- Top of page
Using real-time PCR and immunohistochemistry, we have examined the expression of carbonic anhydrase isozymes (CA) I, II, III, IV, IX, XII, XIII and XIV in the brain, kidney, stomach and colon of the wild-type, CA II-deficient (Car2−/−), and CA IX deficient (Car9−/−) mice. The expression of Car4, Car12, Car13 and Car14 mRNAs did not show any significant deviations between the three groups of mice, whereas both groups of CA deficient mice showed decreased expression levels of Car1 in the colon and Car3 in the kidney. The Car2 mRNA level was greatly reduced but not completely abolished in all four tissues from the Car2−/− mice in which no CA II protein was expressed. Sequencing the Car2 cDNA isolated from C57BL6 Car2−/− mice revealed two nucleotide differences from the wild-type C57BL6 mice. One is a silent polymorphism found in Car2 mRNA from wild-type DBA mice, which is the strain that provided the original mutagenized chromosome. The second change is a mutation that causes prematurely terminated translation at codon 155 (Gln155X). Car9 mRNA and CA IX protein expression levels were up-regulated about 2.5- and 3.6-fold, respectively, in the stomach of the Car2−/− mice. These results suggest that the loss of function of cytosolic CA II in the stomach of Car2−/− mice leads to up-regulation of an extracellular CA, namely CA IX, which is expressed on the cell surface of the gastric epithelium.
Carbonic anhydrases (CAs) are zinc-containing metalloenzymes that catalyse the reversible hydration of carbon dioxide in the reaction CO2+ H2O ⇌ HCO3−+ H+, and participate in various biological processes, including CO2 transport, regulation of pH homeostasis, bone resorption, ureagenesis, gluconeogenesis, production of body fluids, and fertilization (Sly & Hu, 1995; Parkkila, 2000; Lehtonen et al. 2004; Kivela et al. 2005a). The mammalian α-CA family is comprised of 16 different isoforms, among which 13 (CA I, II, III, IV, VA, VB, VI, VII, IX, XII, XIII, XIV and XV) are enzymatically active, whereas the other three (CA-RP VIII, X and XI) appear to lack CA activity because of substitutions in one or more of the functionally important histidine residues (Sly & Hu, 1995; Tashian et al. 2000; Lehtonen et al. 2004; Hilvo et al. 2005; Kivela et al. 2005a). In addition, the receptor-type protein-tyrosine phosphatases (RPTP) β and γ have also been reported to contain ‘CA-like’ domains. The 13 active CA isozymes differ in their subcellular localizations: CA I, II, III, VII and XIII are cytosolic, CA IV, IX, XII, XIV and XV are membrane associated, CA VA and VB are mitochondrial, and CA VI is secreted.
The expression of CAs in mammalian tissues has been intensively studied during the past three decades by means of reverse transcription polymerase chain reaction (RT-PCR), Northern blot, Western blot and immunohistochemical staining (Kumpulainen & Korhonen, 1982; Fleming et al. 1995; Ivanov et al. 1998; Saarnio et al. 1998; Parkkila, 2000; Ivanov et al. 2001; Wykoff et al. 2001; Leppilampi et al. 2003; Kivela et al. 2005b). These studies have provided important background information to understanding the roles of different isozymes in each type of tissue. The most distinctive features of CA expression are: (1) CA II appears to be the most widely expressed isozyme which is present in all major mammalian organs. As a high-activity isozyme, it may represent the most important isozyme for several fundamental biological processes (Sly & Hu, 1995; Parkkila & Parkkila, 1996); (2) two CA isozymes, CA IX and XII, are called cancer-associated isozymes due to their overexpression in certain carcinomas (Pastorekova et al. 2004; Pastorekova & Zavada, 2004); (3) multiple isozymes are often expressed in a particular mammalian tissue or organ, e.g. the murine kidney expresses at least CA II, IV, XII, XIII, XIV and XV (Brion et al. 1997; Kaunisto et al. 2002; Kyllonen et al. 2003; Lehtonen et al. 2004; Hilvo et al. 2005). The co-expression of several isozymes in different cell types and in different subcellular localizations in one organ suggests different functions for the different isozymes, which may or may not be redundant in a given tissue. Loss of one may lead to a compensatory up-regulation of another. Such compensatory mechanisms have never been studied thoroughly. In this study, we used quantitative real-time RT-PCR and immunohistochemical staining to assess the expression levels of CA isozymes I, II, III, IV, IX, XII, XIII and XIV in the brain, kidney, stomach and colon of the wild-type, CA II-deficient (Car2−/−), and CA IX-deficient (Car9−/−) mice.
- Top of page
A number of studies have examined the expression of different CA isozymes in mammalian tissues (Fleming et al. 1995; Kivela et al. 2005a; Purkerson & Schwartz, 2005). These investigations provided important background information for later functional studies using CA inhibitors and various cell culture and animal models. Previous studies made clear that several isozymes could be identified in the same mammalian tissue or organ. In many cases, the different isozymes were expressed in different cell types in these tissues and/or in different subcellular localizations. Examples of tissues which express multiple CA isozymes include the four tissues selected for the present investigation.
The purpose of this study was to determine whether loss of function of CA II or CA IX in the respective knockout mouse models leads to compensatory changes in other CAs. The mRNA expression of Car4, Car12, Car13 and Car14 did not show any significant changes in the brain, colon, stomach and kidney of either the Car2−/− or Car9−/− mice. The unchanged level of CA XII protein expression in these tissues as examined by immunohistochemical staining confirmed the lack of compensatory change in this isozyme (data not shown). In a previous study, CA IV protein levels were reported to be up-regulated in the central nervous system of the Car2−/− mice (Brion et al. 1994). However, no such alteration in the expression of Car4 mRNA in the brain of Car2−/− mice was detected in the present study. We were unable to document the protein levels of CA IV by immunohistochemistry since the antigenicity for CA IV was lost in these samples due to fixation.
The expression of Car1 mRNA in the colon of wild-type mice was extremely high in comparison with other isozymes. Car3 mRNA level was also high in the kidney of wild-type mice. Interestingly, both groups of CA-deficient mice showed a down-regulation of Car1 and Car3 mRNAs in the colon and kidney, respectively. The mechanism for this change is unclear, although it may simply reflect a disturbed pH regulation in these organs due to CA II or CA IX deficiency.
A somewhat surprising finding was the high basal expression of Car14 mRNA in the kidney relative to the mRNAs for the other CAs in the wild-type mice. The relative copy number of Car14 mRNA was approximately 5 or 10 times higher than those of Car2 or Car12, respectively. This finding suggests an important role of CA XIV in mouse kidney function. Recent studies have also indicated that CA XIV has even higher CA activity than murine CA II (Whittington et al. 2004). The same report indicated that CA XIV is nearly five times as active as murine CA IV. These results add further weight to the idea that CA XIV is probably a very important enzyme for renal physiology in the mouse.
Comparing the full-length Car2 cDNAs from the wild-type C57BL and Car2−/− mice revealed two nucleotide changes in the cDNA from the Car2−/− mice. One is the chemically induced mutation that causes an early stop codon in the Car2 transcript. This nonsense mutation at codon 155 (Gln155X) would account for the absence of CA II protein in the Car2−/− mice. It is also likely that the reduced level of mRNA in the Car2−/− mice is explained by the nonsense-mediated decay mechanism, which has been well characterized (Conti & Izaurralde, 2005). The original report of the Car2−/− mouse detected no difference in the expression of Car2 mRNA between the Car2−/− and wild-type mice (Lewis et al. 1988). However, the real-time RT-PCR method used in the present study is more sensitive to quantitative differences.
CA II has been considered the major isozyme in the stomach, where it participates in the production of gastric acid by proton secretion from the parietal cells, and on the other hand, protects the epithelium from acid digestion by bicarbonate secretion from the mucus-producing epithelial cells. To date, no gastrointestinal symptoms have been reported in CA II-deficient humans or mice (Ohlsson et al. 1986; Lewis et al. 1988). CA IX is another isozyme expressed in the gastric epithelium. The absence of functional CA IX enzyme has been linked to abnormal morphogenesis of the gastric mucosa (Ortova Gut et al. 2002). In the present study, the expression of Car9 mRNA and CA IX protein in the Car2−/− mice was found to be significantly increased compared with levels in the wild-type mice. This result is interesting in that it apparently reflects a compensatory up-regulation of a CA expressed on the cell surface of the gastric mucosal cells in response to the loss of the cytosolic isozyme CA II. Comparison of the gastric function in Car2−/− and Car9−/− knockout mice may provide clear insights into the relative contribution of each to gastric acidification and indicate whether the apparent compensatory increase in CA IX in CA II deficiency is effective.