Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M408942200 on September 7, 2004

J. Biol. Chem., Vol. 279, Issue 47, 49160-49171, November 19, 2004
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/47/49160    most recent
M408942200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Inoue, M.
Right arrow Articles by Levi, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Inoue, M.
Right arrow Articles by Levi, M.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Partitioning of NaPi Cotransporter in Cholesterol-, Sphingomyelin-, and Glycosphingolipid-enriched Membrane Domains Modulates NaPi Protein Diffusion, Clustering, and Activity*

Makoto Inoue{ddagger}§, Michelle A. Digman§, Melanie Cheng§, Sophia Y. Breusegem{ddagger}||, Nabil Halaihel{ddagger}, Victor Sorribas**, William W. Mantulin¶, Enrico Gratton¶, Nicholas P. Barry{ddagger}||, and Moshe Levi{ddagger}||{ddagger}{ddagger}

From the Departments of {ddagger}Medicine, ||Physiology, and Biophysics, Division of Renal Diseases and Hypertension, University of Colorado Health Sciences Center and Denver Veterans Affairs Medical Center, Denver, Colorado 80262, Laboratory for Fluorescence Dynamics, Department of Physics, University of Illinois, Urbana-Champaign, Illinois 61801, and the **Department of Toxicology, University of Zaragoza, E-50013 Zaragoza, Spain

Received for publication, August 4, 2004 , and in revised form, September 1, 2004.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
In dietary potassium deficiency there is a decrease in the transport activity of the type IIa sodium/phosphate cotransporter protein (NaPi) despite an increase in its apical membrane abundance. This novel posttranslational regulation of NaPi activity is mediated by the increased glycosphingolipid content of the potassium-deficient apical membrane. However, the mechanisms by which these lipids modulate NaPi activity have not been determined. We determined if in potassium deficiency NaPi is increasingly partitioned in cholesterol-, sphingomyelin-, and glycosphingolipid-enriched microdomains of the apical membrane and if the increased presence of NaPi in these microdomains modulates its activity. By using a detergent-free density gradient flotation technique, we found that 80% of the apical membrane NaPi partitions into the low density cholesterol-, sphingomyelin-, and GM1-enriched fractions characterized as "lipid raft" fractions. In potassium deficiency, a higher proportion of NaPi was localized in the lipid raft fractions. By combining fluorescence correlation spectroscopy and photon counting histogram methods for control and potassium-deficient apical membranes reconstituted into giant unilamellar vesicles, we showed a 2-fold decrease in lateral diffusion of NaPi protein and a greater than 2-fold increase in size of protein aggregates/clusters in potassium deficiency. Our results indicate that NaPi protein is localized in membrane microdomains, that in potassium deficiency a larger proportion of NaPi protein is present in these microdomains, and that NaPi lateral diffusion is slowed down and NaPi aggregation/clustering is increased in potassium deficiency, both of which could be associated with the decreased Na/Pi cotransport activity in potassium deficiency.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Disorders of extracellular inorganic phosphate (Pi) concentration and impairments in renal and gastrointestinal Pi reabsorption are common problems. Aging, diabetes mellitus, malignancy, alcoholism, transplantation, AIDS, and several therapeutic drugs are well known to cause or to be associated with hypophosphatemia or hyperphosphatemia, mainly by affecting renal tubular Pi transport. The kidney plays a critical role in the regulation of Pi homeostasis (1). The evidence to date indicates that regulation of the overall renal tubular Pi transport by dietary, hormonal, or metabolic factors occurs mainly at the level of the proximal tubular apical brush border membrane (BBM)1 Na/Pi cotransport system. Three distinct families of renal Na/Pi cotransporters have been identified: type I, type II, and type III. These Na/Pi cotransporters are expressed in the proximal tubule of humans, rats, mice, and rabbits. The type IIa renal apical BBM Na/Pi cotransport system mediates the majority of the renal proximal tubular BBM Na/Pi transport (1).

Dietary factors, hormones, metabolic factors, and the developmental and aging process regulate Na/Pi cotransport activity by diverse molecular and cellular mechanisms, including transcriptional control, translational control, and most importantly control via acute trafficking (endocytosis or exocytosis) of the type IIa Na/Pi cotransport protein to and from the apical membrane. The net result is that with one known notable exception Na/Pi cotransport activity is directly correlated with the apical BBM type IIa Na/Pi cotransport protein abundance (1).

We have demonstrated that alterations in renal lipid composition, including BBM cholesterol, sphingomyelin, and glycosphingolipid content play an important role in the regulation of renal Na/Pi cotransport activity. Specifically, adaptation to changes in dietary Pi as well as the aging process is associated with alterations in apical BBM cholesterol content, and there is an inverse relationship between BBM Na/Pi transport activity and BBM cholesterol content (2, 3). In addition, in diabetes, in dietary potassium deficiency, and following treatment with glucocorticoids there is an inverse relationship between BBM Na/Pi transport activity and BBM sphingomyelin and glycosphingolipid (glucosylceramide and ganglioside GM3) content (4, 5). We have shown that in isolated BBM in vitro modification of cholesterol composition selectively modulates Na/Pi transport activity and that BBM Na/Pi cotransport activity is directly related to BBM cholesterol content, indicating that changes in membrane cholesterol per se can modulate Na/Pi cotransport activity (6). Similarly, inhibition of glucosylceramide and ganglioside GM3 synthesis also results in modulation of Na/Pi transport activity (4).

An important role for glycosphingolipids in mediating post-translational regulation of Na/Pi cotransport activity is illustrated in dietary potassium deficiency where there is a significant decrease in BBM Na/Pi cotransport activity, which occurs despite an increase in BBM NaPi protein abundance. Treatment of potassium-deficient rats with an inhibitor of glucosylceramide and ganglioside GM3 synthesis normalizes BBM Na/Pi cotransport activity, despite no changes in NaPi protein abundance (5). This study therefore indicates an important and a novel role for lipids in mediating posttranslational regulation of Na/Pi cotransport activity.

Although our previous studies do indicate that alterations in cholesterol and glycosphingolipid composition modulate renal Na/Pi cotransport activity, the mechanisms by which lipids modulate Na/Pi cotransport activity have not been determined and remain unknown. Most interestingly, cholesterol, sphingomyelin, and glycosphingolipids are important components and mediators of formation of specialized membrane microdomains known as lipid rafts (7-12). We have shown previously by using fluorescence spectroscopy and two-photon excitation fluorescence microscopy techniques that cholesterol and glycosphingolipid-enriched membrane microdomains or lipid rafts are present in the renal BBM, that these domains have a marked decrease in lipid dynamics, and that the formation of these domains is modulated by the cholesterol content of the membrane (13-16).

The purposes of the present study are as follows: 1) to determine whether the Na/Pi cotransport protein is present in these cholesterol-, sphingomyelin-, and glycosphingolipid-enriched membrane microdomains or lipid rafts, 2) to determine whether in the model of dietary potassium deficiency there is increased partitioning of the Na/Pi cotransport protein in these membrane microdomains, and 3) to determine whether the partitioning of the NaPi protein in these membrane microdomains modulates its activity and its dynamics in the BBM lipid bilayer. In this work we explore protein dynamics by using fluorescence correlation spectroscopy (FCS). This technique has been shown to be a noninvasive method that allows for the measurement of freely diffusing and bound molecules inside live cells in small volumes (sub-femtoliters) (30-32, 52-54). In the present study we use giant unilamellar vesicles (GUVs) grown from renal BBM, and we measure the diffusion and clustering properties of NaPi in the membrane.

The results indicate the following. 1) By using detergent-free isolation of apical membrane domains, the Na/Pi cotransport protein does preferentially partition into cholesterol-, sphingomyelin-, and glycosphingolipid-enriched membrane microdomains that have a marked decrease in lipid dynamics. 2) In apical brush border membranes isolated from potassium-deficient rats, there is a further increased presence of Na/Pi cotransport protein in these membrane microdomains that correlates with decreased Na/Pi cotransport activity despite increased Na/Pi cotransport protein abundance. 3) By using FCS, we show for the first time that increased partitioning of the NaPi protein in the cholesterol-, sphingomyelin-, and glycosphingolipid-enriched membrane microdomains results in a marked decrease in the lateral diffusion and a marked increase in the aggregation/clustering of Na/Pi cotransport protein that is associated with decreased Na/Pi cotransport activity.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Animals—Six-week-old Sprague-Dawley male rats weighing between 150 and 175 g were obtained from Harlan Sprague-Dawley (Madison, WI) and housed at the animal facility where they were kept on a 12-h light/dark cycle. The animals were fed a control diet (Teklad, TD88238) or a potassium-deficient diet (Teklad, TD95006) for 14 days (5). The diets were otherwise matched for their phosphorus, calcium, magnesium, sodium, protein, carbohydrate, and fat content. We used six rats for each individual experiment, and the experiments were repeated four times. The animal studies were approved by the Institutional Review Board.

Materials—All chemicals were from Sigma except where noted. Polyclonal NaPi rabbit antiserum was generated and used as described before (5, 17). Monoclonal antibody against {beta}-actin was obtained from Sigma. Ezrin antibody was obtained from Cytoskeleton Inc. Antibody to MAP-17 was a gift from Victor Sorribas (University of Zaragoza, Spain). NHERF-1 antibody was a gift from Ed Weinman (University of Maryland, Baltimore). PDZK-1 antibody was a gift from Olivier Kocher (Harvard Medical School, Boston). SGLT-1 antibody was obtained from Alpha Diagnostics International (San Antonio, TX). NHE3 antibody was obtained from Chemicon International, Temecula, CA. Flotillin-1, 5'-nucleotidase (5'NT), and horseradish peroxidase-conjugated antibodies were obtained from Santa Cruz Biotechnology.

Isolation of Brush Border Membrane Vesicles—Rats were anesthetized via an intraperitoneal injection of 100 mg/kg pentobarbital sodium (Pentothal, Abbott). After clamping off the renal vessels, the kidneys were removed, and thin slices from the superficial cortex were dissected on ice. The superficial cortex kidney slices were homogenized in 15 ml of an ice-cold isolation buffer consisting of 300 mM mannitol, 5 mM EGTA, 1 mM phenylmethylsulfonyl fluoride, 16 mM HEPES, and 10 mM Tris, pH 7.5, using a Polytron homogenizer (90 s at 40% power). The BBM were isolated from the homogenate by Mg2+ precipitation followed by differential centrifugation as described previously (3, 4). Briefly, 0.54 ml of 1 M MgCl2 and 21 ml of water were added to each 15 ml of kidney homogenate. After 20 min of shaking, the homogenate was centrifuged at 2,790 x g for 15 min. The supernatant was subjected to another round of Mg2+ precipitation, and the resulting supernatant was centrifuged at 40,000 x g for 30 min. The resulting BBM pellets were resuspended in a buffer containing 300 mM mannitol, 16 mM HEPES, and 10 mM Tris, pH 7.5, and aliquoted for (a) measurement of total protein concentration, (b) Na/Pi cotransport activity measurement, (c) protein electrophoresis and Western blotting, (d) OptiPrep gradient flotation for the isolation of BBM fractions, and (e) preparation of giant unilamellar vesicles (GUVs) for FCS measurements. BBM protein concentration was determined by the method of Lowry et al. (18).

Transport Activity Measurements—Transport activity measurements were performed in fresh isolated BBM vesicles by radiotracer uptake followed by rapid Millipore filtration (4). To measure sodium gradient-dependent 32Pi uptake (Na/Pi cotransport), 10 µl of BBM or membrane fragments preloaded in an intravesicular buffer (in mM) of 300 mannitol, 16 HEPES, and 10 Tris, pH 7.5, was vortex-mixed at 25 °C with 40 µl of an extra-vesicular uptake buffer of 150 mM NaCl, 100 µM K2H32PO4 (PerkinElmer Life Sciences), 16 mM HEPES, and 10 mM Tris, pH 7.5. Sodium gradient-dependent D-[3H]glucose uptake (sodium/glucose cotransport) was measured as above except for using 100 µM D-[3H]glucose. Uptake was terminated after 5 s (representing the initial linear rate) by an ice-cold stop solution. All uptake measurements were performed in triplicate, and uptake was calculated on the basis of specific activity determined in each experiment and expressed as pmol Pi or D-glucose x (5 s)-1 x (mg of BBM protein)-1.

Isolation of BBM Detergent-resistant and Detergent-sensitive Fractions—For isolation of detergent-resistant (DR) and detergent-sensitive (DS) BBM fractions, the BBM sample was first incubated for 30 min on ice in TNET buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 5 mM EDTA, and a complete protease inhibitor mixture) containing 1% Triton X-100 or 1% Lubrol WX (19). Following centrifugation at 100,000 x g at 4 °C for 1 h, the DR (pellet) and DS (supernatant) fractions were collected and analyzed for (a) total protein by BCA protein assay (Pierce) and (b) Western blots for NaPi.

Isolation of BBM Lipid Raft Fractions by Non-detergent Density Gradient Ultracentrifugation Using OptiPrep (Iodixanol)—BBM were incubated in 100 µl of TNET buffer at 4 °C for 30 min. After Dounce homogenization on ice, the extract was adjusted to 50% Optiprep, after which 600 µl was overlaid with 1 ml each of 40, 30, and 20% Optiprep in TNET and then 400 µl of 10% Optiprep in TNET (20, 21). Following centrifugation at 170,000 x g at 4 °C for 4 h, a continuous density gradient was formed. 8x 500-µl fractions were collected from the top to the bottom of the gradient and analyzed for (a) total protein by BCA protein assay (Pierce); (b) Western blots for NaPi, SGLT-1, NHE3, NHERF-1, PDZK-1, MAP-17, {beta}-actin, ezrin, 5'NT, and flotillin-1; and (c) lipid composition.

Protein Electrophoresis and Western Blotting—Intact BBM or BBM membrane fraction samples were denatured for 3 min at 95 °C in 2% SDS, 10% glycerol, 0.5 mM EDTA, and 95 mM Tris-HCl, pH 6.8. Samples corresponding to 10 µg of total protein were loaded on 9% polyacrylamide gels (Criterion gels, Bio-Rad) and separated according to the method of Laemmli (22). Proteins were electrotransferred onto nitrocellulose paper (23). After blockage with 5% nonfat milk powder with 1% Triton X-100 in Tris-buffered saline (20 mM, pH 7.3), blots were incubated with antiserum against the C-terminal amino acid sequence of NaPi at a dilution of 1:4,000. Horseradish peroxidase-linked secondary antibody was used at a dilution of 1:50,000. Primary antibody binding was visualized using enhanced chemiluminescence (Pierce), and the signals were quantified in a PhosphorImager with chemiluminescence detection and densitometry software (Bio-Rad). The protein blots were also probed with antiserum against {beta}-actin (1:5000), ezrin (1:500), MAP-17 (1:1000), NHERF-1 (1:3000), PDZK-1 (1:10,000), SGLT-1 (1:500), NHE3 (1:2000), 5'NT (1:200), and flotillin-1 (1:500).

Lipid Extraction and Measurement of Lipid Composition—Lipids were extracted by the method of Bligh and Dyer (24). Total cholesterol was determined enzymatically using Amplex Red (Molecular Probes, Eugene, OR, or Wako Chemicals, Richmond, VA). Sphingomyelin was determined with a fluorescence assay (25), using reagents supplied in the Amplex Red Sphingomyelinase kit (Molecules Probes). To determine GM1 content, the BBM were incubated with Alexa 555-labeled cholera toxin B (Molecular Probes), which binds to GM1. After separation of the BBM fractions by flotation on an OptiPrep gradient, the relative intensity of the Alexa 555 signal in each fraction was determined in a fluorescence spectrometer (26). The signal from each fraction was corrected against blank fractions from an OptiPrep gradient that was seeded with an identical amount of Alexa 555-cholera toxin B in buffer but no BBM.

Measurement of NaPi Lateral Diffusion and Aggregation/Clustering by FCS and PCH—Fluorescence correlation spectroscopy (FCS) measurements were performed on giant unilamellar vesicles (GUVs) grown from intact BBM. GUVs were prepared by the electroformation method developed by Angelova and Dimitrov (27), as we have described previously (15, 16, 28-30). Briefly, membrane samples were laid onto platinum wires and dried in a temperature-controlled chamber that fits on the microscope stage. The membranes were then hydrated in buffer at about 43 °C (above the liquid-ordered to liquid-disordered phase transition temperature) and in the presence of an alternating electric field (10 Hz, 3V).

A highly specific and affinity-purified anti-NaPi rabbit polyclonal antibody (5, 17) was further purified with an Affi-Gel Protein A column (Bio-Rad) and conjugated with the amine-reactive fluorescent probe Alexa 488 by using an Alexa 488 protein labeling kit (Molecular Probes). Conjugates were labeled with an average of three dye molecules per antibody molecule.

FCS measurements were performed at 37 °C as described previously (28-31) by using two-photon excitation at 785 nm. The two-photon excitation scanning fluorescence microscope used in these experiments was assembled in the Laboratory for Fluorescence Dynamics (University of Illinois, Urbana-Champaign) and has been described previously (31). A mode-locked titanium-sapphire laser with 80-MHz, 100-fs pulse width (Tsunami; Spectra-Physics, Mountain view, CA) was used as the excitation light source. The laser was guided into the microscope by x-y galvano-scanner mirrors (model 6350; Cambridge Technology, Watertown, MA) to achieve beam scanning in both x and y directions. The scanner mirrors were moved by a voltage generated in a computer card. The movement of the x scanner mirror was independent from the y scanner mirror movement. For the laser beam to move in a circular path, the x and y scanner mirrors were driven by two identical sine waves with 90° phase shift. The radius and frequency of the circular scan were controlled by the amplitude and frequency of the sine waves. For a raster scan, the x and y scanner mirrors were driven by two saw-tooth signals at different frequency. A photomultiplier tube (Hamamatsu R7400P, Japan) was used for light detection in the photon counting mode. A BG39 optical filter (Chroma Technologies, Brattleboro, VT) was placed before the photomultiplier for suppression of IR excitation light. A Zeiss C-Apo 40X (1.2 N.A.), water-immersion objective lens was used for the measurement because of its exceptionally long working distance (230 µm).

Two-photon Microscopy, FCS, Scanning FCS, and PCH Analysis— For two-photon excitation fluorescence imaging, the scanning areas ranged from 70 x 70 µm to 18 x 18 µm. Images were 256 x 256 pixels and averages of 10 frames with a pixel residence time of 50 µs. Regions of interest were directly selected from the fluorescence image. For single point FCS, the sampling rate was 64 kHz, and the total measurement lasted less than 5 min. For scanning FCS the orbital scanning frequency was 1 kHz with a sampling rate (pixel rate) of either 40 or 64 kHz and total measurement times of 400 or 255 s, respectively. The laser power at the sample was 10 milliwatts. The average fluorescence intensity of the sample remained constant, indicating that the fluorophore was not photobleached during the measurement. Because of the variation in the laser alignment from day to day, the waist {omega}0 of the excitation beam was calibrated before each day's measurement by measuring an autocorrelation curve of 10 nM fluorescein in 0.01 M NaOH. {omega}0 was derived from a fit to this curve by assuming a fluorescein diffusion coefficient of 300 µm2/s; typically {omega}0 values of 0.3 µmto0.5 µm were obtained.

The autocorrelation function, G({tau}) in Equation 1, was used to analyze the fluorescence fluctuations, where N(t) is the fluorescence intensity at time t; {tau} is some absolute time delay, and the angle brackets denote a time average. For a single molecular species the autocorrelation function at {tau} = 0 (G(0)) is inversely proportional to the average number of molecules in the excitation volume (Equation 2).

(Eq. 1)

(Eq. 2)

For single point FCS data, the experimental autocorrelation function was fitted using a three-dimensional Gaussian-Lorentzian beam profile (32-34) and one- or two-dimensional diffusional components. The concentration (C) of the antibody was calculated from the G(0) value obtained from the autocorrelation fits according to Equation 3, where NA is Avogadro's number, and VPSF is the volume of the point spread function (~1 femtoliter). The factor {gamma} accounts for the illumination of the excitation volume and equals 0.072 for the 2-photon excitation profile.

(Eq. 3)

For the scanning FCS measurement, the center of the circular scanning path was directly selected from the fluorescence image. The data acquisition frequency was 40 or 64 kHz, and the scanning frequency was 1 kHz, yielding 40 or 64 data points in each scanning cycle. The data were plotted on an intensity carpet where the x axis is the position along the scan and the y axis is time (30). The scanning diameter was between 5 and 23 µm depending on the imaging area. Each column on the x axis, which represents the 40 or 64 points along the orbit of the scan as shown by the intensity carpet, was temporally correlated and fitted by using a procedure similar to that for the single point FCS data and is described in Equation 4. The two-dimensional diffusion terms were complemented with an exponential term to account for incomplete relaxations of slowly diffusing species. {tau}c represents the residence time of the molecules in the excitation volume. Each diffusional component contributes with an amplitude G0i, where i is either a slow, fast, or exponential component, subtracted by the background B. The diffusion constants D (in µm2/s) were calculated from {tau}c according to Equation 5.

(Eq. 4)

(Eq. 5)

The PCH analysis is based on the probability distribution of finding molecules in the observation volume as detected by the FCS experiments (33). The amplitude of the intensity fluctuation depends on the intrinsic brightness of the particle and on the location of the particle in the intensity beam profile. From the PCH analysis, the concentration and the molecular brightness ({epsilon}, in counts/molecules/s) can be calculated (33, 35, 36). This information allows for the determination of molecules that are bound together and moving as one unit. Therefore, dimers, trimers, and oligomers can be detected if the molecular brightness increases by 2-fold, 3-fold, etc. The Alexa 488-labeled NaPi antibody was measured in solution after purification to determine the intrinsic molecular brightness before binding. PCH analyses were done for the free antibody inside, outside, and on the GUV from the scanning experiments. Fitting of both autocorrelation functions and PCH data was done by using Laboratory for Fluorescence Dynamics Globals Unlimited software (Champaign, IL).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
In Potassium Deficiency BBM Na/Pi Cotransport Activity Is Decreased Despite Increased BBM NaPi Protein Abundance—In agreement with our earlier publication (5), we have found that in BBM isolated from rats with dietary potassium deficiency Na/Pi cotransport activity is decreased (Fig. 1A) despite a significant increase in NaPi protein abundance (Fig. 1B). Transport kinetic studies showed that the decrease in BBM NaPi transport activity is mediated by a decrease in the Vmax and no change in the Km (5). This effect is quite selective for NaPi as potassium deficiency resulted in no changes in BBM sodium/glucose cotransport activity (Fig. 1C) or BBM sodium/glucose cotransporter protein (SGLT-1) abundance (Fig. 1D). However, we did find increased abundance of the Na/H exchanger protein NHE3 in BBM from potassium-deficient rats compared with control BBM (data not shown). This is in agreement with earlier studies that have shown Na/H exchange activity and NHE3 protein to be up-regulated in potassium deficiency (37).



View larger version (23K):
[in this window]
[in a new window]
  		FIG. 1.
Effect of potassium deficiency on BBM transport activities and abundance of Na/Pi and sodium/glucose cotransporter proteins. A, Na/Pi cotransport activity; B, NaPi protein abundance; C, sodium/glucose cotransport activity; D, sodium/glucose cotransport protein (SGLT-1) abundance. Potassium deficiency causes a significant decrease in BBM Na/Pi cotransport activity despite an increase in BBM NaPi protein abundance. In contrast, potassium deficiency has no significant effects on sodium/glucose cotransport activity or SGLT-1 protein abundance.

 
NaPi Protein Is Partitioned into Cholesterol-, Sphingomyelin-, and Glycosphingolipid-enriched BBM Domains—DR and DS BBM fractions were isolated by high speed centrifugation in the presence of 1% Triton X-100 (Fig. 2A) or 1% Lubrol WX (Fig. 2B). We found that in BBM isolated from control rats approximately 40% of NaPi protein is in the DR membrane fraction, which is highly enriched in cholesterol, sphingomyelin, and glycosphingolipids. We also fractionated BBM isolated from control rats using a non-detergent OptiPrep gradient flotation technique (20, 21). We found that the majority of the NaPi protein localizes in fractions 1-3 (Fig. 3A), which are also highly enriched in cholesterol (Fig. 3B), sphingomyelin (Fig. 3C), and ganglioside GM1 (Fig. 3D), classical lipid raft markers (7-12).



View larger version (15K):
[in this window]
[in a new window]
  		FIG. 2.
Effects of potassium deficiency on BBM NaPi protein partitioning into DS and DR membrane fractions. Control and potassium-deficient BBM were treated with 1% Triton X-100 at 4 °C for 30 min (A) or 1% Lubrol WX at 4 °C for 30 min (B). Following centrifugation at 100,000 x g at 4 °C for 1 h, the DR (pellet) and DS (supernatant) fractions were collected and analyzed for NaPi protein abundance by Western blotting. 40% of the NaPi protein from control BBM is present in the Triton X-100-(40 ± 5%) or Lubrol WX (40 ± 6%)-resistant membrane fraction. In potassium-deficient BBM, the fraction of NaPi present in detergent-resistant membranes is significantly (p < 0.01) increased to 70 ± 8% for Triton X-100 treatment and to 60 ± 8% for Lubrol WX treatment.

 



View larger version (14K):
[in this window]
[in a new window]
  		FIG. 3.
NaPi protein is preferentially partitioned into cholesterol-, sphingomyelin-, and glycosphingolipid-enriched BBM domains. BBM isolated from control rats were fractionated using a non-detergent OptiPrep gradient flotation technique, and the fractions were analyzed for NaPi protein abundance (A), cholesterol content (B), sphingomyelin content (C), and GM1 content (D) as detailed under "Experimental Procedures."

 
In Potassium Deficiency There Is Increased Partitioning of NaPi Protein in Cholesterol-, Sphingomyelin-, and Glycosphingolipid-enriched BBM Domains—In BBM DR and DS fractions isolated by single step high speed centrifugation in the presence of 1% Triton X-100 (Fig. 2A) or 1% Lubrol WX (Fig. 2B), we found that in BBM isolated from potassium-deficient rats there is a further marked and significant increase in the abundance of NaPi protein that is present in the DR membrane fraction. When BBM isolated from control and potassium-deficient rats were simultaneously subjected to non-detergent OptiPrep gradient flotation and the gradients were analyzed for NaPi protein abundance by Western blotting, we found that in BBM isolated from potassium-deficient rats there is a marked increase in the abundance of NaPi protein in fractions 2-4 (Fig. 4A). NHE3 protein also increasingly partitions in OptiPrep fractions 2-4 in potassium-deficient BBM, and compared with control BBM its abundance is higher in fractions 1-4 of potassium-deficient BBM (Fig. 4C). In contrast, the distribution for SGLT-1 is the same for BBM from control and potassium-deficient rats (Fig. 4B).



View larger version (17K):
[in this window]
[in a new window]
  		FIG. 4.
Distribution of NaPi, SGLT-1, and NHE3 in control and potassium-deficient BBM fractionated by detergent-free OptiPrep gradient flotation. BBM isolated from control and potassium-deficient rats were fractionated using a non-detergent OptiPrep gradient flotation technique. The fractions were analyzed for NaPi protein abundance (A), sodium/glucose cotransport protein (SGLT-1) abundance (B), and NHE3 protein abundance (C).

 
Partitioning of NaPi-interacting PDZ Domain-containing Proteins, {beta}-Actin, Ezrin, MAP-17, Flotillin, and 5'-Nucleotidase in Cholesterol-, Sphingomyelin-, and Glycosphingolipid-enriched BBM Domains—NHERF-1, a PDZ domain-containing protein that plays an important role in NaPi trafficking to the BBM (38-39), partitions mostly into fractions 2-3, and there are no differences in the abundance or distribution of NHERF-1 in BBM from control and potassium-deficient rats (Fig. 5A). In contrast, PDZK-1, another PDZ domain containing protein (40-41), has a broader distribution (fractions 1-4) with significantly more PDZK-1 in fraction 4 of potassium-deficient BBM compared with control BBM (Fig. 5B). Most interestingly, MAP17, another protein that interacts with NaPi protein (42, 43), has a similar distribution as NHERF-1; however, MAP17 abundance is decreased in BBM from potassium-deficient rats, especially in fraction 3 (Fig. 5D).



View larger version (34K):
[in this window]
[in a new window]
  		FIG. 5.
Distribution of NaPi-interacting PDZ domain containing proteins, ezrin, MAP17, and lipid raft marker proteins in control and potassium-deficient BBM fractionated by detergent-free OptiPrep gradient flotation. BBM isolated from control and potassium-deficient rats were fractionated using a non-detergent OptiPrep gradient flotation technique, and the fractions were analyzed by Western blotting for NHERF-1 (A), PDZK-1 (B), ezrin (C), MAP17 (D), flotillin-1 (E), and 5'NT (F).

 
Ezrin, a member of the family of ERM (ezrin-radixin-moesin) proteins (44, 45), partitions mostly into fractions 1-3, and there are no differences in the abundance or distribution of ezrin in BBM from control and potassium-deficient rats (Fig. 5C). Similarly to NaPi and NHE3, {beta}-actin was mostly partitioned into fractions 2-4, and no differences were observed in the abundance or distribution of {beta}-actin in BBM from control and potassium-deficient rats (data not shown).

Flotillin-1, a protein that has been reported to be present in lipid rafts (46, 47), partitions into fractions 1-4, and there are no major differences in the abundance of flotillin-1 in BBM isolated from control and potassium-deficient rats (Fig. 5E). In contrast, 5'NT, a glycosylphosphatidylinositol-anchored protein that has been reported to be present in lipid rafts (48-51), partitions mostly in fractions 1-3, and the abundance of 5'NT is significantly decreased in BBM isolated from potassium-deficient rats (Fig. 5F).

In Potassium Deficiency There Is a Marked Decrease in NaPi Protein Lateral Diffusion and an Increase in NaPi Protein Aggregation/Clustering—We performed FCS measurements to determine whether increased BBM sphingomyelin and glycosphingolipid content and increased partitioning of NaPi protein into cholesterol-, sphingomyelin-, and glycosphingolipid-enriched BBM domains results in alterations of its lateral diffusion.

Single Point FCS—The autocorrelation curves obtained from single point FCS measurements were fitted with the two-diffusion component model. Multiple measurements were made on different GUVs. Typically, a fast diffusion component associated with unbound antibody plus a slow component and an exponential component were seen. Over repeated measurements, four characteristic diffusion coefficients emerged. These values are listed in Table I.


View this table:
[in this window]
[in a new window]
  		TABLE I
Average diffusion coefficients of anti-NaPi in solution and membrane-bound on control and potassium-deficient GUVs measured using single point FCS Diffusion coefficients D are expressed as mean ± S.D. in units of µm2/s. n is the number of data points. p values are from unpaired t tests.

 
Representative autocorrelation curves are shown in Figs. 6 and 7. The corresponding values of the diffusion coefficients, as obtained by the fitting procedure, are identified with the region of diffusion relaxation shown in the autocorrelation curves. The fast diffusion component (D1 in Table I) for both control (62 µm2/s) and potassium-deficient (67 µm2/s) BBM is comparable with that measured for freely diffusing antibody (69 µm2/s). Furthermore, three slow diffusion coefficients were observed. These we identify with anti-NaPi bound to the GUV membrane. Autocorrelation curves obtained from GUVs grown from control BBM gave slow diffusion values, D2 = 3.31 µm2/s and D3 = 0.37 µm2/s. In contrast, fits to autocorrelation curves obtained on GUVs grown from potassium-deficient BBM yielded significantly slower diffusion rates (there was about a 2-fold difference) for the membrane-bound antibody (D2 = 1.27 µm2/s and D3 = 0.22 µm2/s). The single point FCS measurements therefore indicate that the lateral diffusion of NaPi protein in GUVs made from potassium-deficient BBM is markedly and significantly decreased when compared with control BBM. A final and very slow diffusion coefficient for the control BBM (D4 = 0.008 µm2/s) and potassium-deficient BBM (D4 = 0.015 µm2/s) was observed.



View larger version (39K):
[in this window]
[in a new window]
  		FIG. 6.
Single-point FCS on GUVs made from control and potassium-deficient BBM. A, fitted autocorrelation curve of Alexa 488-labeled anti-NaPi on a control BBM GUV. The curve was fitted to a two-component and exponential decay model, with a faster diffusion component corresponding to the free antibody in solution (D1) and a slower component corresponding to the membrane-bound antibody (D3). B, fitted autocorrelation curve of anti-NaPi on a potassium-deficient BBM GUV. The fitted curve yielded two diffusing species with longer residence time for the membrane-bound labeled antibody compared with the control. Insets, intensity image of the control (A) and potassium-deficient (B) GUV showing the binding of the antibody to the membrane. The red arrows depict the spots where the FCS measurements were made.

 



View larger version (44K):
[in this window]
[in a new window]
  		FIG. 7.
Comparison of single point FCS measurements on GUVs made from control and potassium-deficient BBM. Shown are normalized Alexa 488-anti-NaPi autocorrelation curves measured on GUVs from control (red circles) and potassium-deficient (blue squares) BBM in single point FCS. Both curves were fitted with a two-component model, yielding the diffusion of the unbound and bound antibody (D1 and D3). Insets, fluorescence intensity images of control (left) and potassium-deficient (right) GUVs labeled with Alexa 488-anti-NaPi.

 
Fig. 7 shows, for example, normalized autocorrelation curves of anti-NaPi bound to control and potassium-deficient BBM, resulting in two diffusing species for each membrane as follows: (i) a fast component (D1) corresponding to the free antibody in solution, and (ii) a slower component (D3) corresponding to the antibody bound to the membrane. Comparison of fluorescence intensity images of the antibody bound to the GUVs (Fig. 7, inset), where there is higher intensity on the potassium-deficient GUV, indicates that there is a higher concentration of NaPi on the potassium-deficient membrane. This result is consistent with the increased abundance of NaPi-IIa in BBM isolated from potassium-deficient rats (Fig. 1).

The above observations of different diffusion coefficients for bound NaPi antibody illustrate that the GUVs formed from rat BBM are highly heterogeneous in their properties. Furthermore, there are consistent differences between the control and potassium-deficient membranes. Typically diffusion in the potassium-deficient BBM (D2 and D3) is slower. We interpret the different diffusion coefficients as indicative of NaPi existing in a number of possible conditions including regions of differing membrane fluidity, association with membrane domains, aggregation of the protein, and motion of membrane domains (see "Discussion").

Scanning FCS—We then performed scanning FCS and PCH analysis in control and potassium-deficient GUVs to better determine the differences in slower diffusing species that we are less able to detect with single point FCS and to determine aggregation states in terms of molecular brightness. Scanning FCS offers the advantage of multipoint data acquisition (in this case, either 40 or 64 points). With appropriate positioning, the orbit crosses the membrane at known points. This allows motion in the membrane to be distinguished from motion of the vesicle as a whole. However, since the scanning rate is 1 kHz, determination of diffusion components faster than about 10 µm2/s is not precise.

Fig. 8A shows a representative image of a GUV formed from control BBM labeled with anti-NaPi. The adjacent intensity "carpet" shows the intensity trace along the beam orbit as a function of time. The intensity fluctuations during the measurement at any particular point on the orbit can be found by extracting the appropriate column from the carpet. Shown in Fig. 8B are autocorrelation curves calculated for columns of interest and fitted using the same model as before (Equation 4). A representative image of a GUV formed from potassium-deficient BBM and the intensity carpet for the indicated orbit is shown in Fig. 9A. The autocorrelation curves for a point on the membrane and a point outside the vesicle are shown in Fig. 9B. Over many measurements, a pattern of diffusion coefficients is seen that substantially matches those of the single point data in terms of their characteristic values and the consistent differences between the control and potassium-deficient BBM (Table II).



View larger version (52K):
[in this window]
[in a new window]
  		FIG. 8.
Scanning FCS intensity carpet and fitted autocorrelation curves of Alexa 488-anti-NaPi at different points along the scanning orbit of a control BBM GUV. The data were collected at a sampling frequency of 64 kHz with 64 data points per scan (1 ms/scan) and a total acquisition time of 255 s. A, FCS intensity carpet (left) and intensity image (right) of a control BBM GUV. In the intensity image the scanning orbit is indicated by the yellow circle, and numbers indicate positions on the scan. The FCS intensity carpet plots the intensity measured at the sampling points versus time. The dashed line corresponds to position 31. B, calculated and fitted autocorrelation curves for columns in the FCS carpet corresponding to Alexa 488-anti-NaPi on the GUV membrane ({blacktriangleup}, column 31), in the interior of the vesicle (x, column 6), and in solution ({triangledown}, column 54).

 



View larger version (50K):
[in this window]
[in a new window]
  		FIG. 9.
Scanning FCS intensity carpet and fitted autocorrelation curves of Alexa 488-anti-NaPi at two points along the scanning orbit of a potassium-deficient BBM GUV. A, FCS intensity carpet (left), plotted for data collected with a sampling frequency of 40 kHz and a total data acquisition time of 400 s, and intensity image (right) of a potassium-deficient GUV, showing the scanning orbit (yellow circle) with numbered sampling positions. B, calculated and fitted autocorrelation curves for columns in the FCS carpet corresponding to Alexa 488-anti-NaPi on the GUV membrane ({diamond}, column 33) and outside the GUV (*, column 5).

 


View this table:
[in this window]
[in a new window]
  		TABLE II
Average diffusion coefficients of anti-NaPi in solution and membrane-bound on control and potassium-deficient GUVs measured by scanning FCS Diffusion coefficients Dsc are expressed as mean ± S.D. in units of µm2/s. n is the number of data points. p values are from unpaired t tests.

 
As in the single point FCS data, a fast diffusion component (Dsc1) is seen, which we identify as free diffusion of the Alexa 488-labeled antibody. However, because of the increased delay between measurements at any particular point, this value is less well resolved in the scanning measurement. The GUVs grown from control BBM show the presence of two slow diffusing species (Dsc2 = 3.18 µm2/s and Dsc3 = 0.35 µm2/s). Similarly, for potassium deficiency, two slow components (Dsc2 = 1.37 µm2/s and Dsc3 = 0.25 µm2/s) were seen. These values correspond to antibody bound to the membrane.

The value of the very slow diffusion coefficient determined by scanning FCS (Dsc4 = 0.010 µm2/s) was again similar to the value obtained from single point FCS in the case of the control BBM; however, for potassium-deficient BBM, the values of Dsc4 = 0.0026 and Dsc5= 0.0004 µm2/s are substantially slower than seen in the single point measurements. Overall the very slow diffusion observed is between 4-fold (Dsc4) and 25-fold slower (Dsc5) in the potassium-deficient BBM compared with control. These observations suggest that there may be protein aggregates on the membrane with higher protein concentrations or that the increased levels of glycosphingolipids form microdomains into which NaPi may partition.

PCH Analysis—To distinguish these possibilities, we made measurements of the molecular brightness using PCH analysis. Using the scanning FCS intensity carpet, as shown for example in Figs. 8A and 9A, we extracted columns of intensity measurements made at particular points in the scan. These intensity traces are shown in Fig. 10 for the control vesicle, on the membrane (Fig. 10A), and outside the vesicle (Fig. 10B), and for the potassium-deficient case, for Alexa 488-anti-NaPi on the membrane (Fig. 10C) and in solution (Fig. 10D). The intensity traces on the membrane show long-lived events significantly brighter than those seen in the solution. These events give rise to the very slow diffusion coefficients (Dsc4 and Dsc5). Furthermore, the intensity of these events is higher in the case of potassium deficiency indicating more bound antibody. From these data, photon counting histograms and fits were calculated, shown in Fig. 11, A (control BBM) and B (potassium-deficient BBM), at the points indicated in the figure legend. A fit to a point outside the vesicle yields the characteristic brightness of a single freely diffusing labeled antibody giving 17 x 103 counts/molecules/s under our experimental conditions. By comparison to the brightness of the entities observed on the membrane (Table III), an estimate of the number of bound anti-NaPi (aggregation and clustering) can be made. The antibody bound on the control GUV is 2-fold brighter than the antibody in solution, indicating that there are dimers on the surface of the control membrane. In the case of the potassium-deficient BBM, the brightness of the bound antibody is 82.7 x 103 counts/molecules/s, which is a difference of ~5-fold. Analysis of data sets that yielded diffusion coefficients corresponding to Dsc2 and Dsc3 did not show an increase in molecular brightness compared with the free antibody in solution, suggesting that these entities were monomeric protein-antibody complexes (data not shown).



View larger version (30K):
[in this window]
[in a new window]
  		FIG. 10.
Intensity profiles of lipid domain structures diffusing in a single spot as a function of time. The intensity profiles were obtained from the scanning FCS intensity carpets in Figs. 8 and 9. A, intensity profile for control BBM column 31 in Fig. 8A. The intensity fluctuates strongly as the fluorescent complex diffuses in and out of the sampling spot over a period of time. B, fluctuations in intensity seen for molecules outside the control BBM GUV (column 54 in Fig. 8A). C, intensity profile for potassium-deficient BBM column 33 in Fig. 9A. The amount of bound anti-NaPi is increased in comparison to control. D, intensity profile of free antibody (column 5 in Fig. 9A).

 



View larger version (16K):
[in this window]
[in a new window]
  		FIG. 11.
PCH of membrane-bound Alexa 488-anti-NaPi measured in GUVs made from control and potassium-deficient BBM. A, PCH of the NaPi antibody on a control BBM GUV ({blacktriangleup}) and outside the GUV ({triangledown}) for the scanning data in Fig. 8A (columns 31 and 54, respectively). The data were binned by 10 for these traces with a sampling frequency of 64 kHz. B, PCH of the NaPi antibody on the potassium-deficient BBM GUV ({diamond}) and outside the GUV (*) for the scanning data in Fig. 9A (columns 33 and 5, respectively). For both A and B, the PCH data were fit to a single species model. The results of the fits are collected in Table III.

 


View this table:
[in this window]
[in a new window]
  		TABLE III
PCH of control versus potassium-deficient GUV The table lists the molecular brightness values ({epsilon}) in units of counts/molecules/s, as obtained from the fit to the data in Fig. 11.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
NaPi Is Present in Cholesterol-, Sphingomyelin-, and Glycosphingolipid-enriched BBM Microdomains and Increasingly So in Potassium Deficiency—Potassium deficiency is a novel model that suggests an important role for lipids in mediating the posttranslational regulation of the major renal phosphate transport protein NaPi. In potassium deficiency, a significant decrease in BBM Na/Pi cotransport activity is observed despite an increase in BBM NaPi protein abundance (Fig. 1), a correspondence that is associated with significant increases in BBM sphingomyelin, glucosylceramide, and ganglioside GM3 content as we showed in an earlier study (5). Treatment of potassium-deficient rats with an inhibitor of glucosylceramide and ganglioside GM3 synthesis normalizes BBM Na/Pi cotransport activity without further changing the NaPi protein abundance (5). However, the cellular and molecular mechanisms by which these lipids may modulate NaPi activity have not been determined.

In the present study we find that the decrease in Na/Pi cotransport activity in potassium deficiency despite the increase in NaPi protein abundance is associated with increased partitioning of NaPi protein into cholesterol-, sphingomyelin-, and glycosphingolipid-enriched BBM microdomains (Figs. 2, 3, 4A). This is the first demonstration of the presence of NaPi protein in membrane lipid microdomains or lipid rafts.

Increased partitioning of proteins into cholesterol-, sphingomyelin-, and glycosphingolipid-enriched membrane microdomains has differential effects on the regulation of the function and activity of proteins. For example, epidermal growth factor and clonidine-stimulated Na/H exchange activity in the ileum is associated with increased BBM NHE3 protein abundance and increased partitioning of NHE3 into lipid rafts (65, 66). NHE3 is also largely present in lipid raft domains of the apical membrane when expressed in opossum kidney cells (67), and we find that in rat kidney BBM NHE3 partitions into cholesterol-, sphingomyelin-, and glycosphingolipid-enriched BBM microdomains (Fig. 4C). Furthermore, in potassium deficiency, which previously had been shown to result in increased renal BBM Na/H exchange activity with a parallel increase in NHE3 protein abundance (37), we find further increased partitioning of NHE3 into "lipid raft" fractions (Fig. 4C). The sodium/glucose cotransporter SGLT-1 has also been shown to be partially present in "lipid rafts" (68). We confirmed the partial presence of SGLT-1 in cholesterol-, sphingomyelin-, and glycosphingolipid-enriched BBM microdomains; however, we saw no effect of potassium deficiency on the partitioning of SGLT-1 in these domains (Fig. 4B). This result is consistent with the absence of significant alterations in BBM sodium/glucose cotransport activity or BBM SGLT-1 protein abundance in potassium deficiency (Fig. 1, C and D).

NaPi Diffuses Slower and Forms Aggregates in Larger Microdomains in Potassium-deficient BBM—By using novel FCS techniques, we found that despite a higher abundance of NaPi protein in potassium-deficient BBM, its lateral diffusion is markedly decreased (Tables I and II and Figs. 6, 7, 8, 9). This is the first demonstration of a decrease in BBM NaPi protein lateral diffusion being associated with a decrease in Na/Pi cotransport activity. Most interestingly, by analysis of the PCH data (Fig. 11 and Table III), we also show evidence for increased aggregation (clustering or oligomerization) of the NaPi protein in potassium deficiency, an observation also reported for other raft proteins (55-57). Unique to our study is our ability to assign the diffusing species that we observe to specific molecular entities through the combination of the diffusion coefficient values obtained from the autocorrelation analysis of scanning FCS data and the molecular brightness of the particles obtained from the corresponding PCH analysis. Of course, we only observe the fluorescent species, which is the antibody, and we use the brightness of the single antibody molecule as obtained from solution experiments to infer the number of molecules in an aggregate.

The molecular brightness measurements are crucial for our interpretation because, contrary to solution experiments, it is impossible to determine the mass of an object diffusing in a membrane from the value of the diffusion coefficient alone, given the range of possible diffusion coefficients observed for membrane proteins and lipids in membranes and the variability of the local viscosity. For example, Kahya et al. (54) made lipid diffusion measurements on GUVs formed from ternary mixtures of cholesterol, sphingomyelin, and dioleoyl phosphatidylcholine. For a number of different fractional compositions, partitioning into liquid-ordered and liquid-disordered phases was seen, with lipid diffusion coefficients of ~0.25 and ~5 µm2/s respectively. With increasing sphingomyelin content, a decrease in the diffusion coefficient was measured (54). This offers insight into the origin of the slow NaPi diffusion components D2 and D3 (or Dsc2 and Dsc3), and the ~2-fold decrease in diffusion for the antibody bound on the potassium-deficient compared with the control BBM is consistent with the increased sphingomyelin content of potassium-deficient BBM. However, BBM has a much more complex composition than the lipid mixtures studied by Kahya et al. (54). Also, our measurements were made at 37 °C (compared with 25 °C in Ref. 54), and a phase separation was not directly observed in the GUVs. Other researchers have determined diffusion coefficients of single proteins diffusing in a membrane to be in the range of 0.5 to 0.05 µm2/s (30, 31). Our values of D2 lie at the upper end of this range, which suggests the existence of regions of increased membrane fluidity in kidney BBM.

When proteins associate with very large structures, such as vesicles or membrane rafts, the apparent diffusion coefficient will decrease by another order of magnitude to 0.005 µm2/s or smaller (70). We observe particles moving with diffusion coefficients in the range 0.01 to 0.0004 µm2/s (Dsc4 and Dsc5). These particles have a larger molecular brightness than the freely diffusing antibody in solution (Table III), i.e. the observed fluorescent entity contains more than one antibody. For control BBM, these aggregates were brighter by a factor of 2, whereas for potassium-deficient BBM the increase in brightness was by a factor of 5. This suggests that there are larger aggregates in the potassium-deficient membranes compared with the control membranes. We can now interpret this result in conjunction with the diffusion measurements. By assuming that the 5-fold increase in brightness on potassium-deficient BBM corresponds to ~5 protein-antibody complexes aggregated on the membrane surface, then this aggregate, if made of NaPi proteins only, should slow the diffusion of the antibody by a factor of about 2. However, in potassium deficiency Dsc4 is slower by a factor of ~100 compared with the diffusion coefficient of the monomeric species Dsc3. This implies that the very slow diffusing object observed in the scanning data of potassium-deficient GUVs is a large body, which includes additional proteins and/or lipids. A similar rationale can be followed for the control BBM data, for which Dsc4 is slower by a factor >30 compared with Dsc3. We believe that these very slowly diffusing molecular entities with larger brightness correspond to membrane rafts and that the intensity profiles of these entities (Fig. 10, A and C) delineate the lateral diffusion of lipid domains.

An alternative explanation is that the observed slow intensity fluctuations are because of particles attaching and detaching from an (almost) immobile object. However, an equilibrium of single antibody molecules occurring on the same time scale would lead to much less distinct fluctuations (corresponding to the gain or loss of single antibodies) than the large steps observed in our intensity profiles (Fig. 10, A and C), indicative of a bright particle moving as a rigid object. In addition, dissociation of the antibody from the NaPi protein is unlikely to occur in the time scale of about 1 s given the high affinity of the antibody. A final possibility is that the entire NaPi-antibody complex detaches from the raft region. Although this equilibrium process cannot be excluded, it cannot explain the slow fluctuation of the very bright objects we have observed.

Similar observations of aggregation have been made for other membrane proteins. For example, a physiological role for alterations in the lateral diffusion of a protein being associated with its function has been shown recently for bacteriorhodopsin, a proton-transporting membrane protein in Halophilic Archaea, which is highly sensitive to its lipid environment (53). By using FCS, a decrease in the lateral diffusion of bacteriorhodopsin was observed upon photoactivation, which was fully reversible upon return to the dark-adapted state. The shifts in lateral diffusion or mobility were determined to be caused by transient photoinduced oligomerization of bacteriorhodopsin (53). In addition, several studies have used the fluorescence recovery after photobleaching (FRAP) imaging technique (58, 59) to show that the diffusion of certain proteins can be modulated by lipids and/or their partitioning into lipid rafts. For example, one study showed that the lateral diffusion of green fluorescent protein-labeled caveolin is highly restricted but increased upon treatment of the cells with methyl-{beta}-cyclodextrin to extract cholesterol or with cytochalasin D to disrupt the actin cytoskeleton (60). Similar results following the disruption of the actin cytoskeleton were obtained with green fluorescent protein-labeled aquaporin 2 (61). Two other studies using high resolution single particle tracking showed markedly decreased diffusion of raft-associated proteins compared with non-raft proteins and increased diffusion of the raft-associated proteins following lowering of the cell cholesterol content (62, 63). However, a recent study that used FRAP to measure the diffusion coefficients of several types of putative raft and non-raft proteins under steady-state conditions and in response to raft perturbations found that raft proteins diffused freely over large distances (>4 µm), exhibiting diffusion coefficients that varied 10-fold (64). Perturbations reported to affect lipid rafts in model membrane systems or by biochemical fractionation had similar effects on the diffusional mobility of raft and non-raft proteins. This study therefore indicated that raft association is not the dominant factor in determining long range protein mobility at the cell surface (64). In contrast to FRAP which measures diffusion over a relatively large area of the membrane, the advantage of FCS is that it has single molecule sensitivity and can measure diffusion of proteins in a very small area/volume of the membrane (30-32, 52-54).

Summary and Conclusion—This study indicates for the first time the important role of the partitioning of the type IIa Na/Pi cotransporter protein into cholesterol-, sphingomyelin-, and glycosphingolipid-enriched microdomains with regard to its function in the model of potassium deficiency. We show that increased partitioning of NaPi into these lipid microdomains is associated with the decreased Na/Pi cotransport activity. By using novel FCS techniques, we not only determine decreased lateral diffusion of single NaPi molecules but also the presence of larger NaPi aggregates in potassium-deficient BBM compared with control BBM, both of which could contribute to the observed decrease in NaPi transport activity. Finally, this study demonstrates the power of combining an autocorrelation analysis and a PCH analysis of scanning FCS data to determine the nature of slowly diffusing proteins and protein complexes in a biological membrane.


   FOOTNOTES
 
* This work was supported by a Veterans Affairs Merit Review, National Institutes of Health Grant 5R01 DK062209-02, Juvenile Diabetes Foundation Grant 1-2003-108 (to M. L.), the Cell Migration Consortium Grant PHSSUBUVGC 10641, National Institutes of Health Public Health Service Grant 5 P41-RRO3155, and by the University of Illinois, Urbana-Champaign (to E. G.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

§ These authors contributed equally to this work. Back

{ddagger}{ddagger} To whom correspondence should be addressed: University of Colorado Health Sciences Center, 4200 East 9th Ave., Denver, CO 80262. Tel.: 303-315-1541; Fax: 303-315-1929; E-mail: Moshe.Levi{at}UCHSC.edu.

1 The abbreviations used are: BBM, brush border membrane; DR, detergent-resistant; DS, detergent-sensitive; FCS, fluorescence correlation spectroscopy; GUV, giant unilamellar vesicle; NaPi, sodium/phosphate cotransporter type IIa; PCH, photon counting histogram; 5'NT, 5'-nucleotidase; FRAP, fluorescence recovery after photobleaching. Back



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Tenenhouse, H. S., and Murer, H. (2003) J. Am. Soc. Nephrol. 14, 240-248[Free Full Text]
  2. Molitoris, B. A., Alfrey, A. C., Harris, R. A., and Simon, F. R. (1985) Am. J. Physiol. 249, F12-F19[Medline] [Order article via Infotrieve]
  3. Levi, M., Jameson, D. M., and van der Meer, B. W. (1989) Am. J. Physiol. 256, F85-F94[Medline] [Order article via Infotrieve]
  4. Levi, M., Shayman, J. A., Abe, A., Gross, S. K., McCluer, R. H., Biber, J., Murer, H., Lotscher, M., and Cronin, R. E. (1995) J. Clin. Investig. 96, 207-216[Medline] [Order article via Infotrieve]
  5. Zajicek, H. K., Wang, H., Puttaparthi, K., Halaihel, N., Markovich, D., Shayman, J., Beliveau, R., Wilson, P., Rogers, T., and Levi, M. (2001) Kidney Int. 60, 694-704[CrossRef][Medline] [Order article via Infotrieve]
  6. Levi, M., Baird, B. M., and Wilson, P. V. (1990) J. Clin. Investig. 85, 231-237[Medline] [Order article via Infotrieve]
  7. Galbiati, F., Razani, B., and Lisanti, M. P. (2001) Cell 106, 403-411[CrossRef][Medline] [Order article via Infotrieve]
  8. Brown, D. A., and Jacobson, K. (2001) Traffic 2, 668-672[CrossRef][Medline] [Order article via Infotrieve]
  9. Simons, K., and Ehehalt, R. (2002) J. Clin. Investig. 110, 597-603[CrossRef][Medline] [Order article via Infotrieve]
  10. Anderson, R. G., and Jacobson, K. (2002) Science 296, 1821-1825[Abstract/Free Full Text]
  11. Edidin, M. (2003) Annu. Rev. Biophys. Biomol. Struct. 32, 257-283[CrossRef][Medline] [Order article via Infotrieve]
  12. Pike, L. J. (2004) Biochem. J. 378, 281-292[CrossRef][Medline] [Order article via Infotrieve]
  13. Levi, M., Wilson, P. V., Cooper, O. J., and Gratton, E. (1993) Photochem. Photobiol. 57, 420-425[Medline] [Order article via Infotrieve]
  14. Parasassi, T., Gratton, E., Yu, W. M., Wilson, P., and Levi, M. (1997) Biophys. J. 72, 2413-2429[Medline] [Order article via Infotrieve]
  15. Dietrich, C., Bagatolli, L. A., Volovyk, Z. N., Thompson, N. L., Levi, M., Jacobson, K., and Gratton, E. (2001) Biophys. J. 80, 1417-1428[Medline] [Order article via Infotrieve]
  16. Dietrich, C., Volovyk, Z. N., Levi, M., Thompson, N. L., and Jacobson, K. (2001) Proc. Natl. Acad. Sci. U. S. A. 98, 10642-10647[Abstract/Free Full Text]
  17. Sorribas, V., Halaihel, N., Puttaparthi, K., Rogers, T., Cronin, R. E., Alcalde, A. I., Aramayona, J., Sarasa, M., Wang, H., Wilson, P., Zajicek, H., and Levi, M. (2001) Kidney Int. 59, 1024-1036[CrossRef][Medline] [Order article via Infotrieve]
  18. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265-275[Free Full Text]
  19. Roper, K., Corbeil, D., and Huttner, W. B. (2000) Nat. Cell Biol. 2, 582-592[CrossRef][Medline] [Order article via Infotrieve]
  20. Liang, X., Nazarian, A., Erdjument-Bromage, H., Bornmann, W., Tempst, P., and Resh, M. D. (2001) J. Biol. Chem. 276, 30987-30994[Abstract/Free Full Text]
  21. Lindwasser, O. W., and Resh, M. D. (2001) J. Virol. 75, 7913-7924[Abstract/Free Full Text]
  22. Laemmli, U. K. (1970) Nature 227, 680-685[CrossRef][Medline] [Order article via Infotrieve]
  23. Towbin, H., Staehelin, T., and Gordon, J. (1979) Proc. Natl. Acad. Sci. U. S. A. 76, 4350-4354[Abstract/Free Full Text]
  24. Bligh, E. G., and Dyer, W. J. (1959) Can. J. Med. Sci. 37, 911-917[Medline] [Order article via Infotrieve]
  25. He, X., Chen, F., McGovern, M. M., and Schuchman, E. H. (2002) Anal. Biochem. 306, 115-123[CrossRef][Medline] [Order article via Infotrieve]
  26. Magnani, F., Tate, C., Wynne, S., Williams, C., and Haase, J. (2004) J. Biol. Chem. 279, 38770-38778[Abstract/Free Full Text]
  27. Angelova, M., and Dimitrov, D. (1986) Faraday Discuss. Chem. Soc. 81, 303-311[CrossRef]
  28. Bagatolli, L. A., and Gratton, E. (1999) Biophys. J. 77, 2090-2101[Medline] [Order article via Infotrieve]
  29. Bagatolli, L. A., and Gratton, E. (2000) Biophys. J. 78, 290-305[Medline] [Order article via Infotrieve]
  30. Ruan, Q., Cheng, M., Levi, M., Gratton, E., Mantulin, W. (2004) Biophys. J. 87, 1260-1267[CrossRef][Medline] [Order article via Infotrieve]
  31. Ruan, Q., Chen, Y., Gratton, E., Glaser, M., and Mantulin, W. W. (2002) Biophys. J. 83, 3177-3187[Medline] [Order article via Infotrieve]
  32. Berland, K. M., So, P. T., and Gratton, E. (1995) Biophys. J. 68, 694-701[Medline] [Order article via Infotrieve]
  33. Chen, Y., Muller, J. D., So, P. T., and Gratton, E. (1999) Biophys. J. 77, 553-567[Medline] [Order article via Infotrieve]
  34. Chen, Y., Muller, J. D., Berland, K. M., and Gratton, E. (1999) Methods (Duluth) 19, 234-252
  35. Muller, J. D., Chen, Y., and Gratton, E. (2000) Biophys. J. 78, 474-486[Medline] [Order article via Infotrieve]
  36. Chen, Y., Muller, J. D., Ruan, Q., and Gratton, E. (2002) Biophys. J. 82, 133-144[Medline] [Order article via Infotrieve]
  37. Elkjaer, M. L., Kwon, T. H., Wang, W., Nielsen, J., Knepper, M. A., Frokiaer, J., and Nielsen, S. (2002) Am. J. Physiol. 283, F1376-F1388
  38. Hernando, N., Deliot, N., Gisler, S. M., Lederer, E., Weinman, E. J., Biber, J., and Murer, H. (2002) Proc. Natl. Acad. Sci. U. S. A. 99, 11957-11962[Abstract/Free Full Text]
  39. Shenolikar, S., Voltz, J. W., Minkoff, C. M., Wade, J. B., and Weinman, E. J. (2002) Proc. Natl. Acad. Sci. U. S. A. 99, 11470-11475[Abstract/Free Full Text]
  40. Gisler, S. M., Pribanic, S., Bacic, D., Forrer, P., Gantenbein, A., Sabourin, L. A., Tsuji, A., Zhao, Z. S., Manser, E., Biber, J., and Murer, H. (2003) Kidney Int. 64, 1733-1745[CrossRef][Medline] [Order article via Infotrieve]
  41. Kocher, O., Pal, R., Roberts, M., Cirovic, C., and Gilchrist, A. (2003) Mol. Cell. Biol. 23, 1175-1180[Abstract/Free Full Text]
  42. Pribanic, S., Gisler, S. M., Bacic, D., Madjdpour, C., Hernando, N., Sorribas, V., Gantenbein, A., Biber, J., and Murer, H. (2003) Am. J. Physiol. 285, F784-F791
  43. Blasco, T., Aramayona, J. J., Alcalde, A. I., Catalan, J., Sarasa, M., and Sorribas, V. (2003) Am. J. Physiol. 285, F799-F810
  44. Huber, T. B., Schmidts, M., Gerke, P., Schermer, B., Zahn, A., Hartleben, B., Sellin, L., Walz, G., and Benzing, T. (2003) J. Biol. Chem. 278, 13417-13421[Abstract/Free Full Text]
  45. Smith, W. J., Nassar, N., Bretscher, A., Cerione, R. A., and Karplus, P. A. (2003) J. Biol. Chem. 278, 4949-4956[Abstract/Free Full Text]
  46. Salzer, U., and Prohaska, R. (2001) Blood 97, 1141-1143[Abstract/Free Full Text]
  47. Kokubo, H., Helms, J. B., Ohno-Iwashita, Y., Shimada, Y., Horikoshi, Y., and Yamaguchi, H. (2003) Brain Res. 965, 83-90[CrossRef][Medline] [Order article via Infotrieve]
  48. Lipardi, C., Nitsch, L., and Zurzolo, C. (2000) Mol. Biol. Cell 11, 531-542[Abstract/Free Full Text]
  49. Draeger, A., Monastyrskaya, K., Burkhard, F. C., Wobus, A. M., Moss, S. E., and Babiychuk, E. B. (2003) Dev. Biol. 262, 324-334[CrossRef][Medline] [Order article via Infotrieve]
  50. Ledoux, S., Runembert, I., Koumanov, K., Michel, J. B., Trugnan, G., and Friedlander, G. (2003) Circ. Res. 92, 848-855[Abstract/Free Full Text]
  51. Oh, P., and Schnitzer, J. E. (2001) Mol. Biol. Cell 12, 685-698[Abstract/Free Full Text]
  52. Haupts, U., Maiti, S., Schwille, P., and Webb, W. W. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 13573-13578[Abstract/Free Full Text]
  53. Kahya, N., Wiersma, D. A., Poolman, B., and Hoekstra, D. (2002) J. Biol. Chem. 277, 39304-39311[Abstract/Free Full Text]
  54. Kahya, N., Scherfeld, D., Bacia, K., Poolman, B., and Schwille, P. (2003) J. Biol. Chem. 278, 28109-28115[Abstract/Free Full Text]
  55. Varma, R., and Mayor, S. (1998) Nature 394, 798-801[CrossRef][Medline] [Order article via Infotrieve]
  56. Friedrichson, T., and Kurzchalia, T. (1998) Nature 394, 802-805[CrossRef][Medline] [Order article via Infotrieve]
  57. Sharma, P., Varma, R., Sarasij, R. C., Ira, Gousset, K., Krishnamoorthy, G., Rao, M., and Mayor, S. (2004) Cell 116, 577-589[CrossRef][Medline] [Order article via Infotrieve]
  58. Nichols, B. J., Kenworthy, A. K., Polishchuk, R. S., Lodge, R., Roberts, T. H., Hirschberg, K., Phair, R. D., and Lippincott-Schwartz, J. (2001) J. Cell Biol. 153, 529-541[Abstract/Free Full Text]
  59. Lippincott-Schwartz, J., Altan-Bonnet, N., and Patterson, G. H. (2003) Nat. Cell Biol. 5 (suppl.), 7-14[CrossRef][Medline] [Order article via Infotrieve]
  60. Thomsen, P., Roepstorff, K., Stahlhut, M., and van Deurs, B. (2002) Mol. Biol. Cell 13, 238-250[Abstract/Free Full Text]
  61. Umenishi, F., Verbavatz, J. M., and Verkman, A. S. (2000) Biophys. J. 78, 1024-1035[Medline] [Order article via Infotrieve]
  62. Pralle, A., Keller, P., Florin, E. L., Simons, K., and Horber, J. K. (2000) J. Cell Biol. 148, 997-1008[Abstract/Free Full Text]
  63. Dietrich, C., Yang, B., Fujiwara, T., Kusumi, A., and Jacobson, K. (2002) Biophys. J. 82, 274-284[Medline] [Order article via Infotrieve]
  64. Kenworthy, A. K., Nichols, B. J., Remmert, C. L., Hendrix, G. M., Kumar, M., Zimmerberg, J., and Lippincott-Schwartz, J. (2004) J. Cell Biol. 165, 735-746[Abstract/Free Full Text]
  65. Li, X., Galli, T., Leu, S., Wade, J. B., Weinman, E. J., Leung, G., Cheong, A., Louvard, D., and Donowitz, M. (2001) J. Physiol. (Lond.) 537, 537-552[Abstract/Free Full Text]
  66. Li, X., Leu, S., Cheong, A., Zhang, H., Baibakov, B., Shih, C., Birnbaum, M. J., and Donowitz, M. (2004) Gastroenterology 126, 122-135[CrossRef][Medline] [Order article via Infotrieve]
  67. Akhter, S., Kovbasnjuk, O., Li, X., Cavet, M., Noel, J., Arpin, M., Hubbard, A. L., and Donowitz, M. (2002) Am. J. Physiol. 283, C927-C940
  68. Runembert, I., Queffeulou, G., Federici, P., Vrtovsnik, F., Colucci-Guyon, E., Babinet, C., Briand, P., Trugnan, G., Friedlander, G., and Terzi, F. (2002) J. Cell Sci. 115, 713-724[Abstract/Free Full Text]
  69. Kwik, J., Boyle, S., Fooksman, D., Margolis, L., Sheetz, M. P., and Edidin, M. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 13964-13969[Abstract/Free Full Text]
  70. Khan, T. K., Yang, B., Thompson, N. L., Maekawa, S., Epand, R. M., and Jacobson, K. (2003) Biochemistry 42, 4780-4786[CrossRef][Medline] [Order article via Infotrieve]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 Am. J. Physiol. Cell Physiol.Home page
A. D. M. Riquier, D. H. Lee, and A. A. McDonough
Renal NHE3 and NaPi2 partition into distinct membrane domains
Am J Physiol Cell Physiol, April 1, 2009; 296(4): C900 - C910.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Renal Physiol.Home page
D. H. Lee, A. D. M. Riquier, L. E. Yang, P. K. K. Leong, A. B. Maunsbach, and A. A. McDonough
Acute hypertension provokes acute trafficking of distal tubule Na-Cl cotransporter (NCC) to subapical cytoplasmic vesicles
Am J Physiol Renal Physiol, April 1, 2009; 296(4): F810 - F818.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Renal Physiol.Home page
C. Kastner, M. Pohl, M. Sendeski, G. Stange, C. A. Wagner, B. Jensen, A. Patzak, S. Bachmann, and F. Theilig
Effects of receptor-mediated endocytosis and tubular protein composition on volume retention in experimental glomerulonephritis
Am J Physiol Renal Physiol, April 1, 2009; 296(4): F902 - F911.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Renal Physiol.Home page
P. Welker, A. Bohlick, K. Mutig, M. Salanova, T. Kahl, H. Schluter, D. Blottner, J. Ponce-Coria, G. Gamba, and S. Bachmann
Renal Na+-K+-Cl- cotransporter activity and vasopressin-induced trafficking are lipid raft-dependent
Am J Physiol Renal Physiol, September 1, 2008; 295(3): F789 - F802.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
R. Murtazina, O. Kovbasnjuk, M. Donowitz, and X. Li
Na+/H+ Exchanger NHE3 Activity and Trafficking Are Lipid Raft-dependent
J. Biol. Chem., June 30, 2006; 281(26): 17845 - 17855.
[Abstract] [Full Text] [PDF]

Home page
 J. Am. Soc. Nephrol.Home page
L. E. Yang, A. B. Maunsbach, P. K.K. Leong, and A. A. McDonough
Redistribution of Myosin VI from Top to Base of Proximal Tubule Microvilli during Acute Hypertension
J. Am. Soc. Nephrol., October 1, 2005; 16(10): 2890 - 2896.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Renal Physiol.Home page
S. Y. Breusegem, N. Halaihel, M. Inoue, H. Zajicek, E. Lederer, N. P. Barry, V. Sorribas, and M. Levi
Acute and chronic changes in cholesterol modulate Na-Pi cotransport activity in OK cells
Am J Physiol Renal Physiol, July 1, 2005; 289(1): F154 - F165.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Gastrointest. Liver Physiol.Home page
A. Stauber, T. Radanovic, G. Stange, H. Murer, C. A. Wagner, and J. Biber
Regulation of Intestinal Phosphate Transport II. Metabolic acidosis stimulates Na+-dependent phosphate absorption and expression of the Na+-Pi cotransporter NaPi-IIb in small intestine
Am J Physiol Gastrointest Liver Physiol, March 1, 2005; 288(3): G501 - G506.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/47/49160    most recent
M408942200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Inoue, M.
Right arrow Articles by Levi, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Inoue, M.
Right arrow Articles by Levi, M.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2004 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement