INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

REGULATION OF SEVERAL transport processes in the kidney have been proposed to occur as a result of regulated trafficking of transporter proteins into and out of the plasma membrane. For example, the osmotic water permeability of the collecting duct is increased by vasopressin, acting via increases in intracellular cAMP, through exocytosis of intracellular vesicles whose limiting membranes contain aquaporin-2, the vasopressin-regulated water channel (22). Addition of aquaporin-2 to the apical plasma membranes increases its water permeability. In addition, vasopressin has also been proposed to decrease the rate of endocytosis of water channels in the collecting duct (24, 28). The activities of other renal epithelial transporters have also been proposed to be modulated through regulation of trafficking to and from the plasma membrane. These include the H⁺-ATPase (10) and the Cl/HCO₃ exchanger (31) of collecting duct intercalated cells, the Na⁺ channel in cultured amphibian collecting duct cells (A6 line) (16), the apical Na⁺/H⁺ exchanger (NHE3) of the proximal tubule (1), and the Na⁺-K⁺-ATPase in the cortical collecting duct (6).

Although relatively little is known yet about regulation of trafficking in epithelia, substantial progress has been made in investigation of mechanisms of regulation of trafficking of neurotransmitter-containing vesicles at synapses in the central nervous system (4, 27). Several proteins have been identified that are believed to play a role in the docking and fusion of these synaptic vesicles with the plasma membrane. Although much remains to be learned about this process, it is clear that several families of membrane proteins called SNAREs (vesicle-targeting proteins) play essential roles in this process. The general view has emerged that SNAREs present in the translocating vesicle (v-SNAREs) bind to SNAREs present in the target membrane (t-SNAREs) and that this interaction is critical to the docking and fusion of vesicles in the correct target membrane domain (4, 27). Two families of t-SNAREs have thus far been identified, the syntaxins and homologs of SNAP-25 ("synaptosomal associated protein"). Among these, we have identified two t-SNAREs in our studies of aquaporin-2 trafficking that are present in principal cells at the correct location to play a potential role in aquaporin-2 trafficking, namely syntaxin-4 (20) and SNAP-23 (13). In addition, two families of v-SNAREs have been identified, the synaptobrevins ("vesicle-associated membrane proteins" or VAMPs) and the synaptotagmins. Previously, we (25) and others (14, 18) have demonstrated that one of the synaptobrevins, namely synaptobrevin-2 (or VAMP-2), is expressed in the principal cells of the collecting duct and, furthermore, is a major integral membrane protein in aquaporin-2-containing vesicles. Whether any of the synaptotagmins is expressed in collecting ducts or in other renal epithelia has not yet been reported.

The synaptotagmins (29) are a family of integral membrane proteins with a single membrane-spanning domain near the amino terminus of the polypeptide chain. The cytoplasmic region, the COOH-terminal portion, contains two calcium-regulatory domains, called C₂ domains, which are structurally similar to the C₂ domains responsible for Ca²⁺-mediated regulation of protein kinase C. Several of the synaptotagmins exhibit Ca²⁺-dependent binding to phospholipids and to syntaxin via the first C₂ domain (C₂-A). Synaptotagmin I, expressed predominantly in the central nervous system, is believed to play a central role in the process by which a local rise in intracellular Ca²⁺ near the synaptic ending triggers exocytosis of neurotransmitter-containing vesicles (26), and it is likely that other members of the synaptotagmin family have similar functions. In addition, the synaptotagmins have been demonstrated to bind to AP2 (clathrin adaptor protein) in a calcium-independent manner and therefore have the potential to play a role in regulation of endocytosis as well as exocytosis (33). Recently, multiple novel members of the synaptotagmin family have been cloned, many of which are expressed at sites outside of the central nervous system including the kidney (12, 17, 32). One synaptotagmin that has been demonstrated to be expressed in the kidney at an mRNA level is synaptotagmin VIII, an isoform that does not undergo Ca²⁺-dependent binding to phospholipids or syntaxins (17). Here, we use immunochemical methods and RT-PCR to localize synaptotagmin VIII expression along the renal tubule.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Immunochemical Studies

Preparation of membrane protein samples for immunoblotting. Tissue samples were homogenized in ice-cold isolation solution (250 mM sucrose, 10 mM triethanolamine) containing protease inhibitors, leupeptin (1 µg/ml; Bachem California, Torrance, CA) and phenylmethylsulfonyl fluoride (0.1 mg/ml; US Biochemical, Toledo, OH), using a tissue homogenizer (Omni 1000 fitted with a micro-sawtooth generator). The homogenates were initially spun at low speed (1,000 g) for 10 min at 4°C using a Sorvall RC2-B centrifuge to pellet incompletely homogenized fragments and nuclei. The pellets were suspended in isolation solution, rehomogenized, and spun again at 1,000 g for 10 min. The supernatants were combined and either 1) spun at 200,000 g for 1 h at 4°C in a Beckman ultracentrifuge (Ti-80 rotor) to obtain a cell membrane pellet or 2) spun sequentially at 17,000 g for 20 min (Sorvall RC2-B centrifuge with SS34 rotor) and 200,000 g for 60 min (Beckman ultracentrifuge with Ti-80 rotor) to obtain pellets enriched in plasma membranes and intracellular vesicles, respectively. These pellets were resuspended in the isolation solution, and the total protein concentration was measured in each sample using the Pierce bicinchoninic acid protein assay reagent kit. These samples were then solubilized by heating at 60°C for 20 min in Laemmli sample buffer.

Electrophoresis and immunoblotting of membrane proteins. SDS-PAGE was carried out on minigels of 12% polyacrylamide or gradient gels of 4-20% polyacrylamide (Novex, San Diego, CA) using the membrane samples solubilized in Laemmli buffer. The separated proteins were transferred from the gels to nitrocellulose membranes electrophoretically using a Bio-Rad transfer apparatus (Bio-Rad Laboratories, Hercules, CA). The membranes were blocked with 5% nonfat dry milk in wash buffer (150 mM NaCl, 50 mM NaH₂PO₄, 0.05% Tween-20, pH 7.5). The nitrocellulose membranes were then probed with the affinity-purified polyclonal antibody to synaptotagmin VIII at an IgG concentration of 0.5 µg/ml in antibody dilution buffer (above wash buffer containing 0.02% sodium azide and 0.1% bovine serum albumin, pH 7.5). Affinity-purified anti-aquaporin-1 antibody (L266) and anti-aquaporin-2 antibody (L126) were used at a concentration of 0.5 µg/ml each. The secondary antibody was donkey anti-rabbit IgG conjugated to horseradish peroxidase (Pierce no. 31458) used at a concentration of 0.16 µg/ml in milk block prepared as above. Sites of antigen-antibody reaction were visualized by chemiluminescence using SuperSignal Substrate (Pierce) before exposing to light-sensitive imaging film (Kodak No. 165-1579 Scientific Imaging Film).

Preparation of IMCD and non-IMCD enriched fractions. Fractions enriched in inner medullary collecting ducts (IMCD) and non-IMCD elements were prepared from the inner medullas of Sprague-Dawley rat kidneys (200-250 g body wt) as described by Chou et al. (5). Briefly, six rats were injected with furosemide (5 mg/animal ip) 20-30 min before decapitation. The kidneys were removed, and the inner medullas were separated quickly and immersed in ice-cold bicarbonate-buffered isotonic solution (composition in mM: 118 NaCl, 25 NaHCO₃, 2.5 K₂HPO₄, 2 CaCl₂, 1.2 MgSO₄, 5 CH₃COONa, 5 urea, and 5.5 glucose). The inner medullas were then minced finely and transferred into tubes containing bicarbonate-buffered isotonic solution with collagenase B (3 mg/ml; Boehringer-Mannheim, Indianapolis, IN) and hyaluronidase (600 U/ml; Worthington Biochemicals, Freehold, NJ). The minced tissue was digested at 37°C with constant bubbling of 95% air-5% CO₂, until the tubule segments were well separated. Halfway through the incubation, DNase I (Boehringer-Mannheim) was added to the digesting solution to obtain a final concentration of 0.001%. The latter helps by destroying stray DNA in the solution which might cause aggregation of separated tubule segments. Approximately 25% of the suspension volume was set aside and kept on ice ("whole inner medulla" sample). The remaining 75% was subjected to repeated low-speed centrifugation (50 g, 10 s) and washes with the bicarbonate-buffered isotonic solution to separate fractions enriched in IMCD (pellet) and non-IMCD (supernatants) elements. Following this separation, all the three samples (whole inner medulla, IMCD-enriched fraction and non-IMCD fraction) were pelleted by centrifugation at 4,000 g for 20 min. These samples were homogenized, and a membrane fraction was prepared as described above.

Immunocytochemistry. Tissue for immunocytochemistry was from 140- to 200-g male Sprague-Dawley rats. Similar localizations were obtained from rat kidneys fixed with 2% paraformaldehyde plus 0.01% glutaraldehyde, 2% paraformaldehyde alone, or 20°C methanol in 20% dimethyl sulfoxide. Fixed kidneys were sectioned and labeled as previously described by Ginns et al. (9). The primary antibodies were used at 10 µg/ml. Secondary antibodies were conjugated either to fluorescein or Texas Red. Labeled sections were examined with a Zeiss LSM410 confocal microscope. Secondary antibodies were chosen to assure single-species specificity among IgG of rabbit, mouse and guinea pig (donkey anti-rabbit-FITC, 711-095-152; donkey anti-mouse-TR, 715-075-150; and donkey anti-guinea pig-TR, 706-075-148; all from Jackson ImmunoResearch Laboratories, West Grove, PA). Appropriate control studies demonstrated that these antibodies do not cross-label the other primary antibodies used in the multilabel localizations.

RT-PCR Experiments

Primers. Sequence-specific primers for rat synaptotagmin VIII were designed on the basis of the following criteria: specificity, T_m close to 60°C, and lack of predicted internal structure. The sequence of the sense primer (primer 1) corresponded to bp 376-395 (5' TGA AGA GAC ATG CTG CTT CC 3'), and the sequence of the antisense primer (primer 2) corresponded to bp 574-593 (5' TAC AGC TTC CAG CTG TCA CG 3') of the rat synaptotagmin VIII cDNA partial sequence submitted by Li and associates (17) to the GenBank database (accession no. U20110). In some cases a second reaction was run using an internal sense primer (primer 3) corresponding to bp 391-401 (5' CTT CCT GCT AAG TCT AGC CTG 3') and primer 2 (above). Primers specific for aquaporin-2 water channel, used as a positive control for reverse transcription step (see below) were described previously (20).

Reverse transcription. The starting material was either total RNA extracted from whole tissue blocks or from microdissected tubules (up to 4 µg per sample) or microdissected tubules prepared as described (20) (8.7 µl of total volume per tube). Reverse transcription was initiated by adding 11.3 µl of a mix containing 50 U avian myeloblastosis virus reverse transcriptase (Boehringer-Mannheim), 50 U RNase inhibitor, 20 nmol of each deoxynucleotide phosphate, and 1.6 µg of poly(dT)₁₅ in Boehringer-Mannheim reverse transcription incubation buffer. For reverse transcription-negative (RT) control reactions, the enzyme was substituted with an equal volume of diethyl pyrocarbonate-treated water. The reaction was carried out for 60 min at 55°C followed by inactivation of the enzyme for 1 min at 95°C.

Polymerase chain reaction. After reverse transcription, the tubes were placed on ice, and 80 µl of a master mix containing 200 µmol of each deoxynucleotide phosphate, 50 pmol each of sense and antisense primers, and 2.5 U DNA polymerase (Ampli-Taq, Perkin-Elmer) in Perkin-Elmer reaction buffer. Sequence-specific primers for aquaporin-2 water channel were run in parallel reactions in each set of reactions as a positive control for detergent permeabilization and/or reverse transcription reaction. The samples were overlaid with mineral oil and processed for 31 cycles [94°C, 1 min (3 min for the initial cycle); 60°C, 1 min; 72°C, 1 min]. The elongation period in the last cycle was extended to 7 min. Ten-microliter aliquots of the final PCR product were electrophoresed directly on 2% agarose gels after mixing with 2 µl of DNA loading buffer. The gels were stained with ethidium bromide, destained, and photographed under UV light. In some cases, 2 µl of the PCR product from the first reaction was used as starting material for a second run of PCR performed using the nested primer pair (primers 3 and 2). In some cases, the PCR products from several reactions were pooled and purified using Aqua-Select-D G-50 Spin Chromatography columns (5 Prime  3 Prime, Boulder, CO) and sequenced commercially (Sequi-net; Div. of Macromolecular Resources, Colorado State Univ., Ft. Collins, CO).

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Immunoblotting Studies

View larger version (20K): [in this window] [in a new window]   View larger version (34K): [in this window] [in a new window]   Fig. 1.   A: immunoblot of renal inner medullary low-density microsome fraction. Low-density microsome fraction was obtained by centrifuging the supernatant from a 17,000 g spin at 200,000 g and solubilizing the pellet in Laemmli sample buffer. SDS-PAGE was run on a 12% polyacrylamide minigel using 20 µg protein from low-density microsome fraction. After transfer of proteins to nitrocellulose, blots were probed with either affinity-purified anti-synaptotagmin VIII antibody (L330) at an IgG concentration of 0.5 µg/ml (left) or with the protein A purified preimmune IgG fraction (0.5 µg/ml; right). The affinity-purified antibody labeled a predominant band at 52 kDa, which was not seen in the blot probed with preimmune IgG fraction. B: immunoblot loaded with a conjugate of the immunizing peptide and ovalbumin (OVA) (first 3 lanes), ovalbumin alone (4th lane), and peptide alone (5th lane). Blot was probed with affinity-purified anti-synaptotagmin-VIII antibody (L330, 0.5 µg/ml). The antibody recognized the conjugate and the peptide but not ovalbumin. The electrophoresis was carried out using a 4-20% polyacrylamide gradient gel. (In conjugate lanes, the lower band is presumably the 1:1 conjugate of peptide and ovalbumin, whereas higher molecular weight bands are presumably conjugates involving two or more ovalbumin molecules.)

Figure 2 is an immunoblot loaded with membrane fractions from three regions of the rat kidney (cortex, outer medulla, and inner medulla), as well as three other tissues (heart, spleen, and cerebral cortex) and probed with affinity-purified anti-synaptotagmin VIII (L330). The membrane fractions were obtained by centrifuging an initial homogenate at 1,000 g and then centrifuging the 1,000 g supernatant sequentially at 17,000 g (low-speed pellet) and 200,000 g (high-speed pellet). The same amount of membrane protein (20 µg) was loaded in each lane. The 52-kDa synaptotagmin VIII band was seen in all three regions of the kidney. It was more prominent in the 17,000 g fraction than in the 200,000 g fraction in all three renal regions. It was more abundant in the cortex and outer medulla than in inner medulla. With the short exposure time used for this blot, the 52-kDa band was barely visible in the 200,000 g fraction from the inner medulla, although longer exposures clearly demonstrated the 52-kDa band as shown in Fig. 1. In the 17,000 g fraction from the inner medulla, in addition to the 52-kDa band, there was a weaker additional band at slightly lower molecular weight. The expression level of synaptotagmin VIII was extremely high in heart as indicated by the gross overexposure of the 52-kDa band on this blot. There was little or no expression in spleen. In cerebral cortex, the 52-kDa band was prominent in the 17,000 g pellet, although in this tissue several unidentified bands were also seen in both fractions. Conceivably, these unidentified bands represent synaptotagmin VIII complexed with other proteins.

View larger version (76K): [in this window] [in a new window]   Fig. 2.   Immunoblot showing regional distribution of anti-synaptotagmin VIII labeling in kidney (left) and anti-synaptotagmin VIII labeling in heart, spleen and cerebral cortex (right). Samples were prepared by homogenizing the indicated tissues and carrying out limited differential centrifugation. Supernatant from an initial 1,000 g spin was subjected to two centrifugations in series at 17,000 g and 200,000 g. The pellets from these two spins were solubilized in Laemmli sample buffer. SDS-PAGE was run on minigels using 20 µg of membrane protein from the two fractions. After transferring the proteins to nitrocellulose membranes, the blots were probed with the anti-synaptotagmin VIII antibody (L330; 0.5 µg/ml).

To determine whether synaptotagmin VIII is present in collecting duct cells, we fractionated a renal inner medullary tubule suspension (whole IM) into a collecting duct-enriched (IMCD) fraction and a non-collecting duct (Non-IMCD) fraction (Fig. 3). Figure 3 shows three immunoblots prepared from SDS-polyacrylamide gels loaded with membrane proteins (200,000 g pellets obtained from the 1,000 g supernatant) from whole IM, IMCD, and non-IMCD fractions. The top blot in Fig. 3 was probed with the affinity-purified anti-synaptotagmin VIII antibody. The middle blot in Fig. 3 was probed with an antibody that recognizes aquaporin-2, a water channel expressed only in the collecting duct. The bottom blot in Fig. 3 was probed with an antibody to aquaporin-1, a water channel expressed in non-IMCD elements (thin limbs of Henle's loop and vasa recta). As expected, relative to the whole inner medulla, aquaporin-2 protein was enriched in the IMCD sample, whereas aquaporin-1 was enriched in the non-IMCD sample, demonstrating the efficacy of the cell separation procedure. If synaptotagmin VIII were expressed only in collecting ducts, then its distribution would be expected to be like that of aquaporin-2, whereas if it were entirely absent from collecting ducts, its distribution may be expected to be similar to that of aquaporin-1. The blot probed with anti-synaptotagmin VIII antibody shows that there were approximately equal amounts of synaptotagmin VIII in the IMCD and non-IMCD samples, indicating that synaptotagmin VIII is expressed in both the collecting duct and non-collecting duct components of inner medulla.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Relative abundance of synaptotagmin VIII protein in fractions enriched in IMCD or non-IMCD elements compared with the whole inner medullary preparations. Inner medullary tubule suspension (whole IM) was fractionated into IMCD and non-IMCD fractions as described in METHODS. After homogenization, the supernatant from an initial 1,000 g centrifugation was subjected to centrifugation for 1 h at 200,000 g, and the pellet was solubilized in Laemmli sample buffer; 20 µg of protein from this fraction were electrophoresed on SDS-PAGE, and transferred to nitrocellulose membranes. The latter were probed with anti-synaptotagmin VIII antibody (L330) at an IgG concentration of 0.5 µg/ml (top). Additional minigels and blots were run in parallel by applying 2 or 5 µg of protein samples. These blots were probed with either anti-aquaporin-2 antibody (L126; middle) or anti-aquaporin-1 antibody (L266; bottom) both at an IgG concentration of 0.5 µg/ml. The distribution pattern of aquaporin-2 (collecting duct marker) and aquaporin-1 (loop of Henle and vasa recta marker) identifies the degree of enrichment obtained with the tubule fractionation procedure. The blot probed with anti-synaptotagmin VIII antibody (top) shows that synaptotagmin VIII is not preferentially enriched in either the IMCD or non-IMCD tubule fractions.

Synaptotagmin VIII mRNA Expression in Rat Kidney

View larger version (86K): [in this window] [in a new window]   Fig. 4.   RT-PCR amplification of synaptotagmin VIII mRNA in microdissected IMCD and total RNA extracted from microdissected IMCD segments. RT-PCR was carried out to determine whether synaptotagmin VIII mRNA is expressed in IMCDs. Lane 1 shows size markers. Total RNA extracted from brain (lane 2) and from the inner medulla (lane 3) was used as starting material (each 4 µg) to provide positive controls for synaptotagmin VIII expression. Lanes 3 and 4 show results for permeabilized microdissected IMCD segments dissected from the renal inner medulla. RT indicates that reverse transcriptase was omitted from the reaction, whereas RT+ indicates that reverse transcriptase was included. RT (lane 3) and RT+ (lane 4) tubes were each loaded with 2 mm of IMCD segments. Lanes 5 and 6 show results for total RNA extracted from microdissected IMCD segments or total RNA corresponding to 2.5 mm of microdissected IMCD segments. RT (lane 5) and RT+ (lane 6) tubes were each loaded with 12.5% of total RNA extracted from 20 mm of microdissected IMCD segments. PCR was run for 31 cycles using sequence-specific primers for synaptotagmin VIII (expected product size 203 bp). Blank (lane 7) contained no RNA or tubule samples. Parallel amplifications for aquaporin-2 water channel target (product size 562 bp) were run using microdissected outer medullary collecting duct segments (2 mm) (lane 8) and total RNA extracted from rat IMCD (12.5% of total yield from 20 mm).

Immunofluorescence Localization of Synaptotagmin VIII in the Renal Medulla

View larger version (145K): [in this window] [in a new window]   Fig. 5.   Immunofluorescence localization of synaptotagmin VIII relative to Na-K-2Cl cotransporter in rat outer medulla. A: localization with anti-synaptotagmin VIII. B: localization of Na-K-2Cl cotransporter (both absorptive and secretory isoforms) using T4 monoclonal antibody. C: colocalization with both antibodies. D: peptide adsorption control in which anti-synaptotagmin VIII was preadsorbed with an excess of the immunizing peptide. In A, synaptotagmin VIII antibody labels thick ascending limbs and collecting ducts. Note that B shows labeling of both the apical region of the thick ascending limb (absorptive isoform) and the basolateral region of collecting duct intercalated cells. In C, synaptotagmin VIII localization is seen as green and the Na-K-2Cl as red. Areas where both antibodies label are is seen as yellow-green (arrows). In D, the peptide-adsorbed anti-synaptotagmin VIII antibody does not display this labeling pattern. Bar = 25 µm.

View larger version (91K): [in this window] [in a new window]   Fig. 6.   Longitudinal section of rat outer medulla displaying localization of synaptotagmin VIII relative to Na-K-2Cl cotransporter. Localization with anti-synaptotagmin VIII antibody alone (A) and colocalization with anti-synaptotagmin VIII and T4 anti-Na-K-2Cl cotransporter antibody (B). In A, synaptotagmin VIII antibody labels the basolateral region of the thick ascending limb (green in B) and the basolateral aspect a subset of collecting duct cells identified in B as intercalated cells (yellow overlap with anti-Na-K-2Cl antibody). As seen in B, synaptotagmin VIII labeling appears in the apical region of principal cells of collecting duct as well (cells unlabeled with T4 antibody). Bar = 10 µm.

Figure 6A shows a longitudinal section of outer medulla labeled with the anti-synaptotagmin VIII antibody, whereas Fig. 6B shows dual labeling with anti-synaptotagmin VIII (green) and the monoclonal antibody to the Na-K-2Cl cotransporter (red). The bright yellow color seen in Fig. 6B demonstrates a region of colocalization indicating anti-synaptotagmin VIII labeling of the basolateral surface of the type A intercalated cells. In addition, cells between the intercalated cells in the collecting duct, i.e., the principal cells, were also labeled with anti-synaptotagmin VIII antibody. However, in contrast to the intercalated cells, principal cells were labeled predominantly in the apical and subapical region in a punctate distribution consistent with labeling of subapical vesicles and/or the apical plasma membrane (green in Fig. 6B).

Figure 7 shows labeling with anti-synaptotagmin VIII (A), labeling with the monoclonal antibody to the Na-K-2Cl cotransporter (B), and dual labeling with both antibodies (C) in adjacent sections from the transition region between the inner medulla (upper portion of each section) and outer medulla (lower portion of each section) of rat kidney. Labeling in the outer medulla was consistent with that seen in Figs. 5 and 6 above, i.e., the anti-synaptotagmin VIII antibody labeled both thick ascending limbs and collecting ducts. As seen in Fig. 7, A and C, in the inner medulla, the anti-synaptotagmin VIII antibody labeled thin limbs of Henle's loops. Presumably, these structures are ascending limbs, since no thin limb labeling was seen in the inner stripe of the outer medulla where thin descending limbs, but not thin ascending limbs, are present.

View larger version (60K): [in this window] [in a new window]   Fig. 7.   Immunofluorescence localization of synaptotagmin VIII relative to Na-K-2Cl cotransporter at the renal inner medulla/outer medulla border of rat. Localization with L330 synaptotagmin VIII antibody (A) and T4 anti-Na-K-2Cl cotransporter antibody (B). C: colocalization with both antibodies. The anti-synaptotagmin VIII antibody labels thin limbs in the inner medulla (presumably thin ascending limbs) and, as demonstrated in Figs. 5 and 6, labels thick ascending limbs and collecting ducts in the outer medulla. Bar = 25 µm.

Finally, although not shown prominently in Fig. 7, IMCD exhibited weak but variable labeling with the anti-synaptotagmin VIII antibody. The results shown in Figs. 5-7 reinforce the conclusion from cell fractionation experiments (Fig. 3) that synaptotagmin VIII is expressed in both collecting duct and other renal tubule segments.

Immunofluorescence Localization of Synaptotagmin VIII in the Renal Cortex

View larger version (89K): [in this window] [in a new window]   Fig. 8.   Immunofluorescence localization of synaptotagmin VIII in rat renal cortex. Labeling with anti-synaptotagmin VIII (A) and anti-synaptotagmin VIII preadsorbed with immunizing peptide (B) in the renal cortex. In A, synaptotagmin VIII labeling is seen in the early proximal tubule, but other structures (e.g., in top right) are unlabeled. In B, the peptide-adsorbed antibody does not display this labeling pattern. The other labeled structure in the photograph is the cortical collecting duct, which has a brighter and more prominent label. Bar = 25 µm.

Figure 9 shows labeling with the synaptotagmin VIII antibody (A), labeling with a guinea pig polyclonal antibody to the Na⁺/Ca²⁺ exchanger (a marker for connecting tubule cells and cells in the late distal convoluted tubule, DCT2; Ref. 3) (B), and dual labeling with both antibodies (C; green, synaptotagmin VIII; red, Na⁺/Ca²⁺ exchanger). The synaptotagmin VIII antibody labeled three tubule profiles in Fig. 9, A and C. The presence of yellow labeling in some of the cells indicates overlap of labeling by the two antibodies, implying that connecting tubule and/or DCT2 cells contain synaptotagmin VIII. In addition, interspersed with the Na⁺/Ca²⁺ exchanger-positive cells are additional synaptotagmin VIII-positive cells that are labeled in a basolateral distribution. These cells are presumably type A intercalated cells in the connecting tubule. Further along the tubule at bottom left of Fig. 9 (arrow), as the Na⁺-Ca²⁺ exchanger labeling is lost, synaptotagmin VIII labeling becomes prominent in the apical portion of principal cells of the initial collecting tubule, with an intracellular distribution similar to that shown in the outer medulla (compare Fig. 6).

View larger version (70K): [in this window] [in a new window]   Fig. 9.   Comparison of synaptotagmin VIII and Na/Ca exchanger localization in rat renal cortex. Immunofluorescence localization with anti-synaptotagmin VIII (A), a guinea pig polyclonal antibody to the Na/Ca exchanger (B), and colocalization of both antibodies (C) in the renal cortex. In A, distal elements can be seen to be labeled with the anti-synaptotagmin VIII antibody. B: the same structures are labeled with the antibody for the Na/Ca exchanger, a marker for the late part of the distal convoluted tubule (DCT2) and connecting tubule. In C, yellow labeling signifies areas where both antibodies label. Arrow in C points to principal cells labeled with synaptotagmin VIII antibody but not Na/Ca exchanger antibody. Bar = 25 µm.

Figure 10 shows labeling with the synaptotagmin VIII antibody (A), labeling with the anti-Na-K-2Cl cotransporter monoclonal antibody (B), and dual labeling with the same two antibodies (C; green, synaptotagmin VIII; red, Na-K-2Cl cotransporter) in the region of a renal glomerulus (G). The base of the extraglomerular mesangial region is in contact with the macula densa cells. Neither the mesangium nor the macula densa cells are labeled by the anti-synaptotagmin VIII antibody (Fig. 10, A and C). In contrast to the macula densa cells, the cortical thick ascending limbs cells surrounding the macula densa are labeled with the anti-synaptotagmin VIII antibody (Fig. 10, A and C). Na-K-2Cl cotransporter labeling can clearly be seen in both the macula densa cells and the cortical thick ascending limb cells in the apical region, as a result of the presence of the type 1 (absorptive) cotransporter.

View larger version (87K): [in this window] [in a new window]   Fig. 10.   Comparison of synaptotagmin VIII and Na-K-2Cl cotransporter localization in rat renal cortex. Immunofluorescence localization with anti-synaptotagmin VIII antibody (A), T4 antibody to Na-K-2Cl cotransporter (B), and both antibodies together (C) in the renal cortex. A: synaptotagmin VIII antibody labeling in the thick ascending limb on both apical and basolateral surfaces. Note the relative lack of synaptotagmin VIII antibody labeling in the macula densa region at site of contact of tubule with glomerulus (G). B: marker antibody T4 labeling on the apical surface of the macula densa and cortical thick ascending limb. C: labeled areas of synaptotagmin VIII antibody (green), Na-K-2Cl cotransporter antibody (red), and areas where both antibodies are colocalized (orange-yellow). G, glomerulus. Bar = 25 µm.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

Thus far, eleven distinct members of the synaptotagmin family have been identified in mammals, including several that are expressed outside the central nervous system (12, 17, 32). The general structure and function of the synaptotagmins has been recently reviewed (29). The structure is schematically summarized in Fig. 11. Most of the synaptotagmins have 400-600 amino acids with predicted molecular weights in the range 48-72 kDa. Each has a single membrane spanning domain near the amino terminus with the carboxy-terminal tail on the cytoplasmic side of the membrane. The intracytoplasmic region contains two C₂ domains analogous to the calcium binding domains found in protein kinase C. Synaptotagmin I, the best characterized synaptotagmin, is expressed chiefly in the central nervous system, where it is found in synaptic vesicles. It has been proposed to function as a calcium sensor, mediating Ca-dependent neurotransmitter release. The phospholipid- and syntaxin-binding function of the C₂-A domain is believed to play a central role in calcium-triggered exocytosis of synaptic vesicles (26). Furthermore, the propensity of its carboxy terminal region to bind clathrin-AP2 suggests a role in endocytosis. Thus, by analogy to synaptotagmin I, synaptotagmins in the kidney can be viewed as potential regulators of both exocytosis and endocytosis. Based on mRNA localization studies using Northern blotting and RT-PCR, several synaptotagmins including synaptotagmin VI, VII, and VIII are believed to be expressed in the kidney (17). We raised a rabbit polyclonal antibody directed to the immediate carboxy terminus of mouse synaptotagmin VIII using a carrier-protein-linked synthetic peptide as the immunogen. The 21-amino-acid sequence chosen is not present in any of the other known synaptotagmins, including the recently cloned synaptotagmin X (2) and XI (32). Furthermore, a BLAST analysis using the 21-amino-acid carboxy-terminal synaptotagmin VIII peptide run against the combined protein sequence data bases did not reveal significant overlap with any other known eukaryotic protein (unpublished observations). An immunoblot probed with the resulting affinity-purified antibody (L330) using a low-density membrane fraction from rat kidney (Fig. 1A) revealed a single 52-kDa band (not seen with preimmune IgG from the same rabbit), compatible with the expected molecular weight of members of the synaptotagmin family. Furthermore, the antibody recognized both the immunizing peptide and a peptide-ovalbumin conjugate on immunoblots, indicating that the antibody recognizes a sequence specific to the peptide and not, for example, an unrelated epitope present in the carrier protein. The conclusion that synaptotagmin VIII is expressed in the kidney was reinforced by RT-PCR experiments using total RNA extracted from kidney homogenates (or from microdissected tubules) which yielded a PCR product whose sequence was consistent with the reported synaptotagmin VIII sequence for rat.

View larger version (9K): [in this window] [in a new window]   Fig. 11.   Diagram of synaptotagmin structure. This diagram is based chiefly on studies of synaptotagmin I. See text for explanation. [Based on a review by Südhof and Rizo (29).]

The 52-kDa synaptotagmin VIII band was seen on immunoblots using membrane fractions from cortex, outer medulla, and inner medulla (Fig. 2), providing initial evidence that this protein is broadly distributed in the kidney. This conclusion was verified by the immunofluorescence results (Figs. 5-10) showing synaptotagmin VIII labeling of several renal tubule segments, namely, the early parts of the proximal tubule, thin ascending limbs of Henle's loop, thick ascending limbs of Henle's loop, connecting tubules, and collecting ducts in both cortex and medulla. This broad distribution, coupled with the proposed role of the synaptotagmins as a membrane trafficking protein, suggests that synaptotagmin VIII could be involved in regulation of protein trafficking at several sites along the renal tubule. The immunofluorescence localization of synaptotagmin VIII in the kidney can be summarized as having three definable components: 1) basolateral localization in the distal nephron (thin ascending limb and thick ascending limb) and in intercalated cells of the collecting duct; 2) apical and subapical labeling in the thick ascending limb, connecting tubule cells, and collecting duct principal cells; and 3) apical or subapical labeling of the early proximal tubule. Further work will be required to determine the functional roles of synaptotagmin VIII at these sites.

Relative to synaptotagmin-1, little is known regarding the function of synaptotagmin VIII. In the present study, immunoblotting reveals that rat synptotagmin VIII has an apparent molecular mass of 52 kDa, similar to that of other synaptotagmins. In contrast to synaptotagmin I, II, III, V, and VII, the first C₂ region of synaptotagmin VIII does not bind phospholipid in a calcium-dependent manner (17). This suggests that synaptotagmin VIII does not mediate calcium-triggered exocytosis. Thus, if synaptotagmin VIII functions as a regulator of exocytosis, the trigger for exocytosis may be some process other than the binding of calcium ions, e.g., a local change in pH or phosphorylation. With regard to the former possibility, synaptotagmin VIII possesses a histidine-rich region (HRHRHRK) not present in the other synaptotagmins in the cytoplasmic region of the protein flanking the transmembrane region. Conceivably, titration of these histidines could alter the binding of synaptotagmin VIII with syntaxins in the plasma membrane. This could be a mechanism whereby intercalated cells adapt their acid-base transport functions to changes in local pH. With regard to the latter possibility, synaptotagmin VIII has two putative protein kinase A phosphorylation sites in the cytoplasmic portion of the protein. Therefore, it seems possible that synaptotagmin VIII plays a role in vasopressin/cAMP-mediated regulation of transporter trafficking in the collecting duct and thick ascending limb via protein kinase A-mediated phosphorylation. For example, based on its localization in collecting duct principal cells and connecting tubule cells in the kidney, synaptotagmin VIII appears to be a potential mediator of vasopressin-induced exocytosis of aquaporin-2 water channels in the collecting duct principal cells. Whether synaptotagmin VIII is present in aquaporin-2 bearing vesicles has not yet been established.