Characterization of CA XIII, a Novel Member of the Carbonic Anhydrase Isozyme Family*

The carbonic anhydrase (CA) gene family has been reported to consist of at least 11 enzymatically active members and a few inactive homologous proteins. Recent analyses of human and mouse databases provided evidence that human and mouse genomes contain genes for still another novel CA isozyme hereby named CA XIII. In the present study, we modeled the structure of human CA XIII. This model revealed a globular molecule with high structural similarity to cytosolic isozymes, CA I, II, and III. Recombinant mouse CA XIII showed catalytic activity similar to those of mitochondrial CA V and cytosolic CA I, with kcat/Km of 4.3 × 107 m–1 s–1, and kcat of 8.3 × 104 s–1. It is very susceptible to inhibition by sulfonamide and anionic inhibitors, with inhibition constants of 17 nm for acetazolamide, a clinically used sulfonamide, and of 0.25 μm, for cyanate, respectively. Using panels of cDNAs we evaluated human and mouse CA13 gene expression in a number of different tissues. In human tissues, positive signals were identified in the thymus, small intestine, spleen, prostate, ovary, colon, and testis. In mouse, positive tissues included the spleen, lung, kidney, heart, brain, skeletal muscle, and testis. We also investigated the cellular and subcellular localization of CA XIII in human and mouse tissues using an antibody raised against a polypeptide of 14 amino acids common for both human and mouse orthologues. Immunohistochemical staining showed a unique and widespread distribution pattern for CA XIII compared with the other cytosolic CA isozymes. In conclusion, the predicted amino acid sequence, structural model, distribution, and activity data suggest that CA XIII represents a novel enzyme, which may play important physiological roles in several organs.

bootstrap runs with different random seeds. The dataset was bootstrapped 1000 times for each run.
Structure Modeling-The model of human CA XIII was based on the structure of the human CA I at 2.0 Å resolution (9) (PDB entry 1CZM). The model was built with the program InsightII (Accelrys Inc., San Diego, CA). Amino acid substitutions were built using a side chain rotamer library. The initial model was refined with Discover™ in a stepwise manner by energy minimization using the Amber™ force field. First, the newly built loops were refined with 500 steps of minimization with a fixed and a free backbone, respectively. Then, all side chains with a constrained backbone were minimized for 500 steps, followed by another 1000 steps of minimization for the whole model.
Kinetic and Inhibition Measurements-Initial rates of CO 2 hydration were measured with an SX.18MV-R Applied Photophysics stopped-flow instrument (Oxford, UK) by the changing pH indicator method (12). Phenol red (at a concentration of 0.2 mM) was used as an indicator, working at the absorbance maximum of 557 nm, with 10 mM Hepes (pH 7.5), 10 mM Tris, and 0.1 M Na 2 SO 4 , following the CA-catalyzed CO 2 hydration reaction for a period of 10 -100 s. Saturated CO 2 solutions in water at 25°C were used as substrate (12). Stock solutions of inhibitors (1 mM) were prepared in distilled-deionized water with 10 -20% (v/v) Me 2 SO (which is not inhibitory at these concentrations), and dilutions up to 0.1 nM were made thereafter with distilled-deionized water. Inhibitor and enzyme solutions were preincubated together for 10 min at room temperature prior to assay, in order to allow for the formation of the enzyme-inhibitor complex. Triplicate experiments were done for each inhibitor concentration, and the values reported throughout the study are the mean of such results. Inhibition constants were calculated as described earlier (13).
Antibodies-The anti-mouse CA XIII serum was raised by Innovagen AB (Lund, Sweden) in a rabbit against a synthetic peptide (AC-) DG-DQQSPIEIKTKEC (-COOH), which was designed based on the predicted amino acid sequence of mouse CA XIII. This peptide was found to be identical in the predicted human CA XIII sequence (Fig. 1), and thus, the same antibody was also used to detect the human orthologue. Both mouse and human CA13 cDNA sequences are available in GenBank TM (accession numbers AF231123 and AK095314). Rabbit antisera against human CA I and II have been produced and characterized earlier (12). Rabbit anti-mouse CA II antibody was raised against mouse CA II produced in E. coli. The specificity of the antibody was confirmed using blood lysates from both normal and CA II-deficient (Car-2 Ϫ/Ϫ ) mice (data not shown).
Western Blotting-30 g of protein from human colon, human liver, and mouse colon total homogenates or 20 l of purified fraction from ProBond™ or inhibitor affinity chromatography were analyzed by SDS-PAGE under reducing conditions according to Laemmli (13). The separated proteins were stained by Colloidal Coomassie Blue (Invitrogen, Carlsbad, CA) or transferred electrophoretically from the gel to a polyvinylidene fluoride membrane (Invitrogen) in a Novex Xcell II blot module (Invitrogen). After the transblotting, the sample lines were detected by WesternBreeze chromogenic immunodetection system (Invitrogen) or ECL method (Amersham Biosciences, Buckinghamshire, UK) according to the manufacturer's instructions. The first antibodies were usually diluted 1:1000. His-tagged CA XIII was detected using anti-CA XIII serum or anti-HisG antibody (1:5000). In the control experiments, preimmune serum (1:1000) or a mixture of anti-human CA I, anti-human CA II, and anti-mouse CA II sera were used instead of the anti-CA XIII serum. Additional control experiments were also performed in the presence of the blocking peptide (1:1000 diluted antibody plus CA XIII peptide, 100 g in 10 ml).
Immunohistochemistry-Immunoperoxidase staining for human CA XIII was performed using the biotin-streptavidin complex method. In the mouse tissues, the localization of CA XIII was examined by both immunoperoxidase and immunofluorescence methods.
Human and mouse tissue samples fixed in Carnoy's fluid and embedded in paraffin were cut at 5 m sections and placed on microscope slides. The biotin-streptavidin complex method included the following steps: (a) pretreatment of the sections with undiluted cow colostral whey (Biotop, Oulu, Finland) for 30 min and rinsing in phosphatebuffered saline (PBS); (b) incubation for 1 h with the primary antiserum or preimmune serum diluted 1:100 in 1% bovine serum albumin-PBS solution (BSA-PBS); (c) incubation for 10 min with biotinylated second antibody (Histostain-plus, Zymed Laboratories Inc., South San Francisco, CA); (d) incubation for 10 min with peroxidase-conjugated streptavidin (Histostain-plus, Zymed Laboratories Inc.); (e) incubation for 3 min with 3,3Ј-diaminobenzidine tetrahydrochloride (DAB) solution (Liquid DAB substrate kit, Zymed Laboratories Inc.). The sections were washed three times for 5 min in PBS after incubation steps b and c and four times for 5 min after incubation step d. All of the incubations and washings were carried out at room temperature. Anti-human CA II serum was used as a positive control. Additional control staining was performed in the presence of the corresponding peptide (1:100 diluted anti-CA XIII serum plus 100 g of peptide/100 l of BSA-PBS). The sections were finally mounted in Neo-Mount (Merck, Darmstadt, Germany). The stained sections were examined and photographed with a Zeiss Axioskop 40 microscope (Carl Zeiss, Göttingen, Germany).
Mouse CA XIII was immunostained by the biotin-streptavidin complex method by employing the following steps: (a) pretreatment of the sections with undiluted cow colostral whey (Biotop) for 40 min and rinsing in PBS; (b) incubation for 1 h with the primary antiserum or preimmune serum diluted 1:100 in 1% BSA-PBS; (c) incubation for 1 h with biotinylated second antibody diluted 1:300 in 1% BSA-PBS (ZyMax goat anti-rabbit IgG, Zymed Laboratories Inc.); (d) incubation for 30 min with peroxidase-conjugated streptavidin diluted 1:750 in PBS (Zy-Max, Zymed Laboratories Inc.); (e) incubation for 1 min with DAB solution (9 mg of DAB in 15 ml of PBS plus 5 l of H 2 O 2 ). The sections were washed three times for 10 min in PBS after incubation steps b and c and four times for 5 min after incubation step d. All of the incubations and washings were carried out at room temperature. Preimmune serum and anti-mouse CA II serum were used for control purposes.
Mouse tissue sections were also stained by the immunofluorescence method and analyzed by confocal laser scanning microscopy. The steps in the immunofluorescence staining were as follows: PCR Methods-The expression of human and mouse CA XIII mRNAs were examined using cDNA kits purchased from BD Biosciences (Palo Alto, CA). The cDNAs included in MTC™ panels were used as templates for polymerase chain reaction (PCR) using CA XIII gene-specific primers. The human MTC™ panel II and mouse MTC™ panel I (BD Biosciences) contained first strand cDNA preparations produced from total poly(A) RNAs isolated from a number of different tissues.
5 ng of total cDNA was used as a template for PCR. All the reagents for PCR were from BD Biosciences. The PCR reactions were amplified on the thermal cycler (Gene Amp PCR system 9700, Applied Biosystems, Foster City, CA). The PCR amplification for human CA13 was performed using a three-step method. The PCR cycling parameters consisted of denaturation at 94°C for 2 min, followed by 33 cycles of denaturation at 94°C for 30 s, annealing at 50°C for 30 s and extension at 72°C for 1 min 30 s, followed by final extension at 72°C for 3 min. The PCR amplification for mouse CA13 with primer F1 and R1 was performed using a two-step method recommended by the manufacturer. The PCR cycling parameters consisted of denaturation at 94°C for 2 min, followed by 33 cycles of denaturation at 94°C for 30 s and extension and annealing at 68°C for 2 min, followed by final extension at 68°C for 3 min. The PCR amplifications for mouse spleen and 17-dayold embryo with primers F1 and R2 were also performed using a three-step method consisting cycling parameters of denaturation at 94°C for 2 min, followed by 33 cycles of denaturation at 94°C for 30 s, annealing at 55°C for 30 s and extension at 72°C for 1 min 30 s, followed by final extension at 72°C for 10 min. The PCR products were analyzed by electrophoresis on 1.2% agarose gel containing 0.1 g/ml ethidium bromide with DNA standard (100 bp DNA Ladder, New England Biolabs, Beverly, MA).
Sequencing the PCR Products-The PCR products from the human thymus, mouse spleen, and mouse 17-day-embryo were purified from agarose gel using GFX™ PCR DNA and Gel Band Purification kit (Amersham Biosciences). The sequencing was performed using ABI PRISM™ Big Dye Terminator Cycle Sequencing Ready Reactions Kit, version 2.0 (Applied Biosystems) following the protocol of the manufacturer. First, 5 l of DNA template was mixed with 4 l of Terminator Ready Reaction Mix (Applied Biosystems) and 1.6 pmol of the primer was added to the reaction mixture. Second, the reactions were amplified by cycle sequencing on the Gene Amp PCR system 9700 thermal cycler (Applied Biosystems). The cycling parameters consisted of 25 cycles of denaturation at 94°C for 10 s, annealing at 50°C for 5 s and extension at 60°C for 4 min. Third, after cycle sequencing, the extension products were purified by ethanol precipitation method recommended by the manufacturer. After purification the samples were resuspended to Template Suppression Reagent (Applied Biosystems) and denatured according to the manufacturer's protocol. The sequencing was finally performed on the ABI PRISM™ 3100 Genetic Analyser instrument (Applied Biosystems).
Expression and Purification of the His-tagged CA XIII-The fulllength PCR product from mouse 17-day-embryo was cloned into the E. coli TOP 10 bacterium using pTrcHis TOPO TA expression kit (Invitrogen) following the manufacturer's instructions. First, the PCR product was cloned into the TOPO vector and then the recombinant vector was transformed into E. coli. The transformed bacteria were incubated overnight at 37°C on LB (Luria Bertani) plates containing 50 g/ml ampicillin and 0.5% glucose. Several colonies were picked for analysis of positive clones. First, the colonies were cultured overnight in LB medium containing 50 g/ml ampicillin and 0.5% glucose at 37°C. Then the plasmids were isolated using QIAprep Spin Miniprep kit (Qiagen, Hilden, Germany) according to the manufacturer's protocol. The isolated plasmids were analyzed by restriction analysis and by sequencing. The restriction analysis was performed using two restriction enzymes, BamHI and HindIII (New England Biolabs), according to the manufacturer's protocol. The sequencing was performed using ABI PRISM™ Big Dye Terminator Cycle Sequencing Ready Reactions Kit, version 2.0 (Applied Biosystems) following the protocol of the manufacturer. The sequence primers, Xpress Forward primer and pTrcHis Reverse primer were from Invitrogen.
E. coli TOP 10 strain transformed with pTrcHis plasmid containing mouse CA XIII cDNA was grown at 37°C in 50 ml of LB medium containing 50 g/ml ampicillin and 0.5% glucose to an O.D of 0.6 at 600 nm. The inducer, isopropyl-1-thio-␤-D-galactopyranoside, was then added to a final concentration of 1 mM and then the culture was incubated at 37°C. The cells were harvested after 5 h by centrifugation, and the pellet was frozen at Ϫ80°C for later use. For large scale expression the bacteria were grown in 1 liter of LB medium.
The bacterial cell pellet was resuspended in 160 ml of lysing buffer (50 mM NaHPO 4 , 0.5 M NaCl, pH 8.0) and 160 mg lysozyme (Sigma) was added. The resuspended cells were incubated on ice 30 min and lysed by pipetting. Bacterial cell components were removed by centrifugation (13,000 rpm, 15 min, ϩ4°C). The supernatant was directly applied onto either InvitrogenЈs ProBond™ -purification system or conventional CA-inhibitor affinity chromatography purification.
The ProBond™ is a nickel-charged agarose resin that has been designed to purify recombinant proteins containing a polyhistidine sequence. The ProBond™ purification was performed under native conditions according to Invitrogen's protocol. The bound enzyme was eluted using 250 mM imidazole, pH 8.0. The inhibitor affinity chromatography was performed using the carboxymethyl (CM) Bio-Gel A (Bio-Rad Laboratories, Richmond, CA) coupled to p-aminomethylbenzenesulfonamide. The gel was washed using 10 mM HEPES buffer (pH 7.5) containing 150 mM NaCl and the bound enzyme was eluted using 0.1 M sodium acetate, pH 5.6, containing 0.5 M sodium perchlorate. The purified His-tagged CA XIII was visualized by two methods: Colloidal Coomassie Blue protein stain (Invitrogen) and Western blotting. Fig. 1 shows the amino acid sequence alignment of human CAs including the novel isozyme, CA XIII. All three histidine residues (His-94, -96, and -119) known to be critical for CA enzymatic activity are conserved in the predicted human and mouse (data not shown) CA XIII sequences. The phylogenetic analysis (Fig. 2) showed that CAs form three distinct clusters. CA I, II, III, V, and VII are clustered together as cytosolic/intracellular CAs. The second cluster includes extracellular and membrane-bound isozymes CA IV, VI, IX, XII, and XIV. The CA-RP VIII, X, and XI belong to the third cluster. The novel CA XIII is most closely related to CA I, II, and III with about 60% sequence identity, and thus, is clustered as cytosolic/intracellular isoform.

Phylogenesis and Structural Model of CA XIII-
In Fig. 3, the surfaces of three CAs (human CA I (1CZM), human CA II (1CNX), and human CA XIII) are colored according to the hydrophobic/hydrophilic properties of amino acids. The superimposition of these structures indicates that the location of the inhibitor binding pocket is conserved. The sulfonamide-type inhibitors are located inside the cavity and are bound to the zinc and the hydrophobic patch of the active site. The sequence similarity between the template (human CA I) and the target (human CA XIII) was 77%. A sequence identity above 50% generally guarantees that the modeled protein structure is correct at a level comparable to that of an x-ray crystal structure (16). The sequence alignment showed that the three Zn 2ϩ coordinating histidine residues are conserved in CA XIII molecule. Therefore, the Zn 2ϩ ion was directly merged into the structural model. The quality of the final model was checked using the three-dimensional profiles verification (17) and PROCHECK (18). The three-dimensional profiles verification method showed that the three-dimensional/one-dimensional scores of our model were always positive, which are within the range of scores for x-ray structure determinations. Additionally, PROCHECK confirmed that the model of human CA XIII had good geometric structural parameters. Table I shows the kinetic parameters for the CO 2 hydration reaction catalyzed by several ␣-CAs including recombinant mouse CA XIII. Inhibition data with acetazolamide, a clinically used CA inhibitor of the sulfonamide type, and cyanate, an anion inhibitor, are also presented.

Catalytic Activity and Inhibition by Sulfonamides and Anions-
CA13 Gene Expression-CA13 gene expression was investigated by PCR amplification of commercially available sets of cDNAs produced for selected human and mouse tissues. Human CA XIII mRNA was expressed in the thymus and small intestine, testis, spleen, prostate, ovary, and colon (Fig. 4). No CA XIII transcript was detected in the human leukocytes. In the mouse tissues, the positive signals were detected in the spleen, lung, kidney, heart, brain, skeletal muscle, testis, 7-day-old embryo, 11-day-old embryo, 15-day-old embryo, and 17-day-old embryo (Fig. 5). No CA XIII transcript was detected in the mouse liver.
Sequencing the PCR Amplification Products-Automated sequencing of the PCR products confirmed that correct CA13 sequences were amplified from the cDNAs of human thymus, mouse spleen, and 17-day-old mouse embryo. The primers for mouse CA13 were designed to obtain a full-length cDNA, while the corresponding human PCR product represented only a partial one. For sequencing the PCR product of human CA13 cDNA was amplified forward from the 5Ј-end, while mouse CA13 was amplified from both the 5Ј-and 3Ј-ends covering the whole coding sequence. The results clearly indicated that the PCR fragments represented CA13 cDNAs in both species.
Localization of CA XIII in Human and Mouse Tissues-Immunoreactivity of anti-CA XIII serum was confirmed by Western blotting. Single, 30-kDa polypeptides were detected in the human and mouse colon, while the signal in the human liver was barely apparent (Fig. 6). The control experiments using preimmune serum or peptide blocking were negative.
The immunolocalization studies for CA XIII revealed three major features on this novel enzyme: First, CA XIII showed an intracellular staining pattern, typical for cytosolic proteins. Second, there were distinct differences in the distribution of CA XIII between human and mouse tissues. Third, CA XIII was found in a number of different tissues in both species. Although the distribution of CA XIII resembles that of CA II in several tissues, we found some differences that make CA XIII unique with respect to its localization.
In the human alimentary tract, CA XIII was found in a number of tissues from the salivary glands to the large intestine. In the submandibular gland, CA XIII was expressed in serous acinar cells and duct epithelial cells (Fig. 7). Compared with CA II, CA XIII showed stronger signal in the duct epithelium. In the gastric mucosa, CA XIII showed only very weak staining in the surface epithelial cells of the body and antrum segments (Fig. 8). By contrast, CA II showed strong positive staining in the surface epithelial cells and parietal cells, which is in line with the fact that it is the major isozyme of the human gastric mucosa (2). Fig. 9 demonstrates the distribution of CA XIII in different segments of the human gut, and the immunostaining results of CA II are shown for comparison. CA XIII was found in enterocytes in every segment, although the variation in the staining intensity was notable between different parts of the gut. CA II showed stronger staining in the duodenum and large intestine, while CA XIII was somewhat more prominent in the jejunum and ileum. Immunostaining revealed no positive signal for CA XIII in the human liver (data not shown). Human pancreas also showed much weaker staining for CA XIII compared with CA II (data not shown). Only faint reactions were observed in the duct epithelium and acini.
CA XIII was highly expressed in the human testis, and in a pattern quite distinct from that of CA II (Fig. 10). Positive immunoreactions were observed in all stages of the developing sperm cells. In contrast, CA II showed strong signal only in the mature sperm cells limited to the most luminal face of the seminiferous tubules. CA XIII is also an abundant isozyme in the female reproductive tract. Based on histochemical demonstration of CA activity, it had been reported that the human endometrium expresses some CA isozyme (19). Recently, Karhumaa et al. (20) demonstrated that CA XII is present at the basolateral plasma membrane of the human endometrial epithelium. However, the presence of CA XII could not explain the cytosolic staining pattern that had been observed in early histochemical staining. Fig. 11 shows that CA XIII is, indeed, highly expressed in the uterine cervix and some endometrial glands were also positive. In contrast, CA II immunoreactivity was limited to the blood capillaries. The cervical and endome- The results show that CA XIII is most closely related to cytosolic CAs (isozymes I, II, and III). trial glands were negative for CA II.
Kidney was one of the human tissues we found positive for CA XIII. In both renal cortex and medulla, strongest immunoreactions were located to the collecting ducts (Fig. 12). Positive, albeit usually weak, immunoreactions were seen in the glomerulus. For comparison, CA II showed similar staining pattern in the collecting ducts in addition to its high expression in the nephron.
CA XIII expression was also investigated in mouse tissues using immunohistochemistry. Fig. 13 shows some representative examples of immunostaining for CA XIII and CA II. In the mouse brain (cerebrum and cerebellum) these isozymes were both expressed in the oligodendrocytes and positive signal was also associated with the nerve fiber bundles. The latter finding conceivably indicates their location in myelin, which is known to contain CA II (21). In the mouse kidney, positive staining for CA XIII was detected in the cortical and medullary collecting ducts. Comparison of immunostaining reactions for CA XIII and -II revealed that CA XIII is most probably expressed in the intercalated cells like CA II. From the various gastrointestinal tissues, the strongest immunoreaction for CA XIII was observed in the colon. The more proximal segments of the gastrointestinal canal showed none or only a faint signal. No CA XIII-specific staining was detected in the mouse testis, whereas CA II was present in the mature spermatozoa locating in the seminiferous tubules. Epithelial cells of the mouse uterus contained CA XIII and CA II, the latter of which has been previously located to those cells (22). Expression for CA XIII was   4. PCR analysis of human CA XIII mRNA expression. The strongest signals were obtained from the thymus and small intestine, followed by the testis, spleen, prostate, ovary, and colon. No signal was seen in the leukocytes.
FIG. 5. Mouse CA XIII mRNA expression analyzed by PCR amplification. The strongest signals were detected in the spleen, lung, kidney, 7-day-old embryo, and 17-day-old embryo, followed by the heart, brain, skeletal muscle, testis, 11-day-old embryo, and 15-day-old embryo. No signal was detected in the liver. All cDNAs were amplified using intrinsic primers (A) and spleen and 17-day-old embryo were selected to amplify a 1008-bp fragment representing full-length CA13 cDNA (B). also detected in the mouse lung, where the staining was most abundant in rounded cells of the alveolar wall, most probably representing the type II pneumocytes, which are known to contain CA II (23).
Cloning and Expression of the Mouse CA XIII in E. coli-The coding region of the mouse CA13 gene was amplified from cDNA and inserted into the bacterial expression vector, pTrcHis, for expression of recombinant CA XIII in E. coli strain TOP 10. CA XIII protein was expressed as a fusion protein containing a His tag at its N-terminal end. The successful cloning was confirmed by double digestion of the pTrcHis vector and automated sequencing.
The fusion protein was purified using two methods, The CA inhibitor affinity chromatography and ProBond™ purification system. The purified fusion protein was visualized by Colloidal Coomassie Blue staining and Western blotting (Fig. 14). Pro-Bond™ purification method was more efficient, but the purified FIG. 6. Western blotting of the human and mouse colon and human liver using anti-CA XIII serum. Negative control experiments were performed using preimmune serum or anti-CA XIII antibody in the presence of the blocking peptide. Positive controls were visualized using a mixture of anti-human CA I and II and anti-mouse CA II antibodies.   fraction contained small amounts of contaminant proteins in addition to a 35-kDa polypeptide representing His-tagged CA XIII. The CA inhibitor affinity chromatography method showed higher specificity in that the purified fraction showed only a single polypeptide of 35 kDa. DISCUSSION The present results indicated that CA XIII belongs to the cluster of cytosolic CA isozymes. Compared with human CA I and -II, the environment of the sulfonamide inhibitor binding pocket in human CA XIII is more hydrophilic due to the amino acid residue Arg-91 (which corresponds to Phe-91 in the struc-   In the cerebellum (A and C) and cerebrum (B), both isozymes are located in the oligodendrocytes (arrows) and nerve fiber bundles (arrowheads). In the kidney, positive staining for CA XIII is seen in the cortical (D) and medullary (E) collecting ducts. CA II shows a similar staining pattern (compare panels E and F). Testis shows no staining for CA XIII (G), whereas CA II is expressed in mature sperm cells locating in the middle of the seminiferous tubules (arrows in panel H). Epithelial cells of the endometrium express both CA XIII and II (I and J). Positive staining for CA XIII is also present in the colonic enterocytes (K) and lung (L). Original magnifications: immunofluorescence images in panels B, D, E, K, and L ϫ630; immunoperoxidase images ϫ200.

FIG. 14. Colloidal Coomassie Blue staining (upper panel) and
Western blotting (lower panel) analyses of purified His-tagged CA XIII protein produced in E. coli. 35-kDa polypeptide was isolated using CA inhibitor affinity chromatography and ProBond™ methods. In the Western blot, both anti-HisG and anti-CA XIII antibodies recognized the bacterially expressed His 6 -tagged CA XIII fusion protein.