Human Molecular Genetics, 2002, Vol. 11, No. 26 3257-3265
© 2002 Oxford University Press
A gene fusion method to screen for regulatory effects on gene expression: application to the LDL receptor
Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX 77030, USA
Received July 15, 2002; Accepted October 10, 2002
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
We have evaluated the feasibility of using fusion genes to link transcriptional promoters of biomedical interest to reporter genes to screen for pharmacological or genetic regulatory effects. Using gene targeting, we generated two lines of embryonic stem (ES) cells expressing human
-fetoprotein (hAFP) under control of the endogenous promoter for the LDL receptor (LDLR). In one line, hAFP was introduced into the first intron after the translational start codon; in the other, hAFP was positioned in the 3'-untranslated region leaving the potential for expression of LDLR intact. In both lines, an internal ribosome entry site (IRES) was included to facilitate translation. Readily measurable levels of hAFP were found in the medium with both targeted ES cell lines, compared with undetectable levels with the starting cell line. The expectation that the level of hAFP would reflect the steady-state level of mRNA for the fusion transcript and secondarily transcriptional control of LDLR was confirmed by correlating hAFP levels with the abundance of LDLR and fusion transcripts. We also generated mice carrying the LDLR–hAFP fusion in the 3'-untranslated region and these mice produced detectable levels of hAFP in serum. Levels of hAFP in culture medium and in serum were increased by simvastatin, a drug known to up-regulate LDLR. These ES cell clones and mice are suitable for pharmacological and genetic screening to detect effects on expression of LDLR. The data demonstrate the feasibility of using gene fusions to screen for drugs and genetic factors that affect expression of a wide variety of genes of interest. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Gene targeting allows for the precise introduction of predetermined modifications into the chromosomal genes. This technology has important and wide-ranging applications in studies of gene function and the development of animal models for human genetic diseases (1,2). In this study, we have used gene targeting to generate a screening system to detect pharmacological, genetic and other effects on a gene of interest. In our screening system, we introduced a reporter/marker under the control of an endogenous promoter by gene targeting and we will refer to this strategy as gene fusion methodology. Mice carrying gene fusions have significant advantages in screening for regulatory effects on a gene promoter. Using gene targeting, the reporter cassette can be precisely introduced into the endogenous locus and the reporter level is likely to accurately reflect normal in vivo control of gene expression. In addition, it is well known that regulatory elements can be located near the promoter or at great distances upstream or downstream, including the sequences within large introns (3). In such instances, constructs can be designed to insert the reporter cassette into various locations in a gene, including in the 3'-untranslated region, so that physiological regulation of the reporter sequence can be achieved.
In this study, we chose human -fetoprotein (hAFP) as the reporter gene. AFP is a major serum protein during fetal development and is produced primarily by the fetal liver and yolk sac. Its synthesis decreases markedly after birth and only a trace amount is present in the serum of adults (4). Human AFP is particularly attractive as a reporter, because it is secreted and easily quantified. This allows the analysis of the kinetics of gene expression without requiring that animals be sacrificed or that tissue culture cells be harvested. Analysis of mice receiving an adenoviral vector expressing hAFP delivered to the airway showed that serum levels of hAFP were a useful marker reflecting its tissue expression (5). The ubiquitous over-expression of hAFP under control of the ß-actin promoter in transgenic mice did not cause any abnormality (6). In addition, modest serum levels of hAFP over a wide range in pregnant women with normal or abnormal fetuses do not have any harmful effects on the mother (4,7).
We have used the promoter for the LDL receptor (LDLR) as a model system to screen for pharmacological and other effects on gene expression. LDLR mediates the clearance of lipoproteins containing apolipoprotein E or apoB-100 from plasma. This receptor-mediated pathway supplies cholesterol to cells in the liver and other organs (8,9). A high-cholesterol diet reduces hepatic uptake of LDL by suppressing production of LDLR and this is accompanied by increased levels of atherogenic LDL (10). Conversely, cholesterol-lowering drugs, such as HMG CoA reductase inhibitors, block cholesterol synthesis, thereby reducing intracellular levels of cholesterol and increasing production of LDL receptors (8). The fusion gene model used in this study would be an effective system to efficiently screen for pharmacological, dietary, genetic or other factors regulating the expression level of LDLR. This gene fusion strategy could be used to study a wide range of biological systems.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Generation of ES cells with hAFP within intron 1 of LDLR
The LDLR gene includes 18 exons with the translational start codon in exon 1 (11). The hAFP cDNA was introduced into intron 1 of the LDLR locus by homologous recombination in embryonic stem (ES) cells. A replacement type vector (Fig. 1) with selection markers flanked by two loxP sites was constructed to insert a splice acceptor site and the hAFP cDNA with a poly adenylation site (pA) into intron 1 of the LDLR gene. An internal ribosome entry site (IRES) (12) was inserted between the splice acceptor site and the coding region for hAFP to allow the reporter gene to be translated from a fusion transcript initiated at the LDLR promoter. Thus, changes in transcription of the promoter for LDLR would be expected to be reflected by changes in the level of hAFP secreted into the medium as is generally assumed for knock-in strategies. The targeting vector with Neo-TK markers was electroporated into AB2.2 ES cells derived from the 129/SvEv mouse strain. Correctly targeted G418-resistant clones were identified by Southern blot analysis using both 5'- and 3'-flanking probes (Fig. 1C) to ensure that homologous recombination had occurred as desired. After screening 192 G418-resistant colonies, we identified 28 targeted clones that we designated as 5'-targeted clones. To eliminate any effect from the promoters of Neo-TK cassette, a plasmid expressing Cre recombinase was transiently transfected into 5'-targeted clones and the Neo-TK markers were deleted via recombination of the two loxP sites. We confirmed the recombination by Southern blotting analysis using both 5'- and 3'-flanking probes (Fig. 1C) as well as a TK probe (data not shown).
|
Generation of ES cells and mice with hAFP in the 3'-untranslated region of LDLR
We also prepared a construct to insert the hAFP reporter cassette into the 3'-untranslated region of the LDLR transcript (Fig. 2). In this 3'-targeting gene fusion strategy, the potential for expression of endogenous LDLR would remain intact. A 4.6 kb EcoRI fragment containing exons 13–18 of LDLR was cloned into the Neo-TK vector for homologous recombination. Similar to the strategy for 5'-targeting, an IRES–hAFP–pA cassette was placed downstream of the fragment containing exons 13–18, but a splice acceptor site was not included for the 3'-targeting construct. This targeting vector was linearized in the middle of the 4.6 kb genomic fragment using SalI endonuclease and the construct was electroporated into ES cells. Both 5'- and 3'-flanking probes were used to identify the correctly targeted clones by Southern blotting analysis. The targeted clones were further transfected with a Cre-expressing plasmid and the recombinant clones with deletion of the Neo-TK cassette were identified by Southern blotting analysis using two flanking probes (Fig. 2C). To generate the mice with the LDLR/hAFP fusion allele, the 3'-targeted ES cells after treatment with Cre were injected into C57BL/6J mouse blastocysts. Male chimeras were bred with C57BL/6J females and germ-line transmission of the LDLR/hAFP fusion transgene was obtained. Mice heterozygous and homozygous for the fusion allele were identified (Fig. 2D) and these mice would be expected to have approximately normal expression of LDLR protein.
|
Expression of hAFP under control of the endogenous LDLR promoter in both 5'- and 3'-targeted ES cells
To examine the expression of hAFP under the control of the endogenous LDLR promoter, multiple individual targeted ES clones were cultured for 1–2 days and hAFP was measured in the medium. Both the 5'- and 3'-targeted lines expressed readily detectable hAFP in the medium with relatively higher expression detected in the 5'-targeted lines (Fig. 3). As expected, the hAFP level in the medium from the parental ES cells was undetectable. Thus, the specific expression of hAFP under the control of the endogenous LDLR promoter was successfully established in the ES cells.
|
Drug effect on the expression of hAFP in targeted ES cells
Next, we assessed if drugs and metabolic alterations known to alter the LDLR promoter activity would change expression of hAFP in the targeted cells. We first examined the response to lipoprotein deficient serum (LPDS), an alteration known to up-regulate the transcription of LDLR. The 5'-targeted cells (after Cre) were cultured in medium with fetal bovine serum (FBS) or LPDS. Higher levels of hAFP were found in the LPDS medium (Fig. 4A), as would be expected if expression of hAFP was under the control of the LDLR promoter. We also tested the effect of a statin drug on expression of hAFP by culturing the 5'-targeted cells (after Cre) in medium with simvastatin, a drug known to inhibit HMG CoA reductase and secondarily to enhance transcription of LDLR. The levels of hAFP were measured after 1–2 days of cell culture at different concentrations of drug in medium with LPDS. The levels of hAFP were increased in a dose-dependant manner with the greatest induction at 2.0 µM simvastatin (Fig. 4A). Simvastatin was cytotoxic to ES cells at higher concentrations and cells died in medium with a concentration of 5 µM. The 3'-targeted cells were also used to test the effect of simvastatin and again expression of hAFP was significantly induced in the presence of 2.0 µM of simvastatin for 2 days (Fig. 4C, bottom). Levels of hAFP remained undetectable in the medium when the parental ES cells were cultured with simvastatin (data not shown).
|
Although expression of hAFP would be expected on general principles to correlate with changes in the transcriptional activity of the LDLR promoter, this correlation was examined directly in heterozygous cells (Fig. 4B). Transcripts with the LDLR coding sequences were examined using RT–PCR with primers in exons 1 and 4. For the 5'-mutation, the PCR product is likely to be derived exclusively from the native rather than the fusion transcript in heterozygous cells since the expression cassette for hAFP lies between exons 1 and 4. For the 3'-mutation, the PCR product for LDLR is likely to be derived both from the native and the fusion transcript. The abundance of the RT–PCR for LDLR was
2-fold higher for the 3'-mutation compared with the 5'-mutation, which is compatible with this interpretation. Whatever the case, relative changes in hAFP transcript and hAFP protein should parallel changes in abundance of the LDLR transcript. This was found to be the case for both the 5- and 3'-mutations, as shown in Fig. 4B. In addition, we compared the 5'- and 3'-targeted lines with or without removal of the Neo-TK cassette from the targeted allele to determine if the promoters from Neo-TK would affect the responses of the LDLR promoter to drug. We found similar drug effects in the cells with or without Neo-TK in both 5'-and 3'-targeted lines (Fig. 4C), suggesting that the Neo-TK promoters had no obvious effect on the expression of hAFP from the fusion transgene.
Regulation of hAFP expression in mice carrying LDLR–hAFP fusion allele
Germ-line transmission was obtained for the 3'-targeted ES cells and mice carrying the LDLR–hAFP fusion allele were viable and fertile with no readily observed phenotypic effects in heterozygotes or homozygotes (Fig. 2D). In heterozygous mice, the hAFP level in the blood was 300 µg/ml which was 100-fold higher than the blank value for wild-type mice (Fig. 5A), indicating that hAFP was expressed in vivo under the control of the LDLR promoter. To evaluate the drug effect on hAFP expression in these mice, 7-week-old heterozygous mice were fed simvastatin daily (0.55 mg/g food). The mice were bled retro-orbitally at different time points to obtain serum for hAFP assay. After treatment with drug for 1 or 2 weeks, the hAFP level in mice was significantly increased compared with the untreated group (n=5 each group, P<0.05 for drug versus no drug). When the drug was withdrawn, the hAFP values returned to pretreatment levels. As for the cultured ES cells, the changes in hAFP levels correlated with the changes for native LDLR and fusion transcripts in mice heterozygous for the gene fusion (Fig. 5B). Finally, we tested the effect of dietary cholesterol on expression of hAFP by feeding mice (n=4 each group) a high-fat diet that contained 21% fat for 2 weeks. Compared with the group with regular chow, the levels of hAFP were lower in the high-fat group (Fig. 5C). Although the difference was not statistically significant (P=0.11), the observed difference was consistent with the expectation that a high-fat diet would result in down-regulation of the promoter activity of LDLR.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Using homologous recombination, we generated two lines of ES cells that express hAFP under the control of the endogenous LDLR promoter. We believe that expression of hAFP can be interpreted to be under the control of the LDLR promoter on general principles and experience with knock-in mutations and parallel determinations of LDLR native transcript, fusion transcript and hAFP protein supported this conclusion. Both 5'- and 3'-targeted lines increased secretion of hAFP in response to LPDS and simvastatin, two conditions expected to increase expression of LDLR. These hAFP-expressing lines allowed us to test drug effects on LDLR expression by examining the hAFP level in the medium without harvesting cells. These cell lines could be used to facilitate screening of drugs that up-regulate expression of LDLR. Insertional mutagenesis using a retroviral gene trap mutagenesis (13,14) could also be applied to these cells to search for novel genes regulating the LDLR pathway.
We also generated the mice carrying an LDLR–hAFP fusion allele from the 3'-targeted ES cells. In these mice, changes in the levels of both fusion transcript and hAFP protein again correlated with changes in the level of native LDLR transcript and hAFP in the serum was induced by simvastatin or reduced by high-fat chow (Fig. 5). The data supported the interpretation that expression of hAFP in these mice accurately reflects transcriptional control of the LDLR promoter. Therefore, we can test for direct or indirect pharmacological effects on the LDLR promoter in vivo by measuring hAFP level in the blood. Although most of the regulation detected with the gene fusions is likely to reflect transcriptional control, post-transcriptional effects are possible and must be considered, as demonstrated by the detection of translation regulation for the Snurf–Snrpn locus in mice using a similar gene fusion strategy (15). In addition, the LDLR–hAFP gene fusion mice provide an excellent animal model to conduct a wide variety of physiological studies of the LDLR pathway without sacrificing the animals.
We chose ES cells to perform the tissue culture studies both because of the extensive experience with homologous recombination in these cells and because of the associated option of germ-line transmission. However, ES cells are not ideal to study regulation of the LDLR promoter, since LDLR is expressed most abundantly in hepatocytes. To overcome this limitation, primary cultured hepatocytes or hepatocyte cell lines derived from the LDLR–hAFP gene fusion mice could be studied and it would be of interest to analyse the secretion of hAFP in these cells in response to drugs.
We believe that this gene fusion strategy could be generalized to place cassettes producing easily detectable markers under the control of regulatory regions of interest. Mice expressing a detectable marker under the control of a promoter of interest could then be mutagenized to search for genetic effects of upstream genes. Our laboratory has established a mouse line expressing the agouti coat-color marker under the control of the Snurf–Snrpn promoter (15). Using these mice in combination with N-ethyl-N-nitrosourea (ENU) mutagenesis, genetic and epigenetic variants affecting genomic imprinting were isolated. In this study, we used hAFP as the reporter gene because it is secreted and easily quantified. Other markers that could easily be measured in the circulating blood such as human growth hormone or human 1-antitrypsin can be used for the same purpose. Cassettes expressing green fluorescent protein (GFP) might also be an excellent marker for studying the effect of upstream mutations on expression patterns. It would be possible to insert bicistronic cassettes such as hAFP and GFP reporters separated by an IRES under the control of a promoter of interest. A major advantage of a bicistronic reporter sequence is that events affecting measurement of a reporter but irrelevant to the expression of the gene fusion transcript are easily distinguished in that they would ordinarily affect one reporter sequence but not the other. The advantages of bicistronic reporters are well documented in the use of gene fusions in bacterial genetics (16). In order to minimize irrelevant effects, it may also be preferable to remove foreign promoters and reporter genes such as the Neo expression cassette in the targeted allele, as shown in this study. Many variations on this strategy can be envisioned in combination with growing use of ENU and insertional mutagenesis in the ES cells and mice (14,17).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Targeting vectors
We isolated overlapping lambda phage clones from a 129/SvEv genomic library provided by A. Bradley. A 5'-targeting vector was constructed using a replacement strategy designed to place a reporter cassette (SA–IRES–hAFP–pA) into intron 1 of LDLR. We cloned a 6.3 kb fragment consisting of a 5' homologous arm (a 3.3 kb EcoRI fragment) and the SA–IRES–hAFP–pA cassette into the pNeo-TK vector (18). A 3.2 kb EcoRI–SalI fragment was subsequently cloned into the same vector as the 3'-homologous arm. A 3'-targeting insertion vector was used to place an IRES–hAFP–pA cassette in the 3'-noncoding region of LDLR by cloning a 4.6 kb EcoRI fragment (homologous sequence) and an IRES–AFP cassette into the pNeo-TK vector (18).
Generation of ES cells and mice expressing hAFP under control of the LDLR promoter
We linearized both 5'- and 3'-targeting vectors using SalI and electroporated each into AB2.2 ES cells. Colonies were selected with 200 µg/ml Geneticin (G418 sulfate; Life Technologies) and analysed by mini-Southern blot hybridization as described previously (19,20). We analysed ES cell DNA of 5'-targeted clones using a 0.4 kb fragment containing exon 1 of LDLR as a 5'-flanking probe and a 0.6 kb SalI–EcoRI fragment as the 3'-flanking probe after digestion of the ES cell DNA with either XbaI or BglII, respectively. For 3'-targeting, we digested ES cell DNA with BglII and analysed the DNA using a 0.27 kb BglII–EcoRI fragment as a 5'-flanking probe and a 0.5 kb fragment containing a part of exon 18 of LDLR as a 3'-flanking probe. Appropriately targeted clones for each vector were further electroporated with a plasmid expressing Cre recombinase (21). We confirmed recombination using the same restriction enzymes and flanking probes as described above. We injected two independent 3'-targeted clones into C57BL/6J blastocysts and reimplanted these into pseudopregnant female mice using standard procedures (20). Chimeric males were mated with C57BL/6J females. Germ-line transmission was confirmed by agouti coat color in F1 animals and by Southern blot analysis of tail DNA using the 3'-flanking probe.
RT–PCR
Total RNA from the indicated cell lines or mouse liver was prepared using the RNAeasy kit from Qiagen (Valencia, CA, USA). Transcript from the fusion allele was amplified using oligonucleotide primers in the IRES (IR-F: GTG ATT TTC CAC CAT ATT GCC GTC) and AFP coding region (hAFP-R: GCC AGG TCA GCT AAA CTT ATC TCT) for 24 cycles (94°C 45 s, 56°C 30 s, 72°C 90 s ). Mouse LDLR mRNA was amplified using primers in exon 1 (E1-F: TCA TCG CCC TGC TCC TTG CT) and exon 4 (E4-R: GAG CCG TCA ACA CAG TCG ACA T) for 25 cycles (94°C 45 s, 56°C 30 s, 72°C 90 s). Mouse glyceraldehyde 3-phosphate dehydrogenase (GAPD) mRNA was amplified using primers mGAPD-F (CCAGTATGACTCCACTCACGG) and mGAPD-R (CACAGCCTTGGCAGCACCAGT) for 16 cycles (94°C 45 s, 55°C 45 c, 72°C 90 s). Reaction products were resolved on a 1% agarose gel and quantified using NIH IMAGE software.
Quantification of hAFP
The targeted ES cells were cultured in ES medium with 15% fetal bovine serum (FBS) as described previously (20). A total of 1.2x106 cells per well were seeded into six-well plates for 1 day in ES cell medium with 15% FBS and changed to medium with 10% FBS or 10% LPDS the next day in the presence or absence of simvastatin. Simvastatin was provided by Merck & Co (West Point, PA, USA) and activated by alkaline hydrolysis of the lactone moiety, according to the manufacturer's protocol. The cultured medium was used directly for assaying hAFP concentration at 24 and 48 h. We measured hAFP using a two-site sandwich immunoassay system from Chiron Diagnostics (catalog no. 672258; Emeryville, CA, USA) following the manufacturer's instruction.
Administration of simvastatin to mice
Seven-week-old mice heterozygous for the LDLR–hAFP fusion allele were fed with 5021 regular diet (powder form, Harlan Teklad; Indianapolis, IN, USA) mixed in the absence or presence of simvastatin (0.55 mg/g food) for 2 weeks. As a control, the wild-type littermates were also treated with simvastatin. The mice were bled retro-orbitally at indicated time points to obtain serum for hAFP assay. To assess the effect of a high-fat diet, 7-week-old mice were fed the standard 5021 diet or a high-fat, Western-type diet that contained 21.22% (g/100 g) fat, 17.01% protein, 48.48% carbohydrate and 0.15% cholesterol (TD88137; Harlan Teklad) for 2 weeks, and blood was collected for hAFP assay. The differences in AFP value between treated groups were compared for statistical significance using unpaired, two-tailed t-test.
|
ACKNOWLEDGEMENT |
|---|
This work was supported by NIH grant R01 HD37283.
|
FOOTNOTES |
|---|
* To whom correspondence should be addressed at: Department of Molecular and Human Genetics, Baylor College of Medicine, One Baylor Plaza, Room T619, Houston, TX 77030, USA. Tel: +1 7137984795; Fax: +1 7137987773; Email: abeaudet{at}bcm.tmc.edu
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
1 Bradley, A., Hasty, P., Davis, A. and Ramirez-Solis, R. (1992) Modifying the mouse: design and desire. Biotechnology (NY), 10, 534–539.[Medline]
2 Melton, D.W. (1994) Gene targeting in the mouse. Bioessays, 16, 633–638.[Web of Science][Medline]
3 Zhou, G., Garofalo, S., Mukhopadhyay, K., Lefebvre, V., Smith, C.N., Eberspaecher, H. and de Crombrugghe, B. (1995) A 182 bp fragment of the mouse pro alpha 1(II) collagen gene is sufficient to direct chondrocyte expression in transgenic mice. J. Cell Sci., 108, 3677–3684.[Abstract]
4 Brock, D.J.H. (1977) Biochemical and cytological methods in the diagnosis of neural tube defects. Prog. Med. Genet., 2, 1.[Medline]
5
O'Neal, W.K., Rose, E., Zhou, H., Langston, C., Rice, K., Carey, D. and Beaudet, A.L. (2000) Multiple advantages of -fetoprotein as a marker for in vivo gene transfer. Mol. Ther., 2, 640–648.[Web of Science][Medline]
6 Yamashita, T., Kasai, N., Miyoshi, I., Sasaki, N., Maki, K., Sakai, M., Nishi, S. and Namioka, S. (1993) High level expression of human alpha-fetoprotein in transgenic mice. Biochem. Biophys. Res. Commun., 191, 715–720.[Web of Science][Medline]
7 Zarzour, S.J., Gabert, H.A., Diket, A.L., St Amant, M. and Miller, J.M., Jr. (1998) Abnormal maternal serum alpha fetoprotein and pregnancy outcome. J. Matern. Fetal Med., 7, 304–307.[Medline]
8
Brown, M.S. and Goldstein, J.L. (1986) A receptor-mediated pathway for cholesterol homeostasis. Science, 232, 34–47.
9 Goldstein, J.L., Hobbs, H.H., and Brown, M.S. (2001) Familial Hypercholesterolemia. In Scriver, C.R., Beaudet, A.L., Valle, D. and Sly, W.S. (eds), The Metabolic and Molecular Bases of Inherited Disease. McGraw-Hill, New York, pp. 2863–2913.
10 Spady, D.K., Woollett, L.A. and Dietschy, J.M. (1993) Regulation of plasma LDL-cholesterol levels by dietary cholesterol and fatty acids. A. Rev. Nutr., 13, 355–381.
11 Hoffer, M.J., van Eck, M.M., Petrij, F., van der, Z.A., de Wit, E., Meijer, D., Grosveld, G., Havekes, L.M., Hofker, M.H. and Frants, R.R. (1993) The mouse low density lipoprotein receptor gene: cDNA sequence and exon–intron structure. Biochem. Biophys. Res. Commun., 191, 880–886.[Web of Science][Medline]
12 Rees, S., Coote, J., Stables, J., Goodson, S., Harris, S. and Lee, M.G. (1996) Bicistronic vector for the creation of stable mammalian cell lines that predisposes all antibiotic-resistant cells to express recombinant protein. BioTechniques, 20 , 102–110.
13 Stanford, W.L., Cohn, J.B. and Cordes, S.P. (2001) Gene-trap mutagenesis: past, present and beyond. Nat. Rev. Genet., 2, 756–768.[Web of Science][Medline]
14 Nolan, P.M., Peters, J., Strivens, M., Rogers, D., Hagan, J., Spurr, N., Gray, I.C., Vizor, L., Brooker, D., Whitehill, E. et al. (2000) A systematic, genome-wide, phenotype-driven mutagenesis programme for gene function studies in the mouse. Nat. Genet, 25, 440–443.[Web of Science][Medline]
15
Tsai, T.F., Chen, K.S., Weber, J.S., Justice, M.J. and Beaudet, A.L. (2002) Evidence for translational regulation of the imprinted Snurf–Snrpn locus in mice. Hum. Mol. Genet., 11, 1659–1668.
16 Slauch, J.M. and Silhavy, T.J. (1991) Genetic fusions as experimental tools. Meth. Enzymol., 204, 213–248.[Web of Science][Medline]
17 Wiles, M.V., Vauti, F., Otte, J., Fuchtbauer, E.M., Ruiz, P., Fuchtbauer, A., Arnold, H.H., Lehrach, H., Metz, T., von Melchner, H. et al. (2000) Establishment of a gene-trap sequence tag library to generate mutant mice from embryonic stem cells. Nat. Genet., 24, 13–14.[Web of Science][Medline]
18 Abuin, A. and Bradley, A. (1996) Recycling selectable markers in mouse embryonic stem cells. Mol. Cell. Biol., 16, 1851–1856.[Abstract]
19 Bressler, J., Tsai, T.-F., Wu, M.-Y., Tsai, S.-F., Ramirez, M.A., Armstrong, D. and Beaudet, A.L. (2001) The SNRPN promoter is not required for genomic imprinting of the Prader–Willi/Angelman domain in mice. Nat. Genet., 28, 232–240.[Web of Science][Medline]
20 Ramirez-Solis, R., Davis, A.C. and Bradley, A. (1993) Gene targeting in embryonic stem cells. Meth. Enzymol., 225, 855.[Web of Science][Medline]
21
Orr-Urtreger, A., Broide, R.S., Kasten, M.R., Dani, J.A., Beaudet, A.L. and Patrick, J.W. (2000) Mice homozygous for the L250T mutation in the 7 nicotinic acetylcholine receptor show increased neuronal apoptosis and die within one day of birth. J. Neurochem., 74, 2154–2166.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||