- Email: [email protected]
Chlorpyrifos targets developing glia: effects on glial fibrillary acidic protein
Developmental Brain Research 133 (2002) 151–161 www.elsevier.com / locate / bres
Research report
Chlorpyrifos targets developing glia: effects on glial fibrillary acidic protein Stephanie J. Garcia, Frederic J. Seidler, Dan Qiao, Theodore A. Slotkin* Department of Pharmacology and Cancer Biology, Box 3813, Duke University Medical Center, Durham, NC 27710, USA Accepted 14 January 2002
Abstract The organophosphate pesticide, chlorpyrifos (CPF), is a developmental neurotoxicant. In cell cultures, CPF affects gliotypic cells to a greater extent than neuronotypic cells, suggesting that glial development is a specific target. We administered CPF to developing rats and examined the levels of glial fibrillary acidic protein (GFAP), an astrocytic marker. Prenatal CPF exposure (gestational days 17–20) elicited an increase in GFAP levels in fetal brain, but the effect was seen only at high doses that elicited maternal and fetal systemic toxicity. Early postnatal (PN) CPF treatment (PN1–4) elicited effects only in the cerebellum of male rats; GFAP was suppressed initially (PN5) and showed a rebound elevation (PN10) before returning to normal values by PN30. In contrast, when we administered CPF during the peak of gliogenesis and glial cell differentiation (PN11–14), GFAP was initially decreased across all brain regions and in both sexes; in males, subsequent elevations were seen on PN30, with the largest effect in the striatum; females also showed an increase in striatal GFAP. Our results indicate that CPF disrupts the pattern of glial development in vivo, with the maximum effect corresponding to the peak period of gliogenesis and glial cell differentiation. As glia are responsible for axonal guidance, synaptogenesis and neuronal nutrition, glial targeting suggests that these late-occurring developmental processes are vulnerable to CPF, extending the critical period for susceptibility into stages of synaptic plasticity, myelination, and architectural modeling of the developing brain. 2002 Elsevier Science B.V. All rights reserved. Theme: Disorders of the nervous system Topic: Neurotoxicity Keywords: Astrocyte; Brain development; Chlorpyrifos; Glia; Glial fibrillary acidic protein
1. Introduction Virtually all children in the US have detectable exposure to chlorpyrifos (CPF), an organophosphate pesticide that is a developmental neurotoxin, and the US Environmental Protection Agency has recently placed restrictions on its use inside the home [50]. Although immature animals recover more rapidly than adults from CPF-induced cholinesterase inhibition [26,27], they are more sensitive to CPF-induced CNS neurotoxicity [39,40,55], suggesting that CPF targets events specific to brain development Abbreviations: ANOVA, analysis of variance; CPF, chlorpyrifos; GD, gestational day; GFAP, glial fibrillary acidic protein; PN, postnatal day *Corresponding author. Tel.: 11-919-681-8015; fax: 11-919-6848197. E-mail address: [email protected] (T.A. Slotkin).
[38,45]. Indeed, at levels below the threshold for systemic toxicity, CPF inhibits neural cell replication, interferes with cell differentiation, evokes oxidative stress, and alters synaptic neurotransmission [19,45]. The sensitive period for CPF-induced alterations in brain development extends beyond the period of neurogenesis, into later phases in which neurons are postmitotic and are establishing synaptic connections [45]. Glia, which proliferate and differentiate later than neurons [7,20], provide nutritional, structural and homeostatic support to neurons and are essential to neuronal development [2,4,20]. Recent evidence suggests that CPF may specifically target glia [1,19,31,41]. In vitro, organophosphates affect glial markers in cell cultures of mixed neurons and glia [31]; studies with transformed gliotypic cells indicate further that CPF inhibits replication / differentiation and elicits oxidative stress [19,41]. In animals, CPF continues to
0165-3806 / 02 / $ – see front matter 2002 Elsevier Science B.V. All rights reserved. PII: S0165-3806( 02 )00283-3
152
S. J. Garcia et al. / Developmental Brain Research 133 (2002) 151 – 161
inhibit neural cell DNA synthesis and to elicit cell loss after the completion of neurogenesis and during the peak of glial differentiation [45], further indirect evidence for glial involvement. Accordingly, the current study addresses the issue of whether CPF specifically affects glial cell development in vivo, focusing on astrocytes, a glial cell type that provides guidance during neuronal migration, axonogenesis and synaptogenesis [2,4,7,20], and that protects neurons from oxidative stress [49]. We evaluated the concentration of glial fibrillary acidic protein (GFAP), an intermediate filament protein widely used as an astroglial marker, and which is known to be responsive to neurotoxicity [34,36]. During development, GFAP levels rise precipitously as neural precursor cells differentiate into astrocytes, with especially rapid development in the 2nd– 3rd postnatal week in most brain regions [20,52,56]. We therefore concentrated on three different CPF exposure paradigms: prenatal treatment on gestational days (GD) 17–20, a period in which major neurogenesis is occurring [43], and on two postnatal (PN) periods, just after birth (PN1–4) and during the peak of glial cell differentiation (PN11–14). In each case, we evaluated effects on brain regions with different maturational timetables [43]: midbrain1brainstem, which develops earliest, forebrain, with an intermediate profile, and cerebellum, which develops last. If glia are indeed a specific site for CPFinduced developmental neurotoxicity, then the fact that glial development continues into late stages of brain maturation would indicate that the window of vulnerability to CPF extends well into the postnatal period in which human exposures are likely to be highest [18,21,25].
at which point their pups were randomized within treatment groups and redistributed to the nursing dams with a litter size of 10 to maintain a standard nutritional status. Randomization was repeated at intervals of several days, and in addition, dams were rotated among litters to distribute any maternal caretaking differences randomly across litters and treatment groups. Animals were weaned on PN21. On PN30, brains were dissected into midbrain1 brainstem, forebrain and cerebellum, and then the first two regions were subdivided into midbrain, brainstem, cerebral cortex, hippocampus and striatum. Determinations utilized no more than one male and one female from each litter. For postnatal CPF treatments, all pups were randomized on the day after birth and redistributed to the dams as already described, and equal numbers of animals from a given litter were assigned to each of the treatment groups. For studies of CPF effects in the first few days after birth, animals were given 1 mg / kg daily on PN1–4. For studies in older animals, which tolerate higher doses [8,39,40,55], daily treatment with 5 mg / kg was given on PN11–14. These doses have been shown previously to alter neural function without eliciting overt systemic toxicity [8,47,55]; behavioral differences remain apparent, or may first emerge, after weaning despite the rapid recovery of cholinesterase activity [14,47]. Neither regimen evokes weight loss or mortality [8,13,23,47] and in the current study we did not observe any changes in suckling or maternal caretaking. Animals were weaned and selected from each litter as detailed above. Tissues were frozen in liquid nitrogen and stored at 245 8C until assayed.
2.2. GFAP determinations 2. Methods
2.1. Animal treatments All experiments were carried out in accordance with the Declaration of Helsinki and with the Guide for the Care and Use of Laboratory Animals as adopted and promulgated by the National Institutes of Health. Timed-pregnant Sprague–Dawley rats (Zivic Laboratories, Pittsburgh, PA) were housed in breeding cages, with a 12-h light–dark cycle and free access to food and water. CPF (Chem Service, West Chester, PA) was dissolved in dimethylsulfoxide to provide rapid and complete absorption [55] and was injected subcutaneously in a volume of 1 ml / kg body weight. For prenatal CPF exposure, dams were injected daily with CPF in doses ranging from 1 to 40 mg / kg of body weight, or with vehicle. Then 24 h after the last injection (GD21), fetuses were removed and brains were separated into forebrain and the rest of the brain; because the cerebellum represents an inappreciable proportion of brain weight on GD21, the remainder was designated as ‘midbrain1brainstem’. Additional dams in the control, 1and 5-mg / kg treatment groups were allowed to reach term,
GFAP was assayed by a modified dot-immunobinding technique [35]. Briefly, tissues were homogenized with a sonic probe (Heat Systems-Ultrasonics, Plainview, NY) in 9 vol. of hot 1% sodium dodecyl sulfate (Bio-Rad, Hercules, CA) and stored at 280 8C until use. Samples were diluted in 120 mM KCl, 20 mM NaCl, 2 mM MgCl 2 , 2 mM NaHCO 3 , 0.7% Triton X-100, 0.2% NaN 3 , 5 mM HEPES (pH 7.4) and 10-ml aliquots, containing 1–2 mg protein, were blotted in triplicate onto prewashed nitrocellulose membranes (0.2 mm; Bio-Rad). Blots were airdried overnight and then for 15 min at 60 8C. Blots were fixed in 25% isopropanol, 10% acetic acid, 65% water for 20 min at ambient temperature, rinsed in water for 10 min, incubated for 5 min in Tris-buffered saline (200 mM NaCl, 50 mM Tris, 0.002% NaN 3 , pH 7.4), and treated for 1 h with a ‘blocking solution’ of 0.5% gelatin (EIA grade; Bio-Rad) in Tris-buffered saline. Blots were then incubated for 2 h in blocking solution containing, in addition, rabbit anti-bovine polyclonal anti-GFAP antibody (1:500; DAKO, Carpentaria, CA) and 0.1% Triton X-100, followed by multiple washes. Blots were rinsed with blocking solution for 30 min, followed by addition of 20 mCi of
S. J. Garcia et al. / Developmental Brain Research 133 (2002) 151 – 161
[ 125 I]Protein-A (specific activity 382 Ci / mmol; PerkinElmer Life Sciences, Boston, MA) and 0.1% Triton X-100 for 1 h. Blots were washed repeatedly, dried overnight, and counted for radioactivity. Each blot included serial dilutions of a single preparation of adult hippocampus, which was then used to construct a standard curve to normalize the values across blots. Thus, although values are reported in relative units, quantitative comparisons across treatments, regions and ages could be carried out.
2.3. Data analysis Data were compiled as means and standard errors. Differences between treatment groups were assessed first by a global ANOVA (data log-transformed because of heterogeneous variance) incorporating all relevant variables: treatment (control, CPF), treatment period (regimen), age, region, and sex. Whenever the initial ANOVA indicated an interaction of CPF treatment with other variables, lower order ANOVAs were conducted, followed by Fisher’s Protected Least Significant Difference to determine which individual treatment effects were significantly different from the corresponding control values. In the absence of interaction terms, only main treatment effects were compiled, without subdivision into individual determinations. For convenience, some data are presented as the percentage change from the corresponding controls, but statistical significance was always assessed on the original data. For presentation purposes, control data were combined across the different treatment regimens (GD17–20, PN1–4, PN11–14), but in all cases, CPF effects were established using only the matched control groups. Significance was assumed at the level of P,0.05 for main effects; however, for interactions at P,0.1, we examined lower-order main effects after subdivision of the interactive variables [46].
3. Results
3.1. Global statistical analyses We evaluated three data groupings for main treatment effects and interactions before undertaking separate analyses for each treatment regimen and brain region. First, we compared the two postnatal regimens (PN1–4, PN11–14) across all brain regions, for the effects across two time points (24 h after the last CPF injection and 5 days later). There was a main treatment effect for CPF (P,0.007) but the effects of the two regimens differed from each other (CPF3regimen, P,0.07). The effects also showed distinct dependence on the post-treatment time (age) at which measurements were made (CPF3age, P,0.009) and were selective for brain region and sex (CPF3regimen3sex, P,0.06; CPF3regimen3age3region, P,0.008; CPF3
153
age3region3sex, P,0.07). Because this first grouping could not include the prenatal CPF regimen, which lacked the cerebellum, the sex variable, and the 5-day time point, we examined a second data grouping of all three treatment regimens (GD17–20, PN1–4, PN11–14) limiting the determinations to two regions shared by all of them (midbrain1brainstem, forebrain), a single time point (24 h post-treatment) and without regard to sex. Again, we found a highly significant difference among the three different treatment regimens (CPF3regimen, P,0.0001). Finally, we compared the subregional effects on PN30, across the two postnatal treatment regimens; again, the effects of CPF differed significantly between regimens and among regions (CPF3regimen, P,0.03; CPF3regimen3region, P, 0.05). In light of the significant differences among the CPF regimens, we subdivided the results according to regimen, and then reexamined age, region and sex differences for presentation.
3.2. Gestational CPF treatment Unlike the situation with postnatal CPF regimens, we did not have prior data available to identify the maximum tolerated dose for this particular treatment period and route of administration. Accordingly, we evaluated a range of doses from 1 to 40 mg / kg per day. Maternal weights among the different treatment groups did not differ at the start of treatment on GD17 (36165 g). By GD21, however, it was evident that the threshold for systemic toxicity, defined as maternal weight loss, was between 5 and 10 mg / kg per day (main effect of treatment, P,0.0001, n57–14 per treatment group). Weights on GD21 were 413613 g in controls, 429611 at 1 mg / kg per day, 418618 at 2 mg / kg per day, 418611 at 5 mg / kg per day, 320619 g at 10 mg / kg per day (P,0.0001 vs. control), 33169 at 20 mg / kg per day (P,0.0001), and 329613 at 40 mg / kg per day (P,0.0001). Although fetal weights were reduced at doses above the threshold for maternal toxicity, there was no reduction in the number of fetuses even at the highest CPF dose (data not shown). Accordingly, although we evaluated the immediate effects of CPF at all doses, we evaluated long-term effects (PN30) only for doses at which there was no maternal or fetal weight loss (1 or 5 mg / kg per day). In keeping with the fact that astrocytes are just beginning their differentiation during late gestation [7,20], levels of GFAP were measurable but extremely low on GD21 and rose substantially during the postnatal period (Fig. 1). CPF administration had a significant effect on GFAP assessed on GD21, 24 h after the last injection, characterized by a significant increase (main treatment effect, P, 0.006) across the two major brain regions (midbrain1 brainstem, forebrain), without distinction between regions (no treatment3region interaction) (Fig. 2). However, effects were limited to high doses (20 and 40 mg / kg per day) that exceeded the threshold for maternal and fetal
154
S. J. Garcia et al. / Developmental Brain Research 133 (2002) 151 – 161
Fig. 1. GFAP levels in brain regions of control rats during perinatal development. Data represent means and standard errors obtained from eight determinations on GD21, and from 12–40 determinations for postnatal ages. Values at the right represent subdivision of the midbrain1brainstem into its two components, and separation of the forebrain into cerebral cortex, hippocampus and striatum. Although there was a significant sex difference (interaction of age3sex), a difference was detectable for only one region at one age (forebrain on PN20, male 4162 U, female 3562 U, P,0.03), so the values are shown for the two sexes combined; the control values were also combined across the three different treatment regimens (GD17–20, PN1–4, PN11–14) but effects of CPF were evaluated only as compared to the matched control groups.
systemic toxicity; there was no effect below the threshold (1, 2, 5 mg / kg per day), nor at a dose just above the threshold (10 mg / kg per day). As presented below, we identified delayed effects of postnatal CPF treatment on GFAP in the striatum on PN30. Accordingly, we also evaluated whether, despite the lack of effect of subtoxic prenatal exposures on GFAP in the immediate post-treatment period, there might be a similar, delayed effect. Striatal GFAP was unchanged on PN30: 4261 U (n524) for control, 4462 U (n58) for 1 mg / kg per day, and 4262 U (n516) for 5 mg / kg per day. The lack of effect after gestational CPF treatment was statistically distinguishable from the positive effect (see below) seen for treatment on PN11–14 (CPF3regimen, P,0.02). There was no effect of gestational CPF exposure on the weight of the striatum on PN30 (Table 1).
3.3. CPF treatment on PN1 – 4 GFAP levels rose by only a small amount between
GD21 and PN5, the time point at the end of the early postnatal CPF treatment regimen (Fig. 1). GFAP, evaluated either 24 h after the final CPF injection (PN5) or 5 days later (PN10), was unaffected in the midbrain1brainstem and forebrain (Fig. 3). However, there were profound effects in the cerebellum, the region undergoing the greatest rate of growth during the CPF treatment period. Cerebellar GFAP was subnormal on PN5 and showed a rebound increase at the margin of statistical significance on PN10, with the effect limited to males. On PN30, there were no overall significant differences across brain subregions, but there was a tendency (P,0.06) toward a decrease across the cerebral cortex and hippocampus in females. The lack of effect on striatal GFAP in this treatment group was statistically distinguishable (CPF3regimen, P,0.005) from the induction seen in the later postnatal treatment group (see below). In keeping with earlier results [45], daily administration of 1 mg / kg per day of CPF had no effect on body or brain region weights of the pups, did not noticeably impair suckling or maternal caretaking, and evoked no discernible
S. J. Garcia et al. / Developmental Brain Research 133 (2002) 151 – 161
155
Fig. 2. Effects of gestational CPF exposure (GD17–20) on GFAP, assessed on GD21, 24 h after the last CPF injection. Data represent means and standard errors obtained from four determinations for each dose and region, presented as the percent change from control values; controls are shown in Fig. 1. ANOVA across all regions appears above the panel and asterisks denote doses at which the CPF group differs significantly from the corresponding control. Separate analyses were not conducted for each region because of the absence of a treatment3region interaction.
signs of systemic toxicity or mortality (data not shown). Previous work did not evaluate brain subregion weights on PN30, so these are shown in Table 1; CPF had no significant effect.
Table 1 Brain region weights on PN30 Region
Control
CPF GD17–20
Midbrain Brainstem Cerebral cortex Hippocampus Striatum Cerebellum
27062 14061 82266 9761 10562 23362
10363
PN1–4
PN11–14
26762 14162 84067 9562 9962 23263
26163 14162 811610 9961 11063 23363
Data represent means and standard errors of values for males and females combined, since there was no interaction of CPF treatment3sex. The values for controls were combined across the different treatment regimens, since there were no differences among them. Values shown for the striatum in the GD17–20 CPF group represent the higher dose (5 mg / kg per day) but values were similar for the lower dose (1 mg / kg per day, 11265 mg). None of the differences was statistically significant. Units are milligrams.
3.4. CPF treatment on PN11 – 14 In control rats, GFAP showed its greatest rate of rise between PN5 and PN15 (Fig. 1), in keeping with the known pattern for gliogenesis and astrocyte differentiation [7,20]. With this later CPF treatment regimen, we obtained both immediate and delayed effects on GFAP (Fig. 4). On PN15, 24 h after the last CPF injection, GFAP was subnormal when evaluated across all brain regions and both sexes. Although values returned to normal by PN20, differences reemerged in the subregional determinations on PN30. There was a main treatment effect (P,0.007) representing an overall increase in GFAP levels measured across all regions. However, there were also significant interactions with sex and brain region. Effects were statistically significant in males (P,0.0007) but not in females, and the effects were individually significant across both sexes in the striatum (P,0.002) but not elsewhere. Again, as in earlier studies [45], the CPF regimen used in this group was below the threshold for growth impairment or signs of systemic toxicity (data not shown), and the new information for subregion weights on PN30 appears in Table 1.
156
S. J. Garcia et al. / Developmental Brain Research 133 (2002) 151 – 161
Fig. 3. Effects of early neonatal CPF exposure (PN1–4) on GFAP. The top panel shows values assessed 24 h after the last injection (PN5) and 5 days later (PN10). The bottom panel shows the detailed subregional effects on PN30. Data represent means and standard errors obtained from six to 14 determinations for each age and sex in the top panel, and four to eight determinations in the bottom panel, presented as the percent change from control values; controls are shown in Fig. 1. ANOVA across all variables appears above the panels and asterisks denote individual values for which the CPF group differs significantly from the corresponding control.
S. J. Garcia et al. / Developmental Brain Research 133 (2002) 151 – 161
157
Fig. 4. Effects of late neonatal CPF exposure (PN11–14) on GFAP. The top panel shows values assessed 24 h after the last injection (PN15) and 5 days later (PN20). The bottom panel shows the detailed subregional effects on PN30. Data represent means and standard errors obtained from four to ten determinations for each age and sex in the top panel, and four to eight determinations in the bottom panel, presented as the percent change from control values; controls are shown in Fig. 1. ANOVA across all variables appears above the panels, with appropriate subdivision by sex and region as dictated by the significant interaction terms. Testing of individual sexes for each region was not carried out because of the absence of a treatment3region3sex interaction.
158
S. J. Garcia et al. / Developmental Brain Research 133 (2002) 151 – 161
4. Discussion We recently reported that CPF targets the development of C6 glioma cells in vitro, by inhibiting cell replication, disrupting signaling cascades and nuclear transcription factors involved in differentiation, and by producing and enhancing the effects of reactive oxygen species [19]. The current results indicate that glial development is disrupted by in vivo CPF exposures that do not produce overt signs of systemic toxicity. Glia control late-occurring events in CNS maturation, such as axonogenesis and synaptogenesis [7,20,51], so that the present findings indicate that effects on glia may contribute to the subsequent alterations noted for neuronal maturation and synaptic function [45]. The pattern of alterations in GFAP and the regional differences seen here are all suggestive of both primary actions of CPF as well as reactive changes to neuronal damage. Typically, when injury involves neurons only, GFAP rises almost immediately after exposure as a forerunner of gliosis [33,34,36]. This is precisely what happened when CPF was given in gestation, during a peak period of neurogenesis but a relatively quiescent period for gliogenesis. In keeping with the concept that induction of GFAP is a reactive change to neuronal damage, the doses required for this effect exceeded the threshold for systemic toxicity. It should be noted that the relative insensitivity of GFAP to gestational CPF exposure is not simply a general inability to change GFAP expression at this stage of development, as other neurotoxic agents do elicit changes at systemically subtoxic doses [53]. Our results indicate either that the fetus is protected from CPF neurotoxicity by maternal factors [27,30], or alternatively that this early treatment window lies outside the critical period for the specific vulnerability of glia. The latter interpretation is almost certainly correct. Apparent fetal protection from cholinesterase-related effects reflects rapid resynthesis of cholinesterase rather than lack of penetration of CPF to the fetus [27]. Furthermore, CPF does affect neuronal development, albeit with apparently less sensitivity than glial development [10,15,19,41,42,45,48], and accordingly, gestational CPF exposure elicits long-term neurobehavioral anomalies [9,32,38]. Our results thus suggest that the effects of CPF on the developing brain may present a shifting target: primary effects on neuronal development in the fetus but selective glial effects during later exposure periods. With postnatal CPF administration, we observed a different pattern: initial deficits of GFAP, followed by subsequent elevations. This pattern is typical of delayed glial cell differentiation [52], as would be expected from our findings with gliotypic cells in vitro [19,41]. Consistent with this interpretation, the most widespread effects of CPF were apparent with administration on PN11–14, during the peak period of astroglial proliferation and differentiation [7,20,52,56]. We did, however, find a more selective effect with earlier administration on PN1–4,
where the delayed differentiation pattern was obtained in one particular region, the cerebellum. The finding was somewhat unexpected, as the cerebellum undergoes later maturation than the midbrain1brainstem or forebrain. However, this earlier stage represents a major cerebellar growth spurt [16,43], during which Bergmann glia proliferate and guide migrating neurons to their proper locations [7,20,37,44]. Bergmann glia, since they are astrocytes, express GFAP [7,44]. Accordingly, effects on the cerebellum during the early postnatal period are likely to represent delayed development of Bergmann glia, leading to later-emerging disruption of neuronal and synaptic development. Indeed, these effects could explain why early postnatal CPF administration has major effects on cerebellar development that are unrelated to cholinergic effects or cholinesterase inhibition [38,45]. In keeping with this interpretation, behavioral deficits that emerge after CPF treatment on PN1–4 involve motor activities that are dependent on cerebellar function [14]. Although at this time we have no mechanistic explanation that could account for the sex differences, the behavioral outcomes display the same preference for males as seen here for alterations in GFAP [14]. When CPF was administered on PN11–14, during the period in which glial proliferation and differentiation are most active, we obtained widespread deficits of GFAP in the immediate post-treatment period, an effect typical for specific repression of glial development, rather than the increase that typifies reactive gliosis after neuronal damage [33,34,36,52]. Later, on PN30, there were rebound elevations of GFAP that again displayed preferential effects in males. However, most notably, there was one specific region, the striatum, that showed the greatest effect, and for this region, the increase was equally evident in both sexes. There are cogent reasons why the striatum may represent a specific target for CPF. First, this region contains by far the highest concentration of dopamine, a neurotransmitter that, when released in excess, can elicit oxidative stress [5,24]. Neonatal CPF administration evokes the release of dopamine, with nearly a ten-fold greater concentration achieved on PN15 as opposed to PN5 [12]. At the same time, CPF itself produces oxidative stress and sensitizes neurons and glia to the effects of other oxidants [3,11,19]. Although glia ordinarily protect neurons from reactive oxygen species [49], we saw impairment of glial development by CPF administration on PN11–14, the period of peak effect on dopamine release [12]. So the net actions of CPF during the later postnatal treatment period comprise oxidative stress and sensitization to other oxidative stressors, excessive release of striatal dopamine, which itself induces oxidative stress, and simultaneous compromise of the development of the glial cells responsible for the handling of oxidative species. The outcome is thus likely to resemble that seen with neurotoxic agents that affect dopamine projections, such as 6-hydroxydopamine [6]. This toxin, like CPF, elicits
S. J. Garcia et al. / Developmental Brain Research 133 (2002) 151 – 161
delayed increases in striatal GFAP, but only when administered after the developmental rise in striatal dopamine levels [6]. Taken together, then, these findings can explain both the regional selectivity seen with late postnatal CPF administration as well as the critical vulnerable period centered on the 2nd postnatal week. In any case, damage to striatal dopaminergic pathways is likely to contribute to motor deficits that emerge in adolescence or adulthood [14,29]. There may also be longer-term liabilities; with the further decline of striatal dopaminergic innervation in senescence [54], the additional burden caused by developmental CPF exposure may lead to the emergence of Parkinson-like disorders, similar to those seen with dopaminergic neurotoxins [5]. Obviously, future work should address this important issue. There are a number of important limitations to the current results. First, although we assessed a specific marker for astrocytes, there is no reason to suppose that CPF affects only astroglia. As already discussed, prenatal treatment is more likely to target developing neurons. Further, because both astrocytes and oligodendrocytes arise from a common glial precursor [7,28], there is every reason to expect that myelination may also be affected by CPF, and we are currently conducting studies to delineate this possibility. A second limitation is that our use of GFAP does not distinguish the actual cellular mechanism underlying CPF’s effects on glial development. Although in vitro work suggests multiple cellular effects, including impaired replication and differentiation, and oxidative stress [19,31,41], we do not know whether the effects seen here represent initial decreases in cell number, interference with astrocyte differentiation, or specific suppression of GFAP expression. Indeed, GFAP content is not constant per cell but rather changes with differentiation and during reactive gliosis [17,20,22]. Third, and perhaps most importantly, negative biochemical findings, such as those for midbrain1brainstem and forebrain after gestational or early postnatal exposure, may be misleading. The cerebellum, which was especially vulnerable to early postnatal CPF treatment, was too small for biochemical characterization during gestation, so we cannot be certain that it was indeed unaffected by prenatal CPF administration; immunocytochemical techniques may be more sensitive in this case. As shown by the control data for PN30, the individual subregions contained within the forebrain have widely disparate levels of GFAP; accordingly, even a robust change in the hippocampus or striatum would not be detected as a change in ‘forebrain’ values, since the bulk of the tissue is cerebral cortex. Indeed, we saw a trend toward delayed decreases in GFAP in the cerebral cortex and hippocampus in females after CPF exposure on PN1–4, and it is possible that effects on smaller regional subdivisions may therefore emerge on detailed examination; with our grouping of heterogeneous regions in the current study, a large, focal change may be obscured by the inclusion of unaffected areas. Again, these types of changes are likely
159
to contribute to sex-selective behavioral effects that emerge long after the cessation of CPF exposure [29]. Our finding that CPF targets glial cell maturation in the developing brain is of critical importance both for models of pesticide neurotoxicity and for the issue of critical periods of human exposure. Because glia regulate neuronal migration, axonogenesis and synaptogenesis, in vitro systems that fail to consider glial–neuronal interactions are likely to underestimate the impact of pesticides on brain development. Indeed, we noted previously that individual neuronal or glial cell culture systems are far less sensitive to CPF than is the developing brain in vivo [10,19,41,48]. More importantly, however, effects on glia, during later phases of development, after the closure of neurogenesis, means that the vulnerable period likely extends into childhood, when CPF exposures may be especially high [18,21,25]. The major damage elicited by CPF is thus likely to be missed if we consider only prenatal exposure paradigms.
Acknowledgements The authors thank Charlotte A. Tate and Mandy M. Cousins for technical assistance. This research was supported by USPHS ES10387 and ES10356, by a STAR fellowship from the US Environmental Protection Agency, and by the Leon Golberg Fellowship in Toxicology.
References [1] M. Aschner, Interactions between pesticides and glia: an unexplored experimental field, Neurotoxicology 21 (2000) 175–180. [2] M. Aschner, J.W. Allen, H.K. Kimelberg, R.M. LoPachin, W.J. Streit, Glial cells in neurotoxicity development, Annu. Rev. Pharmacol. Toxicol. 39 (1999) 151–173. [3] D. Bagchi, M. Bagchi, E.A. Hassoun, S.J. Stohs, In vitro and in vivo generation of reactive oxygen species, DNA damage and lactate dehydrogenase leakage by selected pesticides, Toxicology 104 (1995) 129–140. [4] S. Barone, K.P. Das, T.L. Lassiter, L.D. White, Vulnerable processes of nervous system development: a review of markers and methods, Neurotoxicology 21 (2000) 15–36. [5] D. Blum, S. Torch, N. Lambeng, M. Nissou, A.L. Benabid, R. Sadoul, J.M. Verna, Molecular pathways involved in the neurotoxicity of 6-OHDA, dopamine and MPTP: contribution to the apoptotic theory in Parkinson’s disease, Prog. Neurobiol. 65 (2001) 135–172. [6] G.R. Breese, H.E. Criswell, K.B. Johnson, J.P. O’Callaghan, G.E. Duncan, K.F. Jensen, P.E. Simson, R.A. Mueller, Neonatal destruction of dopaminergic neurons, Neurotoxicology 15 (1994) 149–159. [7] R.S. Cameron, P. Rakic, Glial cell lineage in the cerebral cortex: a review and synthesis, Glia 4 (1991) 124–137. [8] C.G. Campbell, F.J. Seidler, T.A. Slotkin, Chlorpyrifos interferes with cell development in rat brain regions, Brain Res. Bull. 43 (1997) 179–189. [9] S.M. Chanda, C.N. Pope, Neurochemical and neurobehavioral effects of repeated gestational exposure to chlorpyrifos in maternal
160
[10]
[11]
[12]
[13]
[14]
[15]
[16] [17]
[18]
[19]
[20] [21]
[22] [23]
[24]
[25]
[26]
[27]
[28] [29]
S. J. Garcia et al. / Developmental Brain Research 133 (2002) 151 – 161 and developing rats, Pharmacol. Biochem. Behav. 53 (1996) 771– 776. T.L. Crumpton, F.J. Seidler, T.A. Slotkin, Developmental neurotoxicity of chlorpyrifos in vivo and in vitro: effects on nuclear transcription factor involved in cell replication and differentiation, Brain Res. 857 (2000) 87–98. T.L. Crumpton, F.J. Seidler, T.A. Slotkin, Is oxidative stress involved in the developmental neurotoxicity of chlorpyrifos?, Dev. Brain Res. 121 (2000) 189–195. K. Dam, S.J. Garcia, F.J. Seidler, T.A. Slotkin, Neonatal chlorpyrifos exposure alters synaptic development and neuronal activity in cholinergic and catecholaminergic pathways, Dev. Brain Res. 116 (1999) 9–20. K. Dam, F.J. Seidler, T.A. Slotkin, Developmental neurotoxicity of chlorpyrifos: delayed targeting of DNA synthesis after repeated administration, Dev. Brain Res. 108 (1998) 39–45. K. Dam, F.J. Seidler, T.A. Slotkin, Chlorpyrifos exposure during a critical neonatal period elicits gender-selective deficits in the development of coordination skills and locomotor activity, Dev. Brain Res. 121 (2000) 179–187. K.P. Das, S. Barone, Neuronal differentiation in PC12 cells is inhibited by chlorpyrifos and its metabolites: is acetylcholinesterase inhibition the site of action?, Toxicol. Appl. Pharmacol. 160 (1999) 217–230. J. Dobbing, J. Sands, Comparative aspects of the brain growth spurt, Early Hum. Dev. 3 (1979) 79–83. L.F. Eng, R.S. Ghirnikar, Y.L. Lee, Glial fibrillary acidic protein: GFAP—31 years (1969–2000), Neurochem. Res. 25 (2000) 1439– 1451. R.A. Fenske, K.G. Black, K.P. Elkner, C. Lee, M.M. Methner, R. Soto, Potential exposure and health risks of infants following indoor residential pesticide applications, Am. J. Public Health 80 (1990) 689–693. S.J. Garcia, F.J. Seidler, T.L. Crumpton, T.A. Slotkin, Does the developmental neurotoxicity of chlorpyrifos involve glial targets? Macromolecule synthesis, adenylyl cyclase signaling, nuclear transcription factors, and formation of reactive oxygen in C6 glioma cells, Brain Res. 891 (2001) 54–68. C. Guerri, J. Renau-Piqueras, Alcohol, astroglia, and brain development, Mol. Neurobiol. 15 (1997) 65–81. S. Gurunathan, M. Robson, N. Freeman, B. Buckley, A. Roy, R. Meyer, J. Bukowski, P.J. Lioy, Accumulation of chlorpyrifos on residential surfaces and toys accessible to children, Environ. Health Perspect. 106 (1998) 9–16. M.E. Hatten, R.K.H. Liem, M.L. Shelanski, C.A. Mason, Astroglia in CNS injury, Glia 4 (1991) 233–243. D.E. Johnson, F.J. Seidler, T.A. Slotkin, Early biochemical detection of delayed neurotoxicity resulting from developmental exposure to chlorpyrifos, Brain Res. Bull. 45 (1998) 143–147. E. Junn, M.M. Mouradian, Apoptotic signaling in dopamine-induced cell death: the role of oxidative stress, p38 mitogen-activated protein kinase, cytochrome c and caspases, J. Neurochem. 78 (2001) 374– 383. P.J. Landrigan, L. Claudio, S.B. Markowitz, G.S. Berkowitz, B.L. Brenner, H. Romero, J.G. Wetmur, T.D. Matte, A.C. Gore, J.H. Godbold, M.S. Wolff, Pesticides and inner-city children: exposures, risks, and prevention, Environ. Health Perspect. 107 (Suppl. 3) (1999) 431–437. T.L. Lassiter, S. Barone, V.C. Moser, S. Padilla, Gestational exposure to chlorpyrifos: dose response profiles for cholinesterase and carboxylesterase activity, Toxicol. Sci. 52 (1999) 92–100. T.L. Lassiter, S. Padilla, S.R. Mortensen, S.M. Chanda, V.C. Moser, S. Barone, Gestational exposure to chlorpyrifos: apparent protection of the fetus?, Toxicol. Appl. Pharmacol. 152 (1998) 56–65. J.C. Lee, M. Mayer-Proschel, M.S. Rao, Gliogenesis in the central nervous system, Glia 30 (2000) 105–121. E.D. Levin, N. Addy, N.C. Christopher, F.J. Seidler, T.A. Slotkin,
[30]
[31]
[32]
[33]
[34] [35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43] [44]
[45] [46] [47]
[48]
[49]
[50]
Persistent behavioral consequences of neonatal chlorpyrifos exposure in rats, Dev. Brain Res. 130 (2001) 83–89. J.P.J. Maurissen, A.M. Hoberman, R.H. Garman, T.R. Hanley, Lack of selective developmental neurotoxicity in rat pups from dams treated by gavage with chlorpyrifos, Toxicol. Sci. 57 (2000) 250– 263. F. Monnet-Tschudi, M.G. Zurich, B. Schilter, L.G. Costa, P. Honegger, Maturation-dependent effects of chlorpyrifos and parathion and their oxygen analogs on acetylcholinesterase and neuronal and glial markers in aggregating brain cell cultures, Toxicol. Appl. Pharmacol. 165 (2000) 175–183. M.A. Muto, F. Lobelle, J.H. Bidanset, J.N. Wurpel, Embryotoxicity and neurotoxicity in rats associated with prenatal exposure to Dursban , Vet. Hum. Toxicol. 34 (1992) 498–501. J.P. O’Callaghan, Nervous-system specific proteins as biochemical indicators of neurotoxicity, Trends Pharmacol. Sci. 4 (1983) 388– 390. J.P. O’Callaghan, Neurotypic and gliotypic proteins as biochemical markers of neurotoxicity, Neurotoxicol. Teratol. 10 (1988) 445–452. J.P. O’Callaghan, Quantification of glial fibrillary acidic protein: comparison of slot-immunobinding assays with a novel sandwich ELISA, Neurotoxicol. Teratol. 13 (1991) 275–281. J.P. O’Callaghan, Quantitative features of reactive gliosis following toxicant-induced damage of the CNS, Ann. N.Y. Acad. Sci. 679 (1993) 195–210. I. Podkletnova, H. Alho, Neonatal noradrenaline depletion prevents the transition of Bergmann glia in the developing cerebellum, J. Chem. Neuroanat. 14 (1998) 167–173. C.N. Pope, Organophosphorus pesticides: do they all have the same mechanism of toxicity?, J. Toxicol. Environ. Health 2 (1999) 161– 181. C.N. Pope, T.K. Chakraborti, Dose-related inhibition of brain and plasma cholinesterase in neonatal and adult rats following sublethal organophosphate exposures, Toxicology 73 (1992) 35–43. C.N. Pope, T.K. Chakraborti, M.L. Chapman, J.D. Farrar, D. Arthun, Comparison of in vivo cholinesterase inhibition in neonatal and adult rats by three organophosphorothioate insecticides, Toxicology 68 (1991) 51–61. D. Qiao, F.J. Seidler, T.A. Slotkin, Developmental neurotoxicity of chlorpyrifos modeled in vitro: comparative effects of metabolites and other cholinesterase inhibitors on DNA synthesis in PC12 and C6 cells, Environ. Health Perspect. 109 (2001) 909–913. D. Rice, S. Barone, Critical periods of vulnerability for the developing nervous system: evidence from humans and animal models, Environ. Health Perspect. 108 (2000) S511–S533. P.M. Rodier, Structural-functional relationships in experimentally induced brain damage, Prog. Brain Res. 73 (1988) 335–348. A.K. Shetty, D.W. Phillips, Effects of prenatal ethanol exposure on the development of Bergmann glia and astrocytes in the rat cerebellum: an immunohistochemical study, J. Comp. Neurol. 321 (1992) 19–32. T.A. Slotkin, Developmental cholinotoxicants: nicotine and chlorpyrifos, Environ. Health Perspect. 107 (Suppl. 1) (1999) 71–80. G.W. Snedecor, W.G. Cochran, Statistical Methods, Iowa State University Press, Ames, IA, 1967. X. Song, F.J. Seidler, J.L. Saleh, J. Zhang, S. Padilla, T.A. Slotkin, Cellular mechanisms for developmental toxicity of chlorpyrifos: targeting the adenylyl cyclase signaling cascade, Toxicol. Appl. Pharmacol. 145 (1997) 158–174. X. Song, J.D. Violin, F.J. Seidler, T.A. Slotkin, Modeling the developmental neurotoxicity of chlorpyrifos in vitro: macromolecule synthesis in PC12 cells, Toxicol. Appl. Pharmacol. 151 (1998) 182–191. J. Tanaka, K. Toku, B. Zhang, K. Isihara, M. Sakanaka, N. Maeda, Astrocytes prevent neuronal death induced by reactive oxygen and nitrogen species, Glia 28 (1999) 85–96. US Environmental Protection Agency, Administrator’s Announce-
S. J. Garcia et al. / Developmental Brain Research 133 (2002) 151 – 161 ment, http: / / www.epa.gov / pesticides / announcement6800.htm, accessed 11 October, 2000. [51] E.M. Ullian, S.K. Sapperstein, K.S. Christopherson, B.A. Barres, Control of synapse number by glia, Science 291 (2001) 657–660. [52] S. Valles, J. Pitarch, J. Renau-Piqueras, C. Guerri, Ethanol exposure affects glial fibrillary acidic protein gene expression and transcription during rat brain development, J. Neurochem. 69 (1997) 2484– 2493. [53] S. Valles, M. Sancho-Tello, R. Minana, E. Climent, J. RenauPiqueras, C. Guerri, Glial fibrillary acidic protein expression in rat brain and in radial glia culture is delayed by prenatal ethanol exposure, J. Neurochem. 67 (1996) 2425–2433.
161
[54] N.D. Volkow, G.J. Wang, J.S. Fowler, Y.S. Ding, R.C. Gur, J. Gatley, J. Logan, P.J. Moberg, R. Hitzemann, G. Smith, N. Pappas, Parallel loss of presynaptic and postsynaptic dopamine markers in normal aging, Ann. Neurol. 44 (1998) 143–147. [55] K.D. Whitney, F.J. Seidler, T.A. Slotkin, Developmental neurotoxicity of chlorpyrifos: cellular mechanisms, Toxicol. Appl. Pharmacol. 134 (1995) 53–62. [56] N.H. Zawia, G.J. Harry, Developmental exposure to lead interferes with glial and neuronal differential gene expression in the rat cerebellum, Toxicol. Appl. Pharmacol. 138 (1996) 43–47.
Research report
Chlorpyrifos targets developing glia: effects on glial fibrillary acidic protein Stephanie J. Garcia, Frederic J. Seidler, Dan Qiao, Theodore A. Slotkin* Department of Pharmacology and Cancer Biology, Box 3813, Duke University Medical Center, Durham, NC 27710, USA Accepted 14 January 2002
Abstract The organophosphate pesticide, chlorpyrifos (CPF), is a developmental neurotoxicant. In cell cultures, CPF affects gliotypic cells to a greater extent than neuronotypic cells, suggesting that glial development is a specific target. We administered CPF to developing rats and examined the levels of glial fibrillary acidic protein (GFAP), an astrocytic marker. Prenatal CPF exposure (gestational days 17–20) elicited an increase in GFAP levels in fetal brain, but the effect was seen only at high doses that elicited maternal and fetal systemic toxicity. Early postnatal (PN) CPF treatment (PN1–4) elicited effects only in the cerebellum of male rats; GFAP was suppressed initially (PN5) and showed a rebound elevation (PN10) before returning to normal values by PN30. In contrast, when we administered CPF during the peak of gliogenesis and glial cell differentiation (PN11–14), GFAP was initially decreased across all brain regions and in both sexes; in males, subsequent elevations were seen on PN30, with the largest effect in the striatum; females also showed an increase in striatal GFAP. Our results indicate that CPF disrupts the pattern of glial development in vivo, with the maximum effect corresponding to the peak period of gliogenesis and glial cell differentiation. As glia are responsible for axonal guidance, synaptogenesis and neuronal nutrition, glial targeting suggests that these late-occurring developmental processes are vulnerable to CPF, extending the critical period for susceptibility into stages of synaptic plasticity, myelination, and architectural modeling of the developing brain. 2002 Elsevier Science B.V. All rights reserved. Theme: Disorders of the nervous system Topic: Neurotoxicity Keywords: Astrocyte; Brain development; Chlorpyrifos; Glia; Glial fibrillary acidic protein
1. Introduction Virtually all children in the US have detectable exposure to chlorpyrifos (CPF), an organophosphate pesticide that is a developmental neurotoxin, and the US Environmental Protection Agency has recently placed restrictions on its use inside the home [50]. Although immature animals recover more rapidly than adults from CPF-induced cholinesterase inhibition [26,27], they are more sensitive to CPF-induced CNS neurotoxicity [39,40,55], suggesting that CPF targets events specific to brain development Abbreviations: ANOVA, analysis of variance; CPF, chlorpyrifos; GD, gestational day; GFAP, glial fibrillary acidic protein; PN, postnatal day *Corresponding author. Tel.: 11-919-681-8015; fax: 11-919-6848197. E-mail address: [email protected] (T.A. Slotkin).
[38,45]. Indeed, at levels below the threshold for systemic toxicity, CPF inhibits neural cell replication, interferes with cell differentiation, evokes oxidative stress, and alters synaptic neurotransmission [19,45]. The sensitive period for CPF-induced alterations in brain development extends beyond the period of neurogenesis, into later phases in which neurons are postmitotic and are establishing synaptic connections [45]. Glia, which proliferate and differentiate later than neurons [7,20], provide nutritional, structural and homeostatic support to neurons and are essential to neuronal development [2,4,20]. Recent evidence suggests that CPF may specifically target glia [1,19,31,41]. In vitro, organophosphates affect glial markers in cell cultures of mixed neurons and glia [31]; studies with transformed gliotypic cells indicate further that CPF inhibits replication / differentiation and elicits oxidative stress [19,41]. In animals, CPF continues to
0165-3806 / 02 / $ – see front matter 2002 Elsevier Science B.V. All rights reserved. PII: S0165-3806( 02 )00283-3
152
S. J. Garcia et al. / Developmental Brain Research 133 (2002) 151 – 161
inhibit neural cell DNA synthesis and to elicit cell loss after the completion of neurogenesis and during the peak of glial differentiation [45], further indirect evidence for glial involvement. Accordingly, the current study addresses the issue of whether CPF specifically affects glial cell development in vivo, focusing on astrocytes, a glial cell type that provides guidance during neuronal migration, axonogenesis and synaptogenesis [2,4,7,20], and that protects neurons from oxidative stress [49]. We evaluated the concentration of glial fibrillary acidic protein (GFAP), an intermediate filament protein widely used as an astroglial marker, and which is known to be responsive to neurotoxicity [34,36]. During development, GFAP levels rise precipitously as neural precursor cells differentiate into astrocytes, with especially rapid development in the 2nd– 3rd postnatal week in most brain regions [20,52,56]. We therefore concentrated on three different CPF exposure paradigms: prenatal treatment on gestational days (GD) 17–20, a period in which major neurogenesis is occurring [43], and on two postnatal (PN) periods, just after birth (PN1–4) and during the peak of glial cell differentiation (PN11–14). In each case, we evaluated effects on brain regions with different maturational timetables [43]: midbrain1brainstem, which develops earliest, forebrain, with an intermediate profile, and cerebellum, which develops last. If glia are indeed a specific site for CPFinduced developmental neurotoxicity, then the fact that glial development continues into late stages of brain maturation would indicate that the window of vulnerability to CPF extends well into the postnatal period in which human exposures are likely to be highest [18,21,25].
at which point their pups were randomized within treatment groups and redistributed to the nursing dams with a litter size of 10 to maintain a standard nutritional status. Randomization was repeated at intervals of several days, and in addition, dams were rotated among litters to distribute any maternal caretaking differences randomly across litters and treatment groups. Animals were weaned on PN21. On PN30, brains were dissected into midbrain1 brainstem, forebrain and cerebellum, and then the first two regions were subdivided into midbrain, brainstem, cerebral cortex, hippocampus and striatum. Determinations utilized no more than one male and one female from each litter. For postnatal CPF treatments, all pups were randomized on the day after birth and redistributed to the dams as already described, and equal numbers of animals from a given litter were assigned to each of the treatment groups. For studies of CPF effects in the first few days after birth, animals were given 1 mg / kg daily on PN1–4. For studies in older animals, which tolerate higher doses [8,39,40,55], daily treatment with 5 mg / kg was given on PN11–14. These doses have been shown previously to alter neural function without eliciting overt systemic toxicity [8,47,55]; behavioral differences remain apparent, or may first emerge, after weaning despite the rapid recovery of cholinesterase activity [14,47]. Neither regimen evokes weight loss or mortality [8,13,23,47] and in the current study we did not observe any changes in suckling or maternal caretaking. Animals were weaned and selected from each litter as detailed above. Tissues were frozen in liquid nitrogen and stored at 245 8C until assayed.
2.2. GFAP determinations 2. Methods
2.1. Animal treatments All experiments were carried out in accordance with the Declaration of Helsinki and with the Guide for the Care and Use of Laboratory Animals as adopted and promulgated by the National Institutes of Health. Timed-pregnant Sprague–Dawley rats (Zivic Laboratories, Pittsburgh, PA) were housed in breeding cages, with a 12-h light–dark cycle and free access to food and water. CPF (Chem Service, West Chester, PA) was dissolved in dimethylsulfoxide to provide rapid and complete absorption [55] and was injected subcutaneously in a volume of 1 ml / kg body weight. For prenatal CPF exposure, dams were injected daily with CPF in doses ranging from 1 to 40 mg / kg of body weight, or with vehicle. Then 24 h after the last injection (GD21), fetuses were removed and brains were separated into forebrain and the rest of the brain; because the cerebellum represents an inappreciable proportion of brain weight on GD21, the remainder was designated as ‘midbrain1brainstem’. Additional dams in the control, 1and 5-mg / kg treatment groups were allowed to reach term,
GFAP was assayed by a modified dot-immunobinding technique [35]. Briefly, tissues were homogenized with a sonic probe (Heat Systems-Ultrasonics, Plainview, NY) in 9 vol. of hot 1% sodium dodecyl sulfate (Bio-Rad, Hercules, CA) and stored at 280 8C until use. Samples were diluted in 120 mM KCl, 20 mM NaCl, 2 mM MgCl 2 , 2 mM NaHCO 3 , 0.7% Triton X-100, 0.2% NaN 3 , 5 mM HEPES (pH 7.4) and 10-ml aliquots, containing 1–2 mg protein, were blotted in triplicate onto prewashed nitrocellulose membranes (0.2 mm; Bio-Rad). Blots were airdried overnight and then for 15 min at 60 8C. Blots were fixed in 25% isopropanol, 10% acetic acid, 65% water for 20 min at ambient temperature, rinsed in water for 10 min, incubated for 5 min in Tris-buffered saline (200 mM NaCl, 50 mM Tris, 0.002% NaN 3 , pH 7.4), and treated for 1 h with a ‘blocking solution’ of 0.5% gelatin (EIA grade; Bio-Rad) in Tris-buffered saline. Blots were then incubated for 2 h in blocking solution containing, in addition, rabbit anti-bovine polyclonal anti-GFAP antibody (1:500; DAKO, Carpentaria, CA) and 0.1% Triton X-100, followed by multiple washes. Blots were rinsed with blocking solution for 30 min, followed by addition of 20 mCi of
S. J. Garcia et al. / Developmental Brain Research 133 (2002) 151 – 161
[ 125 I]Protein-A (specific activity 382 Ci / mmol; PerkinElmer Life Sciences, Boston, MA) and 0.1% Triton X-100 for 1 h. Blots were washed repeatedly, dried overnight, and counted for radioactivity. Each blot included serial dilutions of a single preparation of adult hippocampus, which was then used to construct a standard curve to normalize the values across blots. Thus, although values are reported in relative units, quantitative comparisons across treatments, regions and ages could be carried out.
2.3. Data analysis Data were compiled as means and standard errors. Differences between treatment groups were assessed first by a global ANOVA (data log-transformed because of heterogeneous variance) incorporating all relevant variables: treatment (control, CPF), treatment period (regimen), age, region, and sex. Whenever the initial ANOVA indicated an interaction of CPF treatment with other variables, lower order ANOVAs were conducted, followed by Fisher’s Protected Least Significant Difference to determine which individual treatment effects were significantly different from the corresponding control values. In the absence of interaction terms, only main treatment effects were compiled, without subdivision into individual determinations. For convenience, some data are presented as the percentage change from the corresponding controls, but statistical significance was always assessed on the original data. For presentation purposes, control data were combined across the different treatment regimens (GD17–20, PN1–4, PN11–14), but in all cases, CPF effects were established using only the matched control groups. Significance was assumed at the level of P,0.05 for main effects; however, for interactions at P,0.1, we examined lower-order main effects after subdivision of the interactive variables [46].
3. Results
3.1. Global statistical analyses We evaluated three data groupings for main treatment effects and interactions before undertaking separate analyses for each treatment regimen and brain region. First, we compared the two postnatal regimens (PN1–4, PN11–14) across all brain regions, for the effects across two time points (24 h after the last CPF injection and 5 days later). There was a main treatment effect for CPF (P,0.007) but the effects of the two regimens differed from each other (CPF3regimen, P,0.07). The effects also showed distinct dependence on the post-treatment time (age) at which measurements were made (CPF3age, P,0.009) and were selective for brain region and sex (CPF3regimen3sex, P,0.06; CPF3regimen3age3region, P,0.008; CPF3
153
age3region3sex, P,0.07). Because this first grouping could not include the prenatal CPF regimen, which lacked the cerebellum, the sex variable, and the 5-day time point, we examined a second data grouping of all three treatment regimens (GD17–20, PN1–4, PN11–14) limiting the determinations to two regions shared by all of them (midbrain1brainstem, forebrain), a single time point (24 h post-treatment) and without regard to sex. Again, we found a highly significant difference among the three different treatment regimens (CPF3regimen, P,0.0001). Finally, we compared the subregional effects on PN30, across the two postnatal treatment regimens; again, the effects of CPF differed significantly between regimens and among regions (CPF3regimen, P,0.03; CPF3regimen3region, P, 0.05). In light of the significant differences among the CPF regimens, we subdivided the results according to regimen, and then reexamined age, region and sex differences for presentation.
3.2. Gestational CPF treatment Unlike the situation with postnatal CPF regimens, we did not have prior data available to identify the maximum tolerated dose for this particular treatment period and route of administration. Accordingly, we evaluated a range of doses from 1 to 40 mg / kg per day. Maternal weights among the different treatment groups did not differ at the start of treatment on GD17 (36165 g). By GD21, however, it was evident that the threshold for systemic toxicity, defined as maternal weight loss, was between 5 and 10 mg / kg per day (main effect of treatment, P,0.0001, n57–14 per treatment group). Weights on GD21 were 413613 g in controls, 429611 at 1 mg / kg per day, 418618 at 2 mg / kg per day, 418611 at 5 mg / kg per day, 320619 g at 10 mg / kg per day (P,0.0001 vs. control), 33169 at 20 mg / kg per day (P,0.0001), and 329613 at 40 mg / kg per day (P,0.0001). Although fetal weights were reduced at doses above the threshold for maternal toxicity, there was no reduction in the number of fetuses even at the highest CPF dose (data not shown). Accordingly, although we evaluated the immediate effects of CPF at all doses, we evaluated long-term effects (PN30) only for doses at which there was no maternal or fetal weight loss (1 or 5 mg / kg per day). In keeping with the fact that astrocytes are just beginning their differentiation during late gestation [7,20], levels of GFAP were measurable but extremely low on GD21 and rose substantially during the postnatal period (Fig. 1). CPF administration had a significant effect on GFAP assessed on GD21, 24 h after the last injection, characterized by a significant increase (main treatment effect, P, 0.006) across the two major brain regions (midbrain1 brainstem, forebrain), without distinction between regions (no treatment3region interaction) (Fig. 2). However, effects were limited to high doses (20 and 40 mg / kg per day) that exceeded the threshold for maternal and fetal
154
S. J. Garcia et al. / Developmental Brain Research 133 (2002) 151 – 161
Fig. 1. GFAP levels in brain regions of control rats during perinatal development. Data represent means and standard errors obtained from eight determinations on GD21, and from 12–40 determinations for postnatal ages. Values at the right represent subdivision of the midbrain1brainstem into its two components, and separation of the forebrain into cerebral cortex, hippocampus and striatum. Although there was a significant sex difference (interaction of age3sex), a difference was detectable for only one region at one age (forebrain on PN20, male 4162 U, female 3562 U, P,0.03), so the values are shown for the two sexes combined; the control values were also combined across the three different treatment regimens (GD17–20, PN1–4, PN11–14) but effects of CPF were evaluated only as compared to the matched control groups.
systemic toxicity; there was no effect below the threshold (1, 2, 5 mg / kg per day), nor at a dose just above the threshold (10 mg / kg per day). As presented below, we identified delayed effects of postnatal CPF treatment on GFAP in the striatum on PN30. Accordingly, we also evaluated whether, despite the lack of effect of subtoxic prenatal exposures on GFAP in the immediate post-treatment period, there might be a similar, delayed effect. Striatal GFAP was unchanged on PN30: 4261 U (n524) for control, 4462 U (n58) for 1 mg / kg per day, and 4262 U (n516) for 5 mg / kg per day. The lack of effect after gestational CPF treatment was statistically distinguishable from the positive effect (see below) seen for treatment on PN11–14 (CPF3regimen, P,0.02). There was no effect of gestational CPF exposure on the weight of the striatum on PN30 (Table 1).
3.3. CPF treatment on PN1 – 4 GFAP levels rose by only a small amount between
GD21 and PN5, the time point at the end of the early postnatal CPF treatment regimen (Fig. 1). GFAP, evaluated either 24 h after the final CPF injection (PN5) or 5 days later (PN10), was unaffected in the midbrain1brainstem and forebrain (Fig. 3). However, there were profound effects in the cerebellum, the region undergoing the greatest rate of growth during the CPF treatment period. Cerebellar GFAP was subnormal on PN5 and showed a rebound increase at the margin of statistical significance on PN10, with the effect limited to males. On PN30, there were no overall significant differences across brain subregions, but there was a tendency (P,0.06) toward a decrease across the cerebral cortex and hippocampus in females. The lack of effect on striatal GFAP in this treatment group was statistically distinguishable (CPF3regimen, P,0.005) from the induction seen in the later postnatal treatment group (see below). In keeping with earlier results [45], daily administration of 1 mg / kg per day of CPF had no effect on body or brain region weights of the pups, did not noticeably impair suckling or maternal caretaking, and evoked no discernible
S. J. Garcia et al. / Developmental Brain Research 133 (2002) 151 – 161
155
Fig. 2. Effects of gestational CPF exposure (GD17–20) on GFAP, assessed on GD21, 24 h after the last CPF injection. Data represent means and standard errors obtained from four determinations for each dose and region, presented as the percent change from control values; controls are shown in Fig. 1. ANOVA across all regions appears above the panel and asterisks denote doses at which the CPF group differs significantly from the corresponding control. Separate analyses were not conducted for each region because of the absence of a treatment3region interaction.
signs of systemic toxicity or mortality (data not shown). Previous work did not evaluate brain subregion weights on PN30, so these are shown in Table 1; CPF had no significant effect.
Table 1 Brain region weights on PN30 Region
Control
CPF GD17–20
Midbrain Brainstem Cerebral cortex Hippocampus Striatum Cerebellum
27062 14061 82266 9761 10562 23362
10363
PN1–4
PN11–14
26762 14162 84067 9562 9962 23263
26163 14162 811610 9961 11063 23363
Data represent means and standard errors of values for males and females combined, since there was no interaction of CPF treatment3sex. The values for controls were combined across the different treatment regimens, since there were no differences among them. Values shown for the striatum in the GD17–20 CPF group represent the higher dose (5 mg / kg per day) but values were similar for the lower dose (1 mg / kg per day, 11265 mg). None of the differences was statistically significant. Units are milligrams.
3.4. CPF treatment on PN11 – 14 In control rats, GFAP showed its greatest rate of rise between PN5 and PN15 (Fig. 1), in keeping with the known pattern for gliogenesis and astrocyte differentiation [7,20]. With this later CPF treatment regimen, we obtained both immediate and delayed effects on GFAP (Fig. 4). On PN15, 24 h after the last CPF injection, GFAP was subnormal when evaluated across all brain regions and both sexes. Although values returned to normal by PN20, differences reemerged in the subregional determinations on PN30. There was a main treatment effect (P,0.007) representing an overall increase in GFAP levels measured across all regions. However, there were also significant interactions with sex and brain region. Effects were statistically significant in males (P,0.0007) but not in females, and the effects were individually significant across both sexes in the striatum (P,0.002) but not elsewhere. Again, as in earlier studies [45], the CPF regimen used in this group was below the threshold for growth impairment or signs of systemic toxicity (data not shown), and the new information for subregion weights on PN30 appears in Table 1.
156
S. J. Garcia et al. / Developmental Brain Research 133 (2002) 151 – 161
Fig. 3. Effects of early neonatal CPF exposure (PN1–4) on GFAP. The top panel shows values assessed 24 h after the last injection (PN5) and 5 days later (PN10). The bottom panel shows the detailed subregional effects on PN30. Data represent means and standard errors obtained from six to 14 determinations for each age and sex in the top panel, and four to eight determinations in the bottom panel, presented as the percent change from control values; controls are shown in Fig. 1. ANOVA across all variables appears above the panels and asterisks denote individual values for which the CPF group differs significantly from the corresponding control.
S. J. Garcia et al. / Developmental Brain Research 133 (2002) 151 – 161
157
Fig. 4. Effects of late neonatal CPF exposure (PN11–14) on GFAP. The top panel shows values assessed 24 h after the last injection (PN15) and 5 days later (PN20). The bottom panel shows the detailed subregional effects on PN30. Data represent means and standard errors obtained from four to ten determinations for each age and sex in the top panel, and four to eight determinations in the bottom panel, presented as the percent change from control values; controls are shown in Fig. 1. ANOVA across all variables appears above the panels, with appropriate subdivision by sex and region as dictated by the significant interaction terms. Testing of individual sexes for each region was not carried out because of the absence of a treatment3region3sex interaction.
158
S. J. Garcia et al. / Developmental Brain Research 133 (2002) 151 – 161
4. Discussion We recently reported that CPF targets the development of C6 glioma cells in vitro, by inhibiting cell replication, disrupting signaling cascades and nuclear transcription factors involved in differentiation, and by producing and enhancing the effects of reactive oxygen species [19]. The current results indicate that glial development is disrupted by in vivo CPF exposures that do not produce overt signs of systemic toxicity. Glia control late-occurring events in CNS maturation, such as axonogenesis and synaptogenesis [7,20,51], so that the present findings indicate that effects on glia may contribute to the subsequent alterations noted for neuronal maturation and synaptic function [45]. The pattern of alterations in GFAP and the regional differences seen here are all suggestive of both primary actions of CPF as well as reactive changes to neuronal damage. Typically, when injury involves neurons only, GFAP rises almost immediately after exposure as a forerunner of gliosis [33,34,36]. This is precisely what happened when CPF was given in gestation, during a peak period of neurogenesis but a relatively quiescent period for gliogenesis. In keeping with the concept that induction of GFAP is a reactive change to neuronal damage, the doses required for this effect exceeded the threshold for systemic toxicity. It should be noted that the relative insensitivity of GFAP to gestational CPF exposure is not simply a general inability to change GFAP expression at this stage of development, as other neurotoxic agents do elicit changes at systemically subtoxic doses [53]. Our results indicate either that the fetus is protected from CPF neurotoxicity by maternal factors [27,30], or alternatively that this early treatment window lies outside the critical period for the specific vulnerability of glia. The latter interpretation is almost certainly correct. Apparent fetal protection from cholinesterase-related effects reflects rapid resynthesis of cholinesterase rather than lack of penetration of CPF to the fetus [27]. Furthermore, CPF does affect neuronal development, albeit with apparently less sensitivity than glial development [10,15,19,41,42,45,48], and accordingly, gestational CPF exposure elicits long-term neurobehavioral anomalies [9,32,38]. Our results thus suggest that the effects of CPF on the developing brain may present a shifting target: primary effects on neuronal development in the fetus but selective glial effects during later exposure periods. With postnatal CPF administration, we observed a different pattern: initial deficits of GFAP, followed by subsequent elevations. This pattern is typical of delayed glial cell differentiation [52], as would be expected from our findings with gliotypic cells in vitro [19,41]. Consistent with this interpretation, the most widespread effects of CPF were apparent with administration on PN11–14, during the peak period of astroglial proliferation and differentiation [7,20,52,56]. We did, however, find a more selective effect with earlier administration on PN1–4,
where the delayed differentiation pattern was obtained in one particular region, the cerebellum. The finding was somewhat unexpected, as the cerebellum undergoes later maturation than the midbrain1brainstem or forebrain. However, this earlier stage represents a major cerebellar growth spurt [16,43], during which Bergmann glia proliferate and guide migrating neurons to their proper locations [7,20,37,44]. Bergmann glia, since they are astrocytes, express GFAP [7,44]. Accordingly, effects on the cerebellum during the early postnatal period are likely to represent delayed development of Bergmann glia, leading to later-emerging disruption of neuronal and synaptic development. Indeed, these effects could explain why early postnatal CPF administration has major effects on cerebellar development that are unrelated to cholinergic effects or cholinesterase inhibition [38,45]. In keeping with this interpretation, behavioral deficits that emerge after CPF treatment on PN1–4 involve motor activities that are dependent on cerebellar function [14]. Although at this time we have no mechanistic explanation that could account for the sex differences, the behavioral outcomes display the same preference for males as seen here for alterations in GFAP [14]. When CPF was administered on PN11–14, during the period in which glial proliferation and differentiation are most active, we obtained widespread deficits of GFAP in the immediate post-treatment period, an effect typical for specific repression of glial development, rather than the increase that typifies reactive gliosis after neuronal damage [33,34,36,52]. Later, on PN30, there were rebound elevations of GFAP that again displayed preferential effects in males. However, most notably, there was one specific region, the striatum, that showed the greatest effect, and for this region, the increase was equally evident in both sexes. There are cogent reasons why the striatum may represent a specific target for CPF. First, this region contains by far the highest concentration of dopamine, a neurotransmitter that, when released in excess, can elicit oxidative stress [5,24]. Neonatal CPF administration evokes the release of dopamine, with nearly a ten-fold greater concentration achieved on PN15 as opposed to PN5 [12]. At the same time, CPF itself produces oxidative stress and sensitizes neurons and glia to the effects of other oxidants [3,11,19]. Although glia ordinarily protect neurons from reactive oxygen species [49], we saw impairment of glial development by CPF administration on PN11–14, the period of peak effect on dopamine release [12]. So the net actions of CPF during the later postnatal treatment period comprise oxidative stress and sensitization to other oxidative stressors, excessive release of striatal dopamine, which itself induces oxidative stress, and simultaneous compromise of the development of the glial cells responsible for the handling of oxidative species. The outcome is thus likely to resemble that seen with neurotoxic agents that affect dopamine projections, such as 6-hydroxydopamine [6]. This toxin, like CPF, elicits
S. J. Garcia et al. / Developmental Brain Research 133 (2002) 151 – 161
delayed increases in striatal GFAP, but only when administered after the developmental rise in striatal dopamine levels [6]. Taken together, then, these findings can explain both the regional selectivity seen with late postnatal CPF administration as well as the critical vulnerable period centered on the 2nd postnatal week. In any case, damage to striatal dopaminergic pathways is likely to contribute to motor deficits that emerge in adolescence or adulthood [14,29]. There may also be longer-term liabilities; with the further decline of striatal dopaminergic innervation in senescence [54], the additional burden caused by developmental CPF exposure may lead to the emergence of Parkinson-like disorders, similar to those seen with dopaminergic neurotoxins [5]. Obviously, future work should address this important issue. There are a number of important limitations to the current results. First, although we assessed a specific marker for astrocytes, there is no reason to suppose that CPF affects only astroglia. As already discussed, prenatal treatment is more likely to target developing neurons. Further, because both astrocytes and oligodendrocytes arise from a common glial precursor [7,28], there is every reason to expect that myelination may also be affected by CPF, and we are currently conducting studies to delineate this possibility. A second limitation is that our use of GFAP does not distinguish the actual cellular mechanism underlying CPF’s effects on glial development. Although in vitro work suggests multiple cellular effects, including impaired replication and differentiation, and oxidative stress [19,31,41], we do not know whether the effects seen here represent initial decreases in cell number, interference with astrocyte differentiation, or specific suppression of GFAP expression. Indeed, GFAP content is not constant per cell but rather changes with differentiation and during reactive gliosis [17,20,22]. Third, and perhaps most importantly, negative biochemical findings, such as those for midbrain1brainstem and forebrain after gestational or early postnatal exposure, may be misleading. The cerebellum, which was especially vulnerable to early postnatal CPF treatment, was too small for biochemical characterization during gestation, so we cannot be certain that it was indeed unaffected by prenatal CPF administration; immunocytochemical techniques may be more sensitive in this case. As shown by the control data for PN30, the individual subregions contained within the forebrain have widely disparate levels of GFAP; accordingly, even a robust change in the hippocampus or striatum would not be detected as a change in ‘forebrain’ values, since the bulk of the tissue is cerebral cortex. Indeed, we saw a trend toward delayed decreases in GFAP in the cerebral cortex and hippocampus in females after CPF exposure on PN1–4, and it is possible that effects on smaller regional subdivisions may therefore emerge on detailed examination; with our grouping of heterogeneous regions in the current study, a large, focal change may be obscured by the inclusion of unaffected areas. Again, these types of changes are likely
159
to contribute to sex-selective behavioral effects that emerge long after the cessation of CPF exposure [29]. Our finding that CPF targets glial cell maturation in the developing brain is of critical importance both for models of pesticide neurotoxicity and for the issue of critical periods of human exposure. Because glia regulate neuronal migration, axonogenesis and synaptogenesis, in vitro systems that fail to consider glial–neuronal interactions are likely to underestimate the impact of pesticides on brain development. Indeed, we noted previously that individual neuronal or glial cell culture systems are far less sensitive to CPF than is the developing brain in vivo [10,19,41,48]. More importantly, however, effects on glia, during later phases of development, after the closure of neurogenesis, means that the vulnerable period likely extends into childhood, when CPF exposures may be especially high [18,21,25]. The major damage elicited by CPF is thus likely to be missed if we consider only prenatal exposure paradigms.
Acknowledgements The authors thank Charlotte A. Tate and Mandy M. Cousins for technical assistance. This research was supported by USPHS ES10387 and ES10356, by a STAR fellowship from the US Environmental Protection Agency, and by the Leon Golberg Fellowship in Toxicology.
References [1] M. Aschner, Interactions between pesticides and glia: an unexplored experimental field, Neurotoxicology 21 (2000) 175–180. [2] M. Aschner, J.W. Allen, H.K. Kimelberg, R.M. LoPachin, W.J. Streit, Glial cells in neurotoxicity development, Annu. Rev. Pharmacol. Toxicol. 39 (1999) 151–173. [3] D. Bagchi, M. Bagchi, E.A. Hassoun, S.J. Stohs, In vitro and in vivo generation of reactive oxygen species, DNA damage and lactate dehydrogenase leakage by selected pesticides, Toxicology 104 (1995) 129–140. [4] S. Barone, K.P. Das, T.L. Lassiter, L.D. White, Vulnerable processes of nervous system development: a review of markers and methods, Neurotoxicology 21 (2000) 15–36. [5] D. Blum, S. Torch, N. Lambeng, M. Nissou, A.L. Benabid, R. Sadoul, J.M. Verna, Molecular pathways involved in the neurotoxicity of 6-OHDA, dopamine and MPTP: contribution to the apoptotic theory in Parkinson’s disease, Prog. Neurobiol. 65 (2001) 135–172. [6] G.R. Breese, H.E. Criswell, K.B. Johnson, J.P. O’Callaghan, G.E. Duncan, K.F. Jensen, P.E. Simson, R.A. Mueller, Neonatal destruction of dopaminergic neurons, Neurotoxicology 15 (1994) 149–159. [7] R.S. Cameron, P. Rakic, Glial cell lineage in the cerebral cortex: a review and synthesis, Glia 4 (1991) 124–137. [8] C.G. Campbell, F.J. Seidler, T.A. Slotkin, Chlorpyrifos interferes with cell development in rat brain regions, Brain Res. Bull. 43 (1997) 179–189. [9] S.M. Chanda, C.N. Pope, Neurochemical and neurobehavioral effects of repeated gestational exposure to chlorpyrifos in maternal
160
[10]
[11]
[12]
[13]
[14]
[15]
[16] [17]
[18]
[19]
[20] [21]
[22] [23]
[24]
[25]
[26]
[27]
[28] [29]
S. J. Garcia et al. / Developmental Brain Research 133 (2002) 151 – 161 and developing rats, Pharmacol. Biochem. Behav. 53 (1996) 771– 776. T.L. Crumpton, F.J. Seidler, T.A. Slotkin, Developmental neurotoxicity of chlorpyrifos in vivo and in vitro: effects on nuclear transcription factor involved in cell replication and differentiation, Brain Res. 857 (2000) 87–98. T.L. Crumpton, F.J. Seidler, T.A. Slotkin, Is oxidative stress involved in the developmental neurotoxicity of chlorpyrifos?, Dev. Brain Res. 121 (2000) 189–195. K. Dam, S.J. Garcia, F.J. Seidler, T.A. Slotkin, Neonatal chlorpyrifos exposure alters synaptic development and neuronal activity in cholinergic and catecholaminergic pathways, Dev. Brain Res. 116 (1999) 9–20. K. Dam, F.J. Seidler, T.A. Slotkin, Developmental neurotoxicity of chlorpyrifos: delayed targeting of DNA synthesis after repeated administration, Dev. Brain Res. 108 (1998) 39–45. K. Dam, F.J. Seidler, T.A. Slotkin, Chlorpyrifos exposure during a critical neonatal period elicits gender-selective deficits in the development of coordination skills and locomotor activity, Dev. Brain Res. 121 (2000) 179–187. K.P. Das, S. Barone, Neuronal differentiation in PC12 cells is inhibited by chlorpyrifos and its metabolites: is acetylcholinesterase inhibition the site of action?, Toxicol. Appl. Pharmacol. 160 (1999) 217–230. J. Dobbing, J. Sands, Comparative aspects of the brain growth spurt, Early Hum. Dev. 3 (1979) 79–83. L.F. Eng, R.S. Ghirnikar, Y.L. Lee, Glial fibrillary acidic protein: GFAP—31 years (1969–2000), Neurochem. Res. 25 (2000) 1439– 1451. R.A. Fenske, K.G. Black, K.P. Elkner, C. Lee, M.M. Methner, R. Soto, Potential exposure and health risks of infants following indoor residential pesticide applications, Am. J. Public Health 80 (1990) 689–693. S.J. Garcia, F.J. Seidler, T.L. Crumpton, T.A. Slotkin, Does the developmental neurotoxicity of chlorpyrifos involve glial targets? Macromolecule synthesis, adenylyl cyclase signaling, nuclear transcription factors, and formation of reactive oxygen in C6 glioma cells, Brain Res. 891 (2001) 54–68. C. Guerri, J. Renau-Piqueras, Alcohol, astroglia, and brain development, Mol. Neurobiol. 15 (1997) 65–81. S. Gurunathan, M. Robson, N. Freeman, B. Buckley, A. Roy, R. Meyer, J. Bukowski, P.J. Lioy, Accumulation of chlorpyrifos on residential surfaces and toys accessible to children, Environ. Health Perspect. 106 (1998) 9–16. M.E. Hatten, R.K.H. Liem, M.L. Shelanski, C.A. Mason, Astroglia in CNS injury, Glia 4 (1991) 233–243. D.E. Johnson, F.J. Seidler, T.A. Slotkin, Early biochemical detection of delayed neurotoxicity resulting from developmental exposure to chlorpyrifos, Brain Res. Bull. 45 (1998) 143–147. E. Junn, M.M. Mouradian, Apoptotic signaling in dopamine-induced cell death: the role of oxidative stress, p38 mitogen-activated protein kinase, cytochrome c and caspases, J. Neurochem. 78 (2001) 374– 383. P.J. Landrigan, L. Claudio, S.B. Markowitz, G.S. Berkowitz, B.L. Brenner, H. Romero, J.G. Wetmur, T.D. Matte, A.C. Gore, J.H. Godbold, M.S. Wolff, Pesticides and inner-city children: exposures, risks, and prevention, Environ. Health Perspect. 107 (Suppl. 3) (1999) 431–437. T.L. Lassiter, S. Barone, V.C. Moser, S. Padilla, Gestational exposure to chlorpyrifos: dose response profiles for cholinesterase and carboxylesterase activity, Toxicol. Sci. 52 (1999) 92–100. T.L. Lassiter, S. Padilla, S.R. Mortensen, S.M. Chanda, V.C. Moser, S. Barone, Gestational exposure to chlorpyrifos: apparent protection of the fetus?, Toxicol. Appl. Pharmacol. 152 (1998) 56–65. J.C. Lee, M. Mayer-Proschel, M.S. Rao, Gliogenesis in the central nervous system, Glia 30 (2000) 105–121. E.D. Levin, N. Addy, N.C. Christopher, F.J. Seidler, T.A. Slotkin,
[30]
[31]
[32]
[33]
[34] [35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43] [44]
[45] [46] [47]
[48]
[49]
[50]
Persistent behavioral consequences of neonatal chlorpyrifos exposure in rats, Dev. Brain Res. 130 (2001) 83–89. J.P.J. Maurissen, A.M. Hoberman, R.H. Garman, T.R. Hanley, Lack of selective developmental neurotoxicity in rat pups from dams treated by gavage with chlorpyrifos, Toxicol. Sci. 57 (2000) 250– 263. F. Monnet-Tschudi, M.G. Zurich, B. Schilter, L.G. Costa, P. Honegger, Maturation-dependent effects of chlorpyrifos and parathion and their oxygen analogs on acetylcholinesterase and neuronal and glial markers in aggregating brain cell cultures, Toxicol. Appl. Pharmacol. 165 (2000) 175–183. M.A. Muto, F. Lobelle, J.H. Bidanset, J.N. Wurpel, Embryotoxicity and neurotoxicity in rats associated with prenatal exposure to Dursban , Vet. Hum. Toxicol. 34 (1992) 498–501. J.P. O’Callaghan, Nervous-system specific proteins as biochemical indicators of neurotoxicity, Trends Pharmacol. Sci. 4 (1983) 388– 390. J.P. O’Callaghan, Neurotypic and gliotypic proteins as biochemical markers of neurotoxicity, Neurotoxicol. Teratol. 10 (1988) 445–452. J.P. O’Callaghan, Quantification of glial fibrillary acidic protein: comparison of slot-immunobinding assays with a novel sandwich ELISA, Neurotoxicol. Teratol. 13 (1991) 275–281. J.P. O’Callaghan, Quantitative features of reactive gliosis following toxicant-induced damage of the CNS, Ann. N.Y. Acad. Sci. 679 (1993) 195–210. I. Podkletnova, H. Alho, Neonatal noradrenaline depletion prevents the transition of Bergmann glia in the developing cerebellum, J. Chem. Neuroanat. 14 (1998) 167–173. C.N. Pope, Organophosphorus pesticides: do they all have the same mechanism of toxicity?, J. Toxicol. Environ. Health 2 (1999) 161– 181. C.N. Pope, T.K. Chakraborti, Dose-related inhibition of brain and plasma cholinesterase in neonatal and adult rats following sublethal organophosphate exposures, Toxicology 73 (1992) 35–43. C.N. Pope, T.K. Chakraborti, M.L. Chapman, J.D. Farrar, D. Arthun, Comparison of in vivo cholinesterase inhibition in neonatal and adult rats by three organophosphorothioate insecticides, Toxicology 68 (1991) 51–61. D. Qiao, F.J. Seidler, T.A. Slotkin, Developmental neurotoxicity of chlorpyrifos modeled in vitro: comparative effects of metabolites and other cholinesterase inhibitors on DNA synthesis in PC12 and C6 cells, Environ. Health Perspect. 109 (2001) 909–913. D. Rice, S. Barone, Critical periods of vulnerability for the developing nervous system: evidence from humans and animal models, Environ. Health Perspect. 108 (2000) S511–S533. P.M. Rodier, Structural-functional relationships in experimentally induced brain damage, Prog. Brain Res. 73 (1988) 335–348. A.K. Shetty, D.W. Phillips, Effects of prenatal ethanol exposure on the development of Bergmann glia and astrocytes in the rat cerebellum: an immunohistochemical study, J. Comp. Neurol. 321 (1992) 19–32. T.A. Slotkin, Developmental cholinotoxicants: nicotine and chlorpyrifos, Environ. Health Perspect. 107 (Suppl. 1) (1999) 71–80. G.W. Snedecor, W.G. Cochran, Statistical Methods, Iowa State University Press, Ames, IA, 1967. X. Song, F.J. Seidler, J.L. Saleh, J. Zhang, S. Padilla, T.A. Slotkin, Cellular mechanisms for developmental toxicity of chlorpyrifos: targeting the adenylyl cyclase signaling cascade, Toxicol. Appl. Pharmacol. 145 (1997) 158–174. X. Song, J.D. Violin, F.J. Seidler, T.A. Slotkin, Modeling the developmental neurotoxicity of chlorpyrifos in vitro: macromolecule synthesis in PC12 cells, Toxicol. Appl. Pharmacol. 151 (1998) 182–191. J. Tanaka, K. Toku, B. Zhang, K. Isihara, M. Sakanaka, N. Maeda, Astrocytes prevent neuronal death induced by reactive oxygen and nitrogen species, Glia 28 (1999) 85–96. US Environmental Protection Agency, Administrator’s Announce-
S. J. Garcia et al. / Developmental Brain Research 133 (2002) 151 – 161 ment, http: / / www.epa.gov / pesticides / announcement6800.htm, accessed 11 October, 2000. [51] E.M. Ullian, S.K. Sapperstein, K.S. Christopherson, B.A. Barres, Control of synapse number by glia, Science 291 (2001) 657–660. [52] S. Valles, J. Pitarch, J. Renau-Piqueras, C. Guerri, Ethanol exposure affects glial fibrillary acidic protein gene expression and transcription during rat brain development, J. Neurochem. 69 (1997) 2484– 2493. [53] S. Valles, M. Sancho-Tello, R. Minana, E. Climent, J. RenauPiqueras, C. Guerri, Glial fibrillary acidic protein expression in rat brain and in radial glia culture is delayed by prenatal ethanol exposure, J. Neurochem. 67 (1996) 2425–2433.
161
[54] N.D. Volkow, G.J. Wang, J.S. Fowler, Y.S. Ding, R.C. Gur, J. Gatley, J. Logan, P.J. Moberg, R. Hitzemann, G. Smith, N. Pappas, Parallel loss of presynaptic and postsynaptic dopamine markers in normal aging, Ann. Neurol. 44 (1998) 143–147. [55] K.D. Whitney, F.J. Seidler, T.A. Slotkin, Developmental neurotoxicity of chlorpyrifos: cellular mechanisms, Toxicol. Appl. Pharmacol. 134 (1995) 53–62. [56] N.H. Zawia, G.J. Harry, Developmental exposure to lead interferes with glial and neuronal differential gene expression in the rat cerebellum, Toxicol. Appl. Pharmacol. 138 (1996) 43–47.