Skip to main content
Download PDF
  • Original Articles
  • Published:

In Vivo Analysis of DNase I Hypersensitive Sites in the Human CFTR Gene

Molecular Medicine volume 5, pages 211–223 (1999)Cite this article

Abstract

Background

The cystic fibrosis transmembrane conductance regulator gene (CFTR) shows a complex pattern of expression. The regulatory elements conferring tissue-specific and temporal regulation are thought to lie mainly outside the promoter region. Previously, we identified DNase I hypersensitive sites (DHS) that may contain regulatory elements associated with the CFTR gene at −79.5 and at −20.5 kb with respect to the ATG and at 10 kb into the first intron.

Materials and Methods

In order to evaluate these regulatory elements in vivo we examined these DHS in a human CFTR gene that was introduced on a yeast artificial chromosome (YAC) into transgenic mice. The 310 kb human CFTR YAC was shown to restore the phenotype of CF-null mice and so is likely to contain most of the regulatory elements required for tissue-specific expression of CFTR.

Results

We found that the YAC does not include the −79.5 kb region. The DHS at −20.5 kb is present in the chromatin of most tissues of the transgenic mice, supporting its non-tissue-specific nature. The DHS in the first intron is present in a more restricted set of tissues in the mice, although its presence does not show complete concordance with CFTR expression. The intron 1 DHS may be important for the higher levels of expression found in human pancreatic ducts and in lung submucosal glands.

Conclusion

These data support the in vivo importance of these regulatory elements.

Introduction

Transcription of the cystic fibrosis transmembrane conductance regulator (CFTR) gene (1) is tightly regulated both spatially and temporally. Its expression is limited primarily to specific epithelial cell types within the respiratory system, the small intestine and colon, pancreas and gall bladder, kidney, genital ducts, and sweat glands (2–6). In the postnatal airway, highest levels of CFTR mRNA are detected in the serous portion of submucosal glands with low levels in the surface epithelium (6). In the intestine, CFTR mRNA is seen primarily in the crypt cells with gradients of expression decreasing along the crypt-villus axis and from the small intestine to the large intestine (5). CFTR mRNA and protein are found in the ductal epithelium of the pancreas and gall bladder and possibly also in pancreatic acini (4). Temporal regulation of CFTR transcription is seen in the airway where abundant CFTR transcription throughout the epithelium of the fetus decreases substantially after birth (7,8). CFTR expression in the reproductive tract is also cyclical in some species (9,10).

Relatively little is known of the mechanisms underlying this strict cellular control of CFTR expression, and there is no consistent evidence for tissue-specific control elements in the promoter (11–13). We searched for regulatory elements in the CFTR genomic DNA outside the immediate promoter region by looking for DNase I hypersensitive sites (DHS) that are often associated with sequence elements involved in the regulation of transcription. We identified three DHS in the CFTR gene, two in the 5′ flanking sequence (−79.5 kb and −20.5 kb with respect to the CFTR ATG), and one in intron 1 (at 185 + 10 kb, where 185 refers to the last base in exon 1) (14,15). These DHS were identified in human cell lines and cultured primary human genital duct cells that express endogenous CFTR. The sites at −79.5 and −20.5 kb showed no correlation with CFTR expression in the cell types investigated, although they may still play a role in the complex series of events involved in the regulation of CFTR transcription. The 185 + 10 kb DHS showed complete correlation with CFTR expression only being evident in cells that express CFTR and not in those that lacked CFTR expression (15,16). In vitro analysis of binding of proteins to this region by electromobility shift assays (EMSA) and DNase I footprinting revealed that some proteins that are only present in CFTR-expressing cells bound to specific elements. When assayed by transient expression in a cell line expressing endogenous CFTR, the element at 185 + 10 kb augmented reporter gene expression under the control of the CFTR promoter but had no effect in nonexpressing cells (15). In vitro evaluation of regulatory elements in a large complex gene such as CFTR can only generate limited data, primarily because chromatin-mediated effects are difficult to reproduce. Our aim here was to use transgenic mice carrying the human CFTR gene, on a yeast artificial chromosome (YAC), to evaluate whether regulatory elements at the −79.5, −20.5, and 185 + 10 kb are important for CFTR expression in vivo.

A 310 kb YAC containing the intact human CFTR gene [37AB12, (17)] has previously been introduced into transgenic mice (18). The transgenic mice were bred with the “Cambridge” CF null mice to produce mice that only express the human CFTR transgene (18). These mice had a normal phenotype, in contrast to the CF null mice, suggesting that the 37AB12 YAC contained regulatory elements to direct CFTR expression in many of the usual sites. Expression of the human transgene showed cell type-specific expression in these mice equivalent to that of the endogenous mouse gene in the intestine and salivary glands. However, in some tissues where human CFTR expression would be expected, it was not observed. In sites where there is a difference in CFTR expression patterns between mouse and humans, the transgene expression followed the mouse pattern. The YAC contains 310 kb of genomic DNA, including 200 kb spanning the coding region of the human CFTR gene and about 70 kb of upstream flanking sequence. Hence, it is likely to contain at least the −20.5 kb and 185 + 10 kb regions, though probably not the −79.5 kb region.

Here we show that the YAC does not extend as far 5′ as the −79.5 kb region and hence this DHS is not an absolute requirement for CFTR expression. We have also evaluated the presence of the DHS at −20.5 kb and 185 + 10 kb in chromatin extracted from the YAC transgenic mouse tissues and found that their appearance often correlates with CFTR expression.

Materials and Methods

Cosmid Clones for YAC Mapping

Probes (Fig. 1) were derived from the cosmid contig described in the original CFTR cloning paper (19) and were kindly donated by Dr. Joanna Rommens.

Fig. 1
figure 1

Map of the region from −80 kb to −40 kb from the start of the CFTR gene (A) and of the region spanning the −20.5 and 185 + 10 kb DHS (B). (A) The extent of cosmids cJ21, phage R14.4, and plasmids pB14E4.9 and pB14E5.3 are shown (top). EcoRI sites (marked R) and fragments are also shown (middle). Fragments used as probes for mapping the 5′ end of the YAC are indicated by bold lines and type (not all data are shown), and distances 5′ of the ATG translation start site are indicated beneath the line. The −79.5 kb DHS is also indicated. The lower section shows HindIII sites and fragments above the line (marked H) and BamHI sites and fragments below it (marked B). The probable 5′ end of the YAC is indicated by an arrow below the line. (B) The extent of the cosmid cW44 is shown (top). The middle section shows HindIII sites (marked H), with HindIII fragments used as probes indicated by bold lines and labeled in bold type. The locations of the −20.5 kb and 185 + 10 kb DHS, and exon 1 are also shown. The position of the ATG translation start site and distances from it are indicated below the line. The lower section shows EcoRI sites (marked R) and sizes of EcoRI fragments above the line and BamHI sites (marked B) and fragments below it.

Full size image

Transgenic Mice

Details of generation of the mice carrying the 37AB12 YAC and crosses with the “Cambridge” null mice have been described previously (18). The mice used in this study were homozygous for the YAC transgene and for the CF knockout. Mouse tissues were collected immediately after death and placed in liquid nitrogen for RNA extraction or processed directly for chromatin extraction. Only data from the T30 strain of transgenic mice are presented here.

Chromatin Extraction

Chromatin was extracted from transgenic mouse tissues, normal mouse ileum, and Calu3 cells as described previously (14,15,20). For most tissues, chromatin was extracted from the whole organ. For the small intestine, the samples were enriched for epithelial cells by washing the contents from the intestine, opening it by cutting along its length, and scraping the surface with a microscope slide.

RNA Extraction

Total RNA was extracted from transgenic mouse tissues, normal mouse ileum, and Calu3 cells by standard methods (21).

Polymerase Chain Reaction (PCR)

Agarose plugs containing 37AB12 YAC DNA were treated with β-agarase to release template DNA for PCR reactions. The primers used to amplify genomic and YAC DNA are shown in Table 1. Southern blot analyses of YACs were carried out by standard methods.

Table 1 Primers used for PCR
Full size table

Reverse Transcriptase-PCR (RT-PCR)

All tissues studied were evaluated for transcription of human and mouse CFTR mRNA expression by RT-PCR (14). The locations of primers used for RT-PCR are shown in Table 2 and described in detail elsewhere (22,23). The β-actin primers were β3′ATGCCATCCTGCGTCTG-GACCTGGC and β5′AGCATTTGCGGTGCGA-CATGGAGGG, producing a 607 bp fragment from mouse RNA and a 581 bp fragment from human RNA.

Table 2 Comparison of human and mouse RT-PCR primer sequencesa
Full size table

DNase I Hypersensitivity Assays

Chromatin from transgenic mouse tissues was probed for DNase I hypersensitive regions by standard methods (24). Each batch of chromatin was tested with the RA2.2 probe that detects a constitutive DHS in the human and mouse α-globin gene cluster (25). The DHS at −20.5 kb was detected with the H1.1 probe and that at 185 + 10 kb with the H4.0 probe (see Fig. 1B). The DHS at 185 + 10 kb was previously referred to as the 181 + 10 kb DHS (15).

Results

Mapping of the 5′ End of the CFTR YAC

Long-range mapping had previously shown that the YAC contained about 70 kb of upstream genomic DNA as well as the intact CFTR gene (18). PCR assays were carried out to determine whether the DHS at −79.5 kb, −20.5 kb and 185 + 10 kb were in fact present in the YAC (Table 1 shows the primers and expected products). The YAC DNA and genomic DNA were amplified in parallel for each DHS (Fig. 2A). Each PCR reaction also contained a control pair of primers that amplified a 438 bp fragment spanning exon 4 which was known to be present in the intact YAC. The PCR assays for the DHS at −20.5 and 185 + 10 kb gave the expected product with the YAC as well as the genomic DNA, indicating that these sites are present in the YAC. The PCR assay for the DHS at −79.5 kb did not give the expected product from the YAC even though the fragment was generated from the genomic DNA and the control exon 4 assay was efficient. Therefore, the CFTR YAC probably does not contain the DHS at −79.5 kb.

Fig. 2
figure 2

Mapping of the 5′ end of the CFTR YAC. (A) PCR amplification of 37AB12 YAC DNA and human genomic DNA with primers specific for the −79.5 kb region (214 bp product, primers MOU4-Ava and MOU8-Eco), the −20.5 kb region (274 bp product, primers HUG20C and HUG20D), and the 185 + 10 kb region (199 bp product, primers TSR7-PacI and TSR8-SacII). Each PCR reaction also contains primers for exon 4 (438 bp product, primers 4i-5 and 4i-3). For each set of three reactions, lane C contains the product of the no-DNA control, lane G contains the PCR products obtained using human genomic template DNA, and lane Y contains those using 37AB12 YAC template DNA. (B, C) Southern blots of 37AB12 DNA cleaved with HindIII and probed with pB14E4.9 (B) and pB14E5.3 (C). In both panels lane 1 contains 37AB12 YAC DNA; lane 2, a control YAC from outside the CFTR locus also cleaved with HindIII; and lane 3, probe DNA. In (B) lane 4 is a 1 kb ladder (Gibco BRL). The multiple bands in the two YAC lanes are probably due to cross-hybridization to repetitive elements in the human DNA.

Full size image

Final mapping data for the 5′ end of the YAC were provided by hybridization of various probes to Southern blots of YAC DNA. The pB14E4.9 probe (Fig. 1A) did not hybridize to HindIII-digested YAC DNA (Fig. 2B), whereas the pB14E5.3 probe (Fig. 1A) hybridized to the expected HindIII fragments of 1.1, 1.5, and 3 kb (Fig. 2C). (There are additional bands on the gel due to cross-hybridization with repetitive human DNA as shown by the bands also present in the lane with DNA from an unrelated YAC.) Since the YAC library was generated by partial EcoR1 digestion of human genomic DNA, the end of the genomic DNA within the YAC would be expected to lie at an EcoRI site. Thus the CFTR YAC probably extends to the EcoRI site at 58.4 kb 5′ to the CFTR ATG (Fig. 1A) and the YAC does not include the DHS at −79.5 kb.

Analysis of Transgene Expression

RT-PCR was used to detect CFTR mRNA from transgenic mouse duodenum, ileum, kidney, liver, lung, pancreas, spleen, and testis and genital ducts combined (Fig. 3). In addition, the RT-PCR was carried out on RNA from normal mouse ileum and from Calu3 cells (Calu3 is a human airway adenocarcinoma cell line that expresses endogenous CFTR). The positive Calu3 reactions verified the efficiency of the primers to amplify human CFTR cDNA (lanes labeled Cal in Fig. 3A, C, and E). The normal mouse ileum negative control reactions showed that the primers were specific for human CFTR mRNA (lanes labeled NM in Fig. 3A, C, and E). All RT-PCR reactions were accompanied by no-RNA and no-reverse transcriptase controls (Fig. 3B, D, and F). Mouse β-actin primers were used to check that RNA in each sample could be amplified by RT-PCR (Fig. 3G).

Fig. 3
figure 3

RT-PCR products generated using human CFTR-specific primers. (A, C, E) RT-PCR products generated from template RNA extracted from transgenic mouse duodenum (D), ileum (I), kidney (K), liver (Li), lung (Lu), pancreas (P), spleen (Sp), testis and genital ducts (T), Calu3 cells (Cal), and normal mouse ileum (NM). (B, D, F) A no-RNA control and no-reverse transcriptase controls for each tissue. Panels A and B show RT-PCR products generated using the primers A1R and 6S1L; C and D show products using the primers C1R and C2L; E and F show products using primers F1R and F2L. RT-PCR products were electrophoresed in a 1% agarose gel. Molecular weight (MW) is the 1 kb DNA ladder (Gibco BRL). (G) The 607 bp RT-PCR products generated from the same transgenic mouse RNA samples using the β-actin primers β-3′ and β-5′.

Full size image

Three sets of human CFTR-specific primers were used for RT-PCR, AIR/6SIL, C1R/C2L, and F1R/F2L (Table 2). The primers were derived from the human CFTR cDNA sequence but have strong homology to the equivalent mouse sequence (see Table 2). However, none of them generated a product from normal mouse ileum RNA. All products obtained using the human-specific primers on transgenic mouse RNA were therefore amplified from the human transgene. This was further validated by complete digestion of the RT-PCR products using enzymes with recognition sites only in the human and not the relevant mouse fragment.

The primers A1R and 6S1L, located in exons 1 and 6, respectively, amplified a product of the expected size in samples from transgenic duodenum, ileum, lung, and pancreas whereas no product was seen in samples from kidney, liver, spleen, and testis and genital ducts (Fig. 3A). The primers C1R and C2L, located in exons 11 and 13, respectively, amplified a product of the expected size in samples from transgenic duodenum, ileum, lung, pancreas, and testis and genital ducts whereas no product was seen in samples from kidney, liver, or spleen (Fig. 3C). The primers F1R and F2L, located in exons 19 and 24, respectively, amplified a product of the expected size in samples from transgenic duodenum, ileum, lung, pancreas, and testis and genital ducts whereas no product was seen in samples from kidney, liver, or spleen (Fig. 3E).

These results are summarized in Table 3 and show that the human CFTR transgene is expressed in duodenum, ileum, lung, and pancreas but not in kidney, liver, and spleen. Expression in the testis and genital ducts was observed with the primers C1R/C2L and F1R/F2L but not with primers A1R/6S1L.

Table 3 Summary of RT-PCR and DHS resultsa
Full size table

Detection of the DNase I Hypersensitive Site at −20.5 kb

The same tissues that had been analyzed by RT-PCR for CFTR RNA expression were also analyzed by DNase I digestion of chromatin to detect DHS. The efficiency of the DNase I digestion was confirmed by hybridizing filters carrying BamHI-cleaved DNA from each tissue with the RA2.2 probe (25) which recognizes a constitutive DHS in the mouse α-globin gene cluster.

The DHS at −20.5 kb was detected with the cW44H1.1 probe (Fig. 1B) as shown previously with chromatin extracted from human primary cells and cell lines (14). This probe hybridizes to an EcoRI restriction fragment of 4.9 kb in intact chromatin and detects a subfragment of 3.5 kb, and occasionally another subfragment of 3 kb, in chromatin treated with DNase I (14). Chromatin from transgenic mouse tissues was examined for the presence of the −20.5 kb DHS (Fig. 4). The intact 4.9 kb EcoRI restriction fragment was seen in all tissues. A 3.5 kb subfragment appeared in chromatin from kidney (Fig. 4A), colon (Fig. 4D), small intestine (Fig. 4E), and stomach (Fig. 4F). Two subfragments, of about 3.5 kb and 2.8 kb, were seen in liver (Fig. 4B). The 2.8 kb fragment could indicate the presence of another DHS in this area. A faint 3.5 kb subfragment was seen in lung (Fig. 4C) and pancreas (Fig. 4G), although its intensity did not fully correlate with the degree of DNase I digestion of the chromatin. No DHS was seen in chromatin from testis (Fig. 4H) or genital ducts (Fig. 4I) or spleen (not shown).

Fig. 4
figure 4

In vivo evaluation of the −20.5 Kb DHS. (A–I) Chromatin extracted from transgenic mouse tissues, digested with DNase I, cleaved with EcoRI, separated on 1% agarose gels. Southern blotted, and probed with cW44 H1.1 to detect the DHS at −20.5 kb. For each tissue, lane 1 shows DNA prepared from nuclei that had not been treated with DNase I and lanes 2–5 show DNA treated with increasing amounts of DNase I: lane 2, 15 units; lane 3, 30 units; lane 4, 60 units; lane 5, 120 units. Molecular weights were estimated with respect to a 1 kb DNA ladder (Gibco BRL).

Full size image

Detection of the DNase I Hypersensitive Site at 185 + 10 kb

The DHS in intron 1 at 185 + 10 kb was detected with the cW44 H4.0 probe (Fig. 1B). This probe detects a 22 kb BamHI restriction fragment in intact chromatin. It detects an intense 8 kb subfragment and fainter 10 kb and 12 kb subfragments in chromatin from CFTR-expressing human primary cells and cell lines that have been treated with DNase I (15). Chromatin from kidney, liver, lung, small intestine, pancreas, and testis was examined for the presence of the 185 + 10 kb DHS (Fig. 5). Two BamHI fragments of 22 kb and 13 kb were seen in all tissues. The extra 13 kb BamHI fragment could arise from a partial copy of the YAC that had integrated into the genome of the T30 mice in addition to the intact YAC. This was supported by Southern blotting data showing that the transgenic mice had more copies of the 5′ than the 3′ end of the CFTR gene, even though there were definitely intact copies of the YAC present (18). A subfragment of 8 kb was seen in kidney (Fig. 5A), liver (Fig. 5B), and small intestine (Fig. 5D). The presence of this 8 kb fragment in intestinal chromatin that had not been treated with DNase I (Fig. 5D, lane 1) was due to the presence of endogenous DNase in this tissue sample; it was not seen in other intestinal chromatin samples (not shown). An additional subfragment of 10 kb was seen in kidney and small intestine. Small intestine also exhibited another subfragment of 4.8 kb. These extra subfragments suggest that there are additional DHS in intron 1 at 185 + 12 kb (indicated by the 10 kb subfragment) and 185 + 6.8 kb (indicated by the 4.8 kb subfragment). No DHS were observed in lung (Fig. 5C), pancreas (Fig. 5E), or testis (Fig. 5F). A summary of the DHS results is presented in Table 3.

Fig. 5
figure 5

In vivo evaluation of the 185 + 10 kb DHS. (A–F) Chromatin extracted from transgenic mouse tissues, digested with DNase I, cleaved with BamHI, electrophoresed on a 1% agarose gel, Southern blotted, and probed with cW44 H4.0 to detect the DHS in intron 1 of the CFTR gene, at 185 + 10 kb. For each tissue, lane 1 shows DNA prepared from nuclei that had not been treated with DNase I and lanes 2–5 show DNA treated with increasing amounts of DNase I: lane 2, 15 units; lane 3, 30 units; lane 4, 60 units; lane 5, 120 units. Molecular weights were estimated with respect to a 1 kb DNA ladder (Gibco BRL).

Full size image

Discussion

The data presented here provide the first in vivo analysis of potential regulatory elements located at −20.5 kb with respect to the ATG and at 10 kb into the first intron of the human CFTR gene. By using a YAC carrying the intact human CFTR gene, introduced into CF knockout mice, we were able to evaluate the involvement of these regulatory elements in CFTR gene expression. If regulatory elements at these sites are involved in driving CFTR gene expression, then the DHS would be expected to appear in chromatin extracted from tissues of these transgenic mice that express the CFTR gene. The absence of the DHS would indicate that they are not needed for expression in these tissues. However, the DHS might still appear even if they are not important in expression. The fact that two independent transgenes with the YAC carrying the intact human CFTR gene (T30 and T57) were both able to complement the null mice and gave roughly the same expression in all tissues examined (18) argues against position effects of neighboring chromatin playing a significant role in CFTR expression and the DHS in these mice.

The DHS that was previously detected at −79.5 kb was not present on the YAC. As the YAC expresses well in the small intestine and salivary glands of transgenic mice, this indicates that the −79.5 kb DHS is not important for expression in these tissues. However, it could be important for expression in such tissues as the kidney, and Brunner’s glands (18), where expression of the gene from the YAC is not found. The two regions at −20.5 kb and 185 + 10 kb in the YAC were both able to form DHS in some tissues in the transgenic mice. This indicates that the mouse transcriptional machinery is able to interact with these sites and that they are probably important for CFTR gene expression.

An initial investigation of human CFTR transgene expression in the CF knockout mice by mRNA in situ hybridization (18) showed abundant expression in the crypts of the small intestine with a decreasing gradient from the duodenum to the colon, as well as in salivary glands within the ductal epithelium and mucous acini. No CFTR transgene expression was seen in the pancreas, lung, or Brunner’s glands. In the present study human CFTR mRNA was detected by RT-PCR in the small intestine, lung, pancreas, testis and genital ducts but not in the kidney, liver, or spleen. The results for the lung and pancreas are of interest as it was not clear from the in situ hybridization data whether there was any expression of the human transgene in these tissues. The positive results for the lung and pancreas by RT-PCR could be due to the increased sensitivity of RT-PCR over mRNA in situ hybridization (18).

We consistently detected human CFTR mRNA in testis and genital ducts using primers for exons 11 to 13 and exons 19 to 24 of the transgene, but no product was obtained using primers in exons 1 and 6. This could be because mRNA transcribed from the transgene in testis and genital ducts lacks the sequence for primer binding in exon 1, exon 6, or both. Testis-specific alternative splicing of CFTR mRNA and the use of testis-specific transcription start sites have been reported in mice and humans (26–29). To date, only events that insert novel sequences into the CFTR mRNA have been reported, and none that result in deletion of exons. However, alternative splicing causing exon loss from the human CFTR transcript has been reported in nasal epithelial cells, among other cell types (22). Such testis-specific alternative splicing of the human CFTR gene might be relevant to the male infertility associated with CF in humans, although CF mice are fertile. However, the defect is generally thought to be within the genital duct epithelium rather than the testis itself.

The −20.5 kb DHS has been previously seen in a wide variety of human epithelial cell lines and primary cells irrespective of their CFTR expression status (14), hence we predict that it might contain a regulatory element involved in basal CFTR expression. As such, it would be expected to appear in a wide variety of tissues in the YAC transgenic mice. Our results confirm this prediction, as the −20.5 kb DHS was seen to some extent in chromatin extracted from all the transgenic mice tissues analyzed, with the exception of the testis, genital ducts, and spleen. The intensity of the appearance of the DHS cannot be correlated with CFTR expression levels in these analyses since chromatin has been extracted from whole tissue (or in the case of the small intestine, enriched for surface epithelial cells), but we have no information on what percentage of cells within a tissue express the CFTR gene.

The presence of the 185 + 10 kb DHS was previously shown to correlate completely with CFTR expression in a range of human cell lines and primary cells (15), hence it might be predicted to show a more restricted pattern of appearance in the transgenic mice. This is indeed the case, with this DHS being evident only in chromatin extracted from small intestine, kidney, and liver, and not seen in lung, pancreas, or testis. The presence of the 185 + 10 kb DHS in small intestine is consistent with its involvement in CFTR expression in the Caco2 colon carcinoma cell line, as shown by transient reporter gene assays (15).

The 185 + 10 kb DHS is not present in chromatin extracted from transgenic mouse pancreas. In humans, CFTR expression levels are very high in pancreatic duct cells (2,4,5) and the 185 + 10 kb DHS has been observed in a human pancreatic adenocarcinoma cell line (Capan1) that expressed low levels of CFTR (15). However, mice show very low levels of pancreatic CFTR expression (9) and the transgenic mice show low levels of human transgene expression in the pancreas (undetectable by mRNA in situ hybridization). It is possible that only a small percentage of cells (for example, certain ductal epithelial cells) in the transgenic mouse pancreas express the human CFTR gene at a high level and that they contribute too small a fraction to the total tissue chromatin for the DHS to be detected. Alternatively, the −20.5 kb DHS, which is present in pancreatic chromatin from the transgenic mice, might be involved in the low-level expression detected in the transgenic mouse pancreas by RT-PCR, while the 185 + 10 kb site might contain a regulatory element involved in the high levels of CFTR gene expression normally seen in human pancreatic duct cells. The lack of the 185 + 10 kb site in the mouse chromatin (probably due to the lack of appropriate transcription factors) would then explain the lack of high-level expression in the mouse pancreas.

Similarly, the 185 + 10 kb DHS is not present in chromatin extracted from transgenic mouse lung. The CFTR gene is normally expressed at low levels in the surface epithelium of the adult human lung, with higher levels in the serous portion of submucosal glands (6). Mice express low levels of CFTR mRNA in the surface epithelium and have very few submucosal glands restricted to the tracheal submucosa (9). Low levels of human CFTR mRNA were detected by RT-PCR in lungs from the transgenic mice. The presence of the −20.5 kb DHS, but not the 185 + 10 kb DHS, in chromatin extracted from the transgenic lung tissue might be accounted for by the −20.5 kb DHS being responsible for the low level of expression in the epithelial cells, while the 185 + 10 kb DHS is important for higher expression levels in the human submucosal glands.

It is not clear why the 185 + 10 kb DHS is seen in chromatin extracted from liver, since neither mouse nor human CFTR genes are expressed at significant levels in liver and no human transgene CFTR mRNA is detected in liver RNA. CFTR expression is seen at high levels in gall bladder epithelium but gall bladders were removed from the transgenic mouse tissues prior to extraction of mRNA and chromatin. Similarly, the 185 + 10 kb DHS was present in chromatin extracted from transgenic mouse kidney but no human CFTR mRNA was detected in the transgenic mouse kidney by RT-PCR. However, endogenous CFTR mRNA is expressed in the normal human and mouse kidneys. These data suggest that, although the 185 + 10 kb DHS appears, the human CFTR transgene is unable to respond to some of the mouse transcription factors that regulate CFTR expression in the kidney.

The results from the testis and genital ducts, in which the human YAC CFTR transgene was expressed but neither the −20.5 kb or 185 + 10 kb DHS were evident, suggest that neither of these regions is important in testis or genital duct expression. Human testis exhibits low and diffuse levels of CFTR expression, which does not appear to be associated with any specific cell type (30). In contrast, CFTR is expressed at higher levels in the epithelium of the seminiferous tubules of mouse testis at all stages of spermatogenesis (9). CFTR is also known to show specific temporal regulation of expression in the male and female reproductive systems in rodents (10). Control of expression in these tissues may be operating through different regulatory elements.

In conclusion, the data provide evidence for the in vivo importance of the DHS at −20.5 kb as this site is present in chromatin from most of the transgenic mouse tissues. The 185 + 10 kb DHS has a more restricted distribution and is not found in lung or pancreatic tissues of the transgenic mice. The lack of formation of this DHS might be responsible for the very low levels of CFTR expression in the mouse lung and pancreas, which would indicate that the 185 + 10 kb DHS is important for the higher levels of expression in these human tissues. Finally, neither DHS is present in chromatin from testis or genital ducts, and expression in these tissues may be operating through other regulatory elements that have not been identified but that are present on the YAC.

However, the results also illustrate the problems of using the mouse as a model for investigation of the subtleties of human CFTR tissue-specific regulation. The CFTR gene shows a divergent pattern of expression in rodents (9) and humans (2–6). The data presented here show that the murine transcriptional machinery can interact with the −20.5 and 185 + 10 kb DHS and give adequate levels of CFTR mRNA expression in many tissues. However, lack of transgene expression in certain other sites, such as the kidney and Brunner’s glands, might result from the incompatibilities between the mouse and human transcriptional control mechanisms for CFTR. Analysis of the promoter regions of the human, mouse, and rat CFTR genes has shown them to be highly divergent (31,32). Although some putative regulatory elements are loosely conserved, most of them are species-specific. A more extensive cross-species comparison of the CFTR promoter region (33) from eight mammalian species including human, monkey, cow, rabbit, mouse, and rat suggests that the cow and rabbit might be more suitable models in which to analyze human CFTR gene regulation in vivo. Recent advances in cloning technology (34–36) will enable this type of analysis to be carried out in ovine models, which are known to show a pattern of CFTR gene expression and regulation that is very similar to that of the human gene (37,38).

References

  1. Riordan JR, Rommens JM, Kerem, B-S, et al. (1989) Identification of the cystic fibrosis gene: cloning and characterisation of complementary DNA. Science 245: 1066–1073.

    Article  CAS  Google Scholar 

  2. Crawford IC, Maloney PC, Zeitlin PL, et al. (1991) Immunocytochemical localization of the cystic fibrosis gene product CFTR. Proc. Natl. Acad. Sci. U.S.A. 88: 9262–9266.

    Article  CAS  Google Scholar 

  3. Denning GM, Ostedgaard LS, Cheng SH, Smith AE, Welsh MJ. (1992) Localization of the cystic fibrosis transmembrane conductance regulator in chloride secretory epithelia. J. Clin. Invest. 89: 339–349.

    Article  CAS  Google Scholar 

  4. Hyde K, Reid CJ, Tebbutt SJ, Weide L, Hollingsworth MA, Harris A. (1997) The cystic fibrosis transmembrane conductance regulator as a marker of human pancreatic duct development and differentiation. Gastroenterology 113: 914–919.

    Article  CAS  Google Scholar 

  5. Strong TV, Boehm K, Collins FS. (1994) Localization of cystic fibrosis transmembrane conductance regulator mRNA in the human gastrointestinal tract by in situ hybridization. J. Clin. Invest. 93: 347–354.

    Article  CAS  Google Scholar 

  6. Engelhart JF, Yankaskas JR, Ernst SA, et al. (1992) Submucosal glands are the predominant site of CFTR expression in the human bronchus. Nat. Genet. 2: 240–247.

    Article  Google Scholar 

  7. Harris A, Chalkley G, Goodman S, Coleman L. (1991) Expression of the cystic fibrosis gene in human development. Development 113: 305–310.

    PubMed  CAS  Google Scholar 

  8. Trezise AEO, Chambers JA, Wardle CJ, Gould S, Harris A. (1993) Expression of the cystic fibrosis gene in human foetal tissues. Hum. Mol. Genet. 2: 213–218.

    Article  CAS  Google Scholar 

  9. Trezise AE, Buchwald M. (1991) In vivo cell-specific expression of the cystic fibrosis transmembrane conductance regulator. Nature 353: 434–437.

    Article  CAS  Google Scholar 

  10. Trezise AE, Linder CC, Grieger D, et al. (1993) CFTR expression is regulated during both the cycle of the seminiferous epithelium and the oestrous cycle of rodents. Nat. Genet. 3: 157–164.

    Article  CAS  Google Scholar 

  11. Chou J-L, Rozmahel R, Tsui L-C. (1991) Characterization of the promoter region of the cystic fibrosis transmembrane conductance regulator gene. J. Biol. Chem. 266: 24471–24476.

    PubMed  CAS  Google Scholar 

  12. Koh J, Sferra TJ, Collins FS. (1993) Characterization of the cystic fibrosis transmembrane conductance regulator promoter region. J. Biol. Chem. 268: 15912–15921.

    PubMed  CAS  Google Scholar 

  13. Yoshimura K, Nakamura N, Trapnell BC, et al. (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.

    PubMed  CAS  Google Scholar 

  14. Smith AN, Wardle CJC, Harris A. (1995) Characterization of DNAse 1 hypersensitive sites in the 120 kb 5′ to the CFTR gene. Biochem. Biophys. Res. Commun. 211: 274–281.

    Article  CAS  Google Scholar 

  15. Smith AN, Barth ML, McDowell TL, et al. (1996) A regulatory element in intron 1 of the cystic fibrosis transmembrane conductance regulator gene. J. Biol. Chem. 271: 9947–9954.

    Article  CAS  Google Scholar 

  16. McDowell TL, Smith DJ, Huthall HN, Moulin DS, Harris A. (Unpublished) Role of a regulatory element in intron 1 of the CFTR gene in airway epithelium.

  17. Anand R, Ogilvie DJ, Butler R, et al. (1991) A yeast artificial chromosome contig encompassing the cystic fibrosis locus. Genomics 9: 124–130.

    Article  CAS  Google Scholar 

  18. Manson AL, Trezise AEO, MacVinish LJ, et al. (1997) Complementation of CF null mice with a human CFTR transgene. EMBO J. 16: 4238–4249.

    Article  CAS  Google Scholar 

  19. Rommens JM, Iannuzzi MC, Kerem B-S, et al. (1989) Identification of the cystic fibrosis gene: chromosome walking and jumping. Science 245: 1059–1065.

    Article  CAS  Google Scholar 

  20. Becker P, Renkawitz R, Schutz G. (1984) Tissue-specific DNAse I hypersensitive sites in the 5′-flanking sequences of tryptophan oxygenase and the tyrosine aminotransferase genes. EMBO J. 3: 2015–2020.

    Article  CAS  Google Scholar 

  21. Chirgwin JM, Przybyla AE, Macdonald RJ, Rutter WJ. (1979) Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 18: 5294–5299.

    Article  CAS  Google Scholar 

  22. Hull J, Shackleton S, Harris A. (1994) Analysis of mutations and alternative splicing patterns in CFTR mRNA derived from nasal epithelial cells. Hum. Mol. Gen. 3: 1141–1146.

    Article  CAS  Google Scholar 

  23. Chalkley GC, Harris A. (1991) Detection of mutations and polymorphisms in the CF gene by analysis lymphocyte mRNA. J. Med. Genet. 28: 777–780.

    Article  CAS  Google Scholar 

  24. Higgs DR, Wood WG, Jarman AP, et al. (1990) A major positive regulatory region located far upstream of the human alpha-globin gene locus. Genes Dev. 4: 1588–1601.

    Article  CAS  Google Scholar 

  25. Vyas P, Vickers MA, Simmons DL, Ayyub H, Craddock CF, Higgs DR. (1992) Cis-acting sequences regulating expression of the human alpha-globin cluster lie within constitutively open chromatin. Cell 69: 781–793.

    Article  CAS  Google Scholar 

  26. Trezise AEO, Buchwald M, Higgins CF. (1993) Testis-specific, alternative splicing of rodent CFTR mRNA. Hum. Mol. Genet. 2: 801–810.

    Article  CAS  Google Scholar 

  27. Delaney SJ, Rich DP, Thomson SA, et al. (1993) Cystic fibrosis transmembrane conductance regulator splice variants are not conserved and fail to produce chloride channels. Nat. Genet. 4: 426–431.

    Article  CAS  Google Scholar 

  28. Delaney SJ, Koopman P, Lovelock PK, Wainwright BJ. (1994) Alternative splicing of the first nucleotide binding fold of CFTR in mouse testes is associated with specific stages of spermatogenesis. Genomics 20: 517–518.

    Article  CAS  Google Scholar 

  29. White NL, Higgins CF, Trezise AE. (1998) Tissue-specific in vivo transcription start sites of the human and murine cystic fibrosis genes. Hum. Mol. Genet. 7: 363–369.

    Article  CAS  Google Scholar 

  30. Tizzano EF, Silver MM, Chitayat D, Benichou JC, Buchwald M. (1994) Differential cellular expression of cystic fibrosis transmembrane regulator in human reproductive tissues. Clues for the infertility in patients with cystic fibrosis. Am. J. Pathol. 144: 906–914.

    PubMed  PubMed Central  CAS  Google Scholar 

  31. Denamur E, Chehab FF. (1994) Analysis of the mouse and rat CFTR promoter regions. Hum. Mol. Genet. 3: 1089–1094.

    Article  CAS  Google Scholar 

  32. Vuillaumier S, Dixmeras I, Messai H, et al. (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.

    Article  CAS  Google Scholar 

  33. Vuillaumier S, Kaltenboeck B, Lecointre G, Lehn P, 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.

    Article  CAS  Google Scholar 

  34. Campbell KH, McWhir J, Ritchie WA, Wilmut I. (1996) Sheep cloned by nuclear transfer from a cultured cell line. Nature 380: 64–66.

    Article  CAS  Google Scholar 

  35. Wilmut I, Schnieke AE, McWhir J, Kind AJ, Campbell KH. (1997) Viable offspring derived from fetal and adult mammalian cells. Nature 385: 810–813.

    Article  CAS  Google Scholar 

  36. Cibelli JB, Stice SL, Golueke PJ, et al. (1998) Cloned transgenic calves produced from nonquiescent fetal fibroblasts. Science 280: 1256–1258.

    Article  CAS  Google Scholar 

  37. Tebbutt SJ, Wardle CJC, Hill DF, Harris A. (1995) Molecular analysis of the ovine CFTR gene. Proc. Natl. Acad. Sci. U.S.A. 92: 2293–2297.

    Article  CAS  Google Scholar 

  38. Harris A. (1997) Towards an ovine model of cystic fibrosis. Hum. Mol. Genet. 13: 2191–2193.

    Article  Google Scholar 

Download references

Acknowledgments

We are grateful to Johanna Rommens for cosmids and plasmids pB14E.9 and pB14E5.3 and to Erick Denamur for helpful discussions. This work was supported by the Cystic Fibrosis Trust, the Wellcome Trust, and AFLM. D.S.M. was a recipient of a Wellcome Trust prize studentship, H.N.N. of a Cystic Fibrosis Trust studentship.

Author information

Authors and Affiliations

  1. Paediatric Molecular Genetics, Institute of Molecular Medicine, Oxford University, John Radcliffe Hospital, Oxford, OX3 9DS, UK

    Danielle S. Moulin, Hugh N. Nuthall, David J. Smith &  Ann Harris

  2. Section of Molecular Genetics, Division of Biomedical Sciences, Imperial College School of Medicine, London, UK

    Ania L. Manson & Clare Huxley

Authors
  1. Danielle S. Moulin
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Ania L. Manson
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Hugh N. Nuthall
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. David J. Smith
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Clare Huxley
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Ann Harris
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Ann Harris.

Additional information

Communicated by D. Weatherall.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Moulin, D.S., Manson, A.L., Nuthall, H.N. et al. In Vivo Analysis of DNase I Hypersensitive Sites in the Human CFTR Gene. Mol Med 5, 211–223 (1999). https://doi.org/10.1007/BF03402118

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF03402118

Keywords

Molecular Medicine

ISSN: 1528-3658