Identification of KRAP-expressing cells and the functional relevance of KRAP to the subcellular localization of IP3R in the stomach and kidney

KRAS-induced actin-interacting protein (KRAP), originally identified as one of the deregulated genes expressed in colorectal cancer, participates under physiological conditions in the regulation of systemic energy homeostasis and of the exocrine system. We have recently found that KRAP is a molecule associated with inositol 1,4,5-trisphosphate receptor (IP3R) and is critical for the proper subcellular localization of IP3R in the liver and the pancreas. However, the expression of KRAP and its precise function in other tissues remain elusive. In this study, we aimed to identify the KRAP-expressing cells in mouse stomach and kidneys and to examine the relevance of KRAP expression in the regulation of IP3R localization in these tissues. In the stomach, double immunohistochemical staining for KRAP and IP3R demonstrated that KRAP was expressed along with the apical regions in the mucous cells and the chief cells, and IP3R3 was dominantly co-localized with KRAP in these cells. Furthermore, IP3R2 was also co-localized with IP3R3 in the chief cells. It is of note that the proper localization of IP3R3 and IP3R2 in the chief cells and of IP3R3 in the mucous cells were significantly abrogated in KRAP-deficient mice. In the kidneys, KRAP was expressed in both the apical and the basal regions of the proximal tubular cells. Intriguingly, KRAP deficiency abrogated the localization of IP3R1 in the proximal tubular cells. Finally, co-immunoprecipitation study in the stomachs and the kidneys validated the physical association of KRAP with IP3Rs. These findings demonstrate that KRAP physically associates with IP3Rs and regulates the proper localization of IP3Rs in the mucous cells and the chief cells of the stomach and in the proximal tubular cells of the kidneys.

KRAS-induced actin-interacting protein (KRAP) was originally identified as one of the deregulated expression gene in the colorectal cancer cell line, HCT116 (16). The previous studies using KRAP-knockout (KRAP-KO) mice demonstrate that KRAP participates in the regulation of systemic energy homeostasis (17) and of exocrine system (18). Among the adult mouse tissues, KRAP is ubiquitously expressed, with high levels in the pancreas, liver, and brown adipose tissues, and KRAP localizes in the restricted apical regions of the liver parenchymal cells and of the pancreatic exocrine acinar cells (19). Our recent findings indicate that KRAP associates with IP 3 R to regulate its proper subcellular localization in the mouse liver and the pancreas (20) as well as in immortalized cultured cell lines (21). Despite these advances, it remains largely unknown which cell types express KRAP among the other tissues including stomach and kidneys.
Herein, we performed immunohistological analysis and identified the exact KRAP-expressing cells in the stomach and the kidneys, and demonstrated that KRAP plays critical role in the regulation of the precise subcellular localization of IP 3 R in the mucous and the chief cells of the stomach and in the proximal tubular cells of the kidneys.

Materials and methods
Animals. All animals used in this study were treated in accordance with the guidelines of Fukuoka University. KRAPknockout mice were generated as described previously (17).

Results
Localization of KRAP protein in the adult mouse stomach. To examine the cellular distribution of KRAP protein in the adult mouse tissues, we performed immunohistochemical staining by using anti-KRAP antibody. In the stomach, strong KRAP immunoreactivity was restricted to the pit regions of gastric glands (Fig. 1A), whereas significant expression of KRAP was not detected in the muscularis mucosae beneath the gastric glands (Fig. 1A, arrows). The specificity of KRAP expression in the stomach was confirmed by using KRAP-KO tissue as a control (Fig. 1B). In the pit region of the gastric gland, where columnar surface mucous cells mainly exist (22), KRAP was localized beneath the apical membranes of the mucous cells (Fig. 1C). In the base region of the gastric glands, where zymogenic chief cells mainly exist, coronal plane of deeper gastric glands showed that KRAP was restricted to the apical regions of the chief cells (Fig. 1D, arrowheads), whereas KRAP was not detected in the parietal cells (Fig. 1D, asterisks). The distinction between the chief and the parietal cells was validated by ZO-1 staining as described (23), indicating that KRAP was expressed in the ZO-1-positive chief cells but not in the ZO-1-negative parietal cells (Fig. 1E).
KRAP co-localized with IP 3 R in the stomach. Since we previously reported that KRAP associates with particular subtypes of IP 3 R in the liver and the pancreas (20), we exam-in the liver and the pancreas (20), we examined whether KRAP in the stomach is also co-localized with IP 3 R. Double-immunostaining of the stomach for KRAP and IP 3 R3 revealed that KRAP was co-localized with IP 3 R3 in the apical regions of both the chief cells ( Fig. 2A, arrows) and the mucous cells (Fig. 2B, arrows). Of note, IP 3 R2 co-existed with IP 3 R3 in the chief cells (Fig. 2C, arrow) but not in the parietal cells (Fig. 2C, asterisks). Furthermore, IP 3 R2 was not detected in the mucous cells (Fig. 2D, arrows). These results indicated that KRAP was co-localized with IP 3 R2 and IP 3 R3 in the chief cells and with IP 3 R3 in the mucous cells.

Impaired localization of IP 3 R in the KRAP-deficient chief cells and the mucous cells.
We addressed the functional relevance of KRAP to the proper localization of IP 3 R by using KRAP-KO mice. IP 3 R3 was located in the apical region of the chief cells (Fig. 3A, arrow) and of the mucous cells (Fig. 3C, arrows) in the wild-type (WT) mouse stomach, whereas the restricted localization of IP 3 R3 appeared to be diminished in the KRAP-KO stomach (Fig. 3B, arrow; 3D, arrows). Furthermore, IP 3 R2 was detected in both the chief cells (Fig. 3E, arrows) and the parietal cells (Fig. 3E, asterisks) in the WT stomach, whereas the localization of IP 3 R2 in the KRAP-KO stomach was impaired in the chief cells (Fig. 3F, arrows) but not in the parietal cells (Fig. 3F, asterisks). Thus, KRAP plays critical role in the regulation of the proper localization of IP 3 R2 and IP 3 R3 in the chief cells and of IP 3 R3 in the mucous cells.
KRAP expression and its contribution to the localization of IP 3 R1 in the proximal tubules of the mouse kidney. To examine the cellular distribution of KRAP protein in the adult mouse kidneys, we performed immunohistochemical staining by using anti-KRAP antibody. The specificities of the signals were validated by comparing the immunoreactivities of WT and KRAP-KO mouse tissues. In the WT kidneys, intense immunoreactivities were observed in the renal proximal tubules (Fig. 4A) but not in the renal distal tubules (data not shown). On the other hand, significant immunoreactive signal was not detected in the proximal tubules in the KRAP-KO mice (Fig. 4B). Taken together, these results indicate that KRAP was exactly expressed in the proximal tubules. The proximal tubules were identified by the presence of the brush-border stained with phalloidin ( Fig. 4A and B). Immunostaining in the proximal region showed that KRAP was accumulated beneath the brush-border (Fig. 4A, arrows) and KRAP was also detected in the basolateral actin bundles (Fig. 4A, arrowheads). We next examined which subtypes of IP 3 R, IP 3 R1, IP 3 R2, and IP 3 R3, expressed in the proximal tubular cells, revealing that IP 3 R1 (Fig. 4C) but not IP 3 R 2 or IP 3 R3 (data not shown) was detected in the beneath the brush-border and in the basolateral actin bundles. Finally, we addressed the functional relevance of KRAP expression in the proximal tubular cells to the regulation of IP 3 R localization. It is of note that the restricted localization of IP 3 R1 detected in the WT mouse kidney (Fig. 4C) was disturbed in the KRAP-KO mouse kidney (Fig. 4D). Thus, KRAP plays critical role in the regulation of the proper localization of IP 3 R1 in the proximal tubular cells.

KRAP interacts with IP 3 R1 in the kidneys and with IP 3 R3
in the stomach. As described above, immunohistochemical signals for particular IP 3 R subtypes in the KRAP-KO mouse kidneys or the stomach were abrogated, leading us to check the expression levels of IP 3 R between the WT and KRAP-KO mouse tissues. Normal expression levels of IP 3 R1 and IP 3 R3 were detected in the KRAP-KO mouse kidney and the stomach, respectively, compared with the WT mouse tissues (Fig. 5A), suggesting that mislocalizations but not deregulated expressions of IP 3 R occur in the KRAP-KO mouse kidneys and the stomach. Next, to examine the physical association of KRAP with IP 3 R, we performed co-immunoprecipitations by anti-KRAP antibody in the kidneys or the stomach, in which we could not evaluate the specific association of IP 3 R2 with KRAP due to lack of IP 3 R2-specific antibody available for western blotting. In the preparations from the WT mouse tissues, KRAP precipitates IP 3 R1 and IP 3 R3 in the kidney and the stomach, respectively (Fig. 5B). The specificity of co-immunoprecipitations of IP 3 R was confirmed by using KRAP-KO mouse tissue as a control (Fig. 5B). Thus, KRAP physically interacts with IP 3 R1 in the kidneys and with IP 3 R3 in the stomach.

Discussion
In this study, we demonstrated that KRAP protein expression and the subcellular localization was restricted beneath the apical and/or basolateral membranes in specific cell types of the stomach and the kidneys, in which KRAP physically associated with particular IP 3 R subtype(s). In the KRAP-KO mouse stomach and the kidneys, the polarized localization of IP 3 R was impaired, indicating that KRAP plays critical roles in the regulation of the proper subcellular localization of IP 3 R in the stomach and the kidneys.
Notably, KRAP as well as IP 3 R3 proteins were polarized beneath the apical membranes facing the gastric gland lumen and were absent in the parietal cells (Fig. 1), suggesting an association of these proteins with chief cell functions including pepsinogen secretion (22)(23)(24). From this view point, KRAP expression and the localization beneath the apical membranes of the pancreatic acinar cells (19), another type of zymogen cells, may suggest a similar role for KRAP in the stomach and the pancreas. Considering the fact that KRAP physically interacts with IP 3 R to regulate its proper localization in these tissues, stomach (Figs. 2, 3 and 5) and pancreas (20), and that double-knockout of IP 3 R2 and IP 3 R3 in mice revealed a failure in secretory function in the pancreas (25), KRAP seems to be involved in the exocrine systems. Actually, the pancreatic acinar cells in KRAP-KO mice showed an increased amount of zymogen granules, although they seemed to maintain the proper physiological agonist-induced exocytosis (18). Thus, exact functional relevance of KRAP and its interaction with IP 3 R to the exocrine systems in the pancreas and the stomach should await future studies.
It is of note that KRAP was restricted to both the apical region and the basolateral region of the proximal tubular cells of the kidneys (Fig. 4), and that KRAP physically associated with IP 3 R1 in the kidneys (Fig. 5). Furthermore, our previous study showed that KRAP was distributed along the bile canaliculi of hepatocytes and underneath the apical membrane of pancreatic acinar cells (19). All these KRAP localizations in the distinct tissues examined are restricted to epithelial cell types bearing well-developed cell polarity, cell-cell junction and microvilli, where transports of various substances between epithelial cells and extracellular spaces, exocrine space or blood stream occur (22,(26)(27)(28). Since KRAP-KO mice displayed profound metabolic disorders after birth without developmental defects, and certain systemic inter-tissue dysregulations appeared to underlie the metabolic phenotypes (17), KRAP might play physiological roles in secretion and/or absorption functions after birth rather than in developmental events.
Renal proximal tubules serve the reabsorption of the bulk of substances filtered in the glomeruli and the excretion (26,29). These two opposite transports are accomplished by the coordinated action of ion channels and transporters located in the brush border membrane and basolateral membrane (29)(30)(31). Thus, the polarized expression of these membrane proteins is crucial for the function of the proximal tubules. Based on the findings that KRAP protein possesses characteristic features like scaffolding protein, such as polarized localization and transporting of IP 3 R, potential functional relevance of KRAP to these processes would be suspected.
In conclusion, we identified the exact KRAP-expressing cells in the stomach and the kidneys, and found that KRAP physically associates with IP 3 R to regulate its proper sub cellular localization in vivo. Considering the KRAP function as an IP 3 R regulator and the importance of KRAP in energy homeostasis in vivo, further research on the exact relevance of the association between KRAP and IP 3 R to the biological phenomena will lead to a better understanding of physiological metabolic processes. Anti-KRAP (αKRAP) immunoprecipitations were performed using mouse kidneys and stomachs from WT or KO mice, followed by western blotting with anti-KRAP, anti-IP 3 R1, or anti-IP 3 R3 antibodies. total, total lysate;IP, immunoprecipitation; α, anti-.