Introduction

The development of optimally designed vectors is critical for achieving safe, efficient, and successful gene therapy and can facilitate therapeutic outcomes^1,2,3. For example, higher cell- or tissue-specific promoter activity could result in decreased doses of vector administered, with consequently lower toxicity or side effects from the vector. Additionally, cell- or tissue-specific promoters may provide a useful way to restrict unwanted transgene expression as well as facilitate persistent transgene expression^3,4,5,6.

Over the past decade, studies from our laboratory have established several potential clinical applications of gene transfer to salivary glands^{7,8,9,10,11,12}. Salivary glands consist of two general types of epithelial cells, acinar and ductal, in primarily a densely packed monolayer¹³. Vector delivery is performed through cannulation of the main excretory ducts whose orifices are accessible directly in the mouth. Based on results in other tissues, such as the liver³, it is critically important to identify components of the expression cassette that will lead to safe and effective gene transfer in salivary glands.

To optimize the expression cassette for salivary gland gene transfer, we have initially evaluated the strength and specificity of promoter constructs in the rat submandibular A5 cell line in vitro and in rat submandibular glands in vivo. Viral promoters chosen were cytomegalovirus (CMV; human immediate early), Rous sarcoma virus (RSV), simian virus 40 (SV40), and the long terminal repeat from Moloney murine leukemia virus (LTR). Five human promoters were chosen: the elongation factor (EF) 1 promoter, widely useful in mammalian cells^14,15; the cytokeratin 18 (K18) promoter, predominantly active in epithelia of internal organs, e.g., lung, liver, kidney, and intestine^16,17; the cytokeratin 19 (K19) promoter, useful in many epithelial cells^18,19; the tissue kallikrein promoter (Kall), active in the liver, kidney, and salivary glands among many tissues^{20,21,22,23,24}; and the amylase promoter (AMY), relatively specific for salivary glands^25,26. The rat aquaporin-5 (rAQP5) promoter is selectively active in the distal lung, salivary glands, lacrimal glands, and corneal epithelium^27,28. In rat salivary glands, AQP5 is expressed mainly in acinar and intercalated duct cells^29,30.

Also, several modifications of the K18, K19, and rAQP5 promoters were tested. For example, the first intron of the K18 gene is an enhancer element³¹ and includes binding sites for the transcription factors AP-1 and ETS, which are associated with induced K18 expression^32,33. Thus, we tested the K18 promoter the first intron¹⁷. Gut-enriched Krüppel-like factor and Sp1 modulate K19 promoter activity and contribute to tissue specificity¹⁸. Important regulatory elements are located from -2249 to -2050 bp and from -732 to -1 bp in this promoter¹⁹. Thus, we tested three fragments of the K19 promoter (0.7-, 2.2-, and 3.0-kb). The rAQP5 promoter²⁸ displays common enhancers (-1716 to -1638 and -224 to -139) and a common repressor (-503 to -385), in lung (MLE-15) and salivary (Pa-4) cell lines. Accordingly, we tested three modifications of the rAQP5 promoter (1.7-, 0.9-, and 0.4-kb).

All promoters initially were placed into a common plasmid, pAC-luc^11,13 and adenoviral vectors generated to evaluate promoter activity in vitro and in vivo using luciferase as a reporter gene. Thereafter, promoters were selected to drive the enhanced green fluorescence protein (EGFP) reporter gene to localize the site of transgenic protein expression.

Results and discussion

Comparison of promoter activities following plasmid transfection in Vitro and in Vivo

In this study, we used pAC-luc (see Supplementary Fig. 1) as a backbone expression plasmid into which different promoters were inserted. In total, we constructed 16 plasmids, each containing a different promoter element, but with the same luciferase transgene and SV40 polyadenylation sequence (see Supplementary Table 1, and Materials and Methods, for more details). We selected male Wistar rats for in vivo studies because of our previous experience with this animal model^7,8,9,13. We used the A5 cell line as an in vitro cell model because it is derived from rat submandibular gland³⁴. Initially, we transfected all 16 plasmids separately into A5 cells at 5 10⁴ plasmid molecules/cell using plasmid, polyethylenimine (PEI), and Adcontrol complexes to compare promoter activities in vitro. We measured luciferase activity 24 h posttransfection (Fig. 1A). The CMV promoter showed the highest activity, while K19 0.7-kb promoter had the least activity, in A5 cells. The human EF1 promoter exhibited the highest activity of all mammalian promoters, only about threefold lower than that seen with the CMV promoter. The two K18 promoters tested possessed similar activity, while the K19 3.0-kb construct showed the highest activity of the three K19 promoters tested. The rAQP5 0.4- and 0.9-kb constructs exhibited the highest activity of four different rAQP5 promoters tested. The Kall promoter showed activity similar to that of the K18 promoters, while the AMY promoter was the weakest of all mammalian promoter constructs, excepting K19 0.7-kb.

Figure 1.

Comparison of promoter activities in vitro and in vivo after transfection using plasmid–PEI–Adcontrol complexes. (A) Promoter activities in vitro in A5 cells. Each cell was transfected with 5 10⁴ plasmid molecules. See text for details. The data shown are the means SD of luciferase activity from three determinations 24 h after transfection and are representative of two separate experiments. For some determinations the error bars are too small to be visualized. (B) Promoter activities in vivo in rat submandibular glands. Each rat submandibular gland was transfected with 4.35 10¹² plasmid molecules. See text for details. The data shown are the means SD of luciferase activity from six submandibular glands of three rats 24 h after transfections. The absence of activity with the K19-0.7 promoter was verified by an additional experiment.

Full figure and legend (109K)

Next, we transfected all 16 plasmids into rat submandibular glands in vivo at 4.35 10¹² molecules/gland using plasmid–PEI–Adcontrol complexes. We measured luciferase activity at 3 days posttransfection (Fig. 1B). For mammalian promoters, the rank ordering of the luciferase activity profile obtained in vivo was essentially the same compared to the in vitro results. However, the luciferase activity profile seen with viral promoters was notably different in vivo. The highest promoter activity in vivo was found with the RSV construct, 10-fold higher than found with the CMV promoter. The activities of EF1 and K19 3.0-kb promoters in vivo were similar in these transfection experiments, 2- to 3-fold less than observed with the RSV promoter. The K19 0.7-kb promoter showed essentially no activity in vivo in rat submandibular glands.

The results of these plasmid transfection experiments were somewhat different from what has been reported with other cell lines and tissue targets. For example, a previous study showed that cytokeratin 18 enhancer/promoter plus intron 1 (K18-3.3) resulted in the highest level of transgene expression of several constructs tested in a human lung epithelial cell line¹⁷. In the current study in salivary epithelial cells in vivo, we found that cytokeratin 18 enhancer/promoter without intron 1 (K18-2.5) led to higher transgene expression. Similarly, earlier reports examining the cytokeratin 19 promoter demonstrated that a 2249-bp element showed the highest activity in a human cholangioma-derived cell line, KMBC¹⁹. However, in salivary epithelial cells in vitro and in vivo our results demonstrate that a 2952-bp fragment of this promoter led to the highest level of transgene expression. Additionally, studies with the rat AQP5 promoter in a rat parotid cell line, Pa-4²⁸, indicated that a 1716-bp element resulted in the highest transgene expression of several constructs tested, while the second highest activity was found in a 385-bp fragment. Our results show that a 391-bp fragment of the rat AQP5 promoter led to the highest transgene expression in both A5 cells and rat submandibular gland. Such findings are not surprising and indicate that the cell or tissue model used will affect the level of promoter activity measured.

Comparison of promoter activities following adenoviral infection in vitro and in vivo

Based on the results of plasmid transfection experiments, we selected eight promoter constructs for further study using first-generation E1^- adenoviral vectors. The promoter constructs selected for these studies were CMV, RSV, EF1, K18-2.5, K19-3.0, rAQP5-0.4, Kall, and AMY (see Table 1). After construction, we used the corresponding vectors to infect A5 cells in vitro, and rat submandibular glands in vivo, to evaluate promoter activity.

Table 1 - Adenoviral vectors constructed for this study.

Full table

We infected A5 cells at 100 particles/cell and measured luciferase activity 1, 2, and 3 days postinfection (Fig. 2). The highest luciferase activities were seen on day 3. The expression pattern observed for these promoters, at all time-points, was essentially the same as that seen in A5 cells with transfection experiments, except for K19-3.0 promoter, which displayed lower activity. The CMV promoter exhibited the highest activity, while the AMY promoter showed the lowest activity. Interestingly, two mammalian cell promoters, EF1 and rAQP5, resulted in relatively high levels of luciferase expression in A5 cells. The overall statistically determined rank order of promoter strength in vitro in A5 cells was as follows: CMV > RSV = EF1 > rAQP5-0.4 > K18-2.5 > Kall > K19-3.0 > AMY (Fig. 2).

Figure 2.

Comparison of promoter activities from eight adenoviral vectors in A5 cells. All adenoviral vectors contained the luciferase cDNA as a reporter gene. Each cell was infected with 100 vector particles. See text for details. The data shown are the means SD of luciferase activity from three determinations, 1, 2, or 3 days postinfection, and are representative of two separate experiments. For some determinations the error bars are too small to be visualized. Differences between results (promoter-driven luciferase activity) obtained at individual time points were determined, as appropriate, with a one-way ANOVA or Kruskal–Wallis one-way ANOVA on ranks. Thereafter, determinations of the rank order of promoter activities used the day 3 data and a Kruskal–Wallis one-way ANOVA on ranks, followed by a multiple pair-wise comparison with the Student–Newman–Keuls method, to assess statistical significance (see text for additional details).

Full figure and legend (144K)

Additionally, we examined whether dexamethasone would affect adenoviral vector-mediated luciferase expression driven by these promoters in A5 cells. We did this because for in vivo delivery of adenoviral vectors to rat salivary glands we administer dexamethasone (1 mg/rat) to blunt vector-induced immune responses^13,35. As can be seen in Supplementary Table 2, in vitro in A5 cells, at the three dexamethasone doses tested, which approximate the dose used in vivo, we saw little to no effect on activity with the CMV, rAQP5, and Kall promoters. However, dexamethasone led to substantial increases in the activity of the K18-2.5, RSV, and EF1 promoters; a modest elevation in the K19-3.0 promoter activity; and a significant inhibition of AMY promoter activity. We are unaware of any previous reports demonstrating that dexamethasone affects the activities of the K18-2.5, K19-3.0, and EF1 promoters. However, this glucocorticoid has been shown to increase markedly the activity of the RSV promoter in vivo in rat liver and in vitro in rat hepatocytes³⁶. While dexamethasone reportedly increases salivary amylase expression in mouse hepatoma cells in vitro³⁷, the specific amylase promoter used here (AMY1C^25,26,38,39) does not contain a glucocorticoid consensus sequence³⁷. Although the general rank order of promoter activity was similar dexamethasone, clearly this immunosuppressive treatment can affect the absolute activities of several promoters tested in vitro and may have done so in vivo in addition to blunting immune responsiveness to adenoviral vectors.

We next use these same eight adenoviral vectors to infect rat submandibular glands (at 10⁹ particles/gland) in vivo. This dose is similar (about threefold less; 100:1 particles to plaque-forming unit) to the dose of this serotype adenoviral vector shown to lead to proportional (scaled) transgene expression in mice and miniature pigs after salivary gland delivery¹¹. We measured luciferase activity at 2, 3, and 7 days after submandibular gland infection (Fig. 3). Luciferase expression was maximal on day 3 and decreased considerably by day 7 with all promoter constructs. As depicted in Fig. 3, on day 3 the statistically determined rank order of promoter strength in vivo in rat submandibular glands was as follows: CMV > RSV = EF1 > K18-2.5 > rAQP5-0.4 = Kall = K19-3.0 > AMY (Fig. 3). This rank order was fairly similar to that seen for these adenoviral vectors with A5 cells in vitro.

Figure 3.

Comparison of promoter activities from eight adenoviral vectors in rat submandibular glands. All adenoviral vectors contained the luciferase cDNA as a reporter gene. Each rat submandibular gland was infected with 10⁹ particles. See text for details. The data shown are the means SD of luciferase activity from six submandibular glands of three rats at 2, 3, or 7 days postinfection. Differences between results (promoter-driven luciferase activity) obtained at individual time points were determined, as appropriate, with a one-way ANOVA or Kruskal–Wallis one-way ANOVA on ranks. Thereafter, determinations of the rank order of promoter activities used the day 3 data and a Kruskal–Wallis one-way ANOVA on ranks, followed by a multiple pair-wise comparison with the Student–Newman–Keuls method, to assess statistical significance (see text for additional details).

Full figure and legend (141K)

Not surprisingly, plasmid transfections required considerably more molecules ( 500 times) than needed with adenoviral vector infection in A5 cells in vitro to achieve comparable levels of luciferase activity (see Figs. 1 and 2). Similarly, with in vivo experiments in submandibular gland plasmid transfections required 4350 times more molecules than adenoviral vector infection to achieve comparable levels of luciferase activity (see Figs. 1 and 3). Interestingly, it appears that adenoviral genomic sequences had relatively little effect on the activities of most individual promoters except possibly for the CMV and K19-3.0 promoters in rat submandibular glands, as the pattern of luciferase activity measured with adenoviral infection was generally the same as that resulting from plasmid transfection experiments (see Figs. 1 and 3). Overall, this indicates that plasmid transfections may frequently predict the general results from adenoviral vector infections in salivary glands.

Localization of promoter activities following adenoviral infection in vitro and in vivo

To localize cellular protein expression directly, we used EGFP cDNA as a reporter gene in adenoviral constructs (see Table 1). We tested the eight adenoviral vectors generated by infection of A5 cells at 100 particles/cell in vitro. After 48 h, infected A5 cells were observed by fluorescence microscopy. The order of promoter activity observed was follows: CMV > RSV = EF1 > K18-2.5 = K19-3.0 > rAQP5-0.4 > Kall > AMY (Fig. 4A). These results demonstrate that the pattern of EGFP expression driven by the eight promoters examined is generally similar to the pattern of activity observed in the luciferase-encoding vectors in A5 cells, with the notable exception of the rAQP5 promoter (Fig. 2). When we used luciferase as a transgene, the rAQP5 promoter was significantly stronger than the K18-2.5 and K19-3.0 promoters (P < 0.001 and P = 0.012, respectively).

Figure 4.

Comparison of promoter activities in A5 cells and rat submandibular glands using adenoviral vectors encoding EGFP as a reporter gene. (A) Results with A5 cells. Fluorescence microscopy was performed 48 h after infection. Each cell was infected with 100 vector particles. This experiment was repeated three separate times with similar results. Pictures in the top row were taken at a low sensitivity setting (Max 200). Pictures in the middle row were taken at an intermediate sensitivity setting (Max 12). Pictures in the bottom row were taken at a higher sensitivity setting (Max 2). (B, C) Results with rat submandibular glands. Rat submandibular glands were infected with 10⁹ vector particles/gland using one of the following promoters: CMV, RSV, EF1, K18-2.5, or K19-3.0. For experiments evaluating the rAQP5-0.4, Kall, and AMY promoters, 10¹⁰ particles/gland were delivered. (B) Hematoxylin and eosin staining of tissue samples from a control gland (no viral infection) and a gland following infection with AdAMY-EGFP. (C) Immunohistochemical staining using an antibody directed at EGFP. Results are representative of four submandibular glands from two rats, 3 days after vector administration, with three sections from each gland examined. In individual images, "a" indicates acinar cells and "d" indicates ductal cells.

Full figure and legend (396K)

There are two general epithelial cell types, acinar and ductal, present in salivary glands. Acinar cells form the secretory endpiece and are at the distal part of the glands. To determine if any of the promoters tested possessed cell-type specificity, we administered the eight adenoviral vectors encoding EGFP into rat submandibular glands at either 10⁹ particles/gland for the CMV, EF1, RSV, K18-2.5, and K19-3.0 promoters or, for the rAQP5-0.4, Kall, and AMY promoters, 10¹⁰ particles/gland, because these promoters were relatively weak based on in vitro studies (Fig. 4A). Three days after vector administration, we observed EGFP fluorescence directly by confocal microscopy (not shown). We observed a pattern of total promoter activity similar to that seen in vitro; however, it was not possible to determine unequivocally the cellular site, ductal or acinar, by confocal microscopy. To determine accurately the cell types transduced by vectors with different promoters, we performed immunohistochemical staining (Figs. 4B and 4C). As is clear from these results, 3 days after vector delivery most promoters studied were active in both acinar and ductal cells, with generally higher activity seen in ductal cells. In adult male rats there are approximately equal numbers of acinar and ductal cells, unlike the situation in humans, in which there are approximately fourfold more acinar cells than ductal cells. Serotype 5 adenoviral vectors can efficiently infect both acinar and ductal cells^7,13. Thus, adenoviral vectors would have greater exposure to ductal cells in rats than in humans.

As can be seen in Fig. 4B, hematoxylin and eosin staining of paraffin-embedded sections from control (no viral infection) submandibular glands and glands treated with 10¹⁰ particles of AdAMY-EGFP were comparable in morphological appearance, with acinar and ductal cells readily distinguished. After immunohistochemical staining, it is clear (Fig. 4C) that the CMV, RSV, EF1, K18-2.5, and K19-3.0 promoters direct EGFP expression in both salivary cell types, i.e., transgenic protein was expressed in both ductal and acinar cells, albeit with somewhat more expression in ductal cells. Based on the physiological expression of AQP5 in rat salivary glands^29,30 we hypothesized that this promoter would show activity mainly in acinar cells and intercalated duct cells (a minor duct cell compartment adjacent to acinar cells). However, it is also clear from results in Fig. 4C that the rAQP5 promoter exhibited activity in many duct cell types. This may be a result of using a markedly smaller rAQP5 promoter (0.4 kb vs 1.7 or 4.5 kb). The construct used herein to drive EGFP expression represented only 391 bp of the known 4.5-kb 5' flanking region of rat AQP5, and most regulatory elements for the rAQP5 promoter were deleted from this region to increase promoter activity (Fig. 1).

Previously, we demonstrated that the AMY promoter retained relative cell-type specificity in rat submandibular glands, i.e., mainly in acinar cells^38,39. The Kall promoter was effective in salivary glands, but also could drive transgene expression in many tissues, particularly other water-transporting epithelia³⁸. However, in salivary cells it showed relative duct-cell specificity. In this earlier in vivo study, we administered adenoviral vectors encoding luciferase driven by the CMV, Kall, or AMY promoter³⁸, and we administered considerably more vector, 10¹¹ particles/gland. Results herein (Fig. 4C) showed that the activity of the Kall promoter was mainly in ductal cells with limited activity in acinar cells, while the activity of AMY promoter was detected in both acinar and ductal cells. It seems likely that at the higher dose used more adenoviral vector particles were able to reach acinar cells, which are located in the terminal secretory endpieces. The AdKall-luc construct used in this study also showed significantly lower activity, relative to AdCMV-luc, than the AdKall-luc used in the previous study³⁸. However, the AdKall-luc vector used herein is different from the vector employed earlier. The present vector, to be consistent with other constructs used herein, contains a 307-bp SV40 poly(A) sequence downstream of the luciferase coding sequence, while the vector used in our previous study included an 851-bp SV40 small antigen intron plus the SV40 poly(A) sequence downstream of the luciferase coding sequence. These differences may, at least in part, explain the different activities of the Kall promoter in the two studies.

The relatively epithelial-cell-specific promoters tested, i.e., K18-2.5, K19-3.0, rAQP5-0.4, Kall, and AMY, were much weaker in vivo than the CMV, RSV, and EF1 promoters. Using exogenous enhancer elements, it may be possible to increase the strength of one or more of these promoters yet retain a relative preference for salivary glands. The AMY, Kall, K18-2.5, and K19-3.0 promoters are derived from human genes, while the rAQP5-0.4 promoter used was from the rat gene. Little is known about the human AQP5 promoter at present. An important finding of the current study was that the human EF1 promoter used was essentially as powerful in driving transgene expression in vivo in salivary glands as the RSV promoter, with both leading to transgene expression levels 20–30% of those seen with the CMV promoter. It seems reasonable to suggest that for applications of salivary gland gene transfer requiring high levels of transgene expression the EF1 promoter could be useful.

In conclusion, in the present study we have evaluated the function of 16 promoter constructs in vitro and in vivo in rat submandibular epithelial cells. We defined a hierarchy of promoter strength and specificity in salivary glands. These results should lead to improved transgene cassette design, and safety, for gene transfer applications to salivary glands.

Materials and methods

Construction of plasmids and recombinant adenoviral vectors

pAC-luc (^11,38; Supplementary Fig. 1) was constructed using pACCMV-pLpA, a gift from Dr. C. Newgard (University of Texas, Southwestern Medical Center, Austin, TX, USA). pACCMV-pLpA was digested with NotI. A 307-bp DNA fragment containing a multiple cloning site (BglII, XhoI, HindIII, EcoRI, SalI, KpnI, BamHI, and XbaI) and the SV40 poly(A) sequence was removed from pEGFP-C3 (BD Biosciences, Palo Alto, CA, USA) with MluI and ScaI, and filled in with Klenow large DNA fragment (Invitrogen, Carlsbad, CA, USA). NotI linkers were added and then ligated to generate pAC. A 1851-bp luciferase cDNA (KpnI/BamHI fragment from pGL2-Basic; Promega, Madison, WI, USA) was inserted to yield pAC-luc. This plasmid was used as a shuttle vector for generation of adenoviral vectors. Different promoters were inserted into pAC-luc as shown in Supplementary Table 1. The human CMV immediate early promoter/enhancer (589 bp) was excised from pEGFP-C3 with AseI/EcoRI. The SV40 promoter (360 bp) was excised from pEGFP-C3 with StuI/SspI. The RSV promoter (412 bp) was excised from pAAVRnLacZ, a gift from Dr. J. Chiorini (NIDCR, NIH), with BglII/HindIII. The LTR promoter (608 bp) was PCR-amplified from the plasmid pXT1 with primers 5'-CCCTCGAGAAGAACAGATGGTCCCCAGATGCG-3' and 5'-CTCGTCGACTTTTTGGACTCAGGTCGGGC-3'. The EF1 promoter (1272 bp), which includes 87 bp of the SV40 small antigen sequence at its 5' end, was excised from pEF-BOS, a gift from Dr. S. Gutkind (NIDCR, NIH), with EcoRI/HindIII. Two K18 promoters were used: a 2565-bp element with the K18 enhancer plus promoter (K18-2.5) and a 3354-bp element with the K18 enhancer, promoter, plus intron 1 (K18-3.3). The K18 promoters were excised from K18EpilongSEAP, a gift from Dr. J. Hu (University of Toronto, ON, Canada), with either KpnI/MluI or KpnI/NcoI, respectively. The K19 promoter (2952 bp; K19-3.0) was excised from pHCK-2952, a gift from Dr. M. Kagaya (Kanazawa University, Japan) using XhoI/HindIII. Four fragments of the rat AQP5 promoter were used: 4500 (rAQP5-4.5), 1791 (rAQP5-1.7), 900 (rAQP5-0.9), and 391 bp (rAQP5-0.4). These were excised from pUC-AQP5-2, a gift from Dr. D. K. Ann (University of Southern California, Los Angeles, CA, USA), with EcoRI/HindIII, FokI/HindIII, Sau3AI/HindIII, and DelI/HindIII, respectively. Each was filled in using Klenow large DNA fragment, HindIII linkers were added, and then each was placed into the HindIII site of pAC-luc. Additionally, two small fragments of the K19 promoter [2244 (K19-2.2) and 705 bp (K19-0.7)] were PCR-amplified with primers K19F60X (5'-CCGTCGACCTCACGAATGGAACTGTTGATTCC-3')/K19B255LS (5'-ATAGTCGACGGCGAGGCGGAGCACGGACGGAGCAACCCTGGTCTCAGAAGCTGCGATTC-3') and K19F192X (5'-CCGTCGACAATCCCAGCACTTTGGACGG-3')/K19B255LS, separately, and ligated into the SalI site of pAC-luc. As reported earlier^38,39, the 1002-bp AMY promoter was amplified from plasmid pH3GH, a gift from Dr. M. Meisler (University of Michigan, Ann Arbor, MI, USA) and then ligated into the SalI site of pAC-luc. The 315-bp Kall promoter was amplified from plasmid pKallikrein, a gift from Dr. L. Chao (Medical University of South Carolina, Charleston, SC, USA), with the 5' primer 5'-TTCTCGAGTTGCCTCACTG-3' and 3' primer 5'-CTCGTCGACGGTGACAGAGGT-3' and ligated into the SalI site of pAC-luc.

After initial studies with these 16 plasmids (Supplementary Table 1), 8 were chosen for detailed testing: CMV, RSV, EF1, K18-2.5, K19-3.0, rAQP5-0.4, Kall, and AMY. To localize protein expression directly, 795 bp of EGFP cDNA was excised from pEGFP-C3 with NheI/BamHI and filled in with Klenow large DNA fragment, EcoRI linkers were added at the NheI site, and this was then ligated into the EcoRI/BamHI sites of all pAC-luc constructs to replace the reporter cDNA. We made E1^- adenoviral vectors corresponding to each of the 8 promoters for both reporter genes (luciferase and EGFP; Table 1). Adenoviral vectors were generated by homologous recombination with pJM17 (Microbix Biosystems, Inc., Toronto, ON, Canada) and subsequently amplified in C7 cells (a gift from Dr. J. Chamberlain⁴⁰) from 20 150-mm plates and then purified by CsCl gradient centrifugation as described⁴¹. Purified vectors were titered by quantitative PCR (QPCR) with primers from the E2 region of adenovirus, E2q1 (5'-GCAGAACCACCAGCACAGTGT-3') and E2q2 (5'-TCCACGCATTTCCTTCTAAGCTA-3') and titers were expressed as particles/ml (Table 1). pACCMV-luc was used as a standard for QPCR; 1 g of pACCMV-luc DNA (10,432 bp) was equivalent to 8.7 10¹⁰ molecules. Standard curves used 10² to 10⁹ pACCMV-luc molecules, and adenoviral vectors were tested at three dilutions over a 100-fold range. QPCR assays used SYBR Green PCR Master Mix and an ABI Prism 7700 sequence detector (Applied Biosystems, Foster City, CA, USA) as follows: stage 1, 95°C for 2 min; stage 2, 95°C for 10 min; stage 3, 95°C for 15 s, 60°C for 1 min, repeated 40 times.

Plasmid delivery method

We used adenovirus–PEI–plasmid complexes to deliver plasmids in vitro and in vivo^42,43. The adenovirus was Adcontrol, generated by cotransfection of pACCMV-pLpA and pJM17 in C7 cells. For in vitro studies, the complexes were formed using 2 10¹⁰ molecules of plasmid DNA/4 10⁵ cells, 0.1 mM PEI, and 3 10¹⁰ particles of Adcontrol in a volume of 20 l. For in vivo studies, complexes were formed using 4.35 10¹² molecules of plasmid DNA/gland, 0.5 mM PEI, and 1 10¹¹ particles of Adcontrol in a volume of 150 l.

Cell culture

A5 cells³⁴ were grown in McCoy's 5A medium plus 10% bovine serum, 100 U/ml penicillin G, 100 g/ml streptomycin (all from BioSource, Camarillo, CA, USA) at 37°C in a humidified 5% CO₂ atmosphere incubator. C7 cells were grown in DMEM with high glucose plus 10% bovine serum, 50 g/ml hygromycin (Invitrogen), 100 U/ml penicillin G, 100 g/ml streptomycin at 37°C in humidified 5% CO₂.

In vitro and in vivo plasmid transfection and virus infection procedures

A5 cells (2 10⁵ cells/well; 96-well plates) were transfected (1 h) with 20 l of adenovirus–PEI–plasmid complexes, and 180 l of fresh growth medium was added. After 24 h luciferase activity was measured. For infections, A5 cells in suspension were infected for 1 h (100 particles/cell) and incubated at 5 10⁵ cells/200 l/well in 96-well plates for 24 h, and luciferase activity was measured. For dexamethasone studies, three doses (62.5, 125, and 250 g/ml) were added to A5 cells in 96-well plates after infection with adenoviral vectors (1 h). These concentrations were based on a potential in vivo dose of 125 g/ml (1 mg dexamethasone/rat, with an estimated blood volume of 8 ml³⁵).

Male Wistar rats (250–350 g, 3 months of age) were anesthetized^7,8,9 and plasmid complexes, or recombinant adenoviral vectors (10⁹ or 10¹⁰ particles), administered by retrograde ductal instillation¹³ into both submandibular glands (n = 3/group). Dexamethasone (im injection) was given at the time of adenoviral vector delivery and each day until tissues were harvested^13,35. After 2, 3, or 7 days, rats were euthanized and glands collected for luciferase assays. For EGFP studies, submandibular glands were collected at days 3 (Fig. 4) and 7 after infection. All animal studies were approved by the NIDCR Animal Care and Use Committee.

Luciferase and protein assays

Luciferase assays were performed as described^38,39. Protein concentrations were measured using the BCA protein assay kit (Pierce, Rockford, IL, USA). Results were expressed as relative light units (RLU)/mg protein.

Data analysis

Data analyses employed SigmaStat version 2.0 (SPSS, Inc, Chicago, IL, USA). Descriptive statistics are reported as means SD. Differences (promoter-driven luciferase activity) obtained after vector administration were determined, as appropriate, with a one-way ANOVA or Kruskal–Wallis one-way ANOVA on ranks. Studies of adenoviral vector infection of cells and glands used day 3 results (highest) and a Kruskal–Wallis one-way ANOVA on ranks and then a multiple pair-wise comparison (Student–Newman–Keuls). For studies evaluating dexamethasone effects (Supplementary Table 2), we used a one-way ANOVA and then a multiple pair-wise comparison with Bonferroni adjustment.

Fluorescence microscopy

A5 cells infected with EGFP encoding adenoviral vectors were observed directly by fluorescence microscopy or fixed with 2% paraformaldehyde in Dulbecco's phosphate-buffered saline for 20 min. Both gave similar results.

Immunohistochemistry

Submandibular glands were cut in half, fixed in 10% formalin (24 h), embedded in paraffin, sectioned (HistoServ, Germantown, MD, USA), and stained with anti-GFP antibody (Abcam, Inc., Cambridge, MA, USA) and the rabbit ImmunoCruz Staining System, sc-2051 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) plus streptavidin and biotin block. Sections were also stained with hematoxylin/eosin.

References

1. Benihoud, K., Yeh, P. and Perricaudet, M. (1999). Adenovirus vectors for gene delivery. Curr. Opin. Biotechnol. 10: 440–447. | Article | PubMed | ISI | ChemPort |
2. Orphanides, G. and Reinberg, D. (2002). A unified theory of gene expression. Cell 108: 439–451. | Article | PubMed | ISI | ChemPort |
3. Linthout, S. V., Collen, D. and Geest, B. D. (2002). Effect of promoters and enhancers on expression, transgene DNA persistence, and hepatotoxicity after adenoviral gene transfer of human apolipoprotein A-I. Hum. Gene Ther. 13: 829–840.
4. Nettelbeck, D. M., Jerome, V. and Muller, R. (2000). Gene therapy: designer promoters for tumour targeting. Trends Genet. 16: 174–181. | Article | PubMed | ISI | ChemPort |
5. Ostrowski, L. E., Hutchins, J. R., Zakel, K. and O'Neal, W. K. (2003). Targeting expression of a transgene to the airway surface epithelium using a ciliated cell-specific promoter. Mol. Ther. 8: 637–645. | Article | PubMed | ISI | ChemPort |
6. Xu, Z. L., Mizuguchi, H., Ishii-Watabe, A., Uchida, E., Mayumi, T. and Hayakawa, T. (2002). Strength evaluation of transcriptional regulatory elements for transgene expression by adenovirus vector. J. Controlled Release 81: 155–163.
7. Mastrangeli, A., O'Connell, B., Aladib, W., Fox, P. C., Baum, B. J. and Crystal, R. G. (1994). Direct in vivo adenovirus-mediated gene transfer to salivary glands. Am. J. Physiol. 266: G1146–G1155. | PubMed | ISI | ChemPort |
8. Delporte, C., et al. (1997). Increased fluid secretion after adenoviral-mediated transfer of the aquaporin-1 cDNA to irradiated rat salivary glands. Proc. Natl. Acad. Sci. USA 94: 3268–3273. | Article | PubMed | ChemPort |
9. He, X., et al. (1998). Systemic action of human growth hormone following adenovirus-mediated gene transfer to rat submandibular glands. Gene Ther. 5: 537–541. | Article |
10. Kok, M. R., et al. (2003). Local adeno-associated virus-mediated interleukin 10 gene transfer has disease-modifying effects in a murine model of Sjögren's syndrome. Hum. Gene Ther. 14: 1605–1618. | Article | PubMed | ISI | ChemPort |
11. Li, J., et al. (2004). Developing a convenient large animal model for gene transfer to salivary glands in vivo. J. Gene Med. 6: 55–63. | Article | PubMed | ChemPort |
12. Voutetakis, A., et al. (2004). Reengineered salivary glands are stable endogenous bioreactors for systemic gene therapeutics. Proc. Natl. Acad. Sci. USA 101: 3053–3058. | Article | PubMed | ChemPort |
13. Baum, B. J., Wellner, R. B. and Zheng, C. (2002). Gene transfer to salivary glands. Int. Rev. Cytol. 213: 93–146. | PubMed | ChemPort |
14. Kim, D. W., Uetsuki, T., Kaziro, Y., Yamaguchi, N. and Sugano, S. (1990). Use of the human elongation factor 1 alpha promoter as a versatile and efficient expression system. Gene 91: 217–223. | Article | PubMed | ISI | ChemPort |
15. Mizushima, S. and Nagata, S. (1990). PEF-BOS, a powerful mammalian expression vector. Nucleic Acids Res. 18: 5322. | PubMed | ISI | ChemPort |
16. Moll, R., Franke, W. W., Schiller, D. L., Geiger, B. and Krepler, R. (1982). The catalog of human cytokeratins: patterns of expression in normal epithelia, tumors and cultured cells. Cell 31: 11–24. | Article | PubMed | ISI | ChemPort |
17. Chow, Y. H., et al. (1997). Development of an epithelium-specific expression cassette with human DNA regulatory elements for transgene expression in lung airways. Proc. Natl. Acad. Sci. USA 94: 14695–14700. | PubMed |
18. Brembeck, F. H. and Rustgi, A. K. (2000). The tissue-dependent keratin 19 gene transcription is regulated by GKLF/KLF4 and Sp1. J. Biol. Chem. 275: 28230–28239. | PubMed | ISI | ChemPort |
19. Kagaya, M., Kaneko, S., Ohno, H., Inamura, K. and Kobayashi, K. (2001). Cloning and characterization of the 5'-flanking region of human cytokeratin 19 gene in human cholangiocarcinoma cell line. J. Hepatol. 35: 504–511.
20. Harvey, T. J., Hooper, J. D., Myers, S. A., Stephenson, S. A., Ashworth, L. K. and Clements, J. A. (2000). Tissue-specific expression patterns and fine mapping of the human kallikrein (KLK) locus on proximal 19q13.4. J. Biol. Chem. 275: 37397–37406. | Article | PubMed | ISI | ChemPort |
21. Scicli, A. G. (1994). Increases in cardiac kinins as a new mechanism to protect the heart. Hypertension 23: 419–421.
22. Figueroa, C. D., MacIver, A. G., Mackenzie, J. C. and Bhoola, K. D. (1988). Localisation of immunoreactive kininogen and tissue kallikrein in the human nephron. Histochemistry 89: 437–442. | Article | PubMed | ISI | ChemPort |
23. Schmaier, A. H., Smith, P. M., Purdon, A. D., White, J. G. and Colman, R. W. (1986). High molecular weight kininogen: localization in the unstimulated and activated platelet and activation by a platelet calpain(s). Blood 67: 119–130.
24. Simson, J. A. and Chao, J. (1994). Subcellular distribution of tissue kallikrein and Na,K-ATPase alpha-subunit in rat parotid striated duct cells. Cell Tissue Res. 275: 407–417.
25. Meisler, M. H. and Ting, C. N. (1993). The remarkable evolutionary history of the human amylase genes. Crit. Rev. Oral Biol. Med. 4: 503–509. | PubMed | ISI | ChemPort |
26. Ting, C., Rosenberg, M. P., Snow, C. M., Samuelson, L. S. and Meisler, M. H. (1992). Endogenous retroviral sequences are required for tissue-specific expression of a human salivary amylase gene. Genes Dev. 6: 1465–1475.
27. Funaki, H., et al. (1998). Localization and expression of AQP5 in cornea, serous salivary glands, and pulmonary epithelial cells. Am. J. Physiol. 275: C1151–C1157. | PubMed | ISI | ChemPort |
28. Borok, Z., Li, X., Fernandes, V. F., Zhou, B., Ann, D. K. and Crandall, E. D. (2000). Differential regulation of rat aquaporin-5 promoter/enhancer activities in lung and salivary epithelial cells. J. Biol. Chem. 275: 26507–26514.
29. He, X., Tse, C. M., Donowitz, M., Alper, S. L., Gabriel, S. E. and Baum, B. J. (1997). Polarized distribution of key membrane transport proteins in the rat submandibular gland. Pflugers Arch. 433: 260–268. | PubMed | ISI | ChemPort |
30. Gresz, V., et al. (2004). Immunolocalization of AQP-5 in rat parotid and submandibular salivary glands after stimulation or inhibition of secretion in vivo. Am. J. Physiol. Gastrointest. Liver Physiol. 287: G151–G161.
31. Oshima, R. G., Abrams, L. and Kulesh, D. (1990). Activation of an intron enhancer within the keratin 18 gene by expression of c-fos and c-jun in undifferentiated F9 embryonal carcinoma cells. Genes Dev. 4: 835–848. | PubMed | ISI | ChemPort |
32. Pankov, R., Neznanov, N., Umezawa, A. and Oshima, R. G. (1994). AP-1, ETS, and transcriptional silencers regulate retinoic acid-dependent induction of keratin 18 in embryonic cells. Mol. Cell. Biol. 14: 7744–7757. | PubMed | ISI | ChemPort |
33. Pankov, R., Umezawa, A., Maki, R., Der, C. J., Hauser, C. A. and Oshima, R. G. (1994). Oncogene activation of human keratin 18 transcription via the Ras signal transduction pathway. Proc. Natl. Acad. Sci. USA 91: 873–877.
34. Brown, A. M., Rusnock, E. J., Sciubba, J. J. and Baum, B. J. (1989). Establishment and characterization of an epithelial cell line from the rat submandibular gland. J. Oral Pathol. Med. 18: 206–213. | PubMed | ISI | ChemPort |
35. Adesanya, M. R., Redman, R. S., Baum, B. J. and O'Connell, B. C. (1996). Immediate inflammatory responses to adenovirus-mediated gene transfer in rat salivary glands. Hum. Gene Ther. 7: 1085–1093. | PubMed | ISI | ChemPort |
36. Malone, R. W., et al. (1994). Dexamethasone enhancement of gene expression after direct hepatic DNA injection. J. Biol. Chem. 269: 29903–29907. | PubMed | ISI | ChemPort |
37. Darlington, G. J., Tsai, C. C., Samuelson, L. C., Gumucio, D. L. and Meisler, M. H. (1986). Simultaneous expression of salivary and pancreatic amylase genes in cultured mouse hepatoma cells. Mol. Cell. Biol. 6: 969–975.
38. Zheng, C., Hoque, A. T., Braddon, V. R., Baum, B. J. and O'Connell, B. C. (2001). Evaluation of salivary gland acinar and ductal cell-specific prom oters in vivo with recombinant adenoviral vectors. Hum. Gene Ther. 12: 2215–2223. | Article | PubMed | ChemPort |
39. Zheng, C., Hoffman, M. P., McMillan, T., Kleinman, H. K. and O'Connell, B. C. (1998). Growth factor regulation of the amylase promoter in a differentiating salivary acinar cell line. J. Cell. Physiol. 177: 628–635.
40. Amalfitano, A. and Chamberlain, J. S. (1996). Isolation and characterization of packaging cell lines that coexpress the adenovirus E1, DNA polymerase, and preterminal protein: implication for gene therapy. Gene Ther. 4: 258–263.
41. Becker, T. C., et al. (1994). Use of recombinant adenovirus for metabolic engineering of mammalian cells. In: Roth M.G. (Ed). Methods in Cell Biology. Academic Press: San Diego, CA. 161–189.
42. Meunier-Durmort, C., Grimal, H., Sachs, L. M., Demeneix, B. A. and Forest, C. (1997). Adenovirus enhancement of polyethylenimine-mediated transfer of regulated genes in differentiated cells. Gene Ther. 4: 808–814. | Article | PubMed | ChemPort |
43. Baker, A., et al. (1997). Polyethylenimine (PEI) is a simple, inexpensive and effective reagent for condensing and linking plasmid DNA to adenovirus for gene delivery. Gene Ther. 4: 773–782. | Article | PubMed | ISI | ChemPort |

Appendices

Appendix A

Supplementary data

Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.ymthe.2005.03.008.

Acknowledgements

We thank Drs. J. Chiorini, B. Lodde, M. Schmidt, and J. Wang for helpful comments on the manuscript.