From the Cover: 7,8-Dihydroxyflavone Rescues Lead-Induced Impairment of Vesicular Release: A Novel Therapeutic Approach for Lead Intoxicated Children

Abstract

Childhood lead (Pb²⁺) intoxication is a public health problem of global proportion. Lead exposure during development produces multiple effects on the central nervous system including impaired synapse formation, altered synaptic plasticity, and learning deficits. In primary hippocampal neurons in culture and hippocampal slices, Pb²⁺ exposure inhibits vesicular release and reduces the number of fast-releasing sites, an effect associated with Pb²⁺ inhibition of NMDA receptor-mediated trans-synaptic Brain-Derived Neurotrophic Factor (BDNF) signaling. The objective of this study was to determine if activation of TrkB, the cognate receptor for BDNF, would rescue Pb²⁺-induced impairments of vesicular release. Rats were chronically exposed to Pb²⁺ prenatally and postnatally until 50 days of age. This chronic Pb²⁺ exposure paradigm enhanced paired-pulse facilitation of synaptic potentials in Schaffer collateral-CA1 synapses in the hippocampus, a phenomenon indicative of reduced vesicular release probability. Decreased vesicular release probability was confirmed by both mean-variance analysis and direct 2-photon imaging of vesicular release from hippocampal slices of rats exposed to Pb²⁺in vivo. We also found a Pb²⁺-induced impairment of calcium influx in Schaffer collateral-CA1 synaptic terminals. Intraperitoneal injections of Pb²⁺ rats with the TrkB receptor agonist 7,8-dihydroxyflavone (5 mg/kg) for 14-15 days starting at postnatal day 35, reversed all Pb²⁺-induced impairments of presynaptic transmitter release at Schaffer collateral-CA1 synapses. This study demonstrates for the first time that in vivo pharmacological activation of TrkB receptors by small molecules such as 7,8-dihydroxyflavone can reverse long-term effects of chronic Pb²⁺ exposure on presynaptic terminals, pointing to TrkB receptor activation as a promising therapeutic intervention in Pb²⁺-intoxicated children.

7,8-dihydroxyflavone (7,8-DHF), vesicular release, lead (Pb²⁺) neurotoxicity, brain derived neurotrophic factor (BDNF)

Childhood lead (Pb²⁺) intoxication is a significant public health problem in the United States and globally (Tong et al., 2000; Toscano and Guilarte, 2005). Recent episodes of Pb²⁺ exposure in children in communities like Flint, Michigan demonstrate the pervasive nature of the problem (Hanna-Atisha et al., 2016). The National Resources Defense Council reports that millions of Americans get drinking water from water systems that have Pb²⁺ violations and the problem could be much larger, because systems known to have violations do not show up in government databases that track such problems (www.nrdc.org/media/2016/160628; last Accessed October 9, 2017). Despite nearly a century of knowledge on the detrimental effects of Pb²⁺ in children’s development, the widespread presence of this neurotoxin in the global environment continues to affect children in the most vulnerable and economically disadvantaged segments of the population.

Studies have consistently demonstrated that one of the most prominent effects of chronic Pb²⁺ exposure in children is decreased capacity to learn, with devastating effects on cognitive and intellectual development (Bellinger et al., 1992; Canfield et al., 2003; Jusko et al., 2008; Lanphear et al., 2005; Schwartz et al., 2000), resulting in poor performance in school (Evens et al., 2015; Magzamen et al., 2013). Early life Pb²⁺ intoxication diminishes the intellectual capacity of children with tremendous cost to society. Several studies have shown that Pb²⁺ exposure in early life is associated with longitudinal declines in cognitive function (Stewart et al., 2006), loss of brain volume (Cecil et al., 2008; Rizzoli and Betz, 2005), and emergence of mental disorders such as major depression and schizophrenia (Bouchard et al., 2009; Guilarte et al., 2012) with aging.

Our laboratory has provided the first working model by which Pb²⁺ exposure during the period of synaptogenesis can affect synapse development and function, that accounts for both presynaptic and postsynaptic effects of Pb²⁺ on the synapse (Neal and Guilarte, 2010; Neal et al., 2010; Stansfield et al., 2012). Using a Pb²⁺ exposure paradigm during the period of synaptogenesis in primary hippocampal neurons, we found that Pb²⁺ inhibition of postsynaptic NMDA receptors (NMDARs) impairs CREB-dependent transcription of activity-regulated genes such as BDNF, and alters the function of its cognate receptor TrkB and downstream signaling (Neal and Guilarte, 2010; Neal et al., 2010; Stansfield et al., 2012). These studies also showed that Pb²⁺-induced impairment of BDNF trans-synaptic retrograde signaling decreases the presynaptic vesicular proteins synaptophysin and synaptobrevin and inhibits vesicular release (Neal et al., 2010). Importantly, the addition of exogenous BDNF to Pb²⁺-exposed hippocampal neurons during the last day of Pb²⁺ exposure, normalized synaptophysin and synaptobrevin levels and reversed the impairment in vesicular release, providing the first direct evidence of the beneficial effects of BDNF on Pb²⁺-induced synaptic dysfunction (Neal et al., 2010). Consistent with these observations using primary hippocampal neurons in culture, we subsequently showed that rats chronically exposed to Pb²⁺in vivo during development, expressed a marked impairment of hippocampal Schaffer-collateral-CA1 synaptic transmission due to inhibition of vesicular release measured by electrophysiological and 2-photon imaging in ex vivo hippocampal slices (Zhang et al., 2015).

In this study, we aimed to determine if pharmacological activation of the BDNF cognate receptor, TrkB, could rescue the Pb²⁺-induced deficits in vesicular release observed in Pb²⁺ exposed animals in vivo. For these studies, we used 7,8-dihydroxyflavone (7,8-DHF), a small, blood brain barrier permeant molecule from the flavonoid family that is a BDNF mimetic and selectively activates TrkB receptors resulting in sustained TrkB autophosphorylation, internalization, and increased downstream signaling (Liu et al., 2014, 2016). Previous work has shown that 7,8-DHF exhibits promising therapeutic efficacy in animal models of neurodegenerative diseases (Devi and Ohno, 2012; Jiang et al., 2013; Liu et al., 2016). Therefore, based on our previous in vitro studies demonstrating a beneficial effect of BDNF on Pb²⁺-induced inhibition of vesicular release, we hypothesized that 7,8-DHF could be useful in reversing the Pb²⁺-induced deficits in vesicular release that we have documented in our in vivo Pb²⁺ exposure rat model (Zhang et al., 2015). Here we demonstrate that 7,8-DHF completely rescues the inhibition of hippocampal Schaffer-collateral-CA1 vesicular release resulting from chronic developmental Pb²⁺ exposure.

MATERIALS AND METHODS

Chemicals

Chemicals for extra- and intracellular solutions were purchased from Sigma-Aldrich (St Louis, Missouri). Neurotransmitter receptor antagonists were purchased from Tocris Cookson (Minneapolis, Minnesota), and FM1-43 from Invitrogen (Grand Island, New York).

Blood Pb²⁺ analysis

Blood Pb²⁺ levels in samples from littermates were measured using the LeadCare system (Magellan Diagnostics, N. Billerica, Massachusetts).

Animals

Adult female Long-Evans rats (250 g) were purchased from Charles River, Inc. (Wilmington, Massachusetts) and randomly placed on diet containing 0 (control) or 1500 ppm lead acetate (PbAc) (Dyets, Bethlehem, Pennsylvania) 10 days prior to breeding with nonexposed Long-Evans males (300 g). Litters were culled to 10 pups on postnatal day 1 (PN1). Dams were maintained on their respective diet and at PN21 male pups were weaned onto the same diet and maintained until PN50. All rats are housed in plastic cages at 22°C ± 2 °C on a 12/12 light:dark cycle. Food and water were allowed ad libitum. Each litter is a single experimental unit for statistical purposes, so that for each experiment only 1 animal per litter was used for 1 data point. All studies were conducted in accordance with the U.S. Public Health Service’s Policy on Humane Care and Use of Laboratory Animals and the Society of Toxicology’s Guiding Principles in the Use of Animals in Toxicology under protocols approved by Institutional Animal Care and Use Committees from each university.

Hippocampal slice electrophysiology

Experiments were conducted as described previously in Stanton et al. (2003, 2005) and Zhang et al. (2015). At 50 ± 2 days of age, rats were deeply anesthetized with isoflurane, decapitated and their brains rapidly removed and submerged in ice-cold artificial cerebrospinal fluid (ACSF, 2°C–4 °C), containing (in mM): 124 NaCl, 4 KCl, 2 MgCl₂, 2 CaCl₂, 1.25 NaH₂PO₄, 26 NaHCO3, 10 glucose; at pH 7.4, gassed continuously with 95% O₂/5% CO₂). Brains were hemisected, the frontal lobes cut off, and individual hemispheres glued using cyanoacrylate adhesive onto a stage immersed in ice-cold ACSF gassed continuously with 95% O₂/5% CO₂ during slicing. We cut 400 µm thick coronal slices using a vibratome (Leica VT1200S, Leica Biosystems, Buffalo Grove, Illinois), and transferred them to an interface holding chamber for incubation at room temperature for a minimum of 1 hr before transferring to a submerged recording chamber on a Zeiss Axioskop microscope and continuously perfused at 3 ml/min with oxygenated ACSF at 32°C ± 0.5 °C.

Whole cell patch-clamp recordings were performed in CA1 pyramidal neurons using standard techniques. Patch pipettes (R = 3–4 MΩ) were filled with recording solution containing (in mM): 135 CsMeSO₃, 8 NaCl, 10 HEPES, 2 Mg-ATP, 0.3 Na-GTP, 0.5 EGTA, and 1 QX-314 (275 mOsm, pH 7.25 adjusted with Cs(OH)₂). Access resistance was carefully monitored, and only cells with stable access resistance (<5% change) were included in analyses. CA1 pyramidal cells were recorded under voltage clamp using a MultiClamp 700B (Axon Instruments, Union City, California) with Clampex (v9). Recording signals were filtered through an 8-pole Bessel low-pass filter with a 3 kHz cutoff frequency, digitized at 10 kHz, and sampled using Clampex (v9). Neurons were clamped at –60 mV, and Schaffer collateral-evoked excitatory postsynaptic currents (EPSCs) were delivered by a bipolar stimulating electrode (FHC, USA, Bowden, Maine, 50–100 pA, 100 µs duration). EPSC slopes were calculated by linear interpolation of the initial downward current from 20% to 80% of the maximum EPSC amplitude. Paired-pulse facilitation (PPF) was assessed by applying a pair of Schaffer collateral stimuli at intervals of 10–125 ms, and the ratio of slopes of the second to the first response was calculated, so that numbers >1.0 represented facilitation, <1.0 inhibition.

Vesicular release using FM1-43 fluorescence measurements

Fluorescence was visualized using a customized 2-photon laser-scanning Olympus BX61WI microscope with a 60×/0.90 W water immersion infrared objective lens and an Olympus multispectral confocal laser scan unit. The light source was a Mai-Tai laser (Solid-State Laser Co., Mountain View, California), tuned to 860 nm for exciting Magnesium Green and 820 nm for exciting FM1-43. Epifluorescence was detected with photomultiplier tubes of the confocal laser scan head with pinhole maximally opened and emission spectral window optimized for signal over background. In the transfluorescent pathway, a 565 nm dichroic mirror was used to separate green and red fluorescence to eliminate transmitted or reflected excitation light (Chroma Technology, Rockingham, Vermont). After confirming the presence of Schaffer collateral-evoked fEPSPs >1 mV in amplitude in CA1 stratum radiatum, and inducing long-term potentiation (LTP), 10 µM 6-cyano-7-nitroquinoxaline-2, 3-dione was bath-applied throughout the rest of the experiment to prevent synaptically driven action potentials in CA3 pyramidal neurons from accelerating dye release. Presynaptic boutons were loaded by bath-applying 5 µM FM1-43 (Molecular Probes, Eugene, Oregon) in hypertonic ACSF supplemented with sucrose to 800 mOsm for 25 sec to selectively load the RRP (Stanton et al., 2003, 2005), then returned to normal ACSF. Stimulus-induced destaining was measured after 30 min perfusion with dye-free ACSF, by bursts of 10 Hz bipolar stimuli (150 µs DC pulses) for 2 s applied once each 30 s. We fitted a single exponential to the first 6 fluorescence time course values, and decay time constants between groups compared by 2-tailed Student’s t-test, as we have shown previously that the early release reflects vesicular release from the RRP prior to recycling and reuse of vesicles (Stanton et al., 2003, 2005).

Presynaptic Ca²⁺ influx fluorescence measurements

Using established methods for measuring [Ca²⁺] transients (Regehr and Tank, 1991), we filled Schaffer collateral presynaptic fibers with Magnesium Green AM. Briefly, an ejection electrode (tip diameter, 5–10 µm) containing Magnesium Green AM (1 mM Magnesium Green AM, 10% DMSO, 1% pluronic acid in ACSF) was lowered into the Schaffer collateral pathway between the stimulating electrode and the presynaptic terminal field to be observed, air pressure pulses (6–9 psi, 100–200 ms) controlled by a Picospritzer (General Valve Corp. USA, Fairfield, New Jersey) were applied to the electrode until a small bright spot (≈10 mm in diameter) was observed. Thirty minutes elapsed to allow dye to sufficiently diffuse into presynaptic boutons prior to commencing imaging. To verify that magnesium green selectively loaded presynaptic terminals, FM4-64 was loaded with high [K⁺]o at the end of the experiment. To measure Ca²⁺ dynamics, stimulus-evoked fluorescence signals were collected by scanning at 200 Hz in surface-scanning mode (XYT). Baseline fluorescence (F₀) was averaged over 4 images, and ΔF/F calculated as (ΔF/F)_(t) = (F_[t]−F₀)/F₀.

Estimation of presynaptic release probability by variance-mean analysis

Variance-mean (VM) analysis according to a binomial model of synaptic transmission is a method that has been employed to study transmitter release at many synapses (Clements and Silver, 2000; Silver, 2003; Zhang et al., 2006). It is mainly applied to steady-state sequences of evoked EPSCs recorded under a variety of conditions by varying extracellular [Ca²⁺], or delivering long repetitive trains of stimulation of different frequencies, each resulting in a range of mean response size (Foster and Regehr, 2004; Oleskevich et al., 2000; Reid and Clements, 1999; Silver et al., 1998).

We used 3 ratios of [Ca²⁺]/[Mg²⁺] in ACSF (4/1, 2/2, and 1/4 mM) to alter release probability at Schaffer collateral synapses. Experiments began by establishing stable whole-cell recording from a CA1 pyramidal neuron, and then perfusing the slice with 4/1 [Ca²⁺]/[Mg²⁺] ACSF. Cells were voltage-clamped at −65 mV, and 100 µs constant-current stimulus pulses were delivered to Schaffer collateral/commissural fiber axons every 10 s to evoke an EPSC. Stable recordings for 8–10 min were made in 4/1 [Ca²⁺]/[Mg²⁺], before replacing the perfusate with 1/4 mM [Ca²⁺]/[Mg²⁺] ACSF. After EPSCs decreased in amplitude and restabilized, which usually took 5–8 min, EPSCs were recorded for an additional 8 min. Slices were then perfused with 2/2 mM [Ca²⁺]/[Mg²⁺] ACSF. After EPSC amplitudes had again stabilized, another 8 min of recordings were made. To induce long-term depression (LTD), slices were exposed to either 10 µM NMDA or 25 µM DHPG in 2/2 mM [Ca²⁺]/[Mg²⁺] ACSF for 3 or 5 min, respectively, durations which reliably induced LTD lasting hours. After drug exposure, slices were perfused with 2/2 mM [Ca²⁺]/[Mg²⁺] ACSF for >30 min, to verify expression of LTD, and then the same sequence of [Ca²⁺]/[Mg²⁺] ACSF applications was repeated. To ensure that postsynaptic AMPA receptors were responding to nonsaturating glutamate concentration, a requirement for VM analysis, experiments were performed in a low concentration of the AMPA receptor antagonist 6, 7-dinitroquinoxaline-2, 3-dione (100 nM).

7,8-DHF administration

7,8-DHF hydrate (Sigma-Aldrich, St Louis, Missouri) was dissolved in phosphate-buffered saline containing 17% dimethylsulfoxide (DMSO). Male rats received daily intraperitoneal injections of 5 mg/kg 7,8-DHF or 17% DMSO vehicle daily for 14–15 consecutive days starting when they were 35–42 days of age. Rats were sacrificed for slice preparation 24 h after the last 7,8-DHF administration.

Statistics

Power analysis showed a group size of 6 animals per treatment group with significance level preset to p < .05 could detect between group differences of 10% for vesicular transmitter release time constants and release probability at a power of 85% with typical parameter standard deviations. Data sets did not deviate significantly from normal distribution (D’Agostino-Pearson omnibus normality test), and did not exhibit significant differences in parameter variances (F-test). Slices and treatments were randomized, with treated and control slices examined in parallel on the same or sequential days. Although all analyses were automated, the investigator was not blinded to treatment group. Student’s t-test was used to determine differences between the control and Pb²⁺ treated groups for each particular measure. In analyses requiring comparisons between multiple groups, a 1-way ANOVA with Sidak’s Multiple Comparisons analysis was used with post hoc Tukey’s test for individual group comparisons. Significance level was preset to p < .05.

RESULTS

Blood Pb²⁺ Levels and Body Weight of Rats

The Pb²⁺ exposure paradigm did not produce any overt toxicity based on body weight gain. Body weights at PN50 rats were: 264.4 ± 11.9 g (n = 10) for control animals ± 7,8-DHF and 231.9 ± 10.0 g (n = 14) for Pb²⁺-exposed animals plus or minus 7,8-DHF (p > .05). Further, blood Pb²⁺ levels of littermates to animals used in this study at PN50 were: 0.6 ± 0.1 μg/dl (n = 67) for control animals and 22.2 ± 0.9 μg/dl (n = 47) for Pb²⁺-exposed animals. This exposure level is environmentally relevant and previous studies using this animal model have shown deficits in synaptic plasticity (Nihei et al., 2000), decreased adult neurogenesis (Verina et al., 2007), and impairments of spatial learning and contextual fear conditioning (Kuhlmann et al., 1997; McGlothan et al., 2008; Nihei et al., 2000).

7,8-DHF Reverses the Increase in PPF at Schaffer Collateral-CA1 Synapses Produced by Pb²⁺ Exposure

Neuronal short-term presynaptic plasticity is often assessed by delivering paired-pulse stimulation, that is, 2 stimuli to the same synaptic pathway in close succession (Andersen and Lømo, 1967; Zucker, 1989). One form of paired-pulse modulation, PPF, is typically attributed to an increase of release probability (Pr) during the second stimulus, arising from prior accumulation of residual Ca²⁺ near active zones and/or lingering effects of Ca²⁺ on a Ca²⁺ sensor (Neher, 1998; Zucker, 1989). This residual Ca²⁺, when present at terminals that fail to release on the first stimulus, will cause them to release and increase response amplitude from the second stimulus. Therefore, if initial Pr is reduced, as by manipulations such as reducing extracellular [Ca²⁺], the magnitude of PPF (the ratio of second to first response amplitude) should increase (Neher, 1998; Zucker, 1989).

In Figure 1A the black trace shows that PPF in CA1 pyramidal neurons was elicited by 2 Schaffer collateral stimuli applied 30 ms apart. The red trace (Figure 1A) shows the larger PPF typical of a CA1 pyramidal neuron in a slice from a Pb²⁺ rat (Pb²⁺/VEH) while the blue trace illustrates rescue of PPF in a slice from a Pb²⁺ rat that also received 7,8-DHF (Pb²⁺/7,8-DHF). Administration of 7,8-DHF to control animals (CON/7,8-DHF; gray trace) did not alter PPF. When paired-pulse stimuli were applied at intervals varying from 20-70 msec, PPF was significantly increased compared with slices from untreated control rats (1-way ANOVA (F[2, 29] = 9.786, p = .0006). PPF at a paired-pulse interval of 30 ms was significantly increased in slices from Pb²⁺-treated rats compared with slices from control rats (post hoc Tukey’s multiple comparison with Duncan’s correction: p = .008). Moreover, 7,8-DHF treatment of lead exposed rats significantly reduced PPF at this 30 ms interval, compared with rats exposed to lead alone (post hoc Tukey’s multiple comparison with Duncan′s correction: p = .029), or to control rats (post hoc Tukey’s multiple comparison with Duncan’s correction: p = .963), as showed in Figure 1C. One-way ANOVA with repeated measures demonstrated a statistically significant increase in PPF at all interpulse intervals (Figure 1B, red circles; p = .0064) in lead treated rats compared with controls, while 7,8-DHF administration completely rescued PPF across the entire paired-pulse profile (Figure 1B, blue circles).

Figure 1.

7,8-DHF reverses the increase in PPF produced by Pb²⁺ exposure at Schaffer collateral-CA1 synapses in rat hippocampus. A, Representative EPSCs in field CA1 in response to Schaffer collateral paired-pulse stimuli at a 30 ms interstimulus interval (ISI) in slices from Control (CON/VEH; black trace), Pb²⁺-treated (Pb²⁺/VEH; red trace) and Pb²⁺ + 7,8-DHF-treated rats (Pb²⁺/7,8-DHF; blue trace), illustrating the ability of 7,8-DHF to reverse Pb²⁺-induced increases in PPF. B, Mean ± SEM EPSC PPF P2/P1 ratio as a function of ISI, where PPF was significantly enhanced for ISI 20–70 ms in slices from Pb²⁺-treated (filled red circles, n = 9 slices) versus controls (open black circles, n = 13 slices), and this effect was rescued by treatment of Pb²⁺ rats with 7,8-DHF (filled blue circles, n = 10 slices) (p < .05). C, Mean ± SEM ratio of P2/P1 (30 ms ISI) at Schaffer collateral-CA1 synapses in slices from control (open bar), Pb²⁺-treated (red bar), and Pb²⁺ plus 7,8-DHF-treated rats (blue bar), showing that increased PPF in Pb²⁺-treated rats (p < .05) was rescued by 7,8-DHF co-administration.

7,8-DHF Reverses the Impairment in Vesicular Release From the Rapidly Recycling Vesicle Pool Produced by Pb²⁺ Exposure

To directly determine whether presynaptic vesicular release is altered by in vivo Pb²⁺ exposure, we used 2-photon excitation to visualize release of the styryl dye FM1-43 from the rapidly recycling pool (RRP) of presynaptic vesicles after selective loading by hypertonic shock in Schaffer collateral-CA1 terminals in hippocampal slices. Presynaptic vesicles in the RRP were first stimulated by a brief hypertonic shock to fuse with the membrane and release their transmitter, which induces them to take up FM1-43 from the extracellular space, followed by endocytosis and recycling back into the RRP for the next evoked release. We have used this method previously to show that generation of LTP and LTD is associated with persistent increases (Stanton et al., 2005) or decreases (Stanton et al., 2003) in the rate of stimulus-evoked FM1-43 de-staining at Schaffer collateral terminals, and that chronic early life Pb²⁺ exposure persistently reduces vesicular release from the RRP (Zhang et al., 2015).

Figure 2 illustrates the effect of Pb²⁺ exposure on vesicular release from Schaffer collateral presynaptic terminals. Figure 2A shows representative pseudocolor images of FM1-43 labeled Schaffer collateral terminals before (Baseline) and after 12 min of 2 Hz stimulation in control slices from rats that received daily injections of vehicle (CON/VEH) or 7,8-DHF (CON/7,8-DHF), versus slices from Pb²⁺ rats with vehicle (Pb²⁺/VEH) or Pb²⁺ rats that received 7,8-DHF (Pb²⁺/7,8-DHF). The slice from the Pb²⁺-exposed rat showed markedly slower stimulus-evoked de-staining compared with the control slice, while rapid de-staining was restored in the slice from the Pb²⁺ rat that received 7,8-DHF. Figure 2B summarizes the time courses of all slices, showing the markedly slower vesicular release evoked by 2 Hz stimulation of Schaffer collateral terminals in field CA1 of slices from Pb²⁺ rats (red filled circles) compared with controls treated with vehicle (black open circles) or controls treated with 7,8-DHF (gray solid circles), and the rescue of this effect in slices from Pb²⁺ rats treated with 7,8-DHF (blue solid circles). Statistics with 1-way ANOVA (F[3, 102] = 50.73, p < .0001) on the initial time constant of release calculated from a single exponential fit of the first 6 times points (Stanton et al., 2003, 2005) exhibited a significant slower decay constant in rats exposed to Pb²⁺ alone (red bar, p = .0001), compared with control rats treated with vehicle (open bar), 7,8-DHF alone (gray bar, p = .504), or Pb²⁺-exposed rats also given 7,8-DHF (blue bar, p = .711).

Figure 2.

Two-photon laser scanning microscopic (TPLSM) images of FM1-43 vesicular release from Schaffer collateral terminals in field CA1 of hippocampal slices show that Pb²⁺-induced persistent reduction in release probability is rescued by 7,8-DHF. A, Representative TPLSM pseudocolor images of FM1-43 loaded presynaptic terminals in stratum radiatum of field CA1 in hippocampal slices from a Pb²⁺ rat (Pb²⁺/VEH), versus a rat treated with Pb²⁺ plus 7,8-DHF (Pb²⁺/7,8-DHF), and one treated with DHF alone (CON/7,8-DHF), or control vehicle (CON/VEH) imaged prior to (Baseline) and after 12 min 2 Hz Schaffer collateral stimulation (Calibration Bar: 5 µm). (B) Time course (Mean ± SEM) of stimulus-evoked FM1-43 de-staining from puncta in field CA1 of hippocampal slices in response to 2 Hz Schaffer collateral stimulation in slices from control rats (open circles, n = 8 slices, 35 puncta) versus Pb²⁺-treated (red circles, n = 6 slices, 30 puncta), and Pb²⁺ plus 7,8-DHF-treated rats (gray diamonds, n = 8 slices, 36 puncta). C, Mean ± SEM of initial fluorescence decay time constant in slices from Pb²⁺-treated rats (Pb²⁺/VEH), versus rats treated with Pb²⁺ plus 7,8-DHF (Pb²⁺/7,8-DHF), and control rats treated with 7,8-DHF alone (CON/7,8-DHF) or vehicle (CON/VEH). All slices from Pb²⁺ rats showed significantly slower de-staining of Schaffer collateral terminals (p < .05) compared with control slices, and this reduction was completely rescued by 7,8-DHF.

7,8-DHF Reverses Pb²⁺-Induced Reductions in Presynaptic Schaffer Collateral Release Probability Measured by VM Analysis

VM analysis using a binomial model of synaptic transmission has been employed to study neurotransmitter release probability at a variety of synapses (Clements et al., 2000; Silver, 2003). It is typically applied to steady-state sequences of single evoked EPSCs recorded while varying extracellular [Ca²⁺], or delivering long repetitive trains of stimulation of different frequencies, each resulting in a range of mean response size with variance that is a parabolic function of Pr (Foster and Regehr, 2004; Oleskevich et al., 2000; Reid and Clements, 1999; Silver et al., 1998). In this method, low extracellular [Ca²⁺] yields low Pr, release failures and low EPSC variance, high extracellular [Ca²⁺] yields high Pr, few failures and, again, low EPSC variance, and physiological extracellular [Ca²⁺] yields intermediate Pr and higher EPSC variance. We have applied this method to directly estimate presynaptic Pr at Schaffer collateral-CA1 synapses, comparing normal slices to slices from Pb²⁺-exposed rats using the identical protocol as in this study (Zhang et al., 2015).

Figure 3A demonstrates that the VM relationship obtained by varying extracellular [Ca²⁺] was parabolic, with maximum variance at the peak of the parabola. In pyramidal hippocampal neurons from Pb²⁺ rats (Pb²⁺/VEH), individual slice data point (Figure 3A, red circles) and mean amplitudes (Figure 3B, red circles) at different [Ca²⁺], converted to Pr, were reduced along the same parabolic fit at all 3 [Ca²⁺], consistent with a reduction in presynaptic release probability compared with CON/VEH/7,8-DHF (black open circles). It should be noted that the CON/VEH/7,8-DHF is the combined data from CON/VEH and CON/7,8-DHF animals since there was no significant differences between these 2 groups. The groups were combined in order to make panel A and B graphs easier to understand. The actual data for each group is provided in Table 1.

Table 1.

Presynaptic Release Probability Estimated From VM Analysis as a Function of Ratio of [Ca²⁺]/[Mg²⁺]

Ca2+/Mg2+	CON/VEH		CON/7,8-DHF		Pb2+/VEH		Pb2+/7,8-DHF
	Mean ± SD	n	Mean ± SD	n	Mean ± SD	n	Mean ± SD	N
1/4	0.050 ± 0.009	8	0.049 ± 0.002	3	0.037 ± 0.003a	8	0.058 ± 0.022	8
2/1	0.451 ± 0.053	8	0.445 ± 0.052	3	0.288 ± 0.075a	8	0.480 ± 0.058	8
4/1	0.726 ± 0.073	8	0.753 ± 0.098	3	0.603 ± 0.067a	8	0.764 ± 0.070	8

n, number of slices.

a

p < .05 relative to all other treatment groups, Student’s t-test with Bonferroni correction.

Figure 3.

Chronic Pb²⁺ exposure is associated with reduced presynaptic vesicular release probability at Schaffer collateral-CA1 terminals assessed by VM analysis. A, Individual VM data points, corrected to estimate P_r, at each [Ca²⁺]_o for each CON/VEH/7,8-DHF (open black circles), each slice from a Pb²⁺ rat (Pb²⁺/VEH; open red circles), and each slice from a Pb²⁺ rat administered 7,8-DHF (Pb²⁺/7,8-DHF; blue circles). Data from all groups of slices were well fit by a single parabola forced to pass through 0, 0 with Pb²⁺ synapses shifted to the left, consistent with a presynaptic reduction in P_r from chronic Pb²⁺ exposure. This shift was rescued by 7,8-DHF. B, Mean ± SEM of VM points in slices from CON/VEH/7,8-DHF (black circles; n = 11), Pb²⁺/VEH (red circles; n = 8), and Pb²⁺/7,8-DHF (blue circles; n = 8), normalized to the maximal peak amplitude recorded at 4 mM [Ca²⁺]_o.

Across all experiments (Table 1), Pr calculated by this method was significantly reduced in slices from Pb²⁺/VEH rats at low (1/4 mM, p = .007), medium (2/2 mM, p = .005) and high (4/1 mM, p = .005) [Ca²⁺]/[Mg²⁺] ratios (1-way ANOVA with repeated measures (F[8, 69] = 25.14, p = .001). Figure 4 shows a VM versus mean linear plot, where the line fit from pyramidal neurons from a CON/VEH (black dotted line) versus a Pb²⁺/VEH (red dotted line) rat significantly differed in slope (p = .015), consistent with a presynaptic site of reduced Pr. This shift in slope was rescued in a slice from a Pb²⁺ rat treated with 7,8-DHF (Figure 4, blue dotted line). Again, the CON/7,8-DHF data was not different from CON/VEH and it was not included to simplify the graph, but it is provided in Table 2. These shifts in Pr were not associated with significant changes in number of release sites (N) or quantal size (Q) across all slices (Table 2).

Table 2.

Numbers of Release Sites and Quantal Amplitude Estimated by VM Analysis at Schaffer Collateral Synapses in Field CA1

	CON/VEH		CON/7,8-DHF		Pb2+/VEH		Pb2+/7,8-DHF
	Mean ± SD	n	Mean ± SD	n	Mean ± SD	n	Mean ± SD	n
No. Release	144 ± 57	8	108 ± 60	8	130 ± 53	8	126 ± 58	8
Quantal Amp	2.63 ± 0.48	8	2.56 ± 0.93	8	2.53 ± 0.58	8	2.43 ± 0.79	8

n, number of slices.

Figure 4.

Plot of VM ratio versus mean EPSC amplitude (pA) from a single representative slice, which converts the parabolic relationship between mean and variance to a linear one. The number of release sites (n) was derived by estimating the slope of the linear fit, while the y-intercept denotes quantal size (Q) of the EPSC. The reduction in slope indicates that chronic Pb²⁺ exposure (Pb²⁺/VEH; red dotted line) was associated with a reduction in presynaptic P_r compared with a control slice (CON/VEH; dotted black line), that was partially reversed in a slice from a Pb²⁺-exposed rat administered 7,8-DHF (Pb²⁺/7,8-DHF; dotted blue line).

7,8-DHF Reverses Pb²⁺-Induced Reductions in Presynaptic Calcium Influx Into Schaffer Collateral Terminals

Calcium channels (P/Q and N-type) are the major source of action potential mediated Ca²⁺ influx into presynaptic terminals. Previous studies have shown that Pb²⁺ inhibits calcium channels in cultured cells, an effect that is reversible by washing the cellular preparation (Peng et al., 2002). If Pb²⁺ exposure chronically alters the activity of these channels, this could indirectly contribute to alterations in Pr. To directly test whether chronic Pb²⁺ exposure produces a persistent inhibition of presynaptic Ca²⁺ influx, and to determine if 7,8-DHF can reverse such effects, we injected Mg²⁺ Green-AM, a calcium indicator dye that is membrane-permeable (Brustein et al., 2003), directly into stratum radiatum of field CA1 of hippocampal slices. Mg²⁺ Green positive fluorescent puncta were visualized in field CA1 using 2-photon excitation. Figure 5 demonstrates the kinetics of Mg²⁺ Green fluorescence increases in response to a 100 Hz burst of 4 Schaffer collateral stimuli. We have shown previously that these responses persist in the presence of NMDA and AMPA receptor antagonists, despite the loss of fEPSPs, are blocked by cadmium and omega conotoxin, and co-localize with FM4-64, confirming a presynaptic nature for these calcium transients (Zhang et al., 2011).

Figure 5.

7,8-DHF rescues Pb²⁺-induced reductions in presynaptic Ca²⁺ influx into Schaffer collateral terminals. A, Representative fluorescent transients evoked by Schaffer collateral stimulation in single presynaptic terminals of a Control slice (CON/VEH; black trace), a control slice with 7,8-DHF (CON/7,8-DHF; gray trace), a slice from a Pb²⁺ rat (Pb²⁺/VEH; red trace), and a slice from a Pb²⁺ rat administered 7,8-DHF (Pb²⁺/7,8-DHF; blue trace). B, Mean ± SEM presynaptic stimulus-evoked Mg²⁺-Green fluorescence increases in presynaptic terminals in response to a burst of Schaffer collateral stimuli (4 × 20 Hz) in slices from control (CON/VEH, n = 8 slices, 16 terminals) versus Pb²⁺-exposed (Pb²⁺/VEH, n = 8 slices, 14 terminals) rats, and slices from Pb²⁺-exposed rats co-administered 7,8-DHF (Pb²⁺/7,8-DHF, n = 8 slices, 16 terminals). Ca²⁺ influx transients were significantly smaller in terminals of Pb²⁺-exposed rat slices compared with either controls or 7,8-DHF rescued slices (p < .05).

Comparison of mean fluorescence increases of representative stimulus-evoked presynaptic Ca²⁺ influx transients in Schaffer collateral terminals in slices from vehicle control (CON/VEH; Figure 5a, black trace) or control treated with 7,8-DHF (CON/7,8-DHF; Figure 5A, gray trace), versus Pb²⁺ rats (Pb²⁺/VEH; Figure 5A, red trace) and Pb²⁺ rats treated with 7,8-DHF (Pb²⁺/7,8-DHF; Figure 5a, blue trace), revealed that action potential-dependent Ca²⁺ influx was reduced in amplitude by Pb²⁺ exposure, and that 7,8-DHF was able to reverse this reduction in presynaptic Ca²⁺ influx. Figure 5B summarizes these results across all slices, showing that Schaffer collateral presynaptic terminals in hippocampal slices from Pb²⁺ rats (red bar) exhibited reduced Ca²⁺ influx (1-way ANOVA with repeated measures F(3, 28) = 5.233, p = .0054) compared with vehicle control slices (black bar, p = .0129), 7,8-DHF-treated slices (gray bar, p = .0192), or slices from rats exposed to Pb²⁺ plus 7,8-DHF injections (blue bar, p = .0269). Taken together, our data indicate that chronic exposure to Pb²⁺ during development results in a persistent reduction in presynaptic Pr that may be due to both reduced Ca²⁺ influx, and actions downstream of presynaptic Ca²⁺ influx at the level of SNARE protein-mediated exocytosis (Neal and Guilarte, 2010; Neal et al., 2010; Stansfield et al., 2012). Consistent with its effects in rescuing Pr, daily injection of 7,8-DHF was also able to rescue the effects of Pb²⁺ exposure in reducing presynaptic Ca²⁺ influx at Schaffer collateral terminals in the hippocampus.

DISCUSSION

This study shows that the BDNF mimetic, 7,8-DHF, reverses the synaptic dysfunction resulting from chronic developmental Pb²⁺ intoxication in a preclinical animal model. 7,8-DHF is a naturally occurring small molecule in the flavonoid family of polyphenolic compounds found in Godmania aesculifoloia, Tridax procumbens, and primula tree leaves (Bhutia and Valant-Vetschera, 2012; Taddei and Rosas-Romero, 2000). It has been shown to have no toxic effects and to have neuroprotective properties in preclinical studies (Jang et al., 2010; Zhang et al., 2014) and has been proposed as a pro-neurotrophic treatment for neurodevelopment disorders (Du and Hill, 2015).

In this study, we show that daily postnatal intraperitoneal administration of 7,8-DHF to rats chronically exposed to Pb²⁺ during development, completely rescued Pb²⁺-induced impairments in vesicular release and presynaptic Ca²⁺ influx, supporting the use of 7,8-DHF as a novel therapeutic agent for the treatment of cognitive deficits resulting from early life Pb²⁺ exposure. However, further studies are needed to determine whether the rescue of transmitter release by 7,8-DHF is long-lasting or permanent. Furthermore, while 7,8-DHF has been previously shown to ameliorate cognitive function deficits in several preclinical animal models (Devi and Ohno, 2012; Tan et al., 2016; Yang et al., 2014; Zhang et al., 2014), we still need to determine if it is able to rescue behavioral impairments associated with chronic developmental exposure to Pb²⁺.

Although this work did not focus on the potential beneficial effects of 7,8-DHF on learning deficits induced by chronic developmental Pb²⁺ exposure, previous work from our laboratory has shown that environmental enrichment is not only able to reverse Pb²⁺-induced learning deficits but it also enhanced BDNF gene expression in the hippocampus indicating an association between increased BDNF expression in the hippocampus and reversal of Pb²⁺-induced learning deficits (Guilarte et al., 2003). This previous work and subsequent studies indicating a dysfunctional BDNF-TrkB system in primary hippocampal neurons exposed to Pb²⁺ in culture (Neal et al., 2010; Stansfield et al., 2012) and in the hippocampus of rats exposed to Pb²⁺in vivo (Zhang et al., 2015), formed the basis for the current study. Future studies will aim to determine if 7,8-DHF can also reverse the Pb²⁺-induced learning deficits that we have previously documented in this rodent model of chronic developmental Pb²⁺ exposure (Kuhlmann et al., 1997; Nihei et al., 2000; Guilarte et al., 2003).

The mechanism of action of 7,8-DHF has been extensively studied. Seminal studies indicate a selective and long-lasting activation of the TrkB receptor by 7,8-DHF. Liu et al. (2014) have shown the specific interaction of 7,8-DHF to the extracellular domain of the TrkB receptor with low nanomolar affinity (10 nM). They also show that 7,8-DHF produces TrkB receptor phosphorylation and internalization that is sustained relative to BDNF. Further, TrkB-ECD-Fc recombinant proteins suppress the agonist effect of 7,8-DHF confirming that the TrkB receptor is the specific target for 7,8-DHF. These findings are supported by other studies as reviewed in Liu et al. (2016) and Moosavi et al. (2016).

Previous studies from our laboratory on the effects of Pb²⁺ exposure in TrkB receptor phosphorylation, show a decrease in TrkB phosphorylation at Tyr816 (Stansfield et al., 2012) altering synapsin I phosphorylation, which leads to inhibition of vesicular release. Published studies indicate that treatment with 7,8-DHF increases Tyr816 phosphorylation of TrkB receptors (Liu et al., 2014; Luo et al., 2016; Garcia-Diaz Barriga et al., 2017), therefore, we believe that the beneficial effects that we have observed in this study are directly relevant to the direct effects of 7,8-DHF on TrkB activation. Further support for the beneficial effects of 7,8-DHF on TrkB activation are supported by the fact that the TrkB receptor antagonist K252a ameliorates the beneficial effects of 7,8-DHF (Agrawal et al., 2015).

Previous studies using acute exposure of cultured cells to Pb²⁺ have shown inhibition of Ca²⁺ channels, an effect that is reversed upon washout of Pb²⁺ from the cells (Peng et al., 2002). However, using a chronic Pb²⁺ exposure protocol in primary hippocampal neurons results in a persistent impairment of vesicular release including changes in the levels of SNARE proteins (Neal and Guilarte, 2010; Neal et al., 2010). In this study, fluorescent imaging of presynaptic Ca²⁺ influx showed that Pb²⁺ exposure was associated with reductions in voltage-dependent Ca²⁺ channel-mediated Ca²⁺ entry that were completely reversed by 7,8-DHF. Previously, we found that presynaptic Ca²⁺ fluorescent signals evoked by brief 20 Hz bursts of stimulation showed only a small, early reduction in amplitude of presynaptic Ca²⁺ signals in slices from Pb²⁺ rats (Zhang et al., 2015), leading us to use higher frequency 100 Hz bursts of stimulation in this study. Our present findings suggest that developmental Pb²⁺ exposure can persistently impair presynaptic Ca²⁺ entry, and have additional downstream effects at the level of vesicular SNARE protein-mediated docking, recycling, and long-term stability of the release complex, consistent with our previous findings in hippocampal neuronal cultures (Neal et al., 2010), and hippocampal slices from Pb²⁺ rats (Zhang et al., 2015).

In this study, rats were exposed to Pb²⁺ chronically during gestation, postnatally and continuing through to young adulthood when vesicular release was assessed. Notably, the administration of 7,8-DHF occurred while the rats were being exposed to Pb²⁺ indicating that the beneficial effects of 7,8-DHF on vesicular release occur in the presence of sustained Pb²⁺ exposure. It should be noted that the ACSF used to maintain hippocampal slice viability during all experiments did not contain Pb²⁺, indicating that the effects observed on vesicular release were the result of the in vivo Pb²⁺ exposure.

Studies from our laboratory have previously shown that BDNF synthesis and release are decreased in cultured hippocampal neurons exposed to Pb²⁺, and are associated with reductions in levels of SNARE proteins and inhibition of vesicular release (Neal et al., 2010). These effects of in vitro Pb²⁺ exposure, were rescued by exogenous BDNF, consistent with our present findings. Stansfield et al. (2012), using the same Pb²⁺ exposure paradigm in cultured neurons, showed that Pb²⁺ may impair the transport of BDNF-containing vesicles, possibly by altering Huntingtin phosphorylation at a site promoting anterograde BDNF vesicle movement. This effect of Pb²⁺ resulted in impaired BDNF release, decreasing TrkB activation, and phosphorylation of synapsin I. Our current findings further support the hypothesis that BDNF receptor agonists and treatments such as enriched environments that increase BDNF levels and release (Ickes et al., 2000; Rossi et al., 2006), may be able to rescue the effects of chronic Pb²⁺ exposure we observed in hippocampal synaptic networks. Further, previous studies from our laboratory have shown that environmental enrichment can reverse Pb²⁺-induced impairments of spatial learning in rats of similar age and Pb²⁺ treatment as in this study (Guilarte et al., 2003). That study also showed that Pb²⁺-exposed rats placed in an enriched environment that reverses learning deficits, also exhibit increased BDNF gene expression in the hippocampus (Guilarte et al., 2003), supporting our current data implicating the BDNF-TrkB system in Pb²⁺ neurotoxicity.

Collectively, our previous and present studies indicate that chronic developmental Pb²⁺ exposure inhibits NMDAR mediated trans-synaptic BNDF signaling inhibiting vesicular release resulting in impaired synapse formation. We now show that these effects of early life Pb²⁺ exposure can be rescued by the BDNF mimetic 7,8-DHF via activation of the TrkB receptor. We propose that 7,8-DHF may be an effective and safe naturally occurring substance that may benefit the cognitive and intellectual development of Pb²⁺-intoxicated children.

FUNDING

This work was supported by the National Institute of Environmental Health Sciences (RO1 ES006189 and RO1 ES020465 to T.R.G.).