Intraventricular infusion of quinolinic acid impairs spatial learning and memory in young rats: a novel mechanism of lead-induced neurotoxicity

Abstract

Background: Lead (Pb), a heavy metal, and quinolinic acid (QA), a metabolite of the kynurenine pathway of tryptophan metabolism, are known neurotoxicants. Both Pb and QA impair spatial learning and memory. Pb activates astrocytes and microglia, which in turn induce the synthesis of QA. We hypothesized increased QA production in response to Pb exposure as a novel mechanism of Pb-neurotoxicity.

Methods: Two experimental paradigms were used. In experiment one, Wistar rat pups were exposed to Pb via their dams' drinking water from postnatal day 1 to 21. Control group was given regular water. In the second protocol, QA (9 mM) or normal saline (as Vehicle Control) was infused into right lateral ventricle of 21-day old rats for 7 days using osmotic pumps. Learning and memory were assessed by Morris water maze test on postnatal day 30 or 45 in both Pb- and QA-exposed rats. QA levels in the Pb exposed rats were measured in blood by ELISA and in the brain by immunohistochemistry on postnatal days 45 and 60. Expression of various molecules involved in learning and memory was analyzed by Western blot. Means of control and experimental groups were compared with two-way repeated measure ANOVA (learning) and t test (all other variables).

Results: Pb exposure increased QA level in the blood (by ~ 58%) and increased (p < 0.05) the number of QA-immunoreactive cells in the cortex, and CA1, CA3 and dentate gyrus regions of the hippocampus, compared to control rats. In separate experiments, QA infusion impaired learning and short-term memory similar to Pb. PSD-95, PP1, and PP2A were decreased (p < 0.05) in the QA-infused rats, whereas tau phosphorylation was increased, compared to vehicle infused rats.

Conclusion: Putting together the results of the two experimental paradigms, we propose that increased QA production in response to Pb exposure is a novel mechanism of Pb-induced neurotoxicity.

Keywords: Lead; Memory; Neurotoxicity; Quinolinic acid; Spatial learning.

Fig. 1

Serum quinolinic acid levels (ng/mL) in Pb-exposed Wistar rats. Data is mean ± SD (n = 6 in each group); the group means were compared with Student’s t test for two independent samples with unequal variance. Note significant increase in QA in serum at PND60 compared to control, but not at PND45, (*p < 0.05)

Fig. 2

Number of quinolinic acid immunoreactive cells/field in cortex (a), CA1 (b), CA3 (c), dentate gyrus (d), and thalamus (e) regions in Pb-exposed and control groups at PND45 and PND60. Data were analyzed with one-way ANOVA with Bonferroni multiple comparisons test (n = 4 in all groups). Note Pb-exposed rats had significantly higher number of QA-immunoreactive cells in all regions studied at both PND45 and PND60 compared to control group except in thalamus at PND45 (* p < 0.05)

Fig. 3

Photomicrographs of cortex, CA1, CA3, dentate gyrus, and thalamus regions in Control and Pb-exposed groups immunostained for QA at PND45 and PND60. Note higher number of immunoreactive cells in the Pb-exposed group in all regions. Scale bar = 40 μm

Fig. 4

Escape latency of control and Pb-exposed rats (a and b), and vehicle control and QA-infused (c and d) in the Morris water maze: Learning sessions commenced at PND30 (a and c) and PND45 (b and d). a, b Note escape latency decreased from session to session both in control and Pb-exposed rats in PND30 and PND45 groups indicating that learning occurred. In the PND30 group, two-way repeated measures ANOVA revealed that the Pb-exposed groups learned significantly slower than the control group (F = 2.96; p = 0.034). In the PND45 group, the difference in the escape latencies across the sessions was not statistically significant (F = 2.12; p = 0.111). c, d Note escape latency decreased from session to session both in vehicle control and QA infused rat in PND30 and PND45 groups indicating that learning occurred. Two-way repeated measures ANOVA revealed that at PND30, the QA group learned significantly slower compared the vehicle control group (F = 3.52; p = 0.046). At PND45, the difference between VC and the QA groups was not statistically significant (F = 0.66; p = 0.527). e, f Representative video tracks of the probe test in the Morris water maze test of control and Pb-exposed (e) and vehicle control and QA-infused (f) rats. T, Target/platform quadrant

Fig. 5

Expression of NR1 (a) and NR2B (c) subunits of NMDAR and phosphorylation of NR1 (b), and NR2B (d) in the control and QA-infused rats at PND45 and PND60. NMDAR subunit signal was normalized to actin signal. For pNR1 (at serine 897) and pNR2B (at serine 1303), the signal was normalized to the unphosphorylated protein signal. Representative blots are shown above each graph. Data presented as mean ± SD (n = 4); mean data were compared with Student’s t test. Note significantly decreased expression of NR2B at PND60 (*p < 0.05)

Fig. 6

Expression of CREB (a), pCREB (b), synaptophysin (c), and PSD-45 (d) in the control and QA-infused rats at PND45 and PND60. All protein expressions were normalized to actin signal. Representative blots are shown above each graph. Data presented as mean ± SD (n = 4); mean data were compared with Student’s t test. Note significantly decreased expression of PSD-45 at PND45 (*p < 0.05)

Fig. 7

Expression of PP1 (a), PP2A (b), Tau (c), and phosphorylated tau at threonine 231 (d) in the control and QA-infused rats at PND45 and PND60. PP1A, PP2A, and Tau expression was normalized to actin signal; pTau at threonine 231 expression was normalized to Tau signal. Representative blots are shown above each graph. Data presented as mean ± SD (n = 4); mean data were compared with Student’s t test. Note significantly decreased expression of PP1 at PND60, decreased expression of PP2A at both PND45 and PND60, increased expression of AT180 at PND45 (*p < 0.05)

Fig. 8

Schematic representation of probable role of quinolinic acid in lead-neurotoxicity. Pb accumulation in the brain results in decreased NMDAR activity, which results in decreased LTP leading to impaired learning and memory (reviewed by Rahman, 2013 [1]). In addition, Pb induces astroglyosis and microglyosis [38]. Activated astrocytes and microglia produce inflammatory cytokines, which activates the KP, converting tryptophan into QA [11]. High levels of QA are toxic to astrocytes, microglia, and neurons [12, 14]. QA is an NMDAR agonist and causes excitotoxicity in neurons [10]. In addition, it increases synaptic glutamate levels, further increasing excitotoxicity [35]. This increased excitotoxicity in neurons, together with tau hyperphosphorylation [22, 98], eventually results in neurodegeneration. Up arrows indicated upregulation/increase; down arrows indicate downregulation/decrease