Genetics of mouse behavior: Interactions with laboratory environment

Science

Volumn 284, Issue 5420, 1999, Pages 1670-1672

  Crabbe, John C ^a   Wahlsten, Douglas ^b   Dudek, Bruce C ^c  

^a OREGON HEALTH AND SCIENCE UNIVERSITY   (United States)

^b UNIVERSITY OF ALBERTA   (Canada)

^c STATE UNIVERSITY OF NEW YORK   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

ANIMAL BEHAVIOR; ANIMAL EXPERIMENT; ANIMAL MODEL; ARTICLE; BEHAVIOR GENETICS; CONTROLLED STUDY; GENOTYPE; INBRED STRAIN; LABORATORY; LOCOMOTION; MAZE TEST; MOUSE; MUTANT; NONHUMAN; PRIORITY JOURNAL; SEX DIFFERENCE; STANDARDIZATION; STRAIN DIFFERENCE;

ANIMALS; ANIMALS, LABORATORY; ANXIETY; BEHAVIOR, ANIMAL; CONFOUNDING FACTORS (EPIDEMIOLOGY); DRINKING BEHAVIOR; ENVIRONMENT; FEMALE; GENETICS, BEHAVIORAL; GENOTYPE; MALE; MICE; MICE, INBRED STRAINS; MICE, MUTANT STRAINS; MOTOR ACTIVITY; PSYCHOLOGICAL TESTS; REPRODUCIBILITY OF RESULTS;

EID: 0033523043     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.284.5420.1670     Document Type: Article

References (20)

1. M. Sibilia and E. F. Wagner, Science 269, 234 (1995); R. Gerlai, Trends Neurosci. 19, 177 (1996); M. Nguyen et al., Nature 390, 78 (1997).
2. M. Sibilia and E. F. Wagner, Science 269, 234 (1995); R. Gerlai, Trends Neurosci. 19, 177 (1996); M. Nguyen et al., Nature 390, 78 (1997).
3. M. Sibilia and E. F. Wagner, Science 269, 234 (1995); R. Gerlai, Trends Neurosci. 19, 177 (1996); M. Nguyen et al., Nature 390, 78 (1997).
4. It has been known for some time that comparisons of multiple genotypes on learning-related tests do not always yield consistent results across laboratories [D. Wahlsten, in Psychopharmacology of Aversively Motivated Behavior, H. Anisman and G. Bignami, Eds. (Plenum, New York, 1978), pp. 63-118]. For another example, the Crabbe laboratory has reported that C57BL/6 mice show a small enhancement of locomotor activity after low doses of ethanol, while the Dudek laboratory finds no such stimulant response [ J. C. Crabbe et al., J. Comp. Physiol. Psychol. 96, 440 (1982); B. C. Dudek and T. J. Phillips, Psychopharmacology 101, 93 (1990)] . Similar variation has been reported in other measures of activity in various laboratories and apparatus [J. M. LaSalle and D. Wahlsten, in Techniques for the Genetic Analysis of Brain and Behavior: Focus on the Mouse, D. Goldowitz, D. Wahlsten, R. E. Wimer, Eds. (Elsevier, Amsterdam, 1992), pp. 391-406].
5. It has been known for some time that comparisons of multiple genotypes on learning-related tests do not always yield consistent results across laboratories [D. Wahlsten, in Psychopharmacology of Aversively Motivated Behavior, H. Anisman and G. Bignami, Eds. (Plenum, New York, 1978), pp. 63-118]. For another example, the Crabbe laboratory has reported that C57BL/6 mice show a small enhancement of locomotor activity after low doses of ethanol, while the Dudek laboratory finds no such stimulant response [ J. C. Crabbe et al., J. Comp. Physiol. Psychol. 96, 440 (1982); B. C. Dudek and T. J. Phillips, Psychopharmacology 101, 93 (1990)] . Similar variation has been reported in other measures of activity in various laboratories and apparatus [J. M. LaSalle and D. Wahlsten, in Techniques for the Genetic Analysis of Brain and Behavior: Focus on the Mouse, D. Goldowitz, D. Wahlsten, R. E. Wimer, Eds. (Elsevier, Amsterdam, 1992), pp. 391-406].
6. It has been known for some time that comparisons of multiple genotypes on learning-related tests do not always yield consistent results across laboratories [D. Wahlsten, in Psychopharmacology of Aversively Motivated Behavior, H. Anisman and G. Bignami, Eds. (Plenum, New York, 1978), pp. 63-118]. For another example, the Crabbe laboratory has reported that C57BL/6 mice show a small enhancement of locomotor activity after low doses of ethanol, while the Dudek laboratory finds no such stimulant response [ J. C. Crabbe et al., J. Comp. Physiol. Psychol. 96, 440 (1982); B. C. Dudek and T. J. Phillips, Psychopharmacology 101, 93 (1990)] . Similar variation has been reported in other measures of activity in various laboratories and apparatus [J. M. LaSalle and D. Wahlsten, in Techniques for the Genetic Analysis of Brain and Behavior: Focus on the Mouse, D. Goldowitz, D. Wahlsten, R. E. Wimer, Eds. (Elsevier, Amsterdam, 1992), pp. 391-406].
7. It has been known for some time that comparisons of multiple genotypes on learning-related tests do not always yield consistent results across laboratories [D. Wahlsten, in Psychopharmacology of Aversively Motivated Behavior, H. Anisman and G. Bignami, Eds. (Plenum, New York, 1978), pp. 63-118]. For another example, the Crabbe laboratory has reported that C57BL/6 mice show a small enhancement of locomotor activity after low doses of ethanol, while the Dudek laboratory finds no such stimulant response [ J. C. Crabbe et al., J. Comp. Physiol. Psychol. 96, 440 (1982); B. C. Dudek and T. J. Phillips, Psychopharmacology 101, 93 (1990)] . Similar variation has been reported in other measures of activity in various laboratories and apparatus [J. M. LaSalle and D. Wahlsten, in Techniques for the Genetic Analysis of Brain and Behavior: Focus on the Mouse, D. Goldowitz, D. Wahlsten, R. E. Wimer, Eds. (Elsevier, Amsterdam, 1992), pp. 391-406].
8. -/-, which is maintained on the 129/Sv-ter background. Mice were obtained from the Jackson Laboratory (Bar Harbor, ME), Taconic Farms (Germantown, NY), or the colonies of R. Hen (Columbia University, New York, NY). Because many targeted deletions are placed on the 129/SvEvTac background, we included this close relative of 129/Sv-ter. The genealogy of many 129 substrains has recently been discussed [E. M. Simpson et al., Nature Genet. 16, 19 (1997); D. W. Threadgill, D. Yee, A. Matin, J. H. Nadeau, T. Magnuson, Mamm. Genome B, 390 (1997)].
9. -/-, which is maintained on the 129/Sv-ter background. Mice were obtained from the Jackson Laboratory (Bar Harbor, ME), Taconic Farms (Germantown, NY), or the colonies of R. Hen (Columbia University, New York, NY). Because many targeted deletions are placed on the 129/SvEvTac background, we included this close relative of 129/Sv-ter. The genealogy of many 129 substrains has recently been discussed [E. M. Simpson et al., Nature Genet. 16, 19 (1997); D. W. Threadgill, D. Yee, A. Matin, J. H. Nadeau, T. Magnuson, Mamm. Genome B, 390 (1997)].
10. Details of procedures and test protocols are given in the Web site for this study (www.albany.edu/psy/ obssr). Variables explicitly equated across laboratories included apparatus, exact testing protocols, age of shipped and laboratory-reared mice, method and time of marking before testing, food (Purina 5001; Purina 5000 for breeders), bedding (Bed-o-cob, 1/4 inch; Animal Specialties, Inc., Hubbard, OR), stainless steel cage tops, four to five mice per cage, light/dark cycle, cage changing frequency and specific days, male left in cage after births, culling only of obvious runts, postpartum pregnancy allowed, weaned at 21 days, specific days of body weight recording, and gloved handling without use of forceps. Unmatched variables included local tap water, requirement of filters over cage tops in Portland only, variation of physical arrangement of colonies and testing rooms across sites, different air handling and humidity, and different sources of batches of cocaine and alcohol.
11. All breeding stock was shipped on 2 or 3 December 1997, and mating pairs were set simultaneously on 13 January 1998 in all labs to provide "unshipped" mice for testing. On 15 to 17 March 1998, a second batch of mice from each genotype was shipped to each laboratory. These "shipped" mice, age matched with the unshipped cohort already in place, were allowed to acclimate to the laboratory for 5 weeks before testing commenced. We tested 128 mice in each lab, in two groups of 64 separated by 1 week. With an n = 4 mice in each genotype/shipping condition/sex/aboratory condition, we had 16 mice per group for the crucial genotype X laboratory comparisons. This sample size gave us statistical power of 90% to detect modest interactions of genotype X laboratory when Type I error probability was set at 0.01 [J. Cohen, Statistical Power Analysis (Erlbaum, Hillsdale, NJ, 1986); D. Wahlsten, Behav. Brain. Sci. 13, 109 (1990)]. For results of analysis of variance, we report only effects significant at P < 0.01. The Web site in (4) provides detailed protocols used for each test, descriptions of the laboratory conditions rigorously equated across labs, and raw data that may be examined for other interesting patterns.
12. All breeding stock was shipped on 2 or 3 December 1997, and mating pairs were set simultaneously on 13 January 1998 in all labs to provide "unshipped" mice for testing. On 15 to 17 March 1998, a second batch of mice from each genotype was shipped to each laboratory. These "shipped" mice, age matched with the unshipped cohort already in place, were allowed to acclimate to the laboratory for 5 weeks before testing commenced. We tested 128 mice in each lab, in two groups of 64 separated by 1 week. With an n = 4 mice in each genotype/shipping condition/sex/aboratory condition, we had 16 mice per group for the crucial genotype X laboratory comparisons. This sample size gave us statistical power of 90% to detect modest interactions of genotype X laboratory when Type I error probability was set at 0.01 [J. Cohen, Statistical Power Analysis (Erlbaum, Hillsdale, NJ, 1986); D. Wahlsten, Behav. Brain. Sci. 13, 109 (1990)]. For results of analysis of variance, we report only effects significant at P < 0.01. The Web site in (4) provides detailed protocols used for each test, descriptions of the laboratory conditions rigorously equated across labs, and raw data that may be examined for other interesting patterns.
13. AccuScan Digiscan monitors (AccuScan Instruments, Columbus, OH) were generously loaned to D. Wahlsten by R. H. Kant to match those available in the other two laboratories. AccuScan also provided all sites with rotarod apparatus. Mouse-scaled water mazes and elevated plus mazes were constructed by D. Wahlsten and shipped to the other two labs. On the first test day, each mouse was tested for 15 min in a Digiscan open-field monitor in a dark, sound-attenuated chamber. On Day 2, each mouse was videotaped for 5 min in an elevated plus maze. On Day 3, mice were given 10 trials on a rotarod set to accelerate from 0 to 100 rpm in 75 s. After all mice had been tested on the rotarod, mice were pretrained briefly to escape from the water maze. On Day 4, mice were given eight massed trials of escape learning to a visible platform in the water maze. On Day 5, the activity test was repeated immediately following an ip injection of 20 mg of cocaine per kilogram. After 2 days of rest, mice were individually housed, given only tap water for 2 days, and then tested for 4 days for drinking of 6% ethanol in tap water versus tap water alone.
14. J. Flint et al., Science 269, 1432 (1995); S. R. Mitchell, J. K. Belknap, J. C. Crabbe, unpublished observations.
15. J. Flint et al., Science 269, 1432 (1995); S. R. Mitchell, J. K. Belknap, J. C. Crabbe, unpublished observations.
16. G. E. McClearn and D. A. Rodgers, Q. J. Stud. Alcohol 20, 691 (1959); J. K. Belknap, J. C. Crabbe, E. R. Young, Psychopharmacology 112, 503 (1993); L. A. Rodriguez et al., Alcohol. Clin. Exp. Res. 19, 367 (1995).
17. G. E. McClearn and D. A. Rodgers, Q. J. Stud. Alcohol 20, 691 (1959); J. K. Belknap, J. C. Crabbe, E. R. Young, Psychopharmacology 112, 503 (1993); L. A. Rodriguez et al., Alcohol. Clin. Exp. Res. 19, 367 (1995).
18. G. E. McClearn and D. A. Rodgers, Q. J. Stud. Alcohol 20, 691 (1959); J. K. Belknap, J. C. Crabbe, E. R. Young, Psychopharmacology 112, 503 (1993); L. A. Rodriguez et al., Alcohol. Clin. Exp. Res. 19, 367 (1995).
19. 1B +/+ and -/-strains maintained at Columbia University (3). If so, these genes must exert very large epistatic effects on the 1B gene deletion's phenotypic effects on drinking (1). Alternatively, some undetected variable (for example, a change in animal care personnel) may have occurred specifically at the Portland site between the original (1995-96) observations and the current experiments.
20. Supported by the Office of Behavioral and Social Sciences Research, NIH, as supplements to grants AA10760 (J.C.C.) and DA10731 (J. Marley and B.C.D., co-principal investigators), and by the Natural Sciences and Engineering Research Council of Canada Grant # 45825 (D.W.), the Department of Veterans Affairs (J.C.C.), and a K02 Award to B.C.D. AA00170. We thank R. H. Kant at AccuScan for the generous loan of equipment and R. Hen for providing the serotonin receptor mutants. We appreciate the comments of C. Cunningham, R. A. Harris, J. Janowsky, and G. Westbrook on a draft of this manuscript. We also thank 5. Boehm II, S. Burkhart-Kasch, J. Dorow, S. Doerksen, C. Downing, J. Fogarty, K. Henricks, C. McKinnon, C. Merrill, P. Metten, C. Nolte, T. Phillips, M. Schalomon, J. Schlumbohm, J. Sibert, J. Singh, and C. Wenger for valuable assistance.