Carbonic anhydrase XII mRNA encodes a hydratase that isdifferentially expressed - 《American Journal of Physiology》美国生理学杂志干细胞之家



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 ... 24 下一页

返回列表

查看: 484270\|回复: 239	go Carbonic anhydrase XII mRNA encodes a hydratase that isdifferentially expressed [复制链接]

轻羽

新手上路

Rank: 1

积分: 0
威望: 0
包包: 0

楼主

发表于 2009-4-21 13:32 |只看该作者 |倒序浏览 |打印

作者：George J.Schwartz, Anne M.Kittelberger, Richard H.Watkins,  MichaelA.O‘Reilly作者单位：Department of Pediatrics, University of Rochester School ofMedicine, Rochester, New York 14642
   8 t& v% W- T( M0 ?6 ?- b6 o) W- K' e



         ( w, E" g- N9 o" x9 P
         3 T4 i, I2 b& ?* V( d! \


                  ' T" h% \2 N; {5 j" ^9 O' b9 y

      ; H; o" W5 O  `9 Q

      【摘要】6 q9 n( y/ V" [1 w" y5 Z
   Membrane-bound carbonic anhydrase(CA) facilitates acidification in the kidney. Although most hydrataseactivity is considered due to CA IV, some in the basolateral membranescould be attributed to CA XII. Indeed, CA IV isglycosylphosphatidylinositol anchored, connoting apical polarization,but CA IV immunoreactivity has been detected on basolateral membranesof proximal tubules. Herein, we determined whether CA XII mRNA wasexpressed in acidifying segments of the rabbit nephron. The openreading frame of CA XII was sequenced from a rabbit kidney cortex cDNAlibrary; it was 83% identical to human CA XII and coded for a355-amino acid single-pass transmembrane protein. Northern blotanalysis revealed an abundant 4.5-kb message in kidney cortex, medulla,and colon. By in situ hybridization, CA XII mRNA was expressed byproximal convoluted and straight tubules, cortical and medullarycollecting ducts, and papillary epithelium. By RT-PCR, CA XII mRNA wasabundantly expressed in cortical and medullary collecting ducts andthick ascending limb of Henle's loop; it was also expressed inproximal convoluted and straight tubules but not in glomeruli or S3segments. FLAG-CA XII of ~40 kDa expressed in Escherichiacoli showed hydratase activity that was inhibited by 0.1 mMacetazolamide. Unlike CA IV, expressed CA XII activity was inhibited by1% SDS, suggesting insufficient disulfide linkages to stabilize themolecule. Western blotting of expressed CA XII with two anti-rabbit CAIV peptide antibodies showed no cross-reactivity. Our findings indicatethat CA XII may contribute to the membrane CA activity of proximal tubules and collecting ducts.
      【关键词】 in situ hybridization reverse transcriptasepolymerase chainreaction expressed protein proximal tubule medullary collecting duct! p: f" U8 g1 I* w- W
   INTRODUCTION

CARBONIC ANHYDRASE (CA) CATALYZES the reversible hydration of CO 2 anddehydration of carbonic acid. These functions would make this enzymeimportant in renal acid-base transport. Of the 14 isoenzymes of the CAfamily identified thus far, two of the major renal enzymes arecytosolic CA II, which accounts for 95% of activity, and CA IV, whichaccounts for a substantial portion of the membrane-associated CAactivity ( 6, 21, 37, 40 ). CA IV is aglycosylphosphatidylinositol (GPI)-linked protein ( 36, 41 ), indicating that it is likely to be expressed primarily onthe apical membrane ( 8, 26 ). However, we have recently shown functional CA activity on the basolateral membranes of proximal straight tubules ( 34 ), and the identity of thisbasolateral CA activity is unclear.

When the kidneys of CA II-deficient mice were examined forhistochemical activity ( 27 ), intense staining associatedwith the apical and basolateral plasma membranes was detected inproximal convoluted tubules. The absence of cytosolic CA II allowed the clear detection of CA hydratase activity associated with the plasma membranes. In addition, immunocytochemical studies that used three different antibodies to CA IV ( 9, 29, 31 ) consistently revealed both apical and basolateral labeling. The appearance of abasolateral signal suggests that a non-GPI-linked isoform of CA IV oran isoform of CA that cross-reacts with the antibodies to CA IV is present.

Recently, CA XII was identified, cloned, and characterized in renalcell carcinoma and normal human kidney ( 35 ). The sequence predicted a single-pass transmembrane protein with an extracellular CAdomain. Expression was also noted in normal human colon ( 14, 35 ) and in colorectal tumors ( 14 ). Recently,Parkkila et al. ( 24 ), by using an antibody against asecreted form of human CA XII ( 14 ), showed staining in thebasolateral membranes of cells of the thick ascending limb and distaltubules and in principal cells of the collecting ducts of humankidneys. A weak basolateral signal was also noted in proximalconvoluted tubules.# r* E. m1 Q7 r- \+ m9 n# h! H4 D/ P
* n9 s0 t; F, O& Y
The finding of an additional membrane CA was particularly relevant toour recent identification of functional basolateral CA activity in therabbit renal proximal straight tubule ( 34 ). Because muchacid-base physiology has been derived from studies in isolated rabbitnephron segments and CA is a key enzyme in acid-base physiology, wecloned rabbit CA XII, localized its mRNA in the kidney and otherorgans, as well as in some specific regions of the kidney, andcharacterized the expressed protein. Our findings are consistent withthe hypothesis that CA XII is at least partially responsible for somemembrane hydratase activity in the proximal tubule and collecting ducts.+ z( w, g" K# Z4 X* Y& F! ~

METHODS

Animals and preparation of tissue. New Zealand White rabbits (1.5-2.5 kg) were purchased fromHazleton-Dutchland Farms (Denver, PA) and fed standard laboratory chowwith free access to tap water. Each rabbit was anesthetized with anintracardiac injection of pentobarbital sodium (100 mg/kg) afterpremedication with intramuscular xylazine (5 mg/kg) and ketamine (44 mg/kg). The kidneys were rapidly removed and cut into coronal slices of1- to 2-mm thickness. Tissue was dissected from cortex, outer medulla,and inner medulla of fresh kidneys with a razor blade and scraped fromthe colonic mucosa; these samples were snap frozen in liquid nitrogen,as described previously ( 29 ). Small pieces of tissue fromlung, spleen, and heart were also cut and snap frozen. For in situhybridization, a kidney was perfused via the renal artery with PBSuntil it blanched, followed by fixation with neutral buffered formalin;2% paraformaldehyde, 75 mM lysine, and 10 mM sodium periodate; orPrefer (Anatech, Battle Creek, MI).
5 D) o3 ?9 b" N, [
Preparation of total RNA. Frozen tissue was homogenized in five to six short bursts with aTissuemizer (Ultra-Turrax, Janke-Kunkel, Tekmar, Cincinnati, OH) byusing a S25N probe at 24,000 rpm in an acid guanidinium thiocyanate-phenol chloroform solution (TriReagent, Molecular ResearchCenter, Cincinnati, OH), and total RNA was obtained according to theprotocol provided by the manufacturer. The RNA was resuspended indiethylpyrocarbonate-treated water and quantifiedspectrophotometrically by the absorbance at 260 nm. Purity wasdetermined from the ratio of absorbances at 260 and 280 nm and byethidium bromide staining after size fractionation on 1.5% agaroseminigels ( 38 ).' G0 [$ D9 N! l2 }" O. N, ]% u, @5 G

RT-PCR for detecting CA XII in rabbit kidney cortex. From the 3'-end of human CA XII ( 35 ), we prepared adegenerate antisense primer that encoded the last seven amino acids: 5'-GTC GAC TCA NGC RTG NGC YTC NGT YTC-3'. We used an upstream senseprimer encoding five amino acids that are conserved among the membraneCAs (rabbit CA IV, human CA IX, human CA XII, and mouse CA XIV)( 22, 25, 35, 38 ): 5'-AAG CTT CAR YTN CAY YTN CAY TGG-3'.Restriction sites ( Xho I and Hin dIII) were added to the primers to facilitate cloning in other vectors. The predicted size of the PCR product was ~750 bp. The rabbit CA XII probe was obtained by amplifying 250 ng of reverse-transcribed rabbit kidney cortex RNA at 42°C for 5 cycles followed by subsequent amplification at 52°C for 40 cycles. The band was gel purified withlow-melting-point agarose for use as a probe and for sequencing afterligation into pCR 2.1 cloning vector (TA cloning kit, Invitrogen, SanDiego, CA) ( 38 ).  K, X. e6 w" F( S% \

Screening of a rabbit kidney cDNA library for CA XII geneexpression and sequencing of positive controls. The 750-bp RT-PCR fragment of CA XII obtained from rabbit kidney cortextotal RNA was used to screen a quarter of a rabbit kidney cortex cDNAlibrary constructed in Lambda ZAP II vector (Stratagene, La Jolla, CA)and grown on XL1-Blue recA Escherichia coli host strain. Titration of this library revealed 3 × 10 9 plaque-forming units/ml. Duralon-UV membranes(Stratagene) were oriented on 150-mm NZY plates and used to lift DNAfrom the plaques. Prehybridization and overnight hybridization werecarried out in 0.5 M sodium phosphate-1 mM EDTA-7% SDS buffer (pH 7.2)at 55°C according to the manufacturer's instructions (Bio-Rad). The double-strand RT-PCR product was labeled with [ 32 P]dCTPby random hexanucleotide extension (Roche Diagnostics, Indianapolis,IN), and unincorporated radioactivity was removed by separation over aG-50 column. Washes were performed at low stringency [first in 40 mMsodium-1 mM EDTA-5% SDS (pH 7.2) and then as above but in 1% SDS] at55°C. Four rounds of screening yielded two positive clones, fromwhich the inserts and the Bluescript phagemid (Stratagene) were excisedin vivo from Lambda ZAP. The estimated sizes of the two clones were 1.9 and 1.4 kb, and these were sequenced by PCR (Prism DyeDeoxy Terminator,Perkin Elmer, Foster City, CA). Because the smaller clone appearedidentical to the larger one but did not yield a full-length openreading frame, only the larger clone was sequenced with oligonucleotide primers designed every 400 bp along both strands of the sequence. Thenucleotide sequence was assembled and aligned to the human sequence andRT-PCR fragment with MacVector (version 6.5, Accelrys, Princeton, NJ).
7 X$ u2 E% d8 x1 `9 `
Northern blot analysis of total RNA from kidney and other organs. Northern blot analysis was performed as previously described( 38 ). Briefly, 5-20 µg of total RNA were denaturedin 3% formaldehyde, size fractionated on 1.5% agarose gels,transferred to nylon filters (Zeta-Probe, Bio-Rad, Hercules, CA), andhybridized with DNA probes labeled with 32 P byrandom hexanucleotide extension (Roche Diagnostics). Each Northern blotwas run once with one set of tissues or organs. The DNA probes were a750-bp fragment of rabbit CA XII (see below), a 600-bp fragment ofrabbit CA IV ( 38 ), and rat -actin (housekeeping gene)( 23 ), which were labeled to a specific activity of1.5 × 10 9 counts · min 1 · µg 1.Autoradiography was performed with Kodak XAR film at 80°C for 1-4 days.

In situ hybridization. After overnight fixation, the tissue was dehydrated in a series ofgraded alcohols, placed in xylene, and then embedded in paraffinfollowing a standard protocol ( 1 ). Sections were cut to 5- to 8-µm thickness, placed on positively charged slides (Superfrost , VWR Scientific, Piscataway, NJ), and stored at 80°C.- _6 B' |7 @  P
7 K! i. H0 S0 L7 ]2 `
Plasmid DNA containing 1.9 kb of rabbit CA XII cDNA was linearized withrestriction enzymes Xho I or Xba I to makeantisense and sense templates, respectively, for cRNA probes accordingto the protocol of Auten et al. ( 2 ). The identityof these probes was confirmed by sequence analysis. The probes werelabeled with -[ 33 P]UTP to a specific activity of~1.5 × 10 9 counts · min 1 · µg 1.Alkaline hydrolysis reduced the probe length to ~200 bp( 39 ).( C" j# F% G, X% v0 w

Before hybridization, slides were deparaffinized by immersion in xyleneand then dehydrated through a graded ethanol series followed byproteinase K (GIBCO-BRL) digestion for 30 min at 37°C. Slides werethen equilibrated with 100 mM triethanolamine-HCl (pH 8) and treatedwith 0.25% acetic anhydride. The slides were then washed in 2× SSC,dehydrated, and dried. Prehybridization was at 53°C for 3 h;hybridization was performed for 16 h at 53°C by using ahybridization solution containing 30 ng · kb 1 · ml 1 of probe. The hybridization solution was 50% formamide, 300 mM NaCl,10 mM Tris · HCl (pH 8), 1 mM EDTA, 1×Denhardt's reagent, 10% dextran sulfate, and 0.5 mg/ml yeast tRNA.) z3 M: x: B5 A" X9 p7 o6 ~! p
8 ~0 p/ a1 O6 a3 r- l( `
After hybridization, slides were rinsed and digested with RNase A. Thestringent rinse was in 0.1× SSC at 69°C followed by a rinse in 0.1×SSC at room temperature. Then, the slides were dehydrated by passingthrough graded ethanol washes, dipped in 1:1 dilution of NTB-2 emulsion(Eastman Kodak, Rochester, NY) and exposed at 4°C for 17 days,developed as described ( 38, 39 ), and counterstained withhematoxylin and eosin. Sections from at least three different animalswere examined.: M- Z0 c8 z' d2 `& `
/ J( k1 B0 h. O. y! \
RT-PCR detection of CA XII expression in isolated nephronsegments. Tubule segments were isolated from kidney slices under a dissectingmicroscope in chilled PBS (containing calcium and magnesium) plus 10 mMvanadyl ribonucleoside complex (5 Prime 3 Prime, West Chester, PA) toinhibit RNA degradation, as previously described ( 33 ).Some kidney slices were treated at 37°C with 0.1% collagenase (type I, Sigma) for 20-35 min. From the cortex were dissected glomeruli (usually 4-5 glomeruli/tube), proximal convolutedtubules, proximal straight tubules, thick ascending limbs, and cortical collecting ducts. From the outer medulla were obtained S3 proximal tubules and outer medullary collecting ducts primarily from the innerstripe. Inner medullary collecting ducts (IMCDs) were dissected mainlyfrom the initial mid-inner medulla. Segments were measured for length,rinsed twice in PBS, transferred in 2 µl to a 0.5-ml microcentrifugetube containing 1.5 µl of 7% Triton X-100 plus 8 U ribonucleaseinhibitor (RNAsin, Promega, Madison, WI), sonicated 5 min at roomtemperature, and frozen at 80°C. Most segments were 1-2 mm inlength. Dissection of proximal tubules was generally completed within1 h of the animal's death; distal segments could be successfullyobtained for another 30-60 min.6 ?- _8 T5 b/ k1 \
' V: h' D- q' k% N
For each run of five segments, we used a positive control of 1 ng totalRNA from rabbit kidney cortex and a negative control of 0 ng RNA.Reverse transcription was performed for each sample for 1 h at42°C in 20 µl final volume by using 250 ng of random hexamers, 1 µl from each of four 10 mM deoxynucleotide stocks, RT buffer, and 200 U Superscript II (Invitrogen) RNase H RT according to themanufacturer's instructions. PCR was performed in a 50-µl reactionby using 2.5 U Platinum Taq DNA polymerase (Invitrogen). ForCA XII, 8 µl of first-strand cDNA was amplified for 40 cycles byusing 60 pmol of sense and antisense primers in 2 mM MgCl 2 with an annealing temperature of 55°C. For CA II, an identifying gene( 33 ) of proximal tubules and collecting ducts, 8 µl offirst-strand cDNA was amplified for 40 cycles by using 60 pmol of senseand antisense primers in 1.5 mM MgCl 2 with an annealingtemperature of 52°C. For L32, a ribosomal protein used as a generalhousekeeping gene ( 12 ), 1 µl of first-strand cDNA wasamplified for 35 cycles by using 50 pmol of sense and antisense primersin 1 mM MgCl 2 with an annealing temperature of 55°C. EachPCR was completed with a final 7-min extension at 72°C and storage at4°C.
1 V2 v; Z, J! y
The primers for CA XII were 5'-ACA TGT ACC TCC AGG GCC-3' (sense, bases308-325) and 5'-TCA GGC ATG GGC CTC GGT CTC-3' (antisense, bases1045-1066), and the expected size of the PCR product was 758 bp.The primers for CA II ( 33 ) were 5'-ATG TCC CAT CAC TGG GGGTAC-3' (sense, bases 1-21) and 5'-TGG CTC CTC AGG TTC CGC CTCCTT-3' (antisense, bases 718-738), and the expected size of thePCR product was 738 bp. The primers for L32 ( 12 ) were5'-AAG AAG TTC ATC AGG CAC CAG T-3' (sense, bases 691-712) and5'-GCA GCA TGT GCT TGG TTT TCT T-3' (antisense, bases 829-850),and the expected size of the PCR product was 159 bp.5 ~8 m0 c% B2 w0 Z: M/ V
" N0 a6 k) h3 A, _9 ^4 v$ ^
Twenty microliters of each sample were size fractionated on 1.2%agarose minigels (1 for each gene), and the products were visualized byultraviolet fluorescence after ethidium bromide staining. Each gel wasphotographed with identical settings. The products were scored aspresent, absent, or faintly positive. If L32 was faintly positive ornegative, the tubule was discarded. Proximal tubule segments wereconsidered satisfactory only if they expressed CA II mRNA( 33 ). Segments that were faintly positive for CA XII werecounted as positive. Segments (10-25) weremicrodissected from eight rabbits. Collagenase had no effect on theexpression of CA XII (data not shown), and collagenase-treated segmentswere pooled with those obtained without collagenase, as previously reported ( 33 ).( h3 V+ r9 G9 f: x8 U

Transient expression of rabbit CA XII in E. coli and in COScells. The full-length open reading frame of rabbit CA XII (1,068 bp) wasamplified by PCR from the large cDNA clone with restriction sites Xho I and Hin dIII engineered at each end. ThePCR product was directionally subcloned in frame into anNH 2 -terminal pFLAG-cytomegalovirus-1 expression vector(Sigma) via Hin dIII/ Sal I digestions.After transformation into DH5 cells, colonies were selected inampicillin and plasmid preparations were prepared. Some preparationswere induced by 0.5 mM isopropyl thiogalactoside for 24 h. CAXII-FLAG was obtained from the membrane fraction and media and examinedby Western blotting using M2 anti-FLAG monoclonal antibody (Sigma). Inaddition, the protein from 4 liters of media was concentrated toprovide adequate amounts for determination of hydratase activity.: `1 l. r+ X1 A/ \+ }

The full-length open reading frame of rabbit XII was directionallysubcloned in frame into pcDNA3 (Invitrogen), which provides a highlevel of constitutive expression via its cytomegalovirus-enhancer promoter. The CA XII-pcDNA3 (335-670 µg) was transfected into COS-7 cells by using 4 µg lipofectamine (GIBCO-BRL) on 12-well tissueculture plates (20-mm diameter) according to the manufacturer's instructions. After 48 h, the cells were scraped off the well plates and membrane proteins were isolated in Sato's buffer (25 mMtriethanolamine, pH 8.1, 59 mM Na 2 SO 4, and 1 mMbenzamidine chloride) plus the protease inhibitors 1 mM EDTA, 1 mMiodoacetate, 0.1 mg/ml 4-(2-aminoethyl)-benzenesulfonyl fluoridehydrochloride (Pefablock, Boehringer-Mannheim, Indianapolis, IN), 0.1 mg/ml 1,10-phenanthroline, 2 µg/ml pepstatin A, 5 µg/mlchymostatin, 10 µg/ml leupeptin, and 10 µg/ml aprotinin( 29 ).

Stable expression of rabbit CA XII in cultured rat IMCD cells. To obtain larger amounts of material that were posttranslationallymodified by a mammalian system, we expressed these CAs in immortalizedrat IMCD cells. We selected immortalized rat IMCD cells to express CAXII, because they are acidifying epithelia in culture and reasonablyrepresent the IMCD in situ ( 3, 17 ). The full-length openreading frame of rabbit XII was directionally subcloned in frame intopcDNA3 (Invitrogen). This vector has a neomycin resistance site tofacilitate stable expression. In preliminary studies, we determinedthat 600 µg/ml of Geneticin (GIBCO-BRL) killed nontransfected IMCDcells within 2 wk. The IMCD cells were transfected with CA XII in pcDNAand grown in six-well plates in 600 µg/ml of Geneticin. Individualsurviving colonies were obtained with cloning cylinders and eventuallyexpanded into T-75 flasks. Once confluent, the cells were scraped offand homogenized, and membranes were obtained from the fraction that wascentrifuged at 100,000 g for 1 h. The protein wasquantified by the BCA assay (Pierce, Rockford, IL) with BSA as a standard.& l. @6 f0 A" m" g  f

A similar approach was used to stably express the full-length rabbit CAIV open reading frame in IMCD cells ( 30 ). CAIV-transfected cells expressed abundant CA activity and a protein ofsimilar size to rabbit CA IV ( 29 ). These cells were usedas CA IV-positive controls for the IMCD cells expressing CA XII.

CA hydratase activity. Crude membrane samples were solubilized in Sato's buffer (25 mMtriethanolamine, pH 8.1, 59 mM Na 2 SO 4, and 1 mMbenzamidine chloride), which also contained protease inhibitors, asnoted above ( 29 ). Hydratase activity in expressed rabbitCA XII was measured at 1-2°C as a colorimetric end point assayby using imidazole/Tris buffers, CO 2 as a substrate, and p -nitrophenol as the pH indicator ( 4, 6 ). Thisminiature assay utilized 50 µl membrane sample plus water and 50 µl p -nitrophenol in buffer for a total volume of 100 µl. Anenzyme unit (EU), normalized to 1 mg protein, is the amount ofhomogenate necessary to halve the reaction blank time and was correctedfor miniaturization of Maren's micromethod ( 18 ) bydividing the present results by 10 (4). CA hydratase activity was computed from log ( B / S )/log 2 ( 4 ) and normalized to 1 mg protein, wherein B is the time measured for the boiled inactivated enzyme or matchedbuffer devoid of enzyme and S is the time measured for themembrane sample. In practice, we added enough membrane protein suchthat the reaction time was approximately one-half of the blank.
4 y. o, P8 {) Z  e! K- ~, e% `
Characterization of the expressed CA XII hydratase activity wasperformed by examining comparable amounts of membrane that were boiled(denatured) for 3 min, exposed for 20 min to the general CA inhibitoracetazolamide (0.01-500 µM) ( 4 ) or to 0.1-2%SDS; the latter inhibits CA II but not CA IV activity ( 6, 37, 41 ). The expressed CA IV hydratase activity was examined in asimilar fashion to provide a comparison. All hydratase assays wereperformed in duplicate or triplicate.
, g6 X( L/ Q7 A( h1 \
Analysis and statistics. The intensity of the signals in the autoradiograms was analyzed byscanning densitometry (SigmaGel, Jandel, San Rafael, CA). To compensatefor differences in quantity of total RNA on each lane of the membranein the Northern blot analysis, each CA value was normalized to itsrespective value of -actin.  o& ~' Y4 A, O
; }% K) K$ O' e2 z% w$ q" B, _
CA XII protein secondary structure analysis was obtained fromalgorithms contained in MacVector and Gene Runner (version 3.04, Hastings Software, Hastings, NY) programs.  ^! l4 G6 X# r% b

In situ hybridization images were processed by using Spot software(version 3.0.4, Diagnostic Instruments, Sterling Heights, MI) andfurther optimized with Photoshop (version 6.0, Adobe Systems, San Jose,CA). The images were collected and presented in montage form by usingPowerpoint (version 2000, Microsoft, Bellevue, WA).. X" }& V( _9 N3 {7 {( n6 ?

RESULTS

Cloning and sequencing of rabbit CA XII. Using a degenerate antisense primer situated at the end of the codingregion plus an upstream sense primer encoding five amino acids that areconserved among the membrane CAs, we obtained a 720-bp RT-PCRproduct from rabbit kidney total RNA (Fig. 1 ). Sequence analysis of this productrevealed 76% identity to human CA XII ( 35 ). This cDNA wasthen used to screen one-quarter of a rabbit kidney cortex cDNA library.Two clones were obtained after four rounds of screening, and each wassequenced from the ends by using primers derived from the Bluescriptphagemid. Clone 3-2a1 spanned the entire open reading frame of CAXII, and it was sequenced in both directions by using a series ofinternal primers every 300-400 bases. The GenBank accession numberis AF-263367. Comparison with the full-length open reading frame ofhuman CA XII revealed 83% identity at the nucleotide level, but therabbit sequence contained three additional nucleotides at the 3'-endfor a total of 1,068 through the open reading frame. These threeadditional bases were confirmed by sequencing several RT-PCR productsgenerated from the distal half of the molecule.
$ q* n. ^( P9 L+ d# U8 E5 j
Fig. 1. Schematic diagram of rabbit carbonic anhydrase (CA) XIIopen reading frame (ORF) and poly A tail (AAA), the two cDNA clonesobtained from the rabbit kidney cortex library, and the initial RT-PCRproduct obtained from rabbit kidney total RNA. The lengths and positionof the molecules were drawn to scale.

The encoded amino acid sequence revealed 77% identity and anadditional 4% similarity to human CA XII; the rabbit sequence predicted 355 amino acids in the translated polypeptide compared with354 amino acids in the human sequence (Fig. 2 ). There was an addition of one aminoacid (glycine) in the putative cytoplasmic region.
2 E1 \3 @4 z, ~  V! `
Fig. 2. Alignment of rabbit (Rab) and human (Hum) CA XII aminoacid sequences. The dark-gray boxes indicate identical amino acids,whereas the light-gray boxes indicate similar amino acids. Theidentical amino acids are listed below the paired sequences, and a dotindicates a similar amino acid. An absence of gray indicates differentamino acids. There is a one amino acid gap at 332 in human CA XII.
1 ~" I8 Q% a* E8 o
From Kyte-Doolittle hydropathy plots and von Heijne andGoldman, Engleman, and Steitz methods to identify transmembraneregions, we predicted that the rabbit CA XII protein contained anNH 2 -terminal signal sequence of 26 amino acids, followed bya CA domain, a putative transmembrane region (amino acids294-327), and a terminating hydrophilic sequence of 28 aminoacids, which corresponded to a cytoplasmic COOH terminus. Thusrabbit CA XII was predicted to be a single-pass transmembrane proteinwith a cytoplasmic tail. Compared with human CA XII, the predictedsecondary sequences were similar in size and location. The COOHterminus contained putative sites for cAMP and casein kinase II phosphorylation.

Each of the histidines, which bind zinc and/or serve as proton shuttleagents for human CA XII, was encoded by the rabbit sequence (aminoacids 61, 94, 119, 121, 133, 145, and 148). Comparison to each of thehigh-activity rabbit CAs (II and IV) showed conservation of thesehistidines, except for His61 (Fig. 3 ).One pair of cysteines was conserved in rabbit and human CA XII at aminoacids 50 and 230, suggesting the possibility of intramoleculardisulfide bonds, which could confer partial resistance to denaturationof hydratase activity by SDS. CA IV, which has two pairs of cysteines( 38, 41 ), is resistant to denaturation by 0.2-10%SDS ( 6, 37 ). The location of these two cysteines in CA XIIwas reasonably well conserved compared with rabbit CA IV.
& ?1 ^$ a0 r# i5 L, x. S
Fig. 3. Alignment of amino acid sequences of rabbit CA XII, IV,and II, beginning with 1 of CA XII and finishing with 355 of CA XII,which because of gaps numbers 363. The dark-gray boxes indicateidentical amino acids, whereas the light-gray boxes indicate similaramino acids. Identical amino acids are listed below the matchedsequences, and a dot indicates a similar amino acid. An absence of grayindicates different amino acids.& n/ ^/ \' {, s& ]0 E
; d8 {# L' {3 G/ Z" E9 ~
The extracellular portion of the molecule had three potential sites for N -asparagine glycosylation and four for myristolation, whichwould suggest an increase in molecular mass above the 39.6 kDapredicted from the amino acid sequence.

Expression of CA XII in rabbit kidney and other organs. Using the 720-bp RT-PCR product as a probe for CA XII, we examined theexpression of this gene in the zones of the kidney and in a few otherorgans. Figure 4 A shows theexpression of CA XII mRNA in mouse kidney and in outer medulla, innermedulla, and cortex from rabbit kidney. CA XII was seen as a 4.5-kbsingle transcript that was most abundant in kidney cortex and less so in inner medulla and outer medulla. Figure 4 B shows aNorthern blot of cortex, colon, lung, spleen, and heart tissue. CA XII was well expressed by cortex and colon but not by lung, spleen, orheart. This was in contrast to CA IV, which was abundantly expressed incolon, lung, and kidney cortex and much less so in heart. -Actin wasused to compensate for loading differences and was expressed in each ofthese tissues, although the molecular mass was smaller in heart. Whenthese samples were expressed as CA XII/actin ratios, the cortex was3.7, colon 2.0, lung spleen
; b0 [- F- n+ Z6 @& e
Fig. 4. A : Northern blot of 20 µg of total RNA frommouse kidney (MK) and rabbit kidney outer medulla (OM), inner medulla(IM), and cortex (CTX). A single band was seen slightly smaller thanthe 28S marker (~4.5 kb), heaviest in the cortex. B :Northern blot of total RNA from rabbit tissues including cortex (Ctx; 7 µg), colon mucosa (Co; 10 µg), lung (Lu; 20 µg), spleen (Spl; 20 µg), and heart (Hrt; 20 µg) probed for CA XII and CA IV and thenreprobed for actin. Expression of CA XII and CA IV was normalized toactin and expressed as a ratio.2 r* {6 ~1 R. I$ s8 V2 ^  A5 n8 d

In situ hybridization. CA XII was abundantly expressed in the inner cortex, as seen in thelow-power darkfield views of the corticomedullary junction (Fig. 5 A ); proximal convolutedtubules (arrows) but not glomeruli expressed it. The signal was notabundant in the outer stripe of the outer medulla (Fig. 5 B ).In the inner stripe of the outer medulla and the inner medulla, CA XIIwas well expressed over medullary collecting ducts (Fig. 5, B and C ).

Fig. 5. Darkfield views of CA XII in situ hybridization in rabbit kidney. Left : low-power views with antisense probe. Right : high-power views with sense probe; they can becompared with the views from Fig. 6. A : corticomedullaryjunction showing juxtamedullary glomeruli (G) and abundant signal overproximal convoluted tubules (arrows) surrounding these glomeruli; lesssignal was seen in the adjacent outer stripe of the outer medulla. B : view of outer medulla at junction of inner stripe (IS)and outer stripe (OS); signal is observed over tubules (arrows) in theinner stripe. C : inner medulla shows collecting ductsexpressing CA XII (arrows). D : high-power sense view ofmidcortex showing glomerulus (G) in center surrounded by proximaltubules. E : sense view of inner medulla showing collectingducts (arrows). Sense views did not show concentration of signal overany particular structures.
: @% I  [9 k/ g& N( f5 W' L
High-power views over the inner cortex (Fig. 5 D ) and innermedulla (Fig. 5 E ) with a sense probe did not reveal anyspecific increase in grain density over proximal tubules or decreaseover a glomerulus in the cortex (Fig. 5 D ) or over theepithelia of IMCDs (Fig. 5 E, arrows).
- a  B& D, r) V  C1 W- N8 d0 `0 h( v
High-power views showing side-by-side hematoxylin/eosin and dark fieldimages, by using an antisense probe, confirmed a specific CA XII signalover proximal convoluted tubules, but little was observed overglomeruli of the cortex (Fig. 6, A and B ). The medullary rays showedan intermediate signal, and at high power this signal appeared to beprimarily over proximal straight tubules and cortical collecting ducts(C in Fig. 6, C and D ). This signal appeared tobe less than that observed over neighboring proximal convoluted tubules. In the inner stripe of the outer medulla, there was a definitesignal detected over the outer medullary collecting ducts (C in Fig. 6, E and F ). In the initial inner medulla, abundant signal was observed over the epithelia of IMCDs (C in Fig. 6, G and H ) and of collecting ducts in the terminalinner medulla (C in Fig. 6, I and J ). In theselast views, there appeared to be a specific signal noted over thepapillary epithelium, a finding that was not seen by using a senseprobe (not shown).
) p) P/ l+ x) j( O$ c  Y4 m& p) B/ g
Fig. 6. High-power views of rabbit kidney with an antisense probe withhematoxylin and eosin views ( left ) and darkfield( right ). A and B : cortex view withglomerulus (G), proximal convoluted tubules (P), and a collecting duct(C). Signal was seen over proximal tubules and collecting duct andnearly absent from the glomerulus. C and D :cortex view of medullary ray showing proximal straight tubules (S),cortical collecting duct (C), and proximal convoluted tubule (P).Signal appeared heaviest over the proximal convoluted tubules and lessover proximal straight tubules, thick ascending limbs, and collectingduct. E and F : outer medulla view showing signalover outer medullary collecting ducts (C) from the inner stripe. G and H : inner medulla view showing innermedullary collecting ducts (IMCD; C) with abundant signal over theircells. I and J : papillary view showing terminalIMCDs (C) and papillary epithelium (Pa) expressing abundant CA XIImRNA.. q$ f; e1 E$ J0 ?' K
; l! a' \, m) [5 E1 m
Expression of CA XII mRNA by individual nephron segments. RT-PCR was performed on microdissected nephron segments (Table 1 ), and a representative run is shown inFig. 7. Glomeruli did not express CA XII(an average of 5 glomeruli were pooled for each sample). Proximalconvoluted and straight tubules generally expressed CA XII mRNA; 13 of15 convoluted segments averaging 1.4 ± 0.1 mm and 15 of 20 straight segments averaging 1.5 ± 0.1 mm were positive. Only 1 of11 medullary proximal straight tubules, S3 segments averaging 1.2 ± 0.1 mm, expressed CA XII mRNA. Thick ascending limbs of Henle'sloop (averaging 1.6 ± 0.2 mm) expressed CA XII mRNA in more thanone-half of the cases (14/24). Surprisingly, 20 of 24 of these segmentsexpressed CA II as well. CA XII mRNA, as well as CA II, was readilydetected in collecting ducts. Cortical collecting ducts (averaging1.3 ± 0.1 mm) abundantly expressed CA XII (25/25) as well as CAII (25/25), whereas 7 of 9 outer medullary collecting ducts (averaging1.2 ± 0.1 mm) and 12 of 15 IMCDs (averaging 1.2 ± 0.1 mm)expressed CA XII.

Table 1. CA XII mRNA expression by isolated nephron segments
. k0 M, X" k" k/ z) F# y
Fig. 7. Examination of microdissected tubules for expression ofCA XII, CA II, and L32 by RT-PCR. The marker lane (M) shows a 100-bpladder with heavy band for 600 bp. The signal for CA XII is seen at 758 bp, for CA II at 738 bp, and for L32 at 159 bp. Lane 1, 5 glomeruli (Glo); lane 2, thick ascending limb of Henle'sloop (TAL); lane 3, cortical collecting duct (CD); lane 4, medullary proximal straight tubule (S3); lane5, cortical proximal straight tubule (PST); lane 6, 1 ng of total RNA from rabbit cortex (positive control; 1); lane7, 0 ng total RNA (negative control; 0). In this representativerun, both cortical collecting duct and proximal straight tubule expressCA XII, whereas thick ascending limb of Henle's loop, corticalcollecting duct, medullary proximal straight tubule, and corticalproximal straight tubule express CA II; each segment expresses L32.+ k& Y9 I/ L- b
0 g/ M2 M+ e) X1 c4 O! w/ V
Expression of CA XII FLAG protein. To characterize the activity of CA XII and determine whether it reactedwith our anti-CA IV antibodies, we expressed it by using a FLAG-taggedexpression plasmid. CA XII-FLAG was obtained from the membrane fractionof E. coli and expressed as a ~40-kDa protein (arrow) asseen by Western blotting using M2 anti-FLAG antibody (Fig. 8 ). From the 355 amino acids encoded bythe open reading frame, the predicted molecular mass would be 39.6 kDa, and the FLAG tag would add ~1 kDa to this sequence, totaling ~40.6 kDa, similar to what was observed in the Western blot assuming noposttranslational modification by the E coli. Adding 0.05 mM of the inducer isopropyl thiogalactoside for 24 h increased the expression of CA XII-FLAG (not shown).
  v# j$ V7 \0 L6 }5 d& h5 d) l$ Z9 G
Fig. 8. Western blot with M2 anti-FLAG antibody to detect CAXII-FLAG expressed by fractions of Escherichia coli.Left : CA XII was expressed at ~40 kDa in the membranefraction (Memb) but not in the periplasmic (Peri) or secreted (Sec)fractions. Right : when the media was concentrated, secretedCA XII-FLAG was detected in 300 µg of protein (Sec).

The membrane protein was suspended in Sato's buffer plus 0.5% TritonX-100 and assayed for hydratase assay. The membranes yielded 1.5 ± 0.1 EU/mg of hydratase activity (Table 1 ), and this was 69%inhibited by 0.1 mM acetazolamide and 87% by 0.5 mM acetazolamide andcompletely inhibited by 1% SDS or boiling for 3 min. The protein fromthe media contained 1.3 ± 0.1 EU/mg activity, and this was fullyinhibited by 0.1 µM acetazolamide or by boiling and 70% inhibited by1% SDS.4 i8 ?8 }; d8 {8 f6 }
  L" i. V! |+ n# }# s+ |
Transient expression of CA XII in COS cells. COS-7 cells were transiently transfected with CA XII-pcDNA3, and48 h later were scraped, homogenized, and centrifuged to yieldmembrane proteins. Fifteen micrograms of membrane proteins yielded 1 EUof hydratase activity (6.5 ± 0.3 EU/mg, Table 2 ). This activity was completelyinhibited by 0.1 mM acetazolamide or by boiling. Membranes fromnontransfected COS cells had no detectable hydratase activity.. h& ?5 S5 Q  W) g5 d; C! `

Table 2. Expression of CA hydratase activity

Stable expression of CA XII by cultured rat IMCDs. Membrane protein was obtained from rat IMCD cells stably expressing CAXII-pcDNA and assayed for hydratase activity; three bulk preparationsof membranes from several colonies averaged 17.0 ± 1.5 EU/mg(Table 2 ). Seven individual colonies showed ~1 EU ofhydratase activity in 45-220 µg of membrane protein with a meanof 10.1 ± 2.1 EU/mg. In contrast, membranes from IMCD cells transfected with the vector alone showed minimal activity (0.2 ± 0.1 EU/mg, Table 2 ).% P; L- x. r8 D- c* t! g

One clone of IMCD cells (2A9) that expressed CA XII was examined ingreater detail. The hydratase activity was 11.6 EU/mg. The expressedmembrane hydratase activity was completely inhibited by 0.1 µMacetazolamide and by boiling for 3 min and nearly so by 1% SDS. An SDStitration showed 57% inhibition by 0.1% SDS, 82 by 0.5%, 98 by 1%,and 100 by 2% SDS. An acetazolamide titration showed completeinhibition at 1 and 0.1 µM acetazolamide and 48% inhibition at0.01 µM.6 N# D) m( f+ ^& M! J$ K6 |3 n( c. G" i

For comparison, seven independent clones of IMCD cells stablyexpressing CA IV had a mean hydratase activity of 18.9 ± 9.4 EU/mg. One clone studied in depth (2c) showed no inhibition in SDS upto 2% concentration (Table 2 ). For another comparison, membranes fromrabbit kidney cortex expressed ~13 EU/mg ( 6 ). Thishydratase activity was stable in 1% SDS but was completely inhibitedby 2 µM acetazolamide ( 6 ). The resistance to SDS in themembrane fraction is consistent with most of the hydratase activitybeing CA IV.
& Y+ i2 }' V- K6 d$ I
Expression of CA XII does not cross-react with anti-CAIV antibodies. Because anti-CA IV antibodies identify basolateral staining in theproximal tubule, we determined whether CA XII, which may beexpressed basolaterally, was detectable by anti-CA IV antibodies. Membrane proteins were obtained from IMCD cells stably transfected withCA XII and from IMCD cells expressing full-length CA IV. The membraneprotein was transferred to nitrocellulose and probed with each of ouranti-CA IV peptide antibodies ( 38 ). Figure 9 shows that 20 µg of membrane proteinfrom IMCD cells stably expressing CA XII (CA 12) were not detectablewith either of the two anti-CA IV peptide antibodies, anti-KDNV( left ) or anti-YDQR ( right ). In contrast, proteinfrom IMCD cells expressing CA IV (CA 4) was readily detectable witheither antibody and revealed a protein of 46-50 kDa, similar towhat has been observed previously in rabbit kidney ( 29, 31 ).) Z) I- x, T+ S; y
, M: ~9 K8 A' t; C
Fig. 9. CA XII protein stably expressed by IMCD cells (CA12) wasnot detectable with anti-KDNV ( left ) or anti-YDQR( right ) CA IV peptide antibodies. The positive control wasIMCD cells stably expressing CA IV (CA 4).
$ I! [9 ~+ s9 i/ c2 P* F+ u( `0 K
DISCUSSION

With the identification of several new isoforms of CA, it isnecessary to identify each isoform in a specific organ, localize andcharacterize it, and determine its function. Because an extensive physiology of acid-base homeostasis has been derived from isolated perfused segments of the rabbit nephron, we have systematically investigated CA expression in the rabbit kidney ( 5, 6, 29, 32-34, 39 ). With respect to CA, we have previouslylocalized CA II and IV in specific nephron segments of rabbit kidney( 4, 6, 29, 31-33, 39 ). The present study shows thatrabbit CA XII was quite similar to human CA XII (83% identity at thenucleotide level; 76% identical and 83% similar at the amino acidlevel). However, rabbit CA XII had an additional cytoplasmic amino acid for a total of 355. The protein was predicted to be a single-pass transmembrane structure. Each of the three zinc-binding histidines ofthe high-activity CAs was conserved in CA XII.
# {3 I: U& m: P/ E8 ]! y& T  p
Rabbit CA XII mRNA was abundantly expressed in kidney cortex and lessso in medulla. In nonrenal tissues, CA XII was expressed in colon butnot in lung, spleen, and heart. The most abundant expression of CA XIImRNA was detected in kidney cortex and colonic mucosa; other sectionsof the intestine were not investigated. In humans ( 35 ), CAXII mRNA is present in kidney and colon but not in other tissues. ThusCA XII mRNA appears to display a similar pattern of tissue distributionin human and rabbit. By immunocytochemistry, CA XII protein is highlyexpressed in colon ( 14 ) but is absent from the smallintestine; the staining pattern is confined to the basolateralmembranes of the enterocytes.6 a9 \( Q/ F% ^; q* s' o5 P1 o
4 _. ?% y: ^( T8 ?4 ~% [
In addition to examining rabbit CA XII mRNA in various organs, we haveexpressed rabbit CA XII protein in several cell systems. First, byusing an NH 2 -terminal FLAG sequence, we expressed rabbit CAXII in E. coli, and its molecular mass was nearly identical to that predicted for the protein from the amino acid sequence (~40kDa); moreover, substantial hydratase activity was detected. Then, byusing a transient transfection of COS cells, we expressed hydrataseactivity in the membranes derived from these cells. As with most of theactive CAs, the hydratase activity was inhibited by boiling and byacetazolamide, as shown by Tureci et al. ( 35 ), whocharacterized expression of the human isoform.
( v. y; `: h9 ?* t7 k- R3 v
Using the same CA XII construct, we investigated IMCD cells, a cellline that expresses H -ATPase ( 3 ) and CA II( 28 ) but not CA IV ( 28 ). The ample expressionof CA XII mRNA in the IMCD of the rabbit kidney suggested to us that anIMCD cell line would serve as a good model. Indeed, transfected CA XIIwas stably expressed at high levels in the membranes of several clonesof IMCD cells in culture. As a control, membranes from this cell linethat had been transfected with vector alone (Table 1 ) or were nottransfected (not shown) expressed very small amounts of membranehydratase activity, the identity of which is not yet known. Titrationof the CA XII-transfected cells with acetazolamide revealed anI 50 of ~5 × 10 8 M, similar to thatfor partially purified CA IV ( 19, 37 ). Titration with SDSrevealed significant inhibition even at concentrations as low as 0.1%SDS. This study indicated that CA XII had little stability in SDS andtherefore was unlikely to have disulfide linkages that would stabilizeit in detergents, despite the presence of a pair of cysteines in theextracellular domain ( 20, 37 ).

Renal localization of CA XII mRNA. In situ hybridization confirmed the abundant expression in the cortex,with heavy expression by proximal convoluted tubules and less so overproximal straight tubules. Medullary rays, which also include corticalcollecting ducts and cortical thick ascending limbs, also showed CA XIImRNA expression, but there was virtually no expression over glomeruli.The medulla showed labeling over outer medullary collecting ducts butmore abundant signal over collecting ducts from the initial andterminal inner medulla. The proximal tubule in the kidney and the colonare both involved in salt, water, and bicarbonate reabsorption, so itis conceivable that the CA XII is present in the basolateral membraneto provide additional CO 2 to facilitate the reabsorption ofNaCl and water via the Na-3HCO 3 cotransporter. Indeed, wehave shown in a recent physiological study ( 34 ) thatbasolateral CA activity facilitates bicarbonate and fluid absorption inthe isolated perfused rabbit proximal straight tubule.+ s# s4 {+ A' i5 T6 [
! M4 e$ t1 V/ Q, f9 M+ A2 a" `
Examination of CA XII expression in isolated nephron segments by RT-PCRshowed excellent agreement with the in situ hybridization studies.There was abundant expression of CA XII mRNA in cortical, outermedullary, and IMCDs, as well as thick ascending limbs of Henle'sloop. Expression was also noted in proximal convoluted and straighttubules but not in S3 segments or glomeruli.

A recent immunohistochemical study ( 24 ) of human kidneyusing a polyclonal antibody to a COOH-terminally truncated CA XII protein showed distinct labeling over basolateral membranes of thethick ascending limb of Henle's loop and distal convoluted tubulecells and principal cells of the cortical collecting ducts. Abasolateral signal was seen over the proximal convoluted tubules, aswould be anticipated from our data. In the medulla, a signal was notedover basolateral membranes of some cells in the collecting ductepithelium, but localization within the medulla (inner vs. outer) wasnot specified. Double labeling with CA II, a cytosolic marker ofintercalated cells, revealed that the majority of CA XII-positive cellsin collecting ducts showed no reaction for CA II, indicating that thecells expressing CA XII were likely to be principal cells rather thanintercalated cells. This was confirmed by aquaporin-2 staining. In somedistal tubule cells, there was double labeling for CA XII and CA II,but the specific identification of such cells was not performed;intercalated cells are not usually found in distal tubule ( 13, 16 ).

Although our mRNA studies cannot provide information on the polarity ofCA XII expression, they clearly agreed for the most part with theimmunohistochemical study of Parkkilla et al. ( 24 ) thatshowed expression along the distal nephron as well as in proximalconvoluted and straight tubules. Nevertheless, there were some minordifferences between the RT-PCR findings and the in situ hybridizationstudies. The more extensive expression in proximal tubules observed bythe latter method may possibly be attributable to cross-hybridizationof alkaline-hydrolyzed shortened probe containing a common CA domain,with expressed CA II mRNA. In addition, the thickness of the sectionscould result in some lack of specificity. We cannot provide otherexplanations at this time and suggest that additional antibodylocalization studies be performed in the future. Because the antibodyused by Tureci and colleagues ( 35 ) was made in rabbits, itis unlikely to be useful for examining CA XII expression in rabbit kidney.: E! {0 G, L% P1 O. D

Another unexplained finding is the detection of CA II mRNA in thickascending limbs. The rabbit thick ascending limb, in contrast to thatin mice and rats, does not express CA II activity ( 10, 11, 15 ), and we previously found no CA II mRNA in isolated rabbitthick ascending limbs ( 33 ). Perhaps the use of a more potent Taq polymerase contributed to the more recentpositive finding. The fact that this segment in the rabbit does notexpress CA II activity ( 10, 11 ) suggests that theexpression of CA II mRNA by rabbit thick ascending limbs is notphysiologically important.
, b  U3 Q% g9 {# s9 A' e
The molecular identities of the apical and basolateral membrane CAshave been suggested to be CA IV ( 9, 29, 31 ) on the basisof immunocytochemical labeling. This immunocytochemical finding is incontrast to the expectation that CA IV is more likely to bepreferentially apically expressed because of being GPI anchored ( 7 ). It is possible that CA IV is expressed basolaterallyas well, perhaps as a non-GPI-linked single-pass membrane isoform, or abasolateral CA isoform may be cross-reacting with anti-CA IVantibodies. Either of these choices is presently feasible. However,when CA XII was expressed in E. coli or IMCD cells,cross-reacting products were not detected with either of our anti-CA IVpeptide antibodies. Thus it is likely that our anti-CA IV antibodiesare detecting an isoform on the basolateral membrane that is not CA XII.7 Y, x, U5 V' h' Y4 Y9 |

Our studies show that CA XII mRNA is expressed in the rabbit kidney andpredominately over proximal tubules, thick limbs, and collecting ducts.Although CA XII has been considered to be important in salt and watertransport ( 24 ), our sequence analysis showed that each ofthe histidines expressed by high-activity CAs was preserved, and theexpressed protein had substantial hydratase activity. These findingssuggest that CA XII might be involved in CO 2 fluxes (i.e.,H transport) and therefore could be expressed byacidifying segments of the nephron, such as the proximal tubule andmedullary collecting duct. The abundance of the mRNA and thelocalization to these acidifying nephron segments are compellingreasons to generate a specific CA XII antibody that is useful forstudies in rabbit kidney.4 O% x! ]' @+ l, g

ACKNOWLEDGEMENTS5 l; b& s% O* m  Q
; F( o! }* u8 h1 k0 p4 a, z
We appreciate the help of Dr. Lois Arend in reviewing the in situ hybridizations.2 C$ m/ ~5 f% F- _) V3 @) N7 r
      【参考文献】
1. Ausubel, FA,Brent R,Kingston RE,Moore DD,Seidman JG,Smith JA,andStruhl K. Current Protocols in Molecular Biology. New York: Wiley, 1995.
( T* b! _/ C$ N4 ~% }: y. O/ V' t

2. Auten, RL,Watkins RH,Shapiro DL,andHorowitz S. Surfactant apoprotein A (SP-A) is synthesized in airway cells. Am J Respir Cell Mol Biol 3:491-496,1990.1 V# K" V1 P' k/ C) p9 Q9 L

& p9 k7 C' p6 V2 U* X! o
3. Brion, LP,Schwartz JH,Lachman HM,Zavilowitz BJ,andSchwartz GJ. Development of H secretion by cultured renal inner medullary collecting duct cells. Am J Physiol Renal Fluid Electrolyte Physiol 257:F486-F501,1989 .

4 m+ `9 D  S7 L2 f9 O& E- h: G
4. Brion, LP,Schwartz JH,Zavilowitz BJ,andSchwartz GJ. Micro-method for the measurement of carbonic anhydrase activity in cellular homogenates. Anal Biochem 175:289-297,1988  .6 m3 y0 o& F! D9 h3 d& X  I" j

  g# R- ]) Z  S$ r1 q' g8 Y
5. Brion, LP,Zavilowitz BJ,Rosen O,andSchwartz GJ. Changes in soluble carbonic anhydrase activity in response to maturation and NH 4 Cl loading in the rabbit. Am J Physiol Regul Integr Comp Physiol 261:R1204-R1213,1991 .
$ R& [$ A2 W' g. `" {

6. Brion, LP,Zavilowitz BJ,Suarez C,andSchwartz GJ. Metabolic acidosis stimulates carbonic anhydrase activity in rabbit proximal tubule and medullary collecting duct. Am J Physiol Renal Fluid Electrolyte Physiol 266:F185-F195,1994 ." X; f5 q1 X+ q+ J. R
# Z+ a  w, D+ f, n& r
8 K3 ?% n! q, ?9 H1 T

7. Brown, D,andStow JL. Protein trafficking and polarity in kidney epithelium: from cell biology to physiology. Physiol Rev 76:245-297,1996 .( W$ m6 K) Y  K2 `4 H  H% @
2 ]8 d. o. @" e5 k9 A3 w7 T
+ H' o- P& F5 [3 M; S
* [; N& B: }$ z  m1 z
8. Brown, D,andWaneck GL. Glycosyl-phosphatidylinositol-anchored membrane proteins. J Am Soc Nephrol 3:895-906,1992 .6 B( i+ w% T& u! m$ _' i& K' T" d1 O

8 m9 p9 L; D  I- n1 g
" v0 r' i) A  M8 _  ^: G& k7 o
9. Brown, D,Zhu XL,andSly WS. Localization of membrane-associated carbonic anhydrase type IV in kidney epithelial cells. Proc Natl Acad Sci USA 87:7457-7461,1990 .$ l5 \# N7 r# J3 @; R
/ E# d2 d+ m. L. n+ U" ^. l6 C8 j

10. Dobyan, DC,andBulger RE. Renal carbonic anhydrase. Am J Physiol Renal Fluid Electrolyte Physiol 243:F311-F324,1982 .( d) M. b3 M7 A: ~0 _
) e' F  O3 i$ C/ t: u( G% u2 y5 Q/ g

11. Dobyan, DC,Magill LS,Friedman PA,Hebert SC,andBulger RE. Carbonic anhydrase histochemistry in rabbit and mouse kidneys. Anat Rec 204:185-197,1982 .
" Q$ T4 Z5 H4 l6 o4 H" z

12. Dudov, KP,andPerry RP. The gene family encoding the mouse ribosomal protein L32 contains a uniquely expressed intron-containing gene and an unmutated processed gene. Cell 37:457-468,1984  .- r6 q* s+ f8 d+ V" L5 ]

* Q8 R1 i% w) [8 c
13. Kaissling, B,andKriz W. Structural analysis of the rabbit kidney. Adv Anat Embryol Cell Biol 56:1-121,1979 .' y: M2 T" ?7 X+ s

& E+ K4 o$ U& O
14. Kivela, A,Parkkila S,Saarnio J,Karttunen TJ,Kivela J,Parkkila AK,Waheed A,Sly WS,Grubb JH,Shah G,Tureci O,andRajaniemi H. Expression of a novel transmembrane carbonic anhydrase isozyme XII in normal human gut and colorectal tumors. Am J Pathol 156:577-584,2000 .# Q- ~: ~* ]/ X3 A- \* I

15. Lonnerholm, G. Histochemical demonstration of carbonic anhydrase activity in the rat kidney. Acta Physiol Scand 81:433-439,1971  .) q& Z/ U/ A! n, W
- C* z1 @2 ?6 i& d/ e
0 N& s- o3 E( v& ]$ \: W- ?! m) k
! [# j! h: Q/ N* T
16. Madsen, KM,andTisher CC. Structural-functional relationships along the distal nephron. Am J Physiol Renal Fluid Electrolyte Physiol 250:F1-F15,1986 .

, C: f: V2 f- }6 Z& b) Y8 d7 O8 ^- P
: Q4 D* N+ L6 g5 G, n
17. Mankus, R,Schwartz JH,andAlexander EA. Acidification adaptation in cultured inner medullary collecting duct cells. Am J Physiol Renal Fluid Electrolyte Physiol 264:F765-F769,1993 .5 Q9 |( \( a! q2 T
' U2 m8 p& C3 U4 e' S2 i- \
  i2 y4 ]2 I9 X* H! E8 G/ S3 h

18. Maren, TH. A simplified micromethod for the determination of carbonic anhydrase and its inhibition. J Pharmacol Exp Ther 130:26-29,1960 .
2 h# t! t. ~. M& G5 H  |3 z
# @6 _, R1 |% h# @

19. Maren, TH,Wynns GC,andWistrand PJ. Chemical properties of carbonic anhydrase IV, the membrane-bound enzyme. Mol Pharmacol 44:901-905,1993 .# K. _, H" H: q  P) c# R! p- ^

20. Matsumura, M,Signor G,andMatthews BW. Substantial increase of protein stability by multiple disulphide bonds. Nature 342:291-293,1989 .9 s6 y9 _" W% e& u
$ b4 u6 q; Y- Z8 c0 g7 h  F6 m

: R0 I- f  k1 b$ @7 w  e% z
21. McKinley, DN,andWhitney PL. Particulate carbonic anhydrase in homogenates of human kidney. Biochim Biophys Acta 445:780-790,1976 .
% c1 s  O1 C( t: B# X

7 d- `; |1 r3 X  T
22. Mori, K,Ogawa Y,Ebihara K,Tamura N,Tashiro K,Kuwahara T,Mukoyama M,Sugawara A,Ozaki S,Tanaka I,andNakao K. Isolation and characterization of CA XIV, a novel membrane-bound carbonic anhydrase from mouse kidney. J Biol Chem 274:15701-15705,1999 .( Z3 C+ ?0 U- l' w- ~3 v

23. Nudel, U,Zakut R,Shani M,Neuman S,Levy Z,andYaffe D. The nucleotide sequence of the rat cytoplasmic -actin gene. Nucleic Acids Res 11:1759-1771,1983 .
& ]7 n; H  T+ G& F
' O6 X$ Z2 @4 N" g" N

24. Parkkila, S,Parkkila AK,Saarnio J,Kivela J,Karttunen TJ,Kaunisto K,Waheed A,Sly WS,Tureci O,Virtanen I,andRajaniemi H. Expression of the membrane-associated carbonic anhydrase isozyme XII in the human kidney and renal tumors. J Histochem Cytochem 48:1601-1608,2000 .

8 r; r3 H# m. G% J5 j1 m* e
25. Pastorek, J,Pastorekova S,Callebaut I,Mornon JP,Zelnik V,Opavsky R,Zat'ovicova M,Liao S,Portetelle D,Stanbridge EJ,Zavada J,Burny A,andKettmann R. Cloning and characterization of MN, a human tumor-associated protein with a domain homologous to carbonic anhydrase and a putative helix-loop-helix DNA binding segment. Oncogene 9:2877-2888,1994  .; I7 S/ f) |) K' W
) X1 r6 ~' F1 W+ b3 r, X% V

* x& L6 {- Q" F; T
26. Powell, SK,Cunningham BA,Edelman GM,andRodriguez-Boulan E. Targeting of transmembrane and GPI-anchored forms of N -CAM to opposite domains of a polarized epithelial cell. Nature 353:76-77,1991 .* g  G- N; w- a0 a

$ T. ^& w' f& \9 m: O8 _

27. Ridderstrale, Y,Wistrand PJ,andTashian RE. Membrane-associated carbonic anhydrase activity in the kidney of CA II-deficient mice. J Histochem Cytochem 40:1665-1673,1992 .
, h0 J1 Y5 I5 \0 Q4 y
" B8 M) u. u! Q6 r) j8 {! G" o2 P
3 Z* \$ ?- G# i# H' w5 l
28. Schwartz, GJ,Brown D,Mankus R,Alexander EA,andSchwartz JH. Low pH enhances expression of carbonic anhydrase II by cultured rat inner medullary collecting duct cells. Am J Physiol Cell Physiol 266:C508-C514,1994 ." [. C$ E( L0 w' p

  m7 I& a9 T( f9 v/ y3 L# \
29. Schwartz, GJ,Kittelberger AM,Barnhart DA,andVijayakumar S. Carbonic anhydrase IV is expressed in H -secreting cells of the rabbit kidney. Am J Physiol Renal Physiol 278:F594-F904,2000.' [3 U. j# A1 Z. ~

8 l" D4 N+ B2 P0 o2 w' E

30. Schwartz, GJ,Kittelberger AM,Sauer E,andO'Reilly MA. Apical expression of carbonic anhydrase (CA) in kidney cells (Abstract). J Am Soc Nephrol 12:9A,2002.
8 G& s7 _# G/ J' p
# g0 ^9 F6 u( E: Z

31. Schwartz, GJ,Olson J,Kittelberger AM,Matsumoto T,Waheed A,andSly WS. Postnatal development of carbonic anhydrase IV expression in rabbit kidney. Am J Physiol Renal Physiol 276:F510-F520,1999 .
1 P, z$ \3 M; n

: e; D4 ^% [0 ?! Q+ b
32. Schwartz, GJ,Winkler CA,Zavilowitz BJ,andBargiello T. Carbonic anhydrase II mRNA is induced in the rabbit kidney cortex during chronic metabolic acidosis. Am J Physiol Renal Fluid Electrolyte Physiol 265:F764-F772,1993 .. h) P6 b' O) K3 S# h
6 Z+ W8 e+ ]! b

1 D. x$ e2 @8 X/ l+ }: A# r; b
33. Tsuruoka, S,Kittelberger AM,andSchwartz GJ. Carbonic anhydrase II and IV mRNA in rabbit nephron segments: stimulation during metabolic acidosis. Am J Physiol Renal Physiol 274:F259-F267,1998 .

2 z- o2 ~  P5 y6 q+ w0 Q6 S

34. Tsuruoka, S,Swenson ER,Petrovic S,Fujimura A,andSchwartz GJ. Role of basolateral carbonic anhydrase in proximal tubular fluid and bicarbonate absorption. Am J Physiol Renal Physiol 280:F146-F154,2001 .# f1 m% A6 \/ A
; i$ \3 X, {$ K6 I% U# i5 i
/ N+ D/ ?; |# k3 j) ^$ r

35. Tureci, O,Sahin U,Vollmar E,Siemer S,Gottert E,Seitz G,Parkkila AK,Shah GN,Grubb JH,Pfreundschuh M,andSly WS. Human carbonic anhydrase XII: cDNA cloning, expression, and chromosomal localization of a carbonic anhydrase gene that is overexpressed in some renal cell cancers. Proc Natl Acad Sci USA 95:7608-7613,1998 .
/ U8 N- R0 K! g8 C" K

& _9 s* _7 V* z$ ~" |. A4 I/ P( [. V
36. Waheed, A,Zhu XL,andSly WS. Membrane-associated carbonic anhydrase from rat lung: purification, characterization, tissue distribution, and comparison with carbonic anhydrase IVs of other mammals. J Biol Chem 267:3308-3311,1992 .
/ a0 b) f# f; i; x$ y

37. Whitney, PL,andBriggle TV. Membrane-associated carbonic anhydrase purified from bovine lung. J Biol Chem 257:12056-12059,1982 .6 N- @3 |. k. W0 A2 J5 ~2 ]( ^5 U

2 T& x+ T$ d8 h# x8 c1 ]+ }
4 X0 u" C/ o2 Z* y1 x" q  g
38. Winkler, CA,Kittelberger AM,andSchwartz GJ. Expression of carbonic anhydrase IV mRNA in rabbit kidney: stimulation by metabolic acidosis. Am J Physiol Renal Physiol 272:F551-F560,1997 .0 ~; j% u& B5 ?, P; [: z
- c0 d7 }8 z) I& q1 i9 V
' k9 g$ i. d$ B/ I" c
8 c2 M* z6 J) Q9 b/ ?' {" m
39. Winkler, CA,Kittelberger AM,Watkins RH,Maniscalco WM,andSchwartz GJ. Maturation of carbonic anhydrase IV expression in rabbit kidney. Am J Physiol Renal Physiol 280:F895-F903,2001 .

40. Wistrand, PJ,andKnuuttila KG. Renal membrane-bound carbonic anhydrase. Purification and properties. Kidney Int 35:851-859,1989  .2 G$ z+ r  m' X3 ^+ ~

; Z/ h# ?; m$ ?5 @
1 E1 R9 k  \8 W2 e: f
41. Zhu, XL,andSly WS. Carbonic anhydrase IV from human lung: purification, characterization, and comparison with membrane carbonic anhydrase from human kidney. J Biol Chem 265:8795-8801,1990 .

收藏0 分享0 顶0 踩0