Skip Navigation


Human Molecular Genetics Advance Access originally published online on July 6, 2004
Human Molecular Genetics 2004 13(17):1933-1941; doi:10.1093/hmg/ddh200
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
13/17/1933    most recent
ddh200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (1)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Ulatowski, L. M.
Right arrow Articles by Drumm, M. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ulatowski, L. M.
Right arrow Articles by Drumm, M. L.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Human Molecular Genetics, Vol. 13, No. 17 © Oxford University Press 2004; all rights reserved

Strain-specific variants of the mouse Cftr promoter region reveal transcriptional regulatory elements

Lynn M. Ulatowski1, Kirstin L. Whitmore2, Todd Romigh1, Annalisa S. VanderWyden1, Shama M. Satinover1 and Mitchell L. Drumm1,2,*

1Department of Pediatrics and 2Department of Genetics, Case Western Reserve University, Cleveland, OH 44106, USA

Received March 12, 2004; Accepted June 22, 2004


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Regulation of cystic fibrosis transmembrane conductance regulator (CFTR) mRNA levels is not well understood. Mouse Cftr mRNA shows strain-dependent expression differences that cannot be fully explained by variation at non-Cftr loci. Differences in tracheal and colonic expression appear to be due predominantly to elements linked to Cftr. Fifteen single nucleotide sequence variations were found within 1.4 kb 5' to the translation start site between the inbred lines A/J, C57BL/6J and 129/SvJ. In addition, 129/SvJ carries a 100 bp deletion relative to the other two strains. These variants were investigated by sequentially deleting 5' regions and measuring luciferase reporter activity from transfected, mouse epithelial cell lines derived from pancreatic duct, renal collecting duct, salivary gland and trachea. These assays identified a region between –524 and –834 in the C57BL/6J promoter, but not in A/J or 129/SvJ, capable of repressing expression. Sequence analysis and gel mobility shift assays suggest that the transcription factor MZF is involved in the strain-dependent effect. It was also apparent that several reporter constructs displayed expression differences between cell lines, possibly indicating the presence of tissue-specific elements.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
The cystic fibrosis transmembrane conductance regulator (CFTR) is a cyclic AMP (cAMP)-regulated chloride channel found in the epithelia of many organs (reviewed in 1). cAMP not only regulates CFTR at the channel level, but is also critical for transcription of CFTR (2,3). Accordingly, the CFTR promoter region has a consensus cAMP-response element (CRE) conserved across species (4). Other transcription regulatory elements have been suggested on the basis of evolutionary conservation (4), consensus motifs (5–7), and by functional analyses (2,4,7–10). However, these studies have identified elements responsible for basal expression and for transcriptional changes in response to various stimuli, but the apparent tight tissue-specific control of expression is not well understood. The transcriptional start sites of both mouse and human show differences between tissues and also temporal effects in the same tissue (11). DNase hypersensitivity analyses have identified sites that correspond to tissue-specific expression (12–14), but few specific functions of the elements have been established (15).

Phylogenetic comparison of the CFTR promoter region from several species clearly points to conserved regions (4,16,17), suggesting functional conservation of those regions. Here, we extend that concept by comparing the Cftr promoter region in several strains of the same species, Mus musculus, for associations in Cftr mRNA levels. An advantage of concentrating on the mouse model is that functional variation has a much more tractable level of genetic variation than human studies. Thus, interpreting associations between sequence changes and functional differences is more readily achieved.

Nucleotide polymorphisms in exons, or expressed polymorphisms, allow one to assess expression of one allele relative to the other in heterozygous samples. We have found that relative allele expression of Cftr in F1 hybrid mice between C57BL/6J and 129/SvJ displays a tissue-specific profile. In a given cell or tissue, both alleles are expected to be exposed to the same complement of transcription factors, or other trans-acting elements. As the two alleles in heterozygous animals respond differently, we interpret this observation to indicate the existence of cis elements that cause one allele to be expressed at a different level than the other and that these elements are recognized differently in different tissues. To begin to identify the cis elements, we have analyzed a segment of the putative promoter region, where 16 sequence differences exist.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Cftr promoter region variants
Genomic DNA corresponding to 1.4 kb upstream of the translation initiation site from the three inbred strains was sequenced, and 16 sequence variants were identified. As shown in Figure 1, 15 single nucleotide variants were found and a deletion of 100 base pairs in 129/SvJ relative to the other two strains. The 129/SvJ variant is presumed to have arisen by a deletion in 129/SvJ rather than an insertion in the other strains for two reasons. First, alignment with Rattus norvegicus (GenBank accession number L26098) genomic DNA indicates the rat promoter to retain the 100 bases, similar to A/J and C57BL/6J (data not shown). Second, inspection of the A/J and C57BL/6J sequences reveals a tandem duplication of ~84 bp interspersed by 16 nucleotides (Fig. 1). The repeated sequences are 85% identical, but when compared with the 129/SvJ monomeric version there is complete identity to the 5' end of the first repeat and the 3' end of the second repeat, suggesting that the 129/SvJ monomer arose from an unequal crossover between the tandem repeats.



View larger version (43K):
[in this window]
[in a new window]
  		Figure 1. Sequence comparison of the Cftr promoter region between three inbred mouse strains. Genomic DNA from C57BL/6J, A/J and 129/SvJ was amplified and sequenced. The sequence shown represents C57BL/6J, with 15 variant nucleotides shown in bold. The nucleotide immediately below C57BL/6J corresponds to the 129/SvJ variant and the bottom nucleotide corresponds to A/J. Lower case letters designate residues comprising the 84 bp repeat units found in A/J and C57BL/6J with asterisks indicating bases differing between the 5' and 3' copies of the repeat. The underlined residues represent the 100 bp region deleted in 129/SvJ with the hatched region indicating the site of the ancestral recombination leading to the deletion. Nucleotide positions with downward arrows indicate 5' terminal bases of deletion constructs used for luciferase assays. Boxed is the region described as a purine/pyrimidine repeat (4,6) with the putative MZF-binding site indicated. Variants at –621 and –618 are contained in the MZF consensus. Other elements, including CRE (6), and an inverted CAAT, or Y-box (2,3), described and verified by others are shown, labeled above the sequence and with a line over the motif.

 
Expressed sequence variants
mRNAs from C57BL/6J and 129/SvJ were reverse transcribed and polymerase chain reaction (PCR) amplified. The PCR products were sequenced and revealed 13 variants in the transcripts as well. These variants are listed in Table 1. As Table 1 also indicates, three of the variants create restriction site polymorphisms in exons 2, 14a and 17a. These expressed polymorphisms were used to quantify the relative expression from alleles in mice heterozygous at the Cftr locus.


View this table:
[in this window]
[in a new window]
  		Table 1. Sequence differences between C57BL/6J and 129/SvJ. The first nucleotide or amino acid refers to the C57BL/6J sequence
 
Quantifying relative levels of Cftr mRNA
RT–PCR products for each region studied were cloned into plasmid vectors and used as copy controls to develop a quantitative assay for the alleles. When mixed in proportions ranging from 0 to 100% of 129/SvJ and PCR amplified for 30 cycles, we found the assay to be linear over a range of 4 logs of template DNA (Fig. 2A and B).



View larger version (18K):
[in this window]
[in a new window]
  		Figure 2. Relative allele expression assay. Cloned fragments of exon 17a from 129/SvJ and C57BL/6J were mixed in proportions ranging from 0 to 100% 129/SvJ, amplified and proportions measured as described in Materials and Methods. (A) Results of the assay in which the proportion of 129/SvJ is increased in 10% increments. The top band corresponds to 129/SvJ and the bottom to C57BL/6J. (B) A plot of the phosphorimaging quantification of gels like that shown in (A). Repetitions were carried out in which the total amount of DNA amplified was varied from 1 pg to 1 ng. The assay is linear at all DNA concentrations and over the full range of mixtures.

 
To examine allelic effects on mRNA levels, F1 hybrids derived from 129/SvJxC57BL/6J or 129/SvJxA/J crosses were generated and RNA samples were isolated from several tissues, including nasal epithelium, trachea, lung, pancreas, ileum, jejunum, proximal and distal colon, salivary gland and kidney. Products from RT–PCR of the RNA were analyzed for relative Cftr mRNA levels using the conditions described for the plasmid standards. An example of the results from three tissues of three different mice is shown in Figure 3A. Results, expressed as the percentage of the total Cftr mRNA corresponding to the 129/SvJ allele, are shown in Figure 3B. Of note, few tissues show equal quantities of Cftr mRNA from each allele. Expression in the small intestines tended to be equal, whereas airway (nose, trachea and lung), salivary gland, kidney and colon showed significant skewing (P<0.05). Skewing also depended on the allele in which combination was being studied, and was most prominent in the lung, where 129/SvJ predominated over C57BL/6J, but A/J was in greater abundance than 129/SvJ. Pancreatic RNA was also examined but failed to yield reproducible results in this assay.



View larger version (37K):
[in this window]
[in a new window]
  		Figure 3. Relative allele expression from epithelial tissues. Using the assay shown in Figure 2, RNA was RT–PCR amplified and the relative contribution of 129/SvJ and C57BL/6J alleles measured. (A) An example of nasal (N), tracheal (T) and lung (L) samples from three different animals, and the reproducibility of the assay in genetically matched animals. The control (C) corresponds to RNA from a C57BL/6J lung that is included as an indicator that complete digestion has occurred. (B) The measurement of Cftr mRNA from each allele in 129/SvJxC57BL/6J and 129/SvJxA/J crosses. Tissues surveyed include nasal epithelium (N), salivary gland (S), trachea (T), lung (L), duodenum (D), jejunum (J), ileum (I), proximal colon (PC), distal colon (DC) and kidney (K).

 
Genetic evidence for cis and trans-acting effects
The observed tissue-associated expression differences may be due solely to Cftr sequence variants, but there may be variants of other loci contributing as well. We reasoned that if the Cftr variants confer the effects, the expression differences should track with the Cftr allele, independent of the rest of the genome. Alternatively, if loci other than Cftr are involved in the expression differences, mixing the genomes should increase the variation in expression. To begin to address these possibilities, crosses between 129/SvJ and C57BL/6J were carried out in order to examine the inheritance pattern of the allele-specific expression phenotype. For these analyses we focused on trachea and distal colon, the tissues with the greatest expression differences between 129/SvJ and C57BL/6J alleles. Allele expression was compared in mice with various mixtures of the C57BL/6J and 129/SvJ genomes. F1 animals from C57BL/6Jx129/SvJ matings were crossed and Cftr mRNA measured in F2 progeny heterozygous for Cftr. F1 animals were also backcrossed to either C57BL/6J or 129/SvJ inbreds, and the Cftr heterozygotes assayed to determine if the parent of origin for an allele affected the expression. As Figure 4 shows, the percentage of 129/SvJ Cftr mRNA is the same in F1, F2 and backcrossed animals, as is the variation, suggesting that loci unlinked to Cftr are not contributing significantly to the variation in allele expression between trachea and distal colon.



View larger version (13K):
[in this window]
[in a new window]
  		Figure 4. Cftr mRNA expression in 129/SvJxC57BL/6J crosses. Trachea and distal colon were analyzed from 129/SvJxC57BL/6J F1 and F2 animals, as well as progeny of F1 animals backcrossed to 129/SvJ (129BX) or C57BL/6J (B6BX) mice. Neither the proportion nor the variance of 129/SvJ-derived Cftr mRNA differs significantly for trachea or distal colon in any of the crosses.

 
Cftr expression comparison between tissue and cell line
The genetic evidence suggests that the major mechanism controlling strain-specific differences in Cftr mRNA levels lies in cis elements. Although this could involve sequences affecting mRNA turnover, the sequence variants shown in Figure 1 suggest the possibility of differential transcriptional regulation. We have focused here on transcriptional control conferred by the 1429 bp upstream of the translational start site. Not only are there sequence variants in this region, but it is also a site of established cis-acting promoter elements in the mouse (6) and human promoter (2,4,9,10,18,19). Because of the limitations of studying promoter activity and mRNA regulation in intact tissues, we utilized a cell culture model in which reporter constructs could be used to identify and study potential regulatory elements. Epithelial cell lines derived from mouse renal collecting duct (CT-1), pancreatic duct (PEC), trachea (TEC) and salivary gland (SEC) were used for the following studies (20). Using real-time PCR, Cftr mRNA was quantified from reverse-transcribed RNA from each of the cell lines (Fig. 5A). Each of the cell lines was characterized for Cftr mRNA level, both to verify that they retained the ability to express Cftr and to compare the relative expression of Cftr in the various tissues from which they were derived (Fig. 5B).



View larger version (15K):
[in this window]
[in a new window]
  		Figure 5. Cftr mRNA levels from native tissue and corresponding cell lines. Total RNA was extracted from cell lines (A) or tissues (B) and reverse transcribed. Cftr mRNA was quantified by real-time PCR and normalized by comparing to GAPDH levels. Molecules of Cftr per molecule of GAPDH are shown for each cell line (A) and tissue (B). Error bars indicate standard deviation.

 
All four cell lines expressed readily detectable Cftr mRNA levels, although the salivary-derived cells express about half as much as the pancreatic and tracheal cells. Similarly, the four corresponding tissues were analyzed from all three strains' Cftr mRNA levels to determine relative expression levels in the presence of a single allele.

The results shown in Figure 5 demonstrate that all three mouse strains expressed similar levels of Cftr mRNA, relative to GAPDH mRNA, in kidney, but showed significant differences in the salivary gland (Fig. 5B). Relative to GAPDH mRNA, A/J and C57BL/6J salivary glands express similar amounts of Cftr mRNA (0.030±0.006 and 0.026±0.006, respectively, P=0.63), but only about one-third that of 129/SvJ (0.080±0.020, P=0.007 versus A/J and P=0.05 versus C57BL/6J). No other significant differences were observed.

Promoter analyses
To quantitatively assess activity of promoter elements, a 1.4 kb region from A/J and C57BL/6J and 1.3 kb from 129/SvJ was cloned into the luciferase reporter plasmid pGL3-Basic (Promega). These constructs, containing 1429 bp upstream of the initiation of translation, are designated –1429/AJ, –1429/B6 or –1329/129 corresponding to the strains A/J, C57BL/6J and 129/SvJ, respectively, from which they were derived. Using the same nomenclature, deletions to –1014, –834, –524 and –219 were generated and assayed for promoter activity as well (Figs 1 and 6). For comparison, promoter activity of each fragment is shown relative to the –219 C57BL/6J construct expressed in the same cell line in the same experiment.



View larger version (26K):
[in this window]
[in a new window]
  		Figure 6. Comparison of promoter fragments from different strains. Five promoter constructs from each inbred strain were generated and used to drive luciferase expression in the four epithelium-derived cell lines. Corresponding constructs from the three strains are grouped and lengths indicated by the name. Constructs derived from A/J are designated by ‘A’ and white bars, those from C57BL/6J by ‘B’ and black bars and those from 129/SvJ by ‘S’ and grey bars. The numerical part of the label represents the number of nucleotides 5' of the translational start site comprised by the construct. The longest construct, –1429, is generated from C57BL/6J and A/J, whereas its 129/SvJ counterpart is 100 bp shorter and designated as –1329. Error bars indicate SEM.

 
The results, displayed in Figure 6, show several significant differences between the constructs, dependent on strain of origin. Some of these differences are cell line dependent, but not all. In fact, the most apparent effect is seen with the –834 C57BL/6J construct, which has substantially less activity than all other constructs in all four cell lines (P<0.05, ANOVA and Dunnett test). Relative activity of this construct ranged from 0.17 of the –219 C57BL/6J construct in the PEC line, to 0.46 in the TEC line.

Cell line-dependent effects were also observed. Beginning at the region most proximate to the initiation of translation, the activity of the –219 construct is consistently lower than either the A/J or 129/SvJ homologues in CT-1, but not the other three cell lines (P<0.05). In contrast, all three –524 constructs behave similarly in each of the four cell lines. The –834 constructs behave similarly between cell lines, with the exception that SEC cells show higher activity of the A/J construct than 129/SvJ (P<0.05).

Other trends are also apparent, such as the high activity of the 129/SvJ–1014 construct in CT-1 cells, but do not cross the threshold for significance (P>0.05).

Sequence analysis and DNA-binding activity
To begin to determine the identity of the factor acting on the –618/–621 positions, electromobility shift assays (EMSAs) were carried out. As Figure 7A and B shows, a DNA-binding activity for this region exists in lysates from the epithelial cell lines. Analysis of the sequence containing and immediately adjacent to the –618 and –621 positions for transcriptional regulatory motifs was carried out using MatInspector (21), and revealed several potential regulatory elements with consensus binding sites for transcription factors related to Myb (22), myeloid-specific zinc finger protein-1 (MZF-1) (23), GKLF (24) and IK2 (25). Oligonucleotides corresponding to the consensus sequence of these factors were constructed and used as competitors of the native mouse sequence. Figure 7A and B shows that of the five competitors used, the MZF-1 sequence (7 lane 5, Fig. 7A and lanes 4 and 8, Fig. 7B) competes as well as the native sequence (lane 3), whereas the other oligos competed weakly or not at all (lanes 4, 6 and 7, Fig. 7A).



View larger version (60K):
[in this window]
[in a new window]
  		Figure 7. EMSAs and expression profile suggest that MZF influences Cftr promoter activity. (A) Nuclear lysates from the PEC cell line were incubated with a labeled oligonucleotide probe spanning the –621/–618 region and corresponding to the C57BL/6J sequence. Probe and lysate (lane 2) show a mobility shift not seen by probe alone (lane 1). The DNA-binding activity of nuclear lysate proteins is competed by the Cftr sequence (lane 3) and a consensus sequence for MZF-1 binding (lane 5), but not by consensus sequences for Myb (lane 4), KLF (lane 6) or IK2 (lane 7). (B) Comparison of the C57BL/6J (lanes 1–4, 1 h exposure) and A/J (lanes 5–8, 24 h exposure) probe sequences' effect on electromobility shift. Compared to probe alone (lanes 1 and 5), PEC lysate shows strong binding to the C57BL/6J sequence (lane 2) that is competed by itself (lane 3) or an MZF consensus sequence (lane 4), whereas the A/J probe shows apparently weak binding and an additional shifted fragment of slightly lower molecular weight (lane 6). Both of these bands are competed by unlabeled probe (lane 7) or MZF consensus. (C) mRNA from PEC cells (lanes 2 and 3) and a human epithelial cell line, A549 (lanes 4 and 5), was reverse transcribed and amplified with MZF-specific primers matching both human and mouse sequences. Both cell lines show expression (lanes 2 and 4). No amplification was observed in the absence of reverse transcription (lanes 3 and 5).

 
As sequence analysis (21) shows the consensus for MZF-1 to better match C57BL/6J than A/J, we then compared binding of the nuclear lysate factor to the C57BL/6J and A/J sequences (Fig. 7B) and found that affinity for C57BL/6J appears to be higher than to A/J.

Although the EMSAs suggest MZF-1, or a relative, is involved in the observed DNA binding, MZF-1 is reported to be preferentially expressed in myeloid-derived cells (26). We examined the epithelial cell lines for evidence of MZF-1 using RT–PCR. As Figure 7C shows, mRNA for mouse Mzf1 is readily detectable in the pancreatic ductal cells, as well as in a human alveolar epithelial cell line, A549.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
We have provided genetic evidence indicating the variation in Cftr mRNA levels between C57BL/6J and 129/SvJ tissues is predominantly due to elements closely linked to the Cftr gene, as segregation of unlinked loci does not increase the variation in tissue-specific differences. In F1 mice, for example, both alleles should be exposed to the same trans-acting factors, but in F2 animals loci unlinked to Cftr should segregate in a 1 : 2 : 1 ratio and thereby cause increased variation if contributing to the phenotype. As Figure 4 shows, the variances of F1 and F2 mRNA levels are not different.

In the C57BL/6J backcross, all loci will be homozygous or heterozygous for C57BL/6J alleles (BXB6), whereas in the 129/SvJ backcross all loci will be homozygous or heterozygous for 129/SvJ alleles. These crosses should detect dominant or recessive effects that may not have shown up in the F2 crosses owing to the relatively small number of animals studied, but it should also have detected any evidence of epigenetic phenomena, such as imprinting, by switching the parent from which each allele originated. Again, these crosses provided no evidence for factors other than elements cis to Cftr, or linked to Cftr, contributing to the mRNA differences.

As the evidence did not suggest the involvement of variation from loci unlinked to Cftr in the differential regulation between the strains, we reasoned that cis elements, such as the promoter would be a likely location for some of those elements. We used cell lines that express endogenous Cftr mRNA and that are derived from epithelia of different tissues to assay promoter strength. Whereas the longest promoter constructs showed no significant differences between cell lines, and have relatively modest activity, the deletion constructs revealed functional elements that varied between the cell lines expressing them, as well as the strain from which they were derived.

In this survey, we hoped to be able to correlate promoter function with sequence variation and thus identify regulatory elements. Although this may have been accomplished to a limited extent, many of the observations will require extensive follow up experiments. For instance, the shortest promoter fragments, the –219 series, show significant differences between C57BL/6J and the other two constructs. There are only two sequence variants, –31 and –39, in this region and only the –39 discriminates C57BL/6J from A/J and 129/SvJ. Thus, one would predict that the –39C variant found in A/J and 129/SvJ would have higher activity than –39A, the C57BL/6J allele, at least in CT-1 cells. These variants overlap with a consensus for the transcription factor SEF-1, which has been shown to interact with the C/EBP family of factors (27) and a C/EBP-binding element has been identified by Pittman (2) just upstream of this region. In contrast, a much more complex scenario develops in other regions. The region between –834 and –524 of the C57BL/6J –834 construct appears to be repressive, as when this region is deleted (as in the –524 construct) the activity increases between 3- and 10-fold, depending on the cell line. However, the repressive effects of this sequence are not seen when 5' flanking sequences are included, suggesting an interaction between the regions. Nonetheless, five sequence variants reside in the –834 to –524 region, two of which (–621 and –618) are unique to C57BL/6J and therefore likely to confer the repressive effect. The sequence of that region is AGGAACAGACGGGGGAGAAAGAGA, with –621 and –618 in bold. Analysis of this region for transcriptional regulatory motifs by MatInspector (21) reveals several potential regulatory elements with consensus binding sites for transcription factors related to Myb (22), MZF-1 (23), GKLF (24) and IK2 (25) present and competition assays, as well as RT–PCR, suggest MZF-1 is involved. The electromobility shifts suggest the factor involved has stronger binding to the C57BL/6J sequence than to A/J. The core sequence recognized by the MZF family is GGGGA, with the first and fourth guanines most highly conserved. This core sequence is found in the C57BL/6J allele, but the conserved fourth position is an adenine in the A/J allele, which would be expected to diminish its binding. These observations suggest a model in which the activity of this DNA-binding factor is repressive, as promoter reporter constructs carrying the C57BL/6J sequence have substantially lower activity than those with the A/J sequence. Such a repressive activity has been reported by others for MZF-1 (28).

Denamur and colleagues (4,6) originally described the region containing these variants as a purine–pyrimidine repeat that represses transcription, and further showed DNA binding of a 27 kDa protein to the region. Though MZF is predicted to be ~50 kDa, it also displays alternative splicing and thus the DNA-binding activity observed here over a specific motif may be related to that observed for a larger fragment (4). It is interesting that sequence similarity shows the A/J and C57BL/6J promoters to be more closely related than A/J and 129/SvJ, but functionally the A/J and 129/SvJ promoter fragments behave more similarly. Sequence variants between A/J and 129/SvJ, including 10 single nucleotide variants and a 100 bp deletion, have relatively little, if any effect on the activity of these two promoters. In contrast, C57BL/6J differs from A/J at only six single nucleotide positions, whereas 14 single nucleotides and a 100 bp deletion separate it from 129/SvJ.

In studying these variants, our original intent was to determine if elements potentially conferring tissue specificity could be identified using cell line models of the tissues. Cell line-dependent differences in expression were apparent for constructs 834 bp and shorter, but not for the full 1.4 kb region, as shown in Figure 6. Clearly Cftr mRNA regulation includes more than the immediate 5' upstream region, as found by others (29–31) and suggested by conserved regions found by sequence comparison of multiple species (16). Accordingly, it is likely that variation at other positions, other than the proximal sequences investigated here, contribute to the differences we see between tissues. Consistent with that, Cftr mRNA expression in native tissue is substantially greater than that for the corresponding cell lines, especially in pancreas and salivary gland (Fig. 5). Although we believe we have identified sequence variants in regulatory elements by virtue of their effect on Cftr expression, it still remains to be determined whether these residues alter tissue-specific regulatory elements or are involved in basal Cftr expression.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Mice
Mice from the inbred strains A/J, C57BL/6J and 129/SvJ were obtained from Jackson Laboratories at 6–8 weeks of age. Tissues were harvested from 8-week-old animals.

Cftr mRNA comparisons
Mice were sacrificed and tissues were immediately removed and frozen. Tissues were thawed in Trizol reagent (Invitrogen, Carlsbad, CA, USA) and processed according to the manufacturer's instructions. For each allele comparison measurement, 1 µg of total RNA was reverse transcribed according to the manufacture's instructions (First Strand cDNA Synthesis kit, Roche, Indianapolis, USA). One-tenth of the reverse transcription reaction was then used as template for PCR amplification. The primer sequences included: 5'-TATCTCCAAACTCTTCTTCAG-3' (exon 2), 5'-CTTTCTAGTTTTTCAGAAAGT-3' (exon 3), 5'-GACTGTTTTCTTGATGATGTG-3' (exon 14a), 5'-CATAACACAAATAAAGAAGCAG-3' (exon 14b) 5'-TACCATTTTTGACTTCATTCAG-3' (exon 17a), 5'-AGTTGGTGTTCATTGTGATTG-3' (exon 17b) were used to amplify three regions of the Cftr mRNA, corresponding to exons 2–3, 14a and b, and 17a and b, respectively, for 30 cycles of 94°Cx45 s, 55°Cx30 s, 72°Cx30 s. An aliquot of each product was subjected to one more cycle, but in which the forward primer was radiolabeled with 32P. These products were digested by the appropriate restriction endonuclease, fractionated on a 6% acrylamide gel and relative quantities of each fragment determined by phosphorimaging (Storm, Molecular Dynamics). The final radiolabeling cycle was done to remove from the analysis heteroduplexes formed by repetitive denaturing and reannealing that occurs in the PCR reaction.

Cell lines, cell culture and transfection
The epithelial cell lines (mCT-1, mSEC1, mTEC1, mPEC1), derived from renal collecting duct, salivary gland, trachea and pancreatic duct (20), were cultured at 33°C in the exocrine media described for the pancreatic and salivary lines. Transfections with reporter plasmids were carried out by lipofection using Fugene 6 (Roche) according to the manufacturer's instructions. Briefly, 25 000 cells were transfected with 220 ng of total plasmid and 0.88 µl Fugene 6 in a 48-well plate.

Real-time RT–PCR analysis of Cftr mRNA level and MZF RT–PCR
Two microgram of RNA were isolated from tissues and reverse transcribed with random hexamers using the Roche RT kit. PCR was carried out using primers 5'-AAGTGACTCTTCTGATGGGG-3' and 5'-AGGACGATTCCGTTGATGACTG-3' from exons 6 and 7, respectively. PCR was also carried out with the exon 2/exon 3 primer set described earlier. All reactions were carried out in a RotorGene (Phenix Instruments, Hayward, CA, USA) for 35 cycles with SybrGreen (Molecular Probes, Eugene, OR, USA) 1 : 10 000, as the indicator. GAPDH RT–PCR reactions were performed in parallel to normalize the samples for cDNA amounts. GAPDH primer sequences were 5'-TTCCAGTATGACTCCACTCACGG-3' and 5'-TGAAGACACCAGTAGACTCCACGAC-3', and amplify a 169 bp product. MZF cDNA was PCR amplified as earlier, but using primers 5'-CCGGAGATGGGTCACAGTCC-3' and 5'-TTGCTGAACACCTTGCCAC-3', as described in the literature (28).

Plasmid constructs
A 1429 bp fragment from A/J and C57BL/6J genomic DNA and a 1329 bp fragment from 129/SvJ was PCR amplified using the primers 5'-CTGTGAGTTGAGTGACCAGGACACTTT-3' and 5'-GATGTCTCGTGAGGCAATGA-3'. The resulting products were cloned into pCR2.1 (Invitrogen), sequenced, then shuttled into the luciferase reporter vector pGL3-Basic (Promega, Madison, WI, USA) by digesting with EcoRI, blunt-ending by Klenow fill-in and ligating into the SmaI site of pGL3-Basic. Deletions were generated by digesting and re-ligating the longest fragments for each strain with XbaI and EcoRI, AluI and EcoRI, StyI and EcoRI to produce the 1014, –834, –524 reporter plasmids, respectively. These fragments were also blunt-ended and ligated into the SmaI site of pGL3-Basic. The –219 construct was PCR derived with the primers 5'-CGCTGGCTTTAACCTGGGCGG-3' and 5'-GATGTTCCGTGAGGCAATGA-3', and synthesized as described earlier for the longest constructs. Constructs were verified by sequencing, and all were found to have the same vector/insert junction at the 5' end of the promoter fragment.

Luciferase assays
The Dual Luciferase assay system was used (Promega) in order to normalize transfection efficiency. Cells were transfected as described earlier, with 10 ng of pRLTK (Promega), containing the Renilla luciferase coding sequence under control of the Herpes thymidine kinase promoter, and 210 ng of the designated firefly luciferase construct. Firefly constructs were either promoterless (pGL3-Basic) or contained a region of the Cftr promoter. The cells were lysed 24 h post-transfection in 50 µl 1x Passive Lysis Buffer and 20 µl of this volume were assayed on a Molecular Devices Lmax luminometer (Molecular Devices, Sunnyvale, CA, USA) according to Promega's Dual luciferase Reporter Assay System. Cftr promoter activity was assessed as the ratio of firefly luminescence to Renilla, less the ratio of pGL3-Basic to Renilla. An irrelevant plasmid, pBSIIKS–, was also used as a control for the effect of Fugene 6 on the cells.

Electromobility shift assays
Nuclear lysate from mouse epithelial cell lines were harvested according to Panomics (Redwood City, CA, USA) nuclear extraction protocol. Cells were swelled in a hypotonic buffer of 10 mM HEPES, pH 7.9, 10 mM KCl, 10 mM EDTA, 10 mM DTT, protease inhibitor cocktail set I (Calbiochem) and 10% Ipegal. Cells were lysed following resuspension in 20 mM HEPES, pH 7.9, 0.4 M NaCl, 1 mM EDTA and 10% glycerol. Quantification was performed according to Bio-Rad DC protein assay kit.

For each binding reaction, 5 µg nuclear lysate, or for no lysate lanes an equivalent volume of lysate buffer, were incubated with 1 µg poly (dI–dC) and +/– double-stranded cold probe, for 5 min at room temperature. Two µl of EMSA binding buffer (1.5% glycerol, 75 mM KCl, 0.375 mM DTT, 12.5 mM NaCl, 0.375 mM PMSF) and 80 ng of double stranded, 5'-biotinylated probe. The reactions were loaded on a pre-run non-denaturing 6% PAGE mini-gel in 0.5x TBE and electrophoresed at 120 V for 1.5 h. Probes were transferred to Zeta-probe membrane (Bio-Rad) using a Bio-Rad Trans-Blot SD apparatus in 0.5x TBE, for 30 min at 15 V. The membrane was UV-crosslinked on a Stratalinker (Stratagene, LaJolla, CA, USA) at 120 mJ. The EMSA detection method was carried out using Panomics EMSA ‘Gel-Shift’ Kits. The probes encompassed the –635 to –611 sequence of the mouse Cftr promoter. The B6 probe sequence was 5'-GAGAGAGGAACAGACGGGGGAGAAA-3' and AJ was 5'-GAGAGAGGAACAGAAGGAGGAGAAA-3'. Competitor oligos containing consensus sequences for MZF, Myb, KLF and IK2 were 5'-GACGGGGGAG-3', 5'-AGGAACAGAC-3', 5'-GAGGAACAGACGGG-3' and 5'-ACGGGGGAGAAA-3', respectively.

Statistics
Paired comparisons were carried out by t-test, and multiple comparisons were carried out by ANOVA and post hoc analysis by Dunnett's test. All analyses used JMP software (SAS, Cary, NC, USA). Differences were considered significant for P-values ≤0.05.


   ACKNOWLEDGEMENTS
 
We thank Dr Calvin Cotton for the generous gift of providing the epithelial cell lines. This work was supported by the Cystic Fibrosis Foundation and by grant HL-68883 from the National Heart, Lung and Blood Institute.


   FOOTNOTES
 
* To whom correspondence should be addressed. Tel: +1 2163686893; Fax: +1 2163684223; Email: mxd34{at}cwru.edu


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Welsh, M.J., Tsui, L.-C., Boat, T.F. and Beaudet, A.L. (1995) Cystic fibrosis. In Scriver, C.R., Sly, W.S. and Valle, D. (eds), The Metabolic and Molecular Basis of Inherited Disease, 7th edn. McGraw-Hill Inc., pp. 3799–3863.

  2. Pittman, N., Shue, G., LeLeiko, N.S. and Walsh, M.J. (1995) Transcription of cystic fibrosis transmembrane conductance regulator requires a CCAAT-like element for both basal and cAMP-mediated regulation. J. Biol. Chem., 270, 28848–28857.[Abstract/Free Full Text]

  3. Matthews, R.P. and McKnight, G.S. (1996) Characterization of the cAMP response element of the cystic fibrosis transmembrane conductance regulator gene promoter. J. Biol. Chem., 271, 31869–31877.[Abstract/Free Full Text]

  4. Vuillaumier, S., Dixmeras, I., Messai, H., Lapoumeroulie, C., Lallemand, D., Gekas, J., Chehab, F.F., Perret, C., Elion, J. and Denamur, E. (1997) Cross-species characterization of the promoter region of the cystic fibrosis transmembrane conductance regulator gene reveals multiple levels of regulation. Biochem. J., 327, 651–662.

  5. Yoshimura, K., Nakamura, H., Trapnell, B.C., Dalemans, W., Pavirani, A., Lecocq, J.P. and Crystal, R.G. (1991) The cystic fibrosis gene has a ‘housekeeping’-type promoter and is expressed at low levels in cells of epithelial origin. J. Biol. Chem., 266, 9140–9144.[Abstract/Free Full Text]

  6. Denamur, E. and Chehab, F.F. (1994) Analysis of the mouse and rat CFTR promoter regions. Hum. Mol. Genet., 3, 1089–1094.[Abstract/Free Full Text]

  7. Brouillard, F., Bouthier, M., Leclerc, T., Clement, A., Baudouin-Legros, M. and Edelman, A. (2001) NF-kappa B mediates up-regulation of CFTR gene expression in Calu-3 cells by interleukin-1beta. J. Biol. Chem., 276, 9486–9491.[Abstract/Free Full Text]

  8. Rowntree, R.K., Vassaux, G., McDowell, T.L., Howe, S., McGuigan, A., Phylactides, M., Huxley, C. and Harris, A. (2001) An element in intron 1 of the CFTR gene augments intestinal expression in vivo. Hum. Mol. Genet., 10, 1455–1464.[Abstract/Free Full Text]

  9. Koh, J., Sferra, T.J. and Collins, F.S. (1993) Characterization of the cystic fibrosis transmembrane conductance regulator promoter region. Chromatin context and tissue-specificity. J. Biol. Chem., 268, 15912–15921.[Abstract/Free Full Text]

  10. Chou, J.L., Rozmahel, R. and Tsui, L.C. (1991) Characterization of the promoter region of the cystic fibrosis transmembrane conductance regulator gene. J. Biol. Chem., 266, 24471–24476.[Abstract/Free Full Text]

  11. White, N.L., Higgins, C.F. and Trezise, A.E. (1998) Tissue-specific in vivo transcription start sites of the human and murine cystic fibrosis genes. Hum. Mol. Genet., 7, 363–369.[Abstract/Free Full Text]

  12. Smith, A.N., Barth, M.L., McDowell, T.L., Moulin, D.S., Nuthall, H.N., Hollingsworth, M.A. and Harris, A. (1996) A regulatory element in intron 1 of the cystic fibrosis transmembrane conductance regulator gene. J. Biol. Chem., 271, 9947–9954.[Abstract/Free Full Text]

  13. Smith, A.N., Wardle, C.J. and Harris, A. (1995) Characterization of DNASE I hypersensitive sites in the 120 kb 5' to the CFTR gene. Biochem. Biophys. Res. Commun., 211, 274–281.[CrossRef][ISI][Medline]

  14. Smith, D.J., Nuthall, H.N., Majetti, M.E. and Harris, A. (2000) Multiple potential intragenic regulatory elements in the CFTR gene. Genomics, 64, 90–96.[CrossRef][ISI][Medline]

  15. Mouchel, N., Henstra, S.A., McCarthy, V.A., Williams, S.H., Phylactides, M. and Harris, A. (2004) HNF1alpha is involved in tissue-specific regulation of CFTR gene expression. Biochem. J., 378, 909–918.[CrossRef][ISI][Medline]

  16. Thomas, J.W., Touchman, J.W., Blakesley, R.W., Bouffard, G.G., Beckstrom-Sternberg, S.M., Margulies, E.H., Blanchette, M., Siepel, A.C., Thomas, P.J., McDowell, J.C. et al. (2003) Comparative analyses of multi-species sequences from targeted genomic regions. Nature, 424, 788–793.[CrossRef][Medline]

  17. Vuillaumier, S., Kaltenboeck, B., Lecointre, G., Lehn, P. and Denamur, E. (1997) Phylogenetic analysis of cystic fibrosis transmembrane conductance regulator gene in mammalian species argues for the development of a rabbit model for cystic fibrosis. Mol. Biol. Evol., 14, 372–380.[Abstract]

  18. Li, S., Moy, L., Pittman, N., Shue, G., Aufiero, B., Neufeld, E.J., LeLeiko, N.S. and Walsh, M.J. (1999) Transcriptional repression of the cystic fibrosis transmembrane conductance regulator gene, mediated by CCAAT displacement protein/cut homolog, is associated with histone deacetylation. J. Biol. Chem., 274, 7803–7815.[Abstract/Free Full Text]

  19. Romey, M.C., Pallares-Ruiz, N., Mange, A., Mettling, C., Peytavi, R., Demaille, J. and Claustres, M. (2000) A naturally occurring sequence variation that creates a YY1 element is associated with increased cystic fibrosis transmembrane conductance regulator gene expression J. Biol. Chem., 275, 3561–3567.[Abstract/Free Full Text]

  20. Takacs-Jarrett, M., Sweeney, W.E., Avner, E.D. and Cotton, C.U. (2001) Generation and phenotype of cell lines derived from CF and non-CF mice that carry the H-2K(b)-tsA58 transgene. Am. J. Physiol. Cell Physiol., 280, C228–236.[Abstract/Free Full Text]

  21. Quandt, K., Frech, K., Karas, H., Wingender, E. and Werner, T. (1995) MatInd and MatInspector: new fast and versatile tools for detection of consensus matches in nucleotide sequence data. Nucl. Acids Res., 23, 4878–4884.[Abstract/Free Full Text]

  22. Ness, S.A., Marknell, A. and Graf, T. (1989) The v-myb oncogene product binds to and activates the promyelocyte-specific mim-1 gene. Cell, 59, 1115–1125.[CrossRef][ISI][Medline]

  23. Morris, J.F., Hromas, R. and Rauscher, F.J., 3rd (1994) Characterization of the DNA-binding properties of the myeloid zinc finger protein MZF1: two independent DNA-binding domains recognize two DNA consensus sequences with a common G-rich core. Mol. Cell. Biol., 14, 1786–1795.[Abstract/Free Full Text]

  24. Shields, J.M. and Yang, V.W. (1998) Identification of the DNA sequence that interacts with the gut-enriched Kruppel-like factor. Nucl. Acids Res., 26, 796–802.[Abstract/Free Full Text]

  25. Molnar, A. and Georgopoulos, K. (1994) The Ikaros gene encodes a family of functionally diverse zinc finger DNA-binding proteins. Mol. Cell. Biol., 14, 8292–8303.[Abstract/Free Full Text]

  26. Hromas, R., Collins, S.J., Hickstein, D., Raskind, W., Deaven, L.L., O'Hara, P., Hagen, F.S. and Kaushansky, K. (1991) A retinoic acid-responsive human zinc finger gene, MZF-1, preferentially expressed in myeloid cells. J. Biol. Chem., 266, 14183–14187.[Abstract/Free Full Text]

  27. Huang, J.H. and Liao, W.S. (1994) Induction of the mouse serum amyloid A3 gene by cytokines requires both C/EBP family proteins and a novel constitutive nuclear factor. Mol. Cell. Biol., 14, 4475–4484.[Abstract/Free Full Text]

  28. Fujimoto, K., Kyo, S., Takakura, M., Kanaya, T., Kitagawa, Y., Itoh, H., Takahashi, M. and Inoue, M. (2000) Identification and characterization of negative regulatory elements of the human telomerase catalytic subunit (hTERT) gene promoter: possible role of MZF-2 in transcriptional repression of hTERT. Nucl. Acids Res., 28, 2557–2562.[Abstract/Free Full Text]

  29. Nuthall, H.N., Moulin, D.S., Huxley, C. and Harris, A. (1999) Analysis of DNase-I-hypersensitive sites at the 3' end of the cystic fibrosis transmembrane conductance regulator gene (CFTR). Biochem. J., 341 (Pt 3), 601–611.

  30. Nuthall, H.N., Vassaux, G., Huxley, C. and Harris, A. (1999) Analysis of a DNase I hypersensitive site located –20.9 kb upstream of the CFTR gene. Eur. J. Biochem., 266, 431–443.[ISI][Medline]

  31. Phylactides, M., Rowntree, R., Nuthall, H., Ussery, D., Wheeler, A. and Harris, A. (2002) Evaluation of potential regulatory elements identified as DNase I hypersensitive sites in the CFTR gene. Eur. J. Biochem., 269, 553–559.[ISI][Medline]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
13/17/1933    most recent
ddh200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (1)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Ulatowski, L. M.
Right arrow Articles by Drumm, M. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ulatowski, L. M.
Right arrow Articles by Drumm, M. L.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?