BXD Recombinant Inbred Series Research Articles

Short-term selective breeding for High and Low prepulse inhibition of the acoustic startle response; pharmacological characterization and QTL mapping in the selected lines

Selective breeding offers several important advantages over using inbred strain panels in detecting genetically correlated traits to the selection phenotype. The purpose of the current study was to selectively breed for prepulse inhibition (PPI) of the acoustic startle response (ASR), to pharmacologically and behaviorally characterize the selected lines and to use the lines for quantitative trait loci (QTL) mapping. Starting with heterogeneous stock mice formed by crossing the C57BL/6J, DBA/2J, BALB/cJ and LP/J inbred strains and using a short-term selective breeding strategy, animals were selected for High and Low PPI. The selection phenotype was the 80 dB prepulse tone (15 dB above the background noise). After five generations of selection, the High and Low lines differed significantly (78.1 ± 3.1 vs. 45.2 ± 3.9 [percent inhibition], p < 0.00001). The effects of haloperidol and MK-801 on PPI were not different between the High and Low lines. However, at the highest dose tested (10 mg/kg), the High line was more sensitive than the Low line to the disruptive PPI effects of methamphetamine. The lines did not differ in terms of basal activity or methamphetamine-induced changes in locomotor activity. The High and Low lines were genotyped using a panel of 768 SNPs. Significant QTLs (LOD > 10) were detected on chromosomes 11 and 16 that appeared similar to those detected previously [Hitzemann, R., Bell, J., Rasmussen, E., McCaughran, J. Mapping the genes for the acoustic startle response (ASR) and prepulse inhibition of the ASR in the BXD recombinant inbred series: effect of high-frequency hearing loss and cochlear pathology. In: Willott JF, editor. Handbook of mouse auditory research: From behavior to molecular biology. New York: CRC Press; 2001. p. 441–455.; Petryshen, T.L., Kirby, A., Hammer, R.P. Jr, Purcell, S., O'Leary, S.B., Singer, J.B., et al. Two quantitative trait loci for prepulse inhibition of startle identified on mouse chromosome 16 using chromosome substitution strains. Genetics 2005; 171: 1895–1904.]. Overall, the current study illustrates that the heritability of PPI is sufficient for short-term selective breeding and that the lines which are developed can be used to characterize the factors associated with the regulation of PPI.

Dopamine D2 receptor binding, Drd2 expression and the number of dopamine neurons in the BXD recombinant inbred series: genetic relationships to alcohol and other drug associated phenotypes.

It has not been established to what extent the natural variation in dopamine systems contribute to the variation in ethanol response. The current study addresses this issue by measuring D2 dopamine (DA) receptor binding, the expression of Drd2, the number of midbrain DA neurons in the BXD recombinant inbred (RI) series and then compares these strain means with those previously reported for a variety of ethanol and other drug-related phenotypes. Data were collected for 21 to 23 of the BXD RI strains and the parental strains. D2 DA receptor autoradiography was performed using 125I-epidepride as the ligand [ Kanes S, Dains K, Cipp L, Gatley J, Hitzemann B, Rasmussen E, Sanderson S, Silverman S, Hitzemann R (1996) Mapping the genes for haloperidol-induced catalepsy. J Pharmacol Exp Ther 277:1016-1025]. Drd2 expression was measured using the Affymetrix oligoarray system. Immunocytochemical techniques were used to determine the number of midbrain DA neurons [Hitzemann B, Dains K, Hitzemann R (1994) Further studies on the relationship between dopamine cell density and haloperidol response. J Pharmacol Exp Ther 271:969-976]. The range of difference in receptor binding for the RI strains was approximately 2-fold in all regions examined, the core, the shell of the nucleus accumbens (NAc) and the dorsomedial caudate-putamen (CPu); heritability in all regions was moderate--(h2 approximately 0.35). Drd2 expression in forebrain samples from the RI and parental strains ranged 1.5- to h2-fold and was moderate-0.47. Variation in the number of tyrosine hydroxylase (TH) positive neurons was moderate, 41% and 26% and h2 was low--0.19 and 0.15 for the ventral tegmental area (VTA) and substantia nigra compacta (SNc), respectively. Significant correlations were found between D2 DA receptor binding and the low dose (1.33 g/kg) ethanol stimulant response. (p < 0.002) and between expression and conditioned place preference (CPP) (p < 0.0005). No significant correlations were detected between ethanol preference and either receptor binding or Drd2 expression; however, a significant correlation was found between preference and Ncam expression. Ncam is approximately 0.2 Mb from Drd2. Overall, the data suggest ethanol preference and CPP are associated with the expression of Drd2 or closely linked genetic loci.

Dopamine D2 Receptor Binding, Drd2 Expression and the Number of Dopamine Neurons in the BXD Recombinant Inbred Series: Genetic Relationships to Alcohol and Other Drug Associated Phenotypes

Background: It has not been established to what extent the natural variation in dopamine systems contribute to the variation in ethanol response. The current study addresses this issue by measuring D2 dopamine (DA) receptor binding, the expression of Drd2, the number of midbrain DA neurons in the BXD recombinant inbred (RI) series and then compares these strain means with those previously reported for a variety of ethanol and other drug‐related phenotypes.Methods: Data were collected for 21 to 23 of the BXD RI strains and the parental strains. D2 DA receptor autoradiography was performed using 125I‐epidepride as the ligand [Kanes S, Dains K, Cipp L, Gatley J, Hitzemann B, Rasmussen E, Sanderson S, Silverman S, Hitzemann R (1996) Mapping the genes for haloperidol‐induced catalepsy. J Pharmacol Exp Ther 277:1016–1025]. Drd2 expression was measured using the Affymetrix oligoarray system. Immunocytochemical techniques were used to determine the number of midbrain DA neurons [Hitzemann B, Dains K, Hitzemann R (1994) Further studies on the relationship between dopamine cell density and haloperidol response. J Pharmacol Exp Ther 271:969–976].Results and Conclusions: The range of difference in receptor binding for the RI strains was approximately 2‐fold in all regions examined, the core, the shell of the nucleus accumbens (NAc) and the dorsomedial caudate‐putamen (CPu); heritability in all regions was moderate—(h 2∼0.35). Drd2 expression in forebrain samples from the RI and parental strains ranged 1.5‐ to 2‐fold and h2 was moderate—0.47. Variation in the number of tyrosine hydroxylase (TH) positive neurons was moderate, 41% and 26% and h2 was low—0.19 and 0.15 for the ventral tegmental area (VTA) and substantia nigra compacta (SNc), respectively. Significant correlations were found between D2 DA receptor binding and the low dose (1.33 g/kg) ethanol stimulant response. (p &lt; 0.002) and between Drd2 expression and conditioned place preference (CPP) (p &lt; 0.0005). No significant correlations were detected between ethanol preference and either receptor binding or Drd2 expression; however, a significant correlation was found between preference and Ncam expression. Ncam is approximately 0.2 Mb from Drd2. Overall, the data suggest ethanol preference and CPP are associated with the expression of Drd2 or closely linked genetic loci.

Dopamine D2 Receptor Binding, Drd2 Expression and the Number of Dopamine Neurons in the BXD Recombinant Inbred Series: Genetic Relationships to Alcohol and Other Drug Associated Phenotypes

Dopamine D2 Receptor Binding, Drd2 Expression and the Number of Dopamine Neurons in the BXD Recombinant Inbred Series: Genetic Relationships to Alcohol and Other Drug Associated Phenotypes

Identification and time dependence of quantitative trait loci for basal locomotor activity in the BXD recombinant inbred series and a B6D2 F2 intercross.

A complimentary two-phase strategy was used to detect and map quantitative trait loci (QTLs) associated with the basal locomotor response to a saline challenge (10 ml/kg). In phase 1, putative QTLs, significant at p < 0.01 or better, were identified by analysis of the strain means for 25 strains of the B x D recombinant inbred series. QTLs were identified on chromosomes 1, 3, 5, 9, 10, 16, and 18. Some of these QTLs were detected across the entire experimental period (0-20 min), while others were associated with specific 5-min blocks. Eighteen hundred C57BL/6J (B6) x DBA/2J (D2) F2 intercross animals were phenotyped for the basal locomotor response, and of this group, 500 to 700 individuals, pseudo-randomly selected, were used for a genomewide scan to confirm the RI-generated QTLs and to detect new QTLs. No new QTLs were detected but the QTLs on chromosome 1 were confirmed at p < 10(-5) to p < 10(-9), depending on the time interval. In addition, the QTLs on chromosomes 5 and 9 were confirmed at p < 0.001, providing a combined probability (RI + F2) which exceeds the threshold for a significant association. Two additional phenotypes which showed significant RI strain differences were examined--adaptation and thigmotaxis. Adaptation mapped to the same region of chromosome 9 and thigmotaxis to the same region of chromosome 1 as the distance-traveled QTL. Overall, the data presented here and elsewhere (Flint et al., 1995; Gershenfeld et al., 1997) illustrate that QTLs for basal activity are both robust and reliable.

On the relationships of high-frequency hearing loss and cochlear pathology to the acoustic startle response (ASR) and prepulse inhibition of the ASR in the BXD recombinant inbred series.

The measurement of the acoustic startle response (ASR) and prepulse inhibition (PPI) of the ASR in many inbred strains of mice, including C57BL/6 and DBA/2, may be complicated by age-related high-frequency hearing loss (HFHL) and the associated cochlear pathology. Willott and Erway (1998) have recently reported on the age-related changes of the acoustic brain response in the BXD recombinant inbred (RI) series. Based on these data, the RI series was divided into three groups: juvenile-, intermediate-, and adult-onset HFHL. Each of these groups was tested using paradigms which varied the frequency or intensity of the auditory startle and prepulse stimuli. The results obtained in adolescent mice (6-8 weeks) demonstrate that ASR performance is independent of HFHL; there was no group-dependent decline in the ASR amplitudes for high-frequency stimuli. The expected effect of HFHL on PPI is to increase the salience of the still-audible tones. In response to a white-noise prepulse stimulus, the PPI in the juvenile-onset group (which shows marked HFHL at 6 weeks) was similar to that in the adult-onset group. However, when the prepulse stimulus was a pure tone, the juvenile group showed a decrease in salience across all frequencies tested (5-20 kHz). The data point out the need for carefully constructing auditory tasks in the BXD RI series, to avoid the confounding effects of HFHL.

Hypothetical quantitative trait loci (QTL) for circadian period of locomotor activity in CXB recombinant inbred strains of mice.

The locomotor activity of male mice (Mus musculus) of 13 CXB (BALB/cBy x C57BL/6J) recombinant inbred (RI) strains and their progenitor strains was monitored for 4 to 6 weeks by infrared photoelectric beams under constant dark. The circadian period (tau) of locomotor activity was calculated and used in quantitative trait locus (QTL) analysis of strains' means. Results were compared with potential QTL found in a previous study of the BXD RI series. The mean tau of 13 CXB RI mouse strains (three to six animals per strain) in constant dark had a unimodal distribution suggesting polygenic inheritance. A number of potential QTL were found for this trait. There were two associations at p < .001. H23 on chromosome 3 and Pmv16 on chromosome 16. A region of chromosome 1 was associated with tau in both CXB and BXD RI series. There was also a conjunction with a locus determined from QTL analysis of the previously reported tau of wheel running activity in seven CXB RI strains (Schwartz and Zimmerman, 1990).

Quantitative trait loci (QTL) analyses and alcohol-related behaviors.

Recombinant inbred (RI) strains can make an important contribution toward the merger of molecular genetics and quantitative genetics in the quest for quantitative trait loci (QTL). We present preliminary analyses of alcohol-related processes from our ongoing research using the BXD RI series. Issues concerning reliability, genetic correlations, and RI QTL analysis are discussed. Several strategies for replication and extension of QTL candidate regions are considered: F1 crosses between RI strains, F2 crosses, heterogeneous stock, interspecific backcrosses, QTL selection, and the use of murine QTL in chromosomal regions syntenic to human chromosomes as candidate chromosomal regions for human QTL.

Quantitative trait loci associated with brain weight in the BXD/Ty recombinant inbred mouse strains

Adult C57BL/6J (B6) male mice had 37% heavier brains than did DBA/2J (D2) mice, while their body weights did not differ. The BXD recombinant inbred (RI) series of 20 strains, derived from a cross between B6 and D2 inbred strains, was used as the initial screen to determine significant associations between male brain weight and brain:body weight ratio, with allelic variation at 360 known marker gene loci. For brain weight, this yielded five candidate chromosome regions, each reflecting a possible quantitative trait locus (QTL) site affecting brain weight. The second step was to test as many of these five as possible using standard (non-RI) inbred strain data for brain weight previously reported in the literature. For this purpose, only strains possessing the same alleles as the B6 or D2 strains were used. Sufficient data to test two of the five candidate QTL were available. Of these, one was strongly supported as a site affecting brain weight—the D7rp2 region of chromosome 7. For the brain to body weight ratio, four chromosome regions emerged as significantly associated in the BXD series, but none were amenable to testing due to a lack of allelic information for the standard inbred strains. However, two of these regions showed highly significant associations ( p < 0.001, single test) that merit consideration as QTL sites for future testing. These two are the Hba region on chromosome 11 and the D17Tu7 region on chromosome 17. The genetic correlation between brain and body weight was low ( r = 0.28), indicating that these two traits are largely genetically independent in the BXD RI series.

Use of recombinant inbred strains to detect quantitative trait loci associated with behavior

Recombinant inbred (RI) strains are valuable not only for detecting major gene segregation and linkage but also for identifying associations between behavior and quantitative trait loci (QTL) that account for relatively small amounts of variation in behaviors for which strain distribution patterns are not bimodal. When applied to published data on genetic markers and on behavior for BXD RI strains, the RI QTL association approach suggests the presence of QTLs on chromosomes 6 and 12 for open-field activity and on chromosomes 1, 2, and 17 for high-pressure seizure susceptibility. Because the RI QTL approach does not require that the progenitor inbred strains of a particular RI series differ, researchers could focus on the BXD RI series, for which the greatest number of genetic markers are available. Focusing on BXD would capitalize on the cumulative nature of RI research which permits analyses of QTL sources of genetic correlations across studies.

Quantitative trait loci and psychopharmacology: response to commentaries

The theme of our article was that a merger is needed between quantitative genetic and molecular genetic approaches in order to detect genes associated with psychopharmacological processes even when the genes account for small amounts of variance, so-called quantitative trait loci (QTL). The recombinant inbred (RI) QTL approach using the BXD RI series was discussed as a promising approach.The commentaries by Crabbe and by McGuffin and Buckland make several excellent points to which we have little to add. Goldman and Katz, on the other hand, disagree with some of our arguments and for this reason much of the limited space of our response to the commentaries is directed towards the commentary by Goldman and Katz. We focus on three general issues that they raise: the relationship between quantitative genetics and molecular genetics (single genes vs multiple genes), reverse genetics (anonymous markers) vs forward genetics (candidate genes) and heterogeneity (narrow vs broad assessment). Our response to Goldman and Katz is that these are false dichotomies-we do not need to choose sides between major- and multiple- gene approaches, reverse and forward genetics, or narrow and broad assessment. Rather than choosing sides, we should encourage the deployment of multiple research strategies in order to maximize the probability of identifying genes that affect behavior. A major strength of the RI QTL approach is that it can identify both major- and multiple-gene effects, it employs both reverse and forward genetics, and it can be applied to both narrow and broad assessment (and its multivariate extension is ideally suited to understanding the genetic interrelationships among different levels of assessment).After discussing these general issues raised by Goldman and Katz, we address two specific issues raised in the commentaries: multivariate analysis of multiple markers and the use of F(1) crosses between RI strains to increase power. We end by mentioning the establishment of an RI QTL collaborative registry which aims to facilitate RI QTL analyses.