Patterns of Genetic Architecture for Life-History Traits and Molecular Markers in a Subdivided Species

Abstract

Abstract. Understanding the utility and limitations of molecular markers for predicting the evolutionary potential of natural populations is important for both evolutionary and conservation genetics. To address this issue, the distribution of genetic variation for quantitative traits and molecular markers is estimated within and among 14 permanent lake populations of Daphnia pulicaria representing two regional groups from Oregon. Estimates of population subdivision for molecular and quantitative traits are concordant, with Q ST generally exceeding G ST . There is no evidence that microsatellites loci are less informative about subdivision for quantitative traits than are allozyme loci. Characterspecific comparison of Q ST and G ST support divergent selection pressures among populations for the majority of lifehistory traits in both coast and mountain regions. The level of within-population variation for molecular markers is uninformative as to the genetic variation maintained for quantitative traits. In D. pulicaria, regional differences in the frequency of sex may contribute to variation in the maintenance of expressed within-population quantitative-genetic variation without substantially impacting diversity at the genic level. These data are compared to an identical dataset for 17 populations of the temporary-pond species, D. pulex.

Genomic Background and Generation Time Influence Deleterious Mutation Rates in Daphnia

“…Among the differences in mutation parameters observed between populations, the largest change was the rate of decline in clutch size, which may be explained by several factors. First, low levels of expressed genetic variation observed in the Marie population, in conjunction with the high levels of molecular variation (Morgan et al 2001), may indicate that this population maintains substantial levels of hidden quantitative genetic variation. Hidden genetic variation could result from prolonged periods of asexual reproduction leading to the accumulation of mutations (Barton and Charlesworth 1998).…”

“…Specifically, these populations represent the high and low extremes of the phenotypic distribution for life-history traits and body size of natural populations in Oregon (Morgan et al 2001;Baer and Lynch 2003). Not only do phylogenetic analyses identify these populations as among the most genetically divergent in western Oregon, but also quantitative assessments of the expressed levels of genetic variation show that Lake Marie D. pulicaria are low relative to Klamath Lake D. pulicaria, despite similar levels of genetic diversity for both allozyme and microsatellite loci in these populations (Morgan et al 2001). In addition to the population comparison, for one population (Klamath Lake) we experimentally imposed different generation times to assess the impact on the mutation parameters.…”

Distinguishing adaptive from nonadaptive genetic differentiation: comparison of QST and FST at two spatial scales

“…In common garden 1, variance was also partitioned into among-and within-population components for each region separately. The data from common garden 1 (20 populations) allowed us to determine the among-population (s 2 Pop ) and among-region (s 2 Reg ) genetic variance, and to estimate regional subdivision Q RT and population subdivision Q ST as V GB /(V GB þ V GW ), where V GB is the proportion of genetic variance distributed either among populations (s 2 Pop ) or among regions (s 2 Reg ) and V GW is the proportion of genetic variance distributed within populations (Spitze, 1993;Morgan et al, 2001). The population subdivision within each region (Q SR ) was analyzed by conducting nested analysis of variance for each region separately.…”

“…However, interpretation of any discrepancy between Q ST and F ST as evidence of selection on a particular quantitative trait may not always hold. As variation for quantitative traits is introduced by mutation at a higher rate than for molecular markers (Kimura, 1983;Lynch, 1988), the extent of variation for quantitative traits usually exceeds that for molecular markers (Morgan et al, 2001). The population divergence due to stochastic processes such as drift and founder effects may be detectable with quantitative traits but not with molecular markers when the latter possess a low level of polymorphism or when the genes involved in the expression of quantitative traits have significant epistatic effects (Lynch, 1994).…”

“…Several studies in which patterns of quantitative trait variation were strongly associated with climate or distinct habitats also demonstrated a lack of correspondence between molecular and quantitative trait variation patterns (Karhu et al, 1996;Knapp and Rice, 1998;McKay et al, 2001;Steinger et al, 2002). In other studies where genetic subdivisions at quantitative and molecular markers were compared in their relation to geographic subdivision but without tests of local adaptation/ association with environmental parameters, or were compared without any reference to spatial location/ environmental differences, the outcomes varied from a close match (Lagercrantz and Ryman, 1990;Kuittinen et al, 1997;Morgan et al, 2001) to complete disagreement (Podolsky and Holtsford, 1995;Black-Samuelsson et al, 1997;Heaton et al, 1999;Storz, 2002). When only allozyme and RAPD population differentiation patterns based on genetic distances were compared, the correspondence varied from low (Mamuris et al, 1999;Sun and Wong, 2001;Vandewoestijne and Baguette, 2002) to very high (Peakall et al, 1995;Lifante and Aguinagalde, 1996;Jenczewski et al, 1999).…”