Polychlorinated Biphenyls PCB 52, PCB 180, and PCB 138 Impair the

Mar 18, 2010 - Polychlorinated biphenyls (PCBs) are persistent organic pollutants that accumulate in the food chain and are present in human blood and...
0 downloads 0 Views 274KB Size
Chem. Res. Toxicol. 2010, 23, 813–820

813

Polychlorinated Biphenyls PCB 52, PCB 180, and PCB 138 Impair the Glutamate-Nitric Oxide-cGMP Pathway in Cerebellar Neurons in Culture by Different Mechanisms Marta Llansola,† Carmina Montoliu,‡ Jordi Boix,† and Vicente Felipo*,† Laboratory of Neurobiology, Centro de InVestigacio´n Prı´ncipe Felipe, AVda Autopista del Saler, 16, 46012 Valencia, Spain, and Fundacio´n InVestigacio´n Hospital Clı´nico de Valencia, AV. Blasco Iba´n˜ez, 17, 46010 Valencia, Spain ReceiVed December 12, 2009

Polychlorinated biphenyls (PCBs) are persistent organic pollutants that accumulate in the food chain and are present in human blood and milk. Children born to mothers exposed to PCBs show cognitive deficits, which are reproduced in rats perinatally exposed to PCBs. It has been proposed that PCBinduced cognitive impairment is due to impairment of the glutamate-nitric oxide (NO)-cGMP pathway. The aim of the present work was to assess whether chronic exposure to the nondioxin-like PCB52, PCB138, or PCB180 alters the function of this pathway in primary cultures of rat cerebellar neurons and to assess whether different PCBs have similar or different mechanisms of action. PCB180 and PCB138 impair the function of the glutamate-NO-cGMP pathway at nanomolar concentrations, and PCB52 impairs the function of the glutamate-NO-cGMP pathway at micromolar concentrations. The mechanisms by which different PCBs impair the function of the glutamate-NO-cGMP pathway are different. Each PCB affects the pathway at more than one step but with different potency and, for some steps, in opposite ways. Exposure to the PCBs alters the basal concentrations of intracellular calcium, NO, and cGMP. The three PCBs increase NO; however, PCB52 and PCB138 increase basal cGMP, while PCB180 decreases it. PCB52 and PCB138 decrease the activation of soluble guanylate cyclase by NO, and PCB180 increases it. Long-term exposure to PCB52, PCB180, or PCB138 reduces the activation of NO synthase and the whole glutamate-NO-cGMP pathway in response to activation of N-methyl-D-aspartate receptors. The EC50 was 300 nM for PCB52 and 2 nM for PCB138 or PCB180. These results show that chronic exposure to nondioxin like PCBs impairs the function of the glutamate-NO-cGMP pathway in cerebellar neurons by different mechanisms and with different potencies. Impaired function of this pathway would contribute to the cognitive alterations induced by perinatal exposure to PCBs in humans. Introduction 1

Polychlorinated biphenyls (PCBs) are a family of 209 industrial chemicals that are persistent organic pollutants and accumulate in the food chain. Humans are exposed to PCBs through diet, and PCBs are present in human blood and milk (1, 2). PCBs are toxic, endocrine disrupters and induce neurotoxicity (3). Children born to mothers exposed to PCBs show memory deficits and cognitive dysfunctions as well as sensory and motor disorders (4-8). A possible contributor to these neurological alterations could be altered neuronal development. In vitro, PCBs may affect critical developmental periods of the neurons from neurite extension to dendritic growth and spine and synapse formation, resulting in altered developmental trajectory of neurons in vitro. Developmental exposure to PCBs affects dendritogenesis in the brain and maturation of neurons in vitro. Exposure to Aroclor 8 (a mixture of PCBs) of hypothalamic gonadotropin-releasing hormone neurons in culture alters gene expression, cell survival, and neurite outgrowth (9). Exposure of PC12 cells to Aroclor * To whom correspondence should be addressed. Tel: 34 963289680. Fax: 34 963289701. E-mail: [email protected]. † Centro de Investigacio´n Prı´ncipe Felipe. ‡ Fundacio´n Investigacio´n Hospital Clı´nico de Valencia. 1 Abbreviations: PCB, polychlorinated biphenyls; NO, nitric oxide; NMDA, N-methyl-D-aspartate; NOS, nitric oxide synthase; SNAP, S-nitrosoN-acetylpenicillamine.

enhances nerve growth factor-stimulated neurotrophic effects (10). PCB95 promotes dendritic growth in primary cultures of cortical neurons by altering ryanodine receptors (11, 12). Developmental exposure to Aroclor 1254 impairs DNA binding of different transcription factors in the cerebellum and hippocampus of developing rats (13) and alters the expression of synaptic proteins in different cerebral areas (14). These alterations could contribute to PCB-induced altered cerebral function and to cognitive and motor impairment. Perinatal exposure to PCB153 or PCB126 impairs cognitive function in rats when they are 3 months old, reducing the ability of the rats to learn a Y maze conditional discrimination task (15). The mechanisms underlying this PCB-induced impairment of cognitive function were analyzed by these authors. This type of learning is modulated by the glutamate-nitric oxide (NO)-cGMP pathway, associated with N-methyl-D-aspartate (NMDA) receptors. Activation of NMDA receptors increases calcium in the postsynaptic neuron. Calcium binds to calmodulin and activates neuronal nitric oxide synthase (NOS), increasing NO, which activates soluble guanylate cyclase (sGC), increasing cGMP. The function of this glutamate-NO-cGMP pathway and the learning ability decrease in parallel in adults as compared to young rats (16). Also, the function of the pathway and the learning ability are impaired in rats with chronic hyperammonemia (17, 18) and hepatic encephalopathy (19, 20), and both are

10.1021/tx900440q  2010 American Chemical Society Published on Web 03/18/2010

814

Chem. Res. Toxicol., Vol. 23, No. 4, 2010

restored in parallel by treatments with phosphodiesterase inhibitors (18, 20), anti-inflammatories (21), or antagonists of GABAA receptors (22). Perinatal exposure to PDBE99 increases in parallel the function of the pathway and the ability to learn the Y maze task (23). All of these reports support the fact that theabilitytolearnthistaskismodulatedbytheglutamate-NO-cGMP pathway. Piedrafita et al. (15) showed that perinatal exposure to PCB153 or PCB126 impairs the function of this pathway in the brain in vivo, which would be responsible for the impairment of learning. The mechanisms by which the PCBs impair the function of the pathway cannot be analyzed in detail in vivo. Llansola et al. (24) showed that exposure to PCB153 and PCB126 also impairs the function of the glutamate-NO-cGMP pathway in primary cultures of cerebellar neurons, which are therefore a good model to analyze the underlying mechanisms. The mechanisms by which PCB153 and PCB126 affect the pathway were different. PCB153, but not PCB126, reduced the activation of guanylate cyclase by NO. PCB126 reduced the increase of intracellular calcium induced by NMDA. PCBs are classified in two groups depending on their toxicological profiles: dioxin-like PCBs and nondioxin-like PCBs, which have different mechanisms of toxicity. PCB126 is dioxin-like, while PCB153 is nondioxin-like. The aim of present work was to assess whether other nondioxin-like PCB congeners, PCB52, PCB180, and PCB138, also impair the glutamate-NO-cGMP pathway in cerebellar neurons in culture and to assess whether different nondioxin-like PCBs affect the pathway by similar or different mechanisms.

Materials and Methods Primary Cultures of Neurons. Primary cultures of cerebellar neurons were prepared as previously described (25). Briefly, cerebella from 7 day old Wistar rats were rapidly dissected and incubated with 3 mg/mL Dispase (grade II) for 30 min in a 5% CO2 incubator at 37 °C. The supernatant was removed, and the tissue was incubated with basal Eagle medium (BME) containing 40 µg/mL DNase I for 20 min. The cellular suspension was filtered through a mesh with a pore size of 90 µm and rinsed three times with BME. Finally, the cells were resuspended in complete medium [BME containing 10% heat-inactivated fetal bovine serum (Gibco), 2 mM glutamine, 100 µg/mL gentamicin, and 25 mM KCl]. Cells were plated onto polylysine-coated plates, except for viability experiments in which cells were plated in coverslips coated with polylysine (312000 cells/cm2 in both cases, in 35 mm diameter plates). After 20 min at 37 °C, the medium containing unattached cells was removed, and fresh medium was added. The cells were grown at 37 °C in a 5% CO2 atmosphere. To prevent proliferation of non-neuronal cells, 10 µM cytosine arabinoside was added 24 h after plating. Glucose (5.6 mM) was added to the culture medium twice a week. Exposure to Neurotoxicants. PCB180, PCB138, or PCB52 were added to the culture medium 24 h after seeding and were present for 10-13 days. PCBs were dissolved in DMSO, and the same concentration of DMSO was added to the culture medium of the control plates as in ref 24. Determination of Neuronal Viability. Neuronal viability was assessed as previously described (26) by double fluorescence staining with propidium iodide (0.46 µg/mL, 1 min), which labels nonviable cells in red, and fluorescein diacetate (15 µg/mL, 5 min), which stains viable cells in green. At least 1000 neurons were counted for each point directly under the microscope, and the percentage of viable cells was calculated. Neuronal viability was assessed after long-term exposure for 11-14 days to different concentrations of PCB52, PCB138, or PCB180 to choose the doses to be used in the subsequent experiments, avoiding use of doses leading to massive death.

Llansola et al. Determination of Basal cGMP in Cultured Neurons. Primary cultured neurons were used 11-14 days after seeding, and basal intracellular cGMP was determined using the BIOTRAK cGMP enzymeimmunoassay kit from Amersham as described in ref 24. Activation of the Glutamate-NO-cGMP Pathway or sGC in Cerebellar Neurons. To activate the glutamate-NO-cGMP pathway, 0.3 mM NMDA was added for 5 min at 37 °C. To activate sGC, the NO-generating agent S-nitroso-N-acetylpenicillamine (SNAP) (10 µM) was added for 5 min. After these activations, cGMP was determined as described in ref 24. Determination of Nitrites + Nitrates in Cultured Neurons. Primary cultured neurons were used 11-14 days after seeding. Monolayers in tissue culture dishes (35 mm diameter) were washed three times with prewarmed Locke’s solution without magnesium. After treatments, Locke’s solution was removed, and the neurons were resuspended in 100 µL of the assay buffer and disrupted by sonication. Samples were centrifuged (13000g, 5 min), and nitrites + nitrates were measured in the supernatant. Nitrates were transformed in nitrites by the addition of nitrate reductase. Nitrites were detected by Griess reaction (27). Pellets were resuspended in 100 µL of 0.5 N NaOH-1.5 mg/mL sodium deoxicolate, and protein was measured by the bicinchonic acid method using bovine serum albumin as a standard. Determination of Free Intracellular Calcium Concentration. After 10-13 days of PCB exposure, a basal and NMDAinduced increase of intracellular calcium was determined using Fura2-AM as previously described (24). Briefly, neurons were washed three times with prewarmed Locke’s solution without magnesium. Neurons were charged with the fluorescent probe (Fura2-AM, 4 µM) for 45 min and then washed 4-5 times to eliminate the excess of probe and incubated for 15 min to complete probe cutting by intracellular esterases. Plates were put on a chamber with controlled temperature (32 °C) and with continuous superfusion of Locke’s solution. Fluorescence was measured with a fluorescence microscopy connected to a Digital CCD camera (Hamamatsu), using 340 and 380 nm as λ excitation and 510 nm as λ emission. The ratio F340/F380 was determined with the Aquacosmos software (Hamamatsu). To measure the increase of intracellular calcium induced by NMDA, we perfused 12 mL of 0.3 mM NMDA in Locke’s solution followed by Locke’s solution. The free calcium concentration in the neurons was calculated as in ref 18. The value of R is the mean of the ratio F340/F380 for at least 40 neurons for each data. Each point of the graph corresponds to at least five experiments with different cultures with duplicated samples for each experiment and 40 neurons per sample. Determination of the Amount of the Proteins of the Glutamate-NO-cGMP Pathway by Immunoblotting. Cerebellar neurons in culture were washed and homogenized in lysis buffer as described in ref 24. Samples were subjected to SDS-polyacrylamide gel electrophoresis and immunoblotting as previously described (28) using monoclonal antibodies against the NR1 subunit of the NMDA receptor (clone 54.1, Pharmingen, San Diego, CA), calmodulin, and actin (Abcam, Paris, France) and polyclonal antibodies against nNOS (BD Biosciences Transduction Laboratories, Philadelphia, PA), guanylate cyclase R1 subunit (Sigma, St. Louis, MO), and guanylate cyclase β1 subunit made in rabbit in our laboratory. All primary antibodies were used at 1:1000 dilution. Secondary antibodies conjugated with alkaline phosphatase were from Sigma. After development using alkaline phosphatase, images were captured with a Hewlett-Packard scan Scanjet 5300C. The intensities of the bands were measured using the program AlphaImager 2200 (AlphaEaseFC 2200 for Windows, Cambridge, United Kingdom) and normalized for the intensity of the actin band, used as a loading control. Statistical Analysis. The data shown are the means ( standard errors of the mean (SEMs) of 4-7 experiments, with duplicate or triplicate samples in each experiment. Statistical significance was estimated with one-way ANOVA and a Newman-Keuls multiple comparison test, by using GraphPad PRISM software for Windows (GraphPad software Inc., La Jolla, CA).

NDL-PCBs Impair Neuronal Signal Transduction

Figure 1. Effects of chronic exposure to PCB52, PCB180, or PCB138 on neuronal viability. Values are the means ( SEMs of triplicate samples from five different culture preparations. Values that are significantly different from control values are indicated by asterisks (*p < 0.05, **p < 0.01, and ***p < 0.001).

Figure 2. Long-term exposure of cerebellar neurons in culture to PCB52, PCB180, or PCB138 alters basal levels of cGMP. Values are the means ( SEMs of triplicate samples from seven different culture preparations. Values that are significantly different from control values are indicated by asterisks (*p < 0.05, **p < 0.01, and ***p < 0.001).

Chem. Res. Toxicol., Vol. 23, No. 4, 2010 815

Figure 3. Long-term exposure of cerebellar neurons in culture to PCB52,PCB180,orPCB138impairsthefunctionoftheglutamate-NO-cGMP pathway. Values are the means ( SEMs of triplicate samples from five different culture preparations. Values that are significantly different from control values are indicated by asterisks (*p < 0.05, and **p < 0.01).

Results

Figure 4. Long-term exposure of cerebellar neurons in culture to PCB52, PCB180, or PCB138 impairs activation of sGC by NO. Values are the means ( SEMs of triplicate samples from five different culture preparations. Values that are significantly different from control values are indicated by asterisks (*p < 0.05, **p < 0.01, and ***p < 0.001).

Effects of Chronic Exposure to PCB52, PCB138, or PCB180 on Neuronal Viability. Long-term exposure to PCB52 induced some neuronal death at 30 µM (67 ( 2% of survival vs 81 ( 2% in control neurons). The EC50 for PCB52 was 80 µM. Exposure to 30 µM PCB180 induced some neuronal death (74 ( 2% of survival vs 81 ( 2% in control neurons). The EC50 for PCB180 was 98 µM. Long-term exposure to PCB138 induced some neuronal death at concentrations at which PCB52 or PCB180 did not affect viability. For example, at 0.3 µM PCB138, neuronal survival was 69 ( 2%. However, the EC50 for PCB138 (31 µM) was similar to those for the other PCB congeners (Figure 1). PCB52 and PCB138 Increase and PCB180 Decreases Basal Levels of cGMP. PCB52 increased the basal levels of cGMP at concentrations of 1 µM or higher (EC50 ) 1 µM). PCB138 also increased the basal levels of cGMP but at much lower concentrations (EC50 ) 0.35 nM). However, exposure to PCB180 induces the opposite effect. PCB180 decreases the basal level of cGMP at concentrations between 0.1 and 10 µM (EC50 ) 80 nM) (Figure 2). PCB52, PCB138, and PCB180 Impair Activation of the Whole Glutamate-NO-cGMP Pathway by NMDA. When 0.3 mM NMDA was added to activate the glutamate-NO-cGMP

pathway, cGMP increased to 581 ( 94% of basal in control neurons. The increase was significantly lower in neurons treated with PCB52, PCB138, or PCB180 (Figure 3), indicating that all of them impair the function of the glutamate-NO-cGMP pathway. However, the potency to induce this effect was much lower for PCB52 than for PCB138 or PCB180. PCB52 impairs the function of the pathway at micromolar concentrations (EC50 ) 0.3 µM). However, PCB138 or PCB180 induce this effect at very low concentrations. The EC50 was 2nM for both compounds (Figure 3). PCB52 and PCB138 Decrease and PCB180 Increases the Activation of sGC by NO. To assess activation of sGC by NO in intact neurons in culture, we measured the increase in cGMP induced by the addition of the NO-generating agent SNAP (10 µM). The addition of SNAP to control neurons activates sGC and increased cGMP to 897 ( 236% of basal levels. Chronic exposure to PCB52 or PCB138 reduced the increase in cGMP induced by SNAP in cerebellar neurons in culture (Figure 4), indicating that activation of guanylate cyclase by NO is impaired by these PCBs. PCB52 reduced activation of sGC at micromolar concentrations (EC50 ) 0.2 µM) and PCB138 at nanomolar concentrations (EC50 ) 10 nM) (Figure 4). In contrast, exposure to PCB180 enhances the SNAP-induced increase of cGMP, indicating that activation of guanylate cyclase

816

Chem. Res. Toxicol., Vol. 23, No. 4, 2010

Figure 5. Long-term exposure of cerebellar neurons in culture to PCB52, PCB180, or PCB138 alters basal levels of nitrites + nitrates. The indicated concentrations of PCB52, PCB180, or PCB138 were added to the culture medium 24 h after seeding. Experiments were performed 11-14 days later. Neurons were washed three times with prewarmed Locke’s solution without magnesium. The intracellular accumulation of nitrites + nitrates was measured as indicated in the Materials and Methods section. Values are the means ( SEMs of triplicate samples from seven different culture preparations. Values that are significantly different from control values are indicated by asterisks (*p < 0.05, and **p < 0.01).

by NO is increased after long-term exposure to this PCB congener (EC50 ) 0.1 µM) (Figure 4). Long-Term Exposure of Cerebellar Neurons in Culture to PCB52, PCB138, or PCB180 Increases the Basal Levels of NO. We also assessed the effects of in vitro exposure to PCBs on the basal levels of NO. As NO is a very unstable radical, with a very short half-life, it is not possible to get good measures of NO directly. NO is metabolized to nitrites and nitrates, which are stable. We therefore determined nitrites + nitrates (NOx) as a measure of NO. Exposure to any of these nondioxin-like (NDL)-PCBs significantly increases basal levels of NOx in cerebellar neurons in culture (Figure 5). PCB180 or PCB52 increases the basal level of NOx beginning at 0.1 µM (EC50 ) 10 nM for PCB180 and 20 nM and for PCB52). PCB138 is about 100-fold more potent and increases NOx levels with an EC50 ) 0.03 nM (Figure 5). Long-Term Exposure of Cerebellar Neurons in Culture to PCB52, PCB138, or PCB180 Reduces the Activation of NOS in Response to Activation of NMDA Receptors. To determine the effects of the PCBs on the activation of NOS in response to activation of NMDA receptors, we measured the increase in NOx induced by addition of 0.3 mM NMDA to the neurons for 5 min. In control neurons, the addition of NMDA activates the receptor and NOS and increased NOx to 194 ( 37% of basal values. PCB180 and PCB138 reduce the increase in NOx induced by NMDA (Figure 6). PCB180 is more potent to induce this effect, with an EC50 ) 0.8 nM, while the EC50 for PCB138 was ca. 100 nM. The effects of PCB52 depend on the concentration. PCB52 reduces a NMDA-induced increase of NOx at low concentrations (10-100 nM), does not affect it at 1-3 µM, and enhances a NMDA-induced increase of NOx at 10 µM (Figure 6). Long-Term Exposure of Cerebellar Neurons in Culture to PCB52, PCB138, or PCB180 Alters the Basal Concentration of Intracellular Calcium. The basal concentration of intracellular calcium was determined in cerebellar neurons using the Fura 2 fluorescent probe. In control neurons, the intracellular calcium concentration was 129 ( 20 nM (Figure 7). Exposure to PCB52 or PCB180 decreases the basal concentration of intracellular calcium. PCB180 induced this

Llansola et al.

Figure 6. Long-term exposure of cerebellar neurons in culture to PCB52, PCB180, or PCB138 impairs NMDA-induced increase of nitrites + nitrates. The indicated concentrations of PCB52, PCB180, or PCB138 were added to the culture medium 24 h after seeding. Experiments were performed 11-14 days later. Neurons were washed three times with prewarmed Locke’s solution without magnesium. A 0.3 mM concentration of NMDA was added over 5 min, and nitrites + nitrates were measured as indicated in the Materials and Methods section. Values are the means ( SEMs of triplicate samples from five different culture preparations. Values that are significantly different from control values are indicated by asterisks (*p < 0.05, **p < 0.01, and ***p < 0.001).

Figure 7. Long-term exposure of cerebellar neurons in culture to PCB52, PCB180, or PCB138 alters the basal level of intracellular calcium. The indicated concentrations of PCB52, PCB180, or PCB138 were added to the culture medium 24 h after seeding. Experiments were performed 11-14 days later. Neurons were washed three times with prewarmed Locke’s solution without magnesium. Neurons were washed three times with prewarmed Locke’s solution without magnesium and charged with the fluorescent probe for calcium, Fura-2, for 45 min. After 15 additional min, neurons were placed below fluorescence microscopy and basal fluorescence at 340 and 380 nm, and their ratio was registered. The intracellular level of calcium was calculated as indicated in the Materials and Methods section. Values are the means ( SEMs of triplicate samples from seven different culture preparations. Values that are significantly different from control values are indicated by asterisks (*p < 0.05, and **p < 0.01).

effect at concentrations beginning at 0.1 µM (EC50 ) 50 nM) and PCB52 at 1 µM (EC50 ) 0.7 µM) (Figure 7). PCB138 decreases the basal concentration of calcium by more than 50% at 1 nM. However, thereafter, the calcium concentration increased with the concentration of PCB138 added. PCB138 at 1 µM significantly increases the basal concentration of calcium (Figure 7). Long-Term Exposure of Cerebellar Neurons in Culture to PCB52, PCB138, or PCB180 Alters the Increase in Intracellular Calcium Induced by Activation of NMDA Receptors. The addition of 0.3 mM NMDA to control neurons activates the NMDA receptor and increases intracellular calcium to 668 ( 132% of basal value. Long-term exposure to PCB52 decreases the NMDA-induced increase in calcium only at high concentrations (10 µM) (Figure 8). PCB180 and PCB138 have

NDL-PCBs Impair Neuronal Signal Transduction

Figure 8. Long-term exposure of cerebellar neurons in culture to PCB52, PCB180, or PCB138 impairs NMDA-induced increase of calcium. The indicated concentrations of PCB52, PCB180, or PCB138 were added to the culture medium 24 h after seeding. Experiments were performed 11-14 days later. Neurons were washed three times with prewarmed Locke’s solution without magnesium and charged with the fluorescent probe for calcium, Fura-2, for 45 min. After 15 additional min, neurons were placed below fluorescence microscopy. After registration of basal fluorescence, 0.3 mM NMDA was added to perfusion system, and the intracellular concentration of calcium was measured as indicated in the Materials and Methods section. Values are the means ( SEMs of triplicate samples from five different culture preparations. Values that are significantly different from control values are indicated by asterisks (*p < 0.05, and **p < 0.01).

different effects at different concentrations. PCB180 reduces the NMDA-induced increase of intracellular calcium at 0.1 µM (EC50 ) 10 nM) but enhances the NMDA-induced increase in calcium at 1-10 µM (Figure 8). PCB138 strongly reduces the increase of intracellular calcium induced by NMDA at 1 nM concentration. However, thereafter, the NMDA-induced increase of calcium returns at similar levels than induced by NMDA without PCB138 (Figure 8). Effects of Long-Term Exposure to PCB52, PCB138, or PCB180 on the Amount of the Proteins Involved in the Glutamate-NO-cGMP Pathway in Cultured Neurons. It is possible that long-term exposure to NDL-PCBs affects the function of the glutamate-NO-cGMP pathway in cerebellar neurons in culture by altering the content of some of the proteins involved in this pathway. To assess this possibility, we determined the effects of exposure to different concentrations of PCB52, PCB138, or PCB180 on the amount of the NR1, NR2A, and NR2C subunits of the NMDA receptor, calmodulin, neuronal NOS (nNOS), and the R- and β-subunits of sGC. The results obtained are summarized in Table 1. Exposure to PCB52 did not induce relevant effects on the amount of any of these proteins, except a slight decrease of NR2A at 1 µM, but not at other concentrations. Exposure to PCB180 induced a slight increase (around 20%) of the NR1 subunit of NMDA receptors at any concentration between 10 nM and 10 µM. However, the amounts of the NR2C subunit of the same receptor and of calmodulin decrease progressively with the concentration of PCB180, reaching a 37 and 25% reduction, respectively, at 10 µM. Exposure to PCB138 induced a progressive and strong increase of the amount of the NR1 subunit of NMDA receptors, reaching 232% of control at 1 µM PCB138. PCB138 also increases the amount of calmodulin and slightly the amount of the NR2C subunit of the NMDA receptor (Table 1).

Discussion Long-term exposure of primary cultures of cerebellar neurons to the NDL-PCB52, -PCB138, or -PCB138 impairs the function of the glutamate-NO-cGMP pathway. PCB138 and PCB180

Chem. Res. Toxicol., Vol. 23, No. 4, 2010 817

areverypotentandimpairthefunctionoftheglutamate-NO-cGMP pathway at nanomolar concentrations, indicating that very low levels of these NDL-PCBs induce neurotoxic effects. These PCBs induce maximum effects in the range of 1-10 nM and do not show any dose response at higher concentrations, suggesting a saturation of the target of the PCBs. PCB52 is about 1000-fold less potent than PCB138 and PCB180 in impairing the function of the pathway. Micromolar concentrations of PCB52 are required to impair the pathway. One factor that could contribute to the different potency of the effects of the PCBs is their relative solubility in water. The aqueous solubilities of PCB52, PCB180, and PCB138 are 0.0153, 0.0039, and 0.0014 mg/L, respectively (NLM Chemical Database). This could contribute to the lower potency of PCB52 to affect the pathway. In this study, long-term exposure to PCBs covers all of the period of in vitro maturation of neurons; we cannot, therefore, completely exclude the possibility that the effects of PCBs on the glutamate-NO-cGMP pathway could be due to primary alterations in the in vitro development of cerebellar neurons or in the formation of synapses. However, we have no evidence of alterations in the in vitro development of cerebellar granule neurons exposed to the PCBs used. Moreover, the mechanisms by which different NDL-PCBs impair the function of the glutamate-NO-cGMP pathway are different. This cannot be due to the different solubility of the PCBs and must be due to other factors. Each NDL-PCB affects the glutamate-NO-cGMP pathway at more than one step. PCB52 reduces the basal concentration of intracellular calcium (at 1 µM), strongly increases the basal levels of NO (measured as nitrites + nitrates) (at 0.1-10 µM), and slightly increases the basal level of cGMP (at 1 µM). PCB52 reduces the activation of NOS by NMDA (at 0.1-3 µM), the activation of guanylate cyclase by NO (at 0.3-3 µM), the increase in calcium induced by NMDA (at 10 µM), and the whole glutamate-NO-cGMP pathway (at 1-10 µM). PCB180 reduces the basal concentration of intracellular calcium (at 0.1-10 µM), increases the basal levels of NO (measured as nitrites+nitrates) (at 0.1-10 µM), and slightly reduces the basal level of cGMP (at 0.1-10 µM). PCB180 reduces the activation of NOS by NMDA (at 1 nM to 3 µM), strongly increases the activation of guanylate cyclase by NO (at 10 nM to 1 µM), reduces the increase in calcium induced by NMDA at 0.1 µM, but increases it at 1-10 µM, and reduces the function of the whole glutamate-NO-cGMP pathway (at 3 nM to 10 µM). PCB138 increases the basal level of NO (measured as nitrites + nitrates) (at 0.1 nM to 10 µM) and strongly increases the basal level of cGMP (at 1 nM to 10 µM). PCB138 strongly reduces the basal concentration of intracellular calcium at 1 nM; at higher concentrations of PCB138, there is a progressive increase in basal calcium, in parallel with a progressive increase in the amount of NMDA receptors. There is a statistically significant correlation between the amount of the NR1 subunit of NMDA receptors and the concentration in calcium (Figure 9A). This suggests that PCB138 affects basal calcium by two mechanisms: One mechanism, acting at a very low (less than nanomolar) concentration, reduces calcium, and at higher concentrations, the increase in NMDA receptor amounts contributes to progressively increased basal calcium, which may be entering through the NMDA receptor channel. PCB138 reduces the activation of NOS by NMDA (at 0.1-10 µM), the activation of guanylate cyclase by NO (at 10 nM to 10 µM), and the whole glutamate-NO-cGMP pathway (at 10 nM to

818

Chem. Res. Toxicol., Vol. 23, No. 4, 2010

Llansola et al.

Table 1. Effects of Long-Term Exposure to PCB52, PCB180, or PCB138 on the Amount of the Proteins Involved in the Glutamate-NO-cGMP Pathway in Cultured Neuronsa µM

NR1

NR2A

NR2C

CM

NOS

GCR

GCβ

PCB52 0 1 3 10

100 ( 8 101 ( 6 109 ( 5 100 ( 4

100 ( 3 80 ( 8* 94 ( 9 104 ( 3

100 ( 6 82 ( 10 94 ( 8 128 ( 14

0 0.01 0.1 1 10

100 ( 6 112 ( 4* 123 ( 8* 124 ( 6* 126 ( 9*

100 ( 5 98 ( 4 105 ( 7 93 ( 14 96 ( 12

100 ( 5 85 ( 2* 81 ( 6* 78 ( 6** 63 ( 10*

0 0.0001 0.001 0.01 1 10

100 ( 8 116 ( 16 139 ( 16* 168 ( 20** 232 ( 13*** 161 ( 39

100 ( 10 96 ( 12 89 ( 12 119 ( 14 139 ( 11* 80 ( 14

100 ( 7 133 ( 6* 137 ( 14* 128 ( 10* 126 ( 28 126 ( 11

100 ( 7 106 ( 7 106 ( 9 104 ( 7

100 ( 4 85 ( 7 99 ( 5 104 ( 7

100 ( 7 89 ( 6 96 ( 4 98 ( 3

100 ( 4 89 ( 7 97 ( 9 110 ( 8

100 ( 5 87 ( 4* 88 ( 4* 73 ( 8* 75 ( 8*

100 ( 6 100 ( 5 108 ( 5 105 ( 8 114 ( 10

100 ( 7 93 ( 6 98 ( 4 82 ( 6* 103 ( 6

100 ( 8 92 ( 6 99 ( 4 101 ( 7 91 ( 5

100 ( 4 111 ( 17 128 ( 16 132 ( 8** 89 ( 11 100 ( 15

100 ( 2 107 ( 8 103 ( 6 105 ( 8 125 ( 16 125 ( 18

100 ( 4 103 ( 10 110 ( 6 101 ( 8 123 ( 9* 134 ( 8**

100 ( 4 102 ( 7 96 ( 6 100 ( 7 106 ( 6 99 ( 19

PCB180

PCB138

a Values are the means ( SEMs of seven different culture preparations and are expressed as a percentage of the amount in control neurons. Values significantly different from control neurons are indicated with asterisks: *p < 0.05, **p < 0.01, and ***p < 0.001. CM, calmodulin; NR1, NR2A, and NR2C, subunits of the NMDA receptor; GCR and -β, subunits of sGC and NOS.

Figure 9. There is a good correlation between the amount of the NR1 subunit of NMDA receptors and the basal or NMDA-induced intracellular concentration of calcium in neurons exposed chronically to PCB138. Values for the NMDA receptor amount are the same as reported in Table 1, and for basal calcium and a NMDA-induced increase in calcium, those are reported in Figures 7 and 8, respectively. The individual values are shown.

10 µM). PCB138 strongly reduces the increase in intracellular calcium induced by NMDA at 1 nM; at higher concentrations of PCB138, there is a progressive increase in calcium, in parallel with a progressive increase in the amount of NMDA receptors (Figure 9B). This suggests that PCB138 affects a NMDAinduced increase in calcium by two mechanisms: One mechanism, acting at a very low (nanomolar) concentration, reduces calcium, and at higher concentrations, the increase in NMDA

receptor amounts contributes to progressively increase the NMDA-induced increase in calcium. In cerebellar neurons in culture and in brain tissue, in vitro exposure to PCBs for short periods of time alters calcium homeostasis (15, 29-31). The results reported here show that long-term exposure to PCBs also affects calcium homeostasis under basal conditions and, in addition, alters the changes in calcium induced by activation of NMDA receptors. Taking into account the importance of calcium in neuronal function, this suggests that calcium homeostasis can be considered a main target for PCBs neurotoxicity. The increase in basal cGMP induced by PCB52 and PCB138 may contribute to their neurotoxicity. It has been shown that increased intracellular cGMP is neurotoxic in cerebellar neurons in culture (32). The above data clearly show that the three PCBs tested affect the glutamate-NO-cGMP pathway by different mechanisms. Some of the effects are similar for the three PCBs, while others are different. The three PCBs increase basal levels of nitrites + nitrates and reduce the increase in cGMP induced by NMDA. However, other effects are different for different PCBs. For example, PCB180 increases the activation of guanylate cyclase by NO, while PCB52 and PCB138 reduce it (Figure 4). This indicates that PCB180 has mechanisms of action that are not active for PCB52 or PCB138. Also, PCB52 and PCB138 induce different effects on other steps of the pathway. At high concentrations, PCB52 increases the NMDA-induced increase in nitrites, while PCB138 reduces it. Also, at high concentrations, PCB52 reduces basal calcium, while PCB138 does not. This clearly shows that the three PCBs affect the function of the pathway by different mechanisms. The different mechanisms of action cannot be due to different solubilities or toxicokinetics and must be mediated by the action on different targets. The effects of the PCBs on these different steps of the pathway could be due to a direct action of the PCBs on the specific target (e.g., a direct effect on guanylate cyclase) or indirect effects mediated by the action on other targets. For example, the effects of PCB138 on basal calcium could be a consequence of altered expression of NMDA receptors (see above). In any case, the final result is an impairment of the function of the glutamate-NO-cGMP pathway, which could

NDL-PCBs Impair Neuronal Signal Transduction

be involved in the cognitive impairment induced by developmental exposure to PCBs (15, 16). The reasons for the different mechanisms of action of the different PCBs are not clear by now. The three PCBs used in this work are noncoplanar diortho-substituted analogues; PCB52 is a tetra-chlorinated PCB, whereas PCB138 and PCB180 are hexa- and hepta-chlorinated PCBs, respectively. The different amounts of chloro subtitutions may be responsible for the 1000fold lower potency of this PCB on impairment of the glutamate-NO-cGMP pathway. PCB180 is classified in a group (33, 34) less potent than PCB138 in inducing hepatic microsomal aryl hydrocarbon hydroxylase and ethoxyresorufin O-deethylase activities and immunotoxicity. The structure of these compounds is therefore different and seems to have different initial targets, which can lead finally to the different effects on basal levels of cGMP, intracellular calcium, or NOS activity and also on the function of this pathway following activation of NMDA receptors or modulation of activation of sGC by NO. The results reported show that all PCBs induce, through different mechanisms, the same final effect: reduced function of the glutamate-NO-cGMP pathway. As commented in the Introduction,theimpairmentofthefunctionoftheglutamate-NO-cGMP pathway would be responsible for the impairment in at least some types of cognitive function in rats or humans developmentally exposed to PCBs. Acknowledgment. This work was upported by grants from the European Commission (FOOD-CT-2006-022923 ATHON), Ministerio de Ciencia e Innovacion (SAF2008-00062 and CSD2008-00005), and Ministerio de Sanidad (FIS 06/0065) of Spain and from Conselleria de Educacion de la Generalitat Valenciana (ACOMP/2009/191; ACOMP06/005 and ACOMP2009-025; and PROMETEO/2009/027) and AP005/06, EVES 034/2007, AP-024/08, and A-01/08 from Conselleria de Sanitat of Generalitat Valenciana.

References (1) Fangstrom, B., Strid, A., Grandjean, P., Weihe, P., and Bergman, A. (2005) A retrospective study of PBDEs and PCBs in human milk from the Faroe Islands. EnViron. Health 4, 12. (2) Inoue, K., Harada, K., Takenaka, K., Uehara, S., Kono, M., Shimizu, T., Takasuga, T., Senthilkumar, K., Yamashita, F., and Koizumi, A. (2006) Levels and concentration ratios of polychlorinated biphenyls and polybrominated diphenyl ethers in serum and breast milk in Japanese mothers. EnViron. Health Perspect. 114, 1179–1185. (3) Safe, S. H. (1994) Polychlorinated biphenyls (PCBs): Environmental impact, biochemical and toxic responses, and implications for risk assessment. Crit. ReV. Toxicol. 24 (2), 87–149. (4) Jacobson, J. L., and Jacobson, S. W. (1996) Intellectual impairment in children exposed to polychlorinated biphenyls in utero. N. Engl. J. Med. 335, 783–789. (5) Schantz, S. L., Seo, B. W., Moshtaghian, J., Peterson, R. E., and Moore, R. W. (1996) Effects of gestational and lactational exposure to TCDD or coplanar PCBs on spatial learning. Neurotoxicol. Teratol. 18, 305– 313. (6) Faroon, O., Jones, D., and de Rosa, C. (2001) Effects of polychlorinated biphenyls on the nervous system. Toxicol. Ind. Health 16 (7-8), 305– 333. (7) Landrigan, P. J. (2001) Pesticides and polychlorinated biphenyls (PCBs): An analysis of the evidence that they impair children’s neurobehavioral development. Mol. Genet. Metab. 73, 11–17. (8) Gray, K. A., Klebanoff, M. A., Brock, J. W., Zhou, H., Darden, R., Needham, L., and Longnecker, M. P. (2005) In utero exposure to background levels of polychlorinated biphenyls and cognitive functioning among school-age children. Am. J. Epidemiol. 162, 17–26. (9) Gore, A. C. (2001) Environmental toxicant effects on neuroendocrine function. Endocrine 14 (2), 235–246. (10) Angus, W. G., and Contreras, M. L. (1995) Aroclor 1254 alters the binding of 125I-labeled nerve growth factor in PC12 cells. Neurosci. Lett. 191 (1-2), 23–26.

Chem. Res. Toxicol., Vol. 23, No. 4, 2010 819 (11) Lein, P. J., Yang, D., Bachstetter, A. D., Tilson, H. A., Harry, G. J., Mervis, R. F., and Kodavanti, P. R. (2007) Ontogenetic alterations in molecular and structural correlates of dendritic growth after developmental exposure to polychlorinated biphenyls. EnViron. Health Perspect. 115 (4), 556–563. (12) Yang, D., Kim, K. H., Phimister, A., Bachstetter, A. D., Ward, T. R., Stackman, R. W., Mervis, R. F., Wisniewski, A. B., Klein, S. L., Kodavanti, P. R., Anderson, K. A., Wayman, G., Pessah, I. N., and Lein, P. J. (2009) Developmental exposure to polychlorinated biphenyls interferes with experience-dependent dendritic plasticity and ryanodine receptor expression in weanling rats. EnViron. Health Perspect. 117 (3), 426–435. (13) Basha, M. R., Braddy, N. S., Zawia, N. H., and Kodavanti, P. R. (2006) Ontogenetic alterations in prototypical transcription factors in the rat cerebellum and hippocampus following perinatal exposure to a commercial PCB mixture. Neurotoxicology 27 (1), 118–124. (14) Malkiewicz, K., Mohammed, R., Folkesson, R., Winblad, B., Szutowski, M., and Benedikz, E. (2006) Polychlorinated biphenyls alter expression of alpha-synuclein, synaptophysin and parkin in the rat brain. Toxicol. Lett. 161 (2), 152–158. (15) Piedrafita, B., Erceg, S., Cauli, O., Monfort, P., and Felipo, V. (2008) Developmental exposure to polychlorinated biphenyls PCB153 or PCB126 impairs learning ability in young but not in adult rats. Eur. J. Neurosci. 27, 177–182. (16) Piedrafita, B., Cauli, O., Montoliu, C., and Felipo, V. (2007) The function of the glutamate-nitric oxide -cGMP pathway in brain in vivo and learning ability decrease in parallel in mature compared to young rats. Learn. Mem. 14, 254–258. (17) Aguilar, M. A., Minarro, J., and Felipo, V. (2000) Chronic moderate hyperammonemia impairs active and passive avoidance behavior and conditional discrimination learning in rats. Exp. Neurol. 161, 704– 713. (18) Erceg, S., Monfort, P., Herna´ndez-Viadel, M., Llansola, M., Montoliu, C., and Felipo, V. (2005) Restoration of learning ability in hyperammonemic rats by increasing extracellular cGMP in brain. Brain Res. 1036, 115–121. (19) Monfort, P., Corbala´n, R., Martinez, L., Lo´pez-Talavera, J. C., Co´rdoba, J., and Felipo, V. (2001) Altered content and modulation of soluble guanylate cyclase in the cerebellum of rats with portacaval anastomosis. Neuroscience 104, 1119–1125. (20) Erceg, S., Monfort, P., Herna´ndez-Viadel, M., Rodrigo, R., Montoliu, C., and Felipo, V. (2005) Oral administration of sildenafil restores learning ability in rats with hyperammonemia and with portacaval shunt. Hepatology 45, 2–10. (21) Cauli, O., Rodrigo, R., Piedrafita, B., Boix, J., and Felipo, V. (2007) Inflammation and hepatic encephalopathy: Ibuprofen restores learning ability in rats with porto-caval shunts. Hepatology 46, 514–519. (22) Cauli, O., Mansouri, M. T., Agusti, A., and Felipo, V. (2009) Hyperammonemia increases GABAergic tone in the cerebellum but decreases it in the rat cortex. Gastroenterology 136 (4), 1359–1367. (23) Llansola, M., Hernandez-Viadel, M., Erceg, S., Montoliu, C., and Felipo, V. (2009) Increasing the function of the glutamate-nitric oxidecgmp pathway increases the ability to learn a y maze task. J. Neurosci. Res. 87 (10), 2351–2355. (24) Llansola, M., Piedrafita, B., Rodrigo, R., Montoliu, C., and Felipo, V. (2009) Polychlorinated biphenyls PCB 153 and PCB 126 impair the glutamate-nitric oxide-cGMP pathway in cerebellar neurons in culture by different mechanisms. Neurotoxic. Res. 16 (2), 97–105. (25) Minana, M. D., Montoliu, C., Llansola, M., Grisolia, S., and Felipo, V. (1998) Nicotine prevents glutamate-induced proteolysis of the microtubule-associated protein MAP-2 and glutamate neurotoxicity in primary cultures of cerebellar neurons. Neuropharmacology 37, 847– 857. (26) Felipo, V., Min˜ana, M. D., Cabedo, H., and Grisolı´a, S. (1994) L-carnitine increases the affinity of glutamate for quisqualate receptors and prevents glutamate neurotoxicity. Neurochem. Res. 19, 373–377. (27) Verdon, C. P., Burton, B. A., and Prior, R. L. (1995) Sample pretreatment with nitrate reductase and glucose-6-phosphate dehydrogenase quantitatively reduces nitrate while avoiding interference by NADP+ when the Griess reaction is used to assay for nitrite. Anal. Biochem. 224 (2), 502–508. (28) Corbala´n, R., Chatauret, N., Behrends, S., Butterworth, R. F., and Felipo, V. (2002) Region selective alterations of soluble guanylate cyclase content and modulation in brain of cirrhotic patients. Hepatology 36 (5), 1155–1162. (29) Inglefield, J. R., and Shafer, T. J. (2000) Polychlorinated biphenylstimulation of Ca(2+) oscillations in developing neocortical cells: a role for excitatory transmitters and L-type voltage-sensitive Ca(2+) channels. J. Pharmacol. Exp. Ther. 295 (1), 105–113. (30) Inglefield, J. R., Mundy, W. R., and Shafer, T. J. (2001) Inositol 1,4,5triphosphate receptor-sensitive Ca(2+) release, store-operated Ca(2+) entry, and cAMP responsive element binding protein phosphorylation

820

Chem. Res. Toxicol., Vol. 23, No. 4, 2010

in developing cortical cells following exposure to polychlorinated biphenyls. J. Pharmacol. Exp. Ther. 297 (2), 762–773. (31) Mundy, W. R., Shafer, T. J., Tilson, H. A., and Kodavanti, P. R. (1999) Extracellular calcium is required for the polychlorinated biphenylinduced increase of intracellular free calcium levels in cerebellar granule cell culture. Toxicology 136 (1), 27–39. (32) Montoliu, C., Llansola, M., Kosenko, E., Corbala´n, R., and Felipo, V. (1999) Role of cyclic GMP in glutamate neurotoxicity in primary cultures of cerebellar neurons. Neuropharmacology 38, 1883–1891.

Llansola et al. (33) Howie, L., Dickerson, R., Davis, D., and Safe, S. (1990) Immunosuppressive and monooxygenase induction activities of polychlorinated diphenyl ether congeners in C57BL/6N mice: quantitative structureactivity relationships. Toxicol. Appl. Pharmacol. 105 (2), 254–263. (34) Parkinson, A., and Safe, S. (1987) Mammalian biologic and toxic effects of PCBs. In Polychlorinated Biphenyls (PCBs): Mammalian and EnVironmental Toxicology (Sale, S., Ed.) Environmental Toxin Series 1, pp 49-75, Springer-Verlag, New York.

TX900440Q