Developmental neurotoxicity of chlorpyrifos in vivo and in vitro: effects on nuclear transcription factors involved in cell replication and differentiation

Abstract

Chlorpyrifos is a widely used organophosphate insecticide that is a suspected developmental neurotoxin. Although chlorpyrifos exerts some effects through cholinesterase inhibition, recent studies suggest additional, direct actions on developing cells. We assessed the effects of chlorpyrifos on nuclear transcription factors involved in cell replication and differentiation using in vitro and in vivo models. HeLa nuclear protein extracts were incubated with the labeled consensus oligonucleotides for AP-1 and Sp1 transcription factors in the presence and absence of chlorpyrifos. In concentrations previously shown to affect cell development, chlorpyrifos reduced AP-1, but not Sp1 DNA-binding activity. Next, chlorpyrifos was incubated with PC12 cells either during cell replication or after initiation of differentiation with NGF. Chlorpyrifos evoked stage-specific interference with the expression of the transcription factors: Sp1 was reduced in replicating and differentiating cells, whereas AP-1 was affected only during differentiation. Finally, neonatal rats were given apparently subtoxic doses of chlorpyrifos either on postnatal days 1–4 or 11–14 and the effects were evaluated in the forebrain (an early-developing, cholinergic target region) and cerebellum (late-developing region, poor in cholinergic innervation). Again, chlorpyrifos evoked stage-specific changes in transcription factor expression and binding activity, with greater effects on Sp1 during active neurogenesis, and effects on AP-1 during differentiation. The changes were present in both forebrain and cerebellum and were gender-specific. These results indicate that chlorpyrifos interferes with brain development, in part by multiple alterations in the activity of transcription factors involved in the basic machinery of cell replication and differentiation. Noncholinergic actions of chlorpyrifos that are unique to brain development reinforce the need to examine endpoints other than cholinesterase inhibition.

Introduction

Chlorpyrifos is rapidly replacing other organophosphate insecticides both in agricultural and domestic use, largely as a result of its persistence and its relative safety; unlike many other organophosphates, chlorpyrifos does not elicit delayed peripheral neuropathies unless exposure exceeds the threshold for lethality 30, 43. A number of investigations indicate that chlorpyrifos may be more hazardous when exposure occurs during fetal or childhood development. The reference dose for all sources of chlorpyrifos is 3 μg/kg/day [13], but exposures after indoor application are estimated at one to two orders of magnitude above that level 15, 20. In infants with high mouthing behavior, cumulative exposure in a two week post-application period can be as high as 3–5 mg/kg [20]. Although less well documented, unintentional exposure to organophosphate insecticides is equally troublesome. Routine organophosphate exposures of industrial and agricultural workers, or after domestic application, are high enough to cause measurable decreases in plasma cholinesterase in pregnant women [14]. Because inhibition of brain cholinesterase correlates highly with that of plasma cholinesterase [40], it is likely that unintended exposures are often high enough to cause measurable effects. It should be noted that estimated exposure limits are not without challenge [49], as most models are based on extrapolations that involve deposition on surfaces, as well as presumed rates of absorption and metabolism. Nevertheless, recent studies of children exposed to supposedly subtoxic concentrations of pesticide mixtures (including chlorpyrifos) indicate significant neurobehavioral deficits as compared to children in the same community who do not have pesticide exposures [19].

Although chlorpyrifos is neither mutagenic nor carcinogenic and has only weak teratogenic potential 6, 17, 48, studies in neonatal rats indicate up to a 100-fold lower LD₅₀ than in adults 40, 41, 59; pharmacokinetic disparities, although present, cannot account for the large difference in neonatal vs. adult toxicity [3]. Nevertheless, cholinesterase inhibition recovers far more quickly in developing animals [9], largely because of the more rapid synthesis of new cholinesterase molecules [27]. Accordingly, rather than suggesting that the immature organism is “protected” from chlorpyrifos, cholinesterase inhibition, a surrogate marker for toxicity, may simply be inadequate to the task of detecting adverse effects on brain development [27]. This concept is reinforced by recent discoveries indicating the targeting of events that are specific to the immature brain. One of the reasons for greater safety of chlorpyrifos in adult brain is the compensatory downregulation of cholinergic receptors and of presynaptic cholinergic activity, reactive events that are not as prominent in the neonate 10, 39. Perhaps most importantly, the developing brain is undergoing active cellular and architectural modeling, and chlorpyrifos affects these events as well. We found that chlorpyrifos itself, rather than chlorpyrifos oxon, the active metabolite that targets cholinesterase, inhibits DNA synthesis, leading to decreased numbers of neural cells and abnormalities of synaptic activity 8, 11, 12, 23, 45, 55, 59. It is thus not surprising that chlorpyrifos has the potential to evoke behavioral teratogenesis at doses that do not exceed the degree of cholinesterase inhibition necessary to produce systemic toxicity [33].

Accordingly, it is evident that mechanisms other than cholinesterase inhibition are vital to our understanding of neurobehavioral teratogenesis by chlorpyrifos. Using neonatal rats, we identified a biphasic action of chlorpyrifos on cell replication in vivo, with early effects targeting regions with little or no cholinergic innervation, and later effects that did reflect cholinergic hyperstimulation [50]. With PC12 cells as a model system, we showed that mitotic arrest of undifferentiated cells in vitro could be elicited directly by chlorpyrifos itself, without contributions from cholinergic stimulation and without eliciting generalized cytotoxicity [55]. One potential contributor to these direct effects, is the targeting of cell signaling cascades that control gene expression related to cell replication and differentiation 50, 54. If this mechanism is correct, then chlorpyrifos should interfere with the expression and/or action of transcription factors known to be required for these processes. In the present study, we have examined the effects of chlorpyrifos on AP-1, a leucine-zipper transcription factor, and Sp1, a member of the zinc-finger family of transcription factors. Both these factors are major control points for genes required in cell replication/differentiation 1, 22, 28, 29, 36, 46, are highly responsive to trophic signals 24, 47, 52, 57, 58, 61and are targets for developmental neurotoxicants 34, 38, 52, 60, 61. Accordingly, perturbations of the expression or function of AP-1 and Sp1 transcription factors are likely to have deleterious consequences for brain development.

Our studies were carried out at three different organizational levels. First, using cell-free systems, we examined whether chlorpyrifos interferes with the ability of AP-1 and Sp1 transcription factors to bind to their DNA recognition sites. Next, using PC12 cells, a model for neural cell differentiation 18, 56, we determined whether chlorpyrifos exposure affects either transcription factor. Exposures were carried out first in undifferentiated cells and were then examined after initiating neuritic differentiation with NGF. Finally, we administered chlorpyrifos to neonatal rats, using doses below the threshold for growth impairment or other signs of systemic toxicity 8, 12, 23, 54. We examined the effects on two brain regions (forebrain, cerebellum) that differ both in their timetables of maturation and in their magnitude of cholinergic innervation. The forebrain begins its major phase of neurogenesis prenatally and completes cell replication in the first few days postpartum, whereas the cerebellum exhibits its mitotic peak during the second postnatal week [44]. Likewise, the forebrain develops prominent cholinergic innervation but the cerebellum remains poor in cholinergic projections 7, 21, 42, 44, 51. To delineate the critical developmental phases targeted by chlorpyrifos, we also examined effects of treatment during two distinct periods, PN1-4 and PN11-14, a period over which the systemic toxicity of chlorpyrifos decreases by an order of magnitude [59].