Population variation in drought-resistance strategies in a desert shrub along an aridity gradient: Interplay between phenotypic plasticity and ecotypic differentiation

Perspectives in Plant Ecology, Evolution and Systematics 29 (2017) 12–19

Contents lists available at ScienceDirect

Perspectives in Plant Ecology, Evolution and Systematics journal homepage: www.elsevier.com/locate/ppees

Research article

Population variation in drought-resistance strategies in a desert shrub along an aridity gradient: Interplay between phenotypic plasticity and ecotypic differentiation

MARK

⁎

Danny E. Carvajala,b,c, , Andrea P. Loayzaa,b, Rodrigo S. Riosa, Ernesto Gianolia,d, Francisco A. Squeoa,b,e a

Departamento de Biología, Universidad de La Serena, Casilla 554, La Serena, Chile Instituto de Ecología y Biodiversidad (IEB), Chile Doctorado en Biología y Ecología Aplicada (BEA), Chile d Departamento de Botánica, Universidad de Concepción, Casilla 160-C, Concepción, Chile e Centro de Estudios Avanzados en Zonas Áridas (CEAZA), La Serena, Chile b c

A R T I C L E I N F O

A B S T R A C T

Keywords: Atacama desert Arid environments Drought resistance Intraspecific variation Phenotypic plasticity Encelia canescens

Adaptations to drought of deciduous and evergreen species in arid environments are associated with resourceacquisitive (drought avoidance) and resource-conservative (drought tolerance) strategies of water use, respectively. Few studies have addressed whether a single species can exhibit both drought avoidance and drought tolerance strategies along an aridity gradient, and none have evaluated the role of ecotypic differentiation and phenotypic plasticity in shaping such strategies. In the desert shrub Encelia canescens, distributed along an aridity gradient in the Atacama Desert, we hypothesized that populations located in sites with lower and more variable rainfall (northern populations) would exhibit patterns of trait means and plasticity reflecting a water-conservative strategy, while populations in less arid and less variable environments (southern populations) would exhibit a water-acquisitive strategy. We also tested the hypothesis that functional variation in trait means and plasticity are not alternative mechanisms of adaptation to the environment. In a common garden experiment using plants from seeds collected from six populations spanning the species distribution range we found that plants from the northern populations were smaller, had fewer leaves, lower photosynthetic rates and had higher plasticity for root:shoot ratios and lower plasticity for leaf shedding, suggesting a resource-conservative strategy compared to plants from the southern populations, which showed a resource-acquisitive strategy. We found no association between variation in trait means and plasticity, which indicates that these are not alternative mechanisms of plant adaptation to environmental variation. Results suggest that E. canescens populations have evolved different strategies to cope with drought stress depending on their location along the Atacama Desert’s aridity gradient. This gives us a better understanding of the ecological and evolutionary processes that drive phenotypic variation among populations.

1. Introduction Plant species in arid environments show different adaptations to cope with drought (Larcher, 2003). In particular, desert woody plants show features that can be related to either drought avoidance or drought tolerance (Lambers et al., 2008). Deciduous species, which avoid drought through leaf shedding and physiological dormancy, must take advantage of wet periods; consequently, they have traits to acquire and use water quickly: relatively large leaf areas, high growth rates and high photosynthetic rates. In contrast, evergreen species may tolerate ⁎

drought and show traits that secure water uptake from deeper sources (high root:shoot ratio) and decrease water loss (high stomatal control, small leaf area) (Poorter and Markesteijn, 2008; Slot and Poorter, 2007; Ward, 2009). Thus, deciduous and evergreen species show resourceacquisitive and resource-conservative strategies of water use, respectively. Several studies at the interspecific level have reported how traits related to drought avoidance or tolerance vary along aridity gradients (Hallik et al., 2009; Markesteijn and Poorter, 2009; Slot and Poorter, 2007). However, few empirical studies have addressed whether a single species can exhibit both drought avoidance and drought tolerance

Corresponding author at: Departamento de Biología, Universidad de La Serena, Casilla 554, La Serena, Chile. E-mail addresses: [email protected] (D.E. Carvajal), [email protected] (A.P. Loayza), [email protected] (R.S. Rios), [email protected] (E. Gianoli), [email protected] (F.A. Squeo). http://dx.doi.org/10.1016/j.ppees.2017.10.001 Received 31 May 2017; Received in revised form 1 October 2017; Accepted 13 October 2017 Available online 18 October 2017 1433-8319/ © 2017 Elsevier GmbH. All rights reserved.

Perspectives in Plant Ecology, Evolution and Systematics 29 (2017) 12–19

D.E. Carvajal et al.

strategies along a water availability gradient (e.g., Brouillette et al., 2014). In other words, there is scant evidence of changes in the relationships among traits (i.e., trait syndromes) along water availability gradients in such a way that contrasting strategies (acquisitive vs. conservative) are observed. Differences in functional traits related to water use among populations of a single species can result from phenotypic plasticity, ecotypic differentiation or both (Dudley, 1996; Gianoli and Gonzalez-Teuber, 2005; Heschel et al., 2004; Lázaro-Nogal et al., 2015; Ramírez-Valiente et al., 2010). Phenotypic plasticity to water availability is greater in populations with greater temporal heterogeneity in soil moisture (Gianoli, 2004; Gianoli and Gonzalez-Teuber, 2005; Lázaro-Nogal et al., 2015; Liu et al., 2014; Molina-Montenegro et al., 2010), while the magnitude of population differentiation often correlates with environment, that is, increasing differences in selection pressures along environmental gradients may promote increased population differentiation (Akman et al., 2016; Bradburd et al., 2013; Sexton et al., 2014; Shimono et al., 2009; Wang et al., 2013). Phenotypic plasticity and ecotypic differentiation are non-mutually exclusive means of adaptation to environmental variation (Valladares et al., 2014); however, they have been considered as alternative strategies (Hassel et al., 2005; Salamin et al., 2010) and have been compared for their effectiveness in coping with climate change (see Vázquez et al., 2017) and for their contribution to phenotypic divergence in invasive plants (Liao et al., 2016). Little integration of ecotypic differentiation and phenotypic plasticity into the fundamental ecology of plant species, particularly as it relates to resource-use strategy (Reich et al., 1997), has been attempted so far. Plant phenotypic adjustments to water availability, via plastic responses and/or adaptive population differentiation, involve a suite of physiological, morphological and biomass allocation traits, which do not show consistent trends in their relative importance or extent of variation, rather showing idiosyncratic patterns (Bibee et al., 2011; Gianoli and Gonzalez-Teuber, 2005; Heschel et al., 2004; Lázaro-Nogal et al., 2015; Ramírez-Valiente et al., 2010). Typically, plants deal with reduced soil moisture by showing less, smaller, and thicker leaves, reduced photosynthesis and increased water-use efficiency, which reduces evaporative water losses, and increased biomass allocation to roots, which enhances water uptake; plants are also of smaller size, which reduces overall photosynthetic carbon demand (Lambers et al., 2008; Larcher, 2003; Schulze et al., 2005). Importantly, this phenotypic variation is ecologically significant (Gianoli and Valladares, 2012) and it is often associated with plant performance and fitness in natural and experimental populations (Carlson et al., 2016; Geber and Griffen, 2003; Heschel et al., 2002; Maherali et al., 2010). In view of the ideas discussed above, we expected that, for a woody plant species distributed along an aridity gradient, populations located in sites with lower and more variable rainfall would show patterns of trait means and plasticity reflecting a water-conservative (“evergreenlike”) strategy, while populations in less arid and less variable environments would show patterns of trait means and plasticity reflecting a water-acquisitive (“deciduous-like”) strategy. Furthermore, we expected to find at the species level that phenotypic plasticity and ecotypic differentiation would show either a positive association, resulting from similar selective pressures acting locally, or no association, reflecting the idiosyncratic nature of adaptations discussed above, but not a negative correlation, which would suggest that they are alternative mechanisms of adaptation. Here, we tested these hypotheses using as study species the native sunflower Encelia canescens Lam. (Asteraceae), a drought-deciduous shrub distributed along ca. 600 km of latitudinal range in the Atacama Desert (Northern Chile), where aridity and rainfall variability markedly increase from south to north (Carvajal et al., 2015; Rundel et al., 1991; Squeo et al., 1994). Specifically, to determine whole-plant, morphological and physiological responses of E. canescens to experimental drought we used a common garden with plants from six populations.

Table 1 Location and climatic characteristics of E. canescens populations sampled per region. Climate data (Mean annual precipitation, MAP; Mean annual temperature, MAT) were obtained from the WorldClim database (http://www.worldclim.org/). CHA: Chañaral, CAL: Caldera, PAJ: Caleta Pajonales, CHO: Los Choros, ROM: Romeral, and PAL: Puerto Aldea. De Martonne aridity index (DMAI) was calculated as MAP/(MAT + 10); thus, the lower the index value, the greater the aridity. Values of coefficients of variation (CV) in MAP were calculated as (Standard deviation/MAP) x 100. Data for CV were obtained from three different sources: 1The Chilean Meteorological Agency (www.meteochile.cl) for the period 1986–2011, 2CEAZA-Met weather station (www.ceazamet.cl) for the period 1956–2011. 3Weather station from Fray Jorge National Park for the period 1988–2012. Region

Population

Location

North

CHA

26°18′28” 70°26′23” 26°58′05” 70°46′10” 27°46′57” 71°01′36” 29°17′53” 71°17′33” 29°43′48” 71°19′24” 30°18′27” 71°35′26”

North

CAL

North

PAJ

South

CHO

South South

ROM PAL

S– W S– W S– W S– W S– W S– W

MAP (mm)

MAT (°C)

DMAI

CV (%)

18

17.2

0.6

1

214.29 196.93

31

15.3

1.2

1

38

16.0

1.5

1

145.27

64

16.2

2.4

1

129.29

3.2

2

82.91

4.0

3

75.09

80 106

15.3 16.2

These populations span almost the entire distribution range of the study species (Appendix S1) and can be grouped into two regions (north and south) with three populations each. E. canescens is a suitable study model as earlier work with three populations of this species reported population differentiation and phenotypic plasticity to drought in several functional traits (Carvajal et al., 2015). 2. Material and methods 2.1. Seed collection and population characteristics Encelia canescens seeds (pubescent achenes, Olivares and Squeo, 1999) were collected in December 2010 from six coastal sites in the Atacama Desert, from 26° to 30°S: Chañaral (CHA), Caldera (CAL), Caleta Pajonales (PAJ), Los Choros (CHO), Romeral (ROM) and Puerto Aldea (ALD) (Appendix S1, Table 1). In each population, we randomly selected 30 shrubs and collected three to five flower heads with mature seeds per shrub. The first three populations are located in the Northern region, while the other three populations lie within the Southern region of the Coastal Atacama Desert. Mean annual precipitation in these populations ranges between 18 and 106 mm, whereas mean annual temperature is fairly stable, fluctuating between 15 and 17 °C (Table 1). For each site we calculated both the De Martonne’s aridity index (De Martonne, 1926) (DMAI = MAP/[MAT + 10]), where MAP and MAT represent mean annual precipitation and mean annual temperature, respectively, and the coefficient of variation in precipitation (CV = SD/ MAP), where SD is the standard deviation of the precipitation. The former indicated that all populations belong to a hyper-arid desert, and all of them showed significant interannual variability in precipitation; nonetheless, aridity and CV increased from south to north (Table 1). Furthermore, because MAP = 50 mm is the threshold value that separates xeric scrub (MAP ≥ 50 mm) and semi-desertic scrub (MAP < 50 mm) in Chilean vegetation formations (Moreira-Muñoz, 2011), we grouped populations CHA, CAL and PAJ within the north region (MAP < 50 mm) and populations CHO, ROM and ALD into the south region (Table 1). In fact, populations in the north region were significantly more arid (F1,4 = 15.74, P = 0.017; ANOVA) and variable (F1,4 = 11.25, P = 0.029) than those in the south region. 2.2. Common garden experiment To assess the degree to which among-population phenotypic 13

Perspectives in Plant Ecology, Evolution and Systematics 29 (2017) 12–19

D.E. Carvajal et al.

2001). To relate the sites’ aridity (De Martonne’s aridity index) and coefficient of variation in precipitation (CV, standard deviation.mean−1) with phenotypic plasticity and ecotypic differentiation, we calculated a phenotypic plasticity index (PI, Valladares et al., 2000) for those traits that showed a significant treatment and/or population × treatment interaction, and for traits that differed among populations we calculated an ecotypic differentiation index (EDI). The phenotypic plasticity index (PI) was calculated as the difference between the maximum ( X max ) and minimum ( X min ) mean values of water treatments divided by the mean maximum value, that is: PI= (X max − X min)/ X max . PI values range from 0 to 1, where 0 indicates no plasticity and 1 indicates maximum plasticity in the population. The ecotypic differentiation index was calculated as: EDI = (Xpop − Xspecies ) / Xspecies , where Xpop is the mean trait value for a given population and Xspecies represents the mean trait value for the species (i.e., mean of all population means). EDI values close to 0 indicate little or no differentiation from the species mean value, while higher EDI values indicate greater population differentiation. We then used a series of redundancy analyses (RDA; Legendre and Legendre, 1998) to examine how PI and EDI indices vary across populations with aridity and CV in precipitation. For these analyses, we used two response matrices Y, one for PI contained the traits that showed significant P x T effects and other for EDI contained all traits that showed significant P effects. Additionally, we used two environmental constrain matrices, one for the De Martonne’s aridity index and other for the CV in precipitation of each population. We used two independent environmental matrices because De Martonne’s aridity index and CV are correlated (Appendix S4). We evaluated the significance of each model using restricted permutations (9999 permutations) based on the pseudo-F statistic (Legendre and Legendre, 1998). Finally, to compare both EDI and PI indices between North and South regions (three populations per region) we conducted individual one-way ANOVA for each trait, with Region as the main factor and EDI or PI of each trait as the response variable. To evaluate the relationship between ecotypic differentiation and phenotypic plasticity across traits, we calculated the population average of EDI and PI for each trait and then performed a regression analysis between EDI and PI values of all traits. All statistical analyses were performed using the R statistical environment (R Development Core and Team, 2014).

variation in drought-response traits results from ecotypic differentiation and/or phenotypic plasticity, we established a common garden experiment in a greenhouse at Universidad de La Serena (North-Central Chile, 29°54′S - 71°15′W). Sixty seeds (two per plant) from each population were sown individually in 3.5 l pots filled with sandy soil. Populations were the sampling units because the aim of this study was to compare phenotypic responses in plants from different populations. We sampled a number of mother plants in the field in order to get a representative sample of the genotypic diversity within each population, but we did not intend to estimate the genetic variation underlying observed phenotypic variation (Gianoli and Gonzalez-Teuber, 2005). Additionally, because seeds were collected from different plants, each seed was individually weighed before being planted in order to control for potential maternal effects. Once seedlings emerged, they were maintained under well-watered conditions for six months. For each population, plants were randomly assigned to two watering regimes that started six months after seedling emergence and lasted for 45 d. Water treatments were: control (C) and water deficit (W-). Plants in these groups received 76 ml (C) and 10 ml (W-) of water daily. These regimes are equivalent to 100 and 18 mm of annual rainfall, respectively, and correspond to the wettest (ALD) and driest (CHA) populations (Table 1). The final number of replicates per treatment varied among populations (Appendix S2) because not all seeds germinated and some plants died due to unknown causes before the experiment started. 2.3. Functional traits At the end of the experiment, we recorded for each plant: number of leaves, leaf area (cm2), stem diameter (mm) and plant height (cm). Stem diameter was determined with a digital caliper (700 series digital, Mitutoyo Corporation). Leaf area was measured with a leaf area meter (CI-203, CID, Inc.). Plants were oven-dried at 60 °C (Binder FED 53 − 720) for 48 h to measure separately total dry weight of leaves, shoots and roots. We then calculated leaf mass per area (LMA; the ratio of leaf mass to leaf area [g cm−2]), leaf area ratio (LAR; leaf area per total plant mass [cm2 g−1]) and root:shoot biomass ratio (g g−1). We also measured Net photosynthesis (A) and stomatal conductance (gs) from a single leaf of all experimental plants. Gas exchange measurements were performed between 10:00 and 13:00 h; photon flux density during this time period was ca. 1500 μmol photons m−2 s−1. Measurements were taken on fully expanded leaves using an open gas exchange system (Li6400, Li-Cor Inc., NE, USA) set at 1500 μmol photons m−2 s−1 (10% blue), a chamber temperature of 25 °C, and an ambient CO2 concentration of 380 μmol CO2 mol−1. Intrinsic water use efficiency (iWUE) was calculated as the ratio between A and gs.

3. Results All traits showed population differentiation (Table 2) and the factor Region had a significant effect on the overall phenotypic expression of all traits (ΔAIC > 2; Table 2). Four traits showed phenotypic plasticity to water availability: iWUE, LAR, number of leaves and root:shoot ratio (Table 2). Population differentiation for phenotypic plasticity (significant P × T interaction) was detected for number of leaves and root:shoot ratio (Table 2). Seed mass significantly influenced four traits (Table 2). Regarding mean population values, plants from the northern populations (more arid and more variable environments) were, overall, of smaller size, showed higher relative biomass allocation to roots, and had fewer leaves but more leaf area per unit of biomass as compared to plants from the southern populations (Figs. 1, 2). Plants from the northern populations also showed a trend for lower photosynthetic rates and reduced water-use efficiency (Fig. 1). As to those traits showing differential plasticity among populations, it was evident that −in response to drought– plants from the northern populations shed fewer leaves (Appendix S5) and showed a greater increase in root:shoot biomass ratio than plants from the southern populations (Fig. 2). Accordingly, these were the only traits showing significant differences in the magnitude of plasticity (PI) between regions (Fig. 3). The magnitude of ecotypic differentiation (EDI) was greater in northern

2.4. Statistical analyses To examine the effects of water treatment, source population and region (north vs. south) on the phenotype of experimental plants, we conducted an individual linear mixed effect model (LMM) for each trait, except for number of leaves, for which we used generalised mixed effect models (GLMM) with Poisson error distributions (link function “log”). Populations were nested within region (random factor) in the model. The significance of random effects was assessed based on the difference in Akaike's Information Criteria (AICs) between the full model (i.e, model with random factor included) and a model in which the random factor was removed (i.e., generalised linear model − GLM). Seed mass differed among populations (Appendix S3) and was used as a covariate to control for potential maternal effects. The effects of population (P), water treatment (T) and their interaction (PxT), as well as the covariate (seed mass) on plant traits was assessed using likelihood ratio tests (LRT). A significant interaction between water treatment and source population (i.e., reaction norms with different slopes) indicates differences in plasticity among populations. In these cases we compared the slopes of the reaction norms via paired tests of parallelism (Gianoli, 14

Perspectives in Plant Ecology, Evolution and Systematics 29 (2017) 12–19

D.E. Carvajal et al.

Table 2 Results of linear mixed models (LMM) and generalised linear mixed models (GLMM) on morphological and ecophysiological trait values of E. canescens. Significant values (α < 0.05) are in bold. Values of ΔAIC > 2 indicate a significant effect of the variance between regions and are in bold. Trait acronyms: LMA, Leaf mass per area; A, Net photosynthesis; iWUE, intrinsic water-use efficiency; LAR: Leaf area ratio. Functional trait

Plant height

Stem diameter

Number of leaves

Root:shoot ratio

LMA

A

iWUE

LAR

Fixed factors

χ P ΔAIC χ2 P ΔAIC χ2 P ΔAIC χ2 P ΔAIC χ2 P ΔAIC χ2 P ΔAIC χ2 P ΔAIC χ2 P ΔAIC 2

Random factor

Source Population (P)

Water treatment (T)

P × T interaction

Seed mass

85.90 < 0.001

0.006 0.89

2.13 0.37

1.58 0.04

222.42 < 0.001

7.92 0.13

33.21 0.08

24.62 0.007

99.49 < 0.001

411.14 < 0.001

52.23 < 0.001

11.61 < 0.001

14.75 0.001

51.54 < 0.001

13.84 0.003

0.10 0.72

2.85 < 0.001

0.20 0.13

0.92 0.07

0.11 0.27

265.92 0.04

29.13 0.27

226.66 0.09

0.22 0.92

15635.1 0.004

9170.4 0.001

8623.7 0.09

2073.9 0.13

8726.1 < 0.001

4559.9 < 0.001

1477.9 0.08

1038.2 0.008

Variance between Regions

34.7

6.7

4.8

24.8

51.8

20.6

67.9

45.1

Coastal Atacama Desert, leaf pubescence in E. canescens populations increases from south to north (Ehleringer et al., 1981). Therefore, plants from the northern populations may not depend on the modification of LMA −translated at the whole-plant level into LAR– as a strategy to avoid water loss because their leaves are more pubescent, which is an adaptive response to reduce heat load and water loss in desert environments (Ehleringer and Mooney, 1978). It has been reported earlier in a Mediterranean perennial herb that leaf pubescence may act as a compensatory mechanism to avoid water loss under drought conditions in plants with low LMA (Galmés et al., 2007). Importantly, although plants with hairier leaves loose less water, they also capture less light (Ehleringer et al., 1976), in which case having higher LAR could increase light capture at the whole-plant level (Selaya et al., 2008). Regarding water-use efficiency, there are some cases where high iWUE is not favoured under drought conditions (Arntz and Delph, 2001; Donovan et al., 2007; Donovan and Ehleringer, 1994). From an integrated approach to plant functional responses, it has been pointed out that plants with high root:shoot ratios under drought conditions may reduce their need for increased iWUE because of their reduced water deficit (Heschel et al., 2004). This could explain why plants from the northern populations, which showed higher root:shoot ratios than those from the southern populations, did not show greater values of iWUE than their counterparts. Phenotypic plasticity patterns were also consistent with the expected conservative-acquisitive contrast in water-use strategies. Thus, in response to experimental drought, plants from the northern populations exhibited a significantly greater increase in root:shoot ratio (a typical evergreen feature), while plants from the southern populations shed more leaves (a deciduous-like syndrome). These results agree with inter-specific patterns of plant strategies to avoid the effects of drought. For example, Ackerly (2004) showed that plants could avoid drought periods by having deep root systems, which allow them to avoid water tissue deficit. Conversely, species that favour aboveground growth at the expense of their root system are the first to take advantage of water during rainy seasons, but must reduce water loss via transpiration during the dry season through deciduousness or stomatal closure. The deciduous-like syndrome emerges because of the high cost of leaf

populations for root:shoot ratio and iWUE, with no difference in EDI between regions for the other six traits (Fig. 3). The RDA showed that about 74.4% and 78% of the total variation in PI across populations (including root:shoot ratio and number of leaves) was explained by the aridity gradient and the coefficient of variation (CV), respectively. Thus, both aridity and CV are good predictors of among-population variation in phenotypic plasticity (aridity: F1,4 = 11.63, P = 0.02; CV: F1,4 = 8.69, P = 0.004). In contrast, neither aridity (F1,4 = 1.13, P = 0.4) nor CV (F1,4 = 1.51, P = 0.2) significantly accounted for total variation in EDI among E. canescens populations. Finally, when we analyzed the relationship between EDI and PI indices at the species level we found that there was no association across traits (r2 = 0.001, P = 0.94, n = 8 traits; Fig. 4). 4. Discussion Results mostly supported the hypothesis that E. canescens populations located in sites with lower and more variable rainfall, i.e., northern populations, would show patterns of trait means and plasticity reflecting a water-conservative (“evergreen-like”) strategy, while populations in less arid and less variable environments, i.e., southern populations, would show patterns of trait means and plasticity reflecting a water-acquisitive (“deciduous-like”) strategy. Plants from the northern populations were smaller, had fewer leaves, showed lower photosynthetic rates ( = reduced carbon and water demand and slower growth; Bornhofen et al., 2011; Chapin et al., 1990; Poorter et al., 2012) and had higher root:shoot biomass ratios ( = secured water uptake; Ackerly, 2004; Lloret et al., 1999)) compared to plants from the southern populations. However, some results departed from this conservative strategy of water use. Plants from the northern populations also showed more leaf area per unit of biomass (high LAR and low LMA) and reduced wateruse efficiency (iWUE) as compared to plants from the southern populations. This counterintuitive result (see Schulze et al., 2006; Wright et al., 2001), an apparent lack of functional adjustment to drought at the leaf level, may be explained by the occurrence of clinal variation in leaf pubescence, a trait that was not evaluated in this study. In the 15

Perspectives in Plant Ecology, Evolution and Systematics 29 (2017) 12–19

D.E. Carvajal et al.

Fig. 1. Reaction norms of E. canescens populations to water availability. a) Plant height, b) Stem diameter, c) Leaf mass per area (LMA), d) Leaf area ratio (LAR), e) Net photosynthesis (A), and f) intrinsic water-use efficiency (iWUE). Symbols represent Mean ± SE of each population (white symbols = northern populations; grey symbols = southern populations). Black symbols values represent Mean ± SE of each Region. Water treatments were regular watering = control (C) and water deficit (W-). See text for full name of populations.

evergreen vs. water-acquisitive/deciduous) we also detected population differentiation in plasticity. First, plants from the more arid region showed population differentiation in both mean values and phenotypic plasticity for increased root:shoot biomass allocation, a typical feature of evergreen desert shrubs that allow water uptake from deeper soil layers during the dry season. In fact, another desert shrub from the study area, Senna candolleana, also showed higher root:shoot ratios and greater plasticity to drought for this trait in populations from drier sites (Lázaro-Nogal et al., 2015). Second, plants from the less arid region showed constitutively more leaves under regular watering conditions but shed more leaves under experimental drought, compared to plants from the more arid region. Thus, plant from the southern populations showed the typical behaviour of deciduous desert shrubs: an expensive display of leaf area during the wet season followed by a marked leaf loss in the dry season (Ward, 2009), thus reducing leaf area at the wholeplant level and reducing water loss to transpiration (Joly et al., 1989). We found that in E. canescens populations phenotypic plasticity (plasticity index, PI) increased with environmental heterogeneity (coefficient of variation in precipitation, CV) as well as with aridity (De Martonne’s aridity index). The same patterns were observed for another desert shrub (Lázaro-Nogal et al., 2015) and a perennial herb (Gianoli

maintenance as drought increases (Lambers et al., 2008). Thus, in response to drought, before shedding leaves plants store carbohydrate reserves to sustain growth the next season (Givnish, 2002). Most studies addressing how drought adaptation mechanisms vary with water availability (i.e., how trait syndromes determine either a drought avoidance or drought tolerance strategy) have carried out interspecific comparisons (Hallik et al., 2009; Markesteijn and Poorter, 2009; Slot and Poorter, 2007). Here we show at the intraspecific level that a plant species distributed along an aridity gradient can display variation in these mechanisms through ecotypic differentiation, phenotypic plasticity or a combination of both. Few empirical studies have explicitly addressed, in the context of the spectrum of resource-use strategies (Reich et al., 1997), whether a single species can exhibit both drought avoidance and drought tolerance strategies along a water availability gradient (Brouillette et al., 2014; Niinemets, 2015). Moreover, to our knowledge, none has integrated into this approach both ecotypic differentiation and phenotypic plasticity as mechanisms that may determine trait syndromes. We found that E. canescens exhibited population differentiation for all functional traits evaluated and half of these traits showed plasticity to water availability. More importantly, for those key traits representing either strategy (water-conservative/

16

Perspectives in Plant Ecology, Evolution and Systematics 29 (2017) 12–19

D.E. Carvajal et al.

Fig. 4. The relationship between ecotypic differentiation (EDI) and phenotypic plasticity (PI) indices across traits. Each value represents the Mean ± SE for a given index and trait combination. The thin solid line represents a 1:1 relationship between phenotypic plasticity and ecotypic differentiation. Traits in black fit the 1:1 relationship, while traits in grey −which represent traits measured at the whole-plant level– do not fit it. Trait acronyms: LAR: Leaf area ratio, LMA: Leaf mass per area, A: Net photosynthesis, and iWUE: Intrinsic water-used efficiency.

result is in contrast with what has been reported in earlier studies, particularly with regard to the covariation between population differentiation and the environmental gradient (Shimono et al., 2009; Wang et al., 2013). Further studies including parameters of population genetics might explain this rather surprising result (see Bradburd et al., 2013; Sexton et al., 2014). We found that there was no association between EDI and PI indices across traits at the species level. This supported our hypothesis that such a relationship would not be negative, because phenotypic plasticity and ecotypic differentiation are not alternative but complementary mechanisms of plant adaptation to environmental variation. Nonetheless, an inspection of the graph revealed suggestive patterns. Five traits (LAR, LMA, A, iWUE and stem diameter) showed similar magnitudes of ecotypic differentiation and phenotypic plasticity, while

Fig. 2. Reaction norms of E. canescens populations to water availability. The traits shown had a significant Population x Treatment effect in the analysis (Table 2). a) Number of leaves and b) Root:shoot ratio. Symbols represent Mean ± SE of each population (white symbols = northern populations; grey symbols = southern populations). Black symbols values represent Mean ± SE of each Region. Water treatments were regular watering = control (C) and water deficit (W-). Reaction norms with different uppercase letters had significantly different slopes (test of parallelism). See text for full name of populations.

and Gonzalez-Teuber, 2005) from the study area. Moreover, the link between environmental heterogeneity and phenotypic plasticity, largely established from theoretical considerations (Alpert and Simms, 2002; Bradshaw, 1965; Levins, 1968), has been verified in a recent meta-analysis (Vázquez et al., 2017). On the other hand, neither aridity nor CV predicted the magnitude of ecotypic differentiation (EDI). This

Fig. 3. Differences in EDI (ecotypic differentiation index) and PI (phenotypic plasticity index) between regions for: a) Plant height, b) Stem diameter, c) Leaf mass per area (LMA), d) Leaf area ratio (LAR), e) Number of leaves, f) Root:shoot ratio, g) Net photosynthesis (A) and, h) Intrinsic water-use efficiency (iWUE). The mean index value for each region is represented by a circle (northern) and a triangle (southern) (Mean of three populations ± SE). Significant differences between regions in PI or EDI are indicated with asterisk in the corresponding axis (* P < 0.05; ** P < 0.001).

17

Perspectives in Plant Ecology, Evolution and Systematics 29 (2017) 12–19

D.E. Carvajal et al.

Brouillette, L.C., Mason, C.M., Shirk, R.Y., Donovan, L.A., 2014. Adaptive differentiation of traits related to resource use in a desert annual along a resource gradient. New Phytol. 201, 1316–1327. Carlson, J.E., Adams, C.A., Holsinger, K.E., 2016. Intraspecific variation in stomatal traits, leaf traits and physiology reflects adaptation along aridity gradients in a South African shrub. Ann. Bot. 117, 195–207. Carvajal, D.E., Loayza, A.P., Squeo, F.A., 2015. Contrasting responses to water-deficit among Encelia canescens populations distributed along an aridity gradient. Am. J. Bot. 102, 1552–1557. Chapin, F.S., Schulze, E.D., Mooney, H.A., 1990. The ecology and economics of storage in plants. Annu. Rev. Ecol. Syst. 21, 423–447. De Martonne, E., 1926. Une nouvelle fonction climatologique: L’indice d’aridité. La Meteorologie 2, 449-458. Donovan, L.A., Ehleringer, J.R., 1994. Potential for selection on plants for water-use efficiency as estimated by carbon-isotope discrimination. Am. J. Bot. 81, 927–935. Donovan, L.A., Dudley, S.A., Rosenthal, D.M., Ludwig, F., 2007. Phenotypic selection on leaf water use efficiency and related ecophysiological traits for natural populations of desert sunflowers. Oecologia 152, 13–25. Dudley, S.A., 1996. The response to differing selection on plant physiological traits: evidence for local adaptation. Evolution 50, 103–110. Ehleringer, J.R., Mooney, H.A., 1978. Leaf hairs: effects on physiological activity and adaptive value to a desert shrub. Oecologia 37, 183–200. Ehleringer, J., Bjorkman, O., Mooney, H.A., 1976. Leaf pubescence: effects on absorptance and photosynthesis in a desert shrub. Science 192, 376–377. Ehleringer, J., Mooney, H.A., Gulmon, S.L., Rundel, P.W., 1981. Parallel evolution of leaf pubescence in Encelia in coastal deserts of north and south America. Oecologia 49, 38–41. Galmés, J., Abadia, A., Medrano, H., Flexas, J., 2007. Photosynthesis and photoprotection responses to water stress in the wild-extinct plant Lysimachia minoricensis. Environ. Exp. Bot. 60, 308–317. Geber, M.A., Griffen, L.R., 2003. Inheritance and natural selection on functional traits. Int. J. Plant Sci. 164, S21–S42. Gianoli, E., Gonzalez-Teuber, M., 2005. Environmental heterogeneity and population differentiation in plasticity to drought in Convolvulus chilensis (Convolvulaceae). Evolutionary Ecology 19, 603–613. Gianoli, E., Valladares, F., 2012. Studying phenotypic plasticity: the advantages of a broad approach. Biol. J. Linn. Soc. 105, 1–7. Gianoli, E., 2001. Lack of differential plasticity to shading of internodes and petioles with growth habit in Convolvulus arvensis (Convolvulaceae). Int. J. Plant Sci. 162, 1247–1252. Gianoli, E., 2004. Plasticity of traits and correlations in two populations of Convolvulus arvensis (Convolvulaceae) differing in environmental heterogeneity. Int. J. Plant Sci. 165, 825–832. Givnish, T.J., 2002. Adaptive significance of evergreen vs. deciduous leaves: solving the triple paradox. Silva Fennica 36, 703–743. Hallik, L., Niinemets, U., Wright, I.J., 2009. Are species shade and drought tolerance reflected in leaf-level structural and functional differentiation in northern hemisphere temperate woody flora? New Phytol. 184, 257–274. Hassel, K., Pedersen, B., Soderstrom, L., 2005. Changes in life-history traits in an expanding moss species: phenotypic plasticity or genetic differentiation? A reciprocal transplantation experiment with Pogonatum dentatum. Ecography 28, 71–80. Heschel, M.S., Donohue, K., Hausmann, N., Schmitt, J., 2002. Population differentiation and natural selection for water-use efficiency in Impatiens capensis (Balsaminaceae). Int. J. Plant Sci. 163, 907–912. Heschel, M.S., Sultan, S.E., Glover, S., Sloan, D., 2004. Population differentiation and plastic responses to drought stress in the generalist annual Polygonum persicaria. Int. J. Plant Sci. 165, 817–824. Joly, R.J., Adams, W.T., Stafford, S.G., 1989. Phenological and morphological responses of mesic and dry site sources of coastal douglas-fir to water deficit. For. Sci. 35, 987–1005. Lázaro-Nogal, A., Matesanz, S., Godoy, A., Perez-Trautman, F., Gianoli, E., Valladares, F., 2015. Environmental heterogeneity leads to higher plasticity in dry-edge populations of a semi-arid Chilean shrub: insights into climate change responses. J. Ecol. 103, 338–350. Lambers, H., Chapin III, F.S., Pons, T.L., 2008. Plant Physiological Ecology. Springer Science Business Media, LLC, New york. Larcher, W., 2003. Physiological Plant Ecology: Ecophysiology and Stress Physiology of Functional Groups, 4th ed. Springer, New York. Legendre, P., Legendre, L., 1998. Numerical Ecology, 2nd ed. Elsevier Sciences B.V., Amsterdam, The Netherlands. Levins, R., 1968. Evolution in Changing Environments. Princeton University Press Princeton, New Jersey, USA. Liao, H.X., D'Antonio, C.M., Chen, B.M., Huang, Q.Q., Peng, S.L., 2016. How much do phenotypic plasticity and local genetic variation contribute to phenotypic divergences along environmental gradients in widespread invasive plants? A MetaAnalysis. Oikos 125, 905–917. Liu, Y.J., Zhang, L.R., Niu, H.S., Sun, Y., Xu, X.L., 2014. Habitat-specific differences in plasticity of foliar delta C-13 in temperate steppe grasses. Ecol. Evol. 4, 648–655. Lloret, F., Casanovas, C., Penuelas, J., 1999. Seedling survival of mediterranean shrubland species in relation to root: shoot ratio, seed size and water and nitrogen use. Funct. Ecol. 13, 210–216. Maherali, H., Caruso, C.M., Sherrard, M.E., Latta, R.G., 2010. Adaptive value and costs of physiological plasticity to soil moisture limitation in recombinant inbred lines of Avena barbata. Am. Nat. 175, 211–224. Markesteijn, L., Poorter, L., 2009. Seedling root morphology and biomass allocation of 62 tropical tree species in relation to drought- and shade-tolerance. J. Ecol. 97, 311–325.

the other three traits showed either greater plasticity than ecotypic differentiation (root:shoot ratio and number of leaves) or the opposite (plant height) (Fig. 4). Interestingly, whereas most of the traits showing corresponding values of EDI and PI (black symbols in the graph) were morphological or leaf-level traits, all those traits showing contrasting values of EDI and PI (grey symbols in the graph) were whole-plant traits (i.e., traits measured at the individual level). If this pattern is also observed in other study systems, it could suggest that morphological and physiological traits would be more prone to show parallel evolutionary responses −in both mean values and plasticity– to local selective pressures than whole-plant traits, which would be “specialized” in one type of response. Patterns of among-population variation in mean trait expression and plastic responses to water availability indicate that E. canescens populations have evolved different strategies to cope with drought stress depending on their location along the Atacama Desert’s aridity gradient. Specifically, plants from the drier sites of the gradient adopt a conservative (evergreen-like) water-use strategy, while those from the less arid sites show an expensive (deciduous-like) water-use strategy. Both strategies have been previously associated with adaptive plant responses to arid environments at the among-species level. However, few studies have addressed changes in drought adaptation strategies along environmental gradients at the within-species level, and −to our knowledge– none has included the interplay between ecotypic differentiation and phenotypic plasticity in shaping such adaptive patterns, as we report here. Our results highlight the importance of conducting studies of intraspecific variation along environmental gradients because they allow a better understanding of the ecological and evolutionary processes that drive phenotypic variation among populations. In turn, this may provide insights into how intraspecific processes can ultimately translate into interspecific patterns observed at broad spatial scales. Acknowledgements We thank all those who assisted with the field and greenhouse data collection, particularly Patricio García-Guzman and Luis Letelier. We also thank Luis Letelier for making the map and Cristina Armas for her comments to an earlier version of this manuscript. This research was supported by grants from Chilean Millennium Initiative (ICM P05-002) and CONICYT (PFB-23). D. E. Carvajal was supported by a CONICYT doctoral fellowship (21140050). A. P. Loayza was supported by FONDECYT grants 3120123 and 11140400, and R. S. Rios was supported by a FONDECYT postdoctoral grant (3120121). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ppees.2017.10.001. References Ackerly, D., 2004. Functional strategies of chaparral shrubs in relation to seasonal water deficit and disturbance. Ecol. Monogr. 74, 25–44. Akman, M., Carlson, J.E., Holsinger, K.E., Latimer, A.M., 2016. Transcriptome sequencing reveals population differentiation in gene expression linked to functional traits and environmental gradients in the South African shrub Protea repens. New Phytol. 210, 295–309. Alpert, P., Simms, E.L., 2002. The relative advantages of plasticity and fixity in different environments: when is it good for a plant to adjust? Evol. Ecol. 16, 285–297. Arntz, A.M., Delph, L.F., 2001. Pattern and process: evidence for the evolution of photosynthetic traits in natural populations. Oecologia 127, 455–467. Bibee, K., Shishido, K., Hathaway, R.P., Heschel, M.S., 2011. Population differentiation of Impatiens capensis (Balsaminaceae) at the range limit. Int. J. Plant Sci. 172, 599. Bornhofen, S., Barot, S., Lattaud, C., 2011. The evolution of CSR life-history strategies in a plant model with explicit physiology and architecture. Ecol. Model. 222, 1–10. Bradburd, G.S., Ralph, P.L., Coop, G.M., 2013. Disentangling the effects of geographic and ecological isolation on genetic differentiation. Evolution 67, 3258–3273. Bradshaw, A.D., 1965. Evolutionary significance of phenotypic plasticity in plants. Adv. Genet. 13, 115–155.

18

Perspectives in Plant Ecology, Evolution and Systematics 29 (2017) 12–19

D.E. Carvajal et al.

isotope ratios: specific leaf areas and wood growth of Eucalyptus species across a rainfall gradient in Australia. Tree Physiol. 26, 479–492. Selaya, N.G., Oomen, R.J., Netten, J.J.C., Werger, M.J.A., Anten, N.P.R., 2008. Biomass allocation and leaf life span in relation to light interception by tropical forest plants during the first years of secondary succession. J. Ecol. 96, 1211–1221. Sexton, J.P., Hangartner, S.B., Hoffmann, A.A., 2014. Genetic isolation by environment or distance: which pattern of gene flow is most common? Evolution 68, 1–15. Shimono, Y., Watanabe, M., Hirao, A.S., Wada, N., Kudo, G., 2009. Morphological and genetic variations of Potentilla matsumurae (rosaceae) between fellfield and snowbed populations. Am. J. Bot. 96, 728–737. Slot, M., Poorter, L., 2007. Diversity of tropical tree seedling responses to drought. Biotropica 39, 683–690. Squeo, F.A., Ehleringer, J.R., Olivares, N.C., Arancio, G., 1994. Variation in leaf level energy-balance components of Encelia canescens along a precipitation gradient in north-central Chile. Rev. Chil. Hist. Nat. 67, 143–155. Vázquez, D.P., Gianoli, E., Morris, W.F., Bozinovic, F., 2017. Ecological and evolutionary impacts of changing climatic variability. Biol. Rev. 92, 22–42. Valladares, F., Wright, S.J., Lasso, E., Kitajima, K., Pearcy, R.W., 2000. Plastic phenotypic response to light of 16 congeneric shrubs from a Panamanian rainforest. Ecology 81, 1925–1936. Valladares, F., Matesanz, S., Guilhaumon, F., Araujo, M.B., Balaguer, L., Benito-Garzon, M., Cornwell, W., Gianoli, E., van Kleunen, M., Naya, D.E., Nicotra, A.B., Poorter, H., Zavala, M.A., 2014. The effects of phenotypic plasticity and local adaptation on forecasts of species range shifts under climate change. Ecol. Lett. 17, 1351–1364. Wang, I.J., Glor, R.E., Losos, J.B., 2013. Quantifying the roles of ecology and geography in spatial genetic divergence. Ecol. Lett. 16, 175–182. Ward, D., 2009. The Biology of Deserts. Oxford University Press Inc, New York. Wright, I.J., Reich, P.B., Westoby, M., 2001. Strategy shifts in leaf physiology, structure and nutrient content between species of high- and low-rainfall and high- and lownutrient habitats. Funct. Ecol. 15, 423–434.

Molina-Montenegro, M.A., Atala, C., Gianoli, E., 2010. Phenotypic plasticity and performance of Taraxacum officinale (dandelion) in habitats of contrasting environmental heterogeneity. Biol. Invasions 12, 2277–2284. Moreira-Muñoz, A., 2011. Plant Geography of Chile. Springer Science+Business Media, B.V., London, New York. Niinemets, U., 2015. Is there a species spectrum within the world-wide leaf economics spectrum? Major variations in leaf functional traits in the Mediterranean sclerophyll Quercus ilex. New Phytol. 205, 79–96. Olivares, S.P., Squeo, F.A., 1999. Patrones fenológicos en especies arbustivas del desierto costero del norte-centro de Chile. Rev. Chil. Hist. Nat. 72, 353–370. Poorter, L., Markesteijn, L., 2008. Seedling traits determine drought tolerance of tropical tree species. Biotropica 40, 321–331. Poorter, H., Fiorani, F., Stitt, M., Schurr, U., Finck, A., Gibon, Y., Usadel, B., Munns, R., Atkin, O.K., Tardieu, F., Pons, T.L., 2012. The art of growing plants for experimental purposes: a practical guide for the plant biologist Review. Funct. Plant Biol. 39, 821–838. R Development Core Team, 2014. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. Ramírez-Valiente, J.A., Sánchez-Gómez, D., Aranda, I., Valladares, F., 2010. Phenotypic plasticity and local adaptation in leaf ecophysiological traits of 13 contrasting cork oak populations under different water availabilities. Tree Physiol. 30, 618–627. Reich, P.B., Walters, M.B., Ellsworth, D.S., 1997. From tropics to tundra: global convergence in plant functioning. Proc. Natl. Acad. Sci. U. S. A. 94, 13730–13734. Rundel, P.W., Dillon, M.O., Palma, B., Mooney, H.A., Gulmon, S.L., Ehleringer, J.R., 1991. The phytogeography and ecology of the coastal atacama and peruvian deserts. Aliso 13, 1–49. Salamin, N., Wuest, R.O., Lavergne, S., Thuiller, W., Pearman, P.B., 2010. Assessing rapid evolution in a changing environment. Trends Ecol. Evol. 25, 692–698. Schulze, E.-D., Beck, E., Müller-Hohenstein, K., 2005. Plant Ecology. Springer, Verlag Berlin Heidelberg. Schulze, E.D., Turner, N.C., Nicolle, D., Schumacher, J., 2006. Leaf and wood carbon

19