Contributed by William S. Sly, November 6, 1991. ABSTRACT. We have ...... Lloyd, J., McMillan, S., Hopkinson, D. & Edwards, Y. H. (1986). Gene 41, 233-239.
Proc. Natl. Acad. Sci. USA
Vol. 89, pp. 1315-1319, February 1992 Genetics
Human carbonic anhydrase IV: cDNA cloning, sequence comparison, and expression in COS cell membranes (bicarbonate transport/membrane protein/glycosyl-phosphatldylinositol anchor)
TORAYUKI OKUYAMA, SEIJI SATO, XIN LIANG ZHU, ABDUL WAHEED, AND WILLIAM S. SLY* Edward A. Doisy Department of Biochemistry and Molecular Biology, St. Louis University School of Medicine, 1402 South Grand Boulevard, St. Louis, MO 63104
Contributed by William S. Sly, November 6, 1991
With antibodies raised against human CA II and human lung CA IV, we showed that both CA II and CA IV could be demonstrated in membrane vesicles isolated from normal human urine (13). Urinary membranes from patients with CA II deficiency syndrome lacked CA II but had normal levels of CA IV (13). Studies with antibody to rat lung CA IV defined the distribution of CA IV in rat kidney (14). CA IV was confined to apical and basolateral membranes of proximal tubule cells and cells of the thick ascending limb of Henle, both of which are reported to be sites of bicarbonate reabsorption in the rat nephron (15). Another study indicated that CA IV is expressed in many other rat tissues in addition to lung and kidney (A.W., unpublished observation). Studies on tissues from the human eye demonstrated CA IV immunostaining in lens fibers and in a specific capillary bed, the choriocapillaris (16). As a first step in defining the molecular genetics of human CA IV, we isolated and characterized the CA IV cDNA. In this report, we present the cDNA sequence for CA IV, compare its deduced amino acid sequence with the amino acid sequences of other members of the CA gene family, and characterize the enzyme expressed from the cDNA in COS cells.t
We have isolated a full-length cDNA for ABSTRACT human carbonic anhydrase IV (CA IV) from a AgtlO human kidney cDNA library. The 1105-base-pair (bp) cDNA contains a 47-bp 5' untranslated region, a 936-bp open reading frame, and a 122-bp 3' untranslated region. The deduced amino acid sequence is colinear with the N-terminal sequence and the sequence of several tryptic peptides of human lung CA IN. It includes an 18-amino acid signal sequence, a 260-amino acid region that shows 30-36% similarity with the 29-kDa cytoplasmic CAs (CA I, CA H, and CA I), and an additional 27-amino acid C-terminal sequence that ends in a 21-amino acid hydrophobic domain. Of the 17 "active site" residues that are highly conserved in other human CAs, 16 are also present in CA IV. Expression of the cDNA in COS cells produced a 35-kDa enzyme that was membrane associated, resistant to inactivation by SDS, contained no carbohydrate, and reacted on Western blots with antiserum to the 35-kDa CA IV from human lung. Treatment of membranes from transfected COS cells with phosphatidylinositol-specific phospholipase C released 20-30% of the expressed enzyme from membranes, indicating that at least 20-30% of the expressed enzyme was anchored to membranes by a glycosyl-phosphatidylinositol linkage.
The carbonic anhydrases (CAs) are a family of zinc metalloenzymes that catalyze the reversible hydration of CO2 in the reaction CO2 + H20 ;± HCO3 + H+. They vary in physicochemical properties, in sensitivity to various inhibitors, and in their subcellular localization. Cytoplasmic (CA I, CA II, CAIII, and CA VII) (1, 2), cell-surface membrane (CA IV) (3-5), mitochondrial (CA V) (6), and secretory (CA VI) (7-9) isozymes have been described. The physiological importance of the cytosolic isozyme CA II became clear when it was shown that CA II deficiency was the basis for the inherited syndrome of osteopetrosis with renal tubular acidosis and cerebral calcification (10, 11). In an effort to understand fully the pathophysiology of this syndrome, which we proposed was due to a mutation that affected CA II but not CA IV (12), we purified and characterized the 35-kDa membrane-associated CA IV from human lung and kidney, which proved to be indistinguishable (5). CA IV resembled CA II in that it was a high activity isozyme, was relatively sensitive to inhibition by sulfonamides, and was relatively resistant to inhibition by halide ions. However, CA IV was unique in its resistance to inactivation by SDS, a property that greatly facilitated its purification. It was also unique in its association with membranes. The fact that CA IV was released from membranes by treatment with phosphatidylinositol-specific phospholipase C (PI-PLC) indicated that much of it was anchored to membranes by a glycosylphosphatidylinositol (GPI) linkage (5).
MATERIALS AND METHODS The human kidney Agt10 cDNA library was kindly provided by Graeme I. Bell (Howard Hughes Medical Institute, Chicago) (17) and the human lung Agtll cDNA library was purchased from Clontech. The random-primed DNA labeling kit and peptide N-glycosidase F (PNGase F) were obtained from Boehringer Mannheim. Homogeneous human lung CA IV and rabbit anti-human lung CA IV antibody were purified as described (5). Membrane preparations from rat muscle and bovine lung were obtained as described (5). Immobilon-P poly(vinylidene difluoride) membrane was purchased from Millipore. PI-PLC was from Bacillus thuringiensis, [a-32P]dCTP (3000 Ci/ml; 1 Ci = 37 GBq), and [y-32P]dATP (3000 Ci/ml) were from ICN. Screening the cDNA Library. The Agtll human lung cDNA library was screened by using the 53-base-pair (bp) oligonucleotide BP. Oligonucleotide BP (5'-CCCTTCTCCTTCTCGTGCACGATGTGCATCTCCATGGCGAAGTGCTCGCCGTC-3') was synthesized from the cDNA sequence predicted from a codon usage table (18) from a peptide sequence derived from microsequencing tryptic peptides of purified human lung CA IV. A partial cDNA clone (Agtll-L1) was obtained. The insert from Agtll-L1 was used to screen Abbreviations: CA, carbonic anhydrase; PI-PLC, phosphatidylinositol-specific phospholipase C; GPI, glycosyl-phosphatidylinositol; PNGase F, peptide N-glycosidase F. *To whom reprint requests should be addressed. tThe sequence reported in this paper has been deposited in the GenBank data base (accession no. M83670).
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.
1315
Genetics:
1316
Proc. Natl. Acad. Sci. USA 89 (1992)
Okuyarna et A
the AgtlO human kidney cDNA library from which a fulllength clone AgtlO CA IV-1 was obtained. Plaque hybridization was carried out essentially as described by Maniatis and colleagues (19). The prehybridizations were carried out for 4 h at 50°C in a solution of 50% formamide/5x SSPE (ix SSPE is 150 mM NaCl/10 mM NaH2PO4/1 mM EDTA)/5 x Denhardt's solution (ix Denhardt's solution is 0.02% each Ficoll, polyvinylpyrrolidone, and bovine serum albumin fraction V)/0.1% SDS/100 ,ug of denatured salmon sperm DNA per ml. Hybridizations were performed for 16 h at 50°C after adding the 32P-labeled probe at -1 x 106 cpm/ml into the solution used for prehybridizations. [32P]DNA probes were oligonucleotide BP, labeled with T4 polynucleotide kinase (19) and [y-32P]dATP, and the cDNA insert L1, labeled with [a-32P]dCTP by a randomprimer labeling method. After hybridization, filters were washed twice in 2x SSC/0.1% SDS at room temperature and twice in O.lx SSC/0.1% SDS for 10 min each at 60°C. DNA Sequence Analysis. The DNA insert of AgtlO CA IV-1 was sequenced by the Sanger dideoxynucleotide chaintermination method (20). After subcloning into the EcoRI site of M13mpl9 vectors, we used the M13 universal primers and several synthetic oligonucleotides for sequencing. The complete insert was sequenced at least once on both strands. Expression of Human CA IV cDNA in COS-7 Cells. We subcloned the insert of AgtlO CA IV-1 into the EcoRI site of expression vector PMT21, the generous gift of Randy Kaufman (Genetics Institute, Cambridge, MA). COS-7 cells (21) were transfected by the DEAE-dextran procedure (22) followed by treatment with 100 ,uM chloroquine (23). Seventy-two hours posttransfection, cells were homogenized in 50 mM Tris SO4 (pH 7.5) containing 1 mM benzamidine (buffer A). The cell membranes and cytosol were recovered by centrifugation of the cell homogenate at 100,000 X g for 30 min. The membrane pellet was suspended in buffer A.
CA Assay. CA activity was determined by the procedure of Maren (24) as described (25). SDS-resistant CA activity was measured on samples preincubated with 0.2% SDS at room temperature prior to assay. CA activity was also determined in the presence of 1 ,uM acetazolamide. The protein concentrations were determined by the Lowry method (26) using bovine serum albumin as a standard. PI-PLC treatments of COS-7 cell membranes and rat -nuscle membranes were as described (5). Endoglycosidase Jigestion of the membranes was carried out as described (5). Immunoblotting. SDS/PAGE under reducing conditions was carried out as described (27). After SDS/PAGE, the polypeptides were transferred to a poly(vinylidene difluoride) membrane as described (5). The immunoblots were stained with rabbit anti-human lung CA IV antiserum diluted 1:1000 and goat anti-rabbit IgG peroxidase-conjugated second antibodies diluted 1:500. 4-Chloro-1-naphthol and hydrogen peroxide were used to stain for peroxidase activity.
RESULTS Isolation of cDNA for Human CA IV. The clone Agtll-Ll, isolated by screening a Agtll human lung cDNA library with the 53-bp oligonucleotide BP, contained a partial CA IV cDNA. We used the insert from Agtll-L1 to screen 6 x 105 plaques from a AgtlO human kidney cDNA library under stringent hybridization conditions. Thirty-seven clones were identified, 10 of which were analyzed by PCR (28) using AgtlO forward and reverse primers. The two that yielded the largest PCR product (1.2 kilobases) seemed identical on 1% agarose gel analysis. The insert from one of them (AgtlO CA IV-1) was sequenced and shown to contain a full-length cDNA. Nudeotide Sequence and Deduced Amino Acid Sequence. The cDNA (Fig. 1) is 1105 bp long. It contains a 47-bp 5' untranslated region, a 936-bp open reading frame, and a 122-bp 3' untranslated region containing a polyadenylylation
Met Arg Met Leu Leu Ala Leu Leu -11
GCTCGGTGCGCGACCCCGGCTCAGAGGACTCTTTGCTGTCCCGCAAG ATG CGG ATG CTG CTG GCG CTC CTG Ala Leu Ser Ala Ala Arg Pro Ser Ala Ser Ala Glu Ser lis a.n +10 GCC CTC TCC GCG GCG CGG CCA TCG GCC AGT GCA GAG TCA CAC TGG \J GAG GTT CAA Ala (flu 5cr 5cr nTyr Pro GCC GAG TCC TCC AAC TAC CCC
Asp Arg Gln GAC CGC CAG
Leu Val Pro Val Lys Trp Gly Gly Ason TTG GTG CCA GTC AAG TGG GGT GGA AAC
Gln Lys CAG AAG
30
rPro Ile Asn Ile Val Thr Thr Lys Ala Lys Val Asp Lys Lys Leu Gly CCC ATC AAC ATC GTC ACC ACC AAG GCA AAG GTG GAC AAA AAA CTG GGA
50
Arg Phe Phe Phe Ser Gly Tyr Asp Lys Lys Gln Thr Trp Thr Val Gln CGC TTC TTC TTC TCT GGC TAC GAT AAG AAG CAA ACG TGG ACT GTC CAA [
n
Asn Gly His
Ser Val Met Met Leu Leu Glu Asn Lys Ala Ser Ile Ser Gly Gly Gly Leu Pro Ala Pro TCA GTG ATG ATG TTG CTG GAG AAC AAG GCC AGC ATT TCT GGA GGA GGA CTG CCT GCC CCA *
Tyr Gln Ala Lys
TAC CAG GCC AAA CGJTTG
70
AAC GGG CAC
90
*
_ui _ r_ CTG C
His Ph Ser Leu ap G1Y AGC CTC GAT GGG GAG CAC TTT GCC ATG
_CY Lys Gly Ser TGG TCC GAC TTG CCA TAT AAG GGC TCG
His 110
Val His Gbu Lys GI 130 EATG CC ATA GTA CAT GAG AAA GAG AAG GGG
Thr Ser Arg Asn Val Lys Glu Ala Gln Asp Pro Glu Asp Glu Ile Ala V[l Leu Ala Phe 150 ACA TCG AGG AAT GTG AAA GAG GCC CAG GAC CCT GAA GAC GAA ATT GCG CTG GCC TTT Leu Val Glu Ala Gly Thr Gln Val Asn Glu Gly Phe Gln Pro Leu Val Glu Ala Leu Ser 170 CTG GTG GAG GCT GGA ACC CAG GTG AAC GAG GGC TTC CAG CCA CTG GTG GAG GCA CTG TCT Asn Ile Pro Lys Pro Glu Met Ser Thr Thr Met Ala Glu Ser Ser Leu Leu Asp Leu Leu 190 AAT ATC CCC AAA CCT GAG ATG AGC ACT ACG ATG GCA GAG AGC AGC CTG TTG GAC CTG CTC Pro Lys Glu Glu Lys Leu Arg His Tyr Phe Arg CCC AAG GAG GAG AAA CTG AGG CAC TAC TTC CGC T
Lou Gly Ser Leu rhr Pro Thr 210 C CTG GGC TCA CTC [C ACA CG ACC
- IThr- ValPahe Arg Glu Pro Ile Gln Leu His Arg Glu Gln 230 T Asp Glu Lys Val GAT GAG AAG GTC GTC TGG ACT GTG TTC CGG GAG CCC ATT CAG CTT CAC AGA GAA CAG
Ile Leu Ala Phe Ser Gln Lys Leu Tyr Tyr Asp Lys Glu Gln Thr Val Ser Met Lys Asp 250 ATC CTG GCA TTC TCT CAG AAG CTG TAC TAC GAC AAG GAA CAG ACA GTG AGC ATG AAG GAC Asn Val AAT
GTC
Pro Leu Gln Gln Leu Gly Gln Arg Thr Val Ile Lys Ser Gly Ala Pro Gly 270 CTG CAG CAG CTG GGG CAG CGC ACG GTG ATA AAG TCC GGG GCC CCG GGT
CCC
Arg Pro Leu Pro Trp Ala Leu Pro Ala Leu Leu Gly Pro Met Leu Ala CGG CCG CTG CCC TGG GCC CTG CCT GCC CTG CTG GGC CCC ATG CTG GCC
3
Leu Leu Ala 290 CTG GCC
CTG
Gly Phe Leu Arg* GGC TTC CTG CGA TGA TGGCTCACTTCTGCACGCAGCCTCTCTGTTGCCTCAGCTCTCCAAGTTCCAGGCTTCCG GTCCTTAGCCTTCCCAGGTGGGACTTTAGGCATGAZMyAATATGGACATATTTTTGGAG
-
poly
(A)
FIG. 1. DNA sequence and deduced amino acid sequence of the CA IV cDNA. The deduced amino acid sequence is numbered at the end of each row. Amino acids -18 to -1 comprise a putative signal peptide. The arrowhead preceding amino acid + 1 indicates the N-terminal residue found in the mature protein. Amino acid residues that have been determined by protein sequence analysis of purified lung CA IV are underlined (ref. 5; unpublished data). A presumptive
polyadenylylation signal (AT-
TAAA) is underlined. Amino acid residues common to all human CAs (CAs I, II, III, VI, VII) in active sites are boxed. Three potential Zn-liganded histidine residues are designated by an asterisk.
Genetics: Okuyama et al.
Proc. Natl. Acad. Sci. USA 89 (1992)
cleavage signal (ATTAAA) starting 26 bp upstream from the poly(A) tail. The deduced 312-amino acid sequence is colinear with the sequence of three tryptic peptides of CA IV from human lung (5), except for one residue at amino acid position 6. The deduced amino acid sequence of positions 113-130 agreed completely with the peptide sequence determined by microsequencing. Although the 53-bp oligonucleotide probe turned out to differ at 7 positions from the actual nucleotide sequence of the human CA IV cDNA, it proved successful as a hybridization probe. The open reading frame upstream of the N terminus of the mature protein (the first amino acid of which is assigned position 1) includes two methionine residues at positions -16 and -18 (Fig. 1). Translation is assumed to begin at the methionine at position -18, because its surrounding nucleotide sequence better agrees with the consensus sequence surrounding ATG initiator codons in eukaryotic mRNAs (29, 30). In addition, the -18 methionine is followed by an arginine, a positively charged residue in a characteristic position for signal peptides for secretory proteins (31). Additional characteristic features of the 18-amino acid signal peptide include a secondary structure-breaking proline residue in position -4 and a central hydrophobic region from -15 to -9 (32). The sequence of human CA IV contains no N-glycosylation sites (Asn-Xaa-Thr/Ser). Five cysteine residues are present at positions 6, 18, 28, 211, and 286. Some of these are probably involved in intramolecular disulfide bond formation, which stabilizes the CA IV structure (5). The hydrophilicity profile reveals a hydrophobic sequence of 21 amino acids (positions 273-293) followed by an arginine at the C terminus of CA IV. This hydrophobic sequence is of sufficient length to span the membrane and may also contain, or be preceded by, a signal for cleavage and transfer of the CA IV to a GPI (33). 1
HCAI II III VI VII IV
HCAI II III VI VII IV
HCAI II III VI VII IV
HCAI II III VI VII IV
HCAI II III VI VII
IV HCAI II III VI VII IV
a S S A QH V G H AE
P H IC S H S
D WG Y HWG Y ENGY D W TY GWGY H WC Y
D G A S G E
D IC S E Q V
KC N G P CE
P I P L P W G L P.L R F
S S S N E F
V V V M L F
S -Y N S -Y D S -Y D T G Y E S - Y E S G Y D
P Q G T a K
A A G Q C K
T T S A M
G --S G --S G --S G G A S G - - K
H NG P H N G P GALD - D D G QAE S
L L V T
H W H W H W H W H W HW
H G S E H T V QG S E HTV H G S E H T V S G S E H TV V G S E H T V S-H-DLPYKIGSEHSL
D D D D D
G G G A V A
V I I A F F
L F F F L L
G G G Y D G T
Q Q Q S N Q
F Y Y Y Y Y
W T YP GSL T H PP W T|Y|P|G S|L|T|T|P|P W T|Y|Q|G S|F|T|T|P|P Y T|Y|H|G S|L|T|T|P|P W T|Y|P|G S|L|T|T|P|P rR SIL
Q H V D V S - N V N K D
T L S S K
N D D E H
M - - K V L - - K V L - - K I V E V K N E - - T G V - - E A
E G D I D
E S H P E
Q
A A E E H V
O0
JSL|G
K L K E L W
[!T[!JT
N K N N P N
P P G T S E
K G E Y G
L L F Y M F
S IC L Y P H Kt D F P H E L r P P QH Y P H K L Y P P C L V P
I A -N I A -K N a- K A C- G I A V IC W G
~40
I--
GE N - -- -- -- N Q S P V D I K S S E T K H G 15 RQ S P VD I D T H T A K Y G E ---- N Q s P I E L H T K D I R H G Q - - RQS P I N L QRT KV R Y G - - - - D RQ S P I N I I S S Q AV Y G N C Q K D R QSPI N I V T T K A KC V
D D D N S D
T P P P P K
S S S S S K
L L L L L
K K K Q LE G
Q
IT V G H S r H V N F E D N D N R S V L IC G G P r S D S _ R L r Q F H F
E -I R -I T -I F P M S -I T -V
A S A G S QT
Comparison of Human CA IV Amino Acid Sequence with That of Other Human CAs. The amino acid sequences of six isozymes of human CA are compared in Fig. 2. Forty-three residues (17%) are common to all six human isozymes. Sixteen of the 17 highly conserved "active site" residues found in most other CAs (38, 39) are present in CA IV. The exception is one highly conserved proline (position 202 of CA I), which is replaced by a threonine (threonine-210 of CA IV) (Figs. 1 and 2). The percentage similarities between CA IV and other CAs are 31% for CA I, 36% for CA II, 30%o for CA III, 33% for CA VI, and 32% for CA VII (using the MACVECTOR 3.5 program, IBI). CA Activity in COS Cells Transfected with CA IV cDNA. The expression of human CA IV activity in transfected COS-7 cells was determined by measuring the CA activity in the cell homogenate in the presence of 0.2% SDS, which inactivates the soluble CAs (5). The results are shown in Table 1. All of the CA activity was inhibited with 1 AuM acetazolamide (40) (results not shown). There was no increase in CA activity when 0.1% saponin was added (in the absence of SDS), suggesting that all of the CA activity is on the external surface of the membrane vesicles (13). There was '17-fold more SDS-resistant activity in transfected COS-7 cells than in cells transfected with vector only. When the transfected COS-7 cell homogenate was separated into a membrane pellet and cytosol by centrifugation, almost all of the SDS-resistant CA activity (99%o) was associated with the membrane. Fig. 3 shows the results of a Western blot analysis of the total cell extracts of transfected COS cells. Cells transfected with the CA IV cDNA-containing vector contained a 35-kDa immunoreactive CA IV that was not seen in cells transfected with vector only and that migrated like authentic human lung CA IV.
20
W E -H Wf D -H W EAHW P S HW S - N Y
1317
N N N N QN
N N N N N
G G G G G
H A r KC T C
G K G G G
V K V I K
K K K R S
Y Y Y H F
N R H TV Q H S V Q H S VM
V v I V M
E rD D v r D D G L P S D F N D L L E N
S AE LHVAH A A|E|L|H|L V|H| A A|E|L|H|L V|H| V I |E|I|H|I V|H| P S|E|L|H|L V|H|
GEEHHFADMEMLFIPVK K I N R P
160 V V F F L L
L V L I T V
L Y ES V T WI L L|E|C V T|W|I C E|E|C I V|W|L C T|E|N V H|W|F L S|E|S V T|W|I C DEJK,V VIWT
D A LQ D V|L|D D A|L|D S H|L|A D A|L|Y E A LS I V L V V V
S QD K T Y D R TM RM S D D R --K A
C L L L L F
K K K A R R
E E E D E E
I I I I Y I
K K K K R P
T T T Y F K
K K K P K P
S P P F P P
I I M V I I
S S T K C
V V V L I L
S S S S S H
Q
V M V V I
L L A V S
W W W Y W
N N N N N
S A KY T -K Y P -KI Y S -K Y A K K Y -KGTS
K K K Q T M
R S E R K S
A A A T A T
P D P T Q T
F r F L F M
220 S Z QL S E|Q|V S D|Q|M R T|Q|V E R|Q|M R E 0I
A L A W G L
QF
E
A S K N M N
A S T T S
G G G G G E
T T T T S A
K R T G
G G D G G
S S G D N T IC T S T RN N N K G C E
F F F L F S
G P L D G G P L P G G - - - I G P L E G G - L P A
T P V P P
Y Y Y Y
R R I R
L L A L _Q A
I
R Q K K
QF QF
H F H L
QM H F QFHF Q L H L
L A EA A S K AD G LA I r G KCIA V Q Q PIDIG LIA VIL F K E A L K Q RIDIG IIA VII Y D I IAIQ D A PIDIG LIA VIL F G EIAIA S A P|D|G L|A V|G TVKEQDPEEIAVL
1s0 D D D D N S
P P P V P L
S R S Q K L
T G C D C D
L L L M L L
L P S L |P |E F IP |A L|P|R L |P |A L P K
S S C N S E
- R S LL S N V E G D K F - R K|L|N r N G E G E K L - R SILIL S S A E N E K L E N S|L|L D H R N K T K r - R S|L|L r T S E D D A r S Q K L- - Y Y D K E
L - - - D L - - - D R - - - D L Q - - H - - - R H E K L R H N P P I E
A E P H R Q T
V E V I V
240 P M L M P L - -
H M S M
N NEGEP V R A S.Fr T P L K GRTT N WIRIP AIQ P L K NIRIQ I K A S F K N WIRIPPI P I N V RA S rK D Y|R|R T|QP L NIC H|R|V V E S N r P N Q E Y T L G S E r Q r Y L H K I E E I L D Y L R R A L N N FRPP PP KQP V K A S rR A L K GKRVV N V-IJP LIQ L TRT V I K S G A P G R P L P W A L P A L L G P M L A C L L A G L R
NSREV
FIG. 2. Alignment of amino acid sequences of human CAs I, II, III, IV, VI, and VII. Residues that are conserved in all six isozymes are boxed. Gaps were introduced into the sequences to optimize homology. The numbering system used here is based on that of human CA I. The sequence data are from refs. 9, 34-37, and 50.
1318
Genetics:
Okuyama et al.
Proc. Natl. Acad. Sci. USA 89 (1992)
Table 1. CA activity in transfected COS-7 cells CA activity, units per mg of protein SDS Saponin Cell homogenate (0.1%) (0.2%) Cells 0.27 0.37 0.74 PMT21 4.54 4.88 6.24 PMT21 + CA IV cDNA Values are averages of two measurements.
Results of Treating Expressed CA IV with PI-PLC and Endoglycosidase. When a membrane suspension isolated from transfected COS-7 cells expressing CA IV was incubated with PI-PLC for 4 h and analyzed for soluble and membrane-bound enzyme, 20-30o of the enzyme was released from the membrane (Fig. 4A). Increasing the time of incubation to 24 h did not increase the fraction of membranebound enzyme released. In a parallel experiment, CA IV from rat muscle membrane, a control for GPI-anchored enzyme, was completely released from the membrane by this treatment (results not shown). Membranes from transfected COS-7 cells expressing CA IV were also treated with PNGase F and studied for deglycosylation by SDS/PAGE and immunoblotting (Fig. 4B). There was no shift in the relative mobility of the expressed human CA IV polypeptide after PNGase F treatment. This result indicates that the human CA IV expressed in COS cells contains no N-linked oligosaccharide, which is consistent with the deduced amino acid sequence reported here, but exceptional among mammalian CA IVs. The bovine lung CA IV, which is thought to contain five or six N-linked oligosaccharides (3), was reduced from 52 to 36 kDa, indicating removal of oligosaccharide chains when treated by PNGase F under the same conditions (see last two lanes of Fig. 4B). By this criterion, every other mammalian CA IV examined (eight others in addition to bovine) also contained some N-linked oligosaccharide (A.W., unpublished observation).
acid hydrophobic domain and presumably specifies its relationship to membranes. The protein expressed from the cDNA in COS cells has the same mass as CA IV from human lung. Furthermore, the expressed recombinant CA IV also resembles CA IV from human lung in its hydrophobicity, its association with the plasma membrane, its sensitivity to inhibition by acetazolamide, and its relative insensitivity to inhibition by halide anions and to inactivation by SDS. CA IV in human lung and kidney is anchored to membranes by a GPI anchor (5). The GPI linkage has been suggested to be a mechanism for targeting proteins specifically to the apical surface of polarized cells (41). However, CA IV in rat kidney was demonstrated on both apical and basolateral plasma membranes (14). Whether it is anchored to both membranes by the same mechanism is not yet clear. PI-PLC treatment released a fraction (20-30%6) of the CA IV expressed in COS cells, indicating that some of the expressed enzyme is anchored to COS cell membranes by aGPI anchor. However, the percentage released by PI-PLC was less than the amount released from human urinary membranes (95%6) (13), all of
A
4h
24h
S
M +
M _
+
-
s +
-
+
-
CAiN -
,
p-wv~~~-a ISg o
DISCUSSION We have isolated and sequenced a full-length cDNA for CA IV from a human kidney cDNA library. The deduced amino acid sequence is colinear with the partial sequence of human lung CA IV determined by microsequencing (5). The 312amino acid protein includes an 18-amino acid classical signal sequence for a secretory protein, a 260-amino acid central segment with 30-36% similarity to the sequences of the 29-kDa cytosolic CAs (CA I, CA II, and CA III), and a 27-amino acid C-terminal extension that includes a 21-amino S
1
2
B Human --
CAIV +
Bovine 'I
CAIN +
3
92.566.2-45 0
31
G--
21.5-
FIG. 3. Immunochemical detection of CA IV. COS-7 cell extracts were transfected with vector only (lane 1) and cDNA for human
kidney CA IV (lane 2). CA IV purified from human lung (500 ng) (lane 3) was subjected to SDS/PAGE and analyzed by immunoblotting. The apparent molecular mass of standard proteins (lane S) is marked in kDa.
FIG. 4. (A) Solubilization of CA IV from membranes of cDNAtransfected COS-7 cells with PI-PLC. Membrane suspensions from cDNA-transfected COS-7 cells were treated with PI-PLC (lanes +) and with buffer alone (lanes -) for 4 and 24 h and the membrane was sedimented. Membrane-bound (lanes M) and soluble (lanes S) forms of the enzymes were analyzed by SDS/PAGE followed by immunoblotting. (B) Deglycosylation by treatment with PNGase F. Membrane suspensions from COS-7 cells transfected with cDNA of human kidney CA IV (human CA IV) were treated with PNGase F (lanes +) and buffer alone (lanes -) and the membranes were sedimented. For controls, bovine lung microsomes (bovine CA IV) were also treated under identical experimental conditions with PNGase F (lanes +) and with buffer alone (lanes -). The deglycosylation reaction was analyzed by SDS/PAGE followed by immunoblotting.
Genetics:
Okuyarna et A
which are presumed to be apical membranes, and from human lung and kidney microsomes (50-60%o) (5), which contain a mixture of apical and basolateral membranes. Variation in extent of release of GPI-linked proteins from different cell types by PI-PLC has been reported (42). Differential sensitivity to release by PI-PLC could reflect structural differences in the glycan moiety in the linkages, differences in accessibility to the PI-PLC due to surrounding membrane components, or differences in the fraction of the enzyme that is anchored by a transmembrane-spanning domain. The lower level released from COS cell membranes may simply be a property of COS cells, which are nonpolarized. Alternatively, the high level of overexpression in the transfected cells may oversaturate the capacity of the enzymatic machinery for cleavage and transfer of the nascent CA IV to a GPI anchor, leaving a larger fraction anchored by a transmembrane-spanning domain than might be the case at lower levels of gene expression in the same cells. The cDNA sequence indicates that the CA IV gene is a distinct member of the CA gene family (38, 39). The genes for CAs I, II, III, and VII have been characterized and have similar structures in that six introns interrupt the coding sequences at identical positions (38, 39). CA I has an additional 5' noncoding exon (43). The genes for CAs I, II, and III are closely linked on human chromosome 8 (34). The genes for CAs VI and VII are on chromosomes 1 and 16, respectively (44, 45). All seven mammalian CA genes are thought to have evolved from a common ancestral gene by a series of gene duplications (2, 38). Based on nucleotide and amino acid comparisons, CA IV and CA VI are thought to have diverged the earliest, with CA VII somewhat later. CAs I, II, and III diverged more recently, between 300 and 400 million years ago (38, 39). Despite its early divergence in the evolution of the CA gene family, CA IV has retained most of the residues highly conserved in other CAs (Fig. 2). Like CA VI, CA IV has acquired additional N-terminal residues that encode a typical sequence for a secretory protein. Like CA VI, CA IV has also acquired additional C-terminal residues but there is no similarity between the C-terminal extensions in CA IV and CA VI. These residues in CA IV specify its relationship to membranes and presumably contain a signal for cleavage of the nascent CA IV and transfer to a GPI anchor. Prior studies have established that patients with osteopetrosis, renal tubular acidosis, and cerebral calcification have an inherited deficiency of CA II that is explained by mutations in the structural gene for CA II (46). The availability of the cDNA sequence for CA IV should make it possible to define the genomic organization of the CA IV gene, to determine its chromosome localization, and to examine patients with specific abnormalities of bicarbonate transport such as patients with pure proximal renal tubular acidosis (45, 46) for defects in the CA IV gene. We would like to thank Jeffrey Grubb, Donna Roth, Suso Platero, and Robb Hellwig for assistance and helpful suggestions and Elizabeth Torno for editorial assistance. This work was supported by National Institutes of Health Grants GM34182 and DK40163. 1. Deutsch, H. F. (1987) Int. J. Biochem. 19, 101-113. 2. Tashian, R. E. (1989) BioEssays 10, 186-192. 3. Whitney, P. L. & Briggle, T. V. (1982) J. Biol. Chem. 257, 1205612059. 4. Wistrand, P. J. & Knuuttila, K.-G. (1989) Kidney Int. 35, 851-859. 5. Zhu, X. L. & Sly, W. S. (1990) J. Biol. Chem. 265, 8795-8801. 6. Storey, B. T., Dodgson, S. J. & Forster, R. E., 11 (1984) Ann. N. Y. Acad. Sci. 429, 210-211. 7. Feldstein, J. B. & Silverman, D. N. (1984) J. Biol. Chem. 259, 5447-5453. 8. Murakami, H. & Sly, W. S. (1987) J. Biol. Chem. 262, 1382-1388. 9. Aldred, P., Fu, P., Barrett, G., Penschow, J. D., Wright, R. D., Coghlan, J. P. & Fernley, R. T. (1991) Biochemistry 30, 569-575. 10. Sly, W. S., Hewett-Emmett, D., Whyte, M. P., Yu, Y.-S. L. & Tashian, R. E. (1983) Proc. Nati. Acad. Sci. USA 80, 2752-2756.
Proc. Natl. Acad. Sci. USA 89 (1992)
1319
11. Sly, W. S. (1989) in The Metabolic Basis ofInherited Disease, eds. Scriver, C. R., Beaudet, A. L., Sly, W. S. & Valle, D. (McGrawHill, New York), 6th Ed., pp. 2857-2866. 12. Sly, W. S., Whyte, M. P., Krupin, T. & Sundaram, V. (1985) Pediatr. Res. 19, 1033-1036. 13. Sato, S., Zhu, X. L. & Sly, W. S. (1990) Proc. Natl. Acad. Sci. USA 87, 6073-6076. 14. Brown, D., Zhu, X. L. & Sly, W. S. (1990) Proc. Natl. Acad. Sci. USA 87, 7457-7461. 15. Knepper, M. A. & Burg, M. B. (1988) in Diseases of the Kidney, eds. Schrier, J. W. & Gottschalk, C. W. (Little, Brown, Boston), pp. 211-239. 16. Hageman, G. S., Zhu, X. L., Waheed, A. & Sly, W. S. (1991) Proc. Natl. Acad. Sci. USA 88, 2716-2720. 17. Bell, G. I., Fong, N. M., Stempien, M. M., Wormsted, M. A., Caput, D. K. L., Urdea, M. S., Rail, L. B. & Sanchez-Pescador, R. (1986) Nucleic Acids Res. 14, 8427-8446. 18. Maruyama, T., Gojobori, T., Aota, S. & Ikemura, T. (1986) Nucleic Acids Res. 14, Suppl., 151-197. 19. Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual (Cold Spring Harbor Lab., Cold Spring Harbor, NY), 2nd Ed. 20. Sanger, F., Nicklen, S. & Coulson, A. R. (1977) Proc. Natl. Acad. Sci. USA 74, 5463-5467. 21. Gluzman, Y. (1981) Cell 23, 175-182. 22. Lopata, M. A., Cleveland, D. W. & Soliner-Webb, B. (1984) Nucleic Acids Res. 12, 5707-5717. 23. Luthman, H. & Magnusson, G. (1983) Nucleic Acids Res. 11, 1295-1308. 24. Maren, T. H. (1960) J. Pharmacol. Exp. Ther. 130, 26-29. 25. Sundaram, V., Rumbolo, P., Grubb, J., Strisciuglio, P. & Sly, W. S. (1986) Am. J. Hum. Genet. 38, 125-136. 26. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. (1951) J. Biol. Chem. 193, 265-275. 27. Laemmli, U. K. (1970) Nature (London) 227, 680-685. 28. Saiki, R. K., Gelfand, D. H., Stoffel, S., Scharf, S. J., Higuchi, R., Horn, G. T., Mullis, K. B. & Erlich, H. A. (1988) Science 239, 487-491. 29. Kozak, M. (1987) Nucleic Acids Res. 15, 8125-8148. 30. Kozak, M. (1986) Cell 44, 283-292. 31. Austen, B. M. (1979) FEBS Lett. 103, 308-313. 32. von Heijne, G. (1986) Nucleic Acids Res. 14, 4683-4690. 33. Cross, G. A. M. (1990) Annu. Rev. Cell Biol. 6, 1-39. 34. Davis, M. B., West, L. F., Barlow, J. H., Butterworth, P. H. W., Lloyd, J. C. & Edwards, Y. H. (1987) Somat. Cell Mol. Genet. 13, 173-178. 35. Barlow, J. H., Lowe, N., Edwards, Y. H. & Butterworth, P. H. W. (1987) Nucleic Acids Res. 15, 2386. 36. Murakami, H., Marelich, G. P., Grubb, J. H., Kyle, J. W. & Sly, W. S. (1987) Genomics 1, 159-166. 37. Lloyd, J., McMillan, S., Hopkinson, D. & Edwards, Y. H. (1986) Gene 41, 233-239. 38. Tashian, R. E., Hewett-Emmett, D. & Venta, P. J. (1991) in Carbonic Anhydrase: From Biochemistry and Genetics to Physiology and Clinical Medicine, eds. Botre, F., Gros, G. & Storey, B. T. (VCH, Weinheim, F.R.G.), pp. 151-161. 39. Hewett-Emmett, D. & Tashian, R. E. (1991) in Carbonic Anhydrase: Cellular Physiology and Molecular Genetics, eds. Dodgson, S. J., Tashian, R. E., Gros, G. & Carter, N. D. (Plenum, New York), pp. 15-32. 40. Maren, T. H. & Sanyal, G. (1983) Annu. Rev. Pharmacol. Toxicol. 23, 439-459. 41. Powell, S. K., Cunningham, B. A., Edelman, G. M. & RodriguezBoulan, E. (1991) Nature (London) 353, 76-77. 42. Roberts, W. L., Myher, J. J., Kuksis, A., Low, M. G. & Rosenberry, T. L. (1988) J. Biol. Chem. 263, 18766-18775. 43. Lowe, N., Brady, H. J. M., Barlow, J. H., Swoden, J. C., Edwards, M. & Butterworth, P. H. W. (1990) Gene 93, 277-283. 44. Montgomery, J. C., Venta, P. J., Eddy, R. L., Fukushima, Y.-S., Shows, T. B. & Tashian, R. E. (1991) Genomics 11, 835-848. 45. Sutherland, G. R., Baker, E., Fernandez, K. E. W., Callen, D. F., Alfred, P., Coghlin, J. P., Wright, R. D. & Fernley, R. T. (1989) Cytogenet. Cell Genet. 50, 149-150. 46. Venta, P. J., Welty, R. J., Johnson, T. H., Sly, W. S. & Tashian, R. E. (1991) Am. J. Hum. Genet. 49, 1082-1090. 47. Sly, W. S., Sato, S. & Zhu, X. L. (1991) Clin. Biochem. 24,311-318. 48. Winsnes, A., Monn, E., Stokke, 0. & Feyling, T. (1979) Acta Paediatr. Scand. 68, 861-868. 49. Donckerwolcke, R. A., van Stekelenburg, G. J. & Tiddens, H. A. (1970) Arch. Dis. Child. 45, 769-773. 50. Wade, R., Gunning, P., Eddy, R., Shows, T. & Kedes, L. (1986) Proc. Natl. Acad. Sci. USA 83, 9571-9575.
