Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Lab Invest. Author manuscript; available in PMC Apr 1, 2011.
Published in final edited form as:
PMCID: PMC2847636

Conditional overexpression of TGF-β1 disrupts mouse salivary gland development and function


Transforming growth factor-β (TGF-β) signaling is known to affect salivary gland physiology by influencing branching morphogenesis, regulating ECM deposition, and controlling immune homeostasis. To study the role of TGF-β1 in the salivary gland, we created a transgenic mouse (β1glo) that conditionally over-expresses the active TGF-β1 upon genomic recombination by the Cre recombinase. The β1glo mice were bred with a MMTV (mouse mammary tumor virus)-Cre (MC) transgenic line that expresses the Cre recombinase predominantly in the secretory cells of both the mammary and salivary glands. Although most of the double positive (β1glo/MC) pups die either in utero or just after birth, clear defects in salivary gland morphogenesis could be seen such as reduced branching and increased mesenchyme. The β1glo/MC mice that survived into adulthood, however, had hyposalivation due to salivary gland fibrosis and acinar atrophy. Increased TGF-β signaling was observed in the salivary gland with elevated phosphorylation of Smad2 and a concomitant increase in ECM deposition. In particular, aberrant TGF-β1 overexpression caused salivary gland hypofunction in this mouse model because of the replacement of normal glandular parenchyma with interstitial fibrous tissue. These results further implicate TGF-β in pathological cases of salivary gland inflammation and fibrosis that occur with chronic infections in the glands or with the autoimmune disease, Sjögren’s syndrome or with the radiation therapy given to head-and-neck cancer patients.

Keywords: Transforming growth factor-β, fibrosis, salivary glands, saliva


TGF-β1 is a multifunctional cytokine that influences salivary gland development and homeostasis. In particular, TGF-β1 is known to regulate ECM deposition not only by inducing biosynthesis of collagens and fibronectin (1, 2) but also by promoting the expression of protease inhibitors. Furthermore, TGF-β1 is able to encourage epithelial-mesenchymal transition in some cells that can result in more ECM producing myofibroblasts (3, 4). Tissue damage to the salivary glands from inflammation or radiation exposure can result in reparative TGF-β-induced ECM production. ECM deposition by TGF-β1 shapes epithelial-mesenchymal interactions throughout salivary gland organogenesis as well. Along with regulating mesenchymal production of ECM, TGF-β1 can also influence salivary gland development by controlling cellular growth and differentiation. The secretion of TGF-β1 inhibits the proliferation of epithelial cells by downregulating c-myc while simultaneously increasing the expression of cyclin-dependent kinase (cdk) inhibitors such as p15, p21 and p27 (5). Lastly, TGF-β1 affects salivary gland physiology by regulating angiogenesis (6) and by suppressing inflammation (7).

TGF-β1 and its other two mammalian isoforms, TGF-β2 and TGF-β3, are all expressed in the salivary gland during development, which suggests an important role for this cytokine in glandular organogenesis (8). Specifically, the expression of TGF-β1 seems to coincide with salivary gland differentiation (9). TGF-β1 is originally detected in both the epithelium and messenchyme during the initial bud stage but becomes immunolocalized to only the branching epithelia later in development (8). In a 14.5 day post coitum mouse embryo, TGF-β1 mRNA expression is localized in the epithelial end buds, sights of active branching in the developing salivary gland (10). During this stage of development, TGF-β1 may act in a paracrine manner on the mesenchyme and an autocrine manner on epithelial cell growth. Even though the TGF-β1 mRNA is localized at sights of active branching, exogenous TGF-β1 in salivary gland cultures, which mimics overexpression, inhibitis branching morphogenesis (11). Epithelial growth is disrupted and the ducts appear elongated. Following glandular development, TGF-β1 expression, however, is localized to ductal epithelium in the submandibular gland and is absent in the secretory acini (12, 13).

Besides its role in organogenesis, TGF-β also impacts salivary gland physiology by regulating ECM production, particularly in response to tissue injury. Aberrant expression of TGF-β1 is often associated with cases of pathological fibrosis. In the salivary gland, fibrosis specifically causes constriction of secretory components, leading to hyposalivation and xerostomia (14). Salivary gland fibrosis typically occurs after repeated episodes of inflammation such as following chronic infections in the glands or with the autoimmune disease, Sjögren’s syndrome. Fibrosis of the glands also occurs because of tissue damage from radiation, particularly during radiotherapy treatment for head and neck cancer (15). Interestingly, radiation exposure has been shown to induce TGF-β1 expression (16).

We developed a transgenic mouse that conditionally produces TGF-β1 (β1glo) in order to understand the role of TGF-β signaling in salivary gland development and homeostasis. The transgene requires Cre mediated excision of an intervening floxed EGFP gene in order for a ubiquitous promoter to transcribe a TGF-β1 cDNA. In the transgene, the TGF-β1 cDNA is mutated to prevent assembly of the latent associated peptide, in order to allow direct binding of the secreted ligand onto the cell surface receptors. We bred these β1glo mice to a mouse mammary tumor virus (MMTV) Cre (offspring are MC) transgenic line (17) that strongly expresses the Cre recombinase in the both the mammary and salivary glands. The broad expression of Cre in the transgenic mice, however, generated a severe phenotype with most of the double positive (β1glo/MC) pups either dying in utero or within 24 hours after birth. Nonetheless, the effect of TGF-β1 on the salivary gland could clearly be seen in the β1glo/MC pups with increased mesenchyme and disrupted branching. For the β1glo/MC mice that survive into adulthood, the salivary glands were severely fibrotic with signs of atrophy in both the granular convoluted ducts (GCDs) and the acini, and this was associated with hyposalivation.


Construction of pCLE-β1glo transgenic plasmid

To create a transgenic mouse with recombination activated TGF-β1 expression, we generated the pCLE-β1glo vector by subcloning an active hemaglutanin (HA) epitope tagged TGF-β1 cDNA (18) into pCLE (19). The vector pCLE contains a 1.7 kb β-actin promoter combined with a CMV-IE enhancer (CAG promoter) for ubiquitous expression of the transgene. Downstream of the promoter is a 1 kb enhanced green fluorescent protein (EGFP) gene that is flanked by loxP sites. A constitutively active 1.2 kb TGF-β1 cDNA was subcloned into the pCLE vector after EGFP. The cDNA contains the porcine sequence for TGF-β1 from pPK9a, where two cysteines (C223, C225) were mutated to serines to prevent the latent associated peptide, LAP, from assembling around the mature TGF-β1 dimer (20). PCR mutagenesis was employed to add an HA epitope tag onto the ligand to distinguish exogenous TGF-β1 from the endogenous protein (18).

Cell Culture

COS7 cells were transfected with pCLE-β1glo to study not only the efficiency of recombination mediated expression of TGF-β1 from transgene, but also to test the signaling capability of the epitope tagged protein as well. COS7 cells were transfected with either pCLE-β1glo alone or together with the Cre expression vector pBS185 (21) using Lipofectamine LTX (Invitrogen). After overnight incubation, the medium was replaced with serum-free Opti-MEM supplemented with MITO plus (BD Bioscience). Supernatants were collected after 48 h and cells were lysed with T-PER (Pierce) supplemented with a Complete Mini Protease Inhibitor Cocktail (Roche). The COS7 supernatants were treated with 200 mM PMSF (Sigma) and concentrated using an Amicon Ultra centrifugal filter device (Millipore). Supernatants and cell lysates were run on SDS-PAGE to test for expression of the epitope-tagged version of TGF-β1 from the transgene (see below). To study TGF-β signaling, untreated COS7 supernatants from the transfected cells were diluted 1:3 with fresh serum-free medium and plated onto 5 × 105 HepG2 cells in a 6 cm culture dish. The cells were lysed after 30 minutes and cell lysates were run on a Western blot to determine the level of phosphorylation of Smad2 (see below).

Generation and Genotyping of the β1glo Mice

After testing of the pCLE-β1glo plasmid in vitro, the transgenic vector was microinjected to generate transgenic mice. The founder lines were genotyped using Southern blot analysis. Tail DNA was digested using NheI, a unique restriction enzyme site within the transgenic vector, and the whole transgene (4.5 kb) was radiolabeled for use as a probe. β1glo mice were then primarily identified through both PCR and with GFP visualization using a Macro Imaging System from Light Tools Research. Mice were genotyped using the following pair of primers: CAGG; 5’-CTCTAGAGCCTCTGCTAACC-3’ and EGFP; 5’-GGTGCAGATGAACTTCAGGG-3’. To detect recombination within the transgene, a reverse primer in the TGF-β1 cDNA was designed (Beta; 5’-CGCTTTCCACCATTAGCAC). Primers were also generated to detect the HA tag (HA For; 5’-CATACGACGTGCCAGACTAC-3’ and pB1-TGA Rev; 5’-TCAGCTGCACTTGCAGGAACG) and to ensure integration of the SV40 pA (pB1-TGA For; 5’-CGTTCCTGCAAGTGCAGCTGA-3’ and SV40pA; 5’-GATGAGTTTGGACAAACCACAAC-3’). PCR was performed for 40 cycles consisting of 94°C for 30 sec, 57°C for 30 sec, and 72°C for 1 min. The β1glo mice were bred to the MMTV-Cre mice that were generated as previously described (17). All experimental studies and procedures were approved by the Animal Care and Use Committee of the National Institute of Dental and Craniofacial Research, NIH.

Saliva Collection

Saliva production was tested in the resulting adult β1glo/MC progeny by using a subcutaneous injection of 0.5 mg/ml pilocarpine (Sigma) at 0.005 mg per 100 gram body weight to stimulate salivation. The secreted saliva from the pilocarpine injected mice was collected using a micro-hematocrit tube for a total of 30 minutes.

Histopathology and Immunohistochemistry

Initial β1glo/MC mice were often runted and would typically die perinatally. Newborn mice were euthanized along with control littermates for pathological study. Sagittal sections were used for histological staining and immunohistochemistry. β1glo/MC mice that evaded perinatal lethality were sacrificed between 1-week to 10-months of age for analysis. Major organs such as heart, liver, skin, pancreas, kidney, and lung were collected along with glandular tissue (salivary, mammary, lacrimal, etc.) from both wild-type and β1glo/MC mice. Tissues were fixed in 10% buffered formalin and embedded in paraffin. Controls include both β1glo and wild type littermates. For histopathology, 5 µm sections were stained with hematoxylin and eosin (H&E), Masson’s trichrome, or mucicarmine. Sections were deparaffinized in xylene, rehydrated with descending grades of ethanol, treated with 3% hydrogen peroxide for 30 minutes, blocked for 30 minutes, and incubated with primary antibody (see below) in a 1% BSA solution. Next, the sections were treated with Rabbit on Rodent HRP-Polymer (Biocare) or with a biotinylated secondary antibody (Vector Labs) followed by Vectastain ABC reagent (Vector Labs). Antibody binding was detected using liquid DAB (Biogenex). Immunostaining was performed using the following primary antibodies: 1:100 anti-aquaporin 5 (AQP5; Calbiochem #1078615), 1:400 anti-Connective Tissue Growth Factor (CTGF) (Abcam #ab6992), 1:500 anti-Smad2, phospho-specific (Ser465/467) (Millipore #AB3849), 1:500 anti-Smooth Muscle Actin (Millipore # CBL171), and 1:200 anti TGF-Beta1 (Promega G1221). Immunohistochemistry for extracellular TGF-Beta1 (anti-CC 1-30) (22) and immunofluorescence for AQP5 (23) was performed as previously described.

Western Blot and ELISA

Salivary glands were collected from wild type and β1glo/MC mice and tissue lysates were prepared using T-PER (Pierce) supplemented with a Protease Inhibitor Cocktail (Roche). Serum from the mice was analyzed using a TGF-β1 ELISA (R&D Systems #MB100B). Tissue lysates, cell culture lysates, and cultured supernatants were run on a NuPAGE 4-12%BIS-Tris Gels (Invitrogen) using MES buffer. Proteins were transferred onto either an Immobilon – PSQ Transfer Membrane (Millipore) or a 0.45 µm Nitrocellulose membrane (Invitrogen), depending on the molecular weight of the protein. The transferred proteins were incubated at 4°C overnight with primary antibody (see below) in Tris buffered saline containing 5% milk and 0.05% Tween 20. Blots were washed and incubated for one hour with HRP conjugated secondary antibodies (Santa Cruz). Signal was detected using SuperSignal West Pico chemiluminescent substrate (Pierce). For Western blots, the following antibodies were used: 1:1000 anti-Actin (Millipore # MAB1501), 1:5000 anti-Green Fluorescent Protein (Roche #11814460001), 1:100 anti-Glyceraldehyde 3-phosphate dehydrogenase (Abcam #ab9485), 1:50 anti-HA (Santa Cruz #sc-805), 1:500 anti-Smad2 (Invitrogen #51-1300), and 1:500 anti-Smad2, phospho-specific (Ser465/467) (Millipore #AB3849).

Preparation of RNA and RT-PCR

Both RNA and DNA were extracted from wild type and β1glo/MC salivary glands using TRIzol reagent (Invitrogen) according to the manufacturers protocol. Extracted RNA was treated with TURBO DNase (Applied Biosystem). Using random primers (Invitrogen), about 500 ng was reverse-transcribed into cDNA through Super Script III reverse transcriptase (Invitrogen). To detect genomic DNA recombination or RNA expression of the HA tag, PCR amplification was performed using the primers listed above (Generation and Genotyping of the β1glo Mice).


Generation of β1glo mice

To create a mouse model employing conditional overexpression of TGF-β1, a transgenic construct, pCLE-β1glo (Fig. 1A), was engineered by subcloning an active HA epitope-tagged version of the TGF-β1 cDNA (18) into pCLE (19), an expression vector amenable to targeted gene activation through site-specific recombination (24, 25). The transgenic vector pCLE-β1glo contains a global promoter for ubiquitous expression of the TGF-β1 cDNA, but its transcription is blocked by the placement of an intervening floxed EGFP gene. Using the Cre recombinase, however, the EGFP gene can be excised to juxtapose the promoter and the TGF-β1 cDNA together to thereby activate its expression.

Figure 1
Design and testing of the transgenic construct for generation of β1glo mice. (A) Schematic of the β1glo transgene, including location of primers used for genotyping the mice. The vector (pCLE-β1glo) consists of a β-actin ...

To test the transgenic construct for recombination activated TGF-β1 expression, pCLE-β1glo was transfected into COS7 cells with or without pBS185 (21), a plasmid containing the gene for the Cre recombinase. While cells transfected with pCLE-β1glo alone had no TGF-β1 expression, the cells co-transfected with Cre had high levels of TGF-β1 secreted into the culture medium, as determined with both anti-TGF-β1 and anti-HA tag antibodies (Fig. 1B). HepG2 cells were then incubated with transfected cell supernatants to determine if the secreted epitope-tagged TGF-β1 protein could activate cell signaling. As seen with phosphorylation of the downstream messenger protein Smad2 (Fig. 1C), the secreted epitope-tagged ligand from the dually transfected cells could directly activate the TGF-β signaling pathway.

After the transgenic construct was tested, pCLE-β1glo was microinjected to create the β1glo founder lines. The founder lines were genotyped by Southern blot analysis (Fig. 1D), in which a 4.5 kb band was detected corresponding to the size of the transgene. Integration of the transgene was also confirmed using PCR primers to the floxed EGFP gene, the HA tagged TGF-β1 cDNA, and the flanking 2X SV40 pA (data not shown). Three of the β1glo founder lines were selected for further expansion and all of the lines were bred to maintain a heterozygous state (Fig. 1E). All of the β1glo mice were healthy and viable without any toxicity due to the transgene integration. In addition, none of the mice showed evidence of any TGF-β1 induced pathology due to read through transcription past the floxed EGFP attenuator (26).

Breeding of β1glo to MMTV-Cre mice

The β1glo mice were bred with a MMTV-Cre line (17) that strongly, though not uniquely, expressed the Cre recombinase in the salivary glands. The MMTV-LTR used in the Cre transgenic mice predominantly limits its expression to the striated ductal cells of the salivary gland (27, 28). When the β1glo mice were crossed with the MMTV-Cre line, most pups died in utero with only about 10% of the β1glo/MC positive pups born instead of the 25% expected. Of the surviving β1glo/MC pups, most were runted as compared to their littermates. Initially, for all three transgenic lines tested, these β1glo/MC typically would die within 24 hours. However, in one of the β1glo lines, pups were born that survived beyond the perinatal period. These pups could eventually live beyond one year of age. The embryonic and early postnatal lethality, however, seen with most of the β1glo/MC pups may be due to the broad expression pattern of the Cre recombinase in the MMTV-Cre transgenic line (17).

Even though the phenotype of the newborn β1glo/MC mice was more severe than expected, the role of TGF-β1 in salivary gland development could still be studied. In the β1glo/MC mice, the submandibular gland was particularly dysplastic due to TGF-β1 overexpression (Fig. 2A & B). While the submandibular gland was disrupted in the β1glo/MC mice, the sublingual gland in one line, at least, was histologically normal (data not shown). This phenotype corresponds to the reported Cre expression pattern in the MMTV-Cre mice where little recombination occurs in the sublingual gland (17). In the salivary gland of the β1glo/MC mice, activation of the TGF-β signaling pathway was seen with increased downstream phosphorylation of Smad2 (Fig. 2C & D). Branching in the submandibular gland was inhibited and mesenchyme was increased for all three lines of the β1glo/MC mice. The increased mesenchyme in the glands, however, lacked the collagen fibril trichrome staining typical of progressive fibrosis (data not shown). Immunostaining was performed for AQP5, a functional marker of acinar cell polarity, in order to further examine the disrupted morphology of the salivary gland in the β1glo/MC mice. AQP5, a water channel essential for saliva secretion (29, 30), is localized on the apical membrane of the acinar cells of the control mice as seen with the early terminal-web patterning in the salivary gland (31, 32). However, the β1glo/MC mice appeared to have aberrant and mislocalized AQP 5 staining, possibly as an indirect consequence of the dyplasitc development induced by TGF-β1 overexpression (Fig. 2E & F).

Figure 2
Disrupted branching morphogenesis in newborn β1glo/MC mice. A–B: H&E sections showing dysplastic growth of the submandibular gland in the β1glo/MC mice (B) as compared to the β1glo control –line c8 (A). ...

Although the salivary glands in the newborn pups were dysplastic, no other organs in the mice showed significant developmental defects that could be detected histologically. No difference could be detected in the circulating levels of total TGF-β1 between β1glo/MC and control pups with ELISAs (data not shown). Because all of the major organs such as the heart and lungs showed no noticeable defects, we suspect that the rapid perinatal lethality may be due to a skin barrier defect. In particular, Cre mediated recombination is known to occur in the skin using the MMTV-Cre mice (17, 33) and a compromised skin barrier may have been a consequence of the anti-proliferative effect of TGF-β1 on epithelial cells (34).

Lack of Salivation in the Adult β1glo/MC Mice

Although most of the β1glo/MC pups born alive die just after birth, one of our β1glo lines was able to produce offspring that did not succumb to perinatal lethality (n = 12; 7 males and 5 females were studied between 1 week to 1 year of age). A few individual mice were even capable of living beyond one year of age. However, the percentage of pups dying in utero did not change even while more newborn pups survived into adulthood. The adult β1glo/MC mice were generally smaller than their littermates with ruffled fur and malformed ears (Fig. 3A & B). This phenotype was likely due to Cre expression in the skin and hair follicles by the MMTV-LTR.

Figure 3
Adult β1glo/MC mice display acinar atrophy and fibrosis. A–B: The β1glo/MC mice (B) are smaller than their littermates (A) with ruffled fur and deformities in the pinna of the ears. (C) PCR demonstrates transgene recombination ...

DNA and RNA were extracted from the salivary glands to test for both recombination of the transgene and expression of the active TGF-β1 in the β1glo/MC mice. A 600 bp PCR product was amplified in only the β1glo/MMTV-Cre mice using primers in the promoter and the TGF-β1 cDNA. The size of the PCR product suggests that the 1 kb EGFP gene was excised in the salivary gland by proper Cre mediated recombination (Fig. 3C). RT-PCR was then used with a primer specific to the HA in order to confirm the expression of the transgenic epitope tagged TGF-β1 in the β1glo/MC mice. A 400 bp PCR product was generated from the salivary glands with primers from the start of the HA tag to the end of the TGF-β1 cDNA (Fig. 3C).

In the β1glo/MC mice, Cre-mediated TGF-β1 expression resulted in a profound hyposalivation. Although the cholinergic agonist pilocarpine could stimulate a median of 183 ± 64 mg of saliva (5.0 ± 1.8 mg saliva/gram mouse) from the control mice, no measurable amount of salivation could be induced in the β1glo/MC mice at 5 to 10 months of age (Fig. 3D). Most animals displayed a dry mouth even with administration of pilocarpine. To determine the cause for this diminished salivation, the salivary glands of β1glo/MC mice at ages from 1 week to 10 months of age were examined histologically for signs of salivary gland pathology (Fig. 3E–H). At all ages, the induction of TGF-β1 triggered aberrant ECM deposition in the salivary glands of the β1glo/MC mice that lead to progressive fibrosis. The growth of the salivary glands appeared to be severely inhibited in the β1glo/MC mice between 1 to 4 weeks of age. Even at 1 week of age initial signs of fibrotic collagen deposition were in the submandibular gland along with dilated ducts (Fig. 3 E & F). In the older mice (between 5–10 months of age), the TGF-β1 overexpression clearly resulted in marked fibrosis and atrophy of the salivary epithelial cells in the submandibular gland (fig 3G–H). Inflammatory infiltrates could also be seen in the salivary glands in addition to the cellular atrophy. The parotid gland also showed dramatic atrophy in the serous acini with only the ducts still present (data not shown). The sublingual gland, however, was completely unaffected as previously seen in the pups for this line (data not shown). There were no significant pathological lesions seen in most of the other tissues examined. Similar to Pierce et al. (35), the mammary gland in one adult female β1glo/MC mouse (7 months of age) displayed no increase in periductal connective tissue, even though active TGF-β1 expression was induced using MMTV-Cre. Two male mice, however, had mild tubular atrophy of the testes and one of these had severe acinar atrophy in the pancreas. No difference could be detected between the β1glo/MC and the controls in the circulating serum levels of total TGF-β1 (median 67 ± 34 vs. 67 ± 19 ng/ml) as well.

Activated TGF-β Signaling in the β1glo/MMTV-Cre Mice

Cre-mediated TGF-β1 expression was further examined in the salivary glands of the β1glo/MC mice. Activation of the TGF-β signaling pathway by Smad2 phosphorylation was only detectable in the salivary glands from the β1glo/MC mice (Fig. 4A). Two bands were seen using anti-phospho-Smad2 (Ser465/467), an antibody that may also cross react with phosphorylated Smad3 at its equivalent sites (Ser423/425). Immunostaining of the sections revealed elevated TGF-β1 expression in the β1glo/MC salivary glands as compared to the controls (Fig.4 B–C). TGF-β1 staining was mainly detected in the ductal cells and was not seen in the few remaining acinar cells. Increased TGF-β1 expression was additionally confirmed using a specific antibody to extracellular TGF-β1 (22) (Fig.4 D–E). As with the Western blot, increased Smad2 phosphorylation was also seen in the β1glo/MC salivary gland using immunohistochemistry (Fig.4 F–G).

Figure 4
Recombination mediated activation of TGF-β1 in the adult β1glo/MC mice. (A) Western blot showing Smad2 phosphorylation in the salivary glands from the β1glo/MC mice. B–C: Compared to the control (B), increased TGF-β1 ...

Fibrosis and Acinar Cell Atrophy in the β1glo/MMTV-Cre Mice

The salivary glands from the adult β1glo/MC mice were then examined to characterize the extent of fibrosis caused by TGF-β1 overexpression. Masson’s trichrome staining showed abnormal collagen deposition throughout the β1glo/MC submandibular gland (Fig. 5A–B). The blue collagen staining suggests that these glands have increased TGF-β1- induced ECM production (1). Smooth muscle actin (Fig. 5C–D) and connective tissue growth factor (CTGF) (Fig. 5E–F) were also elevated in the β1glo/MC salivary glands. In cases of fibrosis, TGF-β1 often increases expression of these proteins (1) with smooth muscle actin being a common marker for activated myofibroblasts in fibrotic lesions. The smooth muscle actin staining appears periductally either through recruitment of myofibroblasts by TGF-β1 or with the induction of an epithelial-mesenchymal transition through TGF-β signaling (36).

Figure 5
Immunostaining for markers of fibrosis A–B: When compared to the control (A), intense blue collagen fibrils are seen in the β1glo/MC gland (B) using Masson’s trichrome stain. C–D: Arrows indicate altered staining pattern ...

Due to the hyposalivation and fibrosis in the β1glo/MC mice, the salivary glands were examined for histological changes in acinar cells indicative of functional abnormalities that could result from TGF-β1 overexpression. In the salivary glands of adult β1glo/MC mice, immunostaining for the essential water channel AQP5 (Fig. 6A–B) was decreased. Reduced levels of AQP5 lead to decreased saliva secretion and salivary gland hypofunction (27, 28). Interestingly, down-regulation of AQP5 can be caused by radiation-induced injury to the salivary gland (37, 38) and radiation damage can to lead to increased TGF-β1 expression (16). Although AQP5 also plays an important role in pulmonary secretions, the β1glo/MC mice did not show altered expression or trafficking of AQP5 in the lungs (data not shown), suggesting that the MMTV-Cre transgenic line is limiting recombination mediated TGF-β1 expression primarily to the salivary gland. Along with alterations in AQP5 expression, compromised salivary function also was suggested by using mucicarmine, where the decreased red staining denotes reduced mucin production in the β1glo/MC mice (Fig 6C–D).

Figure 6
Immunochemical staining to examine potential salivary gland function in adult mice. A–B: Arrows show more aquaporin 5 staining in a 6 month old male control littermate (A) when compared to the β1glo/MC mouse (B). C–D: Mucicarmine ...


By generating the β1glo mice and breeding them to an MMTV-Cre transgenic line, we set out to study the role of overexpression of TGF-β1 in the development of the salivary glands. TGF-β1 regulates mucosal immunity and deletion of TGF-β signaling in the salivary gland appears to cause an autoimmune disorder that resembles Sjögren’s syndrome (12, 39). Previously, we conditionally deleted the TβRI signaling in mice using MMTV-Cre (40). While male mice were relatively unaffected by the conditional deletion of TβRI, the female mice gradually developed inflammation in the salivary glands. The lack of TβRI, however, did not appear to adversely disrupt salivary gland development in the conditional knockout mice. In contrast, the excess TGF-β signaling seen herein clearly inhibits the normal growth of the salivary gland in β1glo/MC mice. Over production of the active TGF-β1 ligand resulted in dysplastic growth of the submandibular glands and also disrupted branching in the glands. This likely led to the apparent increase in mesenchyme seen in the β1glo/MC mice. This phenotype is consistent with those reported in studies in which exogenous addition of TGF-β1 was able to inhibit branching morphogenesis in the salivary glands (11), lungs (41), and mammary glands (35, 42).

In terms of salivary gland development, the overexpression of active TGF-β1 in the β1glo/MC mice disrupted normal organogenesis by causing decreased cell proliferation and increased mesenchyme in the developing gland. Proper ECM deposition and degradation is necessary for normal salivary gland growth (43) and ECM remodeling is linked to cell proliferation within the gland (44). Both of these processes, cell proliferation and ECM deposition, are regulated by TGF-β signaling. Interestingly, some of the developmental growth defects in β1glo/MC mice are comparable to those seen during salivary gland development in the NOD (non-obese diabetic) mice, a mouse model for Sjogren’s syndrome (44). Furthermore, the changes in glandular homeostasis that occur in the NOD mice appear to mimic reported alterations in ECM turnover that are seen in patients with Sjogren’s syndrome. As a key regulator of ECM production, TGF-β activity may potentially have a role in these observed ECM alterations. In addition to the developmental defects, impaired AQP5 trafficking in β1glo/MC mice is likewise present in Sjogren’s syndrome patients (45), as well as in the NOD mouse model (46, 47). AQP5 is normally located in the luminal membranes of acinar cells of the submandibular gland and is essential for saliva secretion, but expression of AQP5 was mislocalized and reduced in the B1glo/MC mice. In this mouse model, at least, the effects of excess TGF-β signaling in the developing salivary gland probably leads to inhibited proliferation and dysplastic growth of the acinar cells, thereby indirectly resulting in a reduced and altered expression pattern for AQP5. This, in turn, may possibly contribute to the early salivary gland hypofunction seen in the β1glo/MC mice. At later times, much of the normal salivary gland parenchyma is replaced with fibrous tissue.

In the β1glo/MC mice, overexpression of TGF-β1 caused fibrosis of the salivary glands that was accompanied by atrophy of the GCDs and the acini. Overexpression of active TGF-β1 resulted in excessive deposition of ECM proteins that appeared to severely affect the ability of the β1glo/MC mice to secrete saliva. Except for the dilated ducts within the β1glo/MC salivary gland, most of the normal glandular parenchyma was replaced with fibrotic tissue and this most likely caused the lack of saliva secretion in the adult mice. However, TGF-β1 could have concurrently affected saliva secretion by functioning as a negative regulator of growth and by inhibiting the activity of the Na+/K+ ATPase (48). Key hallmarks of TGF-β induced fibrosis were seen in the salivary glands of the β1glo/MC mice. ECM proteins and other indicators of fibrosis like CTGF and smooth muscle actin were upregulated in response to the transgenic expression of TGF-β. While the acinar atrophy reduced the overall size of the salivary glands in our β1glo/MC mice, we were still able to study the effect of excess production of TGF-β1 on pathological glandular fibrosis. In humans, increased fibrosis of the salivary glands is seen in geriatric patients (14), although it is limited and does not affect salivation significantly among healthy elders (49).

The pathology seen in the β1glo/MC mice is similar to submandibular gland atrophy that can be caused by ligation of the salivary gland in animal models or in certain cases of salivary adenitis seen in patients (50, 51). Obstruction of the salivary gland in these cases often leads to fibrosis, degeneration of acinar cells, and dilation of the ducts. Additionally, many patients with Sjögren’s syndrome develop progressive fibrosis in their salivary glands (5254), although analysis of TGF-β production in Sjögren’s syndrome has yielded conflicting results (12, 40, 55, 56). Typically, fibrosis occurs when repeated injury, such as chronic inflammation, triggers the sustained production of TGF-β because of unresolved tissue damage (5760).

Radiotherapy for head and neck cancer can also cause fibrosis of the submandibular gland accompanied by a high density of small-dilated ducts (15) and TGF-β1 can be induced by radiation (16). Such extensive salivary gland fibrosis results in diminished saliva production, which leads to numerous morbidities in patients, including dysphagia, increased oral infections (e.g., Candidiasis, dental caries), as well as generalized oral discomfort. Radiation additionally causes a down-regulation of AQP5 expression in the salivary gland of rats and mice, with similar reductions in AQP5 staining to those we herein have noted in the β1glo/MC mice (37, 38). The β1glo/MC mice, therefore, seem to mimic the process of salivary gland fibrosis seen in pathological conditions such as radiation induced injury and may be a useful model to investigate early interventions to treat fibrosis.

In conclusion, changes in the expression levels of TGF-β can have a profound effect on the physiology of the salivary gland. Lack of TGF-β signaling in the salivary gland appears to trigger autoimmunity (12, 39, 40). In contrast, excess TGF-β resulted in the replacement of the normal salivary gland parenchyma with connective tissue. Therefore, a proper balance of TGF-β expression and signaling appears necessary for normal salivary gland homeostasis.


We would like to thank Drs. Matthew Hoffman, Wan-Jun Chen and Nancy Francis for critical reading of this manuscript, and Drs. Ana Cotrim and Robert Redman for helpful discussions. We would also like to thank Drs. John Letterio and Andrzej Dlugosz for the gift of active hemaglutanin epitope tagged TGF-β1 cDNA and pCLE vector, respectively. These studies were supported by the Division of Intramural Research, National Institute of Dental and Craniofacial Research, National Eye Institute, National Cancer Institute, Division of Veterinary Resources, The National Institutes of Health.


1. Ignotz RA, Massagué J. Transforming growth factor-beta stimulates the expression of fibronectin and collagen and their incorporation into the extracellular matrix. J Biol Chem. 1986;261:4337–4345. [PubMed]
2. Leask A, Abraham DJ. TGF-beta signaling and the fibrotic response. FASEB J. 2004;18:816–827. [PubMed]
3. Zavadil J, Böttinger EP. TGF-beta and epithelial-to-mesenchymal transitions. Oncogene. 2005;24:5764–5774. [PubMed]
4. Kalluri R, Neilson EG. Epithelial-mesenchymal transition and its implications for fibrosis. J Clin Invest. 2003;112:1776–1784. [PMC free article] [PubMed]
5. Lee KY, Bae SC. TGF-beta-dependent cell growth arrest and apoptosis. J Biochem Mol Biol. 2002;35:47–53. [PubMed]
6. Pepper MS. Transforming growth factor-beta: vasculogenesis, angiogenesis, and vessel wall integrity. Cytokine Growth Factor Rev. 1997;8:21–43. [PubMed]
7. Li MO, Wan YY, Sanjabi S, Robertson AK, Flavell RA. Transforming growth factor-beta regulation of immune responses. Annu Rev Immunol. 2006;24:99–146. [PubMed]
8. Jaskoll T, Melnick M. Submandibular gland morphogenesis: stage-specific expression of TGF-alpha/EGF, IGF, TGF-beta, TNF, and IL-6 signal transduction in normal embryonic mice and the phenotypic effects of TGF-beta2, TGF-beta3, and EGF-r null mutations. Anat Rec. 1999;256:252–268. [PubMed]
9. Lourenço SV, Uyekita SH, Lima DM, Soares FA. Developing human minor salivary glands: morphological parallel relation between the expression of TGF-beta isoforms and cytoskeletal markers of glandular maturation. Virchows Arch. 2008;452:427–434. [PubMed]
10. Lehnert SA, Akhurst RJ. Embryonic expression pattern of TGF beta type-1 RNA suggests both paracrine and autocrine mechanisms of action. Development. 1988;104:263–273. [PubMed]
11. Hardman P, Landels E, Woolf AS, Spooner BS. TGF-beta1 inhibits growth and branching morphogenesis in embryonic mouse submandibular and sublingual glands in vitro. Devel Growth Differ. 1994;36:567–577.
12. Kizu Y, Sakurai H, Katagiri S, Shinozaki N, Ono M, Tsubota K, Saito J. Immunohistological analysis of tumour growth factor beta 1 expression in normal and inflamed salivary glands. J Clin Pathol. 1996;49:728–732. [PMC free article] [PubMed]
13. Amano O, Tsuji T, Nakamura T, Iseki S. Expression of transforming growth factor beta 1 in the submandibular gland of the rat. J Histochem Cytochem. 1991;39:1707–1711. [PubMed]
14. Ficarra BJ. Submandibular salivary gland fibrosis. J Med. 1996;27:103–113. [PubMed]
15. Cooper JS, Fu K, Marks J, Silverman S. Late effects of radiation therapy in the head and neck region. Int J Radiat Oncol Biol Phys. 1995;31:1141–1164. [PubMed]
16. Martin M, Lefaix J, Delanian S. TGF-beta1 and radiation fibrosis: a master switch and a specific therapeutic target? Int J Radiat Oncol Biol Phys. 2000;47:277–290. [PubMed]
17. Wagner KU, McAllister K, Ward T, Davis B, Wiseman R, Hennighausen L. Spatial and temporal expression of the Cre gene under the control of the MMTV-LTR in different lines of transgenic mice. Transgenic Res. 2001;10:545–553. [PubMed]
18. Wolfraim LA, Alkemade GM, Alex B, Sharpe S, Parks WT, Letterio JJ. Development and application of fully functional epitope-tagged forms of transforming growth factor-beta. J Immunol Methods. 2002;266:7–18. [PubMed]
19. Bradley SV, Hyun TS, Oravecz-Wilson KI, Li L, Waldorff EI, Ermilov AN, Goldstein SA, Zhang CX, Drubin DG, Varela K, Parlow A, Dlugosz AA, Ross TS. Degenerative phenotypes caused by the combined deficiency of murine HIP1 and HIP1r are rescued by human HIP1. Hum Mol Genet. 2007;16:1279–1292. [PubMed]
20. Samuel SK, Hurta RA, Kondaiah P, Khalil N, Turley EA, Wright JA, Greenberg AH. Autocrine induction of tumor protease production and invasion by a metallothionein-regulated TGF-beta 1 (Ser223, 225) EMBO J. 1992;11:1599–1605. [PMC free article] [PubMed]
21. Sauer B, Henderson N. Targeted insertion of exogenous DNA into the eukaryotic genome by the Cre recombinase. New Biol. 1990;2:441–449. [PubMed]
22. Flanders KC, Thompson NL, Cissel DS, Van Obberghen-Schilling E, Baker CC, Kass ME, Ellingsworth LR, Roberts AB, Sporn MB. Transforming growth factor-beta 1: histochemical localization with antibodies to different epitopes. J Cell Biol. 1989;108:653–660. [PMC free article] [PubMed]
23. Wellner RB, Hong S, Cotrim AP, Swaim WD, Baum BJ. Modifying the NH2 and COOH termini of aquaporin-5: effects on localization in polarized epithelial cells. Tissue Eng. 2005;11:1449–1458. [PubMed]
24. Sauer B, Henderson N. Site-specific DNA recombination in mammalian cells by the Cre recombinase of bacteriophage P1. Proc Natl Acad Sci USA. 1988;85:5166–5170. [PMC free article] [PubMed]
25. Lakso M, Sauer B, Mosinger B, Jr, Lee EJ, Manning RW, Yu SH, Mulder KL, Westphal H. Targeted oncogene activation by site-specific recombination in transgenic mice. Proc Natl Acad Sci USA. 1992;89:6232–6236. [PMC free article] [PubMed]
26. Willimsky G, Blankenstein T. Sporadic immunogenic tumours avoid destruction by inducing T-cell tolerance. Nature. 2005;437:141–146. [PubMed]
27. Wagner KU, Wall RJ, St-Onge L, Gruss P, Wynshaw-Boris A, Garrett L, Li M, Furth PA, Hennighausen L. Cre-mediated gene deletion in the mammary gland. Nucleic Acids Res. 1997;25:4323–4330. [PMC free article] [PubMed]
28. Ewald D, Li M, Efrat S, Auer G, Wall RJ, Furth PA, Hennighausen L. Time-sensitive reversal of hyperplasia in transgenic mice expressing SV40 T antigen. Science. 1996;273:1384–1386. [PubMed]
29. Ma T, Song Y, Gillespie A, Carlson EJ, Epstein CJ, Verkman AS. Defective secretion of saliva in transgenic mice lacking aquaporin-5 water channels. J Biol Chem. 1999;274:20071–20074. [PubMed]
30. Krane CM, Melvin JE, Nguyen HV, Richardson L, Towne JE, Doetschman T, Menon AG. Salivary acinar cells from aquaporin 5-deficient mice have decreased membrane water permeability and altered cell volume regulation. J Biol Chem. 2001;276:23413–23420. [PubMed]
31. Delporte C, Steinfeld S. Distribution and roles of aquaporins in salivary glands. Biochim Biophys Acta. 2006;1758:1061–1070. [PubMed]
32. 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;433:260–268. [PubMed]
33. Qiao W, Li AG, Owens P, Xu X, Wang XJ, Deng CX. Hair follicle defects and squamous cell carcinoma formation in Smad4 conditional knockout mouse skin. Oncogene. 2006;25:207–217. [PubMed]
34. Sellheyer K, Bickenbach JR, Rothnagel JA, Bundman D, Longley MA, Krieg T, Roche NS, Roberts AB, Roop DR. Inhibition of skin development by overexpression of transforming growth factor beta 1 in the epidermis of transgenic mice. Proc Natl Acad Sci USA. 1993;90:5237–5241. [PMC free article] [PubMed]
35. Pierce DF, Jr, Johnson MD, Matsui Y, Robinson SD, Gold LI, Purchio AF, Daniel CW, Hogan BL, Moses HL. Inhibition of mammary duct development but not alveolar outgrowth during pregnancy in transgenic mice expressing active TGF-beta 1. Genes Dev. 1993;7:2308–2317. [PubMed]
36. Guarino M, Tosoni A, Nebuloni M. Direct contribution of epithelium to organ fibrosis: epithelial-mesenchymal transition. Hum Pathol. 2009;40:1365–1376. [PubMed]
37. Li Z, Zhao D, Gong B, Xu Y, Sun H, Yang B, Zhao X. Decreased saliva secretion and down-regulation of AQP5 in submandibular gland in irradiated rats. Radiat Res. 2006;165:678–687. [PubMed]
38. Takakura K, Takaki S, Takeda I, Hanaue N, Kizu Y, Tonogi M, Yamane GY. Effect of cevimeline on radiation-induced salivary gland dysfunction and AQP5 in submandibular gland in mice. Bull Tokyo Dent Coll. 2007;48:47–56. [PubMed]
39. McCartney-Francis NL, Mizel DE, Redman RS, Frazier-Jessen M, Panek RB, Kulkarni AB, Ward JM, McCarthy JB, Wahl SM. Autoimmune Sjögren's-like lesions in salivary glands of TGF-beta1-deficient mice are inhibited by adhesion-blocking peptides. J Immunol. 1996;157:1306–1312. [PubMed]
40. Nandula SR, Amarnath S, Molinolo A, Bandyopadhyay BC, Hall B, Goldsmith CM, Zheng C, Larsson J, Sreenath T, Chen W, Ambudkar IS, Karlsson S, Baum BJ, Kulkarni AB. Female mice are more susceptible to developing inflammatory disorders due to impaired transforming growth factor beta signaling in salivary glands. Arthritis Rheum. 2007;56:1798–1805. [PubMed]
41. Serra R, Pelton RW, Moses HL. TGF beta 1 inhibits branching morphogenesis and N-myc expression in lung bud organ cultures. Development. 1994;120:2153–2161. [PubMed]
42. Robinson SD, Silberstein GB, Roberts AB, Flanders KC, Daniel CW. Regulated expression and growth inhibitory effects of transforming growth factor-beta isoforms in mouse mammary gland development. Development. 1991;113:867–878. [PubMed]
43. Hardman P, Spooner BS. Localization of extracellular matrix components in developing mouse salivary glands by confocal microscopy. Anat Rec. 1992;234:452–459. [PubMed]
44. Cha S, van Blockland SC, Versnel MA, Homo-Delarche F, Nagashima H, Brayer J, Peck AB, Humphreys-Beher MG. Abnormal organogenesis in salivary gland development may initiate adult onset of autoimmune exocrinopathy. Exp Clin Immunogenet. 2001;18:143–160. [PubMed]
45. 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;81:143–148. [PubMed]
46. Konttinen YT, Tensing EK, Laine M, Porola P, Törnwall J, Hukkanen M. Abnormal distribution of aquaporin-5 in salivary glands in the NOD mouse model for Sjögren's syndrome. J Rheumatol. 2005;32:1071–1075. [PubMed]
47. Soyfoo MS, De Vriese C, Debaix H, Martin-Martinez MD, Mathieu C, Devuyst O, Steinfeld SD, Delporte C. Modified aquaporin 5 expression and distribution in submandibular glands from NOD mice displaying autoimmune exocrinopathy. Arthritis Rheum. 2007;56:2566–2574. [PubMed]
48. Streckfus C, Bigler L, O'Bryan T. Aging and salivary cytokine concentrations as predictors of whole saliva flow rates among women: a preliminary study. Gerontology. 2002;48:282–288. [PubMed]
49. Baum BJ, Ship JA, Wu AJ. Salivary gland function and aging: a model for studying the interaction of aging and systemic disease. Crit Rev Oral Biol Med. 1992;4:53–64. [PubMed]
50. Bhaskar SN, Bolden TE, Weinmann JP. Experimental obstructive adenitis in the mouse. J Dent Res. 1956;35:852–862. [PubMed]
51. Teymoortash A, Tiemann M, Schrader C, Hartmann O, Werner JA. Transforming growth factor beta in chronic obstructive sialadenitis of human submandibular gland. Arch Oral Biol. 2003;48:111–116. [PubMed]
52. Koski H, Konttinen YT, Gu XH, Hietanen J, Malmström M. Transforming growth factor beta 2 in labial salivary glands in Sjögren's syndrome. Ann Rheum Dis. 1995;54:744–747. [PMC free article] [PubMed]
53. Skopouli FN, Li L, Boumba D, Stefanaki S, Hanel K, Moutsopoulos HM, Krilis SA. Association of mast cells with fibrosis and fatty infiltration in the minor salivary glands of patients with Sjögren's syndrome. Clin Exp Rheumatol. 1998;16:63–65. [PubMed]
54. Koski H, Janin A, Humphreys-Beher MG, Sorsa T, Malmström M, Konttinen YT. Tumor necrosis factor-alpha and receptors for it in labial salivary glands in Sjögren's syndrome. Clin Exp Rheumatol. 2001;19:131–137. [PubMed]
55. Cauli A, Yanni G, Pitzalis C, Challacombe S, Panayi GS. Cytokine and adhesion molecule expression in the minor salivary glands of patients with Sjögren's syndrome and chronic sialoadenitis. Ann Rheum Dis. 1995;54:209–215. [PMC free article] [PubMed]
56. Mason GI, Hamburger J, Bowman S, Matthews JB. Salivary gland expression of transforming growth factor beta isoforms in Sjogren's syndrome and benign lymphoepithelial lesions. Mol Pathol. 2003;56:52–59. [PMC free article] [PubMed]
57. Border WA, Noble NA. Transforming growth factor beta in tissue fibrosis. N Engl J Med. 1994;331:1286–1292. [PubMed]
58. Border WA, Ruoslahti E. Transforming growth factor-beta in disease: the dark side of tissue repair. J Clin Invest. 1992;90:1–7. [PMC free article] [PubMed]
59. Letterio JJ, Roberts AB. Regulation of immune responses by TGF-beta. Annu Rev Immunol. 1998;16:137–161. [PubMed]
60. Wahl SM. Transforming growth factor-beta: innately bipolar. Curr Opin Immunol. 2007;19:55–62. [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...