Modified aquaporin 5 expression and distribution in submandibular glands from NOD mice displaying autoimmune exocrinopathy.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 56, No. 8, August 2007, pp 2566–2574 DOI 10.1002/art.22826 © 2007, American College of Rheumatology Modified Aquaporin 5 Expression and Distribution in Submandibular Glands From NOD Mice Displaying Autoimmune Exocrinopathy Muhammad S. Soyfoo,1 Carine De Vriese,1 Huguette Debaix,2 Maria D. Martin-Martinez,3 Chantal Mathieu,4 Olivier Devuyst,2 Serge D. Steinfeld,5 and Christine Delporte1 mice and the 8-week-old NOD mice showed AQP5 primarily at the apical membrane of the salivary gland acinus. In contrast, in acini from the submandibular glands (but not the parotid glands) from 24week-old NOD mice, AQP5 staining was reduced at the apical membrane but was increased at the basal membrane. A moderately statistically significant decrease in pilocarpine-stimulated salivary flow was observed in 24-week-old NOD mice compared with that in agematched BALB/c mice. Conclusion. Submandibular glands from 24week-old NOD mice displayed inflammatory infiltrates, increased AQP5 protein expression, and impaired AQP5 distribution. However, the moderately statistically significant decrease in the salivary flow rate in these mice did not match the extent of AQP5 misdistribution. Objective. To investigate the expression and localization of aquaporin 5 (AQP5) in salivary glands and salivary gland function in the NOD mouse. Methods. All experiments were performed using NOD and BALB/c mice (ages 8 weeks and 24 weeks). Real-time reverse transcription–polymerase chain reaction, Western blotting, and immunohistochemical analysis were used to study the expression and distribution of AQP5 in salivary glands. In addition, salivary gland function was determined. Results. Compared with the levels in BALB/c mice, relative AQP5 messenger RNA levels were not significantly modified in the parotid glands from NOD mice of both ages but were significantly increased in the submandibular glands from NOD mice of both ages. Western blot analyses of both salivary gland membranes revealed that the level of AQP5 protein was increased in 24-week-old NOD mice. Important inflammatory infiltrates were observed in the submandibular glands, but not in the parotid glands, from 24-week-old NOD mice. The 8-week-old and 24-week-old BALB/c Sjögren’s syndrome (SS) is an autoimmune disorder characterized by lymphocyte infiltration of salivary and lacrimal glands, leading to glandular hypofunction that often results in characteristic symptoms of dry mouth and dry eyes (1). Typical salivary gland biopsy specimens obtained from patients with SS show lymphocytic infiltrates, with an associated reduction (up to 50%) in the number of acinar cells (2). The roles of reduced numbers of acinar cells and associated glandular dysfunction in the pathogenesis of SS are well characterized (1). Several factors have been proposed to contribute to impaired salivary flow, including the proinflammatory cytokines interleukin-1 and tumor necrosis factor ␣ (which were shown to inhibit both basal and stimulated secretion ), decreased protein kinase C levels in salivary gland acinar cells from patients with SS (3), as well as the presence of antibodies against type 3 muscarinic receptor in patients with primary SS (4). Dr. Soyfoo is recipient of an Erasme Foundation Research Fellowship. Dr. Devuyst’s work was supported by a grant from the Fund for National Scientific Research, Belgium and an Arthritis Research Campaign grant (05/10-328). Dr. Delporte’s work was supported by the Fund for Medical Scientific Research, Belgium (grant 3.4604.05). 1 Muhammad S. Soyfoo, MD, Carine De Vriese, PhD, Christine Delporte, PhD: Université Libre de Bruxelles, Brussels, Belgium; 2 Huguette Debaix, MSc, Olivier Devuyst, PhD: Université Catholique de Louvain Medical School, Brussels, Belgium; 3Maria D. MartinMartinez, MD: Bordet Institute, Brussels, Belgium; 4Chantal Mathieu, PhD: Katholieke Universiteit Leuven, Leuven, Belgium; 5Serge D. Steinfeld, PhD: Erasme University Hospital, Brussels, Belgium. Address correspondence and reprint requests to Christine Delporte, PhD, Laboratory of Biochemistry, CP 611, Université Libre de Bruxelles, Route de Lennik 808, B-1070 Brussels, Belgium. E-mail: firstname.lastname@example.org. Submitted for publication November 7, 2006; accepted in revised form May 11, 2007. 2566 EXPRESSION AND DISTRIBUTION OF AQP5 IN NOD MICE The discovery of a family of water-specific membrane channel proteins, the aquaporins (AQPs), has provided insight into the molecular mechanism of membrane water permeability in a variety of tissue (for review, see refs. 5–8). Among the AQPs identified thus far, AQP5, AQP1, and AQP8 have been localized to normal salivary glands, using immunohistochemical methods (9). In knockout mice, AQP5, which is present in the apical membrane of acinar cells (5,10–12), was shown to participate in saliva secretion (13). In contrast, AQP1, which is present in capillaries and venules adjacent to salivary gland epithelium (14–16), and AQP8, which is located in myoepithelial cells surrounding the acini, were not demonstrated, using knockout mice, to play a physiologic role in saliva secretion (17). In minor salivary glands from patients with SS, a predominance of AQP5 staining intensity was observed at the basal membrane of acinar cells (12). This abnormal distribution of AQP5 suggested a defect in AQP5 distribution that likely contributes to a deficiency in fluid secretion, which is a defining feature of SS. Although a conflicting report had described normal localization of AQP5 in 5 patients with SS (18), similar abnormal localization of AQP5 was reported in lacrimal glands from patients with SS (19). The NOD mouse model is a well-described animal model for the study of the autoimmune exocrinopathy that is prevalent in patients with SS (20). The NOD mouse model allows investigators to study progression of the disease, which is very difficult to achieve in patients with SS. In a relatively short period of time, NOD mice pass from a normal healthy state (at age 8 weeks) to experiencing a severe form of the disease (at age 24 weeks), characterized by histopathologic modifications of their salivary glands (inflammatory lymphocytic infiltrates, acinar cells destruction) and impaired saliva secretion, which also occur in patients with SS (21,22). In NOD mice, hyposalivation was shown to be preceded by inflammatory changes in the salivary glands (23). Recently, confocal laser scanning imaging showed abnormal distribution of AQP5 in parotid and submandibular glands from 20–26-week-old NOD mice compared with that in glands from 9-week-old BALB/c mice. However, no data concerning either AQP5 expression or salivary gland function were provided (24). To further address the possible involvement of AQP5 in exocrine salivary gland dysfunction, we studied the molecular expression and distribution of AQP5 in salivary glands, linked to measurement of salivary gland function, using NOD mice of 2 ages (8 weeks old and 24 2567 weeks old; these ages should represent 2 distinct stages of the disease) and matched control BALB/c mice. MATERIALS AND METHODS Animals. NOD mice, originally obtained from Professor Wu (Beijing, China), were housed and inbred in the animal facility of the Katholieke Universiteit Leuven since 1989. NOD mice were housed under semi-barrier conditions. All experiments were performed using overnight-starved, 8-week-old or 24-week old female NOD mice or BALB/c mice (purchased from Harlan, Horst, The Netherlands). Because autoimmune diabetes also develops spontaneously in NOD mice, mice were checked for glycosuria, and only nondiabetic mice were included in the study. Maintenance of the animals and all experimental procedures were performed in accordance with the guidelines of the ethics committees of the Katholieke Universiteit Leuven (Leuven, Belgium), which are in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Semiquantitative real-time reverse transcription– polymerase chain reaction (RT-PCR). Salivary gland total RNA was extracted with a NucleoSpin RNA II kit (MachereyNagel, Düren, Germany), treated with DNase, and purified using the PureLink Micro-to-Midi Total RNA Purification System (Invitrogen, Merelbeke, Belgium). Prior to reverse transcription using the SuperScript III kit (Invitrogen), the quality of the RNA was checked using the Agilent 2100 BioAnalyzer (Agilent, Palo Alto, CA). The primers used for amplification of mouse AQP5 complementary DNA (cDNA) were as follows: sense primer 5⬘-TGGAGCAGGCATCCTGTACT-3⬘ and antisense primer 5⬘-CGTGGAGGAGAAGATGCAGA-3⬘ (151-bp amplicon); GAPDH sense primer 5⬘TGCACCACCAACTGCTTAGC-3⬘ and antisense primer 5⬘-GGATGCAGGGATGATGTTCT-3⬘ (176-bp amplicon). Changes in target gene messenger RNA (mRNA) levels were determined by semiquantitative real-time RT-PCR with an iCycler iQ System (Bio-Rad, Hercules, CA), using SYBR Green I for detection, as described previously (25). All reactions were performed in duplicate. The PCR conditions were as follows: 94°C for 3 minutes, followed by 31 cycles of 30 seconds at 95°C, 30 seconds at 61°C, and 1 minute at 72°C. To exclude amplification by contaminating genomic DNA, samples of RNA that had not been reverse transcribed were run in a parallel PCR. Standard curves were prepared using serial 4-fold dilutions of cDNA samples for each assay. The efficiency of the reactions was calculated from the slope of the standard curve (efficiency ⫽ [10⫺1/slope]⫺1). The relative mRNA expression of AQP5 was investigated in salivary glands from BALB/c and NOD mice, ages 8 weeks or 24 weeks, after normalization to GAPDH (ratio ⫽ 2⌬Ct [reporter ⫺ target gene]). The relative (mean ⫾ SEM) changes in mRNA levels of AQP5 in salivary glands from NOD mice compared with BALB/c mice of the same age, after adjustment to GAPDH (ratio ⫽ [efficiency target gene]⌬Ct [BALB/c – NOD]/ [efficiency GAPDH]⌬Ct [BALB/c – NOD]), were 0.98 ⫾ 0.01 for AQP5 and 1.04 ⫾ 0.03 for GAPDH. Antibodies to AQP5. An antibody resulting from rabbit immunization using a synthetic peptide corresponding to the C-terminal 25 amino acids of rat AQP5 (this sequence was 2568 100% identical to the corresponding mouse AQP5 sequence) was affinity purified, and its specificity was verified by Western blotting and confocal microscopy (26). Western blot analysis. Crude plasma membranes were prepared from mouse salivary glands, as described previously (27), except that a membrane pellet from 17,000g centrifugation was used, and no ␤-mercaptoethanol was added to the sample buffer for sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE). The samples were analyzed by SDS-PAGE using 12% polyacrylamide gels. Proteins were transferred to polyvinylidene difluoride membranes and immunolabeled using the affinity-purified antibody to AQP5 at a 1:2,000 dilution. The bound antibodies were detected by using the enhanced chemiluminescence method (Amersham, Piscataway, NJ). Band densities (a measure of “volume”) were estimated using image analysis apparatus and Bio-Profil Bio-1D software from Vilbert Lourmat (Marne-la-Vallée, France). Densitometry in image analysis is based on digitalization of the image in pixels, the intensity of which is coded on a scale of 256 levels of grey. The measure of volume is the sum of all pixel intensities comprising the spot and depends on the number of pixels inside the area of the spot and on the intensity of these pixels. Immunohistochemical localization of AQP5. Immediately after being removed, salivary glands were fixed in 4% buffered formaldehyde, paraffin embedded, and sectioned (5 m thick). Some salivary gland sections were stained with hematoxylin and eosin to visualize the infiltrate. Some sections were incubated with the primary antibody, affinity-purified anti-AQP5 (dilution 1:200), overnight at 4°C. The bound anti-AQP5 antibody was revealed by staining with anti-rabbit gamma globulin (Nordic, Tilburg, The Netherlands) followed by peroxidase–antiperoxidase complex (Dako, Glostrup, Denmark). The fixed peroxidase was visualized using diaminobenzidine (Dako). As control, salivary gland sections were incubated either with secondary antibody alone or with anti-AQP5 antibody preadsorbed with the immunizing peptide in 100-fold excess. The labeling index corresponds to the percentage of the acinus area that is immunoreactive for rat AQP5 in each region of the cell. Quantitative analysis of the labeling index of the immunoreactive acini was performed as previously described (28,29), using a SAMBA 2005 computer-assisted microscope system (Unilog, Grenoble, France) equipped with a 20⫻ magnification lens (aperture 0.50 mm). On each pixel, the tissue-integrated optical density due to the hematoxylin counterstaining and the brown diaminobenzidine chromogen was computed on 256 densitometric levels in 2-color channels. Immunoreactivity generated by binding of the histochemical probe was defined by the integrated optical density values in excess of the mean ⫹2 SD of the corresponding negative control. The computer-assisted microscope and related quantitative analyses were standardized as follows. A negative histologic control slide (from which the primary antibody was omitted) was analyzed for each experimental condition. The software used on the computer-assisted microscope automatically subtracted the labeling index and mean optical density values of the negative control sample from the values of each of the 2 positive samples available for each experimental condition. Specific software was included in the computer, to assist the microscope in checking any inherent shading in the CCD camera–based systems, the glare phenom- SOYFOO ET AL enon, and the level of linear precision. The monitoring procedure installed on our computer-assisted microscope showed that neither shading nor glare nor linearity significantly modified our results (data not shown). Ten acini were analyzed in each experimental condition. Salivary flow measurement and protein assay. Mice were anesthetized with ketamine (60 mg/kg body weight) and xylazine (8 mg/kg body weight) and injected subcutaneously with pilocarpine (0.5 mg/kg body weight). Saliva was collected on a cotton swab for 20 minutes. The weight of the cotton swab was measured before and after saliva collection. The weight of the saliva was converted into the volume of saliva, assuming that 1 g represents 1 l. The amount of saliva was normalized to microliters of saliva per gram of body weight per 20 minutes. The protein concentration was determined using the Bradford method (30). Statistical analysis. Data are expressed as the mean ⫾ SEM. The Mann-Whitney U test or the unpaired t-test was used to evaluate statistical differences. All statistical analyses were carried out using GraphPad InStat version 3.00 for Windows 95 (GraphPad Software, San Diego, CA). RESULTS AQP5 mRNA detection by real-time RT-PCR in salivary glands from BALB/c and NOD mice. Real-time RT-PCR analysis of AQP5 mRNA, using GAPDH as reporter gene, was performed on submandibular and parotid glands from BALB/c and NOD mice, ages 8 weeks and 24 weeks. In parotid glands from NOD mice of both ages, as compared with BALB/c mice, the relative AQP5 mRNA levels were not significantly modified. In submandibular glands from 8-week-old and 24-week-old NOD mice, as compared with those from age-matched BALB/c mice, the relative AQP5 mRNA expression levels increased significantly, by 80% and 44%, respectively (Table 1). Western blot quantification of AQP5 expression in salivary glands from BALB/c and NOD mice. Protein immunoblots of membranes from salivary glands of 8-week-old and 24-week-old BALB/c and NOD mice were probed with an affinity-purified anti-AQP5 antibody and an anti–␤-actin antibody. Western blot analysis revealed the presence of an ⬃27-kd band corresponding to AQP5. The band density, also called the measure of volume, is the sum of all pixel intensities composing the spot and depends on both the number of pixels inside the area of the spot and the intensity of these pixels. Linear relationships were determined by plotting separately the values for AQP5 and ␤-actin band volumes versus increasing amounts (0.050–20 g) of salivary gland membrane proteins in BALB/c and NOD mice (data not shown). All further Western blot analyses were EXPRESSION AND DISTRIBUTION OF AQP5 IN NOD MICE Table 1. Relative expression of aquaporin 5 messenger RNA in salivary glands from BALB/c and NOD mice* Age group; salivary gland, mouse Age 8 weeks Parotid gland BALB/c (n ⫽ 4) NOD (n ⫽ 6) Submandibular gland BALB/c (n ⫽ 5) NOD (n ⫽ 6) Age 24 weeks Parotid gland BALB/c (n ⫽ 5) NOD (n ⫽ 5) Submandibular gland BALB/c (n ⫽ 6) NOD (n ⫽ 6) Relative expression, % 100 ⫾ 22 88 ⫾ 13 100 ⫾ 6 180 ⫾ 26† 100 ⫾ 8 116 ⫾ 29 100 ⫾ 6 144 ⫾ 13† * Values are the mean ⫾ SEM. RNA extraction and reverse transcription–polymerase chain reaction were performed as described in Materials and Methods. Messenger RNA expression levels were compared after adjustment to GAPDH as reporter gene. Similar results were obtained when using different reporter genes, including hypoxanthine guanine phosphoribosyltransferase, cyclophilin A, and 18S. † P ⬍ 0.05 versus age-matched BALB/c mice, by unpaired t-test. 2569 performed using 2.5 g of salivary gland membrane protein. AQP5 expression in parotid and submandibular gland membranes from 8-week-old NOD mice was not significantly different from that in membranes from age-matched BALB/c mice (Figures 1A and B). In parotid and submandibular glands from 24-week-old NOD mice, AQP5 expression was significantly increased, by 3.7-fold and 1.7-fold, respectively, compared with that in salivary glands from age-matched BALB/c mice (Figures 1C and D). Histopathologic findings in salivary glands from BALB/c and NOD mice. Parotid and submandibular glands (Figures 2A–C) from 8-week-old and 24-week old BALB/c mice and those from 8-week-old NOD mice were normal. Parotid glands from 24-week-old NOD mice were normal and devoid of any lymphocytic infiltrates (results not shown). Submandibular glands from 24-week-old mice displayed numerous large lymphocytic infiltrates (Figure 2D). Figure 1. Western blot quantification of aquaporin 5 (AQP5) expression in salivary glands from BALB/c and NOD mice. Western blot analysis was performed using anti-AQP5 antibody (1:2,000 dilution) and 2.5 g of membrane (pellet from 17,000g centrifugation) from the parotid (A and C) or submandibular (B and D) glands of 8-week-old (A and B) and 24-week-old (C and D) BALB/c and NOD mice. Bars show the mean and SEM results from 6 mice per group. ⴱ ⫽ P ⬍ 0.05 versus age-matched BALB/c mice, by Mann-Whitney U test. 2570 SOYFOO ET AL Figure 2. Localization of aquaporin 5 (AQP5) in submandibular glands from BALB/c mice at ages 8 weeks and 24 weeks (A and C) and NOD mice at ages 8 weeks and 24 weeks (B and D). AQP5 was localized on submandibular gland sections from BALB/c and NOD mice, as described in Materials and Methods. Images are representative of immunohistochemical staining performed on salivary gland sections from 3 mice per group. Negative control staining was performed in the absence of anti-AQP5 antibody or with anti-AQP5 antibody previously incubated with the immunizing peptide (results not shown). Insets show higher-magnification views. (Original magnification ⫻ 40.) Immunohistochemical localization of AQP5 and quantitative analysis of AQP5-immunoreactive acini in salivary glands from BALB/c and NOD mice. Immunohistochemical analysis revealed AQP5 labeling at both the apical and basolateral membranes of salivary gland acini from BALB/c and NOD mice, at age 8 weeks and age 24 weeks (Figure 3). Computer-assisted microscopy was performed to quantitatively evaluate AQP5 distribution in the immunopositive acini of parotid and submandibular glands from both BALB/c and NOD mice, at age 8 weeks and 24 weeks. A labeling index representing the percent of the acinus area stained with anti-AQP5 antibody was determined for both the apical and basolateral membranes. In 8-week-old mice, the apical and basolateral labeling indices of the parotid glands from NOD mice were not significantly different from those of the parotid glands from BALB/c mice (apical [n ⫽ 6], 35.15 ⫾ 2.01 versus 35.06 ⫾ 1.44% [P ⫽ 0.589]; basolateral [n ⫽ 6], 18.87 ⫾ 1.77 versus 17.42 ⫾ 0.91% [P ⫽ 0.699]) (Figure 3A). In 24-week-old mice, the apical and basolateral labeling indices of the parotid glands from NOD mice were not significantly different from those of the parotid glands from BALB/c mice (apical [n ⫽ 6], 25.09 ⫾ 4.54 versus 14.07 ⫾ 1.60% [P ⫽ 0.065]; basolateral [n ⫽ 6], 6.57 ⫾ 1.56 versus 4.95 ⫾ 0.61% [P ⫽ 0.310]) (Figure 3B). In 8-week-old mice, the apical and basolateral labeling indices of the submandibular glands from NOD mice were not significantly different from those of the submandibular glands from BALB/c mice (apical [n ⫽ 6], 4.03 ⫾ 0.45 versus 3.82 ⫾ 0.43% [P ⫽ 0.699]; basolateral [n ⫽ 6], 0.91 ⫾ 0.14 versus 0.70 ⫾ 0.05% [P ⫽ 0.485]) (Figure 3C). In the submandibular glands from 24-week-old NOD mice, the apical labeling index decreased significantly, by 75% (1.04 ⫾ 0.14 versus 4.13 ⫾ 0.35% in BALB/c mice [n ⫽ 6]; P ⫽ 0.002), while EXPRESSION AND DISTRIBUTION OF AQP5 IN NOD MICE 2571 and 24 weeks of age (Figure 4). At 8 weeks of age, the salivary flow rates in NOD mice (n ⫽ 6) and BALB/c mice (n ⫽ 6) were not statistically significantly different (7.38 ⫾ 0.92 versus 8.90 ⫾ 0.75 l of saliva/gm of body weight/20 minutes; P ⫽ 0.310), while at 24 weeks of age, salivary flow in NOD mice (n ⫽ 6) decreased significantly, by 29%, compared with that in BALB/c mice (n ⫽ 6) (6.95 ⫾ 0.48 versus 9.73 ⫾ 0.88; P ⫽ 0.041). DISCUSSION Figure 3. Quantitative analysis of aquaporin 5 (AQP5)–immunoreactive acini in salivary glands from BALB/c and NOD mice. Quantitative analysis of the labeling index of the labeled acini was performed as described in Materials and Methods. The labeling index was determined for parotid (A and B) and submandibular (C and D) glands from 8-week old (A and C) or 24-week-old (B and D) BALB/c mice (open bars) or NOD mice (solid bars). Bars show the mean and SEM results from 6 mice per group (10 acini for each animal). ⴱ ⫽ P ⬍ 0.05 versus age-matched BALB/c mice, by Mann-Whitney U test. the basolateral labeling index increased significantly, by ⬃50% (2.45 ⫾ 0.11 versus 1.66 ⫾ 0.06% in BALB/c mice [n ⫽ 6]; P ⫽ 0.002) (Figure 3D). Measurement of the salivary flow rate in BALB/c and NOD mice. Pilocarpine-stimulated salivary flow was measured in BALB/c and NOD mice at 8 weeks of age Figure 4. Measurement of the salivary flow rate in BALB/c and NOD mice. Salivary flow was measured in 8-week-old and 24-week-old BALB/c mice (open bars) and NOD mice (solid bars), as described in Materials and Methods. Bars show the mean and SEM results from 6 mice per group. ⴱ ⫽ P ⬍ 0.05 versus age-matched BALB/c mice, by Mann-Whitney U test. In normal salivary glands, localization of AQP5 is restricted to the apical membrane of acinar cells (5,10– 12). The mechanism that leads to saliva secretion involves secretion of a primary isotonic fluid by the acini, from which considerable NaCl is resorbed during passage through the ducts. Ductal cells possess a sodium channel (31,32), a chloride channel (10,31), and a sodium/proton exchanger (10,31) that are likely involved in NaCl reabsorption, as well as a potassium/proton exchanger that may mediate potassium and bicarbonate secretion (31,33). The involvement of AQP5 in saliva secretion was clearly demonstrated using AQP5knockout mice that displayed remarkably reduced salivary flow as well as hypertonic and hyperosmolar saliva (13). Although the biologic mechanisms leading to hyposalivation and hypolacrimation symptoms in SS remain poorly understood, it is increasingly evident that dryness of the eyes and mouth does not result solely from gland destruction. Recent studies have shown that AQP5, a water channel protein, could play a role in the pathogenesis of SS. Indeed, the contribution of impaired AQP5 distribution to the deficiency of fluid secretion, a defining feature of SS, was supported by the abnormal distribution of AQP5 in minor salivary glands (12) and lacrimal glands (19) from patients with SS. Conflicting data describing the normal localization of AQP5 in minor salivary glands from 5 patients with SS (18) could result from the use of distinct antibodies and/or methodologies (12). The NOD mouse, a recognized appropriate model for studying the autoimmune exocrinopathy that is prevalent in patients with SS, was used to investigate the possible involvement of AQP5 in exocrine gland salivary dysfunction. Compared with the parotid and submandibular glands from control BALB/c mice, those from 8-week-old NOD mice had no inflammatory infiltrates, no significant modification in AQP5 protein levels, no significant modification of AQP5 apical and basolateral labeling indices, and no significant salivary 2572 flow decrease. However, the submandibular, but not the parotid, glands demonstrated a significant increase in relative AQP5 mRNA levels. Although a correlation seems to exist between relative AQP5 mRNA levels and AQP5 protein levels in parotid glands from both BALB/c mice and NOD mice, an apparent discrepancy between AQP5 mRNA levels (increased) and protein levels (unchanged) was observed in the submandibular glands and could be attributable to higher protein turnover. Because, from the physiologic point of view, protein levels are more relevant than mRNA levels, it is important to note that in the salivary glands from 8-week-old NOD mice presenting no autoimmune exocrinopathy, AQP5 protein levels and AQP5 localization in the acini were not modified. In contrast, the submandibular glands from 24week-old NOD mice displayed substantial inflammatory infiltrates throughout the glands, a significant increase in AQP5 protein levels, and a significant reduction in the apical labeling index associated with significant augmentation in the basolateral labeling index. In addition, 24-week-old NOD mice displayed a moderately statistically significant decrease in salivary flow (P ⫽ 0.041), which appeared to be marginal, because there was no significant difference between saliva flow rates from 8-week-old and 24-week-old mice. However, although a correlation seems to exist between the increase in the relative AQP5 mRNA levels and the AQP5 protein levels observed in the submandibular glands, the apparent discrepancy between AQP5 mRNA levels (unchanged) and protein levels (increased) in the parotid glands could be attributable to low turnover of protein. In the submandibular acini from 24-week-old NOD mice, impaired AQP5 localization was observed. The 24-week-old NOD mice used in this study displayed submandibular gland lymphocytic infiltrates and a moderately reduced salivary flow rate, which is consistent with results of other studies (22–24). A recent study demonstrated abnormal AQP5 distribution in parotid and submandibular glands from 20–26-week-old NOD mice (using 9-week-old BALB/c mice as controls) (24). In the current study, additional data concerning the distribution of AQP5 in salivary glands from NOD mice of 2 ages were provided by performing quantitative analysis of the labeling index of AQP5-labeled acini. Moreover, at the molecular level, the expression of AQP5 (at both the mRNA and protein levels) was evaluated. At the physiologic level, salivary gland function was determined by measuring saliva flow. The data collected from NOD mice of 2 ages (8 weeks old and 24 weeks old, representing 2 distinct stages of SOYFOO ET AL the disease) were then compared with those collected from matched control BALB/c mice. The unexpected increase in AQP5 protein expression observed in the submandibular glands from 24-week-old NOD mice might reflect a compensatory mechanism, because AQP5 does not distribute properly in acinar cells and is therefore unlikely to play its normal physiologic role. The absence of inflammatory infiltrates in the parotid glands from 24-week-old NOD mice was quite unexpected. Parotid glands from 24-week-old NOD mice showed no increase in mRNA but did have an increase in the AQP5 protein level, which could not reflect a compensatory mechanism, because AQP5 distributes normally in acinar cells and is therefore likely to play its normal physiologic role. Because hyposalivation has been shown to be preceded by inflammatory changes in salivary glands from NOD mice (23), submandibular glands, rather than parotid glands, are likely to contribute to the moderate saliva flow dysfunction observed in our study. A link seems to exist between inflammation and abnormal AQP5 distribution. This hypothesis is based on observations that the parotid glands from 24-week-old NOD mice displayed no inflammatory infiltrates and no changes in AQP5 localization, while the submandibular glands displayed inflammatory infiltrates and abnormal AQP5 distribution. Further studies will be required to understand the mechanisms leading to the increased level of AQP5 protein in parotid glands from 24-week-old NOD mice, and to characterize the possible link existing between inflammatory changes and abnormal AQP5 distribution. In 24-week-old mice, several explanations might account for the discrepancy between the moderate reduction in the salivary flow rate and the extent of AQP5 misdistribution. First, whole saliva (containing saliva from the parotid and submandibular glands) was collected, although parotid glands predominantly contribute to the salivary flow rate (34). Second, increased AQP5 protein expression in the submandibular glands might compensate for AQP5 misdistribution. Finally, the ability of water to permeate properly localized AQP5 might be reduced. Abnormal expression and distribution of AQP5 in acinar cells from submandibular glands from NOD mice and minor salivary glands from patients with SS (12) suggest a defect in AQP5 distribution in autocrine exocrinopathy. Similar observations of altered protein trafficking in patients with SS have been shown for the Na ⫹ -independent chloride–bicarbonate anion exchanger (35). In SS, autoantibodies against AQP5 are unlikely to account for its abnormal distribution, be- EXPRESSION AND DISTRIBUTION OF AQP5 IN NOD MICE cause no antibodies to human AQP5 were detected in sera from patients with SS (12). AQP5 down-regulation and possibly altered distribution are also likely involved in the symptom of dry mouth following radiation therapy for head and neck cancer (36). Consequently, water channel gene transfer has been considered as a potential therapeutic approach in patients with SS, as well as in patients with impaired saliva production resulting from radiation therapy for head and neck cancer (37). Since the discovery that increased fluid secretion could be observed after adenovirus-mediated gene transfer of AQP1 cDNA to irradiated rat salivary glands (14), transfer of human AQP1 cDNA for the correction of radiation-induced hypofunction has been extensively documented in several species (37). Because the NOD mice used in this study demonstrated moderately decreased salivary flow, it would be of particular interest to address in further studies whether impaired localization of AQP5 is also observed in other NOD mouse models demonstrating either no or a very significant salivary flow decrease (24,38). Additional studies investigating the basic mechanisms of AQP5 trafficking may also provide valuable insight into the regulation of AQP5 distribution in salivary glands from both control and NOD mice. In conclusion, submandibular glands from 24week-old NOD mice displayed inflammatory infiltrates, increased AQP5 protein expression, and impaired AQP5 distribution. In 24-week-old NOD mice, however, the moderately statistically significant decrease in the salivary flow rate did not match the extent of AQP5 misdistribution. ACKNOWLEDGMENTS We thank M. Stiévenart for secretarial assistance and Jos Laureys for technical assistance. AUTHOR CONTRIBUTIONS Dr. Delporte had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study design. Soyfoo, De Vriese, Devuyst, Steinfeld, Delporte. Acquisition of data. Soyfoo, De Vriese, Debaix, Mathieu, Devuyst, Delporte. Analysis and interpretation of data. Soyfoo, De Vriese, Debaix, Martin-Martinez, Devuyst, Steinfeld, Delporte. Manuscript preparation. Soyfoo, De Vriese, Debaix, MartinMartinez, Mathieu, Devuyst, Steinfeld, Delporte. Statistical analysis. Soyfoo, Debaix, Delporte. 2573 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. REFERENCES 1. Fox RI. Sjögren’s syndrome [review]. Lancet 2005;366:321–31. 2. Andoh Y, Shimura S, Sawai T, Sasaki H, Takishima T, Shirato K. 23. Morphometric analysis of airways in Sjögren’s syndrome. Am Rev Respir Dis 1993;148:1358–62. Tornwall J, Konttinen Y, Tuomen R, Tornwall M. Salivary gland acinar epithelial cells are deficient in their protein kinase C expression in Sjögren’s syndrome. Lancet 1997;349:1914–5. Bacman S, Sterin-Borda L, Camusso JJ, Arana R, Hubscher O, Borda E. Circulating antibodies against rat parotid gland M3 muscarinic receptors in primary Sjögren’s syndrome. Clin Exp Immunol 1996;104:454–9. Agre P, Brown D, Nielsen S. Aquaporin water channels: unanswered questions and unresolved controversies [review]. Curr Opin Cell Biol 1995;7:472–83. Borgnia M, Nielsen S, Engel A, Agre P. Cellular and molecular biology of the aquaporin water channels [review]. Annu Rev Biochem 1999;68:425–58. King LS, Agre P. Pathophysiology of the aquaporin water channels [review]. Annu Rev Physiol 1996;58:619–48. Verkman AS, van Hoek AN, Ma T, Frigeri A, Skach WR, Mitra A, et al. Water transport across mammalian cell membranes [review]. Am J Physiol 1996;270(1 Pt 1):C12–30. Delporte C, Steinfeld S. Distribution and roles of aquaporins in salivary glands [review]. Biochim Biophys Acta 2006;1758: 1061–70. He X, Tse CM, Donowitz M, Alper SL, Gabriel SE, Baum BJ. Polarized distribution of key membrane transport proteins in the rat submandibular gland. Pflugers Arch 1997;433:260–8. Nielsen S, King LS, Christensen BM, Agre P. Aquaporins in complex tissues. II. Subcellular distribution in respiratory and glandular tissues of rat. Am J Physiol 1997;273:C1549–61. Steinfeld S, Cogan E, King LS, Agre P, Kiss R, Delporte C. Abnormal distribution of aquaporin-5 water channel protein in salivary glands from Sjögren’s syndrome patients. Lab Invest 2001;81:143–8. Ma T, Song Y, Gillespie A, Carlson EJ, Epstein CJ, Verkman AS. Defective secretion of saliva in transgenic mice lacking aquaporin-5 channels. J Biol Chem 1999;274:20071–4. Delporte C, O’Connell BC, He X, Lancaster H, O’Connell A, Agre P, et al. Increased fluid secretion following transfer of the aquaporin-1 cDNA to irradiated rat salivary glands. Proc Natl Acad Sci U S A 1997;94:3268–73. Li J, Nielsen S, Dai Y, Lazowski KW, Christensen EI, Tabak LA, et al. Examination of rat salivary glands for the presence of the aquaporin CHIP. Pflugers Arch 1992;428:455–60. Nielsen S, Smith BL, Christensen EI, Agre P. Distribution of the aquaporin CHIP in secretory and resorptive epithelia and capillary endothelia. Proc Natl Acad Sci U S A 1993;90:7275–9. Yang B, Zhao D, Solenov E, Verkman AS. Evidence from knockout mice against physiologically significant aquaporin 8-facilitated ammonia transport. Am J Physiol Cell Physiol 2006;291: C417–23. Beroukas D, Hiscock J, Jonsson R, Waterman SA, Gordon TP. Subcellular distribution of aquaporin 5 in salivary glands in primary Sjögren’s syndrome. Lancet 2001;358:1875–6. Tsubota K, Hirai S, King LS, Agre P, Ishida N. Defective cellular trafficking of lacrimal gland aquaporin-5 in Sjögren’s syndrome. Lancet 2001;357:688–9. Van Blokland SC, Versnel MA. Pathogenesis of Sjögren’s syndrome: characteristics of different mouse models for autoimmune exocrinopathy [review]. Clin Immunol 2002;103:111–24. Humphreys-Beher MG. Animal models for autoimmune diseaseassociated xerostomia and xerophthalmia. Adv Dent Res 1996;10: 73–5. Humphreys-Beher MG, Peck AB. New concepts for the development of autoimmune exocrinopathy derived from studies with the NOD mouse model. Arch Oral Biol 1999;44 Suppl 1:S21–5. Jonsson MV, Delaleu N, Brokstad KA, Berggreen E, Skarstein K. Impaired salivary gland function in NOD mice: association with 2574 24. 25. 26. 27. 28. 29. 30. changes in cytokine profile but not with histopathologic changes in the salivary gland. Arthritis Rheum 2006;54:2300–5. Konttinen YT, Tensing EK, Laine M, Porola P, Tornwall J, Hukkanen M. Abnormal distribution of aquaporin-5 in salivary glands in the NOD mouse model for Sjögren’s syndrome. J Rheumatol 2005;32:1071–5. Ni J, Cnops Y, McLoughlin RM, Topley N, Devuyst O. Inhibition of nitric oxide synthase reverses permeability changes in a mouse model of acute peritonitis. Perit Dial Int 2005;25 Suppl 3:S11–4. Delporte C, O’Connell BC, He X, Ambudkar IS, Agre P, Baum BJ. Adenovirus-mediated expression of aquaporin-5 in epithelial cells. J Biol Chem 1996;271:22070–5. Dai YS, Ambudkar IS, Horn VJ, Yeh CK, Kousvelari EE, Wall SJ, et al. Evidence that M3 muscarinic receptors in rat parotid gland couple to two second messenger systems. Am J Physiol 1991;261(6 Pt 2):C1063–73. Camby I, Nagy N, Lopes MB, Schafer BW, Maurage CA, Ruchoux MM, et al. Supratentorial pilocytic astrocytomas, astrocytomas, anaplastic astrocytomas and glioblastomas are characterized by a differential expression of S100 proteins. Brain Pathol 1999;9:1–19. Steinfeld S, Penaloza A, Ribai P, Decasetecker C, Danguy A, Gabius HJ, et al. D-mannose and N-acetylglucosamine moieties and their respective binding sites in salivary glands of Sjögren’s syndrome. J Rheumatol 1999;26:833–41. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976;72:248–54. SOYFOO ET AL 31. Cook DI, Van Lennep EW, Roberts ML, Young JA. Secretion by the major salivary glands. In: Johnson LR, editor. Physiology of the gastrointestinal tract. New York: Raven; 1994. p. 1061–117. 32. Moran A, Davis VH, Turner RJ. Na⫹ channels in membrane vesicles from intralobular salivary ducts. Am J Physiol 1995;268: C350–5. 33. Knauf H, Lubcke R, Kreutz W, Sachs G. Interrelationships of ion transport in rat submaxillary duct epithelium. Am J Physiol 1982;242:F132–9. 34. Levy S, Nagler A, Okon S, Marmary Y. Parotid salivary gland dysfunction in chronic graft-versus-host disease (cGVHD): a longitudinal study in a mouse model. Bone Marrow Transplant 2000;25:1073–8. 35. Vazquez JJ, Vazquez M, Idoate MA, Montuenga L, MartinezAnso E, Castillo JE, et al. Anion exchanger immunoreactivity in human salivary glands in health and Sjögren’s syndrome. Am J Pathol 1995;146:1422–32. 36. Li Z, Zhao D, Gong B, Xu Y, Sun H, Yang B, et al. Decreased saliva secretion and down-regulation of AQP5 in submandibular gland in irradiated rats. Radiat Res 2006;165:678–87. 37. Baum BJ, Zheng C, Cotrim AP, Goldsmith CM, Atkinson JC, Brahim JS, et al. Transfer of the AQP1 cDNA for the correction of radiation-induced salivary hypofunction. Biochim Biophys Acta 2006;1758:1071–7. 38. Lodde BM, Mineshiba F, Kok MR, Wang J, Zheng C, Schmidt M, et al. NOD mouse model for Sjögren’s syndrome: lack of longitudinal stability. Oral Dis 2006;12:566–72.