AJP - Renal Watch the video to learn how APS reaches out to developing nations.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Renal Physiol 274: F129-F138, 1998;
0363-6127/98 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
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
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Doctor, R. B.
Right arrow Articles by Mandel, L. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Doctor, R. B.
Right arrow Articles by Mandel, L. J.
Vol. 274, Issue 1, F129-F138, January 1998

Distribution of epithelial ankyrin (Ank3) spliceoforms in renal proximal and distal tubules

R. Brian Doctor1, Jing Chen1, Luanne L. Peters2, Samuel E. Lux3, and Lazaro J. Mandel1

1 Department of Cell Biology, Duke University Medical Center, Durham, North Carolina 27710; 2 Jackson Laboratory, Bar Harbor, Maine 04609; and 3 Department of Hematology/Oncology, Children's Hospital/Harvard Medical School, Boston, Massachusetts 02115

    ABSTRACT
Top
Abstract
Introduction
Methods
Results
Discussion
References

In diverse cell types, ankyrin tethers a variety of ion transport and cell adhesion molecules to the spectrin-based membrane skeleton. In the whole kidney, epithelial ankyrin (Ank3) is the predominantly expressed ankyrin and is expressed as distinct spliceoforms. Antibodies against a portion of the Ank3 regulatory domain detected four major spliceoforms at 215, 200, 170, and 120 kDa. Immunoblotting of the renal cortex, which is 80% proximal tubule (PT), detected all four spliceoforms but showed significantly diminished Ank3200/215. To determine the Ank3 spliceoforms present in the mouse PT cells, PT fragments were purified to 100% from the renal cortex. Isolation was performed by incubating cortical tubule segments with fluorescein and isolating the fluorescein-laden PT fragments or fluorescein-deplete non-PT (distal) fragments under fluorescence microscopy. Distal tubule (DT) fragments displayed abundance of the Ank3200/215 but no Ank3170 or Ank3120. Isolated PT segments contained all four spliceoforms but dramatically diminished Ank3200/215. These larger spliceoforms bind Na-K-ATPase in diverse cell types. Densitometric analysis of Ank3200/215 and Na-K-ATPase abundance measured a lower Ank3200/215-to-Na-K-ATPase ratio in the PT vs. the renal cortex. These proximal vs. distal differences in Ank3 spliceoforms were displayed in LLC-PK1 cells, a proximal cell line, and MDCK cells, a distal cell line. The lower PT content of Ank3200/215 suggests Na-K-ATPase in PT may be organized differently than in DT. Likely reflecting their cell-specific organization, regulation, and function, these studies indicate the different renal cell types express distinct Ank3 spliceoforms.

kidney; sodium-potassium-adenosinetriphosphatase

    INTRODUCTION
Top
Abstract
Introduction
Methods
Results
Discussion
References

MEMBERS OF THE ANKYRIN family of proteins are multipotential adaptors that link integral membrane proteins to the underlying membrane skeleton. Three ankyrin genes have been cloned and sequenced: erythrocyte ankyrin (Ank1 or AnkR; see Refs. 18, 19), brain ankyrin (Ank2 or AnkB), and epithelial or general ankyrin (Ank3 or AnkG; see Refs. 16, 28). These ankyrins share a common domain structure: an NH2-terminal membrane-binding "repeat" domain composed primarily of 24 evolutionarily conserved repeats of 33 amino acids, a central spectrin-binding domain, and a COOH-terminal "regulatory" domain, so named to reflect its ability to modulate the binding affinities of the other domains. Alternative mRNA splicing within the three domains of all three ankyrin genes occurs, which profoundly increases the diversity of protein isoforms.

Ank3 is the most widely distributed ankyrin, with expression observed in a variety of epithelia, liver, muscle, testes, and macrophages (12, 16, 28). In the kidney there is no Ank1 expression, little if any Ank2 expression, but abundant Ank3 expression (27, 28). Furthermore, Ank3 expression exists in multiple Ank3 spliceoforms including proteins of 215, 200, 170, 120, 119, and 105 kDa (4, 28).

The best-described role of ankyrin in the kidney is the retention of Na-K-ATPase along the basolateral membrane of distal tubular (DT) segments. In the DT segments, ankyrin colocalizes with Na-K-ATPase along the tortuous infoldings of the basolateral membrane of thick ascending limb (TAL) cells (6, 15). In Madin-Darby canine kidney (MDCK) cells, a renal DT cell line, ankyrin colocalizes with Na-K-ATPase along the basolateral membrane, coprecipitates with Na-K-ATPase, and increases the retention half-life of Na-K-ATPase in the membrane (11). All of these observations were performed using immunoreagents directed against Ank1 which detects a 190/210-kDa protein. Ank1 antibody reactivity with proximal tubule (PT) cells has been relatively poor, leaving the presence and function of ankyrins within the PT cells largely unstudied (2, 5).

In the present study, antibodies directed against an Ank3-specific region of the regulatory domain, termed Ank3-R1, detected the 215-, 200-, 170-, and 120-kDa spliceoforms. Immunocytochemistry with this antibody specifically stained all tubule segments including the PT cells (28, 29). The distribution of the individual spliceoforms within the different nephron segments is unknown. The present studies demonstrate that DT segments express the known Na-K-ATPase-binding Ank3200/215 spliceoforms. These segments showed no other spliceoforms as detected with Ank3-R1. Conversely, PT segments contained predominantly the Ank3120 and Ank3170 spliceoforms with comparatively little Ank3200/215. As a result, the PT cells had lower ratios of Ank3200/215 to Na-K-ATPase than the whole cortex. The cell type-specific spliceoform expression and lower Ank3200/215-to-Na-K-ATPase ratio were also retained in LLC-PK1 and MDCK cells, cultures of proximal and distal origin, respectively.

    METHODS
Top
Abstract
Introduction
Methods
Results
Discussion
References

The kidney is a highly organized organ with the transitions of the nephron segments defining the different regions of the kidney (Fig. 1). The dissection of the regions of the kidney and isolation of tubule segments from within those regions was used to determine the presence, absence, and relative abundance of epithelial ankyrin spliceoforms. By volume, PT occupy ~80% of the renal cortex. In contrast, PT constitute only a minor portion of the outer medulla and are absent from the inner medulla. To investigate the presence of Ank3 spliceoforms in PT cells, PT were sequentially purified from the kidney by removing the renal cortex, separating cortical tubule segments on a Percoll density gradient, and collecting individual fluorescein-laden PT.


View larger version (34K):
[in this window]
[in a new window]
  		Fig. 1.   Nephron segments of the mouse kidney. Renal epithelia of the nephron and collecting system are highly organized and define the regions of the kidney. These regions include the cortex, outer medulla, which is subdivided into the outer stripe (OS) and inner stripe (IS), and the inner medulla. Proximal tubule (PT) and distal tubule (DT) segments were isolated from the renal cortex (see Figs. 2-4, and 6). Immunofluorescence micrographs (see Fig. 6) are a cross-section of a cortical ray, which contain PT, thick ascending limbs (TAL), and collecting ducts (CD). G, glomerulus; TL, descending and ascending thin limbs; DCT, distal convoluted tubule; CT, connecting tubule. This is a modification of a previously published figure (17).

Whole Kidney and Kidney Cortex Preparation

Whole kidney and kidney cortex samples were solubilized for electrophoresis and immunoblotting. Mice and rats were anesthetized and anticoagulated with an intraperitoneal injection of heparin (4,000 U) and pentobarbital (60 mg/kg body wt), and the kidneys were perfused free of blood. Rats were perfused via retrograde descending aortic perfusion. Mice were perfused via left ventricular/ascending aortic perfusion. The kidneys were removed, frozen in liquid nitrogen, and finely ground under liquid nitrogen. A sample was added to warmed 5× polyacrylamide gel electrophoresis (PAGE) buffer [90°C; 5% sodium dodecyl sulfate, 25% sucrose, 50 mM tris(hydroxymethyl)aminomethane, 5 mM EDTA, 200 µM phenylmethylsulfonyl fluoride (PMSF), 10 µM leupeptin, 2.0 µg/ml aprotinin, and 10 µM pepstatin; pH 8.0]. The protein content was measured by the bicinchoninic acid assay (Pierce Biochemical). Aliquots were reduced with addition of 5% beta -mercaptoethanol.

Mouse Cortical Tubule Suspension

Cortical and PT segments were isolated from the cortex of mouse kidneys (CD57; Charles River) by a method similar to that for rabbit PT and adapted for the mouse by Dr. Kathleen Dickman (VAMC-Northport, Northport, NY; personal communication). Briefly, four male mice were injected with 4,000 U of heparin an euthanized, and the kidneys were removed into ice-cold Dulbecco's modified Eagle's medium-Ham's F-12 medium (DME-F12). The kidneys were sliced into ~4-mm slabs, and the cortex was trimmed and minced. The minced tissue was suspended into 10 ml of collagenase solution [50 ml DME-F12, 50 mg collagenase (Worthington Biochemicals), 5 mg deoxyribonuclease, and 50 mg bovine serum albumin (BSA)] and incubated at 37°C with rapid shaking. After 10 min, the suspension was drawn into a pipette 5-10 times and the suspended tubules are rescued from the chunks. The chunks were resuspended in collagenase solution, and the process was repeated two to four times. The total PT suspension was washed at 150 g for 2 min and then again at 100 g for 2 min.

PT segments are the densest tubular segments. To further concentrate the PT segments, the tubules were resuspended in 25 ml of 45% Percoll solution [22.5 ml Percoll, 2.25 ml 10× Dulbecco's phosphate-buffered saline (PBS), 38.4 mg N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES), 25 ml DME-F12]. The viable PT (~95% PT) were separated at 25,000 g for 30 min and washed twice at 150 g for 2 min. To minimize the potential for DT contamination, only tissue migrating >= 90% the length of the Percoll gradient was retained. For DT isolation, the tubules were resuspended in 25 ml of 35% Percoll solution (22.5 ml Percoll, 2.25 ml 10× Dulbecco's PBS, 38.4 mg HEPES, 25 ml DME-F12). The viable DT were separated at 25,000 g for 30 min and washed twice at 150 g for 2 min. To increase the distal content without including nonviable cells, all segments migrating >50% the length of the Percoll gradient were retained. The suspensions were maintained in ice-cold DME-F12 until needed.

Fluorescein Loading and Isolation of PT

In contrast to all other cell types within the kidney, the PT possesses an avid capacity to transport and accumulate organic acids, including fluorescein (31). This concentrating capacity was exploited to load PT fragments with fluorescein and isolate fluorescein-laden PT or fluorescein-deplete DT under fluorescence microscopy. Percoll-purified PT fragments were resuspended in DME-F12 containing 2 mM fluorescein and incubated for 25 min at room temperature, and ~1 ml was transferred into ice-cold test tubes to allow the tubules to settle out. The supernatant was then aspirated from a single test tube, and the segments were resuspended in DME with 1% goat serum. Tubules were placed in a petri plate at low density and, under fluorescence microscopy (×100 magnification), aspirated into a hematocrit tube and transferred into a microcentrifuge tube containing DME-goat serum with 4 mM ethylene glycol-bis(beta -aminoethyl ether)-N,N,N',N'-tetraacetic acid, and 2× protease inhibitor cocktail (20 µM leupeptin, 4 µg/ml aprotinin, 400 µM PMSF, and 20 µM pepstatin A). Following the collection of the tubule segments, the tubules were pelleted at 2,000 g and solubilized in 5× PAGE buffer. The inclusion of serum or BSA was required to inhibit isolated tubules from sticking to the plates and tubes. This inclusion disallowed the accurate measurement of total isolated protein.

Cell Culture Maintenance and Preparation

MDCK and LLC-PK1 (clone 4) were passaged and maintained as previously described (9, 20). Cell monolayers on semipermeable supports were grown for five days following the establishment of confluence. The cells were solubilized by the addition of 1.0 ml of warmed 5× PAGE and reduced as described above.

Immunoblotting

Gel electrophoresis was performed as previously described (5, 8). Quantities of 20-60 µg protein of the 5× PAGE dissolved tissues were run on a 3.5%-14% polyacrylamide gradient gel (200 mV, 3.5 h). The gels were then transferred onto nitrocellulose (300 mV; 3.25 h).

Western blotting was performed on the transblots as previously described (32). The blots were washed once in blot buffer (150 mM NaCl, 10 mM NaH2PO4, 1 mM EDTA, 0.2% Triton X-100, and 1 mM NaN3; pH 7.4), once in blot buffer with 4% BSA for 20 min or 0.5% dry milk for 60 min, once with blot buffer, and then incubated with the primary antibody for 60 min at room temperature. The blot was then washed four times (4 min/wash) and incubated with a peroxidase-labeled anti-immunoglobulin G (anti-IgG) secondary antibody (1:25,000 dilution; Jackson Immunoresearch Laboratories, West Grove, PA). The reaction products were detected using enhanced chemiluminescence detection (ECL, Amersham).

The Ank3-R1 antibody is specific for the regulatory domain of mouse Ank3. It is used to assess the relative abundance of Ank3 spliceoforms in specific cell types of the mouse kidney. The differences in immunoblot band intensities should represent differences in abundance and not differences in antibody affinities, because the proteins are denatured and directed against antigenic sights generated from the identical gene sequence. Since the sequence of the sites in the different species has not been identified, intensity differences between species are only considered for their qualitative value.

Immunofluorescence

Rat and mouse kidneys were fixed, stained, and examined by immunofluorescence as previously described (5). The kidneys were flushed free of blood with ice-cold PBS and then perfused with 3% paraformaldehyde in PBS. The kidneys were excised and further fixed for 60 min. The cortex was then cut into pyramidal blocks, incrementally infused with 5% (60 min), 10% (60 min), and 25% (overnight) sucrose, and then plunged into liquid nitrogen-cooled isopentane. The blocks were then sectioned (5 µm) and stained. Staining was done by blocking nonspecific interactions (1% BSA and 10% normal goat serum), reducing the block (0.1% BSA and 1% normal goat serum), incubating with the appropriate antibody, washing, incubating with a rhodamine-labeled goat anti-rabbit IgG, washing, postfixing, and mounting in 60% glycerol with 10% diazobicyclooctane. Nonimmune serum replaced the primary antibodies in control sections. The sections were visualized on a Zeiss LSM confocal microscope.

Antibodies and Reagents

Polyclonal antibodies directed against a segment of the regulatory domain of Ank3 (epithelial ankyrin; Ank3-R1) were developed and characterized as previously described (28). Dilutions for immunoblotting were 1:500; dilutions for immunohistochemistry were 1:250. The polyclonal antibody against Ank1 (erythrocyte ankyrin) was graciously provided by Dr. Vann Bennett (Duke University). Dilution for immunohistochemistry was 1:1,000. The polyclonal antibody against the H5 cytoplasmic loop of Na-K-ATPase was graciously provided by Dr. W. James Nelson (Stanford University). Dilutions for immunoblotting were 1:500; dilutions for immunohistochemistry were 1:250. The polyclonal antibodies against aquaporin-2 (AQP2) and bumetanide-sensitive Na-K-2Cl transporter were graciously provided by Dr. Mark Knepper (National Institutes for Health). Dilutions for immunoblotting were 1:3,000 and 1:2,000, respectively. Peroxidase-conjugated and rhodamine-conjugated anti-rabbit IgG secondary antibodies were obtained from Jackson Immunoresearch Laboratories. Dilutions for immunoblotting were 1:25,000; dilutions for immunohistochemistry were 1:50.

    RESULTS
Top
Abstract
Introduction
Methods
Results
Discussion
References

As previously described (28), the mouse kidney contains a number of epithelial ankyrin spliceoforms including a prominent band at 120 kDa and lower intensity bands at 170, 200, and 215 kDa. The 105-kDa band observed in whole kidney crude membranes and macrophages of other mouse strains was only faintly detected in the present study (12, 28). Immunoblots of the mouse renal cortex showed the presence of the same spliceoforms but the intensities of the 200- and 215-kDa band were relatively lower in the cortex vs. the whole kidney. In paired mouse kidney and cortical samples, densitometric analysis showed the 200 kDa-to-120 kDa ratio was significantly higher (P < 0.05, paired t-test) in the whole kidney (0.38 ± 0.11, n = 4) than in the renal cortex (0.18 ± 0.05, n = 4). The lower cortical content of the 200/215-kDa spliceoforms in a region of the kidney containing predominantly PT segments suggests that the diminished Ank3200/215 content reflects a PT property. To ascertain the segment-specific distribution of the Ank3 spliceoforms directly, PT and DT segments were individually isolated.

Ank3 Spliceoforms in Isolated PT and DT

PT and DT were isolated by incubating a cortical tubule suspension with fluorescein and separating either PT (fluorescein laden) or DT (fluorescein deplete) (Fig. 2A). The method does not discriminate between the different DT segments but is effective at rapidly isolating pools of PT and non-PT segments. Preliminary studies demonstrated that the PT isolation, fluorescein incubation, 60-min cold storage, and nonspecific aspiration did not result in a discernible loss of any of the Ank3 spliceoforms (data not shown). To demonstrate the capacity to isolate pure preparations of PT, the content of aquaporin-2 (AQP2) and bumetanide-sensitive Na-K-2Cl transporter (BSC-1) proteins was examined in cortical and isolated PT samples. In the kidney, AQP2 is expressed only in the collecting duct system (26). BSC-1 is expressed exclusively in the TAL (7). Figure 2B shows comparative amounts of cortical and isolated PT tissue, as approximated by Na-K-ATPase levels. Although the renal cortex showed specific staining for AQP2, the isolated PT had no detectable AQP2 reactivity. Similarly, although the signal from mouse cortical tissue was weak, no BSC-1 reactivity was detectable in isolated PT samples (data not shown).


View larger version (45K):
[in this window]
[in a new window]
 

View larger version (40K):
[in this window]
[in a new window]
  		Fig. 2.   PT concentrate fluorescein. A: PT (bright tubule segments) concentrate fluorescein, whereas other segments (arrows) do not concentrate fluorescein. Different distal segments are shown in A and B, which, judging from their size relative to the PT, would include distal convoluted tubules or collecting duct (left) and TAL (right, bottom arrow). Difference in fluorescein concentration allowed either proximal or nonproximal segments to be isolated under fluorescence microscopy. B: capacity to specifically isolate PT segments from a cortical tubule suspension was tested by probing isolated PT (Isol PT) samples for aquaporin-2 (AQP-2; collecting duct tubules) and Na-K-2Cl cotransporter (BSC-1; TAL). Cortical and PT samples with similar Na-K-ATPase levels showed AQP-2 was completely absent from isolated PT. BSC-1 was only faintly present in the cortical sample but also failed to show reactivity with the isolated PT (data not shown).

As described above, the Ank3200 band intensity in renal cortical samples was 18% ± 5% of the Ank3120 band intensity. Isolated mouse PT showed the prominent Ank3120 and Ank3170 bands (Fig. 3A). However, the Ank3200/215 bands were markedly reduced, only seen on blot overexposure, and below the linear range of densitometric analysis. In contrast to the PT, isolated mouse DT displayed prominent Ank3 bands at 200 and 215 kDa and did not display either the Ank3120 or the Ank170 spliceoforms (Fig. 3B).


View larger version (88K):
[in this window]
[in a new window]
 

View larger version (61K):
[in this window]
[in a new window]
  		Fig. 3.   Ank3 spliceoforms are differentially expressed in proximal and distal nephrons. A: as shown above, the mouse renal cortex had varied levels of the Ank3 spliceoforms with the 200/215-kDa proteins being the least expressed of the forms. After the PT were isolated to 100%, the 200/215-kDa spliceoforms were further diminished. Although the Ank3120 and Ank3170 spliceoforms in the isolated PT (Isol PT) sample were equivalent to between 30 and 60 µg of cortical protein, the Ank3200/215 spliceoforms in Isol PT were in less abundance than the level measured in 6 µg of cortical protein. Values on left are in kDa. B: overexposure of isolated PT samples indicated that there was a low level of Ank3200/215 expressed in PT cells. In contrast to PT, isolated distal segments showed the opposite expression pattern. Isolated DT displayed the expression of Ank3200/215 but failed to detect proteins at 120 or 170 kDa. Values on left are in kDa.

Ankyrin Immunofluorescence

To corroborate the relative Ank3200/215 abundance in the distal segments and paucity in proximal segments, mouse cortical sections were fixed and stained for the Ank3200/215 spliceoforms in the mouse kidney. For ease of comparison, sections of cortical rays containing straight PT and TAL (Fig. 1) are shown. Although the Ank3-R1 antibody detects several spliceoforms, the Ank1 antibody cross-reacted with only the Ank3200/215 spliceoforms and failed to detect the Ank3120 and Ank3170 spliceoforms (Fig. 4A). The immunolocalization of the Ank3 spliceoforms, as detected by the Ank3-R1 antibody, has been previously described in detail (28, 29). Ank3-R1 sections, which display the summary localization of the four Ank3 spliceoforms, were included here for reference and comparison (Fig. 5A).


View larger version (34K):
[in this window]
[in a new window]
 

View larger version (43K):
[in this window]
[in a new window]
  		Fig. 4.   Ank1 antibody detects only Ank3200/215 in the kidney. A: the Ank1 polyclonal antibody used in these studies is specific for the Ank3200/215 spliceoforms in the mouse kidney. Mouse kidney samples showed the four prominent Ank3 spliceoforms with Ank3-R1 antibody (left) but only bands at 200 and 215 kDa when probed with Ank1 antibody (right). The Ank1 antibody is used to specifically detect the localization of Ank3200/215 in renal sections (see Fig. 5). B: to observe the comparison of the Ank3200/215 spliceoforms, an isolated PT sample was overloaded onto a gel containing a serial dilution of mouse renal cortex and double stained with Ank3-R1 and Na-K-ATPase (NKA) antibodies. Ank3200/215 content in the loaded PT sample was equivalent to between 27 and 54 µg of cortical tissue. Na-K-ATPase content, however, was greater in the isolated PT sample than in 54 µg of cortical sample. This indicated there were fewer Ank3200/215 moles per mole of Na-K-ATPase in PT cells than in the cortex as a whole.


View larger version (179K):
[in this window]
[in a new window]
 

View larger version (159K):
[in this window]
[in a new window]
 

View larger version (170K):
[in this window]
[in a new window]
  		Fig. 5.   Ank3200/215 is localized to the basolateral membrane and terminal web in PT cells. As previously described in detail, Ank3-R1 (A) reacts with all the renal nephron and collecting duct segments in the mouse renal cortex (28, 29). Na-K-ATPase staining intensity of TAL cells and PT cells correlated with the difference in Na-K-ATPase activity (B). In PT, Na-K-ATPase staining is localized to the basolateral membrane. Detection of the Ank3200/215 spliceoforms with the Ank1 antibody confirm the preponderance of these spliceoforms within the TAL with minor staining of the PT cells (C). Interestingly, the Ank3200/215 spliceoforms in PT cells were observed not only along the basolateral membrane but also in the terminal web region.

Antibodies with spliceoform-specific reactivity would definitively demonstrate the segmental and intracellular distribution for the spliceoforms. The Ank1 antibody, which reacted specifically with only the Ank3200 and Ank3215 spliceoforms, was used in this manner. As previously described in rats (5), the Ank1 antibody strongly stained the TAL and collecting duct cells (Fig. 5B). In mouse PT cells, Ank1 antibody staining was faint but specific. The weakness of the Ank3200/215 signal in the PT cells corroborates the relative decrease of these spliceoforms observed in the immunoblots of the isolated PT. In contrast to the DT segments, the staining was not only along the basolateral membrane but also along the terminal web of the PT cells. Although fodrin also exists within the terminal web region (5, 30), the protein(s) being tethered by ankyrin in this region is unknown.

The Na-K-ATPase staining intensities of the different cortical segments correlate with the known Na-K-ATPase activity levels within these segments (Fig. 5C) (14). TAL cells, which have high Na-K-ATPase activities, stained intensely. Staining of collecting duct and PT cells, which have comparatively lower Na-K-ATPase activities, were comparatively less intense. In TAL cells, Na-K-ATPase colocalized with ankyrin along the basolateral membrane (15). In the mouse PT, Na-K-ATPase was observed along the basolateral membrane. For comparison, the TAL immunofluorescence intensities after Na-K-ATPase and Ank1 antibody staining were adjusted to similar levels. In these sections, Na-K-ATPase staining intensity of PT cells was clearly greater than Ank3200/215 staining intensity of PT cells. This suggested that was comparatively less Ank3200/215 per Na-K-ATPase in the PT vs. the TAL.

Relative Na-K-ATPase and Ank3200 Contents

To quantitatively assess this observation, the immunoblot densities of Na-K-ATPase and Ank3200 in paired isolated PT and renal cortex samples were measured (Fig. 4B). Figure 4B shows an immunoblot of serially diluted renal cortex and isolated PT samples dual-stained for Ank3 and Na-K-ATPase. Ank3200/215 bands of the concentrated PT sample were roughly equivalent to those from 54 µg of loaded cortical sample. From the same blot, Na-K-ATPase was much greater in the isolated PT sample vs. the 54 µg cortical sample. The ratio of Ank3200 to Na-K-ATPase in cortical and isolated PT samples was quantified by densitometry. In cortical samples, the paired sample Ank3200-to-Na-K-ATPase ratio from dual-stained immunoblots was 3.3 ± 1.2 (n = 3). In isolated PT cells, this ratio was 0.8 ± 0.2 (n = 3). These values were statistically significant (P < 0.05) as determined by a paired t-test. Isolated PT had significantly less Ank3200 per unit of Na-K-ATPase than was measured in the renal cortex.

Ank3 Spliceoforms in Proximal and Distal Cell Lines

To verify the observations made in the kidney and assess potential models to study the functions of renal ankyrins, the Ank3 spliceoforms were examined in LLC-PK1 cells, a proximal cell line, and in MDCK cells, a distal cell line. Much like the isolated mouse distal segments, MDCK cells displayed a prominent ankyrin. As previously observed with Ank1 antibodies, these ankyrins presented as a doublet around 200 kDa. Unlike the isolated distal segments, MDCK cells also displayed an ankyrin which migrated around 175 kDa. This ankyrin migrated slower than the 170-kDa ankyrin in the mouse kidney.

Similar to the mouse PT cells, LLC-PK1 cells only showed the presence of the Ank200 doublet after overexposure of the blot (data not shown). Previously, with use of Ank1 antibodies, LLC-PK1 cells were shown to contain an Ank200 doublet (5). The lower Ank3200-to-Na-K-ATPase level measured in isolated PT cells was even more apparent in the LLC-PK1 cell vs. MDCK cell cultures. In the bottom of Fig. 6A, Na-K-ATPase levels were essentially identical between LLC-PK1 and MDCK cells. Ank3200, however, was distinctly prominent in the MDCK sample but unobserved in LLC-PK1 sample.


View larger version (71K):
[in this window]
[in a new window]
  		Fig. 6.   Proximal-distal differences were observed in cultured cell lines. A: lower Ank3200-to-Na-K-ATPase ratio was even more stark in LLC-PK1 cells, a porcine PT cell line, vs. MDCK cells, a canine distal cell line. With roughly equivalent Na-K-ATPase levels (NKA), MDCK cells showed abundant protein at 210 kDa, whereas none was observable in LLC-PK1 cells. Ankyrin spliceoforms in LLC-PK1 cells were different than isolated mouse PT. Displaying a prominent band at 145 kDa, LLC-PK1 cells lacked the Ank3120 and Ank3170 spliceoforms. B: to investigate whether the lack of the Ank3120 and Ank3170 in LLC-PK1 cells was observed in other species, kidney samples for mouse, rat, and rabbits were compared. Similar to mice, rats did possess an abundance of the Ank3120 spliceoform and Ank3200/215 but had comparatively little Ank3170. Kidneys from rabbits, which are phylogenetically more distant from mice than rats, showed a strikingly different Ank3 spliceoform pattern than mice or rats. Like LLC-PK1 cells, rabbit kidneys expressed an Ank3145 protein. Like MDCK cells, the upper spliceoforms were a narrow doublet around 210 kDa.

Ank3 Spliceoforms Vary Between Species

Unlike the mouse PT cells, LLC-PK1 cells did not display an ankyrin at 120 and 170 kDa but, instead, had a prominent ankyrin at 145 kDa. To assess whether the difference in LLC-PK1 cells, derived from porcine kidney, was a product of speciation or arose after culture, Ank3 spliceoforms were compared in mouse, rat, and rabbit kidneys (Fig. 6B). Rat kidneys, phylogenetically close to the mouse, expressed the same Ank3 spliceoforms as the mouse, although the Ank3170 spliceoform was dramatically reduced. In contrast, rabbit kidneys, which are phylogenetically more distant from mice than rats, showed distinct differences. The rabbit kidneys had a higher-molecular-weight ankyrin spliceoform that migrated distinctly from the Ank3200/215 seen in mice and rat kidneys and similar to ankyrin observed in MDCK cells. Like LLC-PK1 cells, rabbit kidneys had a prominent ankyrin around 145 kDa and the complete absence of the 120- and 170-kDa ankyrin spliceoforms. The 145-kDa rabbit ankyrin spliceoform migrated to a distance similar to that of the LLC-PK1 band. Although sequence and functional analysis are ultimately required, these observations suggest that across species, the same cell types express distinct spliceoforms to serve similar functions.

    DISCUSSION
Top
Abstract
Introduction
Methods
Results
Discussion
References

Renal Ank3 Spliceoforms Showed Segmental Specificity

The present study demonstrates that Ank3 spliceoforms are heterogeneously expressed in the different renal epithelial cells. Immunoblotting of the renal cortex vs. the whole kidney with Ank3-R1 antibody provided the initial evidence for segmental specificity of Ank3. In the renal cortex, which is 80% PT by volume, the major isoforms are Ank3170 and Ank3120, with the Ank3200/215 spliceoforms significantly diminished compared with the whole kidney. This demonstrates that Ank3 isoforms are not homogeneously distributed throughout the kidney. Isolated PT segments confirmed this and demonstrated a greatly diminished concentration of the Ank3200/215 spliceoforms (Fig. 3). Furthermore, isolated DT segments expressed only the Ank3200/215 spliceoforms; Ank3120 or Ank3170 were absent.

These observations in freshly isolated tubule segments were mirrored in cultured cells. MDCK cells, a DT cell line, has been shown to possess a 210-kDa ankyrin (5, 25). These ankyrins tether Na-K-ATPase along the basolateral membrane (25). In the present study, the Ank3-R1 antibody also detected the 210-kDa ankyrin in MDCK cells and, like the mouse DT, failed to detect proteins at 120 or 170 kDa. LLC-PK1 cells, derived from PT, also express the 210-kDa ankyrin (5). Immunodetection with the Ank3-R1 antibody demonstrated, like the mouse PT, that expression of this spliceoform pales in comparison to other ankyrin spliceoforms in the PT (Figs. 3 and 6). Taken together, these studies clearly demonstrate that the epithelial ankyrin gene is differentially spliced in the different cell types of the kidney.

The origin of Ank3120 is well documented in macrophages (12, 28). It utilizes a start site within the last repeat of the membrane-binding domain (29 amino acids upstream of the spectrin-binding domain) and includes the entire spectrin-binding and regulatory domains. A 21.5-kDa, highly acidic, alternatively spliced exon is present within the regulatory domain of macrophage Ank3120. Presumably, the kidney Ank3120 isoform arises similarly but definitive proof by cloning and sequencing has not yet been obtained. Northern analysis confirms that kidney Ank3120 contains the spectrin-binding domain (28). Identifying the potential functions of Ank3120 in the modulation of spectrin transport, organization, or binding within PT cells awaits further inquiry.

The 170-kDa ankyrin is not well characterized, and its origins remain obscure (28). Given the enormous diversity of alternative splicing of the Ank3 gene described to date, the Ank3170 is likely to represent a true spliceoform. However, isolated kidney microsomes have a leupeptin-sensitive protease activity capable of clipping the 29-kDa COOH terminus from the regulatory domain of Ank1 (2). Thus the 170-kDa isoform of Ank3 could arise by partial proteolysis of a larger spliceoform. Further analysis will be required to elucidate the origin of Ank3170.

Functions of the Ank3 Spliceoforms in Renal Epithelial Cells

The myriad of segment-specific, polarized transport and adhesion molecules in the kidney indicate the need for specific mechanisms to develop, regulate, and maintain this polarization. The cell type-specific Ank3 spliceoform expression demonstrated above provides one mechanism by which the renal epithelial cells might direct cell type-specific polarization of transporters. Understanding of renal ankyrin spliceoform function and regulation, however, is in its infancy.

Ank3200/215 association with Na-K-ATPase. With reagents developed against Ank1, the association of ankyrin with the Na-K-ATPase was described in renal epithelia, retinal pigmented epithelium, choroid plexus, and parotid gland (1, 10, 21-24). Additionally, two ankyrin-binding sites have been identified on Na-K-ATPase (3, 13). In rat TAL, immunoelectron microscopy showed both ankyrin and Na-K-ATPase were colocalized on the basolateral membrane. The lack of specific immunoreactivity in the rat PT cells, however, left the ankyrin-Na-K-ATPase association in PT cells undetermined. In the mouse cortex, antibodies specific to the Ank3200/215 spliceoforms displayed faint but specific staining of the PT cells (Fig. 5). Similar to the TAL, the ankyrin showed basolateral staining, placing it in the proper location for retaining Na-K-ATPase in PT cells. However, ankyrin staining also appeared in the terminal web region where Na-K-ATPase is not localized. This suggests that one or both of the Ank3200/215 spliceoforms serves an alternative function than to bind Na-K-ATPase.

PT have less Ank3200/215 per unit Na-K-ATPase. TAL and distal convoluted tubules have higher Na-K-ATPase activities per millimeter than the PT (14). Compared with the renal cortex, PT cells had significantly lower levels of Ank3200/215 (Fig. 3A). Thus the lower Ank3200/215 measured in PT vs. the renal cortex could reflect a lower required binding capacity.

As shown previously in rats and predicted by the measured activity levels (5, 14), Na-K-ATPase immunofluorescence in the mouse cortex showed greater staining intensity in the TAL cells than in the PT cells (Fig. 5). Similarly, Ank3200/215 staining intensity was greater in the TAL cells than in the PT cells. When the Ank3200/215 and Na-K-ATPase intensities in TAL cells were set to equivalent levels, qualitatively, the PT staining intensity for Na-K-ATPase was greater than the Ank3200/215 staining. This indicated that there was less Ank3200/215 per Na-K-ATPase in the PT than in TAL cells. Furthermore, the portion of the Ank3200/215 in the terminal web and not along the basolateral membrane would not be tethering Na-K-ATPase in the basolateral membrane. This qualitative observation was quantified when the differences were measured in double-stained immunoblots of isolated PT vs. renal cortical samples (Fig. 4B) and LLC-PK1 cells vs. MDCK cells (Fig. 6A). Despite the fact that the cortex is primarily PT cells, isolated PT had a significantly lower level of Ank3200/215 per unit of Na-K-ATPase than the renal cortex. When cultured proximal cells were compared with cultured distal cells, the lower Ank3200/215 content was even more striking.

Several explanations would account for the lower Ank3200/215 to Na-K-ATPase ratio in PT cells. First, the DT may have a greater number of integral proteins, in addition to Na-K-ATPase, which require Ank3200/215 tethering. Second, Na-K-ATPase in PT may be tethered to fodrin by other, yet discovered, ankyrin spliceoforms. Third, PT cells may have a lower percentage of their Na-K-ATPase tethered to the fodrin cytoskeleton. These hypotheses are currently under investigation.

Ankyrin spliceoform expression diverges between species. A species-specific Ank3 expression in kidneys was apparent with the first descriptions of Ank3 spliceoforms (16, 28). Despite their close phylogenetic relationship, Peters et al. (28) described in 1995 the presence of five Ank3 spliceoforms in the mouse kidney while Kordeli et al. (16) in 1995 reported only two Ank3 spliceoforms present in the rat kidney. The most striking difference was the absence of the relatively abundant 170-kDa spliceoform in rat kidneys. This difference did not result from a difference in applied antibodies, since identical results were obtained in the rat kidney with the Ank3-R1 antibody (Fig. 6B). The species-specific expression was even more pronounced in the rabbit kidney. More phylogenetically distant, the rabbit kidney did not display spliceoforms at 120 and 170 kDa, had a prominent band at 145 kDa, and expressed a narrow doublet around 215 kDa. This is of particular interest since the 120-kDa spliceoform, the most prevalent form in the mouse and rat kidneys, lacks the membrane-binding domain, whereas the 145-kDa spliceoform, the most prevalent form in the rabbit kidney, would be predicted from its size to contain at least a partial membrane-binding domain. Like the rabbit kidney, porcine-derived LLC-PK1 cells also display the 145-kDa band but not the 120- and 170-kDa bands. Furthermore, the 215-kDa doublet in canine-derived MDCK cells was similar to the 215-kDa doublet in rabbits. Although the functional significance remains to be determined, these studies clearly indicate there is a marked species variation in Ank3 spliceoform expression.

In summary, these studies showed that the epithelial ankyrin spliceoforms expressed in the mouse kidney are differentially expressed in the various renal epithelia. Furthermore, within specific renal epithelia, there is an apparent divergence of ankyrin spliceoform expression. Although the complete functional significance will require further study, the apparent tethering of Na-K-ATPase in the basolateral membrane by the Ank3200/215 spliceoform is significantly lower in PT vs. DT cells. The continued identification and localization of novel Ank3 spliceoforms highlight the potential role of ankyrins in mediating the delivery, segregation, and regulation of membrane receptors, transporters, and adhesion molecules.

    ACKNOWLEDGEMENTS

We thank Dr. Blanche Capel for help in optimizing the tubule isolation procedure, and we thank Yan Kuang for technical assistance.

    FOOTNOTES

This work was supported by National Institutes of Health Grants DK-26816 (to L. J. Mandel) and HL-55321 (to L. L. Peters). The confocal microscope was paid for by National Science Foundation Grant BIR-9318118.

Address for reprint requests and present address of R. B. Doctor: Box B-158, 4200 E. 9th Ave., Denver, CO 80262.

Received 23 June 1997; accepted in final form 18 September 1997.

    REFERENCES
Top
Abstract
Introduction
Methods
Results
Discussion
References

1.   Alper, S. L., A. Stuart-Tilley, C. F. Simmons, D. Brown, and D. Drenckhahn. The fodrin-ankyrin cytoskeleton of choroid plexus preferentially colocalizes with apical Na(+)K(+)-ATPase rather than with basolateral anion exchanger AE2. J. Clin. Invest. 93: 1430-1438, 1994.

2.   Davis, J. Q., L. Davis, and V. Bennett. Diversity in membrane binding sites of ankyrins. Brain ankyrin, erythrocyte ankyrin, and processed erythrocyte ankyrin associate with distinct sites in kidney microsomes. J. Biol. Chem. 264: 6417-6426, 1989[Abstract/Free Full Text].

3.   Devarajan, P., D. A. Scaramuzzino, and J. S. Morrow. Ankyrin binds to two distinct cytoplasmic domains of Na,K ATPase alpha -subunit. Proc. Natl. Acad. Sci. USA 91: 2965-2969, 1994[Abstract/Free Full Text].

4.   Devarajan, P., P. R. Stabach, A. S. Mann, T. Ardito, M. Kashgarian, and J. S. Morrow. Identification of a small cytoplasmic ankyrin (AnkG119) in kidney and muscle that binds beta -spectrin and associates with the Golgi apparatus. J. Cell Biol. 133: 819-830, 1996[Abstract/Free Full Text].

5.   Doctor, R. B., V. Bennett, and L. J. Mandel. Degradation of spectrin and ankyrin in the ischemic rat kidney. Am. J. Physiol. 264 (Cell Physiol. 33): C1003-C1013, 1993[Abstract/Free Full Text].

6.   Drenkhahn, D., and V. Bennett. Polarized distribution of Mr 210,000 and 190,000 analogs of erythrocyte ankyrin along the plasma membrane of transporting epithelia, neurons, and photoreceptors. Eur. J. Cell Biol. 43: 479-486, 1987[Medline].

7.   Ecelbarger, C. A., J. Terris, J. R. Hoyer, S. Nielsen, J. B. Wade, and M. A. Knepper. Localization and regulation of the rat renal Na+-K+-2Cl- cotransporter, BSC-1. Am. J. Physiol. 271 (Renal Fluid Electrolyte Physiol. 40): F619-F628, 1996[Abstract/Free Full Text].

8.   Fairbanks, G., T. Speck, and D. Wallach. Electrophoretic analysis of the major polypeptides of the human erythrocyte membrane. Biochemistry 10: 2606-2617, 1971[Medline].

9.   Golenhofen, N., R. B. Doctor, R. Bacallao, and L. J. Mandel. Actin and villin compartmentation during ATP depletion and recovery in renal cultured cells. Kidney Int. 48: 1837-1845, 1995[Medline].

10.   Gundersen, D., J. Orlowski, and E. Rodriguez-Boulan. Apical polarity of Na,K ATPase in retinal epithelium is linked to a reversal of the ankyrin-fodrin submembrane cytoskeleton. J. Cell Biol. 112: 863-872, 1991[Abstract/Free Full Text].

11.   Hammerton, R. W., K. A. Krzeminski, R. W. Mays, T. A. Ryan, D. A. Wollner, and W. J. Nelson. Mechanism for regulating cell surface distribution of Na+,K+-ATPase in polarized epithelial cells. Science 254: 847-850, 1991[Abstract/Free Full Text].

12.   Hoock, T. C., L. L. Peters, and S. E. Lux. Isoforms of ankyrin-3 that lack the NH2-terminal repeats associate with mouse macrophage lysosomes. J. Cell Biol. 136: 1059-1070, 1997[Abstract/Free Full Text].

13.   Jordan, C., B. Puschel, R. Koob, and D. Drenkhahn. identification of a binding motif for ankyrin on the alpha -subunit of Na,K ATPase. J. Biol. Chem. 270: 29971-29975, 1995[Abstract/Free Full Text].

14.   Katz, A. I., A. Doucet, and F. Morel. Na-K-ATPase activity along the rabbit, rat, and mouse nephron. Am. J. Physiol. 237 (Renal Fluid Electrolyte Physiol. 6): F114-F120, 1979[Abstract/Free Full Text].

15.   Koob, R., M. Zimmermann, W. Schoner, and D. Drenkhahn. Colocalization and coprecipitation of ankyrin and Na,K ATPase in kidney epithelial cells. Eur. J. Cell Biol. 45: 230-237, 1987.

16.   Kordeli, E., S. Lambert, and V. Bennett. Ankyrin G: A new ankyrin gene with neural-specific isoforms localized at the axonal initial segment and node of Ranvier. J. Biol. Chem. 270: 2352-2359, 1995[Abstract/Free Full Text].

17.   Kriz, W., and L. Bankir. A standard nomenclature for structures of the kidney. Am. J. Physiol. 254 (Renal Fluid Electrolyte Physiol. 23): F1-F8, 1988[Free Full Text].

18.   Lambert, S., H. Yu, and J. T. Prchal. cDNA sequence for human erythrocyte ankyrin. Proc. Natl. Acad. Sci. USA 87: 1730-1734, 1990[Abstract/Free Full Text].

19.   Lux, S., K. M. John, and V. Bennett. Analysis of cDNA for human erythrocyte ankyrin indicates a repeated structure with homology to tissue-differentiation and cell-cycle control proteins. Nature 344: 36-43, 1990[Medline].

20.   Mandel, L. J., R. B. Doctor, and R. Bacallao. ATP depletion: a novel method to study junctional properties in epithelial tissues. II. Internalization of Na+,K+-ATPase and E-cadherin. J. Cell Sci. 107: 3315-3324, 1994[Abstract].

21.   Masuzawa, T., T. Ohta, K. Kawakami, and F. Sato. Immunocytochemical localization of Na,K ATPase in the canine choroid plexus. Brain 108: 625-646, 1985[Abstract/Free Full Text].

22.   Masuzawa, T., T. Ohta, M. Kawamura, N. Nakahara, and F. Sato. Immunohistochemical localization of Na,K ATPase in the choroid plexus. Brain Res. 302: 357-362, 1984[Medline].

23.   McNeill, H., M. Ozawa, R. Kemler, and W. J. Nelson. Novel function of the cell adhesion molecule uvomorulin as an inducer of cell surface polarity. Cell 62: 309-316, 1990[Medline].

24.   Nelson, W. J., R. W. Hammerton, A. Z. Wang, and E. M. Shore. Involvement of the membrane-cytoskeleton in development of epithelial cell polarity. Semin. Cell Biol. 1: 359-371, 1990[Medline].

25.   Nelson, W. J., and P. J. Veshnock. Ankyrin binding to (Na+ + K+)ATPase and implications for the organization of membrane domains in polarized cells. Nature 328: 533-536, 1987[Medline].

26.   Nielsen, S., S. R. DiGiovanni, E. I. Christensen, M. A. Knepper, and H. W. Harris. Cellular and subcellular immunolocalization of AVP-regulated water channels in the rat kidney. Proc. Natl. Acad. Sci. USA 90: 11663-11667, 1993[Abstract/Free Full Text].

27.   Otto, E., M. Kunimoto, T. McLaughlin, and V. Bennett. Isolation and characterization of cDNAs encoding human brain ankyrins reveal a number of alternatively spliced genes. J. Cell Biol. 114: 241-253, 1991[Abstract/Free Full Text].

28.   Peters, L. L., K. M. John, F. M. Lu, E. M. Eicher, A. Higgins, M. Yialamas, L. C. Turtzo, A. J. Otsuka, and S. E. Lux. Ank3, a widely distributed new member of the ankyrin gene family and the major ankyrin in kidney, is expressed in alternatively spliced forms, including forms that lack the repeat domain. J. Cell Biol. 130: 313-330, 1995[Abstract/Free Full Text].

29.   Piepenhagen, P. A., L. L. Peters, S. E. Lux, and W. J. Nelson. Differential expression of Na+-K+-ATPase, ankyrin, fodrin, and E-cadherin along the kidney nephron. Am. J. Physiol. 269 (Cell Physiol. 38): C1417-C1432, 1995[Abstract/Free Full Text].

30.   Rodman, J. S., M. Mooseker, and M. G. Farquhar. Cytoskeletal proteins of the rat kidney proximal tubule brush border. Eur. J. Cell Biol. 42: 319-327, 1986[Medline].

31.   Sullivan, L. P., J. A. Grantham, L. Rome, D. Wallace, and J. J. Grantham. Fluorescein transport in isolated proximal tubules in vitro: epifluorometric analysis. Am. J. Physiol. 258 (Renal Fluid Electrolyte Physiol. 27): F46-F51, 1990[Abstract/Free Full Text].

32.   Towbin, H., T. Staehlin, and J. Gordon. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedures and some applications. Proc. Natl. Acad. Sci. USA 76: 4350-4354, 1979[Abstract/Free Full Text].

AJP Renal Physiol 274(1):F129-F138
0363-6127/98 $5.00 Copyright © 1998 the American Physiological Society


This article has been cited by other articles:


Home page
 Am. J. Physiol. Renal Physiol.Home page
N. P. Curthoys, L. Taylor, J. D. Hoffert, and M. A. Knepper
Proteomic analysis of the adaptive response of rat renal proximal tubules to metabolic acidosis
Am J Physiol Renal Physiol, January 1, 2007; 292(1): F140 - F147.
[Abstract] [Full Text] [PDF]

Home page
 J. Cell Sci.Home page
S. Ghosh, F. C. Dorsey, and J. V. Cox
CK2 constitutively associates with and phosphorylates chicken erythroid ankyrin and regulates its ability to bind to spectrin
J. Cell Sci., January 11, 2002; 115(21): 4107 - 4115.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Renal Physiol.Home page
J. Chen and M. C. Wagner
Altered membrane-cytoskeleton linkage and membrane blebbing in energy-depleted renal proximal tubular cells
Am J Physiol Renal Physiol, April 1, 2001; 280(4): F619 - F627.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
S. Yu, D. Yu, E. Lee, M. Eckhaus, R. Lee, Z. Corria, D. Accili, H. Westphal, and L. S. Weinstein
Variable and tissue-specific hormone resistance in heterotrimeric Gs protein alpha -subunit (Gsalpha ) knockout mice is due to tissue-specific imprinting of the Gsalpha gene
PNAS, July 21, 1998; 95(15): 8715 - 8720.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
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
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Doctor, R. B.
Right arrow Articles by Mandel, L. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Doctor, R. B.
Right arrow Articles by Mandel, L. J.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online