|Home | About | Journals | Submit | Contact Us | Français|
Further understanding of how prefrontal cortex (PFC) circuit change during postnatal development is of great interest due to its role in working memory and decision-making, two cognitive abilities that are refined late in adolescence and become altered in schizophrenia. While it is evident that dopamine facilitation of glutamate responses occurs during adolescence in the PFC, little is known about the cellular mechanisms that support these changes. Among them, a developmental facilitation of postsynaptic Ca2+ function is of particular interest given its role in coordinating neuronal ensembles, a process thought to contribute to maturation of PFC function. Here we conducted whole-cell patch clamp recordings of deep-layer pyramidal neurons in PFC brain slices and determined how somatic-evoked Ca2+-mediated plateau depolarizations change throughout postnatal day (PD) 25 (juvenile) to adulthood (PD 80). Postsynaptic Ca2+ potentials in the PFC increase in duration throughout postnatal development. A remarkable shift from short to prolonged depolarizations was observed after PD 40. This change is reflected by an enhancement of L-type Ca2+ channel function and postsynaptic PKA signaling. We speculate that such a protracted developmental facilitation of Ca2+ response in the PFC may contribute to improvement of working memory performance through adolescence.
Growing evidence indicates that adolescence is a vulnerable period for onset of some major psychiatric disorders such as schizophrenia and substance abuse (Andersen, 2003, Chambers et al., 2003, Lewis et al., 2004, Lewis and Gonzalez-Burgos, 2006). It is therefore of translational value to gain detailed knowledge of the developmental changes of neural systems and neuromodulators implicated in these neuropsychiatric conditions. Among these, the prefrontal cortex (PFC) is implicated as converging preclinical and clinical findings stressing major deficits in PFC function in schizophrenia (Lewis and Gonzalez-Burgos, 2006), such as working memory and decision-making. Typically, these prefrontal cognitive abilities are refined during adolescence and young adulthood (Casey et al., 2000, Spear, 2000, Luna et al., 2004, Segalowitz and Davies, 2004, Bunge and Wright, 2007), and depend on mesocortical dopamine (Seamans et al., 1998, Goldman-Rakic et al., 2000, Horvitz, 2000, Floresco and Phillips, 2001, Baldwin et al., 2002, Schultz, 2002).
At the cellular level, dopamine action in the PFC also increases during adolescence (Benes et al., 2000, Tseng and O’Donnell, 2005b, Tseng et al., 2006, Tseng et al., 2007, Tseng and O’Donnell, 2007b, Tseng and O’Donnell, 2007a) in a manner that correlates with the delayed acquisition of adult levels of prefrontal dopamine receptors (Verney et al., 1982, Leslie et al., 1991, Rosenberg and Lewis, 1994, Rosenberg and Lewis, 1995, Tarazi et al., 1999, Tarazi and Baldessarini, 2000) and NMDA receptor subunits (Williams et al., 1993, Monyer et al., 1994). Furthermore, studies in rodents and primates have highlighted a widespread structural and molecular refinement in the PFC at late stages of postnatal development (Woo et al., 1997, Casey et al., 2000, Spear, 2000, Gogtay et al., 2004, Pantelis et al., 2005), typically during adolescence, when dramatic processes of synaptic remodeling take place in cortical circuits (Andersen, 2003). Thus, while it is evident that dopamine regulation of glutamate action increases in the PFC during adolescence (Tseng et al., 2005, Tseng and O’Donnell, 2007b, Tseng and O’Donnell, 2007a), little is known about the cellular mechanisms that underlie this facilitation. Here, we assessed how postsynaptic Ca2+ function in PFC pyramidal neurons changes from the juvenile period to adulthood by determining the relative contribution of L-type Ca2+ channel and postsynaptic protein kinase A (PKA) signaling in the regulation of somatic-evoked Ca2+ potentials. L-type Ca2+ channels and PKA signaling are of particular interest given their role in enhancing PFC responses to dopamine (Tseng and O’Donnell, 2004) and in sustaining plateau depolarization in prefrontal output neurons (Tseng and O’Donnell, 2005b). These are related events thought to contribute to maturation of PFC function (O’Donnell and Tseng, 2009) by enabling PFC ensembles to fire synchronously (O’Donnell, 2003, Tseng and O’Donnell, 2005a).
All experimental procedures were performed with the approval of the Rosalind Franklin University of Medicine and Science IACUC, the University of Maryland Baltimore School of Medicine IACUC and in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. All chemicals and drugs were obtained from Sigma-Aldrich and Tocris and they were mixed into oxygenated recording aCSF solution in known concentrations. The PKA inhibitor PKI[5–24] peptide was purchased from Calbiochem.
Brain slices were obtained from male Sprague Dawley rats (Harlan, Madison, WI) housed in groups of 2–4 rats per cage and maintained on a 12 h light/dark cycle with food and tap water available ad libitum. As previously reported (Tseng and O’Donnell 2004,Tseng and O’Donnell 2005), rats were anesthetized with chloral hydrate (400 mg/kg, i.p.) before being decapitated. Brains were rapidly removed into ice-cold artificial cerebrospinal fluid (aCSF) containing (in mM): 125 NaCl, 2.5 KCl, 25 NaHCO3, 1.25 NaH2PO4, 1.0 MgCl2, 2.0 CaCl2, 12.5 glucose, 2.5 kynurenic acid and 0.004 SR-95531 (pH 7.40-7.43, 295-305 mOsm). Coronal slices (300 μm thick) containing infralimbic and prelimbic regions of the medial PFC were obtained with a Vibratome (Leica 1000s, Nussloch GmbH, Germany) in ice-cold aCSF, incubated at room temperature aCSF constantly oxygenated with 95% O2–5% CO2 for at least 60 minutes before recording.
All recordings were conducted at 33–35° C and the recording aCSF was delivered to the recording chamber at the rate of ~2 ml/min. In the recording aCSF, 0.5 μM tetrodotoxin (TTX) and 20 mM tetraethylammonium (TEA) chloride were included to block Na+ and K+ channels, respectively. Patch electrodes (5–8 MΩ) were obtained from 1.5 mm borosilicate glass capillaries (WPI) with a horizontal puller (P-97, Sutter Instrument Co., Novata, CA) and filled with a solution containing 0.125% Neurobiotin and (in mM): 120 CsCl, 10 HEPES, 2.0 MgCl2, 3.0 Na2-ATP, 0.3 Na2-GTP (pH 7.23–7.28, 280–282 mOsm). Medial PFC pyramidal neurons from layers V-VI were identified under visual guidance using infrared-differential interference contrast (IR-DIC) video microscopy with a 40x water-immersion objective (Eclipse E600-FN, Nikon, NY). The image was detected with an IR-sensitive CCD camera (DAGE-MTI) and displayed on a monitor. Whole cell patch-clamp recordings were performed with Axopatch 200B and MultiClamp 700B amplifier (Axon Instruments/Molecular Devices, Sunnyvale, CA), digitized (Digidata 1322A and 1440; Axon Instruments), and acquired with Clampex 8.2 (Axon Instrument) at a sampling rate of 10 KHz. The liquid junction potential was not corrected and electrode potentials were adjusted to zero before obtaining the whole cell configuration. Membrane potential was held at −70 mV in current-clamp mode with DC current throughout the experiment. Ten minutes after obtaining the whole-cell configuration, depolarizing current pulses of 40 ms duration were applied through the recording electrode until plateau potentials were obtained. Suprathreshold current steps (i.e., + 20 pA above the threshold current) were delivered every 2 min while monitoring the input resistance of the neuron with 200 ms hyperpolarizing pulses of 40 pA amplitude. Only neurons that remained stable for at least 30 min after obtaining the whole-cell configuration were included.
The effect of dopamine D1 receptor activation on pyramidal neuron excitability was conducted in TTX- and TEA-free recording aCSF and conventional K+-gluconate-containing (120 mM) patch electrodes as reported previously (Tseng and O’Donnell, 2004, Tseng et al., 2007). Briefly, in each cell, input resistance (measured by hyperpolarizing pulses), membrane potential, the number of evoked spikes, and the latency to the first spike evoked by a 500 ms duration depolarizing current pulse were compared before and after drug administration to the bath solution. Typically, baseline recordings were conducted for 15 min before perfusing with a solution containing the different mixture of drugs for 10 min. All neurons included in the present study showed a similar pattern of spike response and all excitability data comparison were performed from the stable 10 min period of drug administration.
After completion of the recording session, the slices were fixed with 10% formalin overnight at 4 °C and stored in 0.1 M phosphate buffer (PB) until staining. After a series of rinses in 0.1 M PBS, slices were incubated in 3% bovine serum albumin and 2% Triton-X 100 in PBS for 1 hour followed by overnight in Vectastain Elite ABC reagent (Vector Laboratories, Burlingame, CA) at 4°C. Following another series of rinses, slices were reacted with 3,3′ diaminobenzidine and urea-hydrogen peroxide (Sigma FAST DAB set). Slices were then rinsed, mounted on gelatin-coated slides, air-dried for 20 min, cleared in xylene, coverslipped in Permount and examined on a microscope.
Subjects were nine male Sprague-Dawley rats, examined at postnatal day 20, day 30, or day 56 (n=3 per age). They were maintained on a 12hour/12hour light/dark cycle with lights on at 7 a.m.; food and water were available ad libitum. All animals were sacrificed via rapid decapitation and whole brains were removed and immersed in 100 mL of Golgi-Cox solution. Tissue histology followed a procedure described in detail elsewhere (Markham and Juraska, 2002, Markham et al., 2005). Briefly, brains were dehydrated and embedded in a solution of 10% celloiden prior to coronal sectioning at a thickness of 150 μm using a sliding microtome. Free-floating sections were developed according to the procedure described by Glaser and Van der Loos (1981) and mounted onto slides. For each animal, the apical dendritic trees of five neurons with cell bodies residing in layer V of the prelimbic cortex were drawn in their entirety. Only neurons that met the following criteria were selected for measurement: 1) displayed morphological characteristics of a pyramidal neuron, 2) was completely filled with the stain, 3) was not obscured by another cell, and 4) no major dendritic branches were truncated by the section. Apical dendrites were drawn at a magnification of 625X with the aid of a camera lucida (drawing tube) attached to a microscope. Estimates of dendritic complexity were obtained by Sholl ring analysis (Sholl, 1956). For this method, a grid of concentric rings (spaced 20 μm apart) is placed over the camera lucida drawing of the dendritic field, with the center of the concentric circles on the soma, and the number of intersections of the dendrites with the grid rings is counted.
The PKA assay was conducted following the same experimental procedure described in Ford et al 2009 (Ford et al., 2009). Briefly, the medial PFC (bregma +3.2 to +2.7 mm) was dissected and lysed in hypotonic buffer solution containing (in mM; pH 7.9) 10 HEPES, 1.5 MgCl2, 10 KCl, 1 DTT, 0.025 (2)-p-bromotetramisole oxalate, 0.005 cantharidin, 0.005 microcystin-LF, 0.005 cyclosporin A, and supplemented with Complete Protease Inhibitor tablets (Roche Diagnostics). PKA activity in the PFC was determined by estimating the amount of phosphorylated Leu-Arg-Arg-Ala-Ser-Leu-Gly (Kemptide) peptide substrate using the PepTag PKA assay (Promega). In the present study, we used 2 μg of PFC tissue and the amount of phosphorylated kemptide tagged with a UV-fluorescent dye migrating toward the anode was determined. The purified PKA catalytic subunit (10 ng) was used as positive controls, whereas negative controls contained no PKA (data not shown; see Ford et al., 2009).
Overall measures are expressed as mean ± SEM. Drug effects were compared using Student’s t-test or repeated measures ANOVA, and the differences between experimental conditions were considered statistically significant when P<0.05. If data were not normally distributed or had unequal variances, Kruskal-Wallis ANOVA by ranks was preferred for multiple comparisons involving interrelated proportions. Normality and homogeneity of variances were estimated with the Kolmogorov-Smirnov and Levene’s tests, respectively.
Whole-cell current-clamp recordings of pyramidal neurons from layer V of the medial PFC (prelimbic and infralimbic regions) (Gabbott et al., 2005) were conducted using visual guidance under differential interface contrast in the infrared range. All neurons included in the present study (n=175) were further identified as deep-layer pyramidal cells located in the medial PFC by means of Neurobiotin staining.
All recordings were conducted in the presence of K+ and Na+ channel blockers (see Methods section for details) while holding the membrane potential at −70 mV. Under this experimental condition, a typical long-lasting Ca2+ plateau potential can be elicited with a brief, 40 ms depolarizing somatic current step (Figure 1). In agreement with previous reports (Hu et al., 2004, Nasif et al., 2005), bath application of cadmium (CdCl2 400 μM, 5–10 min) distinctly blocked the plateaus in all cell tested (n=8, Figure 1A, B). Thus, the long-lasting somatic-evoked response is dependent on Ca2+ influx.
Somatic-evoked Ca2+ plateau potentials recorded from deep-layer pyramidal neurons increased in duration (measured at the half decay amplitude) from postnatal day (PD) 25 to adulthood (PD>60) (Figure 2). Visual inspection of all data points indicated that PFC plateau duration becomes remarkably prolonged after puberty (PD>40) and throughout adulthood (Figure 2A). While all pyramidal cells exhibited plateau potentials >3.3 s duration after PD 40, only 23% (7/30) of neurons recorded from the prepubertal (PD 25–40) PFC showed prolonged plateaus similar to that in adult rats. The majority of cells (23/30; 77%) in the prepubertal PFC exhibited plateaus <3.3 s of length (0.5–3.1s range) (Figure 2A). To enable statistical analysis, we grouped the data into three developmental groups (Figure 2B): (i) PD 25–40 (n=30); (ii) PD 41–55 (n=17); (iii) PD 56–80 (n=9). No apparent differences in membrane input resistance (PD 25–40: 269.2 ± 16.2 MΩ; PD 41–55: 250.2 ± 18.6 MΩ; PD 56–80: 274.3 ± 34.7 MΩ; mean ± SEM) and current intensity used to elicit plateau potentials (PD 25–40: 246.0 ± 9.8 pA; PD 41–55: 252.3 ± 11.1 pA; PD 56–80: 246.1 ± 10.2 pA) were observed across age groups. Plateau potentials recorded from PD 25–40 PFC neurons were 2.09 ± 0.27 s whereas after PD 40, a shift from short to prolonged plateaus was observed in all neurons sampled (Figure 2A, B). The mean duration of plateau potentials in the PD 41–55 age group was 4.79 ± 0.28 s (n=17; P<0.0002, compared with PD 25–40, Kruskall-Wallis ANOVA). No further increase in plateau duration was found in neurons recorded from PD 56–80 rats (n=9; 4.92 ± 0.38 s; Figure 2B). Together, these results indicate that postsynaptic Ca2+ function in the PFC is developmentally regulated in a manner that correlates with the delayed normal maturation of prefrontal circuits, which typically occurs during the adolescent/postpubertal transition period (Tseng and O’Donnell, 2005b).
We then examined the possibility that the developmental facilitation of Ca2+ plateau potentials observed in the PFC is related to an age-dependent increase in apical dendritic length (Dunia et al., 1996) and/or complexity. Morphological analyses of all recorded cells labeled with Neurobiotin revealed that the overall length of layer V pyramidal neuron apical dendrite is not correlated with the duration of plateau potentials (Figure 3A, B). We next investigated whether the degree of dendritic complexity is different across age in a second cohort of tissue that was exclusively processed for histological measures. Using the Golgi-Cox staining technique, we found an age-dependent increase in apical dendritic complexity of layer V pyramidal neurons in the prelimbic cortex (F2,6=10.1, P<0.01), which occurred during the PD 20 to PD 30 transition period (P<0.01), and not between PD 30 to PD 56 (Figure 3C). Analysis of dendritic ramification as a function of distance from soma revealed the dendritic regions that contributed to the age-dependent increase in overall complexity (distance: F4,24=27.9, P<0.00000001; age x distance interaction F8,24=2.5, P<0.04) (Figure 3D). Dramatic age-dependent ramification was largely restricted to the apical tufts between days 20 and 30 (P<0.04) whereas the middle portions of the tree followed a more modest pace (120–200 μm day 20 < day 56 P<0.01; 220–300 μm day 20 < day 56 P<0.05) (Figure 3D). Together, these results indicate that it is unlikely that changes in overall dendritic length and complexity could account for the developmental facilitation of prefrontal Ca2+ plateau potentials observed after PD 40.
Activation of L-type Ca2+ channels has been shown to mediate somatic-evoked Ca2+ potentials (Hu et al., 2004, Nasif et al., 2005) as well as plateau depolarizations in the PFC (Tseng and O’Donnell, 2005b). We therefore examined the effect of the L-type Ca2+ antagonist nifedipine (10 μM) in a subset of cells exhibiting prolonged plateau depolarizations (Figure 4A). Bath application of nifedipine shortened the plateaus from 4.67 ± 0.38 s (baseline) to 1.04 ± 0.19 s in a time dependent manner (n=13, P<0.0001, repeated measure ANOVA; Figure 4A, B). This effect was observed in all cells tested with long-lasting potentials irrespective of the age group (Figure 4C). The same effect was observed in a subset of pyramidal neurons recorded from the prepubertal PFC exhibiting short plateaus. However, nifedipine exerted a weaker inhibition on short potentials (−32.42 ± 9.56%; n=6, open circles) than those long-lasting plateaus (−74.58 ± 3.49%, P<0.001, unpaired t-Test) (Figure 4C), suggesting a major contribution of L-type Ca2+ channels in sustaining plateau potentials longer in duration. Further analyses revealed a significant correlation between baseline plateau duration and the level of nifedipine-sensitive component of the plateau (correlation coefficient r = 0.95, P<0.0001; Figure 4D). This latter was calculated by subtracting the total (baseline) plateau potential to that left after 15 min of nifedipine (baseline – plateau potential after 15 min of nifedipine). No correlation was observed with the remaining nifedipine-insensitive component of the response (correlation coefficient r = 0.05, P=0.8; Figure 4D, inset). These results indicate that an enhancement of postsynaptic L-type Ca2+ channel function prolongs plateau potentials in PFC pyramidal neurons. Such a facilitation of L-type Ca2+ function is likely to be developmentally regulated as only 23% of the cells sampled in the prepubertal (PD<40) PFC showed long-lasting plateaus (Figure 2).
Several postsynaptic signaling mechanisms have been implicated in the regulation of L-type Ca2+ channel function (De Jongh et al., 1996, Dai et al., 2009). In particular, a role of PKA signaling in mediating D1-NMDA-driven/L-type Ca2+ channel-dependent plateau depolarizations has been observed in the medial PFC, especially in pyramidal neurons recorded from postpubertal rats (Tseng and O’Donnell, 2005b). We therefore examined how PKA signaling blockade impacts the duration of long-lasting plateaus in a subset of PFC pyramidal neurons recorded from PD 56–80. To selectively target postsynaptic PKA, recordings were conducted with patch electrodes containing the PKA inhibitor PKI[5–24] peptide (Tseng and O’Donnell, 2004). Typically, in the absence of the PKA inhibitor (control, n=15) or in the presence of heat-inactivated PKI[5–24] (20 μM, n=6), plateau potentials remained long-lasting even after 30 min of whole-cell recording (Figure 5A). In contrast, loading an effective dose of PKI[5–24] (20 μM, n=12) (Tseng and O’Donnell, 2004) into the recording pipette elicited a profound and progressive shortening of the plateaus during the first 20 min of obtaining the whole-cell configuration, after which a maximum effect was observed (Figure 5A, C). The mean plateau potential duration observed after 30 min of recording with PKI-containing electrodes was 1.19 ± 0.13 s. Bath application of 10 μM nifedipine further reduced the plateaus (Figure 5B, C), suggesting that L-type Ca2+ channel function is not blocked by postsynaptic PKA inhibition. However, the effect of nifedipine was markedly attenuated by PKI[5–24] as PKA blockade alone shortened the plateaus by ~70% relative to control recordings (Figure 5D). A further ~10% reduction of PKI[5–24]-induced short-lasting plateau potentials was observed with the addition of nifedipine. In the absence of postsynaptic PKA inhibition, however, nifedipine alone reduced the duration of long-lasting plateaus by ~80% (Figure 4A & 5D’). Together, the data suggest that postsynaptic PKA signaling is necessary for sustaining L-type Ca2+-mediated plateau potentials of long duration, especially in pyramidal neurons recorded from the postpubertal PFC.
We next asked the question whether postsynaptic PKA contributes to sustaining Ca2+ potentials observed in the prepubertal/juvenile (PD<35) PFC. The impact of PKA signaling blockade was assessed in pyramidal neurons exhibiting both long- and short-lasting plateaus using patch electrodes containing the PKA inhibitor PKI[5–24] peptide (20 μM) as described above. Neurons exhibiting short and long-lasting plateaus can be easily distinguished after 5 min of obtaining the whole-cell configuration due to the slow onset of PKI-induced attenuation of Ca2+ potentials (Figure 6A). We found that Ca2+ plateau depolarizations recorded from the juvenile (PD<35) PFC are also sensitive to postsynaptic PKA inhibition. Interestingly, the magnitude of PKI-induced attenuation is markedly higher in cells showing long-lasting plateaus (n=5) than those exhibiting short-duration potentials (n=12) (Figure 6A, B). Further analyses revealed a significant correlation between baseline potential duration (measured at 5 min of whole-cell) and the PKA-sensitive component of the plateau (baseline plateau potential after 30 min of PKI) (correlation coefficient r = 0.96, P<0.0005; Figure 6C). In addition, PKA inhibition-induced attenuation of long-duration Ca2+ potentials observed in the juvenile PFC resembles to that obtained in adults (Figure 6D). Together, these results indicate that the long duration Ca2+ potentials observed in the prepubertal/juvenile PFC is regulated by the same mechanism as those in adults.
It is very likely that PKA activity in the medial PFC is also developmentally regulated. To address this possibility, a PKA assay was used (Ford et al., 2009) (see methods for details) to determine the amount of PKA activity present in the PFC of pre (PD 32–37) and postpubertal (PD 42–60) rats. Relative to the prepubertal level, a significant 40% increase of PKA activity was found in the postpubertal PFC (n=8 per age group; Figure 7A). We next asked whether postsynaptic activation of PKA is sufficient to shift the characteristic prepubertal short-lasting plateaus into prolonged depolarizations. Towards this goal, we examined the impact of forskolin in a subset of prepubertal (PD 25–30) PFC pyramidal neurons (n=8) exhibiting plateau potentials of ~2 s in duration. Bath application of forskolin prolonged the plateaus from 1.80 ± 0.39 s to 3.28 ± 0.24 s in a dose-dependent manner (Figure 7B). A significant increase in plateau duration was obtained with 10 μM and 50 μM forskolin. However, forskolin failed to prolong the plateaus to the postpubertal level. These results, taken together with those obtained with PKI-containing electrodes (Figures 5 & 6), indicate that a developmental facilitation of postsynaptic PKA signaling is necessary, but not sufficient, to sustain the prolonged plateau depolarizations observed in the adult PFC.
To determine whether such a developmental enhancement of PKA-dependent facilitation of L-type Ca2+ function is directly link to a relevant neurotransmitter, we examined how these developmental changes modulate pyramidal neuron excitability in response to dopamine D1 receptor activation. D1 receptor-induced changes in neuronal excitability were measured in current clamp mode using conventional aCSF (see Material & Methods for details) and by assessing the number of action potentials evoked by a constant-amplitude intracellular current step before and after bath application of an effective dose of the D1 agonist SKF38393 (Tseng and O’Donnell, 2004, Tseng et al., 2007). SKF38393 (10 μM, 10 min) significantly increased the excitability of prefrontal pyramidal neurons, an effect that was completely blocked by the D1 antagonist SCH 23390 (10 μM) (Figure 8). Relative to the response observed in the PFC of PD 25–40 rats, activation of D1 receptors elicited a significantly larger excitability increase in pyramidal neurons recorded from the PD 56–80 age group (79.4 ± 24.2 % vs. 160.0 ± 21.1 %, P=0.008, Mann-Whitney U Test; PD 25–40 (n=8) vs. PD 56–80 (n=10), respectively). Such a D1 action was significantly attenuated when the recordings were conducted in the presence of PKI[5–24] (20 μM in the recording electrode) or nifedipine (10 μM in bath solution) (Figure 8). Together, these results suggest that activation of L-type Ca2+ channels and PKA signaling are required for the enhanced D1 facilitation of pyramidal neuron excitability observed in the PFC of PD 56–80 rats.
In the present study we found that postsynaptic Ca2+ potentials in PFC pyramidal neurons increase in duration throughout postnatal development. A significant shift from short to prolonged plateaus was observed after PD 40. Our electrophysiological, pharmacological and biochemical results converge to indicate that PKA activity is developmentally regulated in a manner that is required to support the gradually-enhanced postsynaptic L-type Ca2+ function observed in the adult PFC. Importantly, this characteristic maturational profile emerges during the adolescent/postpubertal transition period and is required to sustain the enhanced D1 receptor-mediated response observed in the adult PFC.
It is well known that neuronal L-type Ca2+ channels are largely confined in soma and dendrites, and contribute about 50% of the total voltage-gated Ca2+ current in cortical pyramidal neurons recorded from juvenile/prepubertal animals (Mintz et al., 1992, Hell et al., 1993, Simon et al., 2003). A wealth of data also indicate that activation of L-type Ca2+ channels is critical for a variety of activity-dependent processes that regulate synaptic strength and long-term changes in intrinsic and synaptic excitability (Brosenitsch et al., 1998, Hardingham et al., 1998, Kapur et al., 1998, Graef et al., 1999, Weisskopf et al., 1999, Mermelstein et al., 2000, Moosmang et al., 2005). In addition, Ca2+ entry through L-type channels is known to increase short-term synaptic integration (Kim and Connors, 1993, Schiller et al., 1997, Magee et al., 1998, Larkum et al., 1999) and to sustain synaptically-driven NMDA-dependent depolarizations (Hernandez-Lopez et al., 1997, Cepeda et al., 1998, Dudek and Fields, 2001, Tseng and O’Donnell, 2005b). In the PFC, it has been proposed that such a Ca2+-dependent control of glutamatergic drive may be critical for establishing proper temporal coordination between neuronal ensembles, which in turn may be relevant for the production of specific goal-directed behaviors (Tseng and O’Donnell, 2005b, Tseng and O’Donnell, 2005a). Thus, a developmental facilitation of L-type Ca2+-mediated plateau potentials may contribute to maturation of prefrontal network functioning during adolescence (Casey et al., 2000, Spear, 2000, Luna et al., 2004, Segalowitz and Davies, 2004, Bunge and Wright, 2007).
Because the significant age-dependent increases in apical dendritic complexity are between PD 20 to PD 30, and not between PD 30 to PD 56, this factor is unlikely to account for the facilitation of Ca2+ plateaus observed in the PFC after PD 40. Instead, our data highlight that a critical level of postsynaptic PKA is required for sustaining L-type Ca2+ plateaus longer in duration. This is consistent with the well-established facilitatory role of PKA signaling on L-type Ca2+ channel activity, which is mediated by increased phosphorylation of the Cav1.2 subunit and a change in the biophysical property of the channel (Catterall, 2000). Typically, PKA-mediated phosphorylation of the Cav1.2 subunit increases the opening probability of individual channels (Bean et al., 1984, Yue et al., 1990), an effect that is known to increase L-type Ca2+ channel function in pyramidal neurons (Kavalali et al., 1997, Hoogland and Saggau, 2004). It is therefore plausible that a concurrent elevation of postsynaptic PKA signaling and L-type Ca2+ activity observed during the periadolescent transition period (i.e., PD 40) dictates the facilitation of Ca2+ plateau depolarizations in the adult PFC.
In addition to the developmental elevation of postsynaptic PKA activity, which is necessary but not sufficient to sustaining the characteristic long-lasting Ca2+ plateaus in the adult PFC, there are other mechanisms that may contribute to this periadolescent facilitation. It is well known that the Cav1.2 channel subtype accounts for ~80% of the L-type Ca2+ channels in cortical pyramidal neurons, which are often clustered in the soma as well as in dendritic spines (Calin-Jageman and Lee, 2008). Although an overall developmental upregulation of the Cav1.2a1C subunits appears to be after PD40 (Tseng & Hu, unpublished data), an age dependent redistribution of L-type Ca2+ channels along the apical dendritic tree could still account for the facilitation of Ca2+ plateaus observed after PD 40. Such a developmental change may not be restricted to the Cav1.2 subtypes as the remaining 20% of L-type Ca2+ channels are composed by the Cav1.3 subtype (Calin-Jageman and Lee, 2008). Thus, future studies are needed to determine whether an age-dependent change in the relative expression of different L-type Ca2+ channel subtypes and subunits composition (e.g., β subunit) could contribute to alter the inactivation kinetic of the channel and influence the duration of the plateau potential (Dolphin, 2009).
An age-related change in K+ currents may also affect the duration of Ca2+ potentials and contribute to the periadolescent facilitation of Ca2+ plateaus in the PFC. Among these, the TEA-insensitive hyperpolarization-activated cation channels (IH) and the Ca2+-activated K+ channels (i.e., SK) are of particular interest due to their roles in limiting the duration of synaptically-evoked Ca2+ depolarizations (Maher and Westbrook, 2005, Tsay et al., 2007). Given that all recordings in the present study were conducted with cesium-containing electrodes, it is unlikely that IH are implicated as they are blocked by intracellular as well as by extracellular cesium application (Magee, 1998, Krause et al., 2008). We therefore examined the involvement of SK and found that it is also unlikely that these channels contribute to the differential Ca2+ plateau duration observed in the PFC (Figure S1). Bath application of the SK antagonist apamin (300 nM) did not change the repolarization kinetics of the evoked potential irrespective of the duration of the plateau depolarization. These results are consistent with previous reports showing that K+ conductances is not crucial for terminating somatic-evoked Ca2+ plateau potentials (Reuveni et al., 1993, Dunia et al., 1996).
In summary, we have uncovered an important role of PKA signaling in the regulation of developmental facilitation of L-type Ca2+ function in the PFC. The exact role of an enhanced Ca2+ influx via L-type channel in the PFC during the periadolescent transition (i.e., PD 40) period is currently unknown. However, it is possible that such a facilitation of postsynaptic Ca2+ function plays an important role in triggering signaling pathways known to promote dendritic differentiation and to increasing dendritic plasticity (Deisseroth et al., 1998, Shieh et al., 1998, Rajadhyaksha et al., 1999, Tabuchi et al., 2000, Takahashi and Magee, 2009). Thus, the emergence of long-lasting L-type Ca2+-dependent plateau potentials may be critical to enable ensembles of pyramidal neurons into sustained depolarization and synchronous firing, especially in response to dopamine D1 receptors (Lewis and O’Donnell, 2000, Tseng and O’Donnell, 2005b). It is therefore predicted that activation of the mesocortical dopamine system will drive differentially PFC activity depending on the level of cortical maturation, with prepubertal PFC neurons being less likely to exhibit strong plateaus. In fact, in vivo electrophysiological recordings from prepubertal animals (PD<40) have shown PFC plateau depolarizations of smaller amplitude than similar recordings obtained from adult animals (O’Donnell et al., 2002). A similar developmental regulation of dopamine-driven plateau potentials was observed in the PFC when recorded in vitro (Tseng and O’Donnell, 2005b). This latter study further demonstrates that after puberty, L-type Ca2+ channel activation is required for sustaining prolonged plateau depolarizations elicited by co-activation of NMDA and D1 receptors (Tseng and O’Donnell, 2005b). The present study further extends our previous finding by showing that the developmental facilitation of D1 action on pyramidal neuron excitability also requires activation of L-type Ca2+ channels and postsynaptic PKA signaling. At the network level, such developmental facilitation would in turn enhance the signal detection ratio of coincidental glutamatergic drive and enable context-dependent inputs related to attention and salient stimuli (Horvitz, 2000, Cohen et al., 2002) to sustain PFC synchrony. We speculate that a periadolescent facilitation of PFC L-type Ca2+ function may contribute to maturation of mesocortical dopamine actions by enabling the PFC network to be more readily driven and reinforced by dopamine.
This study was supported by Rosalind Franklin University (KYT), National Institute of Drug Abuse (R01 DA004093 to KYT and XTH) and National Institute of Mental Health (R01 MH086507 to KYT). JM is supported by K12HD043489-08. We thank Drs. Adriana Caballero, Amiel Rosenkranz, Thanos Tzounopoulos and Anthony West for helpful comments.
Conflict of Interest: None.
We conducted additional experiments to assess the impact of SK channels blockade on Ca2+ plateau potentials in a subset of pyramidal neurons exhibiting short- and long-lasting depolarizations. We found that bath application of the SK antagonist apamin (300 nM, 10 min) does not change the duration of the evoked plateaus, suggesting that the developmental changes in Ca2+ potentials observed in the PFC is not due to changes in SK-mediated K+ currents. Figure S1 summarizes these results.