternal tissues were dissected for measurement of enzyme inhibition in different tissues. Additionally, fetal brains were placed in RNA later and analyzed for gene expression using microarrays. At GD18 there was greater inhibition of BChE in the plasma of fetuses from PON1^-/- and tgHuPON1_Q192 dams, as compared to the plasma of fetuses from PON1^+/+ and tgHuPON1_R192 dams. Thus, the PON1 status of the dam clearly influenced CPO toxicity to the fetus. The gene expression analysis showed that gene expression in brains of the fetuses of the PON1^-/- mice was more significantly affected than in brains of fetuses of wild type mice. Also, gene expression in the brains of fetuses of the PON1_R192 mice was less affected than expression in the brains of fetuses of PON1_Q192 mice, consistent with the higher efficiency of hydrolysis of CPO by the PON1_R192 alloform. Thus, PON1 status was important in modulating CPO exposure both pre- and postnatally.

Also interesting was the observation that RBC [red blood cell] APH showed a better correlation with brain AChE inhibition than the classic biomarker protein BChE, validating our use of APH as an important and useful biomarker protein for OP exposures. Similar results in fetuses were observed when the four mouse strains were exposed to diazoxon (DZO) during gestation.

Significance

These studies have provided information on the consequences of OP exposure during development and on the role of the human PON1₁₉₂ alloforms in modulating these consequences. The results are relevant to OP sensitivity in humans, because one-half of the U.S. population of Northern European origin is homozygous for the PON1_Q192 allele. The experiments examining the effects of maternal PON1 status are providing important information on the relationship of a mother’s PON1 status to her ability to protect a fetus. There is evidence from studies in various populations that children are exposed, either in utero or postnatally, to different OPs. For example, a study in a New York population identified several diethylphosphates (possible metabolites of diazinon or chlorpyrifos, as well as of other OPs) and malathion dicarboxylic acid (a specific metabolite of malathion) in maternal urine [Engel, et al., 2007 Am J Epidemiol 165(12):1397]. The study reported behavioral alterations in newborns, which were associated with the presence of dialkyl-phosphates and appeared to be modulated by PON1 status. Our prenatal exposure experiments with dams of different PON1 status has demonstrated that the PON1^-/- dams are less well able to protect their fetuses from CPO exposure.

The biomarker project will provide a much better means of quantifying exposures. For example, it is not possible to determine if an individual has been exposed to chlorpyrifos or its breakdown product trichloropyridinol. The protein biomarkers are generated only by exposure to active OP compounds. The development of mass spectrometry (MS) protocols for determining the percentage modification of biomarker proteins will provide a much more accurate determination of target protein inhibition. Currently, a baseline level of activity is required to establish a reasonable estimate of the percentage inhibition of an individual’s BChE. The MS protocols directly determine the percentage modification without the need for a baseline activity determination. The MS analysis of modified protein should be superior to analysis of urinary metabolites for estimating exposures, since it is not possible to know whether a metabolite was taken up directly from the environment or was generated in vivo. Modification of biomarker proteins happens only when active pesticide or its oxon is taken up by an individual. The methods that we have developed for MS analysis of OP adducts will be important for the MS Facility at CDC (Rudy Johnson, Director) in identifying OP related agents of chemical terrorism and cases of pesticide poisoning. The IMB protocols that we have developed require significantly less than 50 μl of plasma and our preliminary experiments with blood spots indicate that the use of blood spot sample collection, shipping and analysis together with the MS analyses will greatly facilitate exposure studies. It was not possible previously to study and document low level exposures using the inaccurate Ellman activity assay determinations due to both inter- and intra-individual differences in BChE activity levels. The inclusion of APH as a biomarker protein will provide a much longer window in which to evaluate OP exposures and if coupled with analysis of BChE modification together with a series of post-exposure time points should allow the extrapolation back to the time of exposure in cases of acute exposures. Our E. coli protein expression system is allowing us to generate heavy isotope-labeled (¹⁵N) recombinant biomarker proteins for use as internal standards in MS analysis of these proteins for our laboratory as well as for CDC’s programs.

The ability to determine whether or not an individual has been exposed to tricresyl phosphate in an airliner cabin as the result of leaky engine seals is very important. Currently, airlines do not inform passengers of any exposures, despite the aircrews often being taken to the hospital at the end of the flight, or in one informative case where the pilot was taken off the airplane on a stretcher and crew members were taken to the hospital in ambulances (see e.g., http://www.telegraph.co.uk/travel/news/fume-incident-hospitalises-american-airlines-crew-and-raises-questions-over-safety-of-cabin-air/). Thus, there is very little information on exposure to children to these toxic fume events, which are estimated to occur with a frequency of one-half percent to one percent of flights. The biomarker research has demonstrated for the first time a physiological response to TCP exposure. Our recent results in analyzing the bioactivation of the meta and para isomers have shown that the active metabolites are different than the CBDP generated from tri-ortho cresyl phosphate. It will be necessary to decorate purified biomarker proteins with the in vitro bioactivated meta and para isomers that will provide adducts with different masses than previously characterized for the tri-ortho isomers.

The relevance of our research on gene-environment interactions involving PON1 and its polymorphisms extends beyond pesticide toxicity to multiple other PON1 substrates, including lactone-containing drugs and prodrugs, quorum sensing factors used by Pseudomonas aeruginosa for virulence and biofilm formation, and oxidized lipids involved in cardiovascular disease and other diseases. Our development of a modified PON1 status assay may allow for the early identification of males who will go on to develop Parkinson’s disease.

Females have higher levels of mitochondrial PON2 and modulate oxidative stress better than males. This finding helps to explain the lower frequency of some neurological diseases in females and provides an understanding as to why we were able to identify 41% of males with PD.

Project-Generated Resources