News Physiol Sci 19: 105-109, 2004;
doi:10.1152/nips.01512.2003
1548-9213/04 $5.00
News in Physiological Sciences, Vol. 19, No. 3, 105-109,
June 2004
© 2004 Int. Union Physiol. Sci./Am. Physiol. Soc.
Use of Recombinant Congenic Strains in Mapping Disease-Modifying Genes
Jana Müllerová1,2 and
Pavel Hozák1,2
1 Department of Cell Ultrastructure and Molecular Biology, Institute of Experimental Medicine, Academy of Sciences of the Czech Republic, 142 20 Prague; and
2 Department of Cellular and Molecular Biology, Third Faculty of Medicine, Charles University, 100 00 Prague, Czech Republic
 |
Abstract
|
|---|
Previous research studies have established much information about single-gene diseases. However, other genes also influencing the outcome of a disease and genes involved in complex disease remain largely unknown. Here we report on recombinant congenic strains of mice, a powerful tool for genetic dissection of a complex trait.
|
Introduction
|
|---|
The latest research results regarding the human genome open the possibility of suggesting an optimal therapy for each individual on the basis of knowledge of the patient’s genotype (12). Many genes whose mutations cause inborn single-gene diseases or cancer have been described in detail. More importantly, additional genes involved in multigenic diseases have been identified. However, it is a particular combination of their alleles that increases the risk and modifies the outcome and efficiency of therapy (2).
It is generally difficult to study these genes. The reasons are in their multiplicity, high number of mutual combinations of their alleles, low penetrance, and extrafamilial occurrence of the diseases. Recombinant congenic strains (RCS) of mice are a very efficient genetic tool, allowing the genetic dissection of genes involved in a complex trait (6). Another possibility, recombinant inbred strains (a cross between two inbred strains of mice) is the most used model to study complex traits. However, recombinant inbred strains are more effective for mapping the genes with stronger penetrance. (Various models are reviewed in Refs. 2, 4, and 7.) Because of the high homology between human and mouse genome, it is possible to expect that the same gene will relate to the same disease in humans and mice; however, the particular polymorphism can be different.
Identification of those genes modifying risk for and course of a disease could lead to an explanation of molecular mechanisms and to detection of new targets for therapy. Although the initial aim of such research is not prevention, it is also possible to assess risk factors for a disease in a healthy person by knowing his/her genotype.
Since cancer is probably the most investigated disease and it perfectly illustrates a complex trait, we will explain the principles of genetic analysis of this trait using RCS. After that, a short review of results obtained so far will be given.
|
Cancer and related genes
|
|---|
Cancer is caused by sequential mutations in one cell, which give this cell and its clones selective advantages such as fast and unregulated division, ability to induce vascularization, or ability to invade other tissue (13). Mutations causing cancer mainly affect well-known protooncogenes and tumor suppressor genes participating in regulation of cell division. In the development of this disease, a large number of nonmutated genes also play a role (1). They can influence the susceptibility to cancer after carcinogen exposition, tumor number and size, the rate of tumor growth, the site of tumor formation, histological type, and progress toward malignancy (6). These genes are called tumor susceptibility genes (6) or tumor modifier genes (TMGs) (2). Similarly, the alleles are called susceptible or resistant, depending on whether they support or inhibit tumorigenesis. Besides carcinogen metabolism, TMGs are associated with tissue’s ability to grow, to invade, to differentiate, to induce appropriate vasculature, to stimulate an immune response, to control genome stability, or to repair damaged DNA. All of them obviously can be critical to tumor progression (2).
With respect to complexity of the disease, the influence of both single gene effects as well as gene interactions can be expected. It even appears that gene interactions are much more frequent than predicted. The high frequency of these interactions suggests the existence of synergic pathways and molecular interactions (e.g., protein-protein interactions) as well as the existence of complex interactions (23). Except for reciprocal interactions, or epistasis, counteracting interactions are very common. In the case where two loci are involved in counteracting interaction, their alleles confer either susceptibility or resistance depending on a specific genetic combination. The same allele can have opposite effects in a different background. The opposite effects of the same allele in combination with the partner locus allele mask each other. Consequently, in spite of the relatively large effects of these loci together, there is no effect if each of them is considered alone. At present, this type of interaction has been detected in many loci and with many biological traits (24).
Originally, it was suggested that most TMGs influence the cellular processes related to neoplastic development, rather than systemic factors (6). However, as the number of identified candidate loci increased, it was found that their chromosomal positions overlap less randomly than expected. This implies that multiple TMGs influence several types of cancer by functioning systematically or by functioning in one type of tissue present in different organs, e.g., in the epithelial tissue of the lung and of the colon (9).
The most difficult problems with mapping TMGs are their low penetrance, phenotypic variability due to the influence of environment, and especially their complex interactions. Frequently, the individual alleles are not intrinsically susceptible or resistant, but their effect is influenced by the genotype at the interacting locus. Since every individual has a unique combination of alleles of TMGs, we could not resolve their interactions and effect without decreasing genetic variability. For that reason, we cannot use any method of genetic mapping such as linkage analysis in affected families, allele-sharing methods, or association studies in human populations. We have to use the fourth method: animal crosses.
The most suitable model is a mouse, because it is possible to cross it to obtain a high number of individuals with a defined genome. The entire experimental group can be bred under identical conditions to eliminate environmental influence on phenotype, and, conveniently, mice have a short generation period.
|
RCS
|
|---|
The RCS can be used to analyze molecular mechanisms of any biological process, such as atherosclerosis or infectious and metabolic diseases, especially those complex traits for which the two parental strains differ (19). Recently, many new series of RCS have been founded especially to study particular complex traits (11, 15).
RCS were first developed in the Netherlands, and they were designed to analyze specifically the traits under multigenic control (8). RCS were created by crossing two inbred strains that strongly differ in reaction to transplacental carcinogen treatment. The former, e.g., strain A, is susceptible because it develops many large tumors with malignant appearance. The latter, strain B, develops few small tumors, and that is why it is called resistant. F1 mice were backcrossed to strain A, and the N1 generation was also backcrossed to strain A. Mice from the following generation had 12.5% of their genome from parent B (donor strain) and the remaining 87.5% of the genome from parent A (background strain). On average, 20 litters were randomly selected, and brother-sister mating over 20 generations followed, so that all genes were in a homozygous state. In this way, a series of 20 inbred strains were founded, each carrying a different random subset of ~12.5% genes from the donor strain and the remaining 87.5% genes from the background strain. In other words, genome of the donor strain was dissected in 20 RCS in 12.5% parts and that part is different in each of 20 RCS (Fig. 1
). Genes of a multigenic trait were also dissected into oligo or single-gene traits (6, 19).
View larger version (15K):
[in this window]
[in a new window]
|
FIGURE 1. Development of recombinant congenic strains (RCS) and a resulting group of RCS. A: development of RCS when donor and background parental strains are crossed to obtain the F1 generation. This generation is then backcrossed to the background parental strain, and the N1 generation is also backcrossed to the background parental strain. Mice from the next generation are brother-sister mated over 20 generations to obtain the inbred strain. B: RCS (Ocb/Dem as an example) as a group of inbred strains, each carrying a different random subset of ~12.5% genes from the donor parental strain and the remaining 87.5% genes from the background parental strain.
|
|
The only weakness of the RCS model is paradoxically connected with its greatest advantage: the very decreased genetic variability. In fact, each set of RCS is derived from two parental strains, and therefore the study is limited to only two alleles of a gene. The candidate allele may also have significant effect in one genetic background but none in another. However, because TMG mapping in humans is difficult and TMGs are still virtually unknown to date, their identification on the basis of murine TMGs is of crucial importance (6).
|
Mapping with RCS
|
|---|
The first step of course is to identify which RCS inherited the part of the donor’s genome where TMGs are. For example, when studying TMGs of lung cancer, RCS as well as parent strains have to undergo transplacental carcinogen treatment. The resulting phenotype is determined by histological analysis of tumor tissue.
More specifically, RCS OcB/Dem are used to study lung cancer. The background susceptible strain is O20/A (O20), and the resistant donor strain is B10.O20/Dem (B10.O20). O20 develops higher number of larger tumors than B10.O20. Moreover, the O20 tumors are mainly of the solid, alveolar type with frequent secondary papillary structures in the central part. Nuclear pleiomorphy may be very extensive, and lymphocyte infiltration is found rather frequently, in contrast to the B10.O20 tumors. B10.O20 mice develop almost exclusively papillary adenomas with homogenous populations of tumor cells with no nuclear pleiomorphy. The majority of the 20 OcB/Dem strains develop tumors, which are for a greater part similar to those of the O20 strain, except two of them, OcB-4 and OcB-6. In strain OcB-4, about half of the lung tumors are of alveolar type without secondary papillary structures and they exhibit little or no nuclear plesiomorphism. Tumors in the strain OcB-6 are predominantly papillary, including the periphery of tumor, in this respect resembling the strain B10.O20; but in contrast to this parent, they have considerable nuclear pleiomorphy (9). Histological appearance of tumors of different strains is summarized in Table 1.
There are so many genes involved in tumorigenesis that one cannot assign the individual effect of each gene to a phenotype absolutely. Since parent strains serve as reference values, phenotype variance is also decreased in RCS. Therefore, due to parental strain we are able to recognize when an active allele is replaced with a nonactive one or when an interaction arises or perishes.
After phenotyping, genotyping allows us to find which part of the donor genome is in which OcB/Dem strain. Variable numbers of tandem repeats (called markers) are used. Individual alleles can be differentiated by PCR followed by agarose gel electrophoresis for determination of fragment length. Besides these molecular markers, the strains were typed for a number of biochemical, serological, and restriction fragment length polymorphisms. The typing data for individual RCS are deposited in the Mouse Genome Database at The Jackson Laboratory that serves as a comprehensive resource available online (http://www.informatics.jax.org). For example, RCS OcB/Dem have already been typed for 611 markers (19).
With the methods described briefly above, it is possible to select an individual strain from RCS, which differs from parent strains, and to determine which part of the donor genome it inherited and that this part is responsible for the phenotype difference. How to find then the particular gene in this region? The methods are again phenotyping and genotyping. Obviously, we cannot use the homozygous RCS; we have to generate different genotypes in loci of donor origin. Therefore, we have to cross this strain with a background strain to obtain the F2 generation.
The major advantage of RCS is that only ~12.5% of the genome segregates in a cross between RCS and its parental background strain, in contrast with crosses in which the whole genome is segregating. The effect on phenotype of each marker, sex, and interactions between pairs are tested by analysis of variance. The ultimate aim of this analysis is to find a marker whose one parent allele is connected with a certain tumor feature while the allele of the second parent origin is not (or even causes an opposite effect). Of course, the marker is not responsible for the phenotype difference. It is a gene and its alleles located close to that marker. Then this candidate locus is shortened as much as possible by detailed mapping and submitted to positional cloning.
|
Major achievements obtained by using RCS
|
|---|
So far, two candidate genes have been identified with RCS. The first is the combined hyperlipidemia gene Hyplip 1, and the second is Ptprj (receptor-type protein tyrosine phosphatase), a candidate gene for the colon cancer susceptibility locus Scc1.
Familial combined hyperlipidemia represents the most common genetic dyslipidemia, with a prevalence of 1.0–2.0%. HcB-19, a strain from RCS HcB/Dem, also has symptoms of this disease, although parent strains are normal. This finding suggests the development of spontaneous mutation de novo. After fine mapping and positional cloning, the sequencing diagnosed the presence of a nonsense mutation in gene called Hyplip 1. It codes Txnip (thioredoxin-interacting protein). Txnip binds reduced thioredoxin, a major regulator of cellular redox state, and inhibits its reducing activity. It is suggested that altered redox status downregulates the citric acid cycle, sparing fatty acids for triglyceride and ketone body production, major symptoms of combined hyperlipidemia. This is a new pathway of plasma lipid metabolism with potential clinical significance (3).
Ptprj was identified by positional cloning. It encodes a receptor-type protein tyrosine phosphatase. The coding sequences of resistant and susceptible alleles show 16 single-nucleotide differences. Furthermore, frequent deletions of Prptj, allelic imbalance in loss of heterozygosity, and missense mutations occur in human colon, lung, and breast cancer (16).
Recently, Czarnomska et al. (5) confirmed that a host genome influences a number of intestinal and mammary tumors as well as the histological type of mammary tumors. Furthermore, they found that susceptibility to both types of cancer had opposite distribution within RCS. The strains most susceptible to intestinal tumor were the most resistant to mammary tumors and vice versa. They used OcB/Dem mice carrying multiple intestinal neoplasia (Min) mutation in the adenomatous polyposis coli (Apc) gene. Other important results are summarized in Table 2.
|
Conclusions
|
|---|
The RCS are a very efficient tool capable of genetic dissection of complex traits and even of identification of gene interactions. The results from lung cancer research on RCS have shown that one locus can be found by testing ~29 F2 mice and one interaction per 35 F2 mice (23). Mapping on RCS is very sensitive; it is not rare for a susceptible allele to be found in a resistant strain and vice versa (24).
The reliability of identification of responsible genes in RCS mapping increases when the same candidate locus is present in more than one strain. In addition, the candidate locus can be further shortened: the recombinant event can occur in a different locus in each strain, but the candidate loci are present only in a segment from the donor parent.
One can expect that the importance of RCS will significantly increase in future. So far, mostly quantitative traits (e.g., size and number of tumors) have been analyzed. However, quantitative aspects (e.g., tumor type, localization) might prove to be of even more importance. Tripodis and Demant (22) and Czarnomska et al. (5) have already published the first studies about qualitative traits.
Identification of all genes involved in the progression of a disease and especially their functions will enable us to understand the molecular mechanisms of these processes and to discover new targets for therapy. This progress will be the basis for future individualized treatment and prevention based on knowledge of a patient’s genotype and for reliable prognosis when the most efficient way of therapy can be determined.
|
Acknowledgments
|
|---|
We thank Dr. Peter Demant for his very helpful comments on the manuscript.
This work was supported by a grant from the Institute of Experimental Medicine, Academy of Sciences of the Czech Republic (AVOZ5039906).
|
References
|
|---|
- Balmain A, Gray J, and Ponder B. The genetics and genomics of cancer. Nat Genet 33, Suppl: 238–244, 2003.
- Balmain A and Nagase H. Cancer resistance genes in mice: models for the study of tumor modifiers. Trends Genet 14: 139–144, 1998.[CrossRef][ISI][Medline]
- Bodnar JS, Chatterjee A, Castellani LW, Ross DA, Ohmen J, Cavalcoli J, Wu C, Dains KM, Catanese J, Chu M, Sheth SS, Charugundla K, Demant P, West DB, de Jong P, and Lusis AJ. Positional cloning of the combined hyperlipidemia gene Hyplip1. Nat Genet 30: 110–116, 2002.[CrossRef][ISI][Medline]
- Brockmann GA and Bevova MR. Using mouse models to dissect the genetics of obesity. Trends Genet 7: 367–376, 2002.
- Czarnomska A, Krysiak E, Piskorowska J, Sitarz M, Pysniak K, Pilcik T, and Demant P. Opposite effects of modifiers of the ApcMin mutation in intestine and mammary gland. Cancer Res 63: 4533–4537, 2003.[Abstract/Free Full Text]
- Demant P. Genetic resolution of susceptibility to cancer–new perspectives. Semin Cancer Biol 3: 159–166, 1992.[Medline]
- Demant P. Cancer susceptibility in the mouse: genetics, biology and implications for human cancer. Nat Rev Genet 4: 721–734, 2003.[ISI][Medline]
- Demant P and Hart AA. Recombinant congenic strains–a new tool for analyzing genetic traits determined by more than one gene. Immunogenetics 24: 416–422, 1986.[CrossRef][ISI][Medline]
- Fijneman RJ, van der Valk MA, and Demant P. Genetics of quantitative and qualitative aspects of lung tumorigenesis in the mouse: multiple interacting susceptibility to lung cancer (Sluc) genes with large effects. Exp Lung Res 24: 419–436, 1998.[ISI][Medline]
- Fortin A, Cardon LR, Tam M, Skamene E, Stevenson MM, and Gros P. Identification of a new malaria susceptibility locus (Char4) in recombinant congenic strains of mice. Proc Natl Acad Sci USA 98: 10793–10798, 2001.[Abstract/Free Full Text]
- Fortin A, Diez E, Rochefort D, Laroche L, Malo D, Rouleau GA, Gros P, and Skamene E. Recombinant congenic strains derived from A/J and C57BL/6J: a tool for genetic dissection of complex traits. Genomics 74: 21–35, 2001.[CrossRef][ISI][Medline]
- Hamet P. Genes and hypertension: where we are and where we should go. Clin Exp Hypertens 21: 947–960, 1999.[Medline]
- Hanahan D and Weinberg RA. The hallmarks of cancer. Cell 100: 57–70, 2000.[CrossRef][ISI][Medline]
- Mori N, van Wezel T, van der Valk M, Yamate J, Sakuma S, Okumoto M, and Demant P. Genetics of susceptibility to radiation-induced apoptosis in colon: two loci on chromosomes 9 and 16. Mamm Genome 9: 377–380.
- Reifsnyder PC and Leiter EH. Deconstructing and reconstructing obesity-induced diabetes (diabesity) in mice. Diabetes 51: 825–832, 2002.[Abstract/Free Full Text]
- Ruivenkamp CA, van Wezel T, Zanon C, Stassen AP, Vlcek C, Csikos T, Klous AM, Tripodis N, Perrakis A, Boerrigter L, Groot PC, Lindeman J, Mooi WJ, Meijjer GA, Scholten G, Dauwerse H, Paces V, van Zandwijk N, van Ommen GJ, and Demant P. Ptprj is a candidate for the mouse colon-cancer susceptibility locus Scc1 and is frequently deleted in human cancers. Nat Genet 31: 295–300, 2002.[CrossRef][ISI][Medline]
- Serreze DV, Bridgett M, Chapman HD, Chen E, Richard SD, and Leiter EH. Subcongenic analysis of the Idd13 locus in NOD/Lt mice: evidence for several susceptibility genes including a possible diabetogenic role for beta 2-microglobulin. J Immunol 160: 1472–1478, 1998.[Abstract/Free Full Text]
- Serreze DV, Prochazka M, Reifsnyder PC, Bridgett MM, and Leiter EH. Use of recombinant congenic and congenic strains of NOD mice to identify a new insulin-dependent diabetes resistance gene. J Exp Med 180: 1553–1558, 1994.[Abstract/Free Full Text]
- Stassen AP, Groot PC, Eppig JT, and Demant P. Genetic composition of the recombinant congenic strains. Mamm Genome 7: 55–58, 1996.[CrossRef][ISI][Medline]
- Szymanska H, Sitarz M, Krysiak E, Piskorowska J, Czarnomska A, Skurzak H, Hart AA, de Jong D, and Demant P. Genetics of susceptibility to radiation-induced lymphomas, leukemias and lung tumors studied in recombinant congenic strains. Int J Cancer 83: 674–678, 1999.[Medline]
- Tripodis N and Demant P. Genetic linkage of nuclear morphology of mouse lung tumors to the Kras-2 locus. Exp Lung Res 27: 185–196, 2001.[ISI][Medline]
- Tripodis N and Demant P. Genetic analysis of three-dimensional shape of mouse lung tumors reveals eight lung tumor shape-determining (Ltsd) loci that are associated with tumor heterogeneity and symmetry. Cancer Res 63: 125–131, 2003.[Abstract/Free Full Text]
- Tripodis N, Hart AA, Fijneman RJ, and Demant P. Complexity of lung cancer modifiers: mapping of thirty genes and twenty-five interactions in half of the mouse genome. J Natl Cancer Inst 93: 1484–1491, 2001.[Abstract/Free Full Text]
- Van Wezel T, Lipoldova M, and Demant P. Identification of disease susceptibility genes (modifiers) in mouse models: cancer and infectious diseases. In: Genotype to Phenotype (2nd ed.). Oxford, UK: BIOS Scientific, 2001, p. 107–129.
- Vladimirov V, Badalova J, Svobodova M, Havelkova H, Hart AA, Blazkova H, Demant P, and Lipoldova M. Different genetic control of cutaneous and visceral disease after Leishmania major infection in mice. Infect Immun 71: 2041–2046, 2003.[Abstract/Free Full Text]
Copyright © 2004 by the Int. Union Physiol. Sci./Am. Physiol. Soc.
Top
Introduction
Cancer and related genes
