| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
NEUROSCIENCE |
1 Neurobiology & Behavior
2 Department of Neurology & Regional Epilepsy Center, University of Washington, Seattle, WA 98195-7290, USA
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
(Received 29 October 2006;
accepted after revision 14 December 2006;
first published online 21 December 2006)
Corresponding author N. Poolos: Box 359745, Harborview Medical Center, 325 9th Ave., Seattle, WA 98104, USA. Email: npoolos{at}u.washington.edu
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
One type of ion channel found in dendrites is the hyperpolarization-activated cation channel (HCN, h-channel), a family of ion channels that is responsible for the current Ih. In CA1 hippocampal pyramidal neurons, Ih is present in a low density at the soma that increases to an over sevenfold higher density in the apical dendrites (Magee, 1998). This pattern is also found in neocortical pyramidal neurons (Berger et al. 2001; Williams & Stuart, 2000). Dendritic Ih in CA1 pyramidal neurons has been proposed to normalize temporal summation of EPSPs (Magee, 1999) and reduce the overall neuronal response to excitatory input (Stuart & Spruston, 1998; Poolos et al. 2002; Nolan et al. 2004). Ih also plays a role in oscillatory firing of thalamocortical relay neurons (McCormick & Pape, 1990), layer II stellate neurons in entorhinal cortex (Dickson et al. 2000), and some interneurons of the hippocampus (Maccaferri & McBain, 1996a; Lupica et al. 2001). In cerebellar Purkinje cells, Ih helps to maintain the resting membrane potential (RMP) in the proper range to maintain high spontaneous firing (Williams et al. 2002).
All known pyramidal neurons have a similar gradient of Ih, which suggests it is essential for pyramidal neuron function. We therefore asked if this gradient of Ih is universal to neurons with pyramidal morphology. We characterized a neuron found in the stratum radiatum (SR) of CA1 hippocampus with a pyramidal morphology to test this hypothesis. We have termed these neurons ‘pyramidal-like principal’ neurons (PLPs). We will show that these neurons, which are probably a subclass of ‘radiatum giant cells’ (RGC; Gulyas et al. 1998), do not stain for interneuronal markers and project to both the septum and olfactory bulb. We found that these neurons have a distinctly different distribution of Ih compared to pyramidal dendrites, with a high density in the soma that declines to a lower density in the apical dendrite. Ih in these neurons had similar biophysical properties to Ih in pyramidal dendrites, and like Ih in pyramidal neurons, modulated temporal summation. PLP neurons therefore violate the correlation between Ih distribution and pyramidal morphology observed in other neurons, and may serve as an additional model with which to test hypotheses regarding dendritic signal integration. Some of these results have been presented previously in abstract form (Bullis et al. 2004).
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Hippocampal slices (400 µm) were made from 6–10-week-old Sprague-Dawley rats using methods previously described (Poolos et al. 2002). Animal protocols were approved by the Institutional Animal Care and Use Committee at the University of Washington. Briefly, rats were anaesthetized with an intraperitoneal injection of a ketamine/xylazine mix (304/19.2 mg kg–1) and perfused intracardially with a modified ACSF, which consisted of (mM): KCl 2.5), NaH2PO4 1.25, NaHCO3 25, CaCl2 0.5, MgCl2 7, dextrose 7, choline chloride 110, ascorbate 1.3, and pyruvate 3. All chemicals were obtained from Sigma (St Louis, MO, USA) or Fisher (Houston, TX, USA) unless otherwise stated. Supplemental doses were administered if the animal responded to a foot pinch. The brain was removed and sliced using a Vibratome 1500 sectioning system (St Louis, MO, USA). Slices were kept at 34°C for 10–20 min immediately following slicing, and at room temperature for at least 1 h before recording. Neurons were visualized with infrared differential interference contrast imaging (IR-DIC) using a Zeiss Axioskop (Oberkochen, Germany) fitted with an Olympus objective (Tokyo, Japan). Whole-cell current-clamp recordings were made with a Dagan BVC-700 amplifier (Minneapolis, MN, USA), sampled at 10 KHz, and filtered at 2 KHz. Cell-attached patch recordings were made with an Axon Instruments Axopatch 200B amplifier (Foster City, CA, USA), sampled at 2 KHz, and filtered at 500 Hz. All data were collected and analysed with custom software written for the Igor Pro analysis environment (Wavemetrics; Lake Oswego, OR, USA). Excitatory postsynaptic potentials (EPSPs) were evoked using extracellular stimulation by a constant current pulse (AM Systems Digital Stimulus Isolator; Carlsborg, WA, USA) via a bipolar tungsten electrode. EPSP amplitude was approximately 2–5 mV during recordings. The electrode was placed from 20 to 50 µm from the visualized dendrite along the border of stratum radiatum/stratum lacunosum moleculare.
For recording, slices were kept at 30–32°C and continuously bathed in oxygenated ACSF consisting of (mM): NaCl 125, NaHCO3 25, dextrose 10, KCl 2.5, CaCl2 2, MgCl2 2, and NaH2PO4 1.25. Pipettes were made from borosilicate glass and pulled with a Sutter P-87 micropipette puller (Novato, CA, USA). Pipette resistance was between 3–5 M
for soma and 4–7 M for dendritic whole-cell recordings. Pipettes were filled with a solution that contained (mM): KMeSO4 120, KCl 20, Hepes 10, Na2-ATP 4, MgCl2 2, Tris-GTP 0.3, EGTA 0.2, pH 7.3 with KOH. For recordings involving synaptic stimulation, 1 mM EGTA was included in the pipette to reduce calcium-related changes in synaptic efficacy during recordings, and 20 µM bicuculline was added to the bath to block inhibitory synaptic input. A cut was made between CA3 and CA1 to minimize spontaneous activity when bicuculline was used. Pipette resistance for cell-attached recordings was between 4–15 M. Pipettes were filled with a solution consisting of (mM): KCl 120, TEA-Cl 20, Hepes 10, 4-aminopyridine 5, CaCl2 2, MgCl2 1, BaCl2 1, pH was adjusted to 7.4 with KOH. For estimations of maximal current density, the area of the patch was obtained from the equation relating patch area to resistance found in Sakmann & Neher (1995), and were elicited at 
–160 mV. Recordings were accepted only if the resting membrane potential was at least –55 mV and Ih showed rundown of less than 25%. Series resistance for whole-cell recordings was less than 40 M. Group data are expressed as mean ± S.E.M. Unless stated, a one-way ANOVA was used to compare three or more conditions, and an unpaired t test for two conditions. Alpha functions were created using the equation:
|
| (1) |
= 1.67. Temporal summation was measured as the fifth EPSP or response divided by the first times 100, which gives the percentage of the response relative to the first event. ZD7288 was obtained from Tocris (Balwin, MO, USA). ZD7288, APV, and bicuculline were made in aqueous stock solutions and bath applied. Histology
To characterize cellular morphology, pyramidal and PLP neurons were filled via recording micropipettes containing a 0.1–0.3% biocytin (Molecular Probes; Eugene, OR, USA) solution. After recording, slices were fixed overnight with a solution of 4% paraformaldehyde in 0.12 M phosphate buffer. Slices were then blocked overnight in 2% BSA with 0.25% Triton X-100 in 0.03 M PBS, incubated overnight with streptavidin/Alexa-488 (1: 200; Molecular Probes) with 0.1% Triton X-100 in PBS, and mounted with Vectashield (Vector Laboratories; Burlingame, CA, USA) on gel-coated slides featuring a custom parafilm (Pechiney Plastic Packaging; Chicago, IL, USA) spacer to allow for sufficient space between the slice and coverslip. Images of the fluorescent neurons were collected in approximately 15 planes through the cells with a Nikon Eclipse TE200/Bio-Rad Radiance 2000 confocal microscope system (El Segundo, CA, USA; Hercules, CA, USA), and processed using Image J software (Wayne Rasband, NIH, USA) into composite projection micrographs. Further anatomical analysis was collected on slices secondarily stained with avidin/horseradish peroxidase (HRP; Vecatastain ABC kit, Vector Laboratories; Burlingame, CA, USA). HRP-treated slices were visualized with a Leica Leitz Laborlux S light microscope (Bannockburn, IL, USA) and photographed using a Nikon Coolpix 4500 digital camera.
Biocytin-filled neurons were examined for colocalization of proteins associated with interneurons, by incubating slices in antibodies for glutamic acid decarboxylase 65/67, calretinin (GAD, CR; Chemicon; Temecula, CA, USA), cholescystokinin-8 (CCK; Immunostar; Hudson, WI, USA), and parvalbumin (PV; Sigma; St. Louis, MO, USA). Fixed and blocked slices were incubated for 24 h in PBS containing 0.3% Triton X-100 and antibodies in concentrations of 1: 1000 (GAD and CCK) and 1: 4000 (PV), followed by 1: 1000 incubation and antirabbit Alexa-568 (GAD and CCK; Molecular Probes; Eugene, OR, USA) and antimouse Alexa-635 (PV; Molecular Probes) for 24 h. These slices were then processed with streptavidin/Alexa-488 as detailed previously. CR immunoreactivity was assayed on fixed hippocampal tissue (without filled neurons) sectioned to 100 µm with a Leica VT 1000 S microtome (Bannockburn, IL, USA). After blocking overnight, slices were incubated for 48 h in 0.05 M TBS with 0.3% Triton X-100 containing antibodies for CR raised in the goat (1: 8000, Chemicon), followed by incubation in antigoat Alexa-568 (1: 1000).
To determine immunoreactivity for proteins associated with excitatory neurons, fixed tissue was sectioned to 100 µm, incubated with goat antibodies for excitatory amino acid transporter 3 (EAAT3, EAAC1; 1: 4000; Chemicon) in TBS (containing no Triton X-100) for 48 h, and finally incubated with antigoat Alexa-568 (1: 1000) in TBS for 24 h. All incubations and washes were performed at 4°C, and slices were mounted and imaged as described above.
Retrograde labelling of PLP neuronal projections
Injections of the retrograde tracer cholera toxin conjugated with Alexa-488 (CTB; Molecular Probes) were made into the olfactory bulb and lateral septum of animals anaesthetized with 4–5% isofurane, to examine the axonal projections of PLP neurons. Using a Hamilton syringe, 2 µl of CTB were administered in the olfactory bulb at three stereotaxic positions previously used by Gulyas et al. (1998) (AP: 9, L: 1.3; D: –3.5; AP: 8, L: 1.1 D: –2.5; AP: 6.5, L: 1.1 D: –1.5 from bregma). Septal injections of 3 µl CTB were made at coordinates AP: 0.7, L: 0.7 D: 6.0 from bregma (Paxinos & Watson, 1998). The animals recovered from surgery for three days, allowing for tracer uptake by axon terminals and retrograde transport. Following recovery, animals were anaesthetized by a similar method to the electrophysiology experiments and intracardially perfused with PBS and 4% paraformaldehyde. Harvested brain tissue was then sectioned to 100 µm, washed in PBS (3 x 60 min), and visualized with fluorescent confocal microscopy.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
We identified a class of neurons in the hippocampus with pyramidal morphology to test whether they had a similar distribution of Ih as pyramidal neurons. We have termed these neurons ‘pyramidal-like principal’ (PLP) neurons, based on our characterization below. We chose these neurons in part because they are easily located in the stratum radiatum (SR) using infrared differential interference contrast microscopy (IR-DIC). Their relative isolation allowed for both easier patch clamping at the soma and tracing of their dendrites for dendritic recordings. Figure 1A shows typical IR-DIC micrographs of both pyramidal and PLP neurons. The soma of the PLP is isolated, while the pyramidal neurons are surrounded by other somata in the stratum pyramidale (SP).
|
Fluorescent staining of a biocytin-filled CA1 pyramidal neuron (Fig. 1B) and PLP neuron (Fig. 1C) revealed a cylindrical or biconical radiation of dendrites common to pyramidal neurons (Fiala & Harris, 1999). The proximal apical dendrites of both types of neurons typically had oblique branches that did not branch extensively. The apical dendrites of both neuron types would often bifurcate after a minimum of 50 µm into branches that themselves branched into many terminal processes. Both neuron types also showed basal dendrites, with several smaller processes emanating from the soma that did not extensively branch. Location, branching, and the projection of PLP dendrites were confirmed with light microscopy and horseradish peroxidase staining of a biocytin-filled PLP neuron in Fig. 1D, which shows a PLP soma located approximately 50 µm from the SP, and demonstrates the extension of the apical dendrite into the SLM. The axons of some PLP neurons were visible projecting to the alveus (Fig. 1C, E and F; arrowheads), while others also bifurcated (Fig. 1E), which suggested that PLP neuron axons project both outside of the hippocampus and to the subiculum and entorhinal cortex, although subicular and entorhinal connections were not verified. In two cases where spines were apparently visible in low-resolution images, higher-resolution images confirmed the presence of spines on secondary dendritic branches (Fig. 1G; arrowheads). These results indicate that PLP and pyramidal neurons share a similar dendritic morphology.
PLPs lack inhibitory interneuronal markers
The presence of spines on the dendrites of PLPs combined with the projection of their axons to the alveus suggested that PLPs are excitatory neurons (but see Hsu & Buzsaki, 1993). We sought to determine if PLPs were excitatory or inhibitory by staining for typical interneuronal markers glutamic acid decarboxylase 65/67 (GAD), parvalbumin (PV), cholescystokinin-8 (CCK), and calretinin (CR; Freund & Buzsaki, 1996, McBain & Fisahn, 2001). When biocytin-filled neurons were counterstained for these markers, PLPs were found to be immunonegative for GAD, PV, CCK, and CR (Fig. 2A and B; CCK data not shown), while neurons that matched morphological descriptions of previously described interneurons were clearly visible (Freund & Buzsaki, 1996; McBain & Fisahn, 2001). In Fig. 2A, numerous neurons were present that stained for GAD. Antibodies for PV primarily stained neurons with somata located in the SP, with some apical processes visible in the SR consistent with the known location of PV-positive basket cells and chandelier cells (Freund & Buzsaki, 1996). In Fig. 2B the dendritic morphology of neurons stained for CR was multipolar, with little branching, while neurons of a pyramidal morphology were not stained. This branching is consistent with the description of CR interneurons as having bipolar dendrites (Freund & Buzsaki, 1996; Gulyas et al. 1996).
|
PLPs project to subcortical regions
It is known that CA1 hippocampal pyramidal neurons project to the septum and olfactory bulb while also projecting within the hippocampal formation to the subiculum and entorhinal cortex in rats (van Groen & Wyss, 1990). Interneurons of the hippocampus, however, project only locally (Freund & Buzsaki, 1996; McBain & Fisahn, 2001). We therefore sought to test whether PLPs have similar projections as CA1 pyramidal neurons by injecting the retrograde tracer cholera toxin Alexa-488 (CTB) into the septum. After three days we examined hippocampal slices to find dense staining of pyramidal neurons and neurons with PLP morphology in the SR (Fig. 3A; arrowhead). We also found neurons with two prominent apical dendrites, which is typical for RGCs and consistent with prior reports of their projection outside the hippocampus (Gulyas et al. 1998; Fig. 3B, arrow). We then administered CTB into the olfactory bulb and again found staining of neurons with PLP morphology in the SR (Fig. 3C and D). Olfactory injections only intermittently labelled pyramidal neurons, which is similar to what was observed in earlier retrograde staining (van Groen & Wyss, 1990; Gulyas et al. 1998). Together, these data suggest that PLPs are a principal neuron of the hippocampus.
|
In summary, we found similarities in morphology between PLP and pyramidal neurons based upon visualization of biocytin-filled neurons. PLPs do not stain for interneuronal markers, but do stain for an excitatory marker, which suggests that PLP neurons are excitatory. Finally, we found that PLP and pyramidal neurons project to the septum and olfactory bulb, although to differing extents.
PLPs have a gradient of Ih density that is reversed from that in CA1 pyramidal neurons
Ih in CA1 pyramidal neurons is known to be present in a gradient pattern, with a low density in the soma that increases sevenfold in the apical dendrites (Magee, 1998). We sought to determine if the PLP neuron had a similar gradient of Ih using cell-attached patch recordings. Figure 4A shows typical current traces from PLP soma and dendrite recordings. Unlike CA1 pyramidal neurons, PLP somata had higher maximal currents than in PLP dendrites. We measured Ih at dendritic distances up to 180 µm from the PLP soma. Because of the location within the SR and larger size of PLP somata, this distance was at or beyond the SR–SLM border, and thus represented distal recording sites for PLP neurons. By normalizing for patch area (Sakmann & Neher, 1995), we found an average density of 14 ± 1.5 pA µm–2 in the PLP somata (n = 20). The average density in the first 60 µm of the dendrites was 7.8 ± 1.9 pA µm–2 (n = 18). For the second 60 µm, the density was 6.4 ± 1.0 pA µm–2 (n = 10) and for the final 60 µm the density was 5.3 ± 1.3 pA µm–2 (n = 6; Fig. 4B). Maximal Ih density was significantly lower in the dendrites compared to the soma (P < 0.05; Kruskal–Wallis ANOVA). Although the means declined through the dendritic compartments, differences between these compartments were not significant. The density of Ih in PLP somata was similar to what we found in the apical dendrites of pyramidal neurons (12 ± 3.4 pA µm–2; n = 12) at a mean distance of 156 µm ± 7.5 µm (n = 12) with a mean patch area of 2.8 ± 0.13 µm2 (n = 11).
|
We measured the biophysical properties of Ih observed in PLP soma and dendrite recordings from the furthest two bins (range between 60 and 180 µm; average 106 ± 9.3 µm), and compared them to those in CA1 pyramidal dendrites. Voltage-dependent activation was found by measuring the normalized tail currents and fitting the current–voltage relationship with a Boltzmann function (Fig. 4C). We found a mean half-activation voltage (V1/2) of –84 ± 3.0 mV for PLP somata (n = 11), –95 ± 2.4 for PLP dendrites (n = 10), and –86 ± 3.4 mV (n = 7) for pyramidal dendrites. The V1/2 was significantly hyperpolarized for PLP dendrites compared to PLP somata or pyramidal dendrites (P < 0.05), but was not significantly different between pyramidal dendrites and PLP somata. A similar trend in V1/2 across the somatodendritic axis has also been observed in pyramidal neurons (Magee, 1998).
Activation time constants were measured by fitting the current evoked near V1/2 with a double-exponential function. Ih shows a double-exponential time course in CA1 pyramidal neurons, reflecting the expression of both HCN1 and HCN2, with HCN1 activation normally an order of magnitude faster than HCN2 (Santoro et al. 2000; Chen et al. 2001; Ulens & Tytgat, 2001; Robinson & Siegelbaum, 2003). Slow activation time constants were 624 ± 94.8 ms for PLP somata (n = 11), 740 ± 189 ms for PLP dendrites (n = 10), and 660 ± 179 ms for pyramidal dendrites (n = 7). There were no significant differences between groups. Fast activation time constants were 79 ± 10 ms for PLP somata, 75 ± 11 ms for PLP dendrites, and 54 ± 8.7 ms for pyramidal dendrites. Again, the groups did not significantly differ. The similarities of both voltage-dependent activation and time constants demonstrate that Ih is equivalent in pyramidal and PLP neurons, thus both neurons probably have a similar composition of HCN1 and HCN2 subunits.
Following cell-attached patch recording, we ruptured the patch to determine the resting membrane potential (RMP). Ih is known to depolarize the RMP, which in some interneurons of the hippocampus influences their spontaneous firing (Maccaferri & McBain, 1996a; Lupica et al. 2001). PLP somata had a RMP of –67 ± 1.7 mV (n = 11), while PLP dendrites had a RMP of –72 ± 2.8 mV (n = 11), with no significant difference between the two locations (P > 0.05). The RMP of pyramidal dendrites was –60 ± 1.3 mV (n = 7), which was significantly depolarized compared to the RMP of PLP somata and dendrites (P < 0.05). This could reflect the reduced Ih in PLP dendrites relative to pyramidal dendrites, which would cause a hyperpolarizing shift in the RMP, although other ion channels also help set the RMP in neurons (Day et al. 2005).
Ih contributes to PLP intrinsic properties
Whole-cell current-clamp recordings were performed at the soma and dendrites to characterize the influence of Ih on the intrinsic properties of PLP neurons. Dendritic recordings were made at a range of 125–175 µm, with an average of 142 ± 6.1 µm (n = 10). Because K-gluconate blocks Ih and some K+ currents (Velumian et al. 1997), we used KMeSO4 in our internal solution to measure the passive properties of PLP neurons. By giving depolarizing current steps in 100 pA increments for 600 ms, we found marked accommodation of action potential (AP) firing in most PLP neurons (Fig. 5A), similar to that in CA1 hippocampal pyramidal neurons. Most of the neurons examined clearly showed a spike after-depolarization (Fig. 5A, arrowhead), which is common in pyramidal and RGC neurons (Gulyas et al. 1998). Hyperpolarizing current injections in all neurons produced a sag in steady-state membrane potential, indicating Ih activation (Fig. 5A; arrow).
|
We sought to verify the contribution of Ih to PLP passive membrane properties by bath-applying ZD7288 (20 µM), a membrane-permeable and specific inhibitor of Ih (Harris & Constanti, 1995), during somatic recordings (Fig. 5C). To control for variability in the RMP, we held the membrane potential at –65 mV. Current injections in the presence of ZD7288 after 30 min produced a greater voltage response than in control conditions, and were sufficient to invoke firing in response to a previously subthreshold current injection. The voltage sag indicative of Ih was also reduced in the presence of ZD7288.
We measured the steady-state input resistance (Rin) in response to 100–300 pA hyperpolarizing current injections (Fig. 5D) and found a Rin of 56 ± 2.8 M for PLPs (n = 25) recorded under control conditions in the soma, and 51 ± 4.9 in the dendrites (n = 10). There was no significant difference between input resistance at the soma and the dendrites. After 30 min of incubation, ZD7288 significantly increased Rin at the soma to 144 ± 8.5 M (n = 14) as expected, indicating a significant contribution of Ih to the input resistance (P < 0.001).
Blockade of Ih by ZD7288 also hyperpolarized the RMP of PLP neurons. The RMP in PLPs hyperpolarized from –66 ± 0.9 to –72 ± 1.0 (n = 16) following application of ZD7288 (P < 0. 001, paired t test). Ih thus helps to set the RMP of PLP neurons, although other ion channels are probably involved as well (Day et al. 2005).
Ih in PLPs alters temporal summation similarly to pyramidal neurons
Ih plays a role in temporal summation (TS) of pyramidal neurons (Magee, 1999; Poolos et al. 2002). This effect is most prominent at a frequency of 20 Hz, and less at other frequencies. To see if the same effects are observed in PLP neurons, we measured the response of PLP neurons to alpha-function current injections at different frequencies to simulate EPSPs. We used a train of five alpha functions at 10 Hz, 20 Hz, and 40 Hz delivered to the soma or dendrites of PLP neurons. Representative voltage responses of a neuron to trains of alpha functions are shown in control conditions and following application of ZD7288 (20 µm; Fig. 5E). Following application of ZD7288, there is a visible increase in the amplitude and time course of the neuronal response to the alpha current injection, leading to an overall increase in TS across frequencies. TS (Fig. 5E) in control conditions was: 103 ± 3.1%, 143 ± 4.5%, and 238 ± 5.4% (n = 19) for 10, 20, and 40 Hz, respectively. Following ZD7288 treatment, TS was: 145 ± 6.8%, 205 ± 9.4%, and 290 ± 14% (n = 10) for 10, 20, and 40 Hz, respectively. For dendrites in control conditions TS was: 93 ± 4.0%, 131 ± 9.8%, and 231 ± 14.5% (n = 10) for 10, 20, and 40 Hz, respectively. For all frequencies, ZD7288 significantly increased TS (P < 0.01, paired t test), but no difference between soma and dendrites was detected. In summary, Ih in PLP neurons decreased temporal summation, with the greatest change coming at 20 Hz, the frequency most sensitive to Ih blockade in pyramidal neurons. Local summation, however, was constant from soma to dendrites.
PLP input is normalized across the somatodendritic axis
To further clarify the role of temporal summation in PLP neurons, we evoked excitatory postsynaptic potentials (EPSPs) in PLP neurons by extracellularly stimulating afferents along the border of the SR–SLM (Fig. 6A). We measured the amount of EPSP summation in the neurons to five stimuli delivered at 10, 20, and 40 Hz, and blocked Ih by ZD7288 (20 µm) to isolate its effects on TS. TS was: 141 ± 6.7%, 263 ± 18.4%, and 455 ± 31.7% for 10, 20, and 40 Hz, respectively, under control conditions (n = 4). Following application of ZD7288, TS increased to: 238 ± 22%, 339 ± 4.1%, and 544 ± 52% for 10, 20, and 40 Hz, respectively, a significant increase for 10 and 20 Hz (P < 0.05, paired t test). These results are higher than TS measured with current injections, and suggest that synaptic events may be a critical factor in TS beyond the intrinsic properties of the neuron.
|
These results show that PLP neurons have Ih that is similar in biophysical properties to that of pyramidal neurons, but found in a reversed somatodendritic gradient. Input resistance and local temporal summation is similar across the somatodendritic axis, with Ih making a significant contribution. As in pyramidal neurons, synaptic TS is also normalized across the somatodendritic axis, with Ih a strong component.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Prior studies have shown that Ih is present at high density in the dendrites of pyramidal neurons in both CA1 hippocampus and the neocortex (Magee, 1998; Berger et al. 2001). The present study demonstrates that this pattern is not a universal feature of neurons with pyramidal morphology. The reversed Ih distribution in PLP neurons compared to CA1 pyramidal neurons is all the more striking considering that the apical dendritic arbors of these two neurons traverse the same hippocampal strata – the stratum radiatum and stratum lacunosum moleculare – and probably receive some of the same excitatory afferent inputs (Maccaferri & McBain, 1996b; Gulyas et al. 1998). This suggests that h-channel somatodendritic distribution is specific to neuronal type and not simply a function of dendritic morphology or excitatory synaptic input. Of note, a different Ih distribution is seen in cerebellar Purkinje cells in which Ih density is uniform from soma to dendrites (Hausser & Clark, 1997), showing that other patterns may be seen in non-pyramidal neurons.
Ih active at rest decreases the temporal summation of EPSPs by gradually deactivating (producing a net outward current), thus opposing the summated depolarization. In pyramidal neurons, Ih tends to ‘normalize’ the temporal summation of EPSPs evoked at distal dendritic locations to the value seen at proximal locations, by compensating for the increase in TS that would occur through passive propagation of distal dendritic potentials to the soma (Magee, 1999). In PLP neurons, the same phenomenon appears to occur, with quantitatively similar summation of distally evoked EPSPs compared to summation of EPSP-like voltage transients at the soma. This suggests, as suspected on theoretical grounds, that normalization of temporal summation may occur with a variety of Ih subcellular distribution patterns (Magee, 1999). Unlike in pyramidal neurons, however, local temporal summation in PLP neurons was not different for dendritic versus somatic recording sites; this probably reflects the shallower Ih gradient (2-fold change versus 7-fold in CA1 pyramidal neurons), and the shorter apical dendritic tree found in PLP neurons.
PLP neurons share many characteristics with the previously described radiatum giant cells, including somatic size and location, lack of staining for interneuronal markers, and extra-hippocampal axonal projections (Gulyas et al. 1998; Kirson & Yaari, 2000; Savic & Sciancalepore, 2001). In light of these similarities, this study shows that PLPs probably represent a subtype of RGCs. In particular, PLP neurons are probably the ‘P-cell-like’ variety described by Gulyas et al. (1998) which comprises approximately 20% of the RGC population. PLPs probably also share RGC patterns of innervation in area CA1 from stratum radiatum afferents, and presumably make excitatory synapses as has been shown for RGCs on an ultrastructural level (Gulyas et al. 1998), although this was not demonstrated in PLP neurons. The typical RGC morphology, however, possesses an inverted pyramidal or shield-shaped soma, with two symmetrical apical dendrites emanating from the apices of the soma (as seen in Fig. 3B). The distinct pyramidal morphology of PLPs compared to that of the typical non-pyramidal, bidendritic morphology of RGCs suggests that it is appropriate to distinguish these two types of neurons. Little is known about the ion channel distributions in typical RGCs, but study of h-channels and other ion channel types in these cells may be useful in understanding how the excitability of these neurons differs from PLPs.
The function of PLPs and RGCs is similarly unknown. That they appear to project more strongly to the olfactory bulb than CA1 hippocampal pyramidal neurons suggests they may subserve a different functional role in the hippocampus. On a more basic level, however, we believe that PLP neurons will be a useful model system for understanding how Ih affects signal integration in pyramidal neurons, given its reversed somatodendritic Ih distribution compared to CA1 pyramidal neurons. Further, Ih in PLP neurons appears to have almost identical biophysical properties to that in CA1 hippocampal pyramidal neurons, and its somatic location in PLP neurons makes patch-clamp recording far more accessible than the comparable dendritic location in CA1 pyramidal neurons. Thus PLP neurons supply an ideal model system for studying pharmacology and modulation of h-channels (Poolos et al. 2006). These considerations underlie our hope that PLPs will emerge as a novel principal neuron of the hippocampus worthy of further investigation.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Bullis JB, Jones TJ & Poolos NP (2004). H-channel distribution in pyramidal-like principal neurons in stratum radiatum of hippocampus differs from pyramidal neurons. Soc Neurosci Abs 29.
Chen S, Wang J & Siegelbaum SA (2001). Properties of hyperpolarization-activated pacemaker current defined by coassembly of HCN1 and HCN2 subunits and basal modulation by cyclic nucleotide. J Gen Physiol 117, 491–504.
Christie BR, Franks KM, Seamans JK, Saga K & Sejnowski TJ (2000). Synaptic plasticity in morphologically identified CA1 stratum radiatum interneurons and giant projection cells. Hippocampus 10, 673–683.[CrossRef][Medline]
Colbert CM, Magee JC, Hoffman DA & Johnston D (1997). Slow recovery from inactivation of Na+ channels underlies the activity-dependent attenuation of dendritic action potentials in hippocampal CA1 pyramidal neurons. J Neurosci 17, 6512–6521.
Day M, Carr DB, Ulrich S, Ilijic E, Tkatch T & Surmeier DJ (2005). Dendritic excitability of mouse frontal cortex pyramidal neurons is shaped by the interaction among HCN, Kir2, and Kleak channels. J Neurosci 25, 8776–8787.
Dickson CT, Magistretti J, Shalinsky MH, Fransen E, Hasselmo ME & Alonso A (2000). Properties and role of I(h) in the pacing of subthreshold oscillations in entorhinal cortex layer II neurons. J Neurophysiol 83, 2562–2579.
Fiala JC & Harris KM (1999). Dendrite structure. In Dendrites, ed. Stuart G, Spruston N & Häusser M, pp. 1–34. Oxford University Press, Oxford, New York.
Freund TF & Buzsaki G (1996). Interneurons of the hippocampus. Hippocampus 6, 347–470.[CrossRef][Medline]
Gulyas AI, Hajos N & Freund TF (1996). Interneurons containing calretinin are specialized to control other interneurons in the rat hippocampus. J Neurosci 16, 3397–3411.
Gulyas AI, Toth K, McBain CJ & Freund TF (1998). Stratum radiatum giant cells: a type of principal cell in the rat hippocampus. Eur J Neurosci 10, 3813–3822.[CrossRef][Medline]
Harris NC & Constanti A (1995). Mechanism of block by ZD 7288 of the hyperpolarization-activated inward rectifying current in guinea pig substantia nigra neurons in vitro. J Neurophysiol 74, 2366–2378.
Hausser M & Clark BA (1997). Tonic synaptic inhibition modulates neuronal output pattern and spatiotemporal synaptic integration. Neuron 19, 665–678.[CrossRef][Medline]
Hausser M & Mel B (2003). Dendrites: bug or feature? Curr Opin Neurobiol 13, 372–383.[CrossRef][Medline]
Hausser M, Spruston N & Stuart GJ (2000). Diversity and dynamics of dendritic signaling. Science 290, 739–744.
Hsu M & Buzsaki G (1993). Vulnerability of mossy fiber targets in the rat hippocampus to forebrain ischemia. J Neurosci 13, 3964–3979.[Abstract]
Johnston D, Magee JC, Colbert CM & Cristie BR (1996). Active properties of neuronal dendrites. Annu Rev Neurosci 19, 165–186.[CrossRef][Medline]
Jung HY, Mickus T & Spruston N (1997). Prolonged sodium channel inactivation contributes to dendritic action potential attenuation in hippocampal pyramidal neurons. J Neurosci 17, 6639–6646.
Kanai Y, Bhide PG, DiFiglia M & Hediger MA (1995). Neuronal high-affinity glutamate transport in the rat central nervous system. Neuroreport 6, 2357–2362.[Medline]
Kirson ED & Yaari Y (2000). Unique properties of NMDA receptors enhance synaptic excitation of radiatum giant cells in rat hippocampus. J Neurosci 20, 4844–4854.
Lupica CR, Bell JA, Hoffman AF & Watson PL (2001). Contribution of the hyperpolarization-activated current (I(h) to membrane potential and GABA release in hippocampal interneurons. J Neurophysiol 86, 261–268.
Maccaferri G & McBain CJ (1996a). The hyperpolarization-activated current (Ih) and its contribution to pacemaker activity in rat CA1 hippocampal stratum oriens-alveus interneurones. J Physiol 497, 119–130.[Medline]
Maccaferri G & McBain CJ (1996b). Long-term potentiation in distinct subtypes of hippocampal nonpyramidal neurons. J Neurosci 16, 5334–5343.
Magee JC (1998). Dendritic hyperpolarization-activated currents modify the integrative properties of hippocampal CA1 pyramidal neurons. J Neurosci 18, 7613–7624.
Magee JC (1999). Dendritic Ih normalizes temporal summation in hippocampal CA1 neurons. Nat Neurosci 2, 508–514.[CrossRef][Medline]
McBain CJ & Fisahn A (2001). Interneurons unbound. Nat Rev Neurosci 2, 11–23.[Medline]
McCormick DA & Pape HC (1990). Properties of a hyperpolarization-activated cation current and its role in rhythmic oscillation in thalamic relay neurones. J Physiol 431, 291–318.
Nolan MF, Malleret G, Dudman JT, Buhl DL, Santoro B, Gibbs E, Vronskaya S, Buzsaki G, Siegelbaum SA, Kandel ER & Morozov A (2004). A behavioral role for dendritic integration: HCN1 channels constrain spatial memory and plasticity at inputs to distal dendrites of CA1 pyramidal neurons. Cell 119, 719–732.[Medline]
Paxinos G & Watson C (1998). The Rat Brain in Stereotaxic Coordinates. Academic Press, San Diego.
Poolos NP, Bullis JB & Roth MK (2006). Modulation of h-channels in hippocampal pyramidal neurons by p38 mitogen-activated protein kinase. J Neurosci 26, 7995–8003.
Poolos NP, Migliore M & Johnston D (2002). Pharmacological upregulation of h-channels reduces the excitability of pyramidal neuron dendrites. Nat Neurosci 5, 767–774.[Medline]
Rall W (1959). Branching dendritic trees and motoneuron membrane resistivity. Exp Neurol 1, 491–527.[CrossRef][Medline]
Rall W (1962). Electrophysiology of a dendritic neuron model. Biophys J 2, 145–167.
Reyes A (2001). Influence of dendritic conductances on the input–output properties of neurons. Annu Rev Neurosci 24, 653–675.[CrossRef][Medline]
Robinson RB & Siegelbaum SA (2003). Hyperpolarization-activated cation currents: from molecules to physiological function. Annu Rev Physiol 65, 453–480.[CrossRef][Medline]
Rothstein JD, Martin L, Levey AI, Dykes-Hoberg M, Jin L, Wu D, Nash N & Kuncl RW (1994). Localization of neuronal and glial glutamate transporters. Neuron 13, 713–725.[CrossRef][Medline]
Sakmann B & Neher E (1995). Single-channel Recording. Plenum Press, New York.
Santoro B, Chen S, Luthi A, Pavlidis P, Shumyatsky GP, Tibbs GR & Siegelbaum SA (2000). Molecular and functional heterogeneity of hyperpolarization-activated pacemaker channels in the mouse CNS. J Neurosci 20, 5264–5275.
Savic N & Sciancalepore M (2001). Electrophysiological characterization of ‘giant’ cells in stratum radiatum of the CA3 hippocampal region. J Neurophysiol 85, 1998–2007.
Segev I & London M (2000). Untangling dendrites with quantitative models. Science 290, 744–750.
Stuart GJ & Sakmann B (1994). Active propagation of somatic action potentials into neocortical pyramidal cell dendrites. Nature 367, 69–72.[CrossRef][Medline]
Stuart G & Spruston N (1998). Determinants of voltage attenuation in neocortical pyramidal neuron dendrites. J Neurosci 18, 3501–3510.
Ulens C & Tytgat J (2001). Functional heteromerization of HCN1 and HCN2 pacemaker channels. J Biol Chem 276, 6069–6072.
van Groen T & Wyss JM (1990). Extrinsic projections from area CA1 of the rat hippocampus: olfactory, cortical, subcortical, and bilateral hippocampal formation projections. J Comp Neurol 302, 515–528.[CrossRef][Medline]
Velumian AA, Zhang L, Pennefather P & Carlen PL (1997). Reversible inhibition of IK, IAHP, Ih and ICa currents by internally applied gluconate in rat hippocampal pyramidal neurones. Pflugers Arch 433, 343–350.[CrossRef][Medline]
Vetter P, Roth A & Hausser M (2001). Propagation of action potentials in dendrites depends on dendritic morphology. J Neurophysiol 85, 926–937.
Williams SR, Christensen SR, Stuart GJ & Hausser M (2002). Membrane potential bistability is controlled by the hyperpolarization-activated current IH in rat cerebellar Purkinje neurons in vitro. J Physiol 539, 469–483.
Williams SR & Stuart GJ (2000). Site independence of EPSP time course is mediated by dendritic Ih in neocortical pyramidal neurons. J Neurophysiol 83, 3177–3182.
|
Acknowledgements |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. J. Sjostrom, E. A. Rancz, A. Roth, and M. Hausser Dendritic Excitability and Synaptic Plasticity Physiol Rev, April 1, 2008; 88(2): 769 - 840. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Angelo, M. London, S. R. Christensen, and M. Hausser Local and Global Effects of Ih Distribution in Dendrites of Mammalian Neurons J. Neurosci., August 8, 2007; 27(32): 8643 - 8653. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Colbert A circuit within a circuit? J. Physiol., March 1, 2007; 579(2): 289 - 289. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS |
), PLP dendrites recorded at an average distance of 106 ± 9.3 µm (
), and pyramidal neuron dendrites (
) recorded at an average distance of 156 ± 7.5 µm. PLP somata showed a similar activation curve to pyramidal dendrites, while PLP dendrites had a hyperpolarized shift of approximately 10 mV.
