The Kv2.1 K+ channel targets to the axon initial segment of hippocampal and cortical neurons in culture and in situ
The Kv2.1 delayed-rectifier K+ channel regulates membrane excitability in hippocampal neurons where it targets to dynamic cell surface clusters on the soma and proximal dendrites. In the past, Kv2.1 has been assumed to be absent from the axon initial segment.
Transfected and endogenous Kv2.1 is now demonstrated to preferentially accumulate within the axon initial segment (AIS) over other neurite processes; 87% of 14 DIV hippocampal neurons show endogenous channel concentrated at the AIS relative to the soma and proximal dendrites. In contrast to the localization observed in pyramidal cells, GAD positive inhibitory neurons within the hippocampal cultures did not show AIS targeting. Photoactivable-GFP-Kv2.1-containing clusters at the AIS were stable, moving <1 μm/hr with no channel turnover. Photobleach studies indicated individual channels within the cluster perimeter were highly mobile (FRAP τ = 10.4 ± 4.8 sec), supporting our model that Kv2.1 clusters are formed by the retention of mobile channels behind a diffusion-limiting perimeter. Demonstrating that the AIS targeting is not a tissue culture artifact, Kv2.1 was found in axon initial segments within both the adult rat hippocampal CA1, CA2, and CA3 layers and cortex.
In summary, Kv2.1 is associated with the axon initial segment both in vitro and in vivo where it may modulate action potential frequency and back propagation. Since transfected Kv2.1 initially localizes to the AIS before appearing on the soma, it is likely multiple mechanisms regulate Kv2.1 trafficking to the cell surface.
KeywordsHippocampal Neuron Glutamic Acid Decarboxylase Surface Cluster Axon Initial Segment Hippocampal Culture
Axon initial segement
Days in vitro
glutamic acid decarboxylase.
Voltage-gated ion channels are often highly localized in electrically excitable cells such as nerve and muscle. As originally noted by Trimmer and colleagues , the Kv2.1 delayed rectifier is expressed primarily in the somatic region of hippocampal neurons where it is found in cell surface clusters that can co-localize with ryanodine receptors and SR-like subsurface cisterns [2, 3]. Interestingly, these clusters also co-localize with cholinergic synapses in spinal motor neurons . Kv2.1 represents the predominant delayed rectifier current in hippocampal neurons where its activity and localization are highly regulated [5, 6]. Glutamate or carbachol treatments induce both Kv2.1 dephosphorylation and declustering [7, 8, 9]. Both treatments also result in a 20 mV hyperpolarizing shift in the activation curve for IK. Chemically-induced ischemia also induces declustering, dephosphorylation, and the hyperpolarizing shift in the activation midpoint [8, 9]. Similar regulation is observed in Kv2.1 transfected HEK cells . These data suggest a strong link between cluster formation, channel phosphorylation, and the voltage-dependence of activation. The increase in channel activity that is linked to declustering has been proposed to be a neuro-protective response to hypoxia/ischemic insult . However, Kv2.1 trafficking to the cell surface is also implicated in cortical neuron apoptosis [11, 12], emphasizing that the trafficking and regulation of Kv2.1 must be under tight physiological control.
While it is commonly assumed that ion channel localization must involve static tethering to scaffolding proteins that in turn are linked directly to the cytoskeleton, our recent studies indicate that the Kv2.1 surface clusters are formed when mobile Kv2.1 channels are corralled behind a cortical actin-based fence . This sub-membrane fence is selective towards only the confined channels, with other membrane proteins being free to cross it. Thus, the Kv2.1-containing surface clusters represent a new mechanism for the stable localization of ion channel proteins to specific cell surface domains. Our previous studies also indicate that the surface clusters are specialized surface sites for the membrane insertion of Kv2.1 channels, functioning as intracellular trafficking vesicle targets . During the course of our studies we often observed GFP-Kv2.1 clusters forming in a single proximal neurite of a transfected hippocampal neuron. While the expression of Kv2.1 within the axon initial segment (AIS) of cultured hippocampal neurons has previously been referred to as a tissue culture artifact , AIS localization was often the only cell surface expression observed in an individual cell. The study presented here was initiated by this apparent contradiction between the literature and our data obtained in hippocampal neurons transfected with GFP-Kv2.1.
We report here that both transfected and endogenous Kv2.1 often show a real preference for the AIS in cultured hippocampal neurons. The Kv2.1 clusters within the AIS are similar to those found on the cell body in that they consist of mobile channels trapped by a perimeter fence. However, perhaps due to the sub-membrane diffusion barriers in the AIS [15, 16, 17], the clusters themselves appear to be more confined than their cell body counterparts . Kv2.1 concentration within the AIS also occurs in both cortical and hippocampal neurons of adult brain, confirming that AIS localization is not a tissue culture artifact. AIS-localized Kv2.1 is predicted to regulate both the frequency and back propagation of the axonal action potential.
Exogenously expressed Kv2.1 targets to the axon initial segment in cultured hippocampal neurons before accumulating on the soma
Endogenous Kv2.1 also preferentially localizes to the AIS
Heterogeneity within the culture system
AIS localized Kv2.1 clusters are stable structures containing mobile, but diffusion limited, channels
Kv2.1 localizes to hippocampal and cortical AIS domains in the intact rat brain
Any Kv2.1 localization to the AIS in cultured hippocampal neurons has been proposed to be a simple tissue culture artifact . However, the data presented above indicate that both transfected and endogenous Kv2.1 often preferentially target to the AIS prior to accumulating at somatic cluster sites. Most importantly, we routinely detected AIS localization of Kv2.1 in the hippocampus and cortex of adult rats, suggesting a role for Kv2.1 in regulating axon excitability in vivo. The AIS-localized Kv2.1 channels reside within clusters similar to those found on the soma in that mobile channels are corralled within a perimeter fence . However, the clusters themselves appear to be more stable within the AIS as compared to their cell body counterparts, consistent with ideas that cytoskeletal structures restrict the diffusion of membrane components within the AIS [16, 17].
Our results indicate that transfected Kv2.1 accumulates at the AIS of hippocampal neurons prior to other cell surface sites (Fig. 1). Endogenous channel also concentrates at the AIS in 7 and 10 DIV neurons (Fig. 3). Given that the Kv2.1 surface clusters are sites where trafficking vesicles deliver channel to the cell surface , it is possible that vesicles containing newly synthesized Kv2.1 are delivered first to the AIS. A fraction of the AIS localized channel could then be internalized and redistributed over the soma. Another possibility is that Kv2.1 is inserted into the soma membrane and then diffuses into the AIS clusters where it becomes trapped. Such a diffusion-trap mechanism has been proposed for voltage-gated Na+ channels that are AIS localized . Alternatively, perhaps delivery occurs equally at the AIS and soma sites but since the AIS is the only neurite receiving the nascent channel, and the newly inserted channel is diluted over all the soma sites, we are left with only the appearance of selective AIS insertion. Future experiments will distinguish between these possibilities.
More than 87% of 14 DIV neurons demonstrate endogenous Kv2.1 concentration at the AIS. Why this localization has not been observed by other investigators is unclear. Perhaps subtle differences in tissue culture conditions could account for the discrepancy. Previous Kv2.1 localizations in adult brain, as performed by other investigators [3, 8] and ourselves (unpublished results), also failed to detect Kv2.1 expression in the AIS. It is possible that the AIS localization was missed because without simultaneous ankyrinG staining and high resolution confocal imaging, it is difficult to identify the AIS structure within the dense layers of neuronal cell bodies in the hippocampus. The finding that not all hippocampal neurons show Kv2.1 localization in the AIS agrees with the Fig. 4 data showing a lack of AIS localization in GAD-positive inhibitory neurons.
While a high concentration of sodium channels at the AIS ensures axonal action potential generation , it is not clear what role Kv2.1 plays in this specialized cell surface domain. Kv2.1 has slow activation kinetics and is therefore unlikely to contribute significantly to repolarization during a single action potential. However, Kv2.1 could regulate axonal excitability following trains of action potentials, for studies by McBain and co-workers indicate that reduction of Kv2.1 in pyramidal neurons results in a broadening of the action potential waveform . Recently, the Kv2.2 channel, which cannot form heterotetrameric complexes with Kv2.1, has been localized to the AIS of neurons in the medial nucleus of the trapezoid body . Kv2.2, which is functionally similar to Kv2.1, is proposed to enhance the recovery of AIS sodium channels from the inactive state by hyperpolarizing the membrane potential during repetitive action potential firing . Kv2.1 could be performing a similar function. In addition, concentration of Kv2.1 at the AIS could regulate the back-propagation of depolarization into the soma. Given that multiple phosphorylation sites within the Kv2.1 C-terminus influence the voltage-dependence of activation , the phosphorylation of AIS-localized Kv2.1 provides a mechanism to regulate axonal excitability.
Primary hippocampal cell cultures
Neurons from embryonic day 18 rat pups were cultured as previously described . Briefly, neurons from cryo-preserved E18 rat hippocampal dissociations were plated at a density of ~15,000 – 30,000 cells/cm2 on poly-D-lysine coated glass-bottom dishes (MatTek) in glial-cell conditioned neurobasal medium (GCM) containing B27 supplement (Invitrogen) . Every 3–4 days after plating, one-half of the culture medium was replaced with GCM. Neurons plated on 35 mm glass bottom dishes and cultured for 7–10 days were transfected with 2.0 μl of Lipofectamine 2000 and 0.75 μg of Kv2.1-expressing plasmid DNA in 100 μl OptiMem (Invitrogen). Two-hours after transfection, the culture medium was replaced with fresh Neurobasal/B27. The expression plasmid constructs containing fluorescent protein and epitope-tagged Kv2.1 have been previously described [13, 14, 22, 29]. In brief, EGFP was placed onto the Kv2.1 N-terminus and two HA epitopes were inserted in the extracellular loop between S1 and S2 transmembrane domains.
Live cell imaging
All image acquisition was performed using an Olympus Fluoview 1000 confocal microscope system as previously described [13, 14]. For live cell acquisition, the objective and microscope stage were maintained at 37°C. Prior to imaging, media was replaced with pre-warmed Hepes-buffered (25 mM) Neurobasal medium containing B-27 supplement (Invitrogen). GFP was excited using the 488 nm line of an Ar laser set at 0.1 – 0.5% transmission and emission collected using the variable bandpass filter set at 500–550 nm. A 60×, 1.4 NA oil immersion objective was used for imaging and the pinhole diameter set for 1 Airy Unit. For each image, the detector voltage was adjusted as necessary to utilize the full 12-bit linear range. For the imaging of individual Kv2.1-containing clusters, an optical zoom of 8–10 × was often used. Images were acquired every 0.5 – 120 seconds as indicated at either 512 × 512 or 1024 × 1024 resolution. Cells were imaged for less than one hour on the microscope stage.
where Ai is the amplitude of each component, t is time and τi is the time constant of each component.
Detection of surface GFP-Kv2.1-HA was performed by incubating live cells with a 1:2000 dilution of anti-HA Alexa-594 conjugated monoclonal antibody (Molecular Probes) in Hepes-buffered Neurobasal/B27 for a minimum of 30 min at 37°C. Only live cells were used for surface channel detection since fixation permits labeling of internal, non-surface channel (data not shown). For endogenous Kv2.1, ankyrinG and MAP2 immuno-staining, neurons were fixed in 4% formaldehyde, 4% sucrose in PBS for 15 min, incubated in 0.5% CHAPS, blocked in 5% non-fat milk and 1% goat serum in PBS, and labeled with the indicated antibody diluted in 1% BSA in PBS. Antibody and dilutions were as follows: affinity-purified rabbit anti-Kv2.1 polyclonal antibody (Upstate Biotechnology, 1:200 dilution, raised against amino acids 837–853); mouse monoclonal anti-MAP-2 (Sigma, 1:2000); mouse monoclonal anti-AnkyrinG (Zymed, 1:100, raised against the spectrin binding domain); and mouse monoclonal anti-α-glutamic acid decarboxylase (Developmental Studies Hybidoma Bank, 1:50). Goat anti-mouse and goat anti-rabbit secondary antibodies conjugated to Alexa 488 or 594 (Molecular Probes) were diluted 1:1000 in 1% BSA/PBS.
Hippocampal slices were prepared from post natal day 21 and adult rats. In accordance with university guidelines, the animals were anesthetized with ketamine and decapitated. After decapitation, the brains were removed and fixed with 4% paraformaldehyde (PFA) for 4 hours at 4°C, replacing the PFA solution after 2 hours. The tissue was cryoprotected in 30% sucrose for 2 days at 4°C. Following cryoprotection and freezing, sagittal sections (16 μm) were prepared on a cryostat microtome. In order to diminish non-specific binding, the tissue sections were incubated overnight in 1 × PBS/5% dry milk/10% goat serum at 4°C. The detection of endogenous Kv2.1 and ankyrinG was performed by incubating the tissue sections with affinity-purified anti-Kv2.1 polyclonal (1:500, Upstate Biotechnology) and anti-Ankyrin G monoclonal (1:500, Zymed) antibodies diluted in 1 × PBS/10% goat serum/0.5% CHAPS overnight at 4°C. Following primary antibody incubation, brain slices were washed three times with 1 × PBS/0.5% CHAPS. The tissue slices were incubated with Alexa 488-conjugated goat anti-rabbit secondary (1:500, Invitrogen) and Alexa 594-conjugated goat anti-mouse secondary (1:500, Molecular Probes) for 1 hour at room temperature. Following secondary antibody incubation, the brain slices were washed three times with 1 × PBS/0.5% CHAPS and mounted using AquaPolymount (Polysciences).
Specificity of antibody binding
The detection of MAP2, ankyrinG, GAD and Kv2.1 in cultured hippocampal neurons has been performed by multiple groups using either the same antibodies as used here or closely related ones [2, 7, 18, 20, 30, 31, 32, 33, 34, 35, 36]. Importantly, the immunolocalization patterns for endogenous Kv2.1 in the hippocampal cultures are identical to the patterns observed with GFP-Kv2.1-HA under GFP optics or with anti-HA immuno-staining. With respect to the anti-HA antibody specificity, cell surface staining was never observed in GFP-Kv2.1-HA free neurons.
An additional control is required for the Kv2.1 localization to the ankyrinG defined AIS in brain tissue sections, as this has not been reported previously. Since immunogen block controls are known to be misleading , we also performed immuno-localization of ankyrinG and Kv2.1 using a monoclonal antibody for Kv2.1 and a polyclonal antibody for ankyrinG. A mouse monoclonal against Kv2.1 (1:500 dilution, Upstate Biotechnology, raised against a GST fusion protein containing amino acids 509–853) was used in conjunction with a rabbit anti-ankyrinG polyclonal (1:500 dilution, Santa Cruz Biotechnology, raised against amino acids 4163–4377). While the immuno-staining was not as robust as that observed with the other antibody combination, the same AIS staining patterns for ankyrinG and Kv2.1 were observed as shown in Additional file 4. It is highly unlikely that these distinct anti-Kv2.1 antibodies both nonspecifically bind AIS structures, especially since GFP-Kv2.1-HA also targets to this region in transfected neurons.
Offline image analysis was done using the Olympus FV1000 software (version 1.03) and Volocity 4.4 (Improvision, Lexington, MA). Data analysis and curve-fitting was performed with SigmaPlot 8 (Systat, Point Richmond, CA) or IgorPro 5.03 (Wavemetrics, Portland, OR). Images were filtered in Volocity using a 3 × 3 median filter. Compilation of images was performed using Adobe Photoshop and contrast and brightness adjustments were made. Both compressed (maximum projection) and single Z-section images are displayed as indicated.
This research was supported by NIH grants NS41542 and HL49330 to MMT and training grant NS43115 to PDS. The anti-GAD was obtained from the Developmental Studies Hybridoma Bank developed under the auspices of the NICHD and maintained by The University of Iowa, Department of Biological Sciences, Iowa City, IA 52242. The authors thank Dr. Noreen Reist for comments on the manuscript.
- 3.Du J, Tao-Cheng JH, Zerfas P, McBain CJ: The K+ channel, Kv2.1, is apposed to astrocytic processes and is associated with inhibitory postsynaptic membranes in hippocampal and cortical principal neurons and inhibitory interneurons. Neuroscience. 1998, 84: 37-48. 10.1016/S0306-4522(97)00519-8.CrossRefPubMedGoogle Scholar
- 10.Misonou H, Mohapatra DP, Trimmer JS: Kv2.1: A Voltage-Gated K(+) Channel Critical to Dynamic Control of Neuronal Excitability. Neurotoxicology. 2005Google Scholar
- 15.Kusumi A, Nakada C, Ritchie K, Murase K, Suzuki K, Murakoshi H, et al: Paradigm shift of the plasma membrane concept from the two-dimensional continuum fluid to the partitioned fluid: high-speed single-molecule tracking of membrane molecules. Annu Rev Biophys Biomol Struct. 2005, 34: 351-378. 10.1146/annurev.biophys.34.040204.144637.CrossRefPubMedGoogle Scholar
- 16.Nakada C, Ritchie K, Oba Y, Nakamura M, Hotta Y, Iino R, et al: Accumulation of anchored proteins forms membrane diffusion barriers during neuronal polarization. Nat Cell Biol 2003 Jul; 5(7): 626–32. 2003, 5: 626-632. 10.1038/ncb1009.Google Scholar
- 19.Boiko T, Vakulenko M, Ewers H, Yap CC, Norden C, Winckler B: Ankyrin-dependent and -independent mechanisms orchestrate axonal compartmentalization of L1 family members neurofascin and L1/neuron-glia cell adhesion molecule. J Neurosci. 2007, 27: 590-603. 10.1523/JNEUROSCI.4302-06.2007.CrossRefPubMedGoogle Scholar
- 23.Fache MP, Moussif A, Fernandes F, Giraud P, Garrido JJ, Dargent B: Endocytotic elimination and domain-selective tethering constitute a potential mechanism of protein segregation at the axonal initial segment. Journal of Cell Biology. 2004, 166: 571-578. 10.1083/jcb.200312155.PubMedCentralCrossRefPubMedGoogle Scholar
- 24.Colbert CM, Pan E: Ion channel properties underlying axonal action potential initiation in pyramidal neurons. Nat Neurosci 2002 Jun; 5(6): 533–8. 2002, 5: 533-538. 10.1038/nn0602-857.Google Scholar
- 25.Johnston J, Griffin SJ, Baker C, Skrzypiec A, Chernova T, Forsythe ID: Initial segment Kv2.2 channels mediate a slow delayed rectifier and maintain high frequency action potential firing in MNTB neurones. J Physiol. 2008Google Scholar
- 28.Meberg PJ, Miller MW: Culturing hippocampal and cortical neurons. Neurons: Methods and Applications for the Cell Biologist. 2003, 71: 111-127.Google Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.