The present address of the T. Kranz is the Psychobiology, Department of Behavioral Genetics, University of Trier, Johanniterufer 15, 54290 Trier, Germany.
Opposite effect of membrane raft perturbation on transport activity of KCC2 and NKCC1
Article first published online: 17 AUG 2009
© 2009 The Authors. Journal Compilation © 2009 International Society for Neurochemistry
Journal of Neurochemistry
Volume 111, Issue 2, pages 321–331, October 2009
How to Cite
Hartmann, A.-M., Blaesse, P., Kranz, T., Wenz, M., Schindler, J., Kaila, K., Friauf, E. and Nothwang, H. G. (2009), Opposite effect of membrane raft perturbation on transport activity of KCC2 and NKCC1. Journal of Neurochemistry, 111: 321–331. doi: 10.1111/j.1471-4159.2009.06343.x
- Issue published online: 23 SEP 2009
- Article first published online: 17 AUG 2009
- Received June 23, 2009; revised manuscript received July 4, 2009; accepted July 7, 2009.
- brain development;
- cation chloride cotransporter;
- chloride homeostasis;
- inhibitory synapse;
- Top of page
- Materials and methods
- Supporting Information
In the majority of neurons, the intracellular Cl− concentration is set by the activity of the Na+-K+-2Cl− cotransporter (NKCC1) and the K+-Cl− cotransporter (KCC2). Here, we investigated the cotransporters’ functional dependence on membrane rafts. In the mature rat brain, NKCC1 was mainly insoluble in Brij 58 and co-distributed with the membrane raft marker flotillin-1 in sucrose density flotation experiments. In contrast, KCC2 was found in the insoluble fraction as well as in the soluble fraction, where it co-distributed with the non-raft marker transferrin receptor. Both KCC2 populations displayed a mature glycosylation pattern. Disrupting membrane rafts with methyl-β-cyclodextrin (MβCD) increased the solubility of KCC2, yet had no effect on NKCC1. In human embryonic kidney-293 cells, KCC2 was strongly activated by a combined treatment with MβCD and sphingomyelinase, while NKCC1 was inhibited. These data indicate that membrane rafts render KCC2 inactive and NKCC1 active. In agreement with this, inactive KCC2 of the perinatal rat brainstem largely partitioned into membrane rafts. In addition, the exposure of the transporters to MβCD and sphingomyelinase showed that the two transporters differentially interact with the membrane rafts. Taken together, membrane raft association appears to represent a mechanism for co-ordinated regulation of chloride transporter function.
cation chloride cotransporters
- HEK293 cells
human embryonic kidney cells
3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide
- PGNase F
50 mM Tris, 150 mM NaCl, 5 mM EDTA
The intracellular Cl− concentration ([Cl−]i) in neurons is mainly set by an interplay of the two cation chloride cotransporters (CCC), Na+-K+-2Cl− (NKCC1) and K+-2Cl− (KCC2) (Delpire 2000; Blaesse et al. 2009). These proteins belong to the family of CCC (Gamba 2005). In many immature neurons, NKCC1 provides the high [Cl−]i required for the depolarizing action of GABA by a coupled Na+-K+-2Cl− import (Yamada et al. 2004; Achilles et al. 2007; Sipila et al. 2009). In contrast, KCC2 mediates the coupled export of K+ and Cl−, thereby generating an inwardly directed, hyperpolarizing electrochemical gradient for Cl− in most adult neurons (Rivera et al. 1999; Huebner et al. 2001; Balakrishnan et al. 2003; Zhu et al. 2005).
KCC2 and NKCC1 are co-expressed in many neurons (Vardi et al. 2000; Balakrishnan et al. 2003; Price and Trussell 2006). Developmental immunoblot analysis demonstrated KCC2 in the immature brain (Blaesse et al. 2006), which also expresses high levels of NKCC1 (Plotkin et al. 1997). Mature cortical neurons express not only KCC2, but also NKCC1 to maintain a high [Cl−]i in the axon initial segment (Khirug et al. 2008). This co-expression of a Cl− uptake (NKCC1) and a Cl− extrusion system (KCC2) requires a coordinated regulation of the two transporters. Previous studies have reported an opposite regulation of their transport activity via phosphorylation (Darman et al. 2001; Song et al. 2002; Gagnon et al. 2006; de Los et al. 2006) or via interaction with cation-chloride cotransporter interacting protein 1 (Caron et al. 2000; Wenz et al. 2009). However, the cellular mechanisms underlying transporter regulation are largely unknown.
In recent years, membrane rafts have been established as a cellular platform for the regulation of protein activity (Allen et al. 2007). Membrane rafts are small (10–200 nm diameter), highly dynamic, cholesterol-, sterol- and sphingolipid-enriched microdomains (Pike 2006). They have been associated with diverse functions, such as intracellular sorting of proteins, membrane trafficking, and signal transduction (Mishra and Joshi 2007). Proteins in membrane rafts resist cold extraction by non-ionic detergents and thereby display a low-buoyant density in sucrose flotation experiments. In contrast, detergent extraction disrupts lipid–protein interactions in non-raft domains, such that these proteins become solubilized and end up in the high-density fractions. In neurons, membrane rafts play an important role in neurotransmission and harbor acetylcholine receptors, glutamate receptors, GABAA receptors, as well as the transporters EAAT2, SERT, and NHE3 (Allen et al. 2007).
Recently, the CCC family members Na+-Cl−-cotransporter and NKCC2 (Welker et al. 2008) as well as KCC3 and KCC4 (Fujii et al. 2008) were identified as membrane rafts-associated proteins. Here, we addressed the question whether KCC2 and NKCC1 are localized to these membrane microdomains. For that purpose, we investigated the role of such rafts in KCC2 and NKCC1 transport activity.
Materials and methods
- Top of page
- Materials and methods
- Supporting Information
Animals were bred and housed in our animal facility and treated in compliance with the current Animal Protection Law. All protocols were approved by the regional animal care and use committees and adhered to the NIH Guide for the Care and Use of Laboratory Animals.
Membrane preparation from rat brain tissue
Rats were killed by decapitation during chloral hydrate-induced anesthesia (7 mg/kg body weight intraperitoneal). Brains were rapidly dissected and stored at −80°C. Membrane fractions were isolated according to a previously published protocol (Guillemin et al. 2005) and either immediately subjected to detergent treatment or stored in aliquots at −80°C.
Isolation of membrane rafts
Membrane homogenates from brain tissue were treated in 3 mL 50 mM Tris, 150 mM NaCl, 5 mM EDTA, pH 7.4 (TNE) with 1% Brij 58 (Sigma, Taufkirchen, Germany) for 30 min on ice. The homogenate was then mixed with 2 mL 70% (w/v) sucrose solution in TNE, transferred into a polyallomer centrifuge tube (14 mm × 95 mm; Beckman Coulter, Krefeld, Germany), then carefully overlaid with 3.5 mL of 30% (w/v) sucrose solution in TNE, and finally with 3.5 mL 5% (w/v) sucrose solution in TNE. Thereafter, discontinuous sucrose density gradient centrifugation was carried out at 4°C for 22 h at 230 000 g, using a SW40 rotor and an Optima L-80 XP ultracentrifuge (Beckman Coulter). Twelve fractions of equal volume were collected from the bottom to the top of the discontinuous sucrose gradient.
Immunoblot analysis was conducted as described previously (Nothwang et al. 2003). Primary antibodies were anti-flotillin-1 (dilution 1 : 500; BD Biosciences, Heidelberg, Germany), anti-cKCC2 Blaesse et al. 2006, dilution 1 : 5000), anti-NKCC1 (T4 antibody, dilution 1 : 1000; Developmental Studies Hybridoma Bank, Iowa, IA, USA), anti-Na+/K+-ATPase alpha subunit (a5, dilution 1 : 1000; Developmental Studies Hybridoma Bank Iowa) and anti-transferrin receptor (dilution 1 : 1000; Invitrogen, Karlsruhe, Germany). The secondary antibodies were goat anti-rabbit IgG or anti-mouse IgG horseradish peroxidase-conjugated (1 : 5000; GE Healthcare, Freiburg, Germany). Bound antibodies were detected using an enhanced chemiluminescence assay (Perkin Elmer, Rodgau-Juegesheim, Germany) and a Las-3000 documentation system (Fuji, Tokyo, Japan).
Manipulation of membrane rafts
Cholesterol from brain membrane homogenates was depleted by four rounds of treatment with 50 mg/mL methyl-β-cyclodextrin (MβCD; Sigma) in 1 mL TNE on ice for 30 min. Solubilized cholesterol was removed by pelleting the membrane suspension at 100 000 g for 30 min. Thererafter, membranes were treated with 1% Brij 58% at a protein to detergent ratio of 1 : 2.5. For cholesterol depletion of human embryonic kidney cells (HEK293), cells were washed once with phosphate-buffered saline (PBS) and then incubated in 10 mg/mL MβCD in serum-free Dulbecco’s modified Eagle’s medium for 30 min at 37°C. For treatment with sphingomyelinase (SMase; Sigma), cells were washed and incubated with 100 mU/mL for 1 h 50 min at 37°C. When combining both treatments, cells were incubated at 37°C first for 80 min with SMase, and then MβCD was added for 30 min. After treatment, cells were immediately processed for flux measurements, confocal imaging, cholesterol content, or cell viability measurements.
Cholesterol content of membranes was determined enzymatically using the Randox Cholesterol reagent kit in comparison with a cholesterol standard (Randox, Krefeld, Germany). Cell viability was tested in 96-well plates using the cytotoxicity assay 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide (MTT) as previously described (Mosmann 1983). After cell treatment with SMase and MβCD, cells were washed twice with PBS and then incubated with 110 μL Dulbecco’s modified Eagle’s medium plus 5 mg/mL MTT reagent per well. After incubation for 4 h, solubilization solution (10% sodium dodecyl sulfate in 0.01 M HCl) was added and the plates were incubated over night at 37°C. The spectrophotometric absorbance of the purple formazan crystals were measured at 595 nm using an ELISA reader (Multiscan Ascent; Lab Systems, Vienna, VA, USA). The reference wavelength was set to 800 nm.
Deglycosylation with EndoH or PGNase F
Deglycosylation experiments were performed using the endoglycosidase H (EndoH) or the N-glycosidase F (PGNase F) (New England Biolabs, Frankfurt, Germany). Twenty-five microgram of membrane proteins were first incubated for 10 min at 40°C in glycoprotein denaturing buffer (0.5% sodium dodecyl sulfate, 40 mM dithiothreitol) and then incubated for 2 h at 37°C in either 50 mM sodium citrate (pH 5.5) at 37°C, supplemented with 1250 units EndoH, or in 50 mM sodium phosphate (pH 7.5), 1% NP-40 (Sigma), supplemented with 1250 units PGNase. After incubation, samples were separated on a 4–12% Bis–Tris NuPAGE system (Invitrogen) for 2.5 h at 150 V.
Generation of expression constructs
Standard PCR and cloning techniques were used to generate KCC2 and NKCC1 expression constructs. The open reading frame of rat KCC2 was cloned into the mammalian expression vector pEF5/FRT/V5-Dest (Invitrogen) and into the pEGFP-N3 vector, encoding the enhanced-green-fluorescent-protein (EGFP) (Clontech Heidelberg, Germany). The open reading frame of rat NKCC1 was cloned into the pDONR/Zeo vector (Invitrogen) and then moved into a Gateway-compatible pEGFP-N3 vector. All three constructs, KCC2-EF5/FRT/V5, KCC2-EGFP-N3, and NKCC1-EGFP-N3 were used to generate HEK293 cell lines stably expressing the transporters. To do so, cells transfected with the different constructs were selected for stable integration by culturing them in the presence of 1.5 mg/mL hygromycin or G418 (Invitrogen).
Determination of transport activity
Transport activity was assessed in HEK293 cells by measuring Cl−-dependent uptake of 86Rb+ (PerkinElmer Life Sciences, Rodgau-Juegesheim, Germany), a congener of K+. HEK293 cells were grown to 100% confluency on poly-l-lysine-coated wells of a 6-well culture dish. To determine KCC2 transport activity, cells were incubated in 1 mL pre-incubation buffer (100 mM N-methyl-d-glucamine-chloride, 5 mM KCl, 2 mM CaCl2, 0.8 mM MgSO4, 5 mM glucose, 5 mM HEPES, pH 7.4, 0.1 mM oubain). After 30 min at 37°C, the pre-incubation buffer was replaced by 1 mL incubation buffer (pre-incubation buffer containing 37 MBq 86Rb+) incubated for another 7 min at 20°C. Following three washing steps in 1 mL ice-cold pre-incubation buffer without oubain, cells were lysed in 500 μL 0.25 M NaOH for 1 h. 86Rb+ uptake was assayed by the determination of Cerenkov radiation and the protein amount was determined by the BCA Assay (Pierce Chemicals, Bonn, Germany). Flux data were corrected for changes in cell viability by subtracting the protein amount corresponding to dead cells from the total protein amount determined after flux measurements.
NKCC1 transport activity was determined using a modified protocol of Dehaye et al. (2003). Cells were washed twice with pre-incubation buffer (135 mM NaCl; 2.5 mM KCl; 1 mM CaCl2; 1 mM MgCl2; 15 mM Na–HEPES; 5 mM glucose; pH 7.4), and then incubated for 10 min at 37°C. The wells were washed with hypertonic chloride-free media (135 mM; Na–gluconate; 2.5 mM K–gluconate; 1 mM CaCl2; 1 mM MgCl2; 15 mM HEPES; 5 mM glucose; 100 mM sucrose; pH 7.4) before they were incubated for 1 h at 37°C. The chloride-free media were then replaced by chloride-free media plus 0.1 mM ouabain for further 10 min at 37°C. Therafter, 86Rb+ uptake was measured in uptake media (135 mM NaCl; 1 mM KCl; 1 mM CaCl2; 15 mM Na–HEPES; 5 mM glucose; 0.1 mM ouabain; 0.3 μCi 86Rb+; pH 7.4) for 7 min at 20°C. Following three washing steps in 1 mL ice-cold termination media (135 mM Na-gluconate; 7.5 mM NaCl; 15 mM HEPES; pH 7.4), cells were lysed in 500 μL 0.25 M NaOH for 1 h. At least six replicas were performed for each experiment. Data are given as mean ± SD. Significant differences between the groups were analyzed by student`s t-test.
To visualize the distribution of the KCC2- and NKCC1-EGFP fusion proteins after membrane raft perturbation, HEK293 cells cultured on sterile glass coverslips pre-treated with 0.1 mg/mL poly-l-lysine were washed once with PBS and fixed with 4% paraformaldehyde in 0.2 M phosphate buffer for 15 min at 20°C. After fixation, cells were washed thrice with PBS and mounted onto glass slides with Vectorshield Hard Set mounting medium (Vector laboratories, Burlingame, CA, USA). Photomicrographs were taken using a SP5 confocal microscope (Leica, Wetzlar, Germany) equipped with a 63 × 1.3 NA glycerol immersion objective. The pinhole diameter was set for 1 Airy Unit.
- Top of page
- Materials and methods
- Supporting Information
Association of KCC2 and NKCC1 with membrane rafts in the mature brain
To investigate whether KCC2 and NKCC1 are associated with membrane rafts, these membrane domains were isolated from P25-30 rat brains by flotation experiments (Pike 2006). We extracted the membranes using 1% Brij 58 at a protein to detergent ratio of 1 : 2.5 and separated detergent-soluble from detergent-insoluble proteins by discontinuous sucrose gradient centrifugation. Collected fractions were assessed by immunoblot analysis for the presence of KCC2 and NKCC1, as well as for the membrane raft marker protein, flotillin-1 (Bickel et al. 1997) and the transferrin receptor (TfR), a non-raft marker protein (Harder et al. 1998) (Fig. 1). The separation of flotillin-1 from TfR demonstrated the successful separation of membrane rafts from non-raft domains (Fig. 1a). We detected two KCC2 populations after Brij 58 treatment. One part of KCC2 was recovered in the low-density fractions 4–6 and co-distributed with flotillin-1. A second population co-distributed with TfR predominantly in the high-density fractions 10–12 (Fig. 1a). The immunoblot analysis showed similar amounts of KCC2 in both populations (Fig. 1a). The higher molecular weight of KCC2 mainly observed in the high-density fractions 10–12 compared to insoluble KCC2 in the lower-density fractions was likely because of the aggregation of the protein in the absence of a lipid environment during centrifugation, as treatment with high concentrations of reducing agents prior to gel electrophoresis did not result in any mobility shift (data not shown). It was therefore not possible to judge whether the Brij 58 soluble KCC2 population represented oligomeric KCC2 and the Brij 58 insoluble population monomeric KCC2 (Blaesse et al. 2006). Unlike KCC2, most NKCC1 was detected in the lower-density fractions 4–6, together with flotillin-1 (Fig. 1a). As the protein to detergent ratio is a key parameter with great influence on the outcome in flotation experiments (Allen et al. 2007; Scandroglio et al. 2008), we varied this ratio from 1 : 1 to 1 : 10. This only slightly affected solubilization of KCC2 and NKCC1 in Brij 58 (data not shown).
The data obtained so far implied that KCC2 is localized both to membrane rafts and non-raft domains, whereas NKCC1 is mainly associated with Brij 58-resistant membrane rafts. As cholesterol is a major component of membrane rafts, we assessed whether perturbing the membrane cholesterol content resulted in increased solubility of the membrane raft-associated transporters. One means to disrupt the organization of cholesterol-rich microdomains is cholesterol depletion by methyl-β-cyclodextrin (MβCD) (Schuck et al. 2003; Butchbach et al. 2004; Magnani et al. 2004; Taverna et al. 2004). We therefore incubated membrane homogenates with MβCD prior to Brij 58 treatment. A 60% reduction of cholesterol did not alter the distribution of KCC2, and a 70% reduction partially shifted KCC2 from the lower-density fractions to higher-density fractions (data not shown). An almost complete depletion of cholesterol resulted in a strong increase in solubilized KCC2, as most immunoreactivity was recovered in fractions 8–12 (Fig. 1b), and thus shifted by four fractions compared to Brij 58 treatment without previous cholesterol depletion. By contrast, NKCC1 was still recovered in lower-density fractions. This observation was reminiscent of the minor shift of the membrane raft-associated NKCC2 after MβCD treatment of cultured renal tubular cells (Welker et al. 2008). We therefore concluded that NKCC1 is associated with membrane rafts, whereas KCC2 forms two populations, with one being located outside rafts.
Finally, we analyzed the distribution of the Na+/K+-ATPase, whose association with transport-active KCC2 has been suggested (Ikeda et al. 2004). The large majority of this protein became solubilized in Brij 58 and was recovered in fractions 9–12 (Fig. 1a). After MβCD treatment, all Na+/K+-ATPase was observed in fractions 8–12 (Fig. 1b). This finding conforms to previous studies reporting the Na+/K+-ATPase as a non-membrane raft protein (Becher et al. 2001; Eckert et al. 2003; Nini et al. 2007).
Glycosylation pattern of raft-associated and non-raft-associated cotransporters
One explanation for the two KCC2 populations that we found is a different localization within specific subcellular compartments. The membrane preparation used for our flotation experiments represents a mixture of the plasma membrane and membranes of intracellular compartments such as the endoplasmic reticulum and the Golgi apparatus. To check whether one population represents intracellular KCC2, we enzymatically deglycosylated the two KCC2 populations obtained by Brij 58 treatment. The EndoH cleaves mainly high-mannose glycosylated forms, which are indicative of proteins within the endoplasmic reticulum (Maley et al. 1989). The amidase PGNase F cleaves also complex oligosaccharides from N-linked glycoproteins, which form during the passage through the Golgi apparatus (Maley et al. 1989). We performed the deglycosylation experiments on pooled fractions 4–6 and 10–12. After incubation with the respective enzyme, proteins were analyzed by immunoblotting. Both KCC2 populations were insensitive to EndoH (Fig. 2a). In contrast, PGNase F treatment resulted in a shift of both KCC2 populations toward lower molecular weights compared to untreated samples (Fig. 2b). The increased molecular weight of KCC2 in the Brij 58-insoluble pool, compared to the untreated fractions (cf. Fig. 1a), probably reflects an aggregation of the transporter caused by the required denaturing protocol prior to enzymatic treatment. The data demonstrate that both the soluble and the insoluble KCC2 population display a mature glycosylation pattern and likely represent KCC2 from compartments distal to the cis-Golgi, and likely from plasma membrane.
Influence of membrane raft perturbation on KCC2 and NKCC1 transport activity
To further investigate the role of membrane rafts in KCC2 and NKCC1 function, we studied the effects of cholesterol depletion on 86Rb+ transport (Williams and Payne 2004; Gagnon et al. 2006; Mercado et al. 2006). To examine the influence of membrane rafts on the functionality of KCC2, we used a HEK293 cell line which stably expresses this transporter (HEK293KCC2). NKCC1 is endogenously expressed in HEK293 cells (Isenring et al. 1998; Dehaye et al. 2003; Delpire et al. 2009). Our previous flotation experiments indicated a strong association of NKCC1 with membrane rafts. We therefore used three different protocols to disrupt membrane rafts. Cell cultures were treated either with 10 mg/mL MβCD for 30 min (Magnani et al. 2004), or with 100 mU/mL SMase (Scheek et al. 1997; Fischer et al. 2007), or with a combination of both agents. SMase hydrolyzes the membrane raft component sphingomyelin to ceramide (Yu et al. 2005). First, we analyzed the effect of the treatments on the effectiveness of the cholesterol depletion and cell viability (Supporting Information Fig. S1). Forty-five percent of cholesterol was sequestered by the MβCD treatment (Supporting Information Fig. S1a). This was similar to a previously reported depletion of ∼50% in HEK293 cells (Magnani et al. 2004). The remaining cholesterol likely reflects the fact that MβCD is membrane-impermeable and does therefore not deplete cholesterol from intracellular membranes. Cell viability was 88% after MβCD treatment, 92% after SMase treatment, and 74% after combined treatment (Fig. 3b). Single treatment thus barely affected viability, and the viability was still high after combined treatment.
Next, we determined the flux activity after perturbation experiments. To do so, we first confirmed pharmacologically the presence of KCC2- and NKCC1-mediated fluxes in the cell lines used for this study. The 86Rb influx was 1.8-fold higher in HEK293KCC2 cells than in native HEK293 cells (14.4 ± 2.4 nmoles × mg/protein × per min vs. 7.99 ± 2.0 nmoles × mg/protein × per min). This increase was highly significant (p = 7.4 × 10−7), and the flux was reduced by 45% in the presence of 2 mM furosemide, similar to a previous study (Strange et al. 2000). Native HEK293 cells displayed Rb+ uptake, which was blocked by 10 μM bumetanide (Fig. 3b; remaining activity: 2.3%), a concentration that is relatively specific for NKCCs (Russell 2000; Blaesse et al. 2009). The activity of KCC2 increased threefold after MβCD treatment (Fig. 3a; 300.1 ± 45.7%; p = 8.4 × 10−19). The combined MβCD and SMase treatment resulted in an increased KCC2 activity of 671.7 ± 152.4% of the control (p = 6.9 × 10−20) (Fig. 3a). SMase alone activated KCC2 only to a small, albeit significant extent (142.9 ± 15.1%; p = 6.1 × 10−12) (Fig. 3a). The observed increase after all treatments was blocked by furosemide (Fig. 3a) and was always significantly different from those observed in native HEK293 cells subjected to the same treatment (data not shown). Unlike KCC2, NKCC1 transport activity after cholesterol depletion by MβCD (144 ± 19.1%; p = 1.1 × 10−6) was only slightly increased compared to untreated cells (Fig. 3b). This increase was completely blocked by 10 μM bumetanide (Fig. 3b). In contrast, breaking down sphingomyelin by SMase reduced NKCC1 activity almost twofold (58.3 ± 9.6%; p = 1.3 × 10−7) (Fig. 3b). Similar to KCC2, the combined action of both agents had a more pronounced effect than either treatment alone and reduced NKCC1 activity to 38.6 ± 17.1% (p = 2 × 10−9) (Fig. 3b).
Effect of membrane raft disruption on KCC2 and NKCC1 localization
The observed changes in transport activity after the disruption of membrane rafts prompted us to ask whether the location of the transporters was altered. To address this issue, we used HEK293 cells which stably express KCC2-EGFP or NKCC1-EGFP fusion proteins; in both constructs, EGFP was fused to the intracellular C-terminus of the transporters. In untreated controls, we observed a granular staining pattern at the plasma membrane of the cells for both, KCC2 and NKCC1 (Fig. 4a,e,i and m). The first and third rows depict cross-sections and the second and fourth rows the basal surface of the same cell. After treatment with MβCD, a slighty more pronounced staining at the plasma membrane was apparent for both transporters (Fig. 4b,f,j and n), which is consistent with the transport results (Fig. 3). SMase alone resulted in no visible changes in KCC2 and NKCC1 distribution (Fig. 4c,g,k and o). The combined MβCD and SMase treatment resulted in large, sharp clusters of the transporters at the plasma membrane (Fig. 4d,h,l and p). These data indicate that perturbation of membrane rafts affects the localization of the transporters at the plasma membrane.
KCC2 association with membrane rafts in the immature brainstem
Our data from the flux measurements in HEK293 cells indicated that the detergent-insoluble KCC2 population, which we observed in flotation experiments of the mature brain, corresponds to transport-inactive KCC2, whereas the detergent-soluble population represents transport-active KCC2. We therefore analyzed KCC2 partitioning in the brainstem at P0-2. At this immature age, most KCC2 is thought to be transport-inactive (Blaesse et al. 2006). According to our conclusion from the flux measurements that the membrane raft-associated KCC2 population of the mature brain represents transport-inactive KCC2, we expected to recover KCC2 in the membrane raft fraction at P0-2. Indeed, after treatment with Brij 58, KCC2 ended up in the low-density fractions 3 and 4, together with the membrane-raft marker protein flotillin-1 (Fig. 5a). Only a minor amount of KCC2 was solubilized. Similar to P25-30, we observed higher molecular weight KCC2 aggregates. They were distributed throughout the gradient mirroring the distribution of monomeric KCC2 at 130 kDa.
To assess whether the presence of KCC2 in membrane rafts in the immature brainstem was because of the segregation into cholesterol-rich membrane rafts, as observed in the mature brain, we depleted cholesterol with MβCD from the membrane homogenate. Under this condition, which extracted ∼70% cholesterol from the membranes, most KCC2 was soluble in Brij 58 and consequently, it recovered at higher densities, mainly in fractions 6 and higher (Fig. 5b). This shift toward higher-density fractions was thus similar to the cholesterol depletion experiments in the mature brain (cf. Fig. 1b). Together, these data support the idea that partitioning into membrane rafts renders KCC2 inactive.
- Top of page
- Materials and methods
- Supporting Information
Interactions between proteins and lipids are essential to many cellular functions. The present work investigated the association of two essential cotransporters for neuronal Cl− homeostasis, KCC2 and NKCC1, with membrane rafts. Two main findings arose from our study: (i) membrane raft perturbations have opposite effects on the activity of the two cotransporters; (ii) the two cotransporters differentially associate with membrane raft microdomains.
Modulation of KCC2 and NKCC1 transport activity by membrane rafts
A defining characteristic of membrane raft proteins is their resistance to solubilization in non-ionic detergents. By means of flotation experiments, about half of KCC2 and the majority of NKCC1 co-distributed with the membrane raft marker flotillin-1 in the lower-density fractions. This provided a strong evidence that a considerable portion of both cotransporters is associated with membrane rafts, similar to the CCC family members KCC3, KCC4 (Fujii et al. 2008), NKCC1, and Na+-Cl−-cotransporter (Welker et al. 2008). We used Brij 58 instead of Triton X-100, as the former was previously shown to be better suited to analyze the membrane raft association of the neuronal transporters SERT (Magnani et al. 2004) and EAAT2 (Butchbach et al. 2004).
To determine the physiologic relevance of this segregation into membrane rafts, we disrupted these domains in HEK293 cells, which are a widely used mammalian expression system for these CCC transporters (Moore-Hoon and Turner 2000; Simard et al. 2004; Wenz et al. 2009). Furthermore, HEK293 cells faithfully reproduce the differential targeting of neuronal potassium channels to distinct membrane rafts (O’Connell and Tamkun 2005). The pivotal result of these experiments was that KCC2 was activated by the perturbation of membrane rafts, whereas NKCC1 was inhibited (Fig. 3). This modulation likely occurred via a change in the distribution of the transporters at the plasma membrane as evidenced by confocal imaging analysis (Fig. 4).
The functional dependence of NKCC1 on the integrity of membrane rafts, together with its location in these microdomains, identifies the cotransporter as a bona fide membrane raft protein. Unlike NKCC1, flotation experiments demonstrated two populations for KCC2 in the brain, a detergent-soluble and a detergent-resistant pool (Fig. 1). Both populations displayed a mature glycosylation pattern, indicating that they originated from compartments distal to the cis-Golgi, and likely from the plasma membrane (Fig. 2). The observed activation of KCC2 after cholesterol depletion suggests that the detergent-soluble population represents constitutively transport-active KCC2 in the plasma membrane of unperturbed cells. In contrast, transport-inactive KCC2 constitutes the insoluble population, which can be activated by perturbation experiments. This is analogous to the two populations of the src-family tyrosine kinase LCK, where the membrane raft-associated enzyme displays low activity and the non-membrane raft population high activity (Rodgers and Rose 1996; Kabouridis et al. 2000). Our analysis of the immature brainstem, which contains mainly inactive KCC2 (Blaesse et al. 2006), is in agreement with this conclusion. At this developmental stage, KCC2 mainly partitioned into membrane rafts. This was evidenced by codistribution of KCC2 with the membrane raft marker flotillin-1 and its shift to higher-density fractions after cholesterol depletion with MβCD.
It is of note that in neuronal cells, the Na+/K+-ATPase is mainly observed in the non-membrane raft fraction (Eckert et al. 2003; Nini et al. 2007). This conforms with the observation that this primary active-transporter interacts with KCC2 to form a larger complex, allowing for efficient functional coupling in the nervous system at the molecular level (Ikeda et al. 2004). The functional role of membrane rafts in KCC3 and KCC4 transport activity has not yet been investigated (Fujii et al. 2008). It is therefore unclear whether other KCC family members require membrane rafts for proper transport activity.
As a result of the opposite effect of membrane rafts on the activity of KCC2 and NKCC1, a shared regulatory mechanism for localization to membrane rafts would allow for efficient and dynamic regulation of [Cl−]i in neurons. It is currently unknown whether a shared mechanism exists and how dynamically the two transporters can shift between different membrane compartments. In many cases, proteins are targeted to membrane rafts either via protein–protein interaction or by protein modification, such as acylation. It might also be worthwhile to investigate whether the reciprocal effect of phosphorylation on the activity of the two transporters (Kahle et al. 2005; Gagnon et al. 2006) involves their redistribution between different plasma membrane compartments. It will also be important to study whether the location of transport-active KCC2 outside membrane rafts and of NKCC1 within membrane rafts participates in the different subcellular distribution of the two transporters, such as the axonal localization of NKCC1 (Khirug et al. 2008).
KCC2 and NKCC1 differentially associate with membrane rafts
We observed strikingly different sensitivities of the two cotransporters toward the two membrane raft disturbing agents MβCD and SMase (Fig. 3). Cholesterol depletion by MβCD activated KCC2, whereas it only slightly affected NKCC1 transport activity. In keeping with this finding, MβCD treatment of brain membranes shifted the distribution of KCC2, but not of NKCC1 (Fig. 1). However, it is of note that we had to deplete cholesterol by ≥ 70% in the brain before observing a visible increase in KCC2 solubility after Brij 58% treatment. This indicates a rather strong association of KCC2 with membrane rafts, as a 60% cholesterol depletion was already sufficient to shift, for instance, the Cav2.1 subunit (Taverna et al. 2004).
In contrast to cholesterol depletion, the breakdown of sphingomyelin by SMase reduced NKCC1 activity by ∼50%, whereas KCC2 was only slightly affected by this treatment (Fig. 3). Insensitivity to SMase has already been observed for other membrane raft proteins. A mass spectrometric analysis revealed only a selective loss of proteins from membrane rafts after SMase treatment (Yu et al. 2005). G-proteins and flotillins remained associated with membrane rafts under this condition, whereas their level in membrane rafts decreased after cholesterol depletion (Foster et al. 2003; Yu et al. 2005). Furthermore, no reduction in membrane raft association was observed for the proteins PLAP, YES, caveolin-1, and VIP17 (Schuck et al. 2003). We attempted to analyze the effect of SMase on the distribution of the two transporters by sucrose gradient centrifugation. However, treatment by SMase resulted in aggregated membrane fractions which prevented further flotation experiments (A.-M. Hartmann and H. G. Nothwang, unpublished data).
The underlying mechanism of this different sensitivity to membrane raft perturbing agents is currently unclear. It might be either caused by an association of KCC2 and NKCC1 with different components within the same membrane rafts or by localization into different membrane rafts. Recent data indicated a rather high degree of heterogeneity within membrane rafts (Pike 2004; Mishra and Joshi 2007). Immunofluorescent probes and immunocytochemical labeling of raft-specific lipids and proteins as well as biochemical analysis will help to distinguish between these two possibilities.
In conclusion, our data identify a cellular mechanism for the coordinated regulation of NKCC1 and KCC2 during neuronal development as well as in subcellular domains.
- Top of page
- Materials and methods
- Supporting Information
We are indebted to K. Ociepka and M. Reents for expert technical assistance and Dr. I. Guillemin for introducing us to the flux measurements. We thank Dr. P. Runeberg-Roos for critical comments on an early version of this manuscript. This work was supported by grants from the Deutsche Forschungsgemeinschaft (No428/2-1 and No428/2-3 to HGN) and the Academy of Finland (KK). KK is a member of the Finnish Center of Excellence (CoE) in Molecular and Integrative Neuroscience Research and of the Nordic CoE, WIRED. The monoclonal antibody α5 from Dr. Fambrough was obtained through 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, USA.
- Top of page
- Materials and methods
- Supporting Information
- 2007) Kinetic properties of Cl uptake mediated by Na+-dependent K+-2Cl cotransport in immature rat neocortical neurons. J. Neurosci. 27, 8616–8627. , , , , , , and (
- 2007) Lipid raft microdomains and neurotransmitter signalling. Nat. Rev. Neurosci. 8, 128–140. , and (
- 2003) Expression and function of chloride transporters during development of inhibitory neurotransmission in the auditory brainstem. J. Neurosci. 23, 4134–4145. , , , , and (
- 2001) The gamma-aminobutyric acid receptor B, but not the metabotropic glutamate receptor type-1, associates with lipid rafts in the rat cerebellum. J. Neurochem. 79, 787–795. , and (
- 1997) Flotillin and epidermal surface antigen define a new family of caveolae-associated integral membrane proteins. J. Biol. Chem. 272, 13793–13802. , , , , and (
- 2006) Oligomerization of KCC2 correlates with development of inhibitory neurotransmission. J. Neurosci. 26, 10407–10419. , , , , , , and (
- 2009) Cation-chloride cotransporters and neuronal function. Neuron 61, 820–838. , , and (
- 2004) Association of excitatory amino acid transporters, especially EAAT2, with cholesterol-rich lipid raft microdomains: importance for excitatory amino acid transporter localization and function. J. Biol. Chem. 279, 34388–34396. , , and (
- 2000) Cloning and functional characterization of a cation-Cl- cotransporter-interacting protein. J. Biol. Chem. 275, 32027–32036. , , and (
- 2001) Modulation of ion transport by direct targeting of protein phosphatase type 1 to the Na-K-Cl cotransporter. J. Biol. Chem. 276, 34359–34362. , and (
- 2003) Identification of a functionally important conformation-sensitive region of the secretory Na+-K+-2Cl- cotransporter (NKCC1). J. Biol. Chem. 278, 11811–11817. , , and (
- 2000) Cation-chloride cotransporters in neuronal communication. News Physiol. Sci. 15, 309–312. (
- 2009) Small-molecule screen identifies inhibitors of the neuronal K-Cl cotransporter KCC2. Proc. Natl Acad. Sci. USA 106, 5383–5388. , , , , , and (
- 2003) Lipid rafts of purified mouse brain synaptosomes prepared with or without detergent reveal different lipid and protein domains. Brain Res. 962, 144–150. , , and (
- 2007) Ceramide as a TLR4 agonist; a putative signalling intermediate between sphingolipid receptors for microbial ligands and TLR4. Cell Microbiol. 9, 1239–1251. , , , , and (
- 2003) Unbiased quantitative proteomics of lipid rafts reveals high specificity for signaling factors. Proc. Natl Acad. Sci. USA 100, 5813–5818. , and (
- 2008) K+-Cl- Cotransporter-3a Up-regulates Na+,K+-ATPase in Lipid Rafts of Gastric Luminal Parietal Cells. J. Biol. Chem. 283, 6869–6877. , , , , , , , , and (
- 2006) Volume sensitivity of cation-Cl- cotransporters is modulated by the interaction of two kinases: Ste20-related proline-alanine-rich kinase and WNK4. Am. J. Physiol. Cell Physiol. 290, C134–C142. , and (
- 2005) Molecular physiology and pathophysiology of electroneutral cation-chloride cotransporters. Physiol. Rev. 85, 423–493. (
- 2005) A subcellular prefractionation protocol for minute amounts of mammalian cell cultures and tissue. Proteomics 5, 35–45. , , , and (
- 1998) Lipid domain structure of the plasma membrane revealed by patching of membrane components. J. Cell Biol. 141, 929–942. , , and (
- 2001) Disruption of KCC2 reveals an essential role of K-Cl cotransport already in early synaptic inhibition. Neuron 30, 515–524. , , , , and (
- 2004) Malfunction of respiratory-related neuronal activity in Na+, K+-ATPase alpha2 subunit-deficient mice is attributable to abnormal Cl- homeostasis in brainstem neurons. J. Neurosci. 24, 10693–10701. , , et al. (
- 1998) Comparison of Na-K-Cl cotransporters. NKCC1, NKCC2, and the HEK cell Na-L-Cl cotransporter. J. Biol. Chem. 273, 11295–11301. , , and (
- 2000) Cholesterol depletion disrupts lipid rafts and modulates the activity of multiple signaling pathways in T lymphocytes. Eur. J. Immunol. 30, 954–963. , , and (
- 2005) WNK3 modulates transport of Cl- in and out of cells: implications for control of cell volume and neuronal excitability. Proc. Natl Acad. Sci. USA 102, 16783–16788. , , , , , , , , and (
- 2008) GABAergic depolarization of the axon initial segment in cortical principal neurons is caused by the Na-K-2Cl cotransporter NKCC1. J. Neurosci. 28, 4635–4639. , , , , and (
- 2006) WNK3 bypasses the tonicity requirement for K-Cl cotransporter activation via a phosphatase-dependent pathway. Proc. Natl Acad. Sci. USA 103, 1976–1981. , , , , , , , , and (
- 2004) Partitioning of the serotonin transporter into lipid microdomains modulates transport of serotonin. J. Biol. Chem. 279, 38770–38778. , , , and (
- 1989) Characterization of glycoproteins and their associated oligosaccharides through the use of endoglycosidases. Anal. Biochem. 180, 195–204. , , and (
- 2006) A C-terminal domain in KCC2 confers constitutive K+-Cl- cotransport. J. Biol. Chem. 281, 1016–1026. , , , and (
- 2007) Lipid raft heterogeneity: an enigma. J. Neurochem. 103 (Suppl. 1), 135–142. and (
- 2000) The structural unit of the secretory Na+-K+-2Cl- cotransporter (NKCC1) is a homodimer. Biochemistry 39, 3718–3724. and (
- 1983) Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J. Immunol. Methods 65, 55–63. (
- 2007) R7-binding protein targets the G protein beta 5/R7-regulator of G protein signaling complex to lipid rafts in neuronal cells and brain. BMC Biochem. 8, 18. , , , , and (
- 2003) Protein analysis in the rat auditory brainstem by two-dimensional gel electrophoresis and mass spectrometry. Mol. Brain Res. 116, 59–69. , , and (
- 2005) Targeting of voltage-gated potassium channel isoforms to distinct cell surface microdomains. J. Cell Sci. 118, 2155–2166. and (
- 2004) Lipid rafts: heterogeneity on the high seas. Biochem. J. 378, 281–292. (
- 2006) Rafts defined: a report on the Keystone Symposium on Lipid Rafts and Cell Function. J. Lipid Res. 47, 1597–1598. (
- 1997) Expression of the Na-K-2Cl cotransporter is developmentally regulated in postnatal rat brains: a possible mechanism underlying GABA’s excitatory role in immature brain. J. Neurobiol. 33, 781–795. , , and (
- 2006) Estimate of the chloride concentration in a central glutamatergic terminal: a gramicidin perforated-patch study on the calyx of Held. J. Neurosci. 26, 11432–11436. and (
- 1999) The K+/Cl- co-transporter KCC2 renders GABA hyperpolarizing during neuronal maturation. Nature 397, 251–255. , , , , , , , and (
- 1996) Exclusion of CD45 inhibits activity of p56lck associated with glycolipid-enriched membrane domains. J. Cell Biol. 135, 1515–1523. and (
- 2000) Sodium-potassium-chloride cotransport. Physiol. Rev. 80, 211–276. (
- 2008) Lipid content of brain, brain membrane lipid domains, and neurons from acid sphingomyelinase deficient mice. J. Neurochem. 107, 329–338. , , , , , , and (
- 1997) Sphingomyelin depletion in cultured cells blocks proteolysis of sterol regulatory element binding proteins at site 1. Proc. Natl Acad. Sci. USA 94, 11179–11183. , and (
- 2003) Resistance of cell membranes to different detergents. Proc. Natl Acad. Sci. USA 100, 5795–5800. , , , and (
- 2004) Characterization of a novel interaction between the secretory Na+-K+-Cl- cotransporter and the chaperone hsp90. J. Biol. Chem. 279, 48449–48456. , , , , , , and (
- 2009) Compensatory enhancement of intrinsic spiking upon NKCC1 disruption in neonatal hippocampus. J. Neurosci. 29, 6982–6988. , , , , , and (
- 2002) Molecular, functional, and genomic characterization of human KCC2, the neuronal K-Cl cotransporter. Mol. Brain Res. 103, 91–105. , , , , , , and (
- 2000) Dependence of KCC2 K-Cl cotransporter activity on a conserved carboxy terminus tyrosine residue. Am. J. Physiol. Cell Physiol. 279, C860–C867. , , and (
- 2004) Role of lipid microdomains in P/Q-type calcium channel (Cav2.1) clustering and function in presynaptic membranes. J. Biol. Chem. 279, 5127–5134. , , , , and (
- 2000) Evidence that different cation chloride cotransporters in retinal neurons allow opposite responses to GABA. J. Neurosci. 20, 7657–7663. , , and (
- 2008) Renal Na+-K+-Cl- cotransporter activity and vasopressin-induced trafficking are lipid raft-dependent. Am. J. Physiol. Renal Physiol. 295, F789–F802. , , , , , , , , and (
- 2009) CIP1 is an activator of the K+-Cl- cotransporter KCC2. Biochem. Biophys. Res. Commun. 381, 388–392. , , and (
- 2004) Cation transport by the neuronal K(+)-Cl(-) cotransporter KCC2: thermodynamics and kinetics of alternate transport modes. Am. J. Physiol. Cell Physiol. 287, C919–C931. and (
- 2004) Cl- uptake promoting depolarizing GABA actions in immature rat neocortical neurones is mediated by NKCC1. J. Physiol. 557, 829–841. , , , , and (
- 2005) Ceramide displaces cholesterol from lipid rafts and decreases the association of the cholesterol binding protein caveolin-1. J. Lipid Res. 46, 1678–1691. , and (
- 2005) Cortical neurons lacking KCC2 expression show impaired regulation of intracellular chloride. J. Neurophysiol. 93, 1557–1568. , and (
- Top of page
- Materials and methods
- Supporting Information
Figure S1. Cell viability and cholesterol depletion in HEK293 cells. (A) Treatment of HEK293 cells with 10 mg/mL MβCD for 30 min depleted 45% of the membrane cholesterol content. The plot shows the mean ± s.d., n ≥ 6. (B) HEK293 cells were treated with 10 mg/mL MβCD to deplete cholesterol, with 100 mU/mL SMase to breakdown sphingomyelin, or with a combination of both agents, and cell viability was determined by an MTT assay. Viability was ~90% after treatment with either of the two agents and 74% after combined action.
As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.
|JNC_6343_sm_figures1.pdf||829K||Supporting info item|
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.