WO1999001577A1 - Reproductive and genetic screening samples of mutagenised animals - Google Patents

Abstract

The invention provides a collection of paired samples from a plurality of mutagenised animals, wherein each paired sample comprises: a first sample comprising genetic screening material from a parent animal; and a second sample comprising reproductive material from that same parent animal. The reproductive material is preferably gametes. It also provides methods for producing and screening the collection, for instance to identify a gamete sample in such a collection known to carry a mutation in a gene of interest. This gamete sample can be used to generate an animal carrying that same mutation. The mutant animal is useful for investigating the function of the gene, and might also be a useful animal model for studying human disease. In effect, therefore, the mutations in a population of mutagenised animals are immortalised and animals carrying particular mutations of interest can be produced easily without the need to maintain a living animal population.

Description

REPRODUCTIVE AND GENETIC SCREENING SAMPLES OF MUTAGEN ISED AN IMALS

This invention relates to collections of biological samples useful for identifying members of an animal population carrying mutations in a gene of interest, said members being useful for the identification and study of phenotypes associated with the gene of interest.

Background to the invention

Historically, much progress has been made in biochemistry by the study of mutants, for instance in early work on the lac operon. Since the introduction of modern DNA manipulation techniques, the specific alterations in genotype responsible for many mutant phenotypes have been investigated and characterised. For instance, the most common cause of the hereditary human disease cystic fibrosis is known to be the deletion of a CTT codon in the CFTR gene.

Since the advent of gene mapping and DNA sequencing, and the initiation of ventures such as the human genome project, a mass of gene sequence information has been produced, and one of the key challenges in modern genetics is the elucidation of functions for these genes [1]. The human genome is thought to contain around 80000 genes, and whilst the human genome project aims to provide detailed sequence information for all of these genes, in most cases this will give no indication of the function of the gene product.

For instance, when the complete sequence of the Haemophilus inβuenzae genome was reported in 1995, over 40% of the genes proved to have no counterparts of known function among the genes already sequenced from other organisms [2]. Even higher figures for the proportion of sequenced human genes with no postulated function have been reported [3].

Typically, workers will start with a known phenotype and try to identify the genotype responsible. Once the genotype and phenotype are correlated, which can be a laborious process, the gene product can be identified and the molecular basis whereby mutations in that gene cause a disease phenotype can be investigated rationally. In other words, the phenotype is usually used as a guide when searching for mutations in genes.

As an alternative, it is possible to start with a mutated gene and then look for associated phenotypes. The effect of mutations can therefore be explored without prior knowledge of the function of the gene. One such an approach has been suggested in reference 4. A large population of mutagenised organisms is established and this population is screened at a genetic level to identify those members of the population carrying a mutation in a known gene of interest, without knowledge of the associated phenotype. Once the organisms in the population carrying a mutation in the specific gene are identified, these can be used for breeding, thus allowing phenotypic studies. However, the population of mutated organisms used for screening in this method is dynamic and maintaining a large population of organisms such as mice is expensive.

Thus it is an object of the invention to provide a population of mutagenised organisms in which the death of an organism does not cause the loss of its genetic information for genome analysis or of its ability to produce progeny for phenotype analysis.

It is also an object of the invention to provide a method for generating an organism known to carry a mutation in a gene of interest.

Disclosure of the invention

According to the invention, there is provided a collection of paired samples from a plurality of mutagenised animals, wherein each of the paired samples comprises: a first sample comprising genetic screening material from a parent animal; and a second sample comprising reproductive material from that same parent animal.

There is further provided a process for producing such a collection, comprising the steps of obtaining pairs of samples from a plurality of mutagenised animals: a first sample comprising genetic screening material; and a second sample comprising reproductive material. The animals may be sacrificed before, during, or after the samples are obtained. The process preferably includes the step of mutagenising a plurality of animals prior to obtaining the samples.

Preferably, the plurality comprises 100 or more mutagenised animals, more preferably 1000 or more animals, more preferably again 10000 or more animals, and most preferably in excess of 100000 animals. Preferably the animals are all of the same species.

The term "mutagenised" means that the animals from which the samples are obtained carry mutations characteristic of exposure to mutagenic conditions (in particular, at a frequency characteristic of exposure to mutagenic conditions). It does not refer to spontaneous or background mutations, which are characterised by their low frequency of occurrence. The animals from which the samples are obtained preferably carry mutations at a frequency substantially above this background frequency. Suitably, the mutation frequency is such that, on average, at a phenotypic level one mutant copy of a gene occurs per 50000 organisms or fewer (eg. one in every 10000 organisms, or one in every 1000 organisms).

The mutations in the samples of genetic screening material and reproductive material must correspond (ie. the location and nature of the mutations must be the same). To ensure this, an animal which has been exposed to mutagenic conditions may be bred in order to obtain progeny (FI, F2, F3 generation etc.) whose mutations are consistent throughout its gametes and somatic tissue. The animals from which samples are obtained might, for instance, be the progeny of an inbred (and therefore genetically identical) population which has been mutagenised. The germline of an animal may be mutagenised, preferably the male germline, and suitable samples can be obtained from its FI generation progeny.

As used herein, "mutation" refers to an alteration in the nucleotide sequence of a given gene (including its regulatory sequences) from its wild-type or normal nucleotide sequence. Thus the term includes point mutations as well as deletions, insertions, or larger substitutions, and also includes chromosomal rearrangements.

Many suitable methods for inducing mutations are known in the art. These include chemical mutagenesis, radiation, and retroviral or transposon insertion. Preferred methods are suitable for in vivo mutagenesis. Preferably the mutagenesis method makes point mutations; insertional mutations are preferably not employed. Preferred methods of chemical mutagenesis involve exposure to alkylating agents such as ethyl- or methyl-nitrosourea (ENU or MNU) [eg. references 5 and 6]. The most suitable method may depend on the animal in question, but the particular choice is routine [methods for the zebrafish, for instance, are described in reference 7, and for the mouse in reference 8]. The particular method used might also depend on the target for mutagenesis, such as germline or somatic cells.

The frequency of mutation chosen may depend on the number of mutagenised animals represented in the collection and the coverage of the genome which is desired. If the collection contains a small number of paired samples, in order to cover the whole genome such that a mutant copy of every gene is represented in the collection it is necessary to use a relatively high mutation frequency. If the collection is very large, a low mutation frequency is sufficient. There is, of course, an upper limit to the frequency of mutation which can be induced without resulting in death of the animal, and a lower limit below which the genome will not be fully covered.

Therefore, large collections require fewer mutations per animal in order to cover the whole genome. A low number of mutations per animal is desirable because possible complications associated with interactions between mutations are reduced and also because it reduces the need for breeding out additional mutations in genes not of interest. On the other hand, as the size of the collection increases, the effort involved in screening the collection also rises.

The mutation frequency is preferably such that, on average, about 10 functionally mutant copies of each gene are present per 10000 samples. At the DNA level the frequency might be more than 10 mutant copies per 10000 samples, but many of these mutations will not affect the function of the gene. The range of these "silent" mutations is diverse, but depending on the gene in question they might be mutations in non-coding regions, point mutations which do not alter the function of a codon (eg. CCU to CCG, or CGG to AGG), and mutations which alter a codon but which ordinarily do not affect the final protein function, such as conservative amino acid substitutions (eg. CUU Leu to AUU He). The functional mutants in the collection preferably carry mutations distributed across the gene of interest, that is to say throughout its coding and regulatory regions.

ENU mutagenesis is particularly suitable for mouse mutagenesis since, in the offspring of mutagenised male mice, a functionally aberrant copy of any given gene will occur at a frequency of approximately 1 per 1000 mice. Thus a mutant copy of each gene will be represented in the collection on average once every 1000 paired samples. Mutation frequencies for a variety of suitable mutagens in mice have been reviewed in reference 9.

A suitable mutagenised population of animals might also be produced by breeding from animals having mutant housekeeping genes such as those coding for DNA repair enzymes or proof-reading enzymes [eg. reference 10].

The animal can be of any type, vertebrate or invertebrate, but is preferably a vertebrate.

Preferably the vertebrate is a mammal or a fish. Suitable mammals include primates, rodents, lagomorphs, guinea pigs, horses, sheep, cattle, goats, pigs, cats, and dogs. Preferred mammals are mice and rats, whilst preferred fish are zebrafish and medaka fish.

The genetic screening material in the first sample in each of the paired samples may be any material suitable for genetic analysis. It may be in a form which is itself suitable or it may be in a form from which suitable material can be derived. Accordingly, this includes intact cells and cellular extracts from which DNA or RNA may be isolated for use in assays, and includes purified DNA or RNA. The DNA may be in the form of genomic DNA or cDNA. Preferably the genetic screening material comprises DNA purified from a diploid cell from an organism. The genetic screening material may be derived from living or dead organisms. The genetic screening material need not be in the same form in each of the paired samples in the collection. For example, it might comprise intact cells in some paired samples and purified DNA in others.

The reproductive material in the second sample in each of the paired samples may be any material which can be used to generate progeny of the parent animal (other than the living parent animal itself), and is preferably gametes (ie. spermatozoa or ova). Rather than having gamete samples in the collection, however, it may be preferable to use gametogenic stem cells from which gametes of the parent animal can be produced. Less preferably, the reproductive material may be an embryo generated using the parent animal's gametes, or embryonic stem cells from such an embryo. The reproductive material might also be cells suitable for nuclear transfer for cloning the parent animal [11], or nuclei from such cells. The reproductive material need not be in the same form in each of the paired samples in the collection. For example, it might comprise spermatozoa in some paired samples and ova in others.

Either or both of the genetic screening material and the reproductive material may be preserved to facilitate storage. The preservation of genetic screening material without affecting the suitability of the material for genetic analysis is routine. For example, freezing can be used but, depending on the nature of the sample, other suitable procedures include cryopreservation and lyophilisation. The preservation of suitable reproductive material is also known. For instance, gamete preservation without affecting viability has been reported for many species. Suitable procedures have long been used for preserving spermatozoa, for example for artificial insemination in mammals such as cattle, horses and humans, and similar approaches for murine gametes have recently been reported [12].

If a mutagenised animal is mated with a non-mutagenised animal, the resulting FI generation will be heterozygous for the mutations. Although samples of reproductive material for the collection can be prepared from this FI generation, meiosis means that only half of these FI gametes will carry any particular mutation present in the parent animal and so not all of them are capable of producing progeny carrying that mutation. In any given aliquot of a gamete sample, however, the mutation will be represented and so each aliquot will contain reproductive material suitable for generating progeny carrying that mutation.

If a gamete carrying a mutation is crossed with a non-mutagenised gamete, the resulting progeny will be heterozygotic for that mutation. If the mutation is recessive, this will not result in a mutant phenotype. In order to produce animals homozygous for the mutation of interest, further breeding may be necessary. This breeding is not required to carry out screening according to the invention, however, since the screening is independent of phenotypic observation; rather it is necessary in order to observe the phenotype caused by a recessive mutation.

The two samples in each of the paired samples in the collection might, in practice, be identical. For instance, gametes can be used as genetic screening material and as reproductive material. However, the process of screening the samples of genetic screening material will destroy the viability of the gametes, whereas the gametes which make up the reproductive material must remain viable in order to fertilise other gametes and generate progeny. It is clear, therefore, that where the two components are identical they do, in fact, have different functions.

Furthermore, the process of screening a collection depletes the supply of genetic screening material. If gametes are used for both samples in a pair, the process of screening the collection will also deplete the available reproductive material. Thus the genetic screening material preferably does not comprise gametes but comprises somatic tissue which is abundantly available.

Indeed, this emphasises the advantages of using paired samples. Each of the paired samples contains two representations of a mutagenised animal's genetic constitution. Readily available material is used for screening the population for mutants. It is not necessary to screen a living population and then breed from individuals in the population, since each sample of genetic screening material is paired with a sample of reproductive material which can be used to generate an animal carrying the same mutation as was identified during screening. In effect, therefore, the mutations in a population of mutagenised animals are immortalised and animals carrying particular mutations of interest can be produced easily without the need to maintain a living animal population.

It is anticipated that the collections according to the invention will be useful where the sequence of a gene or an expressed sequence of interest has been elucidated. This sequence can be used as a basis for screening a collection to identify those paired samples which represent an animal carrying a mutant copy of the sequence. The sample of reproductive material from the paired samples can then be used to produce an animal carrying a mutant copy of the gene of interest.

Although the samples must be "paired" this does not mean, for instance, that the samples must be stored together or even in close proximity. Nor does it mean that there must only be two samples per mutagenised animal. Rather, it means that there must be some way of identifying the sample of reproductive material which corresponds to a given sample of genetic screening material, and vice versa. Whether this is achieved by storing the samples in pairs, for instance, or by storing samples individually in conjunction with an index which can be used to identify the corresponding samples, is a matter of choice and convenience. A mutagenised animal might be represented by more than two different sources of sample (eg. its spleen, its liver, its brain, and its gametes), each of which might be stored in aliquots, but these are still "paired" according to the invention.

Furthermore, it might be preferred to pool samples. For instance, the samples of genetic screening material from more than one mutagenised animal might be stored together, but to retain "pairing" it must be possible to identify the samples of reproductive material which correspond to the pooled material. For instance, if samples of genetic screening material from ten animals are stored together, these samples can be screened together, and if a mutation of interest is identified in the pooled samples then the corresponding samples of reproductive material must be identified. Obviously, where pooling is used it is necessary to deconvolute the pooling step in order to identify the contributions of the individual samples. For instance, the progeny produced from pooled gametes will represent more than one mutagenised parent animal and it will be necessary to screen the progeny to determine which carry the mutation of interest. Therefore pooling concentrates the samples for screening purposes, but dilutes the mutation of interest for reproductive purposes. However, it must always be possible to identify the sample of reproductive material which corresponds to a sample of genetic screening material, and vice versa, even if pooling has been used. According to the invention, there is also provided a collection of samples of reproductive material from a plurality of mutagenised animals, wherein each reproductive material sample is paired with a sample of genetic screening material from the same animal. Similarly, there is provided a collection of samples of genetic screening material from a plurality of mutagenised animals, wherein each sample of genetic screening material is paired with a sample of reproductive material from the same animal.

According to a further aspect of the invention, there is provided a method for identifying samples in a collection according to the invention which carry a mutation in a gene of interest, comprising the step of screening the samples of genetic screening material in the collection to identify those samples which carry said mutation.

The screening method can be any technique for detecting sequence differences. Suitable methods include nucleotide sequencing, single stranded conformation polymorphism (SSCP) [13], denaturing gradient gel electrophoresis, sequencing by hybridisation to an oligonucleotide array [14], chemical cleavage of mismatches, RNase cleavage, and mismatch recognition by DNA repair enzymes such as MutS [15].

Depending on the method used, the genetic screening material is preferably screened using a probe comprising the wild-type sequence of the gene of interest.

The preferred screening method is SSCP, which has proven useful for detection of multiple mutations and polymorphisms [eg. reference 16]. SSCP sensitivity varies with the length of sequence being analysed and the optimal size appears to be 150-300 nucleotides. Preferably the screening method is fluorescence SSCP (fSSCP), which can be analysed using an ABI™ fluorescent DNA sequencing machine.

Where the screening method is optimal for sequences shorter than the complete gene of interest, it may be necessary to screen the gene in a number of segments which span its whole length. Where genomic DNA is being screened, it may be preferable to screen the gene in segments corresponding to the exons.

Usually it will be necessary to amplify the gene of interest, or segment thereof, prior to detecting any sequence differences. Preferably this amplification is by PCR, utilising primers unique to the sequence to be tested. Where the screening method utilises fluorescence or radioactivity, the PCR products should be suitably labelled.

There are three general approaches for screening the samples. Firstly, samples are screened in order to provide mutation detection for a single gene using a probe which is unique to that gene. Secondly, samples are screened using a mixture of unique probes spanning a gene, providing simultaneous mutation detection across the gene examined, for instance where the complete gene is longer than the sensible limit for amplification or for mutation detection. Thirdly, samples are screened using a mixture of probes for different genes, providing simultaneous mutation detection for several different genes. Any of these three approaches can be combined with the pooling of different samples so that a number of samples can be screened simultaneously (multiplexing).

Preferably the screening method is optimised to allow simultaneous screening for a plurality of different mutations and/or simultaneous screening of a large number of samples. Such optimisation accelerates the screening of the collection.

Where the gene is screened in segments, for instance, it is preferable to screen a plurality of those segments for mutations simultaneously. This will involve the use of a plurality of nucleic acid probes collectively spanning the gene of interest, each probe being unique. These probes will usually be produced by amplifying the segments using different sets of amplimers for each segment.

Once the presence of a mutation has been detected, the actual sequence mutation can, if desired, be determined by sequencing nucleic acid derived from the particular mutant gamete sample or from an animal generated therefrom.

According to a further aspect of the invention, there is provided a method for selecting from a collection according to the invention a paired sample which carries a mutation in a gene of interest, comprising the steps of: screening and identifying a sample of genetic screening material as described above; and selecting the paired sample which comprises said identified sample.

According to a further aspect of the invention, there is provided a method for producing an animal carrying a mutation in a gene of interest, comprising the steps of: selecting a paired sample as described above; and using the sample of reproductive material from said paired sample to produce progeny carrying said mutation. According to a further aspect of the invention, there is provided a method for identifying a phenotype associated with a mutation in a gene of interest, comprising the steps of: producing progeny as described above; and examining the progeny for aberrant phenotypes.

According to a further aspect of the invention, there is provided a process for generating a non- human animal model for a human disease known to be caused by a defect in a gene of interest, comprising the steps of: screening a collection according to the invention for a paired sample carrying a mutation in the animal homologue of the gene of interest; using the sample of reproductive material in said paired sample to produce an animal carrying the mutation.

Brief description of the attached figures Figure 1 shows an example of generating of a paired sample library from mice.

Figure 2 shows an example of screening the genetic screening material from a collection of paired samples for mutations.

Figure 3 shows an example of how the reproductive material from a collection of paired samples can be used to regenerate a mouse corresponding to a genetic screening material sample of interest.

Figure 4 shows a mouse breeding scheme and illustrates the steps in breeding mice which contain mutations, in order to study phenotypes associated with mutations.

Examples

Example 1 - Producing a collection from mice Random mutations are induced in the genome of the premeiotic spermatogonia of male mice (1 - see Fig. 1) using ethylnitrosourea (ENU). Three separate doses of 100 mg/kg body weight ENU are injected interperitoneally, with each injection separated by a one week interval. The animals undergo a period of sterility (usually for 8 to 14 weeks), after which they can be mated to non-mutagenised females (2) to produce FI generation offspring (3) which carry heterozygous mutations (of paternal origin) in the genome of their somatic and germ tissue.

At sexual maturity (6 weeks), the FI animals are sacrificed. Gametes (sperm (5) or ova (7)) and a sample of somatic tissue (for instance spleen (4) and (6)) are harvested from each FI mouse. The samples (8, 9, 10, 11) are given bar-code identifiers which link their common origin, and the samples are stored separately. The gametes are stored in an array (13) at -196°C [ref. 12] and DNA extracted from the somatic tissue is stored in a similar array (12) at -20°C. Each well in the gamete array corresponds to a well in the somatic array which contains material taken from the same mouse.

The somatic tissue serves as a source to test for DNA mutations (genetic screening material), and the gametes (reproductive material) comprise an immortal source of material to regenerate a mouse corresponding to any of the somatic samples.

To generate offspring for a large collection of paired samples, 300 males are treated in this way. Approximately 100 of these mice will be permanently sterile; the remaining 200 are mated with 2 females each. This results in the generation of 4000 FI offspring per month. In this way a collection of paired samples comprising over 30000 animals can be produced within a year, including the time required to establish the ENU treated breeding males.

Example 2 - Producing a collection from rabbits

For medium size animals, it is not always feasible to generate all of the offspring in a short period of time, due to the resources required for each animal. A collection of paired samples can be generated over a longer time period for animals such as rabbits.

40 male rabbits are treated with ENU and mated to non-mutagenised females. Females have 3-5 litters per year with a gestation of 1 month and litter size ranges from 4-10 (average 5-6). On average a male mated to a single female will thus produce 24 offspring per year. Using 40 ENU treated males in continuous permanent matings will result in nearly 1000 FI mutated animals per year. Rabbits reach sexual maturity in 4 to 6 months, at which time the FI animal is sacrificed. Gametes and a sample of somatic tissue (for instance liver) are harvested from each animal. This process can be continued to produce a collection which increases in size with time.

Example 3 - screening a collection Members of the SOX gene family have high sequence homology in a portion of each SOX gene. Each member contains an approximately 240 bp DNA sequence corresponding to a 80 amino acid segment (which is called an "HMG box") that shows 60% or greater amino acid similarity between members. This defines the SOX gene family. Outside this region the genes are very different. The HMG box is a DNA binding domain and the SOX genes which have been studied bind to DNA via this region of the protein and are likely modulators of gene expression ie. transcription factors.

There are about 20 different Sox genes known in mouse and a similar number in humans. Several of these have been implicated in disease. SOX9 is a human gene which, when mutated, causes a neonatal lethal chondrodysplasia called campomelic dysplasia (CD). In addition, in CD patients who have XY sex chromosomes and should develop as male, 3/4 develop as female (XY sex reversal). Sox-4 when mutated in mouse leads to problems in B cell formation and a cardiac condition which is the same as a cardiac development defect seen in man.

It is of obvious interest to discover the function of SOX genes, given their likely role as transcription modulators involved in developmental processes. For some SOX genes, little sequence is known; for others the entire cDNA and genomic structure, plus expression patterns are known. One gene which has been studies fairly extensively is Sox-3 (both mouse Sox-3 and the human homologue SOX3). Whilst a mouse Sox-3 knockout (artificially generated null allele) has not been reported, in human there is an individual who has a deletion which removes (minimally) SOX3 and the factor IX (blood clotting) gene and who has mental retardation and haemophilia. SOX3 may be linked to the retardation. Mouse Sox-3 is expressed in the developing central nervous system, adding support to this hypothesis.

A paired sample collection is useful for identification and generation of mice containing mutations within the Sox-3 gene to study the role of Sox-3 in mental retardation. A collection of paired samples is generated as in Example 1 , except that the genetic screening material in each paired sample comprises genomic DNA extracted from the somatic samples.

To accelerate the screening procedure, genetic screening material is pooled (see Figure 2) so that three samples (wells Al, A2 and A3 of DNA array (12)) are screened simultaneously. The Sox-3 gene in all of the pooled samples is amplified by PCR. Rather than amplify the whole ORF using a single pair of primers, overlapping segments (21) of the complete gene (20) are amplified separately, using different pairs of PCR primers labelled with 3 differently coloured fluorochromes. The differently coloured amplified DNA (22, 23, 24) is then combined (25) to give material for fSSCP analysis.

Each pool of amplified DNA (eg. 25) is loaded in a separate lane of a polyacrylamide gel in an

ABI 377 DNA sequencing machine under SSCP conditions. Therefore each lane represents amplified Sox-3 from the screening material from three different animals. Altered fragment mobility of a PCR product indicates that one of the three animals carries a mutation in Sox-3, and the colour of the band indicates which region of Sox- 3 is mutated. The first lane in Figure 2 shows extra bands with altered fragment mobilities (26). For each lane which contains a mutation, the screening material from the three animals which were originally pooled are tested separately by SSCP to identify which of the three carries the mutation. The mutated sequence is also obtained, to define the mutation precisely.

The DNA sample in well A17 of an array (12) was found by fSSCP to carry a mutant copy of Sox-3. The corresponding cryopreserved reproductive material (well A17 of the array (13)) was therefore used [12] to generate mice carrying this mutation (Figure 3). The material from well A17 of the plate of reproductive material (13) was thawed and used for in vitro fertilisation of gametes from a normal mouse. The developing embryo was implanted in a pseudo-pregnant female mouse (31), to produce F2 mice (32).

As the animals from which the samples were taken were heterozygous for the mutation, half of the F2 mice (32) generated from the stored reproductive material will contain the identified mutation. Dominant phenotypes will be exhibited by these F2 mice but, to study recessive phenotypes caused by the mutations, it is necessary to interbreed the F2 mice to generate offspring which are homozygous for the mutation (Figure 4). These animals are examined for a phenotype related to retardation, such as neural defects and impaired learning, as a means to study the function of the Sox-3 gene.

The imtial ENU treatment generates multiple mutations per mouse. In addition to the mutation in the gene of interest, therefore, there exist further "background" mutations. In each animal from which the paired samples were derived there are approximately 100 background mutations. Half of these mutations are lost at random with each mating to a normal mouse, so at the F3 generation, there are 49 random background mutations. Of these, 6 at random will be homozygous. As each sib and half-sib has different background mutations, multiple mice can be examined to insure that the observed phenotype is resulting from the selected mutation (which is homozygous in all of the animals) (Figure 4). Further breeding to normal mice continues to segregate away background mutations, and after approximately 8 generations, the background mutations have been removed.

It will be understood that the invention is described above by way of example only and modifications may be made whilst remaining within the scope and spirit of the invention.

Industrial applicability

Collections according to the invention allow the screening of animal populations for mutations without knowledge of the phenotype caused by the mutation and without the need for maintaining a large population of living animals. Although screening is carried out post mortem, where a mutation of interest is identified in the collection, an animal carrying that mutation can be produced.

Collections according to the invention are useful in the discovery and characterisation of genes of interest, with a view towards the identification and development of therapeutic agents or targets for therapeutic methods. They are also useful for identifying gene defects involved in disease, thus allowing the development of diagnostics.

Screening methods according to the invention are thus useful for identifying biological samples carrying mutations in useful genes and for producing animals carrying mutations in those genes. These animals can be used for phenotypic characterisation of the gene and as models of disease.

Identification of those samples in the collection which carry a mutant gene permits the subsequent assessment of phenotypes resulting from the alteration of gene function and provides a model organism for further disease research. These organisms can be used to model human disease. The mutant genes identified, or their wild-type counterparts, may be useful targets for medical, therapeutic, or diagnostic applications.

References (the contents of which are hereby incorporated)

1. Friedrich (1996) Moving beyond the genome projects. Nature Biotech 14, 1234-1237.

2. Fleischmann RD et al. (1995) Whole-genome random sequencing and assembly of Haemophilus influenzae Rd. Science 269, 496-512.

3. The genome directory (1995) Nature 371, supplement.

4. WO97/44485 (Hexagen Technology Limited). 5. Russell & Hunsicker (1983) Environ Mutagen 5, 498.

6. Russell & Hunsicker (1984) Environ Mutagen 6, 390.

7. Rossant J & Hopkins N (1992) Of fin and fur: mutational analysis of vertebrate embryonic development. Genes Dev 6,1.

8. Rinchik E (1991) Chemical mutagenesis and fine-structure functional analysis of the mouse genome. TIG 7, 15-22.

9. Russell LB et al. (1989) Chlorambucil effectively induces deletion mutations in mouse germ cells. PNAS USA 86, 3704-3708. (See especially Table 2)

10. McWhir J, Selfridge J, Harrison DJ, Squires S, Melton DW (1993) Mice with DNA repair gene (ERCC-1) deficiency gave elevated levels of p53, liver nuclear abnormalities and die before weaning. Nature Genetics 5, 217-224.

11. Wilmut I, Schnieke AE, McWhir J, Kind AJ, Campbell KHS (1997) Viable offspring derived from fetal and adult mammalian cells. Nature 385, 810-813.

12. Nakagata (1995) Studies on cryopreservation of embryos and gametes in mice. Exp. Anim. 44, 1-8.

13. Orita M, Iwahara H, Kanazawa H, Hayashi K, Sekiya T (1989) Detection of polymorphisms of human DNA by gel electrophoresis as single-strand conformation polymorphisms. PNAS USA 86, 2766-2770.

14. Chee M et al. (1996) Accessing genetic information with high-density DNA arrays. Science 274, 610-614.

15. Eng C & Vijg J (1997) Genetic testing: the problems and the promise. Nature Biotech 15, 422-426. (see especially "Best Bets for genetic testing: mutation scanning methods").

16. Glavac & Dean (1993) Optimization of the single-strand conformation polymorphism (SSCP) technique for detection of point mutations. Hum Mut 2, 404-414.

Claims

1. A collection of paired samples from a plurality of mutagenised animals, wherein each of the paired samples comprises: a first sample comprising genetic screening material from a parent animal; and a second sample comprising reproductive material from that same parent animal.

2. A collection according to claim 1, wherein said plurality comprises 10000 or more.

3. A process for producing a collection according to claim 1, comprising the steps of obtaining pairs of samples from a plurality of mutagenised animals: a first sample comprising genetic screening material; and a second sample comprising reproductive material.

4. A process according to claim 3, wherein the animals are sacrificed before, during, or after obtaining the samples.

5. A process according to claim 3 or claim 4, which additionally includes the step of mutagenising a plurality of animals prior to obtaining gamete samples.

6. A process according to claim 5, wherein said mutagenising step is exposure to a chemical mutagen.

7. A process according to claim 6, wherein said chemical mutagen is an alkylating agent.

8. A process according to claim 7, wherein said alkylating agent is ethylnitrosourea.

9. A collection according to claim 1 or a process according to claim 3, wherein said animal is a mouse.

10. A collection according to claim 1, wherein either or both of the first and second samples are preserved.