Local adaptation to an altitudinal gradient: the interplay between mean phenotypic trait variation and phenotypic plasticity in Mimulus laciniatus

Abstract

Organisms can adapt to environmental heterogeneity through two mechanisms: (1) expression of population genetic variation or (2) phenotypic plasticity. In this study we investigated whether patterns of variation in both trait means and phenotypic plasticity along elevational and latitudinal clines in a North American endemic plant, Mimulus laciniatus, were consistent with local adaptation. We grew inbred lines of M. laciniatus from across the species' range in two common gardens varying in day length to measure mean and plastic trait expression in several traits previously shown to be involved in adaptation to M. laciniatus's rocky outcrop microhabitat: flowering time, size-related traits, and leaf shape. We examined correlations between the mean phenotype and phenotypic plasticity, and tested for a relationship between trait variation and population elevation and latitude. We did not find a strong correlation between mean and plastic trait expression at the individual genotype level suggesting that they operate under independent genetic controls. We identified multiple traits that show patterns consistent with local adaptation to elevation: critical photoperiod, flowering time, flower size, mean leaf lobing, and leaf lobing plasticity. These trends occur along multiple geographically independent altitudinal clines indicating that selection is a more likely cause of this pattern than gene flow among nearby populations with similar trait values. We also found that population variation in mean leaf lobing is associated with latitude. Our results indicate that both having more highly lobed leaves and greater leaf shape plasticity may be adaptive at high elevation within M. laciniatus. Our data strongly suggest that traits known to be under divergent selection between M. laciniatus and close relative Mimulus guttatus are also under locally varying selection within M. laciniatus.

Keywords: elevation gradient; heterogeneity; leaf lobing; local adaptation; phenotypic plasticity; trait variation.

Figure 1.

a) Early in the growing season, only lower elevation populations have warmed up enough to experience complete snowpack melt. These populations begin to grow earlier in the spring, and therefore experience a shorter photoperiod. b) By early to mid summer, all snowpack has melted from higher elevation sites, therefore high elevation populations grow during longer photoperiods. c) Snowpack level changes from year to year, and the pace of warming may be slower or faster. This creates fluctuation in experienced photoperiod over time as well as across elevation. Snow depth data collected at Dana Meadows in Yosemite National Park (California Data Exchange Center 2023). Created with BioRender.com.

Figure 2.

A map of the Sierra Nevada mountain range with each population mapped to its original collection location. Populations are colored based on latitude.

Figure 3.

Correlation matrix of mean trait expression and plastic trait expression. Plastic trait expression is denoted by (p). Blank squares had an insignificant correlation (p > 0.05) when tested with a t-test.

Figure 4.

Reaction norms for a selection of focal traits. Trait values of M. laciniatus inbred lines were measured in controlled growth chamber conditions under two different photoperiods: 11 hour and 15 hour days. Reaction norms faceted by population can be found in Figure S2. See Table 2 for the significance of genotype by environment interactions for each trait.

Figure 5.

Trait mean expression per genotype plotted against elevation of origin for each population. See Table 3 for the statistical analysis of each clinal association.

Figure 6.

Trait mean expression per genotype plotted against latitude of origin for each population. See Table 4 for a statistical analysis of the clinal association of each trait with elevation.

Figure 7.

Mean trait plasticity per genotype plotted against elevation of origin for each population. See Table 5 for statistical analysis of the association between trait plasticity and latitude.