X-linked loci influence spatial navigation performance in Dahl rats

Transcription

1 Articles in PresS. Physiol Genomics (December 9, 2003) /physiolgenomics X-linked loci influence spatial navigation performance in Dahl rats Nelson Ruiz-Opazo 1,3 and John Tonkiss 2 1 Section of Molecular Medicine, Department of Medicine 2 Center for Behavioral Development and Mental Retardation Boston University School of Medicine, Boston, Massachusetts 02118, USA Running head: QTLs for spatial navigation 3 Correspondence and requests for reprints to Nelson Ruiz-Opazo, Ph.D. Section of Molecular Medicine, W609 Boston University School of Medicine 700 Albany Street, Boston Massachusetts USA Editorial Office Corresponding author: Nelson Ruiz-Opazo, Ph.D. Tel: (617) Fax: (617) Copyright (c) 2003 by the American Physiological Society.

2 2 Abstract: Elucidation of natural genetic variations underlying strain or individual differences in cognitive function has remained elusive. Here we report the identification of two genetic loci that influence spatial navigation in Dahl rats. In the Morris water maze test, Dahl R rats exhibited efficient spatial navigation, whereas Dahl S rats displayed poor spatial navigation (accuracy). Analysis of F1 male progeny of reciprocal crosses between Dahl S and Dahl R strains implicated the X chromosome with the impairment in spatial navigation observed in Dahl S rats. Quantitative trait locus (QTL) analysis of an (RXS) F2 male population phenotyped for spatial navigation detected two QTLs on chromosome X influencing spatial navigation performance. One QTL (Nav-1, centered at DXRat21, significant for linkage) influenced acquisition performance without affecting spatial accuracy performance and the second QTL (Nav-2, centered at DXRat25, significant for linkage) affected spatial accuracy performance with no detectable effect on acquisition performance. Our results demonstrate X-linkage of spatial navigation performance in Dahl rats and provide evidence for the existence of independent genetic determinants for defined behavioral components of spatial navigation. Key words: genetics, quantitative trait locus, spatial navigation, Morris water maze

3 3 Introduction There is an increasing effort to understand all aspects of cognitive function, including the unraveling of underlying molecular mechanisms contributing to the modulation of learning and memory. This could be of critical importance in deciphering molecular genetic mechanisms involved in pathological processes in which learning and memory function is significantly disrupted or impaired. Two molecular genetic approaches have been employed to gain insights into molecular mechanisms of learning and memory: 1) Gene ablation via specific gene targeting in mice and 2) Standard intercross linkage analysis aimed at identifying natural genetic variations underlying strain differences in cognitive function. Null mutations in several genes that are highly expressed in the brain, including those that encode γ-pkc (1), the metabotropic glutamate receptor mglu R1 (2), CREB (4) (cyclic AMP-responsive element binding protein) and α-ca ++ /CAM kinase II (6), produces deficits in contextual learning in the mouse, thus implicating these proteins in learning and memory function. Recently, two mouse genetic linkage studies have been reported detecting QTLs for contextual learning on chromosomes 1, 2, 3, 10 and 16 in a standard B6/D2 intercross (19) and on chromosomes 1, 3, 7, 8, 9 and 18 in a backcross C3H/B6 linkage analysis (5). QTLs affecting spatial learning have also being reported in mice, on chromosomes 1 and 5 in a study using B x D Ty recombinant inbred mouse strains (10) and on chromosomes 4 and 12 in a DBA2/C57BL6 mouse intercross study (17). The genetic variants accounting for the QTLs effect on contextual and spatial learning remain to be identified. QTL analysis for spatial learning in rats has not been performed and the Dahl rat could offer a valuable model to dissect genetic loci modulating cognitive performance. The Dahl rat is one of the broadly used inbred lines of hypertension, with Dahl S rats being salt-sensitive and their counterpart Dahl R rats being salt-resistant for the development of high blood pressure (15). This animal model has been extensively utilized in studies related to hypertension susceptibility (8).

4 4 We investigated the possibility that distinct behavioral phenotypes could have been segregated in these inbred rat lines during the selective inbreeding for hypertension. We detected significant differences in spatial learning and memory between Dahl S and Dahl R rats. This complex trait was found to be amenable to genetic analysis aimed at identifying natural genetic variations underlying the strain difference in navigational performance. Here we report the identification of two genetic loci on chromosome X influencing spatial learning and memory in Dahl rats using reciprocal F1 intercrosses followed by an F2 intercross linkage analysis. One QTL (Nav-1) affected acquisition performance and the other QTL (Nav-2) influenced spatial accuracy performance.

5 5 Material and Methods Study cohorts. All animal experimentation was conducted in accordance with protocols approved by the Boston University Medical Center, Institutional Animal Care and Use Committee. Dahl S/hsd (n = 12) and Dahl R/hsd (n = 12) male rats (raised on a low sodium diet) were obtained from Harlan (Indianapolis, IN) at 11 weeks of age for characterization of parental strains. Reciprocal mating of the parental strains, Dahl R female x Dahl S male and Dahl S female x Dahl R male, produced two types of F1 male hybrids, F1[RXS] (n = 12) and F1[SXR] (n = 11), respectively. A cohort was derived from brother-to-sister mating of F1 (R female x S male) hybrids to produce an F2 male segregating population (n = 178). In our laboratory, all rats were maintained on LabDiet 5001 rodent chow (Harlan Teklad, Madison WI) containing 0.4 % NaCl available ad lib. Behavioral testing was performed on parental, F1 and F2 cohorts at 12 weeks of age. Morris water maze task. The MWM task was performed essentially as described (18) using a 1.5m diameter, circular water maze (constructed of white plastic and filled with water at 25 o C ± 0.5), and a computer tracking system (Polytrack software program, San Diego Instruments, San Diego, CA). A circular platform, 25cm high and 12cm in diameter, was placed at the center of one of four imaginary quadrants. The water was rendered opaque with 1.5 liters of 2% reduced fat milk. Distance was used to evaluate performance since swim speed was found to be significant different between parental (Dahl R = ± 1.03 cm/sec; Dahl S = ± 1.28 cm/sec, F 1,22 = , P < 0.002) and F1 intercross populations (F1 [RXS] = ± 0.60 cm/sec; F1 [SXR] = ± 1.11 cm/sec, F 1,21 = 9.002, P < 0.007). Hidden Platform version. The platform was submerged 1 cm below the surface of the water and twelve swim trials were given per day (for 2 consecutive days). Animals were placed into the maze (at one of three randomized start positions located adjacent to the wall) and were allowed to traverse the maze in search of the escape platform. On each trial, a maximum swim time of 60 sec was imposed. Between trials, a 35 sec

6 6 interval was imposed with the rat on the platform. At the end of the twenty-fourth trial, the platform was removed (probe trial) and the rat was allowed to search for 1 minute. The distance traveled in the target quadrant [T] and the three other quadrants (adjacent-left [A L ], adjacent-right [A R ] and opposite [O]) was expressed as a percentage of the total distance traveled. Also, the distance traveled within a 30 cm diameter counter centered over the former platform site (in the target quadrant) was expressed as a percentage of the total distance traveled within that counter plus three other counters located at equivalent positions in the remaining quadrants. Visible Platform version. All visual cues were removed from the room, the platform was raised 2 cm above the surface of the water and two 15 cm high dark cylinders were attached to it. Twelve consecutive trials were then administered with a 35s inter-trial interval (as above). The platform was moved in a random fashion between each trial. Intercross linkage analysis. Genotyping was done with 6 chromosome X markers. QTL analysis was performed using the cumulative distance traveled over the 24 trials as an index of acquisition performance (ACQ TD ) and the percent distance traveled in target counter on the probe trial (see above) served as an index of spatial accuracy (SpA). Linkage map, marker regression and interval mapping analyses were done with the Map Manager QTXb17 program (9) which generates a likelihood ratio statistic (LRS) as a measure of the significance of a possible QTL. A backcross analytic design was implemented to perform both the permutation test as well as the QTL analysis. Genetic distances were calculated using Kosambi mapping function (in cm) and were as follows: DXRat8 (21.4, LOD = 29.3) DXRat25 (6.3, LOD = 70.2) DXRat11 (13.5, LOD = 46.2) DXRat32 (19.3, LOD = 32.6) DXRat98 (14.1, LOD = 44.6) DXRat21. Critical significance values (LRS values) for interval mapping were determined by a permutation test (2000 permutations at 10 cm interval) on our 178 informative progeny using Kosambi mapping function. Thus, the minimum LRS values for ACQ TD were for Suggestive linkage = 1.3; for Significant linkage = 6.3; for Highly Significant linkage = 13.6 and for SpA were for Suggestive linkage = 1.3; for Significant linkage = 6.3; for Highly Significant linkage = 16.8 which

7 7 correspond to α values of 0.63, 0.05 and respectively. Statistical analyses. Behavioral data were analyzed by Two-Way Repeated Measures ANOVA, One-Way ANOVA or t-tests (when indicated) using the SigmaStat software, version 2.0 for Windows (SPSS, Chicago). All statistical tests were two-tailed and differences were considered significant at the P < 0.05 level.

8 8 Results Morris water maze testing of parental strains. To examine potential differences in spatial learning and memory between strains, Dahl S and Dahl R male rats were compared directly in the hidden platform version of the MWM task. In this test, subjects are placed in a large tank of water and they must learn to find a hidden escape platform by the flexible use of distal cues (11, 12, 13, 18). Although Dahl S and Dahl R rats exhibited equivalent ability to locate a hidden platform during acquisition performance of the task (Figure 1A, F 1,77 = 0.614, ns), on the probe trial (platform removed), Dahl S subjects had a disorganized search pattern showing absence of target selectivity (Figure 1B) while Dahl R rats demonstrated a clear preference for the target quadrant over all the remaining quadrants (Figure 1C, P < 0.001, derived from Tukey s Pairwise Multiple Comparison Test comparing % search distance in target quadrant versus opposite, target quadrant versus right and target quadrant versus left following One-Way ANOVA). Analysis of the mean percent distance traveled in the target counter (training site) during the probe trial (a better discriminatory measurement of search accuracy for the hidden platform) (14) corroborated the superior performance of Dahl R rats in this task (Figure 1D, Dahl R = 55.6 ± 4.2; Dahl S = 33.8 ± 8.1, F 1,22 = 5.690, P < 0.03). Both groups were equally efficient in locating the escape platform in a visible platform version of the MWM (Figure 1E, F 1,121 = 3.620, ns) demonstrating that the underlying cause of their impairment in navigational performance was unlikely to result from sensori-motor deficits in the Dahl S rats. Morris water maze testing of reciprocal F1 intercrosses. In order to delineate the mode of inheritance (autosomal versus X-linked) of the spatial learning and memory deficit detected in Dahl S rats, we tested F1 (RXS) and F1 (SXR) male progenies in the same apparatus employing the same test conditions used to characterize the parental strains. As shown in Figure 2, the F1 (SXR) subjects demonstrated poor acquisition performance of the task, while F1 (RXS) subjects exhibited efficient location of the hidden platform. Direct

9 9 comparison of the acquisition performance of the two crosses confirmed spatial learning impairment in the F1 (SXR) group as compared with the F1 (RXS) cohort (Figure 2A, F 1,70 = , P < 0.008). The probe trial data affirmed these observations. Similar to rats of the Dahl S genotype, F1 (SXR) subjects showed absence of target selectivity (Figure 2B). Conversely, F1 (RXS) subjects demonstrated a significant preference for the target quadrant over all the remaining quadrants (Figure 2C, P < 0.001, derived from Tukey s Pairwise Multiple Comparison Test comparing % search distance in target quadrant versus each of the other quadrants following One-Way ANOVA) thus resembling the behavior of rats of the Dahl R genotype. Further analysis of the mean percent distance traveled in target counter (training site) during the probe trial corroborated better performance of F1 (RXS) subjects when compared to F1 (SXR) subjects (Figure 2D, F 1,21 = 5.038, P < 0.04). Both experimental groups were equally effective in locating an escape platform when tested on a visible platform version of the MWM task (Figure 2E, F 1,120 = 2.559, ns) demonstrating similar sensori-motor skills. The complete segregation of the spatial learning and memory deficit in the F1 intercross male populations, along with the observation that the impairment of navigational performance detected in Dahl S male rats appears to be maternally transmitted to the male offspring, strongly suggests that this behavioral characteristic has an X-linked inherited component. F2 intercross linkage analysis. Chromosome X was analyzed further based on the results obtained in the reciprocal F1 intercrosses suggesting a chromosome X effect on navigational performance. One hundred and seventy eight F2 male hybrids were genotyped with 6 chromosome X markers. QTL analysis was performed using two measurements associated with navigational performance extracted from the MWM task as quantitative traits: ACQ TD (acquisition, total distance), representing the cumulative distance traveled over the 24 trials by the subjects during acquisition of the task (where the greater the distance traveled the less efficient was the acquisition performance) and SpA (spatial accuracy performance) indexed as the distance traveled in target counter as percentage of the total

10 10 distance traveled over four counters located at equivalent positions within the four quadrants during the probe trial (the larger percentage representing greater spatial accuracy). ACQ TD and SpA are inversely correlated (r = , F 1,176 = 11.48, P < 0.001) reflecting the expected relationship between these two behavioral components of navigational performance: the less distance traveled during acquisition (i.e., better learning performance), the greater percentage of distance traveled in target counter on the probe trial (i.e., greater spatial accuracy). Marker regression followed by interval mapping analyses detected two QTLs on chromosome X, Nav-1 affecting acquisition performance (Figure 3A, centered at DXRat21, LRS = 6.4, significant linkage, Table 3) and Nav-2 influencing spatial accuracy (Figure 3B, centered at DXRat25, LRS = 8.4, significant linkage, Table 3). These results corroborate chromosome X-linkage of spatial navigational performance in Dahl rats as suggested by the reciprocal F1 intercross analysis.

11 11 Discussion Using standard intercross linkage analysis we identified for the first time two QTLs on chromosome X (Nav-1, centered at DXRat21 and Nav-2, centered at DXRat25) influencing spatial navigation in Dahl rats. Nav-1 affects mainly acquisition performance without influencing spatial accuracy performance while Nav-2 influences primarily spatial accuracy performance with no detectable effect on acquisition performance. This finding is consistent with the observed chromosome X effect in the F1 intercross study in which a significant acquisition and spatial accuracy performance difference was observed between the contrasting F1 populations. Although a number of different mechanisms could be postulated to explain the differences between the reciprocal F1 male populations including differences transmitted via the Y chromosome, genomic imprinting or differences in pre- or post-natal maternal care, the results of the F2 cosegregation analysis seem to exclude these possibilities. Two types of experiments support the sex-linked inheritance of spatial navigational performance in Dahl rats. The F1 intercross analysis and the direct chromosome X scan performed on the F2 population. It is noteworthy to mention several earlier studies reporting sex-linked major-gene influence in human spatial visualizing ability (3, 7, 16). Therefore, it is apparent that a major locus on chromosome X exists that influence spatial learning and memory. Two recent studies report QTLs affecting spatial navigation in mice (10, 17). Steinberger et al. (17) found QTLs on chromosomes 4 and 12 influencing variation in spatial learning and Milhaud et al. (10) detected QTLs on chromosomes 1 and 5 modulating spatial navigation performance. The latter study showed that escape latencies and the spatial bias, two distinct components of the task, are controlled by different loci (chromosome 1 QTL influencing escape latencies and chromosome 5 QTL affecting spatial bias). This result is concordant with our findings of two distinct chromosome X loci influencing acquisition performance during the training phase of the task and spatial accuracy performance during the probe trial. The mouse studies did not report

12 12 chromosome X effects on spatial navigation, therefore the chromosome X QTLs detected in our Dahl (RXS) rat intercross must represent new molecular variants affecting the Morris navigation task. Our results provide evidence for independent genetic determinants of different aspects of spatial learning and memory performance. Clearly, new insights into genetic mechanisms underlying cognitive function can be gleaned from such analysis and the Dahl rat appears to provide a powerful model to accomplish those goals. Acknowledgements: We thank Lyle V. Lopez for excellent technical support. This work was supported by grants to N. R-O. from the National Institutes of Health.

13 13 References 1. Abeliovich A, Paylor R, Chen C, Kim JJ, Wehner JM, Tonegawa S. PKC-gamma mutant mice exhibit mild deficits in spatial and contextual learning. Cell 75: , Aiba A, Chen C, Herrup K, Rosenmund C, Stevens CF, Tonegawa S. Reduced hippocampal long-term potentiation and context-specific deficit in associative learning in mglur1 mutant mice. Cell 79: , Bock RD, Kolakowski D. Further evidence of sex-linked major-gene influence on human spatial visualizing ability. Amer. J. Hum. Genet. 25: 1-14, Bourtchuladze R, Frenguelli B, Blendy J, Cioffi D, Schutz G, Silva AJ. Deficient long-term memory in mice with a targeted mutation of the camp-responsive element binding protein. Cell 79: 59-68, Caldarone B, Saavedra C, Tartaglia K, Wehner JM, Dudek BC, Flaherty L. Quantitative trait loci analysis affecting contextual conditioning in mice. Nature Genetics 17: , Chen C, Rainnie DG, Greene RW, Tonegawa S. Abnormal fear response and aggressive behavior in mutant mice deficient for calcium-calmodulin kinase II. Science 266: , Goodenough DR, Gandini E, Olkin I, Pizzamiglio L, Thayer D, Witkin HA. A study of X chromosome linkage with field dependence and spatial visualization. Behavior Genetics 7: , Herrera VLM, Ruiz-Opazo N. Beyond genetic markers: hypertension genes. J. Hypertension 12: , Manly KF, Cudmore Jr RH, Meer JM. Map Manager QTX, cross-platform software for genetic mapping. Mammalian Genome 12: , Milhaud JM, Halley H, Lassalle JM. Two QTLs located on chromosomes 1 and 5 modulate different aspects of the performance of mice of the B x D Ty RI strain series in the Morris navigation task. Behav Genet. 32: 69-78, 2002.

14 Morris RGM. Spatial localization does not require the presence of local cues. Learning and Motivation 12: , Morris RGM, Garrud P, Rawlins JNP, O Keefe J. Place navigation impaired in rats with hippocampal lesions. Nature 297: , Morris RGM. Development of a water-maze procedure for studying spatial learning in the rat. J. Neuroscience Methods 11: 47-60, Paylor R, Baskal L, Wehner JM. Behavioral dissociations between C57BL/6 and DBA/2 mice on learning and memory tasks: a hippocampal-dysfunction hypothesis. Psychobiology 21: 11-26, Rapp JP, Dene H. Development and characteristics of inbred strains of Dahl salt-sensitive and salt-resistant rats. Hypertension 7: , Stafford RE. Sex differences in spatial visualization as evidence of sex linked inheritance. Perceptual and Motor Skills 13: 428, Steinberger D, Reynolds DS, Ferris P, Lincoln R, Datta S, Stanley J, Paterson A, Dawson G, Flint J. Genetic mapping of variation in spatial learning in the mouse. J Neuroscience 23: , Tonkiss J, Shultz P, Galler JR. An analysis of spatial navigation in prenatally protein malnourished rats. Physiology & Behavior 55: , Wehner JM, Radcliffe RA, Rosmann ST, Christensen SC, Rasmussen DL, Fulker DW, Wiles M. Quantitative trait locus analysis of contextual fear conditioning in mice. Nature Genetics 17: , 1997.

15 15 Figure legends Figure 1. Spatial learning and memory in Dahl S and Dahl R male rats. (A D), Hidden version of the Morris water maze (MWM) for Dahl R ( ) and Dahl S ( ) male rats. (A) Acquisition, mean distance (± S.E.M.) to locate the escape platform (F 1,77 = 0.614, ns). (B) Mean percent distance (± S.E.M.) traveled in each quadrant during the probe trial by Dahl S rats. (C) Mean percent distance (+ S.E.M.) traveled in each quadrant during the probe trial by Dahl R rats. (D) Mean distance traveled in target counter as percentage (± S.E.M.) during the probe trial. (E) Mean distance traveled (± S.E.M.) to locate the escape platform during a visible version of the MWM (F 1,121 = 3.620, ns). Quadrants are target (training) quadrant (T), adjacent right (A R ), adjacent left (A L ) and opposite (O). (* P < 0.001, derived from Tukey s Pairwise Multiple Comparison Test comparing % search distance in quadrant T versus O, T versus A R and T versus A L, following One-Way ANOVA, ** F 1,22 = 5.690, P < 0.03). Figure 2. Spatial learning and memory in F1(RXS) and F1(SXR) male rats. (A D), Hidden version of the Morris water maze (MWM) for F1(RXS) ( ) and F1(SXR) ( ) male rats. (A) Acquisition, mean distance (± S.E.M.) to locate the escape platform (F 1,70 = , P < 0.008). (B) Mean percent distance (± S.E.M.) traveled in each quadrant during the probe trial by F1(SXR) rats. (C) Mean percent distance (± S.E.M.) traveled in each quadrant during the probe trial by F1(RXS) rats. (D) Mean distance traveled in target counter as percentage (± S.E.M.) during the probe trial. (E) Mean distance traveled (± S.E.M.) to locate the escape platform during a visible version of the MWM (F 1,120 = 2.559, ns). Quadrants are target (training) quadrant (T), adjacent right (A R ), adjacent left (A L ) and opposite (O). (* P < 0.001, derived from Tukey s Pairwise Multiple Comparison Test comparing % search distance in quadrant T versus O, T versus A R and T versus A L, following One-Way ANOVA, ** F 1,21 = 5.038, P < 0.04).

16 16 Figure 3. Chromosome X QTLs for spatial learning and spatial accuracy in male F2 Dahl (RXS) cohort. Interval mapping with bootstrap for chromosome X (A, B) was performed with the Map Manager QTXb17 program. QTL analyses for two quantitative traits related to navigational performance are presented: ACQ TD (A, acquisition performance) and SpA (B, spatial accuracy performance). Horizontal green lines mark LRS values for significance of linkage. For A from top to bottom: LRS = 13.6 for highly significant, LRS = 6.3 for significant, LRS = 1.3 for suggestive; for B from top to bottom: LRS = 16.8 for highly significant, LRS = 6.3 for significant, LRS = 1.3 for suggestive. Likelihood ratio statistic; regression coefficient. Yellow histograms represent the bootstrap-based confidence intervals for the detected QTLs.

17 17 Table 1. Marker regression for ACQ TD quantitative trait. Locus LRS % P DXRat DXRat DXRat DXRat DXRat DXRat ACQ TD, acquisition performance; LRS, likelihood ratio statistic for the association of the trait with loci; %, the amount of total trait variance that would be explained by a QTL at these loci, as percent; P, nominal significance.

18 18 Table 2. Marker regression for SpA quantitative trait. Locus LRS % P DXRat DXRat DXRat DXRat DXRat DXRat SpA, spatial accuracy performance; LRS, likelihood ratio statistic for the association of the trait with loci; %, the amount of total trait variance that would be explained by a QTL at these loci, as percent; P, nominal significance.

19 19 Table 3. Location of QTLs for spatial navigation detected in F2 intercross Chr QTL Trait Map position LRS % Significance * Effect ** X Nav-1 ACQ TD DXRat Significant X Nav-2 SpA DXRat Significant Chr, chromosome; QTL, quantitative trait locus; ACQ TD, acquisition performance; SpA, spatial accuracy performance; LRS, likelihood ratio statistic for the association of the trait with loci; %, the amount of total trait variance that would be explained by a QTL at these loci, as percent; *, threshold values determined by a permutation test (see Methods) which correspond to α values of 0.63 for suggestive, 0.05 for significant and for highly significant; ** effect of paternal (S) allele on traits;, impairment.

20 20 Distance (cm) A 1800 S 1500 R Search quadrant (%) B Search quadrant (%) C * T O A R A L 0 T O A R A L Block of 3-Trials Distance traveled in target counter (%) D ** Distance (cm) E S R 0 S R Trial Figure 1

21 21 Distance (cm) A F1(SXR) F1(RXS) Search quadrant (%) B Search quadrant (%) C * T O A R A L 0 T O A R A L Block of 3-Trials Distance traveled in target counter (%) D ** Distance (cm) E F1(SXR) F1(RXS) 0 F1(SXR) F1(RXS) Trial Figure 2

22 Figure 3 22