Nature Genetics
25, 444 - 447 (2000)
doi:10.1038/78146
Genome-wide, large-scale production of mutant mice by ENU mutagenesisMartin Hrabé de Angelis1, Heinrich Flaswinkel5, Helmut Fuchs1, Birgit Rathkolb6, Dian Soewarto1, Susan Marschall1, Stephan Heffner1, Walter Pargent7, Kurt Wuensch1, Martin Jung7, André Reis8, Thomas Richter2, Francesca Alessandrini9, Thilo Jakob9, Edith Fuchs10, Helmut Kolb10, Elisabeth Kremmer3, Karlheinz Schaeble1, Boris Rollinski11, Adelbert Roscher11, Christoph Peters12, Thomas Meitinger13, Tim Strom13, Thomas Steckler14, Florian Holsboer14, Thomas Klopstock15, Florian Gekeler15, Catherine Schindewolf1, Thomas Jung16, Karen Avraham17, Heidrun Behrendt9, Johannes Ring9, Andreas Zimmer18, Klaus Schughart19, Klaus Pfeffer5, Eckhard Wolf6
& Rudi Balling41 Institute of Experimental Genetics, GSF Research Center for Environment and Health, Neuherberg, Germany. 2 Institute of Pathology, GSF Research Center for Environment and Health, Neuherberg, Germany. 3 Institute of Immunology, GSF Research Center for Environment and Health, Neuherberg, Germany. 4 Institute of Mammalian Genetics, GSF Research Center for Environment and Health, Neuherberg, Germany. 5 Institute of Medical Microbiology, Immunology and Hygiene, Technical University of Munich, Munich, Germany. 6 Institute of Molecular Animal Breeding, Gene Center, University of Munich, Munich, Germany. 7 Ingenium Pharmaceuticals AG, Martinsried, Germany. 8 Max-Delbrueck-Centre, Molekulare Genetik und Mikrosatellitenzentrum, Berlin, Germany. 9 Division Environmental Dermatology and Allergology, GSF/TUM, Munich, Germany. 10 Institute of Clinical Chemistry, Clinic Harlaching, Munich, Germany. 11 Abteilung fuer Klinische Chemie, Biochemie und Stoffwechsel, Kinderklinik und Kinderpoliklinik im Dr. von Haunerschen Kinderspital, Munich, Germany. 12 Innere Medizin I, Medizinische Molekularbiologie, Universitaetsklinik Freiburg, Freiburg, Germany. 13 Abteilung Medizinische Genetik, Kinderklinik der Ludwig-Maximilians-Universitaet, Munich, Germany. 14 Max-Planck Institute of Psychiatry, Munich, Germany. 15 Department of Neurobiology, Klinikum Gro hadern, Ludwig-Maximilians-Universitaet Munich, Munich, Germany. 16 Bundesamt fuer Strahlenschutz, Institut fuer Strahlenhygiene, Oberschleissheim, Germany. 17 Department of Human Genetics, Sackler School of Medicine, Tel Aviv University, Ramat Aviv, Tel Aviv, Israel. 18 Laboratory of Genetics, NIMH, Bethesda, Maryland, USA. 19 Transgene S.A., 11, Strasbourg Cedex, France.
Correspondence should be addressed to Martin Hrabé de Angelis hrabe@gsf.deIn the post-genome era, the mouse will have a major role as a model system for functional genome analysis. This requires a large number of mutants similar to the collections available from other model organisms such as Drosophila melanogaster and Caenorhabditis elegans. Here we report on a systematic, genome-wide, mutagenesis screen in mice. As part of the German Human Genome Project, we have undertaken a large-scale ENU-mutagenesis screen for dominant mutations and a limited screen for recessive mutations1. In screening over 14,000 mice for a large number of clinically relevant parameters, we recovered 182 mouse mutants for a variety of phenotypes. In addition, 247 variant mouse mutants are currently in genetic confirmation testing and will result in additional new mutant lines. This mutagenesis screen, along with the screen described in the accompanying paper2, leads to a significant increase in the number of mouse models3 available to the scientific community. Our mutant lines are freely accessible to non-commercial users (for information, see http://www.gsf.de/ieg/groups/enu-mouse.html).
To produce new mutant lines, we injected C3HeB/FeJ male mice with ethyl-nitrosourea (ENU) and then mated them with C3HeB/FeJ females to produce F1 founders. F1 animals were analysed for novel phenotypes of dominant and semidominant traits, or bred further to screen for recessive phenotypes (G3). We screened for specific, postnatal abnormalities such as congenital malformations, clinical chemical, biochemical, haematological and immunological/allergological defects, and complex traits such as behaviour. We established a working network with the diagnostic facilities in various departments of medical clinics (Fig. 1). New mutants were produced and kept on an inbred C3HeB/FeJ background, which enables the study of phenotypes without the bias of polymorphic genetic interference. For chromosomal mapping, a microsatellite panel spanning the entire mouse genome was established for polymorphic markers between C3HeB/FeJ and C57BL/6JIco mice with a genome-wide spacing of less than 20 cM. Conserved synteny among mammalian species permits the mapping of the mutant loci to a mouse chromosomal region and allows prediction of where a corresponding human disease gene maps, providing immediate access to potential candidate genes. Currently, more than 30 mutant lines are in mapping crosses. For example, one of the cataract mutants (AEY3) mapped to mouse chromosome 11, distal to marker D11Mit242 and proximal to D11Mit36. As Cryba1 also maps to this region, it was a candidate gene. Subsequently, a T A mutation was detected in Cryba1 that lead to an alteration within the protein4. As complete transcription maps of the human and mouse genomes should be available in the near future, an increasing number of mapped mutant lines can be analysed for mutation detection and gene discovery. Mutant strains are archived by cryopreservation of spermatozoa5,
6. We are also developing MouseNet, a database system that is suitable to support this large-scale project.
 | | |
Characterization of several new mutant lines is described here in more detail. Within the clinical-chemical screen, the mutant ALP-01 was identified and characterized by decreased levels of plasma alkaline phosphatase activity (Fig. 2a). As the mouse mutant ALP-01 does not show clinical symptoms other than hypophosphatasia (detailed X-ray and histological analysis of the skeleton pending), this strain might correspond to human type 3 hypophosphatasia7. Another dominant mutant, HST-01, shows increased levels of plasma urea (Fig. 2b). The causes for azotaemia may be pre-renal (for example extensive protein turnover), renal (reduced kidney function) or post-renal (ureteral or urethral obstruction). As plasma creatinine, another marker of kidney function, is normal in the mutant HST-01, a pre-renal cause of increased plasma urea is most likely. We also identified one recessive hyperglycaemic mutant and another mutant (RP-19; Fig. 2c) characterized by elevated plasma cholesterol and urea concentrations.
| | |
The immunology screen placed particular emphasis on the detection of mouse mutants that are models for primary immune deficiencies and hypersensitivity disorders. We analysed more than 5,500 mice by ELISA and more than 3,500 mice by flow cytometry. Among the 30 mutants that we confirmed, several had Ig and IgG-subclass deficiencies resembling aspects of known primary immune deficiencies of the common variable immunodeficiency disease group8 (CVID). In addition, we identified seven mutants which, as revealed by flow cytometry, have an absolute or relative absence of immunologically relevant cell populations or cell-surface molecules. These phenotypes might encompass a set of T-cell, B-cell or combined cellular immunodeficiencies in humans. One of the mouse lines displayed a recessive phenotype identified in the G3 screen (TUM-011/88). These mice had a decrease in CD21/35 expression on B cells (Fig. 3). This mutation is apparently different from a complete inactivation9,
10, as CD21/35-positive non-B-cells can still be labelled by monoclonal antibodies 7G6, 8C12 and 7E9 (Fig. 3a,b).
| | |
The goal of the allergy screen is to identify mutants with aberrant IgE levels. The most frequent type of known allergic reactions is the IgE-mediated allergy (immediate type hypersensitivity). But animal models for IgE-mediated allergies are rare and often require non-physiological sensitization protocols. We isolated nine mutant lines within the allergy screen (data not shown). We characterized six mutant lines with IgE levels below those that can be detected by ELISA, and three mutant lines with high plasma IgE concentrations compared with wild-type mice.
The main purpose of the dysmorphology screen is to isolate mouse mutants that have phenotypic abnormalities of the sense organs, limbs, axial skeleton, pigmentation and central nervous system. We identified 115 dysmorphology mutants and confirmed them genetically (Table 1). The limb mutant, ALI4, was found in the dominant screen and showed defects in its nails. Heterozygous and homozygous animals are viable and behave normally. Pathological examination showed dysmorphic hyperplastic and pigmented nails (Fig. 4b,c). Defects in heterozygous animals are less severe than those in homozygous ALI4 mutants. A pathological investigation led to the conclusion that ALI4 mutants have a severe disturbance in the matrix of the nail bed with stronger proliferation and hyper-parakeratosis. The ALI4 mutation maps to mouse chromosome 11 between D11Mit263 and D11Mit224. As these markers are closely linked to the keratin complex I, a mutation in one of the keratin genes may be responsible for this phenotype.
| | |
| | |
We have demonstrated the feasibility of large-scale, phenotype-driven mutagenesis screens in mice. The power of this screen lies in its coordinated protocol, in which many screens, with a variety of parameters, are carried out on every mouse. ENU is the most potent mutagen in mice11, but, until recently, a genome-wide multiparameter screen of this scale had not been attempted. The success of this and future screens will depend heavily on the exploitation of interdisciplinary collaborations in the design of new phenotyping assays that are fast, cheap, robust and preferably non-invasive. Moreover, ENU mutagenesis is not necessarily restricted to large screening centres, but can be set up efficiently in smaller labs12,
13. We found dominant-mutation frequencies to be in the range of 2%, whereas preliminary data show that every second pedigree gives rise to the recovery of a mutant in recessive screens. The combination of region-specific deletion screens with the use of dominant coat colour markers and balancer stocks should further improve the overall efficiency of mutant recovery12. Even when chromosomal mapping has reduced the candidate region to a few hundred kilobases or a few candidate genes, one of the major rate-limiting steps is still the actual identification of the individual causative mutations. But once cheap, high-throughput mutation detection technology becomes available, the DNA of offspring from mutagenized mice can be directly analysed for mutations in specific genes, permitting the convergence of phenotype and gene-driven mutagenesis. A major, long-term goal of the human genome project is to understand multigenic and multifactorial diseases. In most cases, the mutants reported here will be monogenic. Crossing them onto different genetic backgrounds, however, will reveal modifying genes and permit us to study their corresponding complexities. Suppressor and enhancer screens will then become routine in mouse genetics. The ENU mutagenesis programmes will be an important tool for the discovery of gene function in the worldwide efforts of genome research.
Methods
Animals.
We carried out the animal studies under the license of the Regierung von Oberbayern and the Staatliches Veterinäramt München, reference no. 211-2531-28/96. For this study, we used C3HeB/FeJ (The Jackson Laboratory) and C57BL/6Jico (Ifacredo) mice. We injected male C3HeB/FeJ mice intraperitoneally with 3 weekly doses of 90 mg/kg ENU (Serva) at  10−14 weeks of age (some injection groups were treated with 3 80 mg/kg or 1160 mg/kg ENU). Groups of 100 C3H males were injected at 3 monthly intervals. We screened a maximum number of 100 offspring per ENU-C3H male. We performed the dominant screen on F1 animals derived from mutagenized C3HeB/FeJ males crossed with wild-type C3HeB/FeJ females. We tested the inheritance on this C3H inbred background in which 20 G2 animals were assessed for the phenotype of the founder animal. We carried out the recessive screen on G3 animals that had been produced in a two-step breeding scheme: first, we crossed an F1 male, excluded from carrying a potentially dominant mutation, with 4 wild-type C3H females producing G2 offspring; and second, we backcrossed 8 G2 females with the F1 male to produce 40 G3 offspring per micropedigree. We tested the inheritance of phenotypes on outcross-intercross or outcross-backcross offspring of the G3 founder. Names used for mutant lines are for internal use only and do not reflect official nomenclature. Mutant lines derived from our screen will be assigned official gene symbols and names in the near future.
Clinical-chemical screen.
We obtained blood samples (550  l) from 3-month-old mice that had fasted overnight by puncturing the retro-orbital sinus under ether anaesthesia. We used EDTA-blood (50 l) for basic haematology (Vet abc Animal Blood Counter, ABX Diagnostics). Plasma (130 l) from Li-heparin-blood was analysed for clinical-chemical parameters (Table 2, see http://genetics.nature.com/supplementary_info/) using a Hitachi 717 autoanalyzer and adapted reagents (Roche). Values were considered as abnormal if being below the first or above the ninety-ninth percentile, respectively.
Immunology screen.
We used an isotype-specific sandwich ELISA for IgM, IgG1, IgG2a, IgG3 and IgA. Anti-mouse Ig antibodies and alkaline phosphatase (AP) conjugates were purchased (Pharmingen, Becton & Dickinson). We measured anti-DNA antibodies on ELISA plates coated with calf thymus DNA (Pharmacia) and detected autoreactive antibodies with AP-conjugated goat anti-mouse IgM and IgG (Dianova). We determined rheumatoid factor using the mouse monoclonal antibody 6G9 (IgG isotype) for coating and an AP-conjugated anti-mouse IgM monoclonal antibody (Pharmingen).
For flow cytometry, we prepared PBL from peripheral blood (600 l/mouse) by erythrocyte lysis (NH4Cl-TRIS buffer). Cells were washed, and Fc-receptors were blocked (Fc-block, Pharmingen) and labelled with fluorochrome-coupled Ab cocktails (Table 3, see http://genetics.nature.com/supplementary_info). We performed the analysis using a FACSCalibur flow cytometer (Becton & Dickinson) and scored 10,000 cells per sample. Semi-automated analysis was based on the Attractors software package (Becton & Dickinson). Screening data were imported into a FILEMaker Database (FileMaker) and variants were graphically visualized using Excel (Microsoft).
Allergy screen.
We analysed plasma (12 l) from 12-week-old mice by ELISA for mouse IgE using a standard microwell ELISA reader and reagents (Pharmingen and The Binding Site).
Dysmorphology screen.
This screen evaluated defects of the CNS, sense organs, limbs, axial skeleton and pigmentation. We performed the dysmorphology screen after birth, after weaning, at 12 weeks and for a fraction of the colony at 6 months. Once a litter was identified, all recognized pups were examined carefully. From week 3 to week 9, mice were weighed every other week. A baseline was created for the weight data including 20,000 records of individual F1 mice (20−160 d, strain C3HeB/FeJ). The information stored in the records was sex, weight and age at the day of data collection. For each day, between 20 and 600 observations for each sex were available. For every day we calculated the mean value and the standard deviation for male and female weight data. An analysis of regression for the model Weight = a + b*Age + c*Age2 + d*Age3 + e*Age4 was performed for the mean weights of males and females. The model for females has an r2 of 0.98 and that for the males has an r2 of 0.97. The increase of the standard deviation with age was estimated using male and female data together in the model standard deviation = f + g*age. The model has an r2 of 0.59. A variant (a mouse with phenotype that must be genetically confirmed in a confirmation cross) will be detected if |v|  3; vmales=(W-(-9.487025+1.4385603*t-0.021653*t2+0.00015285*t3-0.0000003937*t4))/(1.26865119+0.02151744*t) vfemales=(W-(1.532645+0.688995*t-0.008334*t2+0.000054345*t3-0.0000001302*t4))/(1.26865119+0.02151744*t) where W means weight of the mouse that is t days old. Weight data of mice at the same age did not differ significantly from normal distribution. In the eleventh week we examined mice for obvious dysmorphologies (Table 4, see http://genetics.nature.com/supplementary_info/).
Cryopreservation of spermatozoa.
We carried out cryopreservation of spermatozoa as described6. We collected both caudae epididymides and vasa deferentia and dispersed the spermatozoa in a cryoprotectant solution (18% raffinose, 3% skim milk). The sperm suspension was divided into ten aliquots (15 l each) and each specimen was loaded into a French-type straw. After cooling in vapor (10 min), we directly immersed the straws in liquid nitrogen and stored them in liquid nitrogen tanks. For thawing, the frozen straws were removed from liquid nitrogen and put into a waterbath maintained at 37 °C for not more than 15 min.
Recovery of archived mutants.
For recovery of archived mutant males, in vitro fertilization was carried out as described6. We obtained oocyte-cumulus complexes from the oviducts of superovulated females 14 h after injection of human chorionic gonadotropin and introduced them into human tubal fluid (HTF) medium containing thawed spermatozoa. After incubation (37 °C, 4−5 h), we washed the eggs and cultured them overnight in KSOM medium. The next day, oocytes that developed into two-cell embryos were transferred into the oviducts of recipient females (day 0.5 of pseudo-pregnancy).
Genetic mapping.
Mapping was done by an outcross-backcross breeding scheme of affected C3HeB/FeJ and C57BL/6JIco mice. We took tail clips from backcross animals and performed DNA extraction using the DNAeasy kit (Qiagen). A genome-wide microsatellite panel was established for polymorphic markers between C3HeB/FeJ and C57BL/6JIco mice with a genome-wide spacing of less than 20 cM. A genomic scan was performed on 50 affected G3 backcross animals to exclude the possibility of false-negative results.
Database.
A central database is currently under development at the core facility (MouseNet). MouseNet will be accessible from different client platforms via the Internet, will be a full-featured, multi-user system (including access restriction and data locking mechanisms), will rely on an industrial class RDBMS (relational database management system) running on a UNIX server platform (which provides all basic features needed for maximum data security and consistency), and will supply workflow functions and a variety of plausibility checks.
Note: Supplementary information is available on the Nature Genetics web site (http://genetics.nature.com/supplementary_info/).
Received 31 January 2000; Accepted 1 May 2000
REFERENCES
- Hrabé de Angelis, M. & Balling, R. Large scale ENU screens in the mouse. Genetics meets genomics. Mutat. Res. 400, 25–32 (1998). | Article | PubMed | ISI | ChemPort |
- Nolan, P.M. et al. A systematic, genome-wide, phenotype-driven mutagenesis program for gene function studies in the mouse. Nature Genet. 25, 440–443 (2000). | Article | PubMed | ISI | ChemPort |
- Brown, S. & Peters, J. Combining mutagenesis and genomics in the mouse—closing the phenotype gap. Trends Genet. 12, 433–435 (1996). | Article | PubMed | ISI | ChemPort |
- Graw, J. et al. Mutation in the bA3/A1-crystallin encoding gene Cryba1 causes a dominant cataract in the mouse. Genomics 62, 67–73 (1999) | Article | PubMed | ISI | ChemPort |
- Marschall, S. & Hrabé de Angelis, M. Cryopreservation of mouse spermatozoa–double your mouse space. Trends Genet. 15, 128–131 (1999). | Article | PubMed | ISI | ChemPort |
- Marschall, S., Huffstadt, U., Balling, R. & Hrabé de Angelis, M. Reliable recovery of inbred mouse lines using cryopreserved spermatozoa. Mamm. Genome 10, 773–776 (1999). | Article | PubMed | ISI | ChemPort |
- Whyte, M.P., Teitelbaum, S.L., Murphy, W.A., Bergfeld, M.A. & Avioli, L.V. Adult hypophosphatasia. Clinical, laboratory, and genetic investigation of a large kindred with review of the literature. Medicine 58, 329–347 (1979). | PubMed | ISI | ChemPort |
- Spickett, G.P. et al. Common variable immunodeficiency: how many diseases? Immunol. Today 18, 325–328 (1997). | Article | PubMed | ISI | ChemPort |
- Ahearn, J.M. et al. Disruption of the Cr2 locus results in a reduction in B-1a cells and in an impaired B cell response to T-dependent antigen. Immunity 4, 251–262 (1996). | Article | PubMed | ISI | ChemPort |
- Molina, H. et al. Markedly impaired humoral immune response in mice deficient in complement receptors 1 and 2. Proc. Natl Acad. Sci. USA 93, 3357–3361 (1996). | Article | PubMed | ChemPort |
- Russell, W.L. et al. Specific-locus test shows ethylnitrosourea to be the most potent mutagen in the mouse. Proc. Natl Acad. Sci. USA 76, 5918–5922 (1979).
- Justice, M.J., Noveroske, J.K., Weber, J.S., Zheng, B. & Bradley, A. Mouse ENU mutagenesis. Hum. Mol. Genet. 8, 1955–1963 (1999). | Article | PubMed | ISI | ChemPort |
- Nolan, P.M., Kampfhamer, D. & Bucan, M. Random mutagenesis screen for dominant behavioral mutations in mice. Methods 13, 379–395 (1997). | Article | PubMed | ISI | ChemPort |
Acknowledgments
We thank A. Servatius, S. Prettin, G. Bergter, N. Hirsch, A. Mayer, S. Manz, S. Hoffmann, F. Golla, B. Beneckenstein, K. Lobenwein, A. Wolf and D. Kreitz for technical assistance; C. Schindewolf for comments and revision of the manuscript; and H. Wagner for support. Part of this project was supported by grants from the German Human Genome Project to R.B., E.W. and M.H.d.A. (01KW9610/1), and to K.P., J.R. and H.B. (01KW9636).
|
