Characterization of Epididymal Epithelial Cell: RESULTS
Localization of EGFP Expression after In Vivo Electroporation in the Initial Segment
To test the ability of the in vivo electroporation method to express genes in the initial segment, a plasmid-encoding EGFP under the control of the cytomegalovirus (CMV) promoter was injected intraluminally into the initial segment and electroporated. After 72 h, the initial segments were removed and analyzed with integrated confocal immunofluorescence to assess the expression of EGFP As seen in Figure 2B, intense fluorescence was observed in the cytoplasm of the epithelial cells. This expression was observed in all tubules that had been injected with the plasmid and could be visualized as much as 1 wk after electroporation (data not shown). Next, a plasmid containing the full 1976-bp GGT promoter IV controlling EGFP expression was used. Intense staining was observed in what appeared to be a vesicular compartment (Fig. 2C). This construct contains the first 11 amino acids of firefly luciferase attached to EGFP, which may be responsible for the vesicular targeting in these cells. Minor background fluorescence can be seen in the sections in which no DNA was injected (Fig. 2A). With the intraluminal injection method, we frequently experienced tubule blockages that would result in little or no flow of fluid from the testis to the epididymis. As this defeated the purpose of analyzing these promoters by this method, we adopted an interstitial method of plasmid injection. Results from the interstitial injection of the GGT promoter IV-EGFP construct are shown in Figure 2D and reveal a similar fluorescence pattern as seen with the intraluminal injection. To avoid the complications with the intraluminal injections, the remainder of the experiments were carried out with interstitial injections.
Analysis of Initial Segment-Specific Promoters by In Vivo Electroporation
GGT promoter IV and the cres promoter have been analyzed previously in primary initial segment epithelial cell culture and anterior pituitary cell lines, respectively. However, because of their dependence on LTFs for expression in the initial segment, analysis of these promoters in the context of the whole tissue was performed utilizing the in vivo electroporation method. Electroporation of the previously characterized minimal cres promoter (Fig. 1A) resulted in transcriptional activity from the parental plasmid (135 in Fig. 3A). Mutation of the 3′ C/EBP site (1M) had no effect on transcriptional activity, whereas mutation of the 5′ C/EBP site (2M) resulted in a 54% reduction of transcriptional activity. Mutation of both sites (DM) produced no statistically significant difference in luciferase expression compared to the 2M construct even though there may have been a significant decrease with a larger number of experiments.
Next, equal concentrations of the 135-bp GGT promoter IV construct (pGL3-b135), the 135 construct with a deleted PEA3 site (pGL3-b135del), and the 135 construct with a mutated PEA3 site (pGL3-b135mut) were electroporated into the tissue. Whereas deletion of 20 bp containing a putative PEA3 site abrogated luciferase expression, specific mutation of the PEA3 site had no significant effect on the expression from the 135-bp GGT promoter IV construct (Fig. 3B).
Analysis of the 2-kb GGT Promoter IV by InVivo Electroporation
Because of the unexpected results obtained with pGL3-b135, experiments were carried out with the 2-kb promoter to determine if sequences 5′ to the 135 bp were responsible for GGT promoter IV activity in vivo. Equimolar concentrations of pGL3-b1976 and the 5′ deletion constructs of pGL3-b1976 (Fig. 1B) were electroporated into the proximal half of the initial segment. Although each construct was able to generate some level of transcriptional activity, electroporation of the pGL3-b530 construct resulted in a significant increase in transcriptional activity compared to the other constructs (Fig. 4). Truncation of pGL3-b530 to —250 bp decreased the activity of the promoter by almost 95%, indicating the presence of a positive cis-acting regulatory element(s) between —250 and —530 bp. Additionally, when the promoter region was extended to — 903 and— 1976 bp, transcriptional activity decreased by 94% and 97%, respectively, suggesting the presence of a cis-acting negative regulatory element between —530 and —903 bp. To further define the region containing the cis-acting negative regulatory sequence, an additional construct, pGL3-b681, was created. When this construct was used for in vivo electroporation, expression was reduced by 79% compared to pGL3-b530, suggesting that the major repressor activity was between —530 and —681 bp (Fig. 4). This sequence (-530 to -681 bp) was then subcloned in front of the SV40 early promoter in pGL3-control to determine if it could function as a general repressor. No difference was observed between pGL3-control with and without this sequence in in vivo electroporations (data not shown).
Analysis of pGL3-b530 Mutants
Analysis of the sequence between -250 and -530 bp reveals the presence of three consensus PEA3 DNA-binding motifs found at -452 to -447 bp, -399 to -394 bp, and -369 to -364 bp in GGTpromoter IV (Fig. 1B). To determine if any of these PEA3 sites were responsible for the activity of the pGL3-b530 construct in vivo, PCR-based mutagenesis was performed to change the consensus 5′-AGGAAG-3′ to 5′-AGAGAG-3′ beginning with the most 5′ site, yielding pGL3-b530m1, pGL3-b530m2, and pGL3-b530m3, respectively. In vivo, mutation of any one of the three PEA3 sites resulted in an 88%-90% decrease in promoter activity (Fig. 5). An additional construct, pGL3-b530c, was made by mutating the sequence 5′-AGGTAG-3′ at -385 to -380 bp, which closely resembles a PEA3 DNA-binding site, to 5′-AGAGAG-3′. When this construct was used for in vivo electroporation, no significant difference compared to the wild-type pGL3-b530 construct was observed (Fig. 5).
Cloning and Analysis of Additional GGT Promoter IV Sequence
The results shown in Figure 3 suggested that pGL3-b1976 does not represent the entire GGT promoter IV. To examine more 5′ sequences, a 4.5-kb fragment from a X phage clone-containing rat genomic DNA between exon V and promoter IV was cloned into pGL3-b1976. In vivo electroporations were carried out with this construct, pGL3-b6500, and compared to electroporations done with equi-molar concentrations of pGL3-b530. As shown in Figure 6, pGL3-b6500 expression was only 7% of pGL3-b530. These results were similar to pGL3-b1976, which had only 3% the expression of pGL3-b530 (Fig. 5), suggesting that additional 5′ sequences may be involved in the control of GGT promoter IV.
FIG. 1. A) Schematic representation of the minimal cres promoter. Black boxes denote the two C/EBP sites found within the 135-bp cres promoter. B) Schematic representation of the 2-kb promoter for GGT mRNA IV. Black boxes denote consensus PEA3 sites (5′-AGGAAG-3′). White boxes denote nonconsensus PEA3 sites (5′-AGGAAc/t-3′). Asterisks denote those PEA3 sites that are in the opposite orientation with respect to the GGT locus.
FIG. 2. Integrated confocal fluorescence images of initial segments injected with no DNA control (A), intraluminal CMV-EGFP
(B), intraluminal GGT promoter IV-EGFP
(C), and interstitial GGT promoter IV-EGFP
(D). The scale bars represent 50 |m in all panels.
FIG. 3. In vivo electroporation of initial segment-specific promoters. A) Equal concentrations of the cres promoter constructs 135, 1M, 2M, or DM were coinjected with pRL-TK into the initial segment and electroporated. Dual luciferase assays were performed 72 h after electroporation. B) Equal concentrations of the GGT promoter IV constructs pGL3-b135, pGL3-b135del, or pGL3-b135mut were coinjected with pRL-SV40 into the proximal half of rat initial segments and electroporated. Dual luciferase assays were performed 24 h after electroporation. Results from both experiments are expressed as the ratio of firefly luciferase units (FLU) to renilla luciferase units (RLU). Each promoter construct was tested in three to five rats and expressed as the mean FLU-to-RLU ratio ± SEM. Mean ratios with different numbers are significantly different (P < 0.05) as assessed by one-way ANOVA followed by Tukey test.
FIG. 4. In vivo electroporation of GGT promoter IV constructs. Equi-molar concentrations of the various GGT promoter IV constructs were coinjected with pRL-SV40 into the proximal half of rat initial segments and electroporated. Dual luciferase assays were performed 24 h after electroporation. Results are expressed as the ratio of FLU to RLU. Each promoter construct was tested in at least five rats and expressed as the mean FLU to RLU ± SEM. Mean ratios with different numbers are significantly different (P < 0.001) as assessed by one-way ANOVA followed by Tukey test.
FIG. 5. In vivo electroporation of pGL3-b530 and pGL3-b530m1-3 constructs. Equal amounts of either pGL3-b530, pGL3-b530m1-3, or pGL3-b530c were coinjected with pRL-SV40 and electroporated into the proximal half of rat initial segments. Dual luciferase assays were performed 24 h after electroporation. Results are expressed as the ratio of FLU to RLU. Each promoter construct was tested in five rats and expressed as the mean FLU-to-RLU ratio ± SEM. Mean ratios with different numbers are significantly different (P < 0.001) as assessed by one-way ANOVA followed by Tukey test.
FIG. 6. In vivo electroporation of pGL3-b6500. Equimolar concentrations of either pGL3-b530 or pGL3-b6500 were coinjected with pRL-SV40 and electroporated into the proximal half of the initial segment. Dual luciferase assays were performed 24 h after electroporation. Results are expressed as the ratio of FLU to RLU. Each promoter construct was tested in four rats and expressed as the FLU-to-RLU ratio ± SEM. The mean ratio difference between pGL3-b530 and pGL3-b6500 is statistically significant (P < 0.0001) as assessed by an unpaired, two-tailed t-test.