Localization of Mg 2 -sensing shark kidney calcium receptor SKCaR in kidney of



免疫细胞治疗专区	欢迎关注干细胞微信公众号

12 3 4 5 6 7 8 9 10 ... 26 下一页

返回列表

查看: 547265\|回复: 258	go Localization of Mg 2 -sensing shark kidney calcium receptor SKCaR in kidney of

轻羽

新手上路

Rank: 1

积分: 0
威望: 0
包包: 0

发表于 2009-4-21 13:44 |显示全部帖子

作者：Hartmut Hentschel,, Jacqueline Nearing, H. William Harris, Marlies Betka, Michelle Baum, Steven C. Hebert, and Marlies Elger,作者单位：1 Max Planck Institute for Molecular Physiology,D-44229 Dortmund; 3 MariCal, Incorporated, Portland04101; 4 Children‘s Hospital, Boston, Massachusetts02115; 5 Yale University School of Medicine, New Haven,Connecticut 06520; 6 Department of Nephrology, HannoverMedical School, 30625 Hannover, Ge 8 V2 ]: C, Z% \: T& _/ j

   ; m1 r: A* i. ?/ `* R
      & ~% |% @7 I/ [* d) P

         ! K$ K0 N; O8 I- u. t6 n9 [

         & k$ K* ~, u8 }" p, l7 G
         , a, T0 q* ]* I5 P, `; b



      2 _8 x2 G% ^$ {
      【摘要】) y% _; j# ^5 V8 ~4 g9 L- o
   We recently cloned a homologue of the bovine parathyroid calcium receptorfrom the kidney of a spiny dogfish ( Squalus acanthias ) and termedthis new protein SKCaR. SKCaR senses alterations in extracellularMg 2  after its expression in human embryonic kidney cells (NearingJ, Betka M, Quinn S, Hentschel H, Elger M, Baum M, Bai M, Chattopadyhay N,Brown E, Hebert S, and Harris HW. Proc Natl Acad. Sci USA 99:9231-9236, 2002). In this report, we used light and electron microscopicimmunocytochemical techniques to study the distribution of SKCaR in dogfishkidney. SKCaR antiserum bound to the apical membranes of shark kidneyepithelial cells in the following tubular segments: proximal tubules (PIa andPIIb), late distal tubule, and collecting tubule/collecting duct as well asdiffusely labeled cells of early distal tubule. The highly specificdistribution of SKCaR in mesial tissue as well as lateral countercurrentbundles of dogfish kidney is compatible with a role for SKCaR to sense localtubular Mg 2  concentrations. This highly specific distribution ofSKCaR protein in dogfish kidney could possibly work in concert with thepowerful Mg 2  secretory system present in the PIIa segment ofelasmobranch fish kidney to affect recycling of Mg 2  from putativeMg 2  -sensing/Mg 2  -reabsorbing segments. These data provide support for the possible existence of Mg 2  cycling in elasmobranch kidney in a manner analogous to that described for mammals.
      【关键词】 renal handling of magnesium transmembrane receptor protein immunohistochemistry
   ----------3 E1 v# x( Q# y2 I/ ~$ X# i7 n! m" p9 R
, N, J3 d$ |) a, i! ^! k0 f
HOMEOSTASIS OF DIVALENT MINERAL ions in body fluids is sustained by the vertebrate kidney (for a review, see Refs. 14 and 32 ). Marine elasmobranchsabsorb constitutent ions of seawater and must excrete them to maintain ionichomeostasis ( 1, 36 ). In this regard,Mg 2  and Ca 2  are excreted principally by the kidney. Inspiny dogfish ( Squalus acanthias ), urine contains 3 mMCa 2  (a value almost identical to plasma), whereas Mg 2  values have been reported to reach 40 mM (plasma 1 mM)( 8 ). The steep gradient ofMg 2  concentration between plasma and urine demonstrates that thekidneys of marine elasmobranchs possess a powerful epithelial Mg 2  transport system ( 5 ), a majorcomponent of which is the second segment of the proximal tubule (PII), whereMg 2  secretion is thought to be performed ( 26, 37 ).

Previous work by our group has focused on a detailed characterization ofphysiologically relevant characteristics of the elasmobranch kidney( 12, 18, 23, 25 ). The excretory portion ofdogfish kidney consists of multiple, metamerically formed lobules that growtogether during organogenesis( 19 ). Each lobule possesses its own vasculature where renal arteries supply perfusion to glomeruli. Themultiple efferent arterioles merge with the sinusoid capillaries of the renalportal system, which, in turn, are drained via the cardinal veins to theheart. The tubules from each lobule are drained by a single large collecting duct. The lobules are separated into a mesial zone and a zone of lateralbundles. Each nephron is composed of tubular segments that travel in bothzones, forming two hairpin loops in the bundles and two extended convolutionsin mesial tissue. A schematic drawing of the anatomic organization of a singledogfish kidney nephron is shown in Fig.1.
6 u! c7 u5 c5 R% ^* @  k6 c1 m0 x
Fig. 1. Schematic diagram of dogfish nephron. Schematic presentation is shown ofthe nephron, collecting tubule-collecting duct system, and central vessel withreference to the renal zones of mesial tissue and lateral bundles. The courseof the nephron and the anatomy of the countercurrent bundle are greatlysimplified: the mesial convolutions are shown as single loops, and the tubularprofiles in the bundle are drawn apart from each other. In reality, thecollecting tubule is located near the central vessel and is simultaneously incontact with the 2 loops of the nephron. For more details as revealed by3-dimensional reconstruction of the bundle, see Ref. 25. The nephron segments (PIand PII) are indicated by various designs in black and white. The localizationof spiny dogfish kidney cation-sensing receptor (SKCaR) is indicated in lightgrey. a and b, Subsegments of PI and PII.
1 u) t# V: |" o7 z& H2 G
Individual nephrons in dogfish kidney possess well-developed proximal anddistal segments similar to those described for teleost kidney( 21 ). The proximal tubule issubdivided into two major portions that are designated PI and PII( 21 ). The epithelial cells ofPI are endowed with an elaborate apical tubulovesicular apparatus and anextended lysosomal compartment that are similar to those possessed by theentire proximal tubule of mammals (segments S1-S3). By contrast, PIIepithelial cells, which apparently have no counterpart in the mammalian kidney, possess an apical compartment filled with clear smooth vesicles thatcontain high concentrations of Mg 2  ( 26 ). Although presentevidence is very limited, data suggest that the PII segment of marineelasmobranchs may engage in net reabsorption of fluid( 36 ), although the PII segmentis capable of fluid and salt secretion when proximal tubules from teleosts and elasmobranches are incubated in vitro( 5 ). In dogfish, the luminalcontents of these proximal tubule segments are delivered to the early distaltubule (EDT; the homologue of the thick ascending limb of Henle's loop) and alate distal tubule (LDT; the homologue of the distal convoluted tubule inmammals) and, finally, to the collecting tubule (CT)/collecting duct (CD).
9 Y* r1 _! a+ s$ S) _+ n  B: T2 c
Because Mg 2  uptake in dogfish will vary when they migrate between seawater of different salinities as well as during periods ofexcessive feeding, Mg 2  excretion has to be balanced with alterations in Mg 2  uptake to maintain overall Mg 2  balance. However, the mechanisms that control either Mg 2  secretion in the PII segment or possible reabsorption in the distal tubule segments ofelasmobranch kidney are unknown. In this regard, it has been suggested on thebasis of histological evidence using 26 Mg 2  that renalMg 2  excretion in the euryhaline marine teleost, the killifish( Fundulus heteroclitus ), is the result of a two-step process whereproximal tubule cells secrete Mg 2  and Mg 2  reabsorptionoccurs in the CD/CT system( 9 ).. v" E7 }+ v( A2 _& x
5 w; H' X' @8 @% t: c( Y/ _4 A
Molecular cloning and characterization of the calcium/polyvalent cation-sensing protein (CaR) in the nephron segments of mammals have opened anew window to an understanding of renal handling of Mg 2  ( 14, 15 ). Antibody and cDNA probesderived from the sequence of CaRs cloned from mammals have been utilizedextensively to identify patterns of cell-specific CaR expression in multiple mammalian tissues ( 10, 31, 33 ). These studies havesuggested that CaRs possess the ability to "sense" localconcentrations of divalent cations and regulate transepithelial ion transport in response to such changes. CaRs are localized to specific cell types inmultiple tissues, where they serve as key integrators of divalent mineral ionhomeostasis in terrestrial mammals ( 7, 10, 14, 15, 31, 33 ). To determine whether CaRsserve similar roles in elasmobranch and teleost fish, we isolated a 4.16-kbshark kidney CaR (SKCaR) cDNA from a dogfish kidney cDNA library( 29 ). SKCaR is a 1,027-aminoacid (AA) protein possessing overall 74% AA identity with rat kidney CaR( 29 ). The shark kidneycontains two major SKCaR poly A transcripts of 7 and 4 kbthat are similar to those in the mammalian kidney ( 33, 34 ). A combination of RNAblotting and immunocytochemistry reveals significant SKCaR expression in othershark organs besides the kidney, including rectal gland, stomach, intestine, gill, olfactory lamellae, brain, and ovary( 29 ). Functional expression ofSKCaR protein in human embryonic kidney cells shows that it possesseshalf-maximal activation (EC 50 ) values for Ca 2  andMg 2  of 7.5 and 30 mM, respectively, under mammalian physiological ionic conditions( 3, 29 ). These data suggest thatSKCaR likely serves as a Mg 2  receptor in the shark kidney. In thisstudy, we hypothesized that the SKCaR protein might possess a highly specificpattern of cellular expression, possibly reflecting its role as an ion sensorin the shark kidney. Using SKCaR-specific antiserum, we report here that SKCaRexhibits a highly specific subsegmental nephron distribution in the shark kidney that is compatible with a role as a principal regulatory sensor forMg 2  homeostasis in elasmobranch kidney.

MATERIALS AND METHODS, W. s" L- @+ l8 p
0 F. g$ g( a2 C* }
Animals. Male dogfish ( S. acanthias ) were captured bylocal fishermen in Frenchman Bay for the Mount Desert Island Biological Laboratory during July and August. Ten fish were kept overnight (12-24 h) inlarge (2,000 liter) tanks with running aerated seawater (average temperature15°C) before use. Alternatively, four fish were maintained for 1 wk in thetanks. Because the dogfish were wild-caught and did not feed in captivity, it was not possible to determine exactly when they last fed and the nature oftheir last meal, the recent site of their occupancy in the ocean, or detailsof their lives immediately before capture.

Tissue preparation. After anesthesia with tricaine (MS 222, Sigma), dogfish were perfused via the heart and truncus arteriosus withheparinized dogfish Ringer solution (in mM: 280 NaCl, 6 KCl, 3MgCl 2, 5 CaCl 2, 0.5 Na 2 HPO 4, 1.0NaH 2 PO 4, 330 urea, 5 glucose, 72 TMAO, and 8NaHCO 3 in 1 liter). The measured osmolality of this solution was1,000 ± 30 mosM. Approximately 400-500 ml were perfused ( 5-10 min)at a temperature of 0-4°C and a pressure of 120 cmH 2 O. Withouta change in pressure and flow, the fixation fluid was added and perfused for5-10 min./ C+ w# x# W2 o
, s6 H) P: u  N# h8 p
The fixation fluid contained 2% formaldehyde freshly prepared fromparaformaldehyde, 0.05% glutaraldehyde, and 0.5% picric acid in Sorensen'sphosphate buffer, pH 7.4. Sucrose was added to the buffer vehicle (Sorensen'splus 150 mM NaCl) to obtain a blood hyposmotic value of 800 mosM. Afterfixation, small tissue pieces were excised from the organ and thoroughlyrinsed in Sorenson's buffer plus 150 mM NaCl, which was adjusted with sucroseto 850 mosM.3 e6 i2 p  w% ]  l7 F
" }% u, z$ C( G
Small pieces of tissue were processed by four different methods: sampleswere 1 ) postfixed with 1% OsO 4 in Sorensen's buffer, dehydrated via ethanol and acetone, and embedded in Spurr's medium; 2 ) embedded in OCT compound, shock-frozen in melting isopentane, andstored in liquid nitrogen; 3 ) dehydrated via ethanol and xylene andembedded in Paraplast (56°C); and 4 ) dehydrated via ethanol andembedded in LR-White resin.

Histology. Sections (0.5 µm) and thin sections (60 nm) wereobtained from tissue blocks (Spurr's embedding medium) and viewed with a lightmicroscope after being stained with toluidine blue or with an electronmicroscope (Zeiss EM 902 or Philipps EM 301) after being stained with uranylacetate and lead citrate. The nomenclature of renal structures, i.e., nephronsegments, blood vessels, and interstitial cells, was used in accordance withprevious results with spiny dogfish and other marine elasmobranch fish( 12, 18, 20, 25 ). In addition tostructures, which are generally characteristic of the elasmobranch kidney,spiny dogfish feature a specific subdivision of the second segment of theproximal tubule PII (PIIa and PIIb)( 12 ).$ s+ c. |9 ]+ A
. K* _3 h6 R! a* y
Antibody preparation. A 17-mer peptide (ARSRNSADGRSGDDLP C for COOH-terminal conjugation), corresponding to 965-980 of the putative 1,027-AASKCaR polypeptide, was synthesized by standard automated solid-phasetechniques, conjugated to keyhole limpet hemocyanin via a cysteine sulfhydryllinkage, and used to immunize rabbits as reported previously( 29 ). Test bleedings werescreened by immunoblots against shark kidney tissue.( g6 L1 v: R% f
8 I8 i8 e/ M1 {$ B: Q! I
For selected experiments, SKCaR immune antiserum was affinity purified asdescribed previously ( 30 ).Purified immune IgG fractions were absorbed to a column containing covalentlyattached peptide conjugated via a 5-thio-2-nitro-benzoic acid-thiol-agarose linkage. After extensive washing, the purified anti-SKCaR antiserum was elutedat pH 2.5, followed by rapid titration to pH 8.0. Both raw andaffinity-purified antisera were stored at -80°C in multiple aliquots.

Immunoblot analyses. To provide further evidence for thespecificity of the antiserum for SKCaR in addition to comparisons ofimmunoblots and immunocytochemical sections exposed to immune antiserum vs.preimmune serum as previously reported( 29 ), immunoblots containingkidney membranes were probed using affinity-purified antibodies before andafter their preincubation with a 100-fold molar excess of competing peptide( 30 ). In all blots, weroutinely observed that the dye front of all lanes containing shark tissue fractions displayed reactivity (see Fig.2 ) that was not ablated by preincubation with an appropriatepeptide. This apparent reactivity of the dye front was due to binding ofsecondary antiserum because dye front reactivity was present without the addition of primary anti-SKCaR antiserum (either affinity purified, immune, orpreimmune) but not when secondary antiserum was omitted (data not shown).

Fig. 2. Specificity of anti-SKCaR antibody. A : tissue section near tip ofa bundle in dogfish kidney. Note presence of reddish reaction product denotingthe presence of bound antibody. B : adjacent section to A after incubation with preimmune serum. The chromogenic reaction with thedetection system is absent. Asterisks present in the lumens of 2 identicaltubules immediately adjacent to each other provide section orientation. C : immunoblot of shark kidney membranes with affinity-purifiedantiserum. Representative immunoblot containing 40 µg of kidney membranesshows 3 major immunoreactive bands (Immune) of 240, 140, and 91 kDa. Incontrast, inclusion of a 100-fold molar excess of immunizing peptide withprimary antiserum results in the ablation of these bands (Immune Peptide). TG,top of gel; DF, gel dye front. Note that apparent immunoreactivity present atthe DF of both lanes derives from binding of secondary goat anti-rabbitantiserum and not anti-SKCaR antibody (data not shown).. l& J6 |+ S& k6 c2 q* Z
7 ]: p5 {! @/ I7 V) a) }
Immunohistochemistry. Cryosections (4 µm) and Paraplast sections (2-4 µm) were treated with 0.1% H 2 O 2 andused for indirect immunolabeling. After the blocking of unspecific bindingwith a mixture of 0.2% coldwater fish gelatin, 2% BSA, and 2% fetal calfserum, sections were incubated with primary antiserum (SKCaR antibody; seeabove), and specific antibody binding sites were revealed with theimmunoperoxidase technique, involving biotin-streptavidin amplification (ABCElite Kit, Vector Laboratories, Burlingame, CA) with methyl green as a counterstain. Routine controls were performed 1 ) with the omission ofprimary antiserum and 2 ) with incubation with preimmune serum.Permanent mounts (Gelmount) were viewed and photographed with an Axiophotlight microscope (Carl Zeiss, Göttingen, Germany).
- s- m7 L  O( R( ]8 W
Immunoelectron microscopy. Thin sections (60 nm) were obtained from LR-White blocks with an ultramicrotome (Ultracut E, Leica) and incubatedwith anti-SKCaR antibody, followed by anti-rabbit IgG-colloidal gold conjugate(10-nm gold particles, Aurion-Gent). The sections were counterstained withuranyl acetate and viewed with Zeiss EM 902 and Philipps 301 electronmicroscopes.$ n9 l& H& a6 [) V! B
6 u( D$ [" X4 @0 p
RESULTS# p) j7 h5 I$ c6 p5 s% V  P9 p
  g+ P' I! |+ q. b9 I8 u* n
Specificity of anti-SKCaR antiserum. Chromogenic reaction product from bound antibody was present in dogfish kidney sections exposed to immuneanti-SKCaR antiserum but not preimmune serum( Fig. 2, A and B ). Anti-SKCaR antiserum labeled the membranes of epithelial cellsand renal tubule cells of distinct nephron subsegments as well as thecytoplasm of selected interstitial cells. By contrast, glomeruli displayed noimmunoreactivity. The pattern of anti-SKCaR immunoreactivity in all nephronsubsegments was independent of either tissue fixation or preparation of cryosections, paraffin sections, and thin sections of LR-White blocks. Foreach of the nephron segments described below, labeling by immune anti-SKCaRantiserum was present, whereas no chromogenic product was present afterexposure to corresponding preimmune serum. A summary of labeling by anti-SKCaRantibody is provided in Table1.& w# K4 }1 l2 u  O: J* G
% @5 \1 y* I5 Z) h
Table 1. Immunoreactivity of SKCaR in the kidney of Squalus acanthias6 x7 j; x6 W) T* W( G  e
" \; s* D3 B, m4 K8 g4 a
When immunoblots containing crude membranes isolated from dogfish kidneywere probed with affinity-purified anti-SKCaR antibody, prominent bands of240, 140, and 91 kDa were present ( Fig. 2 C ). These bands were completely ablated afterpreincubation with an excess of competing peptide. These bands displaymolecular masses similar to those reported previously for SKCaR protein expressed in human embryonic kidney cells( 29 ) as well as anti-CaR-reactive proteins present in a variety of mammalian tissues( 33, 34 ).

Anti-SKCaR labeling of specific nephron subsegments. Anti-SKCaR antibody labeled specific nephron subsegments in both mesial tissue andlateral bundle zones of dogfish kidney( Fig. 3 ). The PIa segment inthe lateral bundle zone displayed apical SKCaR staining within the region ofits brush border ( Fig. 3 andsee Fig. 6 ). The PIb segment,which displays multiple bands within mesial tissue near the glomeruli, showedSKCaR-specific staining only at the base of the microvilli of a few cells ( Fig. 3 ).% ~  {7 h4 A) R- \# ]: D" U

Fig. 3. Part of a cross section through the kidney. Mesial tissue displays largeprofiles of PIIa with luminal brush border, small profiles of PIIb withluminal brush border and cilia of multiciliary cells, and small profiles oflate distal tubules (LDT). The small profiles (PIIb and LDT) showimmunoreactivity (brownish red) at the luminal side (apex of epithelialcells). A large glomerulus (GL) exhibits close contact between the collectingtubule (CT) and afferent arteriole (AA). The CT at this vascular field of theglomerulus was labeled by chromogenic reaction. In the lateral bundle zone( left ), several early distal tubules (EDT) show immunostaining. PIa,CT, and collecting ducts (CD) in the bundle zone show marked immunostaining ofthe apical cell region. Intermediate segment (IS) and central vessel (CV) areonly stained by counterstain methyl green.
# a/ d8 o; A) A3 t# U( q0 P4 g$ [
Fig. 6. Cross section through a countercurrent bundle. The bundle is sectioned nearthe tip, where the CD leaves and a small CT is entering, coming from aneighbouring bundle (see also Ref. 25 ). The apical cell membraneof proximal tubule PIa cells (first hairpin loop) is labeled with immunostain(brownish red). Strong binding occurs at the apex of CT and CD cells. EDT ofthis profile reacts with antibody along the "intracellularstriations," which represent amplifications of the basolateral cellmembrane. IS, bundle vein (BV), CV, and bundle artery (BA) appear negative forSKCaR.

The PIIa and PIIb segments of the proximal tubule present in mesial tissuedisplayed markedly different patterns of anti-SKCaR staining characteristics(Figs. 3, 4, 5 ): PIIa cells exhibited noSKCaR staining. In contrast, PIIb cells of all 14 animals studied displayedspecific SKCaR immunoreactivity that was observed at their apical membranes(brush border). Interestingly, the intensity of SKCaR labeling of PIIb apicalmembrane varied greatly among various individual animals studied despite the fact that consistent SKCaR labeling was observed in most other shark nephronsegments (see below).

Fig. 4. Cross sections of segments in mesial tissue. Immunostaining of LTD revealsdistinct binding sites of antiserum against SKCaR at the apical cell membrane(red). A faint staining can be seen at the basolateral cell membrane forming"intracellular striations." PIIa shows no reaction. SC, sinuscapillaries of the renal portal system.
) E0 O* d' Q( j, u9 v
Fig. 5. Section through mesial tissue in the vicinity of glomeruli. Proximal tubulesegment PIb with distinct brush border and the 2 portions of the secondproximal tubule, PIIa and PIIb, are shown. These segments are in close contactwith the LDT. In this animal, pronounced staining (red) for SKCaR is confinedto the LDT and a few portions of the brush border of PIIb (arrow).

The EDT, which is present exclusively in the lateral bundle zone, iscontiguous with the LDT, which thereafter performs numerous bands in mesialtissue. EDT cells were diffusely labeled by anti-SKCaR antiserum (Figs. 3 and 6 ). The LDT is present inmesial tissue, where it courses along the pathway of PIIa tubules and isfrequently in close proximity to both PIIa and PIIb cells (Figs. 3, 4, 5 ). LDT cells in all animalsexamined showed sharply defined SKCaR staining at their apical cell membranes. In addition, only very weak immunostaining was observed at the LDT basolateralmembrane in two animals ( Fig.4 ).
5 Y% F/ Z) T! {0 R
Electron microscopy of LDT cells revealed that they possess short, stubbymicrovilli with a marked asymmetry of the apical cell membrane, where theextracellular (luminally facing) side was thickened and had a fuzzy coat(glycocalyx). Immunoelectron microscopy of the apical region of LDT cellsshowed that anti-SKCaR immunoreactivity protein was 1 ) in theimmediate vicinity of the cell membrane; 2 ) in the apical cytoplasmicregion, presumably at apical vesicles; 3 ) associated withmembrane-bound granules located in close proximity to the apical membrane; and 4 ) outside the cell in the glycocalyx( Fig. 7 ).

Fig. 7. Electron micrograph of thin section with immunogold staining of apicalregion of the LDT. Numerous gold particles (10 nm) are present at the cellmembrane, in the fuzzy coat, at a large granule in close proximity to theapical membrane (arrow), and at small apical vesicles, indicating a largeamount of SKCaR antigen.
# C9 M6 [4 y: D" E% G
Significant anti-SKCaR immunoreactivity was observed in the CT as well asin the CD (Figs. 3 and 6 ). The CT at the vascular field of the glomerulus was labeled by the chromogenic reaction. In CT and CDcells, SKCaR antibody binding was confined to the region of the apical cellmembrane and its adjacent cytoplasmic zone, where membrane-bound granulesabound.9 g5 f0 Y0 V$ U; e
' k7 H: V: ]# K4 s) V- a
SKCaR staining was also observed in the cytoplasm of small, round cellswith large spherical nuclei that were arranged in islets in the interstitiumof the lateral bundle zone ( Fig.3 ). These cells belong to the renal lymphomyeloid tissue that isinvolved in hematopoiesis of elasmobranch fish( 22 ). Although the function ofthese cells is presently unknown, they may correspond to hematopoietic cellsthat possess CaR proteins in mammals( 15 ).# r7 y3 O4 j; G9 q; V% N

In summary, we consistently found SKCaR labeling in nephron segments PIa,PIb, PIIb, EDT, the apical membrane of the LDT, the CT/CD system, and in asubpopulation of cells of the interstitial tissue. However, we observed thatSKCaR reactivity was less pronounced in PIIb of four animals, and with theexception of two animals, the basolateral membrane of LDT was not labeled.

DISCUSSION

As summarized in Table 1 andschematically in Fig. 1,significant antibody binding against SKCaR was present on the apical cell membranes of the following epithelial cells of dogfish kidney: 1 ) inthe very early portions of PIa at the first hairpin loop in the countercurrentbundles, 2 ) at the end of PIIb in mesial tissue, 3 ) in theLDT in mesial tissue, 4 ) in CT at the vascular field of theglomerulus and inside the countercurrent bundles, and 5 ) in CD.Moderate binding was observed in several cells of PIb. The EDT was alsodiffusely labeled by anti-SKCaR antibody. It is noteworthy that, inelasmobranchs, cells of the EDT possess extensive amplifications of theirbasolateral membranes where they form extensive lateral interdigitationsrunning from the cell base to the apical cell junctions( 20, 21, 23 ). Therefore, it is likelythat the diffuse labeling pattern observed in dogfish may be due to SKCaRlabeling of these extensive basolateral membrane amplifications. However,careful immunoelectron microscopic studies will be necessary to firmlyestablish this possible SKCaR subcellular localization.  m( a( ^! w8 ^. e3 ?
$ L$ @/ E/ `7 Y; Z9 A
The reason for the variability of anti-SKCaR staining that we observed inthe PIIb and LDT tubular segments is not presently known. We speculate thatthis variability could possibly result from wild-caught animals that were indifferent stages of acclimatization to either captivity or conditions beforebeing captured for study. Further studies are also required to more carefully define any physiological variables that might contribute to differences inSKCaR staining in these tubular segments.2 b8 {4 U4 Y% Y$ W, o/ J; h

The significance of the observed pattern of SKCaR, a Mg 2  sensor, for the handling of Mg 2  by marine fish will be discussedbelow, with particular reference to the elaborate organization of the elasmobranch kidney. Due to the paucity of transport studies in theelasmobranch nephron ( 17, 37 ), these specific SKCaR staining patterns provide the basis for us to propose a functional model ofMg 2  homeostasis in the elasmobranch kidney that highlights specific aspects of putative recycling of Mg 2  betweenMg 2  -sensing and/or -reabsorbing cells andMg 2  -secreting cells of the shark's multisegmental nephron. Inturn, this model provides a series of testable hypotheses that form the basisof future experiments directed at a more detailed understanding of divalentmineral transport and recycling by the elasmobranch kidney.

Overall renal handling of Mg 2  in marine fish. Althoughrenal Mg 2  excretion is essential for the survival of elasmobranchs in the marine environment because seawater contains a high concentration ofMg 2  ( 50-60 mM) relative to their plasma Mg 2  (plasma 1-3 mM), functional data quantifying the specific contribution of various nephron segments to this process of net Mg 2  excretion arescarce ( 17, 37 ). The few measurements of S. acanthias kidney show that urinary flow in conscious, restrainedspiny dogfish is low ( 0.3ml·h -1 ·kg -1 ), ; K# |4 a5 T* q3 p& y: ?
: ~# u0 z. V" ?
Mg 2  homeostasis and functional evidence from the proximal and distal tubules. Secretion of Mg 2  occurs in isolateddogfish proximal tubules ( 35 ).In previous work, we used quantitative transmission electron microscopy todemonstrate that Mg 2  is sequestered at high concentrations withinsecretory apical vesicles of PII cells in European dogfish ( Scyliorhinuscaniculus ) ( 26 ).Micropuncture studies ( 37 ) ina related elasmobranch, the little skate ( Raja erinacea ), show thatthe final urinary Mg 2  concentration is 10-fold higher comparedwith its plasma value. Transtubular concentration differences (TF/P) forMg 2  reveal that large increases in TF/P Mg 2  occur inPII, suggesting that significant Mg 2  secretion occurs within thissegment. Mg 2  transport in elasmobranchs also appears to be verysimilar to Mg 2  secretion in the proximal tubule of two othermarine and euryhaline teleost fish species, the flounder Pleuronectes americanus and the killifish ( F. heteroclitus )( 11 ). Transport kinetics withisolated tubules show saturation far below plasma concentration, suggestingthat Mg 2  secretion by proximal tubules cannot alone be heldresponsible for the tight regulation of Mg 2  concentration in theplasma of marine fish. Accordingly, Beyenbach and co-workers( 5, 6 ) have suggested that proximal tubules in marine fish can be considered to work as devices for clearance ofMg 2  from blood akin to the clearance of plasma solutes byglomerular filtration.- ~+ s. [% Q# y2 [5 `( t
% G) @- a8 R: K
Evidence for Mg 2  reabsorption in distal tubule segments of marine elasmobranchs is very limited. TF/P values obtained from micropuncturestudies of proximal segments vs. CD and final urine of the little skatesuggest that more distal nephron segments may be involved in both tubularreabsorption and secretion of Mg 2  ( 37 ). Although studies ofisolated distal tubules of S. acanthias have revealed greatsimilarity to the diluting segment of mammals and other vertebrates( 13, 16 ), no studies ofMg 2  transport have been performed to quantify Mg 2  transport in this dogfish nephron segment.
. v; D' b' O. y
Selected aspects of elasmobranch kidney morphology relevant toMg 2  homeostasis. The morphological investigations of the kidneys of Chondrichthyes (sharks, skates, and chimeras) have uncovered anextremely complex organization (for references and reviews, see Refs. 12, 18, 20, 25, and 28 ). Previous comparativeanatomic studies in elasmobranchs have revealed the presence of nephronsegments that are homologous to corresponding segments in other vertebrates( 20 ). Moreover, it isinteresting to note that proximal and EDT segments in kidneys of both mammals and elasmobranchs are spatially separated and their renal tissue is zonated.These patterns of kidney tissue zonation are present in many othervertebrates, including agnathean lampreys, archaic fish such as Polypteridaeand lungfish, and amphibian and higher vertebrates (Sauropsida and Mammalia)( 21 ), and might have originated early in vertebrate evolution.

Thus despite the fact that mammalian and cartilaginous fish kidneys areorganized differently, it is intriguing that the kidney of spiny dogfishpossesses two renal zones that superficially resemble those of the mammalianrenal cortex and medulla, respectively. Mesial tissue of dogfish kidneycontains a close association of proximal tubules (segments PIb, PIIa, andPIIb) as well as LDT. This complement of dogfish nephron segments present inmesial tissue appears to be similar to the mammalian renal cortex with itsassembly of proximal tubules, distal convoluted tubule, CT, and cortical CD.By contrast, the lateral bundles of dogfish kidney bear a similarity to themammalian renal medulla in that the EDT and the CT/CD of dogfish kidney are specially associated like that of the thick ascending limb of Henle's loop andmedullary CD of the mammalian kidney. As discussed below, these anatomicassociations might be functionally relevant in renal Mg 2  handlingin the elasmobranch kidney in a manner that is important, as specific nephronsegments appear to do so in the mammalian kidney.

Functional considerations for SKCaR localization in individual elasmobranch nephron segments. The specific distribution pattern of SKCaRprotein in individual cell types present in dogfish nephron segments suggestsa unifying hypothesis whereby Mg 2  excretion in dogfish iseffectively regulated by SKCaR, which senses local Mg 2  concentrations. Based on the combination of our previous research on thestructural features and Mg 2  transport of elasmobranch nephronsegments, SKCaR localization data presented here, and published reports ofCaRs in mammalian kidneys and cells, we can speculate on the performance ofthe different segments in dogfish kidney as described below.( y" g% M8 F% F$ V2 i0 m8 I
& F9 j6 q9 e7 x- w% C8 o8 ~! r
The SKCaR protein localized to the apical membrane of dogfish kidneyproximal tubule segments PIa and PIb may be involved in the modulation ofvarious cellular transport functions, including transepithelial fluxes ofdivalent cations as well as perhaps modulating proton secretion. Thisinteresting possibility is suggested by studies of CaR function in ratproximal tubule, where it has been suggested that CaRs might regulate proton secretion and other ion fluxes( 33 ). In this regard,elasmobranch nephrons are well known to be capable of robust proton secretion, resulting in a low pH value of 4 that is maintained throughout the renaltubule. Urinary acidification begins as early as the PIa segment in thecountercurrent bundle according to intravital microscopy( 27 ), and low pH is thought tobe important in preventing stone formation within the elasmobranch nephron, especially the CT/CD system, where high concentrations of Mg 2  salts exist.

PIIa cells likely perform the primary step in Mg 2  excretion, yet they do not contain SKCaR protein. PIIa is an exceptionally long segmentin mesial tissue of marine Chondrichthyes and provides a significant mass ofspecialized large epithelial cells. Without the presence of SKCaR, extractionof Mg 2  from the surrounding peritubular circulation may possiblybe regulated by local Mg 2  availability. Alternatively, PIIa cellfunction could be modulated by cross talk with adjacent SKCaR-containing cellssuch as PIa or PIIb cells located in neighboring segments. Future experimentswill be necessary to distinguish between these possibilities or possibleregulation of PIIa Mg 2  secretion by autocoids.0 w3 {; w0 `5 `% ?% y

Present studies of mammalian distal tubule have emphasized the importanceof CaRs to modulate cellular transport of both divalent (Mg 2  andCa 2  ) as well as monovalent (Na ) cations. Hebert ( 14 ) has summarized thesalient features of the regulation of thick ascending limb function by CaRs.Bapty and co-workers ( 2 )further defined renal Mg 2  handling by the study of cultured immortalized mouse distal convoluted tubule cells that possess anextracellular polyvalent cation-sensing mechanism responsive toMg 2 , Ca 2 , and neomycin. In mammals, distal tubulesreabsorb 15% of filtered Mg 2  ( 32 ). Thus these data suggestthat mammalian distal convoluted tubule exhibits a renal cell type that bothsenses and transports Mg 2 .- D( M$ r& B5 S) }; O! r; ?6 G8 r
0 K. T5 H9 k& m) @$ \' ]7 N; K  d
It is intriguing to note that the diffuse distribution of SKCaR in sharkEDT cells (perhaps localized to the basolateral membrane) and apicallocalization in shark LDT might correspond to the presence of CaR proteins onthe mammalian medullary thick ascending limb and distal convoluted tubule( 33 ). By analogy to the function of the thick ascending limb of Henle's loop and distal convolutedtubule in rats, we suggest that EDT and LDT are very likely segments whereMg 2  is reabsorbed in dogfish. Moreover, Riccardi et al.( 33 ) have reported thepresence of apical punctuate CaR antibody staining in some type A intercalated cells in rat CD. Interestingly, dogfish LDT cells display cytoplasmic studsand apically located H -K -ATPase, which arecharacteristically found in mammalian type A intercalated cells( 23 ).
  L  A2 f1 l8 k- T) w+ |
In a manner similar to the mammalian inner medullary CD( 10, 34 ), the apical membrane ofthe dogfish kidney CT/CD system stains intensely with anti-SKCaR antibody.While the role of SKCaR in dogfish CT/CD is uncertain, it may modulateNaCl-coupled water reabsorption, as has been proposed for the flounder urinary bladder ( 29 ), and preventexcessively high concentrations of Mg 2  within shark urine, as hasbeen proposed for both flounder bladder( 29 ) and mammalian innermedullary CD ( 10, 34 ). The overall hypothesisoutlined above can be tested by multiple experimental paradigms. A combinationof subcellular immunolocalization efforts together with detailed transportstudies of both divalent and monovalent ions including protons will befacilitated by the labeling pattern of SKCaR and together provide a better understanding of divalent cation metabolism in elasmobranch and teleost fishas well as Mg 2  wasting and stone formation in mammals.
' E% ]6 n' C' x' ~
DISCLOSURES, \: e0 q% ?3 Y" C* b' l

The study was partially funded by Max-Planck-Gesellschaft and DeutscheForschungsgemeinschaft (El 92/6.).* f' i9 U0 _7 O% F+ F

H. Hentschel was the recipient of a New Investigator Award of the MountDesert Island Biological Laboratory, Salsbury Cove, ME.
3 A9 B' e5 D/ A1 p
A portion of the results has been presented at the 31st Annual Meeting ofthe American Society of Nephrology, Philadelphia, PA, 1998, and was publishedin abstract form ( 3, 24 ).
) r" t( ?0 p& t! z
ACKNOWLEDGMENTS# U, w9 B+ e: U9 ?4 ]. b

We thank C. Pieczka and G. Schulte, ZE-DOC Max Planck Institute forMolecular Physiology, for expert help with the photographic work. F. Draegerprovided the graphical artwork.
      【参考文献】( l3 D/ Z" k$ x: P7 o* @5 o! ^( F
Anderson WG,Takei Y, and Hazon N. Osmotic and volaemic effects on drinking rate inelasmobranch fish. J Exp Biol 205: 1115-1122,2002.7 B5 A+ K2 g5 ^  p* ?1 O

Bapty BW, DaiLJ, Richie G, Canaff L, Hendy GN, and Quamme GA. Mg 2  /Ca 2  sensing inhibits hormone-stimulatedMg 2  uptake in mouse distal convoluted cells. Am JPhysiol Renal Physiol 275:F353-F360, 1998.+ Y" D/ I5 O3 x4 _
% }# E8 S" G+ Q/ k5 a9 ]) j$ @, w

Baum MA, FloresF, Elger M, Hentschel H, Brown EM, Hebert SC, and Harris HW. An apicalextracellular calcium/polyvalent cation sensing receptor (CaR) present in theosmoregulatory organs of saltwater (SW) and freshwater (FW) fish likely playsa role in salinity adaptation (Abstract). J Am SocNephrol 7: 1276,1996., v# B1 m  s: k' W0 }

Benyayati S andYokota S. Renal effects of atrial natriuretic peptide in a marineelasmobranch. Am J Physiol Regul Integr Comp Physiol 258: R1201-R1206,1990.
  B6 \# Z! r' ~' k) w
  [3 `) x; G3 n
$ \% |0 p# W6 Z% @
Beyenbach KW. Secretory electrolyte transport in renalproximal tubules of fish. In: Cellular and Molecular Approaches toFish Ionic Regulation, edited by Wood CM and Shuttleworth TJ. NewYork: Academic, 1995.4 e; R) e2 `" u/ L

3 z) D/ B6 X6 y0 ~4 {
Beyenbach KW,Freire CA, Kinne RKH, and Kinne-Saffran E. Epithelial transport ofmagnesium in the kidney of fish. Miner ElectrolMetabol 19:241-249, 1993.8 ]  e. Y/ I* f7 J

# L+ c7 n& M( q5 T& q+ L- J
Brown EM, GambaG, Riccardi D, Lombardi M, Butters R, Kifor O, Sun A, Hediger MA, Lytton J,and Hebert SC. Cloning and characerization of an extracellular Ca-sensingreceptor from bovine parathyroid. Nature 366: 575-580,1993.. R+ ]' R  o7 ^# c% l

0 n6 `" p/ n3 i
Burger JW. Problems in the electrolyte economy of the spiny dogfish, Squalusacanthias. In: Sharks, Skates and Rays, edited byGilbert PW, Mathewson RF, and Rall DP. Baltimore, MD: Johns Hopkins,1967.
, H5 r* Y9 X) t  ]$ ]

- g& M$ g4 q6 e$ t6 U7 L+ \
Chandra S,Morrison GH, and Beyenbach KW. Identification of Mg-transporting renaltubules and cells by ion microscopy imaging of stable isotopes. AmJ Physiol Renal Physiol 273:F939-F948, 1997.

# a9 Z( d  q/ @3 a/ ?. R
% m" l, E8 w7 ~+ Y. D# c
Chattopadhyay N, Baum M, Bai M, Riccardi D, Hebert SC, Harris HW,and Brown EM. Ontogeny of the extracellular calcium-sensing receptor inrat kidney. Am J Physiol Renal Fluid ElectrolytePhysiol 271:F736-F743, 1996.
5 k7 Q# L1 x* B

Cliff WH,Sawyer DB, and Beyenbach KW. Renal proximal tubule of flounder. II.Transepithelial Mg secretion. Am J Physiol Regul Integr CompPhysiol 250:R616-R624, 1986.
7 o6 q7 E# [+ |4 d- m6 h2 m
2 H  \; D3 c. t8 S/ j" t

Elger M andHentschel H. Microdissection of kidney zones and renal tubule segments of Squalus acanthias, with emphasis on renal histology and tubulesegmentation of Elasmobranchs. Bull Mount Desert Isl BiolLab 32: 24-27,1993.

5 o( Y; w& |0 x' Q
Friedman PA andHebert SC. Diluting segment in kidney of dogfish shark. I. Localizationand characterization of chloride absorption. Am J Physiol RegulIntegr Comp Physiol 258:R398-R408, 1990.

: r  o! f  `2 i0 |' U! B1 G9 B

Hebert SC. Extracellular calcium-sensing receptor: implications for calcium and magnesiumhandling in the kidney. Kidney Int 50: 2129-2139,1996.- v' Y* Y/ ]9 l* d, m' }2 e
( `) n( B0 o" s

+ X+ ^3 F" \! d' t8 v9 R
Hebert SC,Brown EM, and Harris HW. Role of the Ca 2  -sensing receptor indivalent mineral ion homeostasis. J Exp Biol 200: 295-302,1997.

! [& f' N' t- @* y$ L
4 a' K& Z" k7 U
Hebert SC andFriedman PA. Diluting segment in kidney of dogfish shark. II.Electrophysiology of apical membranes and cellular resistances. AmJ Physiol Regul Integr Comp Physiol 258:R409-R417, 1990.2 g0 ~- Q& J+ _: C6 c

( t" E7 ]) A7 w/ `) V6 z1 j

Henderson IW,O'Toole BO, and Hazon N. Kidney function. In: Physiology ofElasmobranchs, edited by Shuttleworth TJ. New York: Springer,1988.9 `& e- N0 x8 c
3 G' i' B0 \5 E0 u0 c+ A
: R8 B% h% q+ @! E" f5 Q6 ?

Hentschel H. Renal blood vascular system in theelasmobranch, Raja erinacea Mitchill, in relation to kidney zones. Am J Anat 183:130-147, 1988.
. W; ^' ^8 f+ |( y* y4 k
) [& n% Z- V, S, [" z

Hentschel H. Developing nephrons in adolescent dogfish, Scyliorhinus caniculus, with reference to ultrastructure of earlystages, histogenesis of the renal countercurrent system and nephronsegmentation in marine elasmobranches. Am J Anat 190: 309-333,1991.
! n8 M/ t6 H' U& ^# S! b7 V

Hentschel H andElger M. The distal nephron in the kidney of fishes. Adv AnatEmbryol Cell Biol 108:1-151, 1987.

* w/ _9 f* {* J8 I- B
Hentschel H andElger M. Morphology of glomerular and aglomerular kidneys. In: Structure and Function of the Kidney, edited by KinneRKH. Basel: Karger, 1989.
+ T$ @0 Z9 }  A* ~! h

7 w! D% t6 _3 @1 m, x: @
Hentschel H andElger M. The kidney of Squalus acanthias contains lymphomyeloidtissue. Bull Mount Desert Isl Biol Lab 40: 112-113,2001." e  m* Y# ^  H
5 ]9 U% q5 Z7 J1 S1 }0 a* o  j4 f" m
7 X, o6 q$ i5 r2 T) U! f8 Z8 X4 t

Hentschel H,Mähler S, Herter P, and Elger M. Renal tubule of dogfish, Scyliorhinus caniculus : a comprehensive study of structure withemphasis on intramembrane particles and immunoreactivity forH -K -adenosine triphosphatase. AnatRec 235: 511-532,1993.0 T; _  I. e' o
% m4 s! b7 i2 O

$ B+ s, g1 E5 V
Hentschel H,Nearing J, Harris HW, and Elger M. Cellular localization ofcalcium/polyvalent cation receptor protein SKCaR in the kidney of spinydogfish, Squalus acanthias (Abstract). J Am SocNephrol 9:554A-555A, 1998." L" Q+ W. f+ T, W& G  v% c& \

Hentschel H,Storb U, Teckhaus L, and Elger M. The central vessel of the renalcountercurrent bundles of two marine elasmobranches-dogfish( Scyliorhinus caniculus ) and skate ( Raja erinacea )-asrevealed by light and electron microscopy with computer-assistedreconstruction. Anat Embryol 198: 3-89,1998.; v) j5 D3 Y! W# g1 f( i. m3 e8 B
( m( s; ?9 w' K1 X+ Q. ]
: J8 |, u- z& t. ]( U! L

Hentschel H andZierold K. Morphology and element distribution of magnesium-secretingepithelium: the proximal tubule segment PII of dogfish, Scyliorhinuscaniculus (L.). Europ J Cell Biol 63: 32-42,1994.
6 d' G, ~( f  L8 r, ]

Kempton RT. Studies on the elasmobranch kidney. I. The structure of the renal tubule ofthe spiny dogfish ( Squalus acanthias ). JMorph 73:247-263, 1943.

: I: Y; e" j$ L% E
Lacy ER andReale E. Functional morphology of the elasmobranch nephron and retentionof urea. In: Cellular and Molecular Approaches to Fish IonicRegulation, edited by Wood CM and Shuttleworth TJ. New York:Academic, 1995.
' \( H3 e$ {: N+ B. w
2 |) T5 u0 f& {! m$ j) S

Nearing J,Betka M, Quinn S, Hentschel H, Elger M, Baum M, Bai M, Chattopadyhay N, BrownE, Hebert S, and Harris HW. Polyvalent cation receptor proteins (CaRs) arethe salinity sensors in fish. Proc Natl Acad Sci USA 99: 9231-9236,2002., N; Y8 Y0 x& {- [/ u
( A) e8 x$ E2 F0 c9 F4 h1 C. E

, S" i5 ]0 _* d+ A0 M3 @7 |8 `1 L9 U: G
Nielsen S,DiGiovanni SR, Christensen EI, Knepper MA, and Harris HW. Cellular andsubcellular immunolocalization of vasopressin-regulated water channel in ratkidney. Proc Natl Acad Sci USA 90: 11663-11667,1993.

Pollak MR,Brown EM, Chou YW, Hebert SC, Marx SJ, Steinmann B, Levi T, Seidman CE, andSeidman JG. Mutations in the human CaR gene cause familial hypocalciurichypercalcemia and neonatal severe hyperparathyroidism. Cell 75:1297-1303, 1993.3 U3 O2 R! P) o& C. E& g- k
: ?! {% }# w4 }  P, C* ?( K. E

Quamme GA. Renal magnesium handling: new insights in understanding old problems. Kidney Int 52:1180-1195, 1997.% J  v* K  d; f5 F" u7 Z9 s9 T
( a; W' W3 ^' D5 ?6 m

Riccardi D, HalAE, Chattopadhyay N, Xu JZ, Brown EM, and Hebert SC. Localization of theextracellular Ca 2  /polyvalent cation-sensing protein in rat kidney. Am J Physiol Renal Physiol 274:F611-F622, 1998.

' b8 r. U& f5 v1 B
Sands JM,Naruse M, Baum M, Jo I, Hebert SC, Brown EM, and Harris HW. Apicalextracellular calcium/polyvalent cation-sensing receptor regulatesvasopressin-elicited water permeability in rat kidney inner medullarycollecting duct. J Clin Invest 99: 1399-1405,1997.. d' X) ^) k7 t; |2 ^# V* S& M
# r, b! N5 }/ E' o9 R" c0 v
$ V' S3 T6 j( c! _; [5 K# b. t

Sawyer DB andBeyenbach KW. Mechanism of fluid secretion in isolated shark renalproximal tubules. Am J Physiol Renal Fluid ElectrolytePhysiol 249:F884-F890, 1985.

. Q8 b( K( i- @! }' S+ ^7 v2 p7 n

Schmidt-Nielsen K. Water and osmotic regulation. In: Animal Physiology, edited by Schmidt-Nielsen K.Cambridge, UK: Cambridge University Press, 1997.- E9 F9 v: p' Y+ S7 {$ j0 C2 V8 g2 t( o
, r' S8 p) T, V  {( p5 v
3 o0 F- ~7 z. u) Q3 k

Stolte H,Galaske RG, Eisenbach GM, Lechene C, Schmidt-Nielsen B, and Boylan JW. Renal tubule ion transport and collecting duct function in the elasmobranchlittle skate, Raja erinacea. J Exp Zool 199: 403-410,1977.
  e6 E+ o2 M5 w

2 n8 y! E+ p) n) T4 g, ]
Yokota S andBenyayati S. Regulation of glomerular filtration rate in a marineelasmobranch, the dogfish ( Squalus acanthias ). Bull MountDesert Isl Biol Lab 26:87-90, 1986.