INTRODUCTION

Calcium-conducting TRP channels are multimodal ion sensors, which, responding to various physical and chemical stimuli, integrate many signals entering the cell. TRP channels are involved in the vast majority of body reactions, including visual, olfactory, pain, and temperature receptions (Montell et al., 2002; Clapham, 2003; Flockerzi, 2007). They are involved in neurogenesis, brain development, and synaptic transmission (Vennekens et al., 2012). A number of TRP channels are involved in the body’s immunity and metabolic processes (Fernandes, 2012). Therefore, it is not surprising that dysfunctions of many channels of the TRP superfamily lead to the development of various diseases (Nilius, 2007).

Being integral proteins, TRP channels are sensitive to the lipid composition of the membrane into which they are embedded. Until recently, it was believed that lipids play only a passive role in the cell, being simple building blocks for membranes, delimiting intracellular compartments and separating the internal environment from the extracellular environment of the cell. However, over the past decades, a new understanding of the role of lipids in the life of cells has emerged. It became known that the main lipids of the plasma membrane, such as cholesterol and sphingolipids, can be tightly packed together, forming microdomains (lipid rafts), which are involved in many functional reactions of cells (Simons and Toomre, 2000; Pizzo and Viola, 2003). Rafts are involved in such processes as transport of membrane proteins, assembly of signal complexes, transmission of nerve impulses, and endo- and exocytosis, as well as regulation of the activity of ion channels (Levitan et al., 2000; Lundbaek et al., 2003; Schengrund, 2010; Sheng et al., 2012).

The composition of lipids largely determines the structural and physical properties of the plasma membrane and its fluidity, curvature, and stiffness (Yeagle, 1985), which critically affect ion channel gating (Pucadyil and Chattopadhyay, 2004). Various signaling proteins associated with lipid rafts can also influence channel properties. In addition, some lipids themselves can act as specific ion channel agonists. Important data regarding the regulation of channel functions with cholesterol were obtained using methyl-β-cyclodextrin (MbCD) (Heino et al., 2000; Slimane et al., 2001; Barbuti et al., 2004). Thus, the extraction of membrane cholesterol showed that the activity of TRPM8 channels is critically dependent on the integrity of lipid rafts, since it led to a significant shift in the activation threshold of thermally sensitive TRPM8 channels towards higher temperatures (Morenilla-Palao et al., 2009). In the same way, the destruction of lipid rafts by depletion of any of its main components by pharmacological tools blocked the agonist-dependent activation of TRPV1 channels (Szoke et al., 2010). Input of Ca2+ stopped either when cholesterol was depleted with MbCD (Kilsdonk et al., 1995; Liu et al., 2006;) or when sphingomyelin molecules were destroyed by sphingomyelinase (Kobayashi et al., 2006).

Two members of the TRP superfamily—the TRPV5 and TRPV6 channels (originally known as CaT2 and CaT1, or ECaC1 and ECaC2)—were first cloned from rat rabbit kidney and small intestine epithelial cells (Hoenderop et al., 2001; Peng and Hediger, 2002). They were later assigned to the vanilloid receptor subfamily (TRPV) and have been identified in cells of other mammals, including human beings (Nijenhuis et al., 2003). It was shown that these channels can provide a strictly dosed intake of Ca2+ and participate in active (re)absorption of Ca2+ in the epithelial cells of the kidneys, small intestine, and placenta (Hoenderop, 2002; Nijenhuis et al., 2003). Earlier, using real-time polymerase chain reaction and Western blot analysis, we showed the expression of the TRPV5 and TRPV6 channels in Jurkat human leukemia T cells (Vassilieva et al., 2013).

The TRPV5 and TRPV6 calcium channels are constitutively active; therefore, the selective association of ion channels with a specific lipid microenvironment in the membrane may be critically necessary for their activation. Despite this, the question of the interaction of TRPV5 with the lipid microenvironment and lipid rafts practically has not been studied. Earlier, using electrophysiological and immunofluorescence methods, as well as the method of immunoelectronic microscopy, we showed that membrane cholesterol regulates the activity of TRPV6 channels in human T cells of the Jurkat line (Kever et al., 2019).

In this work, we studied the role of lipid rafts in the localization of TRPV5 calcium channels in the plasma membrane of Jurkat human leukemia T cells.

MATERIALS AND METHODS

The Jurkat cell line was obtained from the Russian collection of vertebrate cell cultures of the Institute of Cytology, Russian Academy of Sciences RAS (St. Petersburg). Cells were cultured at 37°C in RPMI 1640 medium containing 10% fetal calf serum and 0.008% gentamicin at a temperature of 37°C in an atmosphere of 5% CO2.

Immunofluorescence assay was performed on cells that were previously seeded on glass coated with poly-L-lysine. Cells were incubated for 40 min (at 37°C and 5% CO2) in a serum-free medium in the presence and absence (control cells) of 1% methyl-β-cyclodextrin (MbCD, Sigma-Aldrich). Then, the cells were fixed in a solution of 3.7% paraformaldehyde and permeabilized in the presence of 0.25% Tween 20. Then, control and treated cells were incubated first with primary antibodies (dilution of 1 : 100) developed against TRPV5 (H-99, Santa Cruz Biotechnology, United States) during the night (at 4°C), and then with secondary antirabbit antibodies (1 : 200) conjugated to Cy3 (Jackson ImmunoResearch Europe Ltd., United Kingdom) for 40 min (at 37°C). To visualize lipid rafts rich in cholesterol, a specific marker of lipid rafts, the cholera toxin β-subunit (5 mg/mL) conjugated to FITC (FITC-CTB, Sigma-Aldrich), was used. After staining the FITC-CTB cells, the coverslips were enclosed in a Vectashield Mounting Medium resin (Vector Laboratories, United States) and examined using an Olympus FV3000 confocal microscope (Olympus Corporation, Japan) using a 60× oil lens. The fluorescence intensity (at least 7–15 cells) in each experiment) was evaluated using the ImageJ program (NIH, United States) and averaged. The fluorescence intensities in the control and treated cells were compared using the standard t-Student’s criterion. Data are presented as mean values and their errors. The differences were considered significant when p < 0.05.

Immunoelectronic microscopy. Control and MbCD-treated cells were fixed in a PBS solution containing 2% paraformaldehyde and 0.5% glutaraldehyde for 1 h and then dehydrated in alcohols and poured into LR-White acrylic resin. Ultrathin sections were placed on nickel grids coated with formvar film. For immunostaining, sections were exposed for 20 min in 1% BSA on PBS, and then incubated for 1 h first with primary ones (at a dilution of 1 : 100) anti-TRPV5 antibodies (H-99, Santa Cruz Biotechnology, United States), then with secondary (1 : 50) antirabbit antibodies conjugated to colloidal gold (Electron Microscopy Sciences, Hatfield, PA, United States). We used colloidal gold with a particle size of 10 nm. Cell sections treated with antibodies were contrasted with uranyl acetate and lead citrate and analyzed using a Libra-120 electron microscope (Germany).

Preparation of membrane fragment preparations for immunoelectron microscopy. The procedure described in (Sanan and Anderson, 1991) was used. Coverslips with preseeded cells were quickly cooled by immersion in ice buffer (25 mM Hepes, 25 mM KCl, 2.5 mM Mg (CH3COO)2) and flipped onto nickel nets with a form film precoated with poly-L-lysine. Using a rubber stopper, coverslips with cells were pressed against nickel nets for 20 s. The glasses were then sharply torn off from nickel nets, while sections of the plasma membrane of cells remained on the nets. Nickel nets with cell debris were washed gently in ice-cold buffer and fixed in 2% paraformaldehyde (10 min). The cell membranes attached to the grids were then inverted into droplets containing primary anti-TRPV5 and secondary (conjugated to colloidal gold) antibodies (indicated above) and incubated for 30 min. After washing in PBS, the samples were fixed in a solution of 2% glutaraldehyde (1 h) and left to wash overnight in PBS at 4°C. After that, the samples were postfixed in 1% OsO4 (10 min) and washed with buffer and distilled water. Next, the samples were treated with a 1% aqueous solution of tannic acid (10 min) and stained with 1% uranyl acetate (10 min). After drying, membrane fragments attached to the grids were analyzed using a Libra-120 electron microscope (Germany).

RESULTS

Lipid microdomains in Jurkat cells were evaluated by the level of fluorescence of the FITC-CTB conjugate, which selectively binds to ganglioside GM1 (one of the marker components of rafts). GM1 ganglioside staining revealed areas of intense luminescence on the plasma membrane of Jurkat T cells (Figs. 1, green FITC-CTB fluorescence).

Fig. 1.
figure 1

Immunofluorescence distribution of TRPV5 calcium channels in control and MbCD-treated Jurkat T cells. Cell images (from left to right): in transmitted light, green FITC-CTB fluorescence (marker of lipid rafts); red fluorescence of TRPV5 stained with Cy3 tagged antibodies and image combining. The scale is 10 µm. Down below histogram of the distribution of fluorescence intensity (IF) on the membrane and in the cytoplasm of cells before and after extraction of cholesterol using MbCD, respectively. Scan along the cell diameter. The data are presented as mean and its error. It is clear that IF corresponding to ganglioside GM1 and TRPV5 channels. After cell treatment, MbCD significantly decreases on the cell membrane, while the corresponding IF TRPV5 increases in cytosol.

Full size image

The localization of TRPV5 channels in cells was studied by immunofluorescence staining with specific antibodies against TRPV5 (as indicated in Materials and Methods). As is shown in Fig. 1, TRPV5 immunoreactivity was mainly concentrated on the plasma membrane, and it was also partially present in the intracellular space of cells. For the extraction of membrane cholesterol and the destruction of lipid rafts, an MbCD cholesterol acceptor was used. Incubation of cells with 1% MbCD did not change the morphology of T cells (Fig. 1), but significantly reduced the staining of the plasma membrane compared to control cells. In cells treated with a lipid raft destructor, a more diffuse distribution of TRPV5 channels was observed than in control cells (Fig. 1). As the results showed, depletion of membrane cholesterol led to a significant decrease in the intensity of the fluorescent signal on the membrane of cells stained with anti-TRPV6 antibodies, but at the same time increased the signal from antibodies inside the cells (Fig. 1).

Imaging of discrete proteins by immunofluorescence is a difficult issue, given their small size. Therefore, an electron microscope was used to overcome the diffraction limit of light microscopy and to analyze the spatial distribution of TRPV5 proteins in T cells of human leukemia. For electron microscopic analysis, pretreated MbCD and control cells were incubated with antibodies against TRPV5, and secondary antibodies conjugated with gold (as indicated in Materials and Methods). Analysis of the images showed that some of the colloidal gold particles (10 nm) are individually distributed over the cell membrane, while others are in clusters of about 50–100 nm in size in the immediate vicinity of the plasma membrane (Fig. 2, I) The presence of clusters of gold particles suggests a high local concentration of TRPV5 molecules, this being consistent with our electrophysiological (Vassilieva et al., 2013) and immunofluorescence data, which revealed the clustering of TRPV5 channels in the plasma membrane of cells (Fig. 1). It is very important that the distribution of gold particles differed in control cells and treated with MCD. In fact, gold labeling was observed both in the plasma membrane and in the cytosol of control cells (Fig. 2, I). However, after depletion of membrane cholesterol using MbCD, gold particles were rarely found on the plasma membrane of cells. Moreover, local clusters of gold particles labeling TRPV5 in the form of clusters were practically not observed in the plasma membrane of cells treated with MbCD (Fig. 2, I).

Fig. 2.
figure 2

Microphotographs of TRPV5 channel localization in (I) Jurkat cells and (II) their membrane fragments in the control and after the action of MbCD according to immunoelectron microscopy. Channels are labeled with specific antibodies conjugated to colloidal gold particles (10 nm). Gold particles (arrows) are highlighted in a (I) frame or (II) oval. In control cells, gold particles are localized in clusters on the plasma membrane; after cholesterol extraction, the particles are diffusely distributed in the cells (arrow). On the histogram (on right), number of gold particles in the control and after the action of MbCD; vertical segments, error of the mean. Differences are significant at p < 0.05.

Full size image

To obtain additional data on the membrane distribution of TRPV5 channels in control cells and cells with low cholesterol, we modified the method of isolation of fragments of the plasma membrane of cells on nickel networks (Sanan and Anderson, 1991). For this purpose, cells fixed, on the one hand, on a nickel mesh and, on the other, on a coverslip were quickly torn apart by abrupt removal of the coverslip (see Materials and Methods). The cytoplasm with intracellular contents was washed, and membrane fragments were treated with antibodies and then analyzed using an electron microscope.

The distribution of gold particles marking TRPV5 in the control cells is illustrated in the form of micrographs and quantitatively shown in Fig. 2, II. The micrograph shows that the gold particles marking TRPV5 are distributed singly or as small scattered clusters, which rarely exceed a diameter of 100 nm (scale size in Fig. 2, II). The average number of particles gold labeling TRPV5 (manually counted) in control cells was 124.9 ± 9.0 (n = 12) (Fig. 2, II). About half of the gold particles were distributed diffusely, while the rest were in clusters, which most often contained two to four particles, and less often up to ten particles (Fig. 2, II). After extraction of membrane cholesterol with MbCD, an almost 10-fold decrease in the number of gold particles was observed in the plasma membrane of cells (13.0 ± 0.88, n = 11), which were diffusely distributed over the membrane (Fig. 2, II) It is important to note that MbCD clusters of gold particles in the form of clusters were practically not detected in the treated cells.

DISCUSSION

Selective association of a number of ion channels with certain lipids of the plasma membrane and lipid rafts is critical for the regulation of their activity. In the present work, a relationship was established between the localization of TRPV5 calcium channels in the plasma membrane of human leukemia T cells and the integrity of lipid rafts. Using immunofluorescence and immunoelectron microscopy, colocalization of channels with lipid rafts is shown. Extraction of membrane cholesterol led to the destruction of lipid rafts and the diffuse distribution of channels in the cytoplasm of cells. The results indicate that cholesterol is critical for the localization of TRPV5 channels in the plasma membrane of Jurkat T cells. Moreover, the observed association of channels into clusters indicates the important role of lipid rafts in the regulation of channel activity. Membrane rafts lipids can affect channel activity either through direct protein-lipid interactions or by affecting the physical properties of a bilayer membrane (Martens et al., 2000; Hering et al., 2003). In addition, rafts, apparently, can selectively recruit the desired signal molecules for efficient and selective signal transmission. Direct interaction with raft-associated proteins, such as caveolin, can also affect channel functions by changing the ionic conductivity or by affecting the transport of channels and their exposure in the plasma membrane.

A number of studies have shown that lipids of various metabolic pathways, including metabolites of cyclooxygenases, lipoxygenases, phospholipids, and lysophospholipids (Szallasi and Blumberg, 2007; Bang et al., 2010). In addition, signaling proteins that accumulate in lipid rafts can affect the function of ion channels. Thus, in lipid rafts, a number of receptors associated with G proteins, various classes of G proteins, adenylate cyclase, protein kinase C, nitric oxide synthase, tyrosine kinase, Ras proteins, and mitogen-activated protein kinases, etc., were found (Patel et al., 2008).

According to the results, lipid rafts regulate TRPV5 channel activity in human T cells, apparently affecting their localization and density in the plasma membrane. The data obtained are quite important, since it is known that lipid rafts play a key role in receptor signaling and lymphocyte activation (Montixi et al., 1998; Xavier et al., 1998; Zhang et al., 1998; Ilangumaran et al., 1999). The participation of lipid rafts in the early stages of T-cell receptor stimulation was demonstrated on T cells (Schade and Levine, 2002). There is evidence of the participation of lipid rafts in autoimmunity. Some studies suggest a correlation between the level of lipids forming lipid rafts and the regulation of the immune system (Jury et al., 2007). There is evidence that the signaling processes in the lipid T-cell microdomains of healthy people and patients with autoimmune diseases such as systemic lupus erythematosus and rheumatoid arthritis have significant differences (Jury et al., 2004, 2007).

We suggest that signaling platforms formed by lipid rafts in T cells can recruit calcium channels along with signaling proteins. TRPV5, in providing a local increase in calcium, performs many signaling functions, including protein phosphorylation.