Effect of genetic background on growth of mice hemizygous for wild-type or dwarf mutated bovine growth hormone transgenes

The effects of a high-growth genetic background on the growth of mice hemizygous for one of two growth hormone transgenes were examined. Male mice hemizygous for wild-type (W) and dwarf mutant (M) bovine growth hormone (bGH) transgenes were crossed with females of a high-growth selected (S) and control (C) line as follows: W x S, W x C, M x S and M x C. Body weights of progeny were recorded weekly from 2 to 10 weeks of age. F₁ progeny were classified as carriers (P) or non-carriers (N) of the transgene by assaying tail DNA for bGH using the polymerase chain reaction and agarose gel electrophoresis. A deficiency in the number of f₁ progeny carrying the W (P<0.05) and M (P<0.01) bGH transgene was most likely due to differential prenatal and early postnatal mortality. Bodyweight means of wild-type transgenic mice were larger (P < 0.05) than those of non-transgenic littermates by 3 weeks of age in a C background in contrast to 5 weeks in S. The wild-type bGH transgene increased adult body weights more in the C (155%) than in the S (136%) background, indicating transgene expression by selection background interaction (P < 0.05). However, the growth response to the wild-type transgene in the S background was still large. The dwarf mutant transgene had a greater effect on growth reduction in the S (70%) than in the C (84%) background, thus causing transgene expression by selection background interaction (P < 0.05). Gender by wild-type transgene effect interactions (P < 0.001) for adult body weight were caused by the transgene reducing the gender difference for body weight in C and eliminating it in S. The dwarf mutant caused a larger negative effect on growth in males than in females, resulting in a gender by dwarf mutant transgene interaction (P < 0.001) for adult body weights. Results indicate that the effect of a GH transgene on growth can be affected both by a high-growth genetic background and the gender of progeny.     

Effect of selection for spot size on reproduction and body weight in mice

Summary Spot size in descendants from the Goodale white-spotted stock of mice responded to selection for increased spot size. The realized heritability estimate was 0.52. However, no correlated response of reproduction to spot size selection was found in the present study, nor was there any correlated response among body weight variables.Joint project of Purdue University and USDA-SEA-ARS, North Central Region. Journal paper number 8279 from the Purdue Agricultural Experiment Station     

Heterosis among lines of mice selected for body weight

Summary To examine the effect of selection on levels of heterosis, crosses were made between three groups of six lines of mice, one group unselected (controls) and the other two selected for high (large lines) and low (small lines) 6-week body weight, respectively. The coefficient of inbreeding of each line was about 0.60. Each line was crossed reciprocally to one line from each of the parental groups, as well as producing purebred progeny. Heterosis for 3-week weight, 6-week weight and 3–6 week gain averaged 0.0%, 2.4% and 4.2%, respectively, and was higher for males than for females. Heterosis was more extensive in crosses involving large or control lines than in crosses with small lines. There was no detectable heterosis in several measures of developmental rate, such as age at vaginal opening. Food conversion efficiency and carcass composition were measured on a sample of the animals. Food consumption, gonadal fat pad weight, and hindquarters weight, each expressed as a proportion of body weight, exhibited –4.0%, 5.6%, and 2.3% heterosis, respectively.     

Direct and correlated responses to selection for weaning weight,post-weaning weight gain and six-week weight in mice

Summary Four lines of mice were formed from a common base population and selected for 37 generations for either increased 3-week weight (weaning weight), 6-week weight, 3–6 week gain, or maintained as a randomly bred control line. Realised heritability estimates for short-term (long-term) responses were 0.33±0.20 (0.07±0.10), 0.46±0.14 (0.26±0.09), 0.36±0.14 (0.24±0.11) for 3-week weight, 6-week weight and 3–6 week gain, respectively. Realised genetic correlations estimated from short-term (long-term) responses were 0.23±0.08 (0.35±0.10) between 3-week weight and 3–6 week gain; 0.82±0.04 (0.58±0.08) between 3-week weight and 6-week weight; and 0.81±0.04 (0.97±0.04) between 3–6 week gain and 6-week weight. The genetic correlation between 3-week weight and 6-week weight was asymmetric with a greater correlated response for 3-week weight when selecting for 6-week weight (1.06) than vice versa (0.63).     

Restricted selection index in mice designed to change body fat without changing body weight: correlated responses

Summary Correlated responses were studied in lines of mice selected for eight generations based on the criterion of a restricted selection index. Two replicate lines were selected in each treatment as follows: HE, high epididymal fat pad weight (EF) with zero change in body weight (BW) at 12 weeks of age; LE; low EF with zero change in BW; and RS, randomly. Correlated responses showed considerable variation between replicates, suggesting that genetic drift was important. Further, correlated responses for most traits were relatively small, probably because of low selection intensity. The HE line responded as expected in component traits of the restricted index. Associated compositional traits in HE responded as predicted since traits correlated with adiposity increased and hind carcass weight did not change significantly. Feed intake increased and feed efficiency (weight gain/feed intake) decreased in HE, as predicted. In contrast, the LE line did not respond in component traits as predicted since EF did not decrease and BW increased. Consequently, LE exhibited little change in traits associated with adiposity, but hind carcass weight, feed intake and feed efficiency increased. Of the correlated responses scored for fitness traits (littering rate, number of days from pairing of mate to littering, litter size and preweaning pup survival rate), significant effects were found for decreased littering rate in LE and increased prenatal survival rate in HE. In summary, correlated responses to restricted index selection generally agreed with expectation when responses in component traits of the index were considered.The research reported in this publication was funded by the North Carolina Agricultural Research Service (NCARS), Raleigh, NC 27695-7643, USA. The use of trade names in this publication does not imply endorsement by the NCARS, nor criticism of similar ones not mentioned     

Environmental maternal influences on body composition in mice selected for body weight

Summary The effect of the postnatal maternal environment, simulated by rearing mice in litters of three, six or nine, on body weight and body composition was investigated in three lines of mice differing widely in growth rate. The lines were selected for high (H₆) and low (L₆) 6-week body weight while the control line was maintained by random selection. Body weight and weights and percentages of ether extract, water, ash and protein at 21, 42, 63 and 84 days were recorded. With few exceptions, there were positive correlated responses to selection in body weight and in weights of body components. At 21 and 42 days the correlated responses were larger in L₆ mice than in H₆ mice. Body weight and weights of body components were larger for mice reared in litters of three than for those reared in litters of nine. Also, mice reared in litters of six were intermediate in body weight and weights of some of the body components between those reared in litters of three and nine. Differences in body weight and weights of body components due to postnatal maternal environment were small by comparison with differences due to genetic line. There were significant line by maternal environment interactions in body weight at 21 days and in ether extract weight at 21 and 63 days. Line and maternal environment differences in percentages of body components did not follow any consistent trend. The results for percentages of body components were further complicated by line x maternal environment interactions. In general, both line and postnatal maternal environmental differences in percentages of body components diminished with age.Paper No. 5670 of the Journal Series of the North Carolina Agricultural Experiment Station, Raleigh, North Carolina 27650. The use of trade names in this publication does not imply endorsement by the North Carolina Agricultural Experiment Station of the products named, nor criticism of similar ones not mentioned     

Restricted index selection in mice designed to change body fat without changing body weight: direct responses

Summary Replicated within full-sib family restricted index selection was conducted for eight generations in mice for high or low epididymal fat pad weight (EF) holding body weight (BW) constant. Pooled realized heritability estimates of index units based on high, low and divergent selection were 0.42±0.20, 0.44±0.19 and 0.42± 0.05, respectively, which were not different from the base population estimate of 0.33±0.10. Realized responses per generation pooled across replicates in the high-fat restricted index lines were in the expected directions for EF (17.5±7.2 mg; P<0.05) and BW (0.03±0.58 g; P>0.05), but responses in the low-fat restricted index lines were discrepant for EF (0.3±5.1 mg; P>0.05) and BW (0.38±0.01 g; P<0.01). Consequently, the realized responses in component traits were decidedly asymmetric (P<0.05). A technique for estimating realized genetic parameters from index selection lines gave realized heritabilities for BW and EF of 0.68±0.04 and 0.45±0.05, respectively, and a realized genetic correlation between BW and EF of 0.93±0.01 compared with base population estimates of 0.43±0.08, 0.49±0.10 and 0.78±0.05, respectively. Possible explanations for the disparity between observed and expected responses in the low-fat restricted index lines include genetic drift, poor estimates of base population parameters, changes in genetic parameters with selection, linkage disequilibrium resulting from selection and asymmetric realized relative index weights.The research reported in this publication was funded by the North Carolina Agricultural Research Service (NCARS), Raleigh, NC 27695-7643, USA. The use of trade names in this publication does not imply endorsement by the NCARS, nor criticism of similar ones not mentioned     

The effects of selection for different combinations of weights at two ages on the growth curve of mice

Summary The weights of mice in lines selected for different combinations of high and low body weights at 5 and at 10 weeks of age were recorded from 3 to 21 weeks of age. The average growth curve for each line was computed using the Gompertz function. The growth curves of lines selected for high or low weight at a single age (ST lines) showed large differences in estimates of mature size and small differences in estimates of maturing rate, i.e. of the relative rate of growth to maturity. The growth curves of lines selected by independent culling for divergent combinations of deviations of opposite sign in 5- and 10-week weights (ICL lines) showed little difference in estimates of mature size and a large difference in estimates of maturing rate. The growth curves of lines selected by index for divergence in 5-week weight with no change in 10-week-weight or for divergence in 10-week-weight with no change in 5-week weight showed large differences in estimates of mature size and large differences in estimates of the maturing rate. The relationship between mature size and maturing rate was affected in different ways by the three types of selection.     

Chiasma frequency in strains of mice selected for litter size and for high body weight

Summary Chiasma frequency was measured in male mice of three outbred lines: FZt:DU (control); DU:6, selected for increased body weight; and DU:C, selected for high fertility. Chiasma frequency was seen to increase in the high body weight line, but decrease in the high fertility line. In both selected lines the intragroup variance in chiasma frequency increased while in DU:C the intracell variance was lower than in the control.     

Direct and correlated responses to selection for post-weaning weight gain on ad libitum or restricted feeding in mice

Summary Selection for post-weaning weight gain in mice from 21 to 42 days, on either a full or restricted feeding level during this period was carried out for seven generations. Control lines were maintained for each feeding level. The rate of selection response was higher on full feeding due to a higher heritability and a larger phenotypic variance. Realised heritabilities of 0.29±0.05 and 0.19±0.04 for selection on full and restricted feeding respectively, were in close agreement with base population estimates.Selection on full feeding led to positive correlated responses in 21 day weight, 42 day weight, food intake and efficiency between 21 and 42 days, and 42 day tail length, but with little change in reproductive performance.Correlated responses to selection on restricted feeding were reduced 21 day weight, but an increase in 42 day weight and increased efficiency from 21 to 42 days. However, overall reproductive performance fell.     

Energy expenditure and locomotor activity in mice selected for food intake adjusted for body weight

Summary The aim of this study was to examine the differences in physical activity and their contribution to differences in energy utilization in mice, selected either high or low for food intake, adjusted for body weight, which show correlated responses in lean content and metabolic rate. Simultaneous measurements of fasting metabolic rate and activity were made in lines of mice selected at either: a young age, 4-to 6-week food intake corrected for 4-week body weight; or an older age, 8- to 10-week food intake corrected for mean weight at 8 and 10 weeks of age. Correlated response in metabolic rate was found to have been accompanied by changes in locomotor activity near the ages at selection in both sets of lines. Activity, however, accounted for only a small proportion of variation in fasting heat production, generally less than 5%, although a highly positive correlation (r=0.63) between the two traits was found. It was concluded that selection for food intake adjusted for body weight has led to correlated response in physical activity. In consequence, mice selected in the upward direction expend some of the excess energy intake rather than assimilating it as body mass and are, therefore, slightly leaner than their counterparts selected in the downward direction.     

Growth and body composition changes in mice selected for high post-weaning weight gain on two levels of feeding

Summary Two lines of mice were selected for high post-weaning weight gain (3 to 6 weeks) adjusted for 3 week weight. One line (F) was grown on freely available food and the other (S) on a feeding scale set at the same level for all mice. Food intake of the S line averaged 80% of the F line. The realised heritabilities after 6 generations of selection were 0.38±0.06 and 0.33±0.07 for the F and S lines, respectively. In generation 7, mice from the F and S lines and from an unselected control line (C) were compared on both free and set levels of feeding from 3 weeks to 9 weeks of age. Measurements taken were growth rate, appetite, food conversion efficiency (weight gain/food intake) and body composition (fat, protein, ash, water). The F and S lines grew more rapidly and efficiently than the C line on both levels of feeding, each line performing best on the level of feeding on which it was selected. The average genetic correlation between growth rates of the same line on the two feeding levels was 0.54±0.10. The F line grew 19% faster and was 9% more efficient than the S line on free feeding but the S line grew 15% faster and was 15% more efficient than the F line on set feeding. Relative to the C line, food intake per day on free feeding was 4% higher in the F line and 6% lower in the S line. There was no difference between the lines in food intake/g body weight. The rate of deposition of all body components increased in both selection lines. In the F, S and C lines respectively, efficiencies of gains in body components (10²x gain/food) were 1.79, 1.31 and 1.06 for fat, 1.53, 1.63 and 1.22 for protein and 5.88, 6.45 and 4.98 for protein + water. Apparently energy lost as heat was reduced in both the F and S lines. The partitioning of energy retained was altered in favour of more fat in the F line and more protein in the S line.     

Correlated responses of growth traits to selection for high and low plasma alkaline phosphatase in mice

Summary The effectiveness of two way selection for plasma alkaline phosphatase (ALP) was investigated in order to determine its influences on growth traits through thirteen generations. The responses of the two lines selected for high (HP) and low (LP) ALP at 45 days of age were compared to that of the mice selected for large (L) and small (SM) body size. The selection responses of plasma ALP were very effective for both HP and LP lines, with average responses per generation calculated from linear regressions of 0.227±0.037 and –0.088±0.022 respectively. The final levels of ALP in HP and LP were 5.54±0.71 and 1.27±0.20 in the thirtheenth generation, while the SM, L and base population had levels of 3.49±0.08, 0.86±0.55 and 2.77±0.56 respectively. The body weight at 45 days of age in LP (31.4±1.4 g) as a correlated response was significantly higher than HP (23.4±1.8 g) at generation 10. The correlated response of milk yield, measured by weight gain up to 12 days of age, was significantly greater in the LP line than in HP, but the correlated response of gains after weaning was not so different as the response of milk yield. The response of litter size and weight in LP showed significant higher levels than that of HP, but pups' birth weight did not differ between LP and HP. It is suggested that the correlated response of milk yield contributed more to the divergence of body size between HP and LP than the gain after weaning.Realized heritabilities of ALP were 0.335±0.059 (HP) and 0.279±0.051 (LP). Realized genetic correlations between ALP and 45 days' body weight were –0.27±0.13 (HP with SM) and –0.52±0.19 (LP with L). Realized genetic correlations between ALP and milk yield were –0.95±0.03 (HP) and –0.37±0.29 (LP). Correlations between ALP and postweaning gains were fairly low.     

Trait mean depression for second-generation inbred strawberry populations with and without parent selection

 Strawberry genotypes selected for superior fruit yield or chosen at random from first-generation self, full-sib, and half-sib populations were crossed to provide second-generation inbred progenies and composite cross-fertilized control populations. Mean yields for inbred offspring from crosses among selected parents exceeded those from the offspring of unselected parents by 87%, 23%, and 37% for self, full-sib, and half-sib populations, respectively; yields for offspring from unrelated crosses among selected parents were 54% larger than those for crosses among unselected parents. Selection for yield also resulted in significant correlated response for fruit number and plant diameter. Mean yields for second-generation half-sib and full-sib offspring from selected parents were greater than those for offspring from the unselected but non-inbred control population. This suggests that selection can be a powerful force in counteracting most of the inbreeding depression expected in cross-fertilized strawberry breeding programs. Selection treatment× inbreeding rate interactions were non-significant for all traits; thus, selection among partially inbred offspring did not have a large effect on the rate of genetic progress. Differential realized selection intensity among individuals with differing levels of homozygosity accumulated due to inbreeding is suggested as the most likely explanation for the absence of association between pedigree inbreeding coefficients and cross performance detected previously in strawberry. Received: 21 July 1996 / Accepted: 7 March 1997     

Replicated selection for insulin-like growth factor-1 and body weight in mice

Summary Five generations of divergent selection for plasma concentration of insulin-like growth factor-1 (IGF-1) and for 12-week body weight were carried out in mice, including randomly selected control lines for each trait. All lines were replicated once (12 lines in total). Each replicate line consisted of eight male and eight female parents per generation. Litter size was standardized to eight pups at birth. Mass selection was applied in the selected lines and within-family random selection in the control lines. Blood was taken from the orbital sinus of individual mice at 12 weeks of age for IGF-1 assay. Realized heritabilities were 0.10±0.01 for IGF-1 and 0.41 ± 0.02 for 12-week weight. The realized genetic correlation between IGF-1 and 12-week weight was 0.58 ± 0.01, with a phenotypic correlation of 0.38. Although the genetic correlation between IGF-1 and body weight in mice is moderately positive, 12-week weight responded 3.5 times as fast to weight selection as to selection for IGF-1.     

Divergent selection for in vitro developmental capacity of preimplantation mouse embryos

Summary Replicated divergent selection was conducted for two generations in ICR mice for in vitro developmental capacity (IVDC; percentage of fertilized one-cell zygotes developing to blastocysts in vitro per female donor). Realized heritabilities based on high and low selection were 0.03±0.08 and –0.11±0.09 in replicate 1, and 0.10±0.11 and 0.08±0.10 in replicate 2. No differences were detected between selection lines (P>0.2) or replicates (P>0.1). Estimate of heritability in the base population based on 332 daughter-dam pairs was 0.14±0.18. These results indicate that additive genetic variance contributes little to the phenotypic variance in this trait. Considerable phenotypic variation in IVDC was observed (mean=49.3; SD=31.0), with a range of IVDC from 0%–100%. Utilization of donor female as a blocking factor is suggested for designs of experiments with preimplantation embryos to increase precision and power of statistical analyses.     

Responses to selection on genotypic or phenotypic values in the presence of genes with major effects

Summary Average genotypic responses were compared after selection for genotypic values and for phenotypic values on the basis of single-gene models and multigene models in simulated livestock populations. Single-gene models dealt with single gene control of the genetic differences between animals, while multigene models considered a collection of genes with various magnitudes of effects on a trait. In each case, selection lasted through discrete generations until the fixation of the gene frequencies occurred. Generations to reach fixation were used to compare various models, and the two criteria for selection, for their efficiency in selection. Implications of using these models versus using infinitesimal models for selection in practice are presented.     

Conditions for the initial increase of new alleles with frequency dependent selection

The paper presents the derivation of conditions for the initial increase of a new allele in a fairly general model of a one locus genetic system. Incorporated within the model are frequency dependent selection coefficients which allow the study of viability differences, fertility differences, segregation distortion and certain forms of non-random mating. Since in many models it is necessary to go beyond linear terms, quadratic, cubic and quartic terms are also derived explicitly and details given of their extension to higher degrees.     

On the probability of correct selection for large k populations, with application to microarray data

One frontier of modern statistical research is the problems arising from data sets with extremely large k (>1000) populations, e.g. microarray and neuroimaging data. For many such problems the focus shifts from testing for significance to selecting, filtering, or screening. Classical Ranking and Selection Methodology (RSM) studied the probability of correct selection (PCS). PCS is the probability that the "best" (t = 1) of k populations is truly selected, according to some specified criteria of best. This paper extends and adapts two selection goals from the RSM literature that are suitable for large k problems (d-best and G-best selection). It is then shown how estimation of PCS for selecting multiple (t > 1) populations with d-best and G-best selection can be implemented to provide a useful measure of the quality of a given selection. A simulation study and the application of the proposed method to a benchmark microarray data set show it is an effective and versatile tool for assessing the probability that a particular gene selection or gene filtering step truly obtains the best genes. Moreover, the proposed method is fully general and may be applied to any such extremely large k problem.