Localization of ATP-sensitive K⁺ Channel Subunits in Rat Submandibular Gland

Abstract

ATP-sensitive K⁺ (K_ATP) channel subunits were investigated in rat submandibular gland (SMG). RT-PCR detected the presence of mRNA transcripts of the Kir6.1, Kir6.2, SUR2A, and SUR2B in the SMG, whereas SUR1 mRNA was barely detected. Western blot analysis provided the evidence that these four K_ATP channel subunits are expressed in rat SMG. Immunostaining detected that these four K_ATP channel subunits are widely distributed, with different intensities, in myoepithelial cells, epithelial cells of intercalated ducts, granular convoluted tubules, striated ducts, and excretory ducts. Immunofluorescence double staining showed that Kir6.1 and Kir6.2 colocalized with SUR2A in the myoepithelial cells, granular convoluted tubules, striated ducts, and excretory ducts. Kir6.1 and Kir6.2 also colocalized with SUR2B, mainly in the duct system, e.g., the granular convoluted tubules, striated ducts, and excretory ducts. Taken together, these results indicate that the K_ATP channels in SMG may consist of Kir6.1, Kir6.2, SUR2A, and SUR2B, with various combinations of colocalization with each other, and may play important roles in rat SMG during salivary secretion. (J Histochem Cytochem 58:499–507, 2010)

ATP-sensitive K⁺ (K_ATP) channels, originally discovered in heart (Noma 1983), are widely distributed in many tissues and cell types, including pancreatic β-cells (Ashcroft et al. 1984; Cook et al. 1988), neurons and brain (Levin and Dunn-Meynell 1998; Zhou et al. 2002; Acosta-Martinez and Levine 2007), skeletal muscle (Inagaki et al. 1996) and smooth muscle (Isomoto et al. 1996; Kubo et al. 1997), and kidney (Hunter et al. 1988; Zhou et al. 2008). K_ATP channels are inhibited by intracellular ATP and activated by MgADP (Noma 1983; Ashford et al. 1988). They couple the metabolic state of the cell to membrane potential by sensing changes in intracellular adenine nucleotide concentration (Aguilar-Bryan and Bryan 1999).

Functional K_ATP channels consist of two subunits in a hetero-octameric compound (Clement et al. 1997; Nestorowicz et al. 1997): the pore-forming subunit Kir6.x (Kir6.1 or Kir6.2), which belongs to the inward-rectifying K⁺ channel family that has two transmembrane domains (Inagaki et al. 1995a,b; Inoue et al. 1997), and the regulatory subunits that are sulfonylurea receptors (SURs), which belong to the ATP-binding cassette protein superfamily (Seino and Miki 2003). There is ∼71% amino acid identity between Kir6.1 and Kir6.2 (Inagaki et al. 1995a). There are two isoforms of the SURs, SUR1 and SUR2, which are derived from two different genes. Several variants of SUR2A were derived from alternative splicing (Chutkow et al. 1996,1999; Isomoto et al. 1996); the major variant is SUR2B. SUR2A and SUR2B differ only in their carboxyl-terminal 42 amino acids, due to alternative splicing (Isomoto et al. 1996).

K_ATP channels play important roles in insulin secretion and signal transmission, as well as neurotransmitter release (Amoroso et al. 1990; Inoue et al. 1997; Levin and Dunn-Meynell 1998; Mukai et al. 1998). We are interested in whether the submandibular gland (SMG) contains K_ATP channels, because SMG is an important salivary organ that maintains homeostasis of the upper gastrointestine and oral cavity by generating a barrier to microbial, chemical, and mechanical insults through salivary secretion (Nakamoto et al. 2008). SMG secretes not only digestive enzymes but also various growth factors and bioactive substances. In rodents, granular convoluted tubules, which develop between striated and intercalated ducts, are known to secrete heat-shock protein, fibroblast growth factor, insulin-like growth factor, and nerve growth factor (Yamamoto et al. 1992; Amano and Iseki 1993; Amano et al. 1993; Takahashi-Horiuchi et al. 2008).

To date, little is known about the cellular localization, the role, and the regulation of K_ATP channels in the SMG, although a great deal of data have shown that the genes Kir6.1 and SUR2B are ubiquitously expressed in different tissues and cells (Inagaki et al. 1995a; Isomoto et al. 1996). Thus, it is necessary to determine whether these K_ATP channel subunits are localized in the SMG.

With the questions noted above in mind, the present study attempted to clarify the expression and localization of the K_ATP channel subunits in rat SMG by RT-PCR, immunoblotting, and immunohistochemistry to gain insight into the functional roles of K_ATP channels in the SMG.

Materials and Methods

Animal and Tissue Preparation

Male Wistar rats (8–10 weeks) were used (Japan SLC; Hamamatsu, Japan). The protocols for animal experimentation described here were previously approved by the Animal Research Committee, Akita University. All subsequent animal experiments completely followed the University guidelines for animal experimentation.

Rats were deeply anesthetized by diethyl ether inhalation. Specimens for RT-PCR were quickly put into liquid nitrogen and kept at −80C until use. Specimens for immunoblot were cut into pieces, homogenized in a buffer as previously described (Zhou et al. 2005), and stored at −80C until use. Specimens for immunohistochemistry were fixed by transcardial perfusion with cold physiological saline, then further fixed with Zamboni fixative (100 ml 2% paraformaldehyde with 15 ml of saturated picric acid in 0.1 M PBS), pH 7.4. After perfusion, specimens were quickly dissected out, immersed in the same fixative for 6 hr at 4C, and subsequently transferred into 30% sucrose in PBS. Cryosections were cut at a thickness of 8 to 10 μm and thaw-mounted on MAS-coated glass slides (Matsunami Glass Industries; Kishiwada, Japan).

RT-PCR

RT-PCR was carried out as previously described (Zhou et al. 2008). Briefly, total RNA was extracted from SMG or skeletal muscle using the RNeasy Mini Kit (QIAGEN, GmbH; Hilden, Germany) according to the manufacturer's instructions. First-strand cDNA was synthesized by using the total RNA as a template with oligo(dT)_12–18 primer and superscript II reverse transcriptase (Invitrogen; Tokyo, Japan) according to the manufacturer's instructions.

PCR was performed as previously described (Zhou et al. 2008) with primers for rat Kir6.x and SURs (see below) as Kir6.1 (GenBank accession no. D42145), Kir6.2 (GenBank accession no. D86039), SUR1 (GenBank accession no. L40624), SUR2A (GenBank accession no. D83598), and SUR2B (GenBank accession no. AF019628), or the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (G3PDH, GenBank accession no. AF106860). PCR reactions (50 μl volume) contained 5 μl of 10 × buffer, 3 μl of 25 mM MgCl₂, 4 μl of dNTP mixture (2.5 mM each), and 3 U Taq DNA polymerase (Biotech International; Tokyo, Japan). The specimens were heated to 94C for 1 min, then subjected to 30 cycles of denaturation (94C, 30 min), annealing (30 min; 55C for Kir6.1, SUR2A, and G3PDH; 60C for SUR1; 58C for Kir6.2 and SUR2B), and extension (72C, 30 min). A final extension phase (72C, 3–5 min) was included for all samples. PCR products were separated on 1.5% agarose gels and visualized by ethidium bromide staining under ultraviolet light. To test for any residual contamination of the RNA samples by genomic DNA, control RT-PCR reactions were performed in a manner identical to that described above, but with the omission of the reverse transcriptase enzyme. All PCR products were extracted from the gel and purified with a gel extraction kit (Omega Bio-Tek; Doraville, GA), then labeled with BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems; Foster City, CA) and analyzed by sequencing test with an ABI Prism 3100-Avant Genetic Analyzer (Applied Biosystems; Hitachi, Tokyo, Japan) to confirm correct amplification.

K_ATP Channel Primers

Primers for all K_ATP channel subunits were designed as follows: primers for Kir6.1 were designed to reside 212 bp from 1323 to 1534, 5′AAGCGCAACTCTATGAGAAG-3′ (forward) and 5′-ACCAGAACTCAGCAAACTGT-3′ (reverse); primers for Kir6.2 were designed to reside 297 bp from 942 to 1238, 5′-CGCATGGTGACAGAGGAATG-3′ (forward) and 5′-GTGGAGAGGCACAACTTCGC-3′ (reverse). Primers for SUR1, SUR2A, and SUR2B were designed as previously described (Zhou et al. 2008). The primers 5′-CTCAAGATTGTCAGCAATGC-3′ (forward) and 5′-CAGGATGCCCTTTAGTGGGC-3′ (reverse) were designed to reside 393 bp from 507 to 899 for G3PDH as an internal control.

Western Blot Analysis

SDS-PAGE was carried out as previously described (Zhou et al. 2007) using 8% or 10% polyacrylamide gels. Proteins extracted from whole SMG were denatured in a modified sample buffer (125 mM Tris-HCl buffer, pH 6.8, 2% SDS, 25% glycerol, 0.01% bromophenol blue, and 10% 2-mercaptoethanol) and electrophoresed (10 μg per lane). After electrophoresis, proteins were electrophoretically transferred to a polyvinylidene difluoride membrane (Bio-Rad; Hercules, CA) by use of a semi-dry transfer unit (Hoefer TE70 series; Amersham Pharmacia Biotechnology, Buckinghamshire, UK). After blocking with 5% (w/v) BLOT-QickBlocker reagent (Chemicon International; Temecula, CA) in PBS-T (PBS containing 0.1% Tween-20), the membranes were incubated with goat anti-human Kir6.1 (Sc-11224; Santa Cruz Biotechnology, Inc., Santa Cruz, CA) (Zhou et al. 2005), or goat anti-human Kir6.2 (Sc-11228; Santa Cruz Biotechnology, Inc.) (Zhou et al. 2005), or rabbit anti-rat SUR2A (Zhou et al. 2007), and/or rabbit anti-rat SUR2B (Zhou et al. 2007), diluted 1:1000 for 60 min. After rinsing with PBS-T, the membranes were exposed to horseradish peroxidase (HRP)-conjugated donkey anti-goat IgG (AP180P; Chemicon International), or HRP-conjugated donkey anti-rabbit IgG (NA9340; Amersham Pharmacia Biotechnology), diluted 1:5000 for 30 min. The antigen–antibody complexes were visualized with chemiluminescence detection reagents (Amersham Pharmacia Biotechnology), according to the manufacturer's instructions.

Immunohistochemistry

Cryosections of SMG were kept in PBS containing 0.3% Tween-20 for 45 min. Prior to the incubation with first antibodies, sections were treated with a 0.3% solution of H₂O₂ in methanol and the ABC Blocking Kit (Vector Laboratories, Inc.; Burlingame, CA) to reduce the endogenous peroxidase reaction as well as nonspecific binding with avidin–biotin complex. After incubation with 5% normal goat serum or 5% normal rabbit serum for 1 hr, the sections were reacted with antibodies of goat anti-human Kir6.1, goat anti-human Kir6.2, rabbit anti-rat SUR2A, or rabbit anti-rat SUR2B, at a dilution of 1:200 to 1:500 for 12 hr at room temperature. After thorough rinsing with PBS containing 0.05% Tween-20, the sections were exposed to biotinylated rabbit anti-goat IgG (BA-5000; Vector Laboratories, Inc.) or biotinylated goat anti-rabbit IgG (BA-1000; Vector Laboratories, Inc.), diluted 1:200 for 30 min, and then with ABC complex (Vectastain ABC Kit; Vector Laboratories, Inc.) according to the manufacturer's instructions. Reaction sites were visualized by incubating the sections in 0.001–0.005% DAB (3,3′-diaminobenzidine tetrahydrochloride) reaction with 0.003% H₂O₂, and counterstaining with methyl green. Negative control was carried out either by omitting the first antibody or by adding corresponding immunizing peptide antigen against which the antibody was raised.

Immunofluorescence Double Staining

Double staining was performed as previously described (Zhou et al. 2005), although with some modification in relation to double labeling. After preincubation with 5% normal donkey serum in PBS for 1 hr, the sections for double labeling with Kir6.1 and SUR2A were incubated with goat anti-human Kir6.1 antibody (1:200) and rabbit anti-rat SUR2A antibody (1:500) diluted together in PBS. Those for Kir6.2 and SUR2A were incubated with goat anti-human Kir6.2 antibody (1:200) and rabbit anti-rat SUR2A antibody (1:500) diluted together in PBS. Those for double labeling with Kir6.1 and SUR2B were incubated with goat anti-human Kir6.1 antibody (1:200) and rabbit anti-rat SUR2B antibody (1:500) diluted together in PBS, and those for Kir6.2 and SUR2B were incubated with goat anti-human Kir6.2 antibody (1:200) and rabbit anti-rat SUR2B antibody (1:500) diluted together in PBS for 12 hr at room temperature. After rinsing with PBS, the sections were reacted with Alexa 488–conjugated donkey anti-goat IgG (A11055; Molecular Probes, Inc., Eugene, OR) and Alexa 594–conjugated donkey anti-rabbit IgG (A21207; Molecular Probes, Inc.), diluted together at 1:500 in PBS for 30 min. The sections were then coverslipped with PermaFluor aqueous mounting medium (Thermo; Pittsburgh, PA) after counterstaining with 4′,6-diamidino-2-phenylindole, dihydrochloride (DAPI). Fluorescence immunolabeling signals were detected by a laser-scanning microscope (LSM510; Carl Zeiss, Oberkochen, Germany).

Results

RT-PCR

During the course of the studies, reactions were replicated at least twice. The PCR products were sequenced, and positive control (skeletal muscle) was also used. RT-PCR of rat SMG RNAs using the primers for Kir6.1 generated a 212-bp product specific for Kir6.1; using the primers for Kir6.2 generated a 297-bp product specific for Kir6.2; using primers for SUR1 generated a 558-bp product specific for SUR1 in skeletal muscle, but no product for SMG; using the primers for SUR2A generated a 155-bp product specific for SUR2A; and using the primers for SUR2B generated a 152-bp product specific for SUR2B and a 328-bp product specific for SUR2A, respectively (Figure 1). It is suggested that rat SMG contains all K_ATP channel subunits except SUR1.

Figure 1.

RT-PCR of rat submandibular gland (SMG) and skeletal muscle (SKM) total RNA. The 1.5% agarose gels show bands of DNA fragments amplified with primer pairs specific for Kir6.1, Kir6.2, SUR1, SUR2A, and SUR2B, as well as glyceraldehyde-3-phosphate dehydrogenase (G3PDH). Primers were tested to amplify fragments of 212 bp for Kir6.1, 297 bp for Kir6.2, 558 bp for SUR1, 155 bp for SUR2A (primers for SUR2A only), 328 bp for SUR2A, 152 bp for SUR2B (primers for both SUR2A and SUR2B), and 393 bp for G3PDH. The result shows that all ATP-sensitive K⁺ channel subunits, but not SUR1, are expressed in the SMG, compared with the positive control of the SKM RNA.

Western Blot Analysis

In SMG extractions, the anti-Kir6.1 antibody recognized a prominent ∼50-kDa band, with a greater band of ∼100 kDa detected as a dimer (Figure 2, Lane Kir6.1), the anti-Kir6.2 antibody recognized a prominent ∼48-kDa band (Figure 2, Lane Kir6.2), the anti-SUR2A antibody recognized a prominent ∼140-kDa band (Figure 2, Lane SUR2A), and the anti-SUR2B antibody recognized a prominent ∼120-kDa band (Figure 2, Lane SUR2B), respectively.

Figure 2.

Immunoblot analysis of Kir6.1, Kir6.2, SUR2A, and SUR2B from rat SMG. Anti-Kir6.1 antibody recognized a prominent polypeptide of ∼50 kDa with a dimer of ∼100 kDa (Lane Kir6.1), and anti-Kir6.2 antibody recognized a ∼48-kDa polypeptides (Lane Kir6.2) in the SMG. Anti-SUR2A antibody recognized a remarkable polypeptide of ∼140 kDa (Lane SUR2A), and anti-SUR2B antibody revealed a polypeptide of ∼120 kDa (Lane SUR2B) in the SMG.

Immunohistochemistry

In the acinus, immunoreactivity with anti-Kir6.1, anti-Kir6.2, or anti-SUR2A and/or SUR2B antibody was not observed (Figures 3A–3D and 4A–4C). In the myoepithelial cells, faint immunoreactivity with those antibodies of K_ATP channel subunits was observed (Figures 3A, 3B, 4A, and 4B). In the intercalated ducts, weak immunoreactivity was detected with these antibodies of K_ATP channel subunits (Figures 3A, 3C, 4A, and 4C). In the granular convoluted tubules, moderate immunoreactivity with these antibodies was observed in the cytoplasm of these epithelial cells (Figures 3A, 3C, 4A, and 4C). In the striated ducts, moderate immunoreactivity with anti-Kir6.1 and/or with anti-Kir6.2 antibody was observed (Figures 3A and 3C); moderate to intense immunoreactivity with anti-SUR2A antibody (Figure 4A) and moderate immunoreactivity with anti-SUR2B antibody (Figure 4C) were detected in the cytoplasm of the parallel striation to vertical orientation of mitochondria in slender compartments formed by infolding of the basal cell membrane. In the excretory ducts, moderate immunoreactivity with Kir6.1 and/or Kir6.2 antibody (Figures 3B and 3D), and intense immunoreactivity with SUR2A and/or SUR2B antibody were observed (Figures 4B and 4D). In the smooth muscles of blood vessels, weak immunoreactivity with those antibodies of K_ATP channel subunits was also detected (Figures 3B, 3D, 4B, and 4D). Omission of the first antibody, or adding corresponding immunizing peptide antigen against which the antibody was raised, resulted in no immunoreactions (not shown). The relative intensities of distribution of these proteins of K_ATP channel subunits are summarized in Table 1.

Figure 3.

(A) Immunoreactivity with anti-Kir6.1 was observed to be weak in intercalated ducts (Ic) and moderate in the granular convoluted tubules (G) and striated ducts (St), and faint immunoreactivity was also seen in the myoepithelial cells (arrow). (B) Immunoreactivity with anti-Kir6.1 was observed to be moderate in the excretory ducts (Ex). Smooth muscle of blood vessel also showed weak immunoreactivity for Kir6.1 (Bv). (C) Immunoreactivity with anti-Kir6.2 was observed to be weak in the intercalated ducts (Ic) and moderate in granular convoluted tubules (G) and striated ducts (St). Myoepithelial cells showed faint immunoreactivity with anti-Kir6.2 (arrow). (D) Immunoreactivity with anti-Kir6.2 was observed to be moderate in excretory ducts (Ex). Blood vessels (Bv) also showed weak immunoreactivity of Kir6.2. Bar = 20 μm.

Figure 4.

(A) Immunoreactivity with anti-SUR2A was observed to be faint in myoepithelial cells (arrow), weak in intercalated ducts (Ic), moderate in granular convoluted tubules (G), and moderate to intense in striated ducts (St). (B) Immunoreactivity of SUR2A was observed to be intense in the excretory duct (Ex) and weak in the smooth muscles of blood vessels (Bv). (C) Immunoreactivity with anti-SUR2B was observed to be faint in myoepithelial cells (arrow), weak in the intercalated ducts (Ic), and moderate in granular convoluted tubules (G) and striated ducts (St). (D) Immunoreactivity of SUR2B was observed as moderate to intense in the striated (St) and excretory ducts (Ex) and weak in the blood vessels (Bv). Bar = 20 μm.

Table 1.

Localization of K_ATP channel subunits in rat SMG

	Ac	Myc	Ic	G	St	Ex	Sm
Kir6.1	−	±	+	++	++	++	+
Kir6.2	−	±	+	++	++	++	+
SUR2A	−	±	+	++	++–+++	+++	+
SUR2B	−	±	+	++	++	+++	+

−, negative; ±, faint; +, weak; ++, moderate; +++, intense. K_ATP, ATP-sensitive K⁺; SMG, submandibular gland; Ac, acinar cells; Myc, myoepithelial cells; Ic, intercalated ducts; G, granular convoluted tubules; St, striated ducts; Ex, excretory ducts; Sm, smooth muscle of blood vessels.

Immunofluorescence Double Staining

To determine whether the Kir6.1 or Kir6.2 is colocalized with SUR2A or SUR2B in the epithelial cells of the rat SMG, immunofluorescence double staining was performed. Immunoreactivity with Kir6.1 or Kir6.2 antibody was detected as green fluorescence (Alexa 488) (Figures 5A, 5D, 5G, and 5J), whereas immunoreactivity with SUR2A or SUR2B antibody was detected as red fluorescence (Alexa 594) (Figures 5B, 5E, 5H, and 5K). When the images were merged, yellow fluorescence was detected in the cytoplasm (Figures 5C, 5F, 5I, and 5L).

Figure 5.

Immunofluorescence double staining shows the expression of Kir6.1 and Kir6.2 in the SMG as green (Alexa 488, A,D,G,J), SUR2A and SUR2B as red (Alexa 594, B,E,H,K). The merged images (C,F,I,L) indicate that SUR2A colocalizes with Kir6.1 (C) and/or with Kir6.2 (F), and that SUR2B colocalizes with Kir6.1 (I), and/or with Kir6.2 (L) in the myoepithelial cells (thin arrows), granular convoluted tubules (G), striated ducts (St), and excretory ducts (Ex), as well as blood vessels (bold arrows). Bar = 50 μm.

Results showed that Kir6.1 overlapped with SUR2A in the granular convoluted tubules around the nucleus, at both the basal and apical portions. In the striated and excretory ducts, the Kir6.1 and SUR2A were widely colocalized. Colocalization with Kir6.1 and SUR2A was also observed in myoepithelial cells and small blood vessels (Figure 5C). Kir6.2 and SUR2A are widely distributed in myoepithelial cells, striated and excretory ducts, and blood vessels (Figures 5D and 5E). In the granular convoluted tubules, Kir6.2 was mainly localized in the apical portion and slightly localized in the basal portion (Figure 5D), whereas SUR2A was mainly localized in the basal portion rather than the apical portion (Figures 5B and 5E). Thus, colocalization with Kir6.2 and SUR2A was observed mainly in the basal portion and to some extent in the apical portion (Figure 5F). The colocalization with Kir6.2 and SUR2A was also observed in the striated and excretory ducts, in the myoepithelial cells at the base of the acinus, and in blood vessels (Figure 5F). Kir6.1 and SUR2B overlapped in both the basal and the apical portion in the granular convoluted tubules. In the striated and excretory ducts, as well as blood vessels, colocalization with Kir6.1 and SUR2B was also observed (Figure 5I). Kir6.2 overlapped with SUR2B mainly in the apical portion and to some extent in the basal portion in the granular convoluted tubules (Figure 5L). In the striated ducts and smooth muscle of the blood vessels, colocalization with Kir6.2 and SUR2B was also observed (Figure 5L). There was no clear colocalization of Kir6.1 and SUR2B or Kir6.2 and SUR2B in the myoepithelial cells observed.

Discussion

To our knowledge, the present observation provides the first evidence for the localization of K_ATP channel subunits Kir6.1, Kir6.2, SUR2A, and SUR2B in the rat SMG using biochemical and immunohistochemical methods.

ATP-activated Ca²⁺-dependent K⁺ channel (K_Ca) and ROMK-type K_ATP channel (Kir1.1) were detected in human submandibular gland ductal cells, but failed to show the SUR subunit gene using the RT-PCR method (Liu et al. 1999), whereas in mouse embryo, in situ hybridization histochemistry showed that SUR gene expression is apparent in the SMG (Hernandez-Sanchez et al. 1997). But the report did not specify which type of SUR was expressed in the mouse SMG. In the present study, we showed clearly that the regulatory subunits of K_ATP channels in rat SMG are SUR2A and SUR2B.

As in other exocrine secretion, the saliva secretion proceeds in two steps; the acinar cell produces primary isotonic, NaCl-rich saliva, which is then modified during its transportation through the ductal system, changes in electrolyte composition, and to some extent, the osmolarity of the primary fluid, by absorbing the NaCl and secreting KHCO₃− to the mouth as hypotonic final saliva (Zeng et al. 1997; Chaib et al. 1999). During passage through the excretory ducts of the salivary glands, the electrogenic component of duct transport consists of Na⁺ influx through the Na⁺ channel, Cl⁻ flux through the Cl⁻ channels, and K⁺ flux through the K⁺ channels (Chaturapanich et al. 1997). In general, the concentration of inorganic ions, such as Na⁺ and Cl⁻ in the saliva, changes from the acinar cavity to the duct cavity, because they are reabsorbed during their passage through the duct system. The K⁺ ion is secreted from the duct cells into the saliva when it passes through these ducts (Nakamoto et al. 2008).

Containing K_ATP channels in the duct system of the SMG may support their cellular functions. When the secretion speed increases, the reabsorption activity for Na⁺, Cl⁻, and HCO₃⁻ would also increase when the saliva passes through the ducts. Consequently, the ATP consumption increases and then the K_ATP channels open, protecting these cells by keeping membrane potential stable.

It is well known that there are different types of K_ATP channels constituting different combinations of Kir6.x and SURs in native tissues, e.g., Kir6.2 with SUR1 forms an insulin secretion β-cell–type K_ATP channel (Sakura et al. 1995), Kir6.2 with SUR2A forms a cardiac-type K_ATP channel (Inagaki et al. 1996), Kir6.2 with SUR2B forms a smooth-muscle–type K_ATP channel (Isomoto et al. 1996), and Kir6.1 with SUR2B forms a vascular smooth-muscle–type K_ATP channel (Yamada et al. 1997). However, recent research has supplied more evidence that K_ATP channels in different tissues and cells are not restricted, as they are usually described. For example, cardiomyocytes contain not only Kir6.2 and SUR2A but also Kir6.1, SUR1, and SUR2B (Schnitzler et al. 2000; van Bever et al. 2004; Zhou et al. 2005,2007). Furthermore, different combinations of Kir6.x and SUR consist of K_ATP channels with distinct electrophysiological properties and nucleotide and pharmacological sensitivities that reflect the various K_ATP channels in native tissues (Seino and Miki 2003). Even the same combination of K_ATP channels shows different physiological properties in different cell types. For example, coexpression of Kir6.2 and SUR1 in COS-1 cells reconstitutes a unitary conductance of ∼76 pS (Inagaki et al. 1995a), but coexpression of Kir6.2 and SUR2A reconstitutes an inward current with channel conductance of ∼80 pS (Inagaki et al. 1996). Glibenclamide, a type of K_ATP channel blocker, blocks the Kir6.2/SUR1 channel, and only slightly inhibits the Kir6.2/SUR2A channel (Inagaki et al. 1995a,1996; Gribble et al. 1998), whereas tolbutamide inhibits Kir6.2/SUR1 currents with high affinity, but does not inhibit Kir6.2/SUR2A (Gribble et al. 1998). The Kir6.2/SUR2A K_ATP channels coexpressed in Xenopus oocytes are more sensitive to both ATP and glibenclamide than those coexpressed in COS-1 cells (Inagaki et al. 1996; Gribble et al. 1998).

Thus, the present study revealed that four subunits of the K_ATP channel localized in the different epithelial cells of the rat SMG with different combinations, implying that variable K_ATP channels may have certain functions in those epithelial cells. In the granular convoluted tubules, the expression of those K_ATP channel subunits with different combinations in different portions, such as Kir6.1/SUR2A or Kir6.1/SUR2B in both the basal and the apical portion, Kir6.2/SUR2A mainly in the basal portion and Kir6.2/SUR2B mainly in the apical portion, indicated that the K_ATP channels have various types of combinations with different subunits in these tubules for their functions. They are also different from other K⁺ channels, such as the apical maxi-K channel, which localized in the striated and excretory ducts but not in the granular convoluted tubules (Nakamoto et al. 2008), giving a hint that the K_ATP channels may have some relationship to secretion function in granular convoluted tubules, which contain many bioactive substances in their granules (Barka 1980; Amano et al. 1991; Amano and Iseki 1993; Gresik 1994).

In conclusion, K_ATP channel subunit Kir6.1, Kir6.2, SUR2A, and SUR2B localization in the epithelial cells of the rat SMG indicated that variable combinations of these subunits were formed in these cells, and that they may play important roles in those cells during salivation and regulation of ion concentration, as well as in the bioactive substance secretion of saliva through the ductal system to the mouth cavity.