Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
4 Genetic Management of Breeding Colonies Different breeding systems and genetic-engineering methods have been used to produce strains and stocks of rodents for particular experimental purposes inbred strains; coisogenic, congenic, and transgenic strains; re- combinant inbred strains; hybrid strains; and outbred stocks. Outbred stocks are used primarily when genetic heterogeneity is desired and are not useful when a controlled genotype is required. However, the loss of heterozygos- ity cannot be completely avoided in propagating outbred stocks, because the breeding population is necessarily finite. GENETICALLY DEFINED STOCKS Regardless of the breeding system or genetic manipulation used to pro- duce a particular strain, some practices are recommended to maintain high genetic quality. Details of breeding systems used to develop various types of strains can be found elsewhere (Bailey, 1981; Green, 1981a). Here we describe the management of breeding colonies of already-developed strains. . Pedigrees Using a pedigree method allows the parentage of individual experimental animals to be traced; aids in selection of parental pairs to avoid the inadvertent fixation of unwanted mutations, especially mutations that would affect repro- ductive performance; and maximizes genetic uniformity within a strain. 35
36 Traceability RODENTS: LABORATORYANIMAL MANAGEMENT Mutations occur continually in any breeding stock. Many of these mutations are recessive and, when hon~ozygous, will be expressed as unde- sirable traits. When such a mutation is expressed, it is necessary to rid the breeding colony of copies of the mutation that might be carried as a het- erozygous gene by individuals that are normal in phenotype. Use of a pedigree system that records the parents of each individual makes it pos- sible to identify relatives of the affected individual, and they can be tested for the presence of the mutation or eliminated from the colony. It is also desirable to mark the animals with their pedigree identification. Selection of Parental Pairs Reproductive performance, even within a highly inbred strain, can vary greatly. Environmental factors undoubtedly cause much of that variation, but spontaneously occurring mutations that adversely affect breeding per- formance are also contributing factors. To avoid extinction of a strain, the individuals selected for propagating it should be those with the best repro- ductive performance. Reproductive performance can be evaluated retroac- tively by examining a pedigree, that is, the reproductive performance of several generations of offspring can be used in evaluating the breeding performance of the original pair and can aid in avoiding the accidental incorporation or accumulation of deleterious recessive mutations. To en- sure continuation of a strain, several families or lines should be maintained for two to three generations until one pair in each generation is retroactively chosen as the pair from which breeders in all subsequent generations will be derived. This practice not only ensures selection of reproductively fit indi- viduals to propagate the strain but also maximizes genetic uniformity, as described below. Genetic Uniformity The purpose of producing an inbred strain is to achieve genetic unifor- mity among individuals. That allows a greater degree of reproducibility in experiments than is possible if heterogeneous individuals are used. How- ever, total genetic uniformity is never achieved, because new mutations occur. Each new mutation has a 25 percent chance of becoming fixed in an inbred strain (Bailey, 19791. The gradual accumulation of such mutations and the resulting genetic changes are called genetic drift. Because of the random occurrence of mutations, genetic drift will involve different genes in two separately maintained sublines of a strain. Over time, the sublines will become increasingly different from each other; this tendency is called
GENETIC MANAGEMENT OF BREEDING COLONIES 37 subline divergence. Bailey has estimated that separately maintained sub- lines will diverge at the rate of approximately one new mutation every two generations (Bailey, 1978, 1979, 1982~. Even within one breeding colony, subline divergence can occur if the propagation of family branches is al- lowed to continue indefinitely. Another source of subline differences is the genetic heterogeneity present in a strain at the time of subline separation. Many of the early substrains of common inbred strains were separated before the strain had been highly inbred; for example, mouse substrains C57BL/6 and C57BL/10 were sepa- rated from the C57BL strain when it had been inbred for only about 30 generations. That is more than the 20 generations conventionally accepted as the definition of an inbred strain, but the amount of heterogeneity, al- though small in comparison with the total number of genes, is still suffi- cient to result in subline differences. For example, according to Bailey's estimates, one would expect about 14 fixed differences between substrains C57BL/6 and C57BL/10 caused by the presence of unfixed genes at the time of separation. Bailey also showed that the probability of there being no heterogeneity within an inbred strain does not reach 0.99 until after 60 generations of brother x sister inbreeding (Bailey, 1978~. The practical consequence of subline divergence for research is that animals from differ- ent sublines might respond differently in identical experiments, and the difference in responses could lead to misinterpretation of the experimental results. A corollary is that no subline (or substrain) can be considered a reference standard, because all sublines undergo changes with time. Cryopreservation might offer the only means to arrest such changes. Nev- ertheless, it is wise to obtain breeders periodically from the original source colony, to maximize homogeneity between two colonies. A general prac- tice is to do that after 10 generations of separation. Within a breeding colony, pedigree management can be used to maxi- mize genetic uniformity. One pair in each generation can be selected on the basis of breeding performance, to be the common ancestral mating for all progeny. So that all animals at any time can be traced to a single ancestral pair, the number of generations of any branch other than the common ances- tral branch is limited, depending on the number of animals that are pro- duced for experimental use, the productivity or the average number of breeding pairs of progeny expected from a single mating, and the reproductive life span of breeders. Because most commonly used inbred strains today are highly inbred, breeding selection is not effective in increasing reproductive performance. Rather, selection is made to avoid deleterious mutations that would cause a decrease in reproductive performance. The prevalence and rate of such mutations are unknown, but distinct reductions in reproductive performance within family branches have been observed in large breeding colonies. Be
38 R ODENTS: LAB ORA TOR Y ANIMAL MANA CEMENT cause increases in reproductive performance are rare, mutations that are advantageous to reproduction are probably extremely rare. Pedigree identification of animals used as parents for the production of hybrids is advised so that mutations or irregularities can be traced. How- ever, pedigree management is not necessary, because there is no propaga- tion of lines beyond that of the F1 generation. Foundation or Nucleus Colonies, Expansion Colonies, and Production Colonies In large breeding operations, it is often practical for management pur- poses to subdivide the breeding colony of each strain into separate groups- a foundation colony (sometimes called a nucleus colony), an expansion colony, and a production colony that are maintained in separate facilities. A foundation colony is a breeding colony of sufficient size to propagate the strain (following the selection procedures described previously) and to pro- vide breeding stock to an expansion colony. The purpose of an expansion colony is to increase the number of breeding pairs to a quantity adequate to support a production colony. A production colony is made up of breeders from an expansion colony; offspring are distributed for research, not used for breeding. It is more practical to be rigorous about selection practices and genetic monitoring in a foundation colony, which is relatively small, than in the larger expansion and production colonies. It is also more important to carry out those activities in the foundation colony because all the stock in the expansion and production colonies is ultimately derived from it and any change occurring in the foundation colonies will eventually be propagated throughout the entire strain. An advantage of using a separate facility for foundation colonies is that it permits microbiologic status of the foundation colony to be maintained with fewer pathogens than the other colonies. Of- ten, foundation colonies are maintained in a separate building from expan- sion and production colonies to protect against loss of a strain due to dis- ease outbreak or other catastrophe. Cryopreservation and storage of embryos can also fulfill that security requirement. In an expansion colony, it might not be practical or cost-effective to maintain detailed pedigree records or devote much time to selection. It is relatively easy, however, to keep track of the number of generations that a family or subline has been separated from the foundation stock by making a notation on the cage card each time a new mating group is made up. By limiting the number of generations outside the foundation nucleus, maximal genetic uniformity can be achieved. Unnoticed mutations (e.g., those af- fecting reproductive performance) that occur in either an expansion or a production colony will ultimately be purged because of the constant infu
GENETIC MANAGEMENT OF BREEDING COLONIES 39 sion of highly scrutinized breeding stock from the foundation colony. Trio matings (i.e., two females mated to a sibling male) are often used in expan- sion colonies for efficiency. In a production colony, especially a large one, the use of non-sib matings increases efficiency. The probability that recessive, mutated alleles will come together and be expressed in an individual is much decreased when non-sib matings are used. However, it is also less likely that such muta- tions will be detected and eliminated; therefore, it is not recommended that strains be propagated for more than a few generations by non-sib matings. Normally, breeders in a production colony represent the last generation of family lines created in enlarging the colony. NONGENETICALLY DEFINED STOCKS The goal of breeding programs for nongenetically defined stocks is to maintain the diversity in genotypes that is present in the founding animals of that stock. Ideally, no selection pressures should be placed on the popu- lation; however, in practice, there is often a conscious or unconscious selec- tion for reproductive performance, and great care should be taken to elimi- nate this bias. Ideally, a purely random mating structure should be used so that each animal has an equal chance of participating in the breeding pro- gram and of mating with any of the animals of the opposite sex within the colony with no attention to relationship, genotype, phenotype, or any other characteristic; this requires accurate identification of individual animals, extensive record-keeping, and structured randomization in which random- ization tables or computer-generated randomized numbers are used to select breeding pairs. An important limitation on any random breeding program is the size of the population that can be maintained within a facility. Even for commer- cial breeders, populations are limited in size; therefore, without a system- atic method for ensuring that inbreeding does not occur, chance matings between relatives will gradually cause a decrease in heterozygosity within the population. The rate of decrease of heterozygosity is proportional to the population size; very small populations experience a more rapid decrease. For example, a population of 50 will underdo a decrease in heterozygosity at the rate of about 1 percent per generation. _ After 20 generations, this population will have only 82 percent of the heterozygosity with which it started (Green, 198lb). To minimize that loss of heterozygosity, one can use a structured sys- tem of mating that is not completely random but is designed to avoid in- breeding. Several such systems exist. In very small populations (up to 32 animals), systematic mating of cousins can be used to avoid brother x sister mating. When the number of animals exceeds 32, that system becomes too
40 RODENTS: LABORATORY ANIMAL MANAGEMENT cumbersome to use. In larger colonies, either a circular or circular-paired mating system can be used effectively to minimize inbreeding; both systems slow the loss of heterozygosity and require regular pairing of progeny from individual cages or groups of cages with animals in adjacent cages or groups. Detailed descriptions of these systems are available (Kimura and Crow, 1963; Poiley, 1960~. Alternatively, a computerized system of tracking the coefficient of inbreeding of all breeders can be used to set up matings of the least-related animals. Loss of heterozygosity by inadvertent inbreeding and acquisition and fixation of spontaneous mutations can cause considerable genetic diver- gence between populations of the same nongenetically defined stock main- tained at different locations. To minimize the process, there should be a regular exchange of breeding stock between populations. The number of animals that are transferred and the frequency of transfer will depend on many factors. including colony size. breeding system used. and rate at which ~ ~ - The success of such measures can be divergence is anticipated to occur. assessed with population-genetics techniques to calculate the degree of re- sidual heterozygosity in individual populations. These methods usually entail surveying a large number of biochemical or immunologic markers that display polymorphism in a relatively large sample of the population. In addition to the classic nongenetically defined populations maintained by random breeding or outbreeding, populations of rodents with substantial genetic diversity, as evidenced by heterozygosity at a large number of loci, can be developed by making systematic multiple inbred-strain crosses. In such a system, four or more inbred strains are regularly crossed in a circular fashion to yield F1 progeny that are systematically mated with a rotational system to provide F2 animals for use in experimental procedures. F2 ani- mals will show greater genetic diversity than most common nongenetically defined stocks that have been maintained for many years as closed colonies (Green, 1981b'. Overall, the maintenance of nongenetically defined stocks is complex if inbreeding is to be minimized. These populations are unique, dynamic, and diverse and require regular characterization unless they are linked by ex- change of breeding stock. CRYOPRESERVATION Cryopreservation, in the form of freezing of cleavage-stage embryos, offers a means to protect a stock or strain against accidental loss or genetic contamination. It also provides a genetic advantage in retarding genetic changes caused by accumulated mutations and an economic advantage in lowering the costs of strain maintenance. In some circumstances, as when quarantine regulations impede the importation of adult animals, the trans
GENETIC MANAGEMENT OF BREEDING COLONIES 41 portation of frozen embryos, which do not have to be quarantined, is effec- tive. Cryopreservation of embryos has been possible since 1972 (Whittingham et al., 1972; Wilmut, 1972) and has now been successfully carried out for at least 16 mammalian species, including mice and rats (Hedrich and Reetz, 1990; Leibo, 1986; Whittingham, 1975; Whittingham et al., 1972~. Not all stocks warrant cryopreservation. If a strain is preserved with scant information on its characteristics, for example, it is unlikely that it will be of much use in the future. The ILAR Committee on Preservation of Laboratory Resources has recommended the following criteria for identify- ing valuable laboratory animals: the importance of the disease process or physiologic function, the validity or genetic integrity of the stock, the diffi- culty of replacing the stock, versatility of the stock, and current use (NRC, 1990). To obtain embryos of a predetermined stage for freezing, exogenous gonadotropins are administered to induce synchronous ovulation and permit timed matings. Exogenous gonadotropins also often induce superovulation (i.e., the production of more eggs than normal). A combination of pregnant mares' serum, which contains follicle-stimulating hormone, and human chorionic gonadotropin, which contains luteinizing hormone, is commonly used (Gates, 1971~. Freezing eight-cell embryos generally produces the most reliable results, at least in the mouse, but other preimplantation embryo stages can also be used. There are many methods for cryopreserving embryos (Leibo, 1992; Mazur, 19901. Generally, they are in two categories: equilibrium methods and nonequilibrium methods; the distinction depends on the osmotic forces en- countered in the presence of cryoprotectant during the freezing process (Mazur, 19903. Equilibrium methods use low concentrations (1.SM) of cryoprotectants and slow, controlled cooling (approximately 0.5°C/min). Nonequilibrium methods generally use a higher concentration of cryoprotectants (about 4-S M) and fast cooling (more than 200°C/min). The two kinds of methods are equally successful, but nonequilibrium methods have the ad- vantage of not requiring controlled-rate freezers. In mice, SOO is generally considered a safe number of embryos to store. Mouse embryos show no deterioration with time when stored at-196°C, and their viability is not affected by the equivalent of 2,000 years of expo- sure to background radiation (Glenister et al., 1984, 19903. Mice have been born from embryos stored for 14 years with no observable differences in rates of birth from recently frozen embryos. An advantage of liquid-nitro- gen storage systems is that electricity and motors are not required; only a periodic, and preferably routine, replenishment of liquid nitrogen is neces- sary. Alarms and automatic filling devices need electricity, but all mainte- nance and monitoring of liquid-nitrogen storage containers can be carried out manually if necessary.
42 RODENTS: LABORATORY ANIMAL MANAGEMENT To recover animals from frozen embryos, the embryos are thawed and transferred to pseudopregnant females, that is, females in which the hor- mones required to support implantation and pregnancy are induced by mat- ing them to vasectomized or genetically sterile males. The overall rate of live births from frozen mouse embryos of inbred and mutant strains is 20 percent. The rate is usually higher for hybrid and outbred embryos, but there is extreme variability, and the rate from a given attempt can range from 0 to 100 percent. For security, embryos from one strain would ideally be stored in separate cities; at a minimum they should be stored in two containers. Before a strain is considered safely cryopreserved, it should have been re-established at least once from frozen embryos by recovering live born, raising them to maturity, ~ . . . and breeding them to produce the next generation. To avoid genetic contami- nation of a strain, genetic monitoring procedures should be used to verify that animals born from frozen embryos have the expected genotype. RECORD-KEEPING In maintaining pedigrees, the most critical records are those of parent- age. One should be able to identify and trace all relationships through these records. In addition to parental information, which might include indi- vidual identification numbers and mating dates, it is useful to record the generation number, birthdate, number born, weaning date, number weaned, and disposition of progeny. The latter information is useful in evaluating the reproductive performance of a colony. A bound, archive-quality pedi- gree ledger or a secure computer system might be used for recording infor- mation. A computer program for colony record-keeping has been described (Silver, 1993~. If ledgers are used in a colony that includes many strains, it is useful to maintain a separate book for each strain. Each book should identify the book that preceded it or, if it is the first pedigree record for its colony, the origin of the animals. In colonies that have only a few strains, it might be more practical to maintain one general ledger. In this case, it is important to identify each entry accurately according to its strain, as well as its parental and other information. For pedigree management, it is also useful to maintain a pedigree chart, at least for foundation breeders; this helps to avoid unnecessary proliferation of family branches by allowing visualization of individual animal relationships. Marking of each animal with its pedigree identification will preserve identity throughout its lifetime (see Chapter 51. That can be useful when animals from different sibships are housed in the same cage. The advantage of recording individual identifications of animals used in research is that retrospective analysis of such characteristics as age and family relationship can sometimes help to explain unexpected results.
GENETIC MANAGEMENT OF BREEDING COLONIES REFERENCES 43 Bailey, D. W. 1978. Sources of subline divergence and their relative importance for sublines of six major inbred strains of mice. Pp. 197-215 in Origins of Inbred Mice, H. C. Morse III, ed. New York: Academic Press. Bailey, D. W. 1979. Genetic drift: The problem and its possible solution by frozen-embryo storage. Pp. 291-299 in The Freezing of Mammalian Embryos, K. Elliott and J. Whelan, eds. CIBA Foundation Symposium 52 (New Series). Amsterdam: Excerpta Medical Bailey, D. W. 1981. Recombinant inbred strains and bilinear congenic strains. Pp. 223-239 in The Mouse in Biomedical Research. Vol. I: History, Genetics, and Wild Mice, H. L. Foster, J. D. Small and J. G. Fox, eds. New York: Academic Press. Bailey, D. W. 1982. How pure are inbred strains of mice. Immunol. Today 3(8):210-214. Gates, A. H. 1971. Maximizing yield and developmental uniformity of eggs. Pp. 64-75 in Methods in Mammalian Embryology, J. C. Daniel, Jr., ed. San Francisco: Freeman. Glenister, P. H., D. G. Whittingham, et al. 1984. Further studies on the effect of radiation during the storage of frozen 8-cell mouse embryos at -196 degrees C. J. Reprod. Fertil. 70:229-234. Glenister, P. H., D. G. Whittingham, et al. 1990. Genome cryopreservation A valuable contribution to mammalian genetic research. Genet. Res. 56:253-258. Green, E. L. 1981a. Genetics and Probability in Animal Breeding Experiments. New York: Oxford University Press. 271 pp. Green, E. L. 1981b. Breeding systems. Pp. 91-104 in The Mouse in Biomedical Research. Vol. I.: History, Genetics and Wild Mice, H. L. Foster, J. D. Small, and J. G. Fox, eds. New York: Academic Press. Hedrich, H. J., and I. C. Reetz. 1990. Cryopreservation of rat embryos. Pp. 274-288 in Genetic Monitoring of Inbred Strains of Rats: A Manual on Colony Management, Basic Monitoring Techniques, and Genetic Variants of the Laboratory Rat, H. J. Hedrich, ed. Stuttgart: Gustav Fischer Verlag. Kimura, M., and J. F. Crow. 1963. On maximum avoidance of inbreeding. Genet. Res. 4:399-415. Leibo, S. P. 1986. Cryobiology: Preservation of mammalian embryos. Pp. 251-272 in Genetic Engineering of Animals An Agricultural Perspective. J. W. Evans and A. Hollaender, eds. New York: Plenum Press. Leibo, S. P. 19-92. Techniques for preservation of mammalian germ plasm. Anim. Biotechnol. 3(1): 139- I 53. Mazur, P. 1990. Equilibrium, quasi-equilibrium, and nonequilibrium freezin,, of mammalian embryos. Cell Biophys. 17:53-92. N1lC (National Research Council), Institute of Laboratory Animal Resources, Committee on Preservation of Laboratory Animal Resources. 1990. Important laboratory animal re- sources: Selection criteria and funding mechanisms for their preservation. ILAR News 32(4) :A 1 -A32. Poiley, S. M. 1960. A systematic method of breeder rotation for non-inbred laboratory animal colonies. Proc. Anim. Care Panel 10: 159- 166. Silver, L. M. 1993. Recordkeeping and database analysis of breeding colonies. Pp. 3-15 in Guide to Techniques in Mouse Development, P. M. Wassarman and M. L. DePamphilis, eds. Methods in Enzymology, Volume 225. San Diego: Academic Press. Whittingham, D. G. 1975. Survival of rat embryos after freezing and thawing. J. Reprod. Fertil. 43:575-578. Whittingham, D. G., S. P. Leibo, and P. Mazur. 1972. Survival o-f mouse embryos frozen to - 196°C and -269°C. Science 178:411 -414. Wilmut, I. 1972. The effect of cooling rate, warming rate of cryoprotective agent, and stage of development on survival of mouse embryos during freezing and thawing. Life Sci. (II),11:1071 - 1079.