Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Neurochem. Author manuscript; available in PMC 2010 March 1.
Published in final edited form as:
PMCID: PMC2821660

A Novel Choline Cotransporter Sequestration Compartment in Cholinergic Neurons Revealed by Selective Endosomal Ablation


The sodium-dependent, high affinity choline transporter -choline cotransporter- (ChCoT, aka: cho-1, CHT1, CHT) undergoes constitutive and regulated trafficking between the plasma membrane and cytoplasmic compartments. The pathways and regulatory mechanisms of this trafficking are not well understood. We report herein studies involving selective endosomal ablation to further our understanding of the trafficking of the ChCoT. Selective ablation of early sorting and recycling endosomes resulted in a decrease of ~75% of [3H]choline uptake and ~70% of [3H]hemicholinium-3 (HC-3) binding. Western blot analysis showed that ablation produced a similar decrease in ChCoTs in the plasma membrane subcellular fraction. The time frame for this loss was approximately 2 hrs which has been shown to be the constitutive cycling time for ChCoTs in this tissue. Ablation appears to be dependent on the intracellular cycling of transferrin conjugated horseradish peroxidase (Tf-HRP) and the selective deposition of Tf-HRP in early endosomes, both sorting and recycling. Ablated brain slices retained their capacity to recruit via regulated trafficking ChCoTs to the plasma membrane. This recruitment of ChCoTs suggests that the recruitable compartment is distinct from the early endosomes. It will be necessary to do further studies to identify the novel sequestration compartment supportive of the ChCoT regulated trafficking.

Keywords: choline transport, compartmental ablation, recruitment, transporter trafficking

The presynaptic high affinity choline transporter found at the cholinergic synapse has an obligatory requirement for the cotransport of sodium (Yamamura and Snyder 1972; 1973; Haga and Noda 1973; Simon et al. 1976), thus it is a symporter of choline and sodium (i.e. choline cotransporter, ChCoT aka: cho-1, CHT1, CHT; Okuda et al. 2000; Apparsundaram et al. 2000). Choline transporter activity has been shown to be rate-limiting in acetylcholine (ACh) synthesis (Birks and MacIntosh 1961; Roskoski 1978; Newkirk et al. 1980). High frequency stimulation or its simulation with pre-exposure to depolarizing concentrations of potassium of cholinergic tissues results in a substantial increase in choline transport (Murrin and Kuhar 1976; Higgins and Neal 1982; Ferguson et al. 1994; Ford et al. 1999). Hemicholinium-3 (HC-3) binding studies in tissues have been used to confirm that the increase in transport is paralled by a comparable increase in the numbers of ChCoTs in the plasma membrane. In our laboratory use of a plasmalemmal, impermeant, irreversible HC-3 mustard derivative demonstrated that the Limulus brain ChCoT traffics constitutively into and out of the plasma membrane (Ivy et al. 2001). Cell-surface biotinylation and confocal microscopy were used in both SN56 and HEK293 cultured cell lines, to show that the choline cotransporter constitutively and rapidly internalized from the cell surface in clathrin-coated vesicles and was directed to early endosomes via a clathrin-dependent endocytic trafficking mechanism (Ribeiro et al. 2005). Moreover, recent studies by Ribeiro and colleagues (Ribeiro et al. 2007) provide evidence that the choline transporter recycles back to the plasma membrane both in a constitutive and a regulated mode.

The selective endosomal ablation strategy takes advantage of the known trafficking pattern of the transferrin receptor (TfR) through the early endosomal pathway (Mellman 1996). TfRs at the cell surface upon binding its target, transferrin (Tf) protein, undergoes endocytosis. The TfR-Tf complex traffics via the early sorting and early recycling (trafficking) endosomes back to the plasmalemma. Selective endosomal ablation is based on the specific accumulation of TfR-Tf-horseradish peroxidase (HRP) complex in the early endosomal pathway (Sheff et al, 2002). The selective endosomal ablation protocol has been used in the study of the insulin responsive glucose transporter (GLUT4) trafficking (Livingstone et al. 1996; Hashiramoto and James 2000). More recent studies have implicated a sequestration compartment within adipocytes for GLUT4 that survives TfR-Tf-HRP mediated endosomal ablation (Zeigerer et al. 2002; Karylowski et al. 2004). In the studies reported here, we evaluated the effect on the surface distribution of the ChCoTs following the application of the selective endosomal ablation protocol in Limulus brain hemi-slices. Ablated brain slices were used to examine indirectly, both constitutive and regulated trafficking of the ChCoT. Choline uptake measurements, HC-3 binding determinations, and Western blotting were done to assess changes in choline uptake and the redistribution of the ChCoT.

Materials and methods

Animals and tissue preparation

Horseshoe crabs (Limulus polyphemus) of either sex were purchased from the Marine Biological Laboratory (Woods Hole, MA), and were maintained in moist excelsior at 4-8°C until the time of use. Animals averaging 15-30 cm across the carapace were used in all experiments. The corpora pedunculata and protocerebrum complex were removed, placed in normal Chao's (440 mM NaCl; 9 mM KCl; 37 mM CaCl2; 10 mM HEPES; pH 7.4) (Chao, 1933), divided into hemispheres, mounted with Tissue Tek onto a Campden vibroslicer stage, and placed at -10 to -15°C for 5 min before slice preparation. Brain hemi-slices (500 μ thickness) were prepared using a vibroslicer, blotted, weighed, and allowed to equilibrate in normal Chao's at room temperature (RT) prior to either choline transport or HC-3 binding experiments. Data points for all experiments were obtained in at least triplicate, and statistical significance was assessed using a Student's t-test.

Compartmental ablation strategy

The TfR appears to be a ubiquitous integral membrane protein that has been shown to bind its ligand, Tf, and the complex is rapidly endocytosed. This complex traffics via the early sorting (ES) endosome and the early recycling (ER) endosome compartments back to the plasma membrane (Hopkins and Trowbridge, 1983; Presley et al., 1993; Mellman, 1996). Notably, Sheff et al., (2002) have elucidated the detailed trafficking of the TfR using both intact and microsurgically manipulated cytoplasts. In these studies, they show that the primary trafficking of TfR back to the plasma membrane occurs directly from the ES endosome (74%), while a minor component (26%) travels from the ES endosome to the ER endosomal compartment enroute to the surface membrane. A unique aspect of either the TfR-Tf or the substituted TfR-Tf-HRP complex exploited in the endosomal ablation strategy is that the TfR-Tf-HRP complex remains intact until it resurfaces at the plasma membrane. At equilibrium both early endosomal compartments are saturated with the TfR-Tf complex (Sheff et al., 2002). The strategy for selective endosomal ablation relies upon the saturation of these compartments with TfR-Tf-HRP complex and the subsequent treatment with 3, 3′-diaminobenzidine (DAB, 2 mg/ml, Vector) and hydrogen peroxide (H2O2, 0.02%) in a buffer solution (Livingstone et al. 1996). As stated in the paper of Livingstone et al., the technique cross-links and renders insoluble (“ablate”) intracellular compartments containing the TfR by means of a Tf-HRP conjugate. The premise of the selective endosomal ablation strategy is presented in Fig. 1. In our study, selective endosomal ablation was achieved by an initial pre-incubation of brain hemi-slices in the presence Tf-HRP conjugate in normal Chao's solution for 30 min at 37°C. Following this 30 min incubation, one ml of ice-cold quenching buffer (sodium citrate in buffered Chao's) was added to each slice. The slices were aspirated and subsequently washed with several (1 ml) volumes of ice-cold buffered Chao's and then incubated in Chao's containing DAB plus H2O2 for 1 h in the dark at 4-8°C. Control slices were incubated in Chao's containing DAB without H2O2. Fig. 1a models the predicted trafficking pattern of the TfR-Tf-HRP complex, and Fig. 1b models the consequence of the ablation of the ES endosome and ER endosome on subsequent trafficking involving these early endosomes.

Figure 1Figure 1
Intracellular trafficking model for the Transferrin receptor. (a) The Transferrin receptor (TfR) with bound Transferrin (Tf) cycles via the early sorting (ES) endosome and the early recycling (ER) endosome back to the surface membrane. Additional endosomes ...

Choline transport and Hemicholinium-3 specific binding measurements

The brain hemi-slices were prepared as described above. The initial set of experiments were designed to measure [3H]choline uptake for determination of the time course following completion of the 1 hr ablation or control (DAB with no H2O2) treatment. After this treatment, the hemi-slices were washed in normal Chao's for different times (10, 30, 60, 90, 120 and 180 min) and transferred to microtubes containing either normal Chao's or normal Chao's + 200 μM hemicholinium-3 (HC-3) in the presence of 1 μM [3H]choline (86.6 Ci/mmole, New England Nuclear) and incubated for 30 min at RT. Subsequently each slice was washed, solubilized, and Polyfluor (Packard) was added to determine the radioactive content by liquid scintillation spectrometry. Low affinity choline transport was defined as uptake observed in the presence of 200 μM HC-3. The post-ablation high affinity choline transport was calculated by subtracting the low affinity choline uptake from total choline uptake. Transport is expressed as picomoles of [3H]choline per mg of tissue per unit time.

In experiments designed to assess HC-3 binding, following the 1 hr ablation or control treatment, hemi-slices were washed with normal Chao's for either 10 or 120 min at RT. Next, the slices were incubated in normal Chao's containing either 10 nM [3H]HC-3 (120 Ci/mmol) in the absence (total binding) or presence of 200 μM unlabeled HC-3 (non-specific binding) for 30 min at RT. Incubations were terminated by rapidly washing the tissues with ice-cold normal Chao's, and samples were prepared for liquid scintillation counting as described above. Specific binding was determined as the difference between total binding and non-specific binding. Post-ablation specific binding is expressed as femtomoles of [3H]HC-3 bound per mg of tissue per unit time.

The recruitment of ChCoTs to the plasma membrane was achieved by the established protocol in which brain hemi-slices were subjected to antecedent exposure to depolarizing concentrations of potassium chloride (AnEK) in Chao's. Hemi-slices were pre-incubated in either normal Chao's (control) or 120 mM K+ Chao's for 15 min at RT, washed in two five minute exchanges in normal Chao's, before incubation with [3H]choline. After washing slices were incubated in 1 μM [3H]choline in normal Chao's for 30 min at RT. Subsequently, the radioactive content of each slice was measured as previously described. Uptake was calculated in pmoles/mg of tissue/30 min. HC-3 binding was assessed in either control or AnEK pretreated slices washed in Chao's, and then incubated in [3H]HC-3 (10 nM) in Chao's to determine the surface population of transporters. Specific binding was determined by assessing [3H]HC-3 binding in the presence and absence of 200 mM unlabelled HC-3. Following normal Chao's washes, the radioactive content of each slice was determined by liquid scintillation spectrometry.

To determine whether or not the recruitable pool of ChCoTs resides within the set of early endosomes, the AnEK challenge and compartmental endosomal ablation were used. Measurement of [3H]choline uptake and the determination of [3H]HC-3 binding were assessed in both ablated brain hemi-slices and ablated slices challenged subsequently with AnEK. Compartmental ablation was achieved as previously described. Ablated hemi-slices were washed and incubated with either [3H]choline or [3H]HC-3. Choline uptake and HC-3 binding were measured as mentioned above. The effect of AnEK challenge on ablated slices was determined following a 2 hr wash. Ablated and washed slices were incubated for 15 min in Chao's containing 120 mM KCl, and subsequently placed in two 5 min washes of normal Chao's. In separate preparations of ablated and AnEK challenged hemi-slices, either [3H]choline uptake or [3H]HC-3 binding was measured. In a parallel set of experiments, these measurements were done in hemi-slices subjected to both a pre-ablation AnEK challenge followed by a post-ablation challenge.

Subcellular fractionation

Limulus brain hemi-slices were subjected to a modified fractionation procedure (Simpson et al., 1983). Brain hemi-slices were homogenized in 20 mM Hepes, 250 mM sucrose buffer (pH 7.4) containing protease inhibitor cocktail (Sigma-Aldrich) with a Dounce homogenizer using 12 up and down strokes on ice. The homogenate was subjected to a centrifugation of 940 × g for 10 min at 4°C. The supernatant was then recentrifuged at 16,100 × g for 15 min to produce crude plasma membrane (PM) and microsomal membrane (MM) fractions. From these fractions proteins were extracted at 4-8°C for 1 h using 20 mM Hepes buffer containing 1% Triton X-100. All fractions were frozen quickly in isopentane and stored at -80°C until used.

Custom antibody for Limulus choline cotransporter

An antibody raised against the Limulus ChCoT was custom developed by Invitrogen Life Technologies Corporation (Zymed Laboratories; Carlsbad, CA). The epitope used in raising the antibody consisted of an 18 amino acid (aa) sequence from the carboxyl-terminal (C-18) of the cloned Limulus ChCoT (Wang et al. 2001). The polyclonal antibody was raised in rabbit and was affinity purified. A fusion peptide (FP) was constructed as a positive control. The FP consisted of 77 aa from the C-terminal of the Limulus ChCoT. The recombinant peptide was expressed in pGEX-6P-1 vector (Amersham), subcloned via restriction enzyme Bam H1 and Not 1. Purification was completed using Glutathione Sepharose 4B. Samples of the Limulus ChCoT fusion peptide (FP) were dissolved in homogenate buffer and placed on isopentane for storage until further analysis. These samples were used as positive controls in assessing the specificity of the custom antibody. The primary antibody dilution was made by placing anti-ChCoT a volume 1:5000 in SuperBlock Blocking TBS buffer (Pierce) and maintained at 0-4°C until used. The preabsorbed antibody dilution was made by combining equal volumes of anti-ChCoT and C-18 peptide at a ratio of 1:20, and shaking gently overnight at 0-4°C. One volume of the preabsorbed anti-ChCoT was diluted at 1:2500 in SuperBlock, and maintained at 0-4°C until used. Secondary antibody (goat anti-rabbit-IgG-HRP; Vector Laboratories, Inc.) dilution was 1:300,000 in SuperBlock, and maintained at 0-4°C until used.

FP samples and protein extracts from Limulus brain tissues (PM and MM fractions) stored in the presence of protease inhibitor cocktail (Sigma-Aldrich), were separated by SDS-PAGE using 10% polyacrylamide pre-cast gels. Proteins were electrophoretically transferred to nitrocellulose membranes (0.22 μm, Osmonics Inc.). The nitrocellulose membranes were blocked for 1 hr in SuperBlock buffer. Antigens were detected by incubation of the membranes with diluted primary antibody or diluted preabsorbed antibody. The membranes were incubated for 1 hr with shaking at RT or at 4-8°C overnight. Membranes were subsequently placed in six 5 min washes of blocking buffer, following by incubation for 45 min in secondary antibody solution with gentle shaking at RT. After five washes with TBS, the immune complexes were visualized by chemiluminescence (Supersignal West Femto Maximum Sensitivity Substrate, Pierce) using an UVP imaging system. To quantify the amount of ChCoT detected in the PM and MM fractions, the band densities were measured with the UVP imaging software for optical density determinations.


Time course for the loss of surface choline transporters following ablation

As previously presented, the selective endosomal compartment ablation strategy is predicated on disrupting the equilibrium between the exocytotic insertion and the endocytotic retrieval of ChCoTs in the plasma membrane. Thus, the expectation was that the reduction of surface transporters would be time-dependent. To determine the time course of the loss or retrieval of plasmalemmal ChCoTs following the ablation treatment, we examined the rate of [3H]choline uptake at varying times following the ablation paradigm. The results are presented in Fig. 2. Surface transporters, as measured by changes in [3H]choline uptake, declined exponentially with the maximal effect (~80% reduction) occurring at approximately two hours post-ablation. This two hour time course agrees well with the constitutive cycling time previously reported for the ChCoT (Ivy et al. 2001). To assess the possible immediate effect of the ablation protocol on the distribution of the surface transporters, we measured [3H]HC-3 specific binding as well as [3H]choline uptake at two time points post-ablation, 10 and 120 min (Table 1). The results show that at 10 min, there were no measurable differences in [3H]HC-3 binding nor [3H]choline uptake between controls and tissues subjected to the ablation protocol. Conversely, at two hours, there was a highly significant reduction in both HC-3 binding (67%) and choline uptake (76%). These results show that the decrease in choline uptake in selective ablated brain hemi-slices is accompanied by a comparable reduction in the number of surface transporters. Moreover, these results are consistent with the constitutive cycling of ChCoTs as previously reported by this laboratory (Ivy et al. 2001).

Figure 2
Time course for loss of plasma membrane choline cotransporters after ablation. Brain hemi-slices were pretreated with either Tf-HRP conjugate or normal Chao's solution (control) for 30 min at 37°C, after which hemi-slices were treated with DAB ...
Table 1
[3H]Choline Uptake and [3H]Hemicholinium-3 Specific Binding 10 and 120 minutes Post-Ablation

The requirement of transferrin conjugated with HRP for effective ablation

The selective compartmental ablation strategy relies upon the presence of the TfR which specifically transports Tf conjugated HRP (Tf-HRP) to the early endosomal compartments (Sheff et al. 2002). The Limulus TfR has not been cloned nor identified. Thus, a set of experiments were done to show the essential requirement of tranferrin conjugated with HRP, therefore implicating a TfR or a TfR-like protein in Limulus brain hemi-slices. Brain tissues were treated with various configurations with and without HRP conjugated transferrin. Additionally, DAB was applied with and without H2O2. The results seen in Fig. 3 demonstrate the unequivocal requirement for the Tf-HRP conjugate to achieve compartmental endosomal ablation. These findings suggest that the HRP conjugated to transferrin is internalized and is accumulated selectively in the early endosomal compartments known to be involved in TfR trafficking back to the plasma membrane. Therefore, the chemical interaction between DAB and H2O2 with Tf-HRP aggregated selectively within the early endosomal compartments which results in the ablation of traffic through these cytosolic compartments (Sheff et al. 2002). Also, Sheff et al. used a kinetic analysis in determining that the TfR traffics through both the early sorting endosomes and early trafficking (recycling) endosomes enroute back to the plasma membrane.

Figure 3
The effects of HRP conjugated Transferrin on high affinity choline transport. Brain hemi-slices were pretreated with either Tf-HRP conjugate, unconjugated HRP, or normal Chao's solution (control) for 30 min at 37°C. After quenching and washing, ...

Choline transporter recruitment in brain hemi-slices confirmed by Western blotting

It is well established that high frequency stimulation results in the doubling of choline transport in cholinergic neurons (Birks and MacIntosh, 1961; Potter, 1970). High frequency stimulation can be simulated by a paradigm of a brief exposure (approximately 15 min) of nervous tissue to a depolarizing concentration of KCl followed by a 10 min wash, and a return of tissue to its normal physiologic bath. Following this treatment, choline uptake is doubled as are the number of ChCoTs in the plasma membrane. This experimental paradigm is referred to as the antecedent elevated potassium challenge (AnEK).

In order to visualize the recruitment source of the ChCoT to the cell surface, we used Western blot analysis of plasma and microsomal membrane extracts. The custom synthesized antibody was tested for specificity as described in the Materials and methods, and the results are shown in Fig. 4. The ChCoT antibody recognized a protein of approximately 70 kDa in both PM and MM fractions. It also interacted with the positive control fusion peptide (FP, ~35 kDa). Preabsorption of the antibody with the C-18 peptide blocked antibody interaction with the 70 kDa proteins in membrane extracts and also with the FP. These results confirm the specificity of the custom antibody. Furthermore, boiling the custom antibody for 10min produced the same blockage of antibody interaction with the 70 kDa proteins in membrane extracts as well as the FP (data not shown).

Figure 4
Specificity of custom choline cotransporter antibody. Western blot analysis of choline cotransporter (ChCoT) expression in Limulus brain hemi-slice extracts of plasma membrane (PM) and microsomal membrane (MM) fractions were performed using a polyclonal ...

We and others have shown that prior sustained depolarization of cholinergic nervous tissues lead to an increase of both the numbers of ChCoTs in the plasma membrane as well as a complementary enhancement of [3H]choline uptake by these tissues (Roskoski 1978; Newkirk et al. 1980; Ribeiro et al. 2003). Antecedent elevated KCl Chao's exposure results in a doubling of [3H]choline uptake compared to controls with a corresponding increase in [3H]HC-3 binding as shown in Fig 5a. Western blot analysis of PM and MM fractions confirm the increase in surface ChCoTs resulting from AnEK challenge (Fig. 5b). Others (Ferguson et al. 2003; Nakata et al. 2004) have shown and we confirm that the majority of the ChCoT distribution is intracellular. The apparent effect of the AnEK challenge is to redistribute the transporter from intracellular stores to the PM.

Figure 5
The effects of antecedent elevated potassium on high affinity choline transport and hemicholinium-3 specific binding. Brain hemi-slices were subjected to separate 15 min pretreatments in either normal Chao's (basal/control conditions) or 120mM potassium ...

The relationship of presumed early endosomal compartments to the recruitable compartment of choline transporters

Western blot analysis shows that the compartmental endosomal ablation treatment of Limulus brain hemi-slice preparations caused a significant decrease (~67%), in ChCoT levels in the PM fraction (Fig. 6). In an effort to assess the relationship of the ablated compartments with the recruitable population of ChCoTs, we subjected ablated tissues to an AnEK challenge. In tissues that were two hours post-ablation, significant reductions in both [3H]choline uptake (~80%) and [3H]HC-3 binding (~75%) were observed (Fig. 7, column 1). The AnEK challenge of these two hour post-ablated tissues resulted in a full restoration of ChCoTs to the surface membranes as determined by [3H]choline uptake and [3H]HC-3 binding (Fig. 7, column 2). To further examine a possible relationship of the storage compartment responding to the AnEK challenge with the endosomal compartments, we altered the experimental protocol. In these experiments, the tissues were initially challenged with AnEK, ablated, and following a two hour hold, subjected to a second AnEK challenge. These results are shown in Fig. 7, column 3. As can be seen in these results, the second AnEK challenge failed to restore the ablated preparations to their control or pre-ablated levels of either [3H]choline uptake or [3H]HC-3 binding. The apparent failure of a second AnEK challenge to recruit ChCoTs to the surface membrane could be explained by the presence of a sequestration compartment specific for the recruitable population of ChCoTs with limited storage capability, and thus requiring a kinetic link to early endosomes for replenishment.

Figure 6
The effect of ablation on the choline cotransporter subcellular distribution. Western blots shown for brain slices following either the control (no ablation) or ablation protocol in which tissues were treated with only DAB or DAB + H2O2 for 1 h at 4°C, ...
Figure 7
The effect of antecedent elevated potassium challenge on choline cotransport and hemicholinium-3 specific binding in ablated Limulus brain hemi-slices. First, brain hemi-slices were incubated with the Tf-HRP for 30 min at 37°C followed by DAB ...


Previously, the paradigm employing transferrin conjugated with HRP and DAB plus H2O2 cytochemistry to produce selective endosomal compartmental ablation has been used in studies of the trafficking of the insulin responsive GLUT4 transporter (Livingstone et al. 1996; Hashiramoto and James 2000; Zeigerer et al. 2002; Sheff et al. 2002). In the studies reported herein, we have employed this selective endosomal ablation strategy to assess ChCoT trafficking in Limulus brain hemi-slices. As depicted in Fig. 1, the experimental strategy is predicated on the deposition of Tf-HRP in selective endosomal compartments resulting from the established intracellular trafficking of the TfR upon binding its natural ligand Tf. The transferrin-HRP mediated ablation protocol applied to Limulus brain slices resulted in a maximal reduction (~80%) in [3H]choline uptake at two hours post ablation (Fig. 2). Furthermore, the results presented in Table 1 demonstrate that the endosomal ablation protocol has no immediate effect on the surface population of ChCoTs. The TfR family of proteins appears to be rather ubiquitious, having been found extensively in both vertebrates and invertebrates (Lambert et al. 2005; Lambert and Mitchell 2007). The TfR has not been cloned or identified in Limulus. The experimental protocol used to generate the data summarized in Fig. 3 was designed to assess the strict requirement of HRP conjugated to the transferrin (Tf) protein to yield results reflective of selective endosomal ablation. The data presented in Fig. 3 clearly establishes the essential requirement of HRP conjugated transferrin to achieve selective compartmental ablation as determined by cellular uptake of [3H]choline in Limulus brain slices. These results argue strongly for the presence of a transferrin receptor or a transferrin-like receptor protein in Limulus, and thus validate the use of this selective endosomal ablation paradigm in this preparation.

Numerous laboratories have confirmed an activity driven doubling of the transporter numbers in terminal membranes of cholinergic neurons (Mulder et al. 1974; Murrin and Kuhar, 1976; Roskoski, 1978; Newkirk et al. 1980; Higgins and Neal, 1982). High frequency stimulation can be simulated by antecedent exposure of cholinergic nerves to depolarizing concentrations of KCl (AnEK), and produces the same effects. To aid in the resolution of the source of the ChCoTs recruited to the PM, we employed a custom antibody for the Limulus ChCoT in Western blots of cellular membranes. The specificity of the antibody as resolved by antibody preabsorption and is shown in Fig. 4. The results shown in Fig. 5a confirm the doubling of transport activity following AnEK. These results reflecting the doubling of transport activity agrees well with both the change in [3H]HC-3 binding activity (~90%; Fig. 5a), and the changes in the relative optical density measurements in the Western blots (~89%; Fig. 5b). Moreover, the data presented in Fig. 5b confirm that the increase in the plasmalemmal distribution of the ChCoTs results from the redistribution of these transporters from a cytoplasmic population of membranes (i.e. microsomal membranes) to the plasma membrane (PM) fraction. Western blot analyses of ablated preparations reveal a 67% decrease in the PM distribution of ChCoTs (Fig. 6). These results agree well with the reductions in specific [3H]HC-3 binding as shown in Fig. 7, column 1. The reduction in the transporter distribution in the PM following selective compartment ablation is consistent with the view that these ChCoTs are continuously trafficking from the PM via cytoplasmic compartments back to the PM. The ablation strategy targets early endosomal compartments, and thus the selective loss of these compartments would agree with the results depicted in Fig. 6 and Fig. 7, column 1. On the other hand, the results summarized in Fig. 7, column 2, demonstrate that in ablated tissues in which the presumed early endosomes, central to continuous constitutive trafficking, are rendered non-functional, ChCoTs are recruitable from a yet to be described compartment. In a variation of the initial experimental paradigm in which an AnEK challenge preceded compartmental ablation treatment, a second AnEK challenge following ablation failed to recruit ChCoTs to the PM (Fig. 7, column 3). The results presented in Fig. 7, column 3, suggest a kinetic relationship between the endosomal compartments and the special compartment that houses the recruitable population of ChCoTs. Ribeiro and colleagues used confocal microscopic observations of a choline transporter-hemagglutinin to show that the complex traffics from the PM to endocytic organelles (Ribeiro et al. 2003; 2005). Our laboratory employed an irreversible ligand of the ChCoT, a mustard derivative of HC-3, to show that ChCoTs constitutively traffics from the PM and back in approximately two hours (Ivy et al., 2001). The recruitment of the ChCoT with high frequency stimulation (using AnEK challenge) represents the regulated trafficking of the transporter. The results summarized in Fig. 7, are indicative of separate pathways serving the dual trafficking of ChCoTs, a pathway for constitutive trafficking that appears to involve early endosomes, and a regulated trafficking pathway that is primarily distinct from the early endosomes.

The observation that high frequency stimulation or antecedent depolarization with elevated potassium results in the recruitment of choline transporters to the PM has led to various speculations as to the underlying mechanism. One view expressed to explain high frequency stimulation and/or the AnEK challenge recruitment of choline transporters to surface membranes is that the transporter resides in synaptic vesicles. High frequency stimulation or antecedent exposure of cholinergic tissue to depolarizing concentrations of KCl would cause the increased release of neurotransmitter resulting from an increased fusion of synaptic vesicles with the terminal membrane. Concomitantly, the increased fusion of synaptic vesicles with the terminal membrane would also result in the deposition of transporter proteins into the plasma membrane (Ferguson et al. 2003; Ribeiro et al. 2003; 2006). Several studies have been reported that suggests that the choline transporter traffics in synaptic vesicles (Ferguson et al. 2003; Ribeiro et al. 2003; Nakata et al. 2004). On the other hand there are reports that argue for caution in the interpretation of morphological evidence that all vesicles at the presynaptic terminal fit the classical definition of synaptic vesicles (i.e. vesicles containing neurotransmitter molecules). In studies of the plasma membrane associated GABA transporter (GAT1), Deken et al. (2003), isolated the trafficking vesicle containing GAT1, and showed that these vesicles had the physical appearance of synaptic vesicles, and contained several proteins common to synaptic vesicles. However, these vesicles were not synaptic vesicles and did not contain synaptophysin, a protein common to synaptic vesicles. Neither did these vesicles contain GABA, the neurotransmitter associated with GAT1. These authors concluded that GAT1 resides in a vesicle morphologically similar to neurotransmitter-filled vesicles, however, they are not the classical synaptic vesicles, but rather synaptic-like vesicles.

The insulin responsive GLUT4 transporter is agruably the most extensively studied membrane transporter. The ChCoT has been shown to be a member of the sodium dependent glucose transporter family, and thus related to GLUT4 (Wang et al. 2001). Selective endosomal ablation experiments were used in the study of the trafficking of the GLUT4 (Livingstone et al. 1996). Results revealed two GLUT4 intracellular compartments, one is TfR positive and one is devoid of TfR. From these studies it was concluded that the sequestration compartment surviving ablation is distinct from early endosomes. Bogan et al. (2003) published evidence for a non-endosomal cytoplasmic tethering mechanism involving a protein they termed TUG (tethering for the ubiquitin domain of the GLUT4). Our results are consistent with a model in which a “specialized” sequestration compartment, containing the activity driven pool of ChCoTs, survives the ablation protocol (Fig. 8). The results that the pre-ablation antecedent elevated potassium challenge prevents a post ablation AnEK challenge from restoring the Limulus brain slices to the pre-ablation levels suggests that the “specialized” compartment for the ChCoT has a limited capacity, and requires intact, functional early endosomes for replenishment. In summary, we report that a storage/recruitment ChCoT pool persists following compartmental endosomal ablation driven by the early endosomal trafficking of the TfR-Tf-HRP complex. This result implies that a special compartment different from the classical early endosomes exists in cholinergic neurons and may be responsible for the sequestration of the activity driven ChCoTs. This population of transporters is required for adjusting cholinergic neuron output to meet activity driven demands. It is interesting to speculate on the status of this essential compartment in pathological states. Additional studies will be required to determine the physiological properties as well as the cellular location and structural nature of this compartment.

Figure 8
Model for the dual (constitutive and regulated) trafficking of the choline cotransporter. The model depicts pathways through which the ChCoT may traffic from and back to the cell surface. In the basal state, endocytosis of ChCoT vesicles fuse with early ...


The authors thank Dr. Purnima Ghose for technical assistance. Mrs. Carolyn Davis and Mr. Reginald Cannon for manuscript preparation. Also, we thank Drs. Brenda S. McAdory, and Sanika Chirwa for useful comments. This work was supported by NIH grants RR03032, MH57067 and P20MD000261.

Abbreviations used

antecedent elevated potassium
cho-1, CHT1, CHT, choline transporter
early recycling endosome
early sorting endosome
fusion peptide of ChCoT
gamma amino butyric acid
GABA transporter-1
glucose transporter-4
hydrogen peroxide
horseradish peroxidase
insulin-regulated aminopeptidase
late endosome
microsomal membranes
plasma membrane
transferrin receptor
tethering protein for ubiquitin domain of GLUT4
vesicular ACh transporter


  • Apparsundaram S, Ferguson SM, George AL, Jr, Blakely RD. Molecular cloning of a human hemicholinium-3-sensitive choline transporter. Biochem Biophys Res Commun. 2000;276(3):862–867. [PubMed]
  • Birks R, MacIntosh FC. Acetylcholine metabolism of a sympathetic ganglion. J Biochem Physiol. 1961;39:787–827.
  • Bogan JS, Hendon N, McKee AE, Tsao T, Lodish HF. Functional cloning of TUG as a regulator of GLUT4 glucose transporter trafficking. Nature. 2003;425:727–733. [PubMed]
  • Chao I. Action of electrolytes on the dorsal median nerve cord of the Limulus heart. Biol Bull Mar Biol Lab Woods Hole. 1933;64:358–381.
  • Deken SL, Wang D, Quick MW. Plasma membrane GABA transporters reside on distinct vesicles and undergo rapid regulated recycling. J Neurosci. 2003;23:1563–1568. [PubMed]
  • Ferguson SSG, Rylett RJ, Collier B. Regulation of rat brain synaptosomal [3H]hemicholinium-3 binding and [3H]choline transport sites following exposure to choline mustard aziridinium ion. J Neurochem. 1994;63:1328–1337. [PubMed]
  • Ferguson SM, Savchenko V, Apparsundaram S, Zwick M, Wright J, Heilman CJ, Yi H, Levey AI, Blakely RD. Vesicular localization and activity-dependent trafficking of presynaptic choline transporters. J Neurosci. 2003;23(30):9697–9709. [PubMed]
  • Ford BD, Ivy MT, Mtshali CP, Townsel JG. The involvement of protein kinase C in the regulation of choline cotransport in Limulus. Comp Biochem Physiol. 1999;123A:255–261. [PubMed]
  • Haga T, Noda H. Choline uptake systems of rat brain synaptosomes. Biochim Biophys Acta. 1973;291:564–575. [PubMed]
  • Hashiramoto M, James DE. Characterization of insulin-responsive GLUT4 storage vesicles isolated from 3T3-L1 adipocytes. Mol Cell Biol. 2000;20:416–427. [PMC free article] [PubMed]
  • Higgins AJ, Neal MJ. Potassium activation of [3H]choline accumulation by isolated sympathetic ganglia of the rat. Brit J Pharmac. 1982;77:573–580. [PMC free article] [PubMed]
  • Hopkins CR, Trowbridge IS. Internalization and processing of transferrin and the transferrin receptor in human carcinoma A431 cells. J Cell Biol. 1983;97:508–521. [PMC free article] [PubMed]
  • Ivy MT, Newkirk RF, Karim MR, Mtshali CMP, Townsel JG. Hemicholinium-3 mustard reveals two populations of cycling choline cotransporters in Limulus. Neurosci. 2001;102:969–978. [PubMed]
  • Karylowski O, Zeigerer A, Cohen A, McGraw TE. GLUT4 is retained by an intracellular cycle of vesicle formation and fusion with endosomes. Mol Biol Cell. 2004;15:870–882. [PMC free article] [PubMed]
  • Lambert LA, Mitchell SL. Molecular evolution of the transferrin receptor/glutamate carboxypeptidase II family. J Mol Evol. 2007;64:113–128. [PubMed]
  • Lambert LA, Perri H, Halbrooks PJ, Mason AB. Evolution of the transferrin family: conservation of residues associated with iron and anion binding. Comp Biochem Physiol B. 2005;142:129–141. [PubMed]
  • Livingstone C, James DE, Rice JE, Hanpeter D, Gould GW. Compartment ablation analysis of the insulin-responsive glucose transporter (GLUT4) in 3T3-L1 adipocytes. Biochem J. 1996;315:487–495. [PubMed]
  • Mellman I. Endocytosis and molecular sorting. Ann Rev Cell Dev Biol. 1996;12:575–625. [PubMed]
  • Mulder AH, Yamamura HI, Kuhar MJ, Snyder SH. Release of acetylcholine from hippocampal slices by depolarization: dependence on high affinity choline uptake. Brain Res. 1974;70:372–376. [PubMed]
  • Murrin LC, Kuhar MJ. Activation of high-affinity choline uptake in vitro by depolarizing agents. Molec Pharmac. 1976;12:1082–1090. [PubMed]
  • Nakata K, Okuda T, Misawa H. Ultrastructural localization of high-affinity choline transporter in the rat neuromuscular junction: enrichment on synaptic vesicles. Synapse. 2004;53:53–56. [PubMed]
  • Newkirk RF, Maleque MA, Townsel JG. Choline uptake, acetylcholine synthesis and release by Limulus abdominal ganglia. Neurosci. 1980;5:303–311. [PubMed]
  • Okuda T, Haga T, Kanai Y, Endou H, Ishihara T, Katsura I. Identification and characterization of the high-affinity choline transporter. Nature Neurosci. 2000;3:120–125. [PubMed]
  • Potter LT. Synthesis, storage and release of [14C]acetylcholine in isolated rat diaphragm muscles. J Physiol. 1970;206:145–166. [PubMed]
  • Presley JF, Mayor S, Dunn KW, Johnson LS, McGraw TE, Maxfield FR. The end2 mutation in CHO cels slows the exit of transferrin receptors from the recycling compartment but bulk membrane recycling is unaffected. J Cell Biol. 1993;122:1231–1241. [PMC free article] [PubMed]
  • Ribeiro FM, Alves-Silva J, Volknandt W, Martins-Silva C, Mahmud H, Wilhelm A, Gomez MV, Rylett RJ, Ferguson SSG, Prado VF, Prado MAM. The hemicholinium-3 sensitive high affinity choline transporter is internalized by clathrin-mediated endocytosis and is present in endosomes and synaptic vesicles. J Neurochem. 2003;87:136–146. [PubMed]
  • Ribeiro FM, Black SAG, Cregan SP, Prado VF, Prado MAM, Rylett RJ, Ferguson SSG. Constitutive high-affinity choline transporter endocytosis is determined by a carboxyl-terminal tail dileucine motif. J Neurochem. 2005;94:86–96. [PubMed]
  • Ribeiro FM, Pinthong M, Black SAG, Gordon AC, Prado VF, Prado MAM, Rylett RJ, Ferguson SSG. Regulated recycling and plasma membrane recruitment of the high-affinity choline transporter. Eur J Neurosci. 2007;26:3437–3448. [PubMed]
  • Roskoski R. Acceleration of choline uptake after depolarization induced acetylcholine release in rat cortical synaptosomes. J Neurochem. 1978;30:1357–1361. [PubMed]
  • Sheff D, Pelletier L, O'Connell CB, Warren G, Mellman I. Transferrin receptor recycling in the absence of perinuclear recycling endosomes. J Cell Biol. 2002;156:797–804. [PMC free article] [PubMed]
  • Simpson IA, Yver DR, Hissin PJ, Wardzala LJ, Karnieli E, Salans LB, Cushman SW. Insulin-stimulated translocation of glucose transporters in isolated rat adipose cells: characterization of subcellular fractions. Biochim Biophys Acta. 1983;763:393–407. [PubMed]
  • Simon JR, Atweh S, Kuhar MJ. Sodium-dependent high affinity choline uptake: a regulatory steip in the synthesis of acetylcholine. J Neurochem. 1976;26:909–922. [PubMed]
  • Wang Y.,
  • Yamamura HI, Snyder SH. Choline high-affinity uptake by rat brain synaptosomes. Science. 1972;61:626–628. [PubMed]
  • Yamamura HI, Snyder SH. High affinity transport of choline into synaptosomes of rat brain. J Neurochem. 1973;21:1355–1374. [PubMed]
  • Zeigerer A, Lampson MA, Karylowski O, Sabatini DD, Adesnik M, Ren M, McGraw TE. GLUT4 retention in adipocytes requires two intracellular insulin-regulated transport steps. Mol Biol Cell. 2002;13:2421–2435. [PMC free article] [PubMed]