Journal of Heredity, 2016, 42–50
doi:10.1093/jhered/esv070
Symposium Article
Advance Access publication August 30, 2015
Symposium Article
Limits to Future Adaptation in the Invasive
Plant Polygonum cespitosum: Expression of
Functional and Fitness Traits at Elevated CO2
Tim Horgan-Kobelski, Silvia Matesanz, and Sonia E. Sultan
From the Biology Department, Hall-Atwater Laboratories, Wesleyan University, Middletown, CT 06459-0170 (HorganKobelski and Sultan); and the Área de Biodiversidad y Conservación, Departamento de Biología y Geología,
Universidad Rey Juan Carlos, c/ Tulipán s/n, Móstoles 28933, Spain (Matesanz).
Address correspondence to Sonia E. Sultan at the address above, or e-mail: [email protected].
Received December 17, 2014; First decision February 6, 2015; Accepted July 22, 2015.
Corresponding editor: Robin Waples
Abstract
For organisms to adapt to future environments, they must both evolve appropriate functional
responses and phenotypically express those responses under future climatic and CO2 conditions.
We examined these 2 components of future adaptation in an invasive annual plant (Polygonum
cespitosum) by performing a “resurrection” experiment under field conditions simulating a future
environment. Resurrection experiments reveal recent evolution by comparing genotypes from
natural populations sampled across a multigeneration interval. We collected genotypes from
the same 3 North American populations in 1994 and 2005 and raised inbred lines from these
collections under free air CO2 enrichment to examine functional and fitness traits expressed in hot,
dry conditions at both ambient and elevated CO2 (N = 295 plants). The species has rapidly evolved
in its introduced range to increase photosynthetic rate (collection year effect P ≤ 0.011) and delay
senescence (P = 0.017) under full-sun, dry field conditions, but these adaptive changes were not
expressed when the field environment included elevated CO2 (within-treatment year effect P ≥
0.20 for both traits). Populations showed different levels of reproductive output and its genetic
variance in these novel, stressful conditions. These findings illustrate constraints on evolutionary
adaptation to predicted future conditions at both the species and population levels.
Subject areas: Quantitative genetics
Key words: FACE (free air CO2 enrichment), novel environments, Persicaria longiseta, Polygonum cespitosum, rapid evolution, resurrection experiment
Understanding species’ responses to global environmental change
has become a crucial focus of evolutionary ecology and biodiversity
studies. Of particular importance are the responses of invasive species, because their ultimate impact will depend on their success in the
novel environments being created by human activity (Mooney and
Hobbs 2000; Brook et al. 2008; Walther et al. 2009; Bradley et al.
2010; Ziska and Dukes 2011). These environments are expected to
include multiple climatic and atmospheric changes, such as increased
CO2 concentration, higher temperatures, and drought, leading to complex new combinations of factors (Sala et al. 2000; Reusch and Wood
2007). It is well recognized that environmental conditions influence
many aspects of phenotypic expression. Consequently, the long-term
persistence of invasive species will depend not only on the ability
of populations in a given species to evolve appropriate functional
responses to these predicted conditions, but also on the ability of individuals of the species to phenotypically express such newly evolved
© The American Genetic Association. 2015. All rights reserved. For permissions, please e-mail: [email protected]
42
Journal of Heredity, 2016, Vol. 107, No. 1
adaptations in future multifactorial environments (Etterson and Shaw
2001; Jump and Peñuelas 2005; Pinto-Marijuan and Munné-Bosch
2013).
This focus on phenotypic outcomes under real-world conditions, and studies of the environmental context dependency of trait
expression in general, have been termed “ecological developmental
biology” (Gilbert 2001; Sultan 2007, 2015; Gilbert and Epel 2008,
2015). Individual physiological and developmental responses to
predicted aspects of global change such as elevated CO2 levels have
been studied in many plant taxa (Etterson and Shaw 2001; Parmesan
2006; Steinger et al. 2007). However, although adaptive responses to
future environmental stresses are considered key to the future success of invasives, trait expression in such taxa has been relatively little studied (Dukes et al. 2011; Zenni et al. 2014). Even less is known
about the potential in either invasive or native systems to rapidly
evolve functional or fitness traits that are adaptive to novel conditions, or the ability of individuals to express such newly evolved
individual responses under realistic environmental circumstances
(Ward and Kelly 2004; Lau et al. 2007; Vannette and Hunter 2011).
To investigate the evolutionary potential and phenotypic
expression of adaptive responses to predicted future conditions,
we performed a “resurrection” study at a free-air CO2 enrichment (FACE) site that realistically simulated such conditions.
Resurrection experiments directly track microevolutionary change
by sampling genotypes from a given population before and after
a multigeneration time interval, and then raising them in a common environment; before-versus-after differences in the resulting
phenotypes reflect evolutionary changes that have occurred during the interval (Franks et al. 2008). For this experiment, we used
the newly invasive Asian shade annual Polygonum (s.l.) cespitosum
(Blume). This highly plastic, evolutionarily labile invasive plant
provides an excellent system in which to study potential adaptation
to future environments. Individuals exhibit pronounced adaptive
plasticity for functional and fitness traits (Matesanz et al. 2012),
and populations are known to contain substantial neutral as well
as quantitative genetic variation (Matesanz and Sultan 2013;
Matesanz et al. 2014a; Matesanz et al. 2014b). During the past 2
decades, P. cespitosum has expanded its ecological range in northeastern North America to include more open, potentially drier sites
and has also shifted its distribution within existing populations to
sunnier microsites (Mehrhoff et al. 2003; Matesanz et al. 2015). At
the same time, introduced-range populations of the species have
rapidly evolved adaptive responses to high-light conditions, including enhanced ecophysiological function and greater reproductive
output (Sultan et al. 2012).
We collected genotypes from 3 northeastern North American
P. cespitosum populations in 1994 and again in 2005, following the
onset of this species’ invasive spread in the region (Mehrhoff et al.
2003). After a “refresher generation” of growth under uniform conditions, we grew replicate inbred seedlings of 1994- and 2005-collected genotypes in 2 simulated multifactorial future environments
at a FACE site (Ainsworth and Long 2005). Both experimental field
environments were exposed to full sun and were hotter and drier
than sites currently inhabited by the species (Matesanz and Sultan
2013), simulating predicted climate change in parts of the species’
introduced North American range (Karl et al. 2009). One of the
2 experimental environments also included the elevated CO2 level
expected to occur by the year 2050 (Christensen et al. 2007), whereas
the other environment remained at current ambient CO2 concentration. We compared the expression of key functional, anatomical, and
life-history traits by 1994 and 2005 genotypes to test 1) whether
43
the species has recently evolved functional traits that may be adaptive to predicted future combinations of sun, heat, and dry soil and
2) whether individuals expressed any such recently evolved changes
when the novel multifactorial environment included elevated CO2.
The experimental design also allowed us to compare reproductive
output and its among-genotype variation expressed by the 3 populations in the simulated future environments, to assess populationlevel differences in fitness and further evolutionary potential.
Materials and Methods
Experimental Material and Design
Polygonum cespitosum is a primarily self-fertilizing herb introduced to North America from eastern Asia in the early 20th century (Paterson 2000; further references and details in Matesanz et
al. 2012). Plants continue to produce single-seeded fruits (achenes)
indeterminately until they senesce, so timing of senescence strongly
affects reproductive output. We sampled genotypes in 1994 and
2005 from the same 3 northeastern North American populations
of P. cespitosum (ARL = Arch Road, Leeds, MA; ORD = Katherine
Ordway Preserve, Weston, CT; WEI = Weir Farm, Wilton, CT).
These are well-established, geographically separate populations that
occupy at least 100 m2 and present abundances of Polygonum reproductive individuals ranging from 75 to 150 plants/m2. All 3 populations occupy disturbed sites where both light and soil moisture
vary spatially as well as temporally; however, the ARL site is more
heavily and consistently shaded (site details in Sultan et al. 1998).
Light and moisture conditions at these sites have remained generally
unchanged since 1994 (Sultan et al. 2012; see Matesanz et al. 2015
for information regarding the species’ current ecological range in
this region).
We collected achenes from 8–10 field plants per population
(located ≥1 m apart along linear transects) in 1994 and again in
2005, for a total of 55 families (8–10 field plants × 3 populations
× 2 collection years). Achenes were germinated, raised to maturity,
and self-fertilized under uniform glasshouse conditions for 2 generations of inbreeding to produce 55 inbred (selfed full-sib) genetic
lines, eliminating potentially confounding effects of both seed age
and parental environment (Sultan et al. 2012). Because P. cespitosum
is highly self-fertilizing and exists in nature as multiple homozygous
lines, selfing for 2 generations did not decrease genetic variation
(based on microsatellite markers, the inbreeding coefficient (FIS) in
North American populations ranges from 0.75 to 0.98; Matesanz
et al. 2014b).
Achenes of each highly inbred line (genotype) were stratified at
4 °C for 5 weeks and germinated in moist vermiculite. Seedlings were
transplanted in a glasshouse at the first true leaf stage (29–30 May
2008) into SB 300 Universal Mix (SunGro Products, Los Angeles,
CA), then after 2 weeks into a 1:1 mixture of SB 300: sterilized FACE
site soil, and finally outplanted (20–22 June) at a naturalistic density
(45 individuals/m2; Matesanz et al. 2015). One replicate seedling per
genotype was outplanted into each of 3 spatially separate, replicate
ambient and elevated CO2 field plots located at the University of
Illinois FACE site (details in Horgan-Kobelski 2010; Ainsworth and
Long 2005). Each experimental block consisted of a separate FACE
enrichment unit and an ambient plot, paired based on similar soil
moisture and height of adjacent vegetation. Genotypes were randomly distributed across each of these blocked treatment plots (see
Sultan 2000 for a similar design).
Journal of Heredity, 2016, Vol. 107, No. 1
44
After transplant losses (15 seedlings), the final sample consisted
of 315 plants (8–10 genotypes × 3 populations × 2 collection years ×
2 experimental treatments × 3 genotypic replicates/treatment).
Experimental Field Treatments
Field treatments represented more extreme drought-stress and insolation conditions than the species’ current habitats within the introduced range. Soil moisture (% weight) in the experimental plots
averaged 11.4 ± 3.0% moisture, compared to the species’ current
ecological range in northeastern North America of 13.5% (± 4.4%)
to 59.1% (± 11.3%) moisture. Light intensity at Polygonum canopy
height was c. 100% of available photosynthetically active radiation
(PAR) during midday hours, compared with means of 5.5–48.7%
PAR in introduced-range field habitats (Matesanz et al. 2015.). One
plot in each of the 3 blocks was maintained at 550 ppm CO2 (estimated global CO2 level in 2050; Christensen et al. 2007), and the
other plot remained at current ambient CO2 (c. 380 ppm; site details
in Leakey et al. 2004). Plants were grown for 8 weeks in the field
treatments, until all plants had begun to senesce (≥10% senescent
leaves).
Data Collection
Instantaneous photosynthetic rate and stomatal conductance were
calculated from gas exchange measurements made on the most recent
fully expanded leaf on a primary branch of each experimental plant
of this C3 species, at midday (10:30–14:30 h) during sunny midsummer days (between 23 July and 11 August). Measurements were made
using a Licor 6400 photosynthesis system (Licor, Inc., Lincoln, NE)
with red/blue light-emitting diode light at photosynthetic photon flux
density = 1700 µmol m−2 s−1 and reference CO2 concentrations of 380
ppm (ambient treatment) or 550 ppm (elevated treatment). During
gas exchange measurements, temperature in the leaf chamber was
controlled (at 31 °C), and relative humidity was maintained by holding constant the ambient level (within a specified range of 50–80%).
Data were excluded if chamber temperature or humidity varied during the measurement. Leaf tracings were scanned on an LI-3100
leaf area meter (Licor, Inc.) for per-unit area rate calculations. Gas
exchange data from plants that were wilted or senescent were
excluded; one genotype was excluded because only 2 data points met
the inclusion criteria. Whole plant senescence was estimated after 8
weeks of growth in the experimental plots as the percentage of leaves
(to the nearest 10%) with >50% of their area chlorotic.
To measure leaf anatomical (stomatal) traits, epidermal impressions were made by applying a layer of clear nail varnish (Sally
Hansen™) on the abaxial surface of the most recent full-expanded
leaf from each plant in 2 experimental blocks. Air-dried impressions were mounted on slides, and epidermal cells counted in a
340 × 445 µm frame on a compound microscope to calculate the
density of stomatal guard cells or “stomatal index” (½ × [guard cells/
(guard cells + epidermal cells)]) (Royer 2001). Ten guard cells from
each frame were individually measured to determine mean guard
cell length.
Reproductive output was measured by manually collecting all
mature achenes every 7–10 days during weeks 3–8 of growth in the
field environments. Total achene number at week 8 was determined
for each plant either by counting or by dividing the total mass of
achenes collected by the average individual achene mass for that
plant (based on a weighed subset of 10–20 air-dried achenes). Both
methods yielded equivalent results according to a random subsample
of 20 plants (Pearson pairwise correlation r = 0.946; P < 0.0001).
This measurement underestimates total lifetime reproductive output
(particularly for plants in the ambient CO2 plots) because plants had
not undergone complete senescence (depending on collection year,
plants in the ambient CO2 plots had 78–85% nonsenescent shoot
tissue and plants at elevated CO2 61–66% nonsenescent tissue).
Data Analysis
Mixed-model analysis of variance (ANOVA) with type III sums of
squares (JMP Version 7; SAS Institute, Cary, NC) was used to test for
the (fixed) effects of CO2 treatment, population, collection year, and
their interactions; the (random) effect of genotype (nested within
population and year); and the (fixed) effect of block on instantaneous photosynthetic rate, stomatal conductance, stomatal index,
guard cell length, and total achene number. Genotype × CO2 interaction was nonsignificant and hence was pooled with the error term
for all traits; genotype was nonsignificant and hence pooled with
error for the 2 stomatal traits. To protect against the possibility of
spurious significance due to multiple tests in this group of 4 traits, we
compared the number of observed significant results (at an α level of
< 0.05) to the total number of significance tests performed (Waples
R, personal communication). According to this assessment of the
dataset, 10 significance tests out of a total of 34 tests performed
(= 29.4%) fell below the 5% significance level, compared with 1.7
tests (5% of the tests performed) that would be expected to reach
this significance level by chance alone, confirming that the dataset
departs quite substantially from a null model. The possibility of
inflated significance tests across the dataset is also minimized by the
independence of these particular traits (as opposed to highly correlated growth traits such as height, branch number, and biomass). We
used 1-way ANOVA to test for the effect of collection year within
each treatment. Senescence data were analyzed with Kruskal–Wallis
nonparametric ANOVA on ranks (Zar 1999). We conducted Brown–
Forsythe tests on genotypic means in each treatment to compare
genetic variance for achene number among populations.
Data were excluded from 20 individuals (≈7% of the total sample size) with trait values ≥3 standard deviations from the mean
for that population, collection year, and treatment, indicating aberrant microsites (Griffith and Sultan 2012); data from 7 plants were
excluded due to experimental error.
In fulfillment of data archiving guidelines (Baker 2013), we have
deposited the primary data underlying these analyses with Dryad.
Results
All 3 study populations of P. cespitosum from the species’ introduced
North American range have recently evolved to express higher photosynthetic rates under full sun, dry field conditions that are outside
the previous selective history of the species. In the ambient CO2 field
treatment, genotypes collected in 2005 had 13% higher instantaneous photosynthetic rates on average than 1994 genotypes from the
same populations (P = 0.011 for effect of collection year in withintreatment 1-way ANOVA). However, when plants were grown in
the elevated CO2 field treatment, 2005 genotypes did not express
this increased photosynthetic rate, but instead had a slightly (nonsignificantly) lower rate than 1994 genotypes (P = 0.22 for effect
of collection year in 1-way ANOVA, cf. significant year × treatment
interaction, Table 1; Figure 1a). This was largely due to an absolute reduction of photosynthetic rate in 2005 genotypes at elevated
CO2 (Figure 1a). As expected, P. cespitosum plants reduced stomatal
conductance at the elevated CO2 treatment (by an average of 19%:
treatment effect P ≤ 0.001; collection year and population main and
Journal of Heredity, 2016, Vol. 107, No. 1
45
Table 1. Effects of CO2 treatment, population, and collection year on expression of functional and reproductive traits of Polygonum cespitosum in simulated future environments
Source of variation
Photosynthetic rate
(R2 = 0.361; N = 251)
Stomatal index
(R2 = 0.140; N = 188)
Guard cell length
(R2 = 0.178; N = 187)
Total achene number
(R2 = 0.629; N = 295)
CO2 treatment (df =1)
Population (df = 2)
Year (df = 1)
CO2 treatment × population (df = 2)
CO2 treatment × year (df = 1)
Population × year (df = 2)
CO2 treatment × Population × year (df = 2)
Genotype (population, year) (df = 48)
Block (df = 2)
Error
0.995 (0.001)
0.534 (14.40)
0.657 (4.529)
0.979 (0.317)
0.042 (61.58)
0.223 (34.99)
0.029 (52.83)
0.017 (23.03)
0.009 (71.00)
189
0.366 (3.289)
0.079* (20.62)
0.963 (0.009)
0.488 (5.771)
0.101 (10.85)
0.387 (7.635)
0.001 (68.62)
—
0.698 (0.606)
175
0.348 (67.79)
0.085* (383.9)
0.003 (691.4)
0.357 (158.4)
0.422 (49.49)
0.0099 (725.3)
0.174 (270.5)
—
0.005 (620.2)
174
0.147 (1.603)
0.001 (56.78)
0.323 (3.029)
0.436 (0.629)
0.698 (0.114)
0.250 (4.342)
0.684 (0.288)
<0.001 (3.091)
0.072* (2.009)
232
P values and mean square values (in parentheses) are shown from full ANOVA (details in Materials and Methods). Degrees of freedom are shown for each
model term. Note that the genotype term was not tested for leaf anatomical traits. Significant values (P < 0.05) are highlighted in bold. Marginally significant
values (P < 0.10) are marked with the symbol *.
Figure 1. Expression of recently evolved changes in physiology and life history of Polygonum cespitosum in novel environments. Means are shown ± 1 SE for
26–29 genotypes per collection year for (a) photosynthetic rate (µmol CO2 m−2 s−1) and (b) percent reduction in senescence (% fewer senescent leaves of 2005 vs.
1994 genotypes). P values are given from 1-way ANOVA (a) or Kruskal–Wallis nonparametric ANOVA (b) testing for year effect within each treatment. *P < 0.05;
NS P > 0.20.
interaction effects on conductance were nonsignificant). 2005 genotypes of all 3 populations also delayed senescence in sunny, dry conditions compared with 1994 genotypes, with a significantly lower
proportion of senescent leaves after 8 weeks of growth in the field
(Figure 1b; P = 0.017 for effect of collection year in within-treatment
Kruskal–Wallis nonparametric ANOVA). However, this significant
delay in senescence was not expressed by 2005 genotypes grown
at elevated CO2 (P = 0.266 for effect of collection year in withintreatment Kruskal–Wallis nonparametric ANOVA).
For both stomatal traits, evolved changes in environmental response from 1994 to 2005 as well as expression differences
between ambient and high CO2 differed among study populations.
Each of the 3 populations showed unique evolved changes in stomatal index (density) that varied in expression depending on CO2 treatment (treatment × population × year interaction effect P = 0.0003;
Table 1; Figure 2a–c). As a result of among-population response differences, there was no significant main effect of CO2 treatment on
stomata size (NS treatment effect on guard cell length, Table 1); populations showed different patterns of evolutionary change for this
trait as well (population × year interaction, Table 1; Figure 2d–f).
There was no effect of year (or its interactions with treatment
and/or population) on reproductive output (total achene number as of week 8, the onset of senescence). The lack of a treatment
or treatment × year effect most likely resulted from measuring
achene production as of this constant age, which did not capture
the reproductive-output effects of delayed senescence and greater
lifetime net assimilation in the 2005 genotypes grown at ambient
CO2. However, the constant-age reproductive output measurement
revealed marked differences among the 3 populations across both
hot, dry field environments and collection years (note that since
populations did not differ in photosynthetic rate or senescence timing responses, these population differences are expected to remain
consistent with respect to lifetime achene number). On average, both
1994 and 2005 genotypes from the ARL and WEI populations had
produced c. 2.5× more achenes by week 8 in both treatments than
genotypes from the ORD population (significant effect of population, NS main and interaction effects of year and treatment; Table 1;
Figure 3). Populations also contained different amounts of genetic
variation for this component of fitness. While there was significant
genetic variation for week 8 reproductive output across all 3 populations (significant main effect of genotype, Table 1), genetic variation
in reproductive output was (marginally nonsignificantly) greater in
the ARL and WEI populations than in the ORD population (P < 0.06
in Brown–Forsythe tests for all pairwise comparisons of ARL or WEI
vs. ORD in each treatment). Norm of reaction plots showed that
the 2005 samples from the former 2 populations included certain
genotypes that expressed high reproductive output at elevated CO2,
while this was less true in the ORD population (Figure 4). Many
genotypes from ARL and WEI achieved substantially higher reproductive output than any genotypes from ORD. This suggests that
46
Journal of Heredity, 2016, Vol. 107, No. 1
Figure 2. Expression of recently evolved changes in leaf anatomical traits of Polygonum cespitosum in 2 novel field environments. Means are shown ± 1 SE for
8–10 genotypes per collection year for 3 field populations for (a–c) stomatal index (frequency of stomates as a proportion of epidermal cells) and (d–f) mean
guard cell length (µm). P values are given from 1-way ANOVA testing for year effect within each treatment (see Materials and Methods). NS P > 0.20.
Figure 3. Total number of achenes produced by individual plants of
Polygonum cespitosum (as of week 8) in 2 novel field environments.
Population means ± 1 SE in each environment are shown, based on 3
replicates per environment of 18–20 genotypes per population (pooled
across 1994 and 2005 collection years).
ARL and WEI have the potential to evolve higher fitness in these
novel environments than ORD.
Discussion
We compared phenotypic expression by genotypes of the invasive
annual P. cespitosum that were collected from populations in the
species’ introduced range across a recent, 11-generation evolutionary interval (1994 to 2005) and grown at a FACE site in simulated
future conditions. This field environment was sunnier, more exposed,
and drier than the species’ existing (and previous) habitat range, so
this design tested both recent evolutionary changes and their phenotypic expression under novel environmental demands.
In the novel field environment that included current ambient
CO2 concentration (c. 380 ppm), plants from all 3 study populations
expressed shared, recent changes in 2 functional traits: 2005-collected
genotypes increased instantaneous photosynthetic rate (per unit leaf
area), and delayed senescence, compared to 1994 genotypes. Such
rapid evolution of traits of clear functional importance has been
documented in a number of introduced species (Moczek and Nijhout
2003; Bossdorf et al. 2005; Matesanz et al. 2010; Grossman and
Rice 2014). Both of the changed responses documented here would
likely benefit plants in natural populations encountering sunny, dry
conditions. Higher photosynthetic rates could facilitate P. cespitosum’s spread into sunny, resource-rich sites by increasing individual
growth rate and consequently fitness in such habitats.
The rapid, recent evolution of increased photosynthetic performance in full-sun field conditions confirms previous results from a
series of glasshouse resurrection studies documenting both recent
evolutionary change in photosynthetic response to high light and
the adaptive value of that change, in these same P. cespitosum populations (Sultan et al. 2012). In the glasshouse experiments, 2005
genotypes increased instantaneous photosynthetic rate by 20%
more in high light than did 1994 genotypes; in the present study, the
plants expressed a more modest 13% increase, possibly reflecting
the greater moisture and heat stress at the FACE site. The glasshouse
data further showed that 2005 genotypes had significantly higher
reproductive output than 1994 genotypes at high-light growth treatments with both moist and dry soil, showing that introduced-range
populations of this previously shade-distributed species had evolved
to express greater fitness in open conditions (Sultan et al. 2012).
Intriguingly, this evolutionary change in individual patterns of environmental response (or norms of reaction) appears to have resulted
from selection for greater adaptive plasticity in the variable light and
moisture environments that characterize these populations (discussion and references in Sultan et al. 2012).
The present study also revealed that plants in the study populations had evolved norms of reaction that delay senescence at the
FACE site under ambient CO2. In a species with indeterminate
growth and flowering such as P. cespitosum (and indeed most if not
all invasives), this phenological change would allow for an extended
period of assimilation, shoot growth, and fruit production, leading
to greater reproductive output (Arntz and Delph 2001; Gregersen
et al. 2013). Comparative studies have shown that invasives often
show greater shifts in phenology in response to climate change than
Journal of Heredity, 2016, Vol. 107, No. 1
47
Figure 4. Within-population quantitative genetic variation for total achene number of Polygonum cespitosum plants expressed in 2 novel field environments.
Individual norms of reaction are shown for 2005-collected genotypes in each population (genotypic means for each environment are based on 3 inbred replicates
per environment).
do native taxa, allowing them to extend their growing season in
ways that contribute importantly to their success in the introduced
range (Willis et al 2010; Wolkovich et al 2013).
The expression by 2005 plants of increased photosynthetic rate
and delayed senescence at the sunny, dry FACE site exemplifies the
way previously evolved functional trait plasticity may contribute to
phenotypes that are adaptive under novel, future conditions. Indeed,
existing adaptive norms of reaction are expected to play a critical
role in allowing organisms to initially accommodate the rapid environmental shifts that characterize global change (Carroll et al. 2007;
Sultan 2007; Nicotra et al. 2010; Vedder et al. 2013). However, neither the increased photosynthetic rate nor the delay in senescence
expressed by recently sampled P. cespitosum plants at the FACE
site was found when the novel field conditions included the predicted high future level of atmospheric CO2. This finding exemplifies a potentially critical limit on adaptation to future environments:
adaptive patterns of environmental response may not be expressed
at elevated CO2, which directly or indirectly alters numerous regulatory elements affecting development and gene expression in plants
(e.g. Ainsworth et al. 2006; Song et al. 2009) and other organisms
as well (for instance bacteria, Hoffmaster and Koehler 1997, and
marine animals, Portner et al. 2004; Todgham and Hofmann 2009).
Moreover, combinations of future environmental conditions may
affect phenotypic expression in ways that are difficult to predict
based on single factors. For instance, high CO2 can initially increase
photosynthetic rates in many herbaceous C3 plants such as P. cespitosum, but such responses are much diminished at high temperatures
(Poorter and Navas 2003; Wang et al. 2012).
The fact that these recently evolved physiological and life-history traits were not expressed by P. cespitosum plants grown at
elevated CO2 in a sunny, hot site shows that even in species capable
of rapidly evolving adaptive norms of reaction (and therefore predicted to successfully accommodate changing environments), success under global change can be constrained by the effect of future
environments on phenotypic expression. Species-specific phenotypic
responses to predicted global changes in CO2 level, as well as in temperature and precipitation, may either promote or reduce the functional and competitive success of invasive plants, and hence their
future impact (Bradley et al. 2010; Dukes et al 2011). Even more
broadly, the ability of both introduced and native taxa to produce
phenotypes that are functionally adaptive to future conditions may
be limited in unpredictable ways, because a novel environmental factor or combination of factors can disrupt the transduction pathway
leading to expression of adaptive phenotypes in a given developmental system (Sultan 2007; Ghalambor et al. 2007; Gilbert and Epel
2008). It is important to note, however, that organisms have not yet
been subjected to prolonged directional selection under the very high
levels of atmospheric CO2 that are anticipated. The extent to which
such selection may lead to the evolution of altered developmental
systems able to express adaptive phenotypes under elevated CO2 is
not yet known.
Note that the adaptive importance of specific traits may also differ in future conditions, since the contribution to fitness of a given
trait varies depending on the environment and its particular functional demands (Sultan 2015; e.g. Dudley 1996). For instance, when
grown at elevated CO2, plants that use C3 photosynthesis (the vast
majority of species) almost universally reduce stomatal aperture
and consequently water conductance (while still admitting the same
quantity of carbon for fixation; Ainsworth and Long 2005). As a
result, such plants may lose less water relative to carbon gain, which
could reduce drought stress and therefore selection for droughtresistance traits. Regardless of collection year, P. cespitosum plants
grown at elevated CO2 showed this expected reduction in stomatal
conductance. This result suggests that the increased plasticity for
allocation to water-collecting root tissues expressed by 2005 genotypes in response to dry or mesic, full-sun conditions (Sultan et al.
2012) may be less important for future adaptive success if such conditions also include elevated CO2.
Two further limits to future adaptation can occur at the intraspecific level: as a result of their prior evolutionary histories, all populations of a given species may not be equally able to 1) maintain or
2) selectively improve fitness in novel conditions. These populationlevel limits also arise from patterns of phenotypic expression in specific environmental conditions. The first point is illustrated by the
consistently low reproductive output of genotypes from the ORD
population in both of the sunny, dry FACE site treatments. (Although
we found no effect of CO2 treatment on reproductive output, it is
likely that all populations would have shown a reduction in lifetime
achene number at the elevated CO2 treatment compared with the
ambient treatment if measured following complete senescence; see
Results). This pronounced among-population fitness difference was
unexpected, as these genotypes had reproduced no less than those
from the other populations in a range of glasshouse environments
that simulated the species’ current North American habitats (Sultan
et al. 2012). Evidently, existing genetic elements resulted in a very
different fitness outcome under the novel, stressful conditions at the
FACE site.
Performance in novel environments may be influenced by “cryptic” genetic variation that is either not expressed or expressed with
no fitness consequences, in more common environments (Weinig and
Schmitt 2004; Ghalambor et al. 2007; Le Rouzic & Carlborg 2008;
Paaby and Rockman 2014). Such cryptic, “conditionally neutral”
Journal of Heredity, 2016, Vol. 107, No. 1
48
variants can persist or accumulate in populations that lie within
those environmental parameters, yet lead to the expression of new
phenotypes—whether adaptive or maladaptive—in novel or extreme
conditions (Moczek et al. 2011; Anderson et al. 2013; and references
therein). Such variation may originate by mutation, either in source
populations or in situ, or be introduced via gene flow. Depending on
the prevalence and distribution of such conditionally neutral loci in
natural systems (Ledón-Rettig et al. 2014), population differences in
performance and evolutionary potential in future environments may
be dramatic and unpredictable (e.g. Al-Hiyaly et al 1988; Travisano
et al 1995; Kelly et al. 2011). That simulated future environments
exposed dramatic, previously cryptic differences in reproductive
output suggests that the long-term persistence of local P. cespitosum
populations, and their relative contributions to the species’ invasive
spread, may vary across the species’ North American range.
Finally, populations may differ in their evolutionary potential
to further adapt to the novel environmental challenges predicted to
comprise global change. Such selective change requires the presence
of genotypic variants able to express phenotypes that realize high
fitness in future conditions. Those conditions will certainly include
high atmospheric CO2 as well as climatic changes to temperature
and precipitation patterns. The three introduced-range populations
of P. cespitosum revealed different amounts of quantitative-genetic
variation for reproductive output as expressed in the FACE site
environments. Based on norms of reaction for our modest genotypic sample, 2 of the populations included several genotypes that
reproduced well (i.e. produced ≥150 achenes) in the hot/dry field
conditions, and in certain cases maintained or exceeded this level
of reproductive output under elevated CO2. The third study population, which had significantly lower average reproductive output
(ORD, discussed above), lacked such genotypes.
As a result of their particular evolutionary history, certain populations of a given species can lack the genetic variation that might
possibly fuel the evolution of adaptive responses to future environments (e.g. Kelly et al. 2011). Polygonum cespitosum is a largely
inbreeding species yet outcrossing takes place as well; although most
achenes simply fall near the parent plant, occasional dispersal via
animals or humans can initiate new populations in distant sites and,
depending on patterns of movement, contribute to genetic heterogeneity. In such systems, the impact of founder effects, genetic drift,
and low gene flow are likely to vary considerably among populations, leading to differences in the amount of genetic variation,
including quantitative genetic variation for phenotypic responses to
novel environments. Significant population-level differences in this
aspect of evolutionary potential were confirmed in a larger study
of northeastern North American populations of P. cespitosum, in
which functional and fitness norms of reaction were determined for
c. 20 genotypes from each of 6 populations grown in 2 contrasting
glasshouse “habitats” including a sunny, dry treatment (Matesanz
et al. 2014a; for further information on population structure and
quantitative genetic variation in these introduced-range populations,
see Matesanz and Sultan 2013; Matesanz et al. 2014b).
The population-specific evolutionary changes from 1994 to 2005
in the size and frequency of stomata produced by plants in the 2
FACE-site environments also are likely reflections of local genetic
constraints (see Lee 2002). These leaf anatomical traits are generally expected to evolve slowly (on the scale of thousands of years or
more) and predictably, tracking long-term changes in atmospheric
CO2 concentrations (Royer 2001). It is intriguing that in this invasive, environmentally plastic annual plant, the expression of stomatal traits appears to be quite evolutionarily labile.
Conclusions
As a result of global change, future environments within the North
American range of the invasive P. cespitosum are expected to include
high light, greater heat, drought stress, and elevated CO2 concentration, conditions that the species has not encountered in its previous
selective history. Our findings revealed three potential limits to successful adaptation to these future conditions, even in an evolutionarily labile species such as this invasive annual plant. 1) Although
North American populations of P. cespitosum have recently evolved
to increase photosynthetic rate and delay senescence under full-sun,
dry conditions, this adaptive plasticity was not expressed under
elevated CO2. 2) The populations differed markedly in reproductive output in these novel, stressful conditions, possibly reflecting
the accumulation of genetic variation that is cryptic (conditionally
neutral) in the species’ current habitats; and 3) quantitative genetic
variation for fitness response to simulated future conditions differed
among populations, indicating differences in evolutionary potential
to cope with predicted environmental change through further adaptive evolution. The ability of invasive species and other organisms
to succeed in future environments depends on both the expression
of functionally adaptive phenotypes from existing genotypes, and
the availability of genetic variation for further adaptive evolution,
in those environments. Norm of reaction studies under naturalistic
future conditions can provide this critically important information.
Funding
Mellon Foundation Environmental Fellowship awarded
T. H.-K. and a Wesleyan University Project Grant to S.E.S.
to
Acknowledgments
We thank Prof. Stephen P. Long, R. Shekar, and the Long Lab at University of
Illinois for generously providing housing and access to the FACE site and laboratory facilities. We thank J. Spring and E. Vogel for their assistance with data
collection. We also thank Dr Robin Waples and the organizers of the American
Genetics Association 2014 Presidential Symposium.
Data Availability
Data deposited at Dryad: http://dx.doi.org/doi:10.5061/dryad.22hb2
References
Ainsworth EA, Long SP. 2005. What have we learned from 15 years of freeair CO2 enrichment (FACE)? A meta-analytic review of the responses of
photosynthesis, canopy properties and plant production to rising CO2.
New Phytologist 165:351–371.
Ainsworth, EA, Rogers A,Vodkin LO, Walter A, Schurr U. 2006. The effects of
elevated CO2 concentration on soybean gene expression. An analysis of
growing and mature leaves. Plant Physiology. 142:135–147.
Al-Hiyaly SA, McNeily T, Bradshaw AD. 1988. The effects of zinc contamination from electricity pylons - evolution in a replicated situation. New
Phytologist. 110:571–580.
Anderson JT, Lee CR, Rushworth CA, Colautti RI, Mitchell-Olds T. 2013.
Genetic trade-offs and conditional neutrality contribute to local adaptation. Molecular Ecology. 22:699–708.
Arntz AM, Delph LF. 2001. Pattern and process: evidence for the evolution
of photosynthetic traits in natural populations. Oecologia. 127:455–467.
Baker CS. 2013. Journal of heredity adopts joint data archiving policy. Journal
of Heredity. 104:1–1.
Bossdorf O, Auge H, Lafuma L, Rogers WE, Siemann E, Prati D. 2005. Phenotypic and genetic differentiation between native and introduced plant
populations. Oecologia. 144:1–11.
Journal of Heredity, 2016, Vol. 107, No. 1
Bradley BA, Blumenthal DM, Wilcove DS, Ziska LH. 2010. Predicting plant
invasions in an era of global change. Trends in Ecology & Evolution.
25:310–318.
Brook BW, Sodhi NS, Bradshaw CJ. 2008. Synergies among extinction drivers
under global change. Trends in Ecology & Evolution. 23:453–460.
Carroll, SP, Hendry AP, Reznick DN, Fox DW. 2007. Evolution on ecological
time‐scales. Functional Ecology. 21:387–393.
Christensen JH, Hewitson B, Busuioc A, Chen A, Gao X, Held I, Jones R,
Kolli RK, Kwon W-T, Laprise R, Rueda VM, Mearns L, Menéndez CG,
Räisänen J, Rinke A, Whetton ASP. 2007. Regional climate projections.
Contribution of Working Group I to the fourth assessment report of the
Intergovernmental Panel on Climate Change. In: Solomon S, Qin D, Manning M, Chen Z, Marquis M, Averyt KB, Miller MTHL, editors. Climate
Change 2007: The Physical Science Basis. Cambridge (UK); New York
(NY): Cambridge University Press. p. 847–943.
Dudley SA. 1996. Differing selection on plant physiological traits in response
to environmental water availability: a test of adaptive hypotheses. Evolution. 50:92–102.
Dukes JS, Chiariello NR, Loarie SR, Field CB. 2011. Strong response of an
invasive plant species (Centaurea solstitialis L.) to global environmental
changes. Ecological Applications. 21:1887–1894.
Etterson JR, Shaw RG. 2001. Constraint to adaptive evolution in response to
global warming. Science. 294:151–154.
Franks SJ, Avise JC, Bradshaw WE, Conner JK, Etterson JR, Mazer SJ, Shaw
RG, Weis AE. 2008. The resurrection initiative: storing ancestral genotypes to capture evolution in action. Bioscience. 58:870–873.
Ghalambor CK, McKay JK, Carroll SP, Reznick DN. 2007. Adaptive versus
non-adaptive phenotypic plasticity and the potential for contemporary
adaptation in new environments. Functional Ecology. 21:394–407.
Gilbert SF. 2001. Ecological developmental biology: developmental biology
meets the real world. Developmental Biology. 233:1–12.
Gilbert SF, Epel D. 2008. Ecological Developmental Biology: Integrating Epigenetics, Medicine, and Evolution. Sunderland (MA): Sinauer Associates Inc.
Gilbert SF, Epel D. 2015. Ecological Developmental Biology: The Environmental Regulation of Development, Health, and Evolution 2nd ed. Sunderland (MA): Sinauer Associates Inc.
Gregersen PL, Culetic A, Boschian L, Krupinska K. 2013. Plant senescence and
crop productivity. Plant Molecular Biology 82:603–622.
Griffith, T, Sultan SE. 2012. Field‐based insights to the evolution of specialization: plasticity and fitness across habitats in a specialist/generalist species
pair. Ecology and Evolution. 2:778–791.
Grossman JD, Rice KJ. 2014. Contemporary evolution of an invasive grass in
response to elevated atmospheric CO(2) at a Mojave Desert FACE site.
Ecology Letters. 17:710–716.
Hoffmaster AR, Koehler TM. 1997. The anthrax toxin activator gene atxA
is associated with CO2-enhanced non-toxin gene expression in Bacillus
anthracis. Infection and Immunity. 65:3091–3099.
Horgan-Kobelski, TP. 2010. Contemporary evolution, response to novel environments, and ecological breadth in the invasive annual Polygonum cespitosum. [master’s thesis]. Middletown, CT: Biology Department, Wesleyan
University.
Jump AS, Peñuelas J. 2005. Running to stand still: adaptation and the response
of plants to rapid climate change. Ecology Letters. 8:1010–1020.
Karl TR, Melillo JM, Peterson TC, Hassol SJ. 2009. Global Climate Change
Impacts in the United States. New York: Cambridge University Press.
Kelly MW, Sanford E, Grosberg RK. 2011. Limited potential for adaptation to climate change in a broadly distributed marine crustacean. Proceedings of the Royal Society of London. Series B, Biological Sciences.
279:349–356.
Lau JA, Shaw RG, Reich PB, Shaw FH, Tiffin P. 2007. Strong ecological but
weak evolutionary effects of elevated CO2 on a recombinant inbred population of Arabidopsis thaliana. New Phytologist. 175:351–362.
Leakey ADB, Bernacchi CJ, Dohleman FG, Ort DR, Long SP. 2004. Will photosynthesis of maize (Zea mays) in the US Corn Belt increase in future
CO2 rich atmospheres? An analysis of diurnal courses of CO2 uptake
under free-air concentration enrichment (FACE). Global Change Biology.
10:951–962.
49
Ledón-Rettig CC, Pfennig DW, Chunco AJ, Dworkin I. 2014. Cryptic genetic
variation in natural populations: a predictive framework. Integrative and
Comparative Biology. 54:783–793.
Lee CE. 2002. Evolutionary genetics of invasive species. Trends in Ecology &
Evolution. 17:386–391.
Le Rouzic A, Carlborg O. 2008. Evolutionary potential of hidden genetic variation. Trends in Ecology & Evolution. 23:33–37.
Matesanz S, Gianoli E, Valladares F. 2010. Global change and the evolution
of phenotypic plasticity in plants. Annals of the New York Academy of
Sciences 1206:35–55.
Matesanz S, Horgan-Kobelski T, Sultan SE. 2012. Phenotypic plasticity and
population differentiation in an ongoing species invasion. PLoS One.
7:e44955.
Matesanz S, Horgan-Kobelski T, Sultan SE. 2014a. Contrasting levels of evolutionary potential in populations of the invasive plant Polygonum cespitosum. Biological Invasions. 16:455–468.
Matesanz S, Horgan-Kobelski T, Sultan SE. 2015. Evidence for ecological
range expansion in a newly invasive plant. AoB Plants. 7:plv038.
Matesanz S, Sultan SE. 2013. High-performance genotypes in an introduced
plant: insights to future invasiveness. Ecology. 94:2464–2474.
Matesanz S, Theiss KE, Holsinger KE, Sultan SE. 2014b. Genetic diversity
and population structure in Polygonum cespitosum: insights to an ongoing
plant invasion. PLoS One. 9:e93217.
Mehrhoff LJ, Silander JAJ, Leicht SA, Mosher ES, Tabak NM. 2003. IPANE.
Invasive plant atlas of New England. Department of Ecology & Evolutionary Biology, University of Connecticut, Storrs, CT, USA. [cited September 2014]. Available from: URL http://www.ipane.org.
Moczek AP, Nijhout HF. 2003. Rapid evolution of a polyphenic threshold.
Evolution & Development. 5:259–268.
Moczek AP, Sultan SE, Foster S, Ledon-Rettig CC, Dworkin I, Nijhout HF,
Abouheif E, Pfennig DW. 2011. The role of developmental plasticity in
evolutionary innovation. Proceedings of the Royal Society of London.
Series B, Biological Sciences. 278:2705–2713.
Mooney HA, Hobbs RJ. 2000. Invasive Species in a Changing World. Washington (DC): Island Press.
Nicotra AB, Atkin OK, Bonser SP, Davidson AM, Finnegan EJ, Mathesius U,
Poot P, Purugganan MD, Richards CL, Valladares F, et al. 2010. Plant phenotypic plasticity in a changing climate. Trends in Plant Science. 15:684–
692.
Paaby AB, Rockman MV. 2014. Cryptic genetic variation: evolution’s hidden
substrate. Nature Reviews Genetics. 15:247–258.
Parmesan C. 2006. Ecological and evolutionary responses to recent climate
change. Annual Review of Ecology and Systematics 37:637–669.
Paterson AK. 2000. Range expansion of Polygonum caespitosum var. longisetum in the United States. Bartonia. 60:57–69.
Pinto-Marijuan M, Munné-Bosch S. 2013. Ecophysiology of invasive plants:
osmotic adjustment and antioxidants. Trends in Plant Science 18: 660–
666.
Portner HO, Langenbuch M, Reipschlager A. 2004. Biological impact of elevated ocean CO2 concentrations: Lessons from animal physiology and
earth history. Journal of Oceanography. 60:705–718.
Poorter H, Navas M-L. 2003. Plant growth and competition at elevated CO2:
on winners, losers and functional groups. New Phytologist. 157: 175–198.
Reusch T, Wood T. 2007. Molecular ecology of global change. Molecular Ecology. 16:3973–3992.
Royer DL. 2001. Stomatal density and stomatal index as indicators of paleoatmospheric CO(2) concentration. Review of Palaeobotany and Palynology.
114:1–28.
Sala OE, Chapin FSI, Armesto JJ, Berlow E, Bloomfield J, Dirzo R, HuberSanwald E, Huenneke LF, Jackson RB, Kinzig A, et al. 2000. Global biodiversity scenarios for the year 2100. Science. 287:1770–1774.
Song L, Wu J, Li C, Li F, Peng S, Chen B. 2009. Different responses of invasive and native species to elevated CO2 concentration. Acta Oecologica.
35:128–135.
Steinger T, Stephan A, Schmid B. 2007. Predicting adaptive evolution under
elevated atmospheric CO2 in the perennial grass Bromus erectus. Global
Change Biology. 13:1028–1039.
50
Sultan SE. 2000. Phenotypic plasticity for fitness components in Polygonum
species of contrasting ecological breadth. Ecology. 82:328–343.
Sultan SE. 2007. Development in context: the timely emergence of eco-devo.
Trends in Ecology & Evolution. 22:575–582.
Sultan SE. 2015. Organism and Environment: Ecological Development, Niche
Construction, and Adaptation. London: Oxford University Press.
Sultan SE, Horgan-Kobelski T, Nichols LM, Riggs CE, Waples RK. 2013. A
resurrection study reveals rapid adaptive evolution within populations of
an invasive plant. Evolutionary Applications. 6:266–278.
Sultan SE, Wilczek AM, Hann SD, Brosi BJ. 1998. Contrasting ecological
breadth of co-occurring annual Polygonum species. Journal of Ecology.
86:363–383.
Todgham AE, Hofmann GE. 2009. Transcriptomic response of sea urchin larvae Strongylocentrotus purpuratus to CO2-driven seawater acidification.
Journal of Experimental Biology. 212:2579–2594.
Travisano M, Vasi F, Lenski RE. 1995. Long-term experimental evolution in
Escherichia coli. 3. Variation among replicate populations in correlated
responses to novel environments. Evolution. 49:189–200.
Vannette RL, Hunter MD. 2011. Genetic variation in expression of defense
phenotype may mediate evolutionary adaptation of Asclepias syriaca to
elevated CO2. Global Change Biology. 17:1277–1288.
Vedder O, Bouwhuis S, Sheldon BC. 2013. Quantitative assessment of
the importance of phenotypic plasticity in adaptation to climate
change in wild bird populations. Public Library of Science Biology.
11:1001605.
Journal of Heredity, 2016, Vol. 107, No. 1
Walther G-R, Roques A, Hulme PE, Sykes MT, Pysek P, Kuehn I, Zobel M,
Bacher S, Botta-Dukat Z, Bugmann H, et al. 2009. Alien species in a
warmer world: risks and opportunities. Trends in Ecology & Evolution.
24:686–693.
Wang D, Heckathorn SA, Wang X, Philpott SM. 2012. A meta-analysis of
plant physiological and growth responses to temperature and elevated
CO(2). Oecologia. 169:1–13.
Ward JK, Kelly JK. 2004. Scaling up evolutionary responses to elevated CO2:
lessons from Arabidopsis. Ecology Letters. 7:427–440.
Weinig C, Schmitt J. 2004. Environmental effects on the expression of quantitative
trait loci and implications for phenotypic evolution. Bioscience. 54:627–635.
Willis CG, Ruhfel BR, Primack RB, Miller-Rushing AJ, Losos JB, Davis CC.
2010. Favorable climate change response explains non-native species’ success in Thoreau’s woods. PLoS One. 5:e8878.
Wolkovich EM, Davies TJ, Schaefer H, Cleland EE, Cook BI, Travers SE, Willis
CG, Davis CC. 2013. Temperature-dependent shifts in phenology contribute to the success of exotic species with climate change. American Journal
of Botany. 100:1407–1421.
Zar JH. 1999. Biostatistical Analysis. Upper Saddle River, NJ: Prentice Hall.
Zenni RD, Lamy J-B, Lamarque LJ, Porte AJ. 2014. Adaptive evolution and
phenotypic plasticity during naturalization and spread of invasive species: implications for tree invasion biology. Biological Invasions. 16:
635–644.
Ziska LH, Dukes JS. 2011. Weed Biology and Climate Change. Ames (IA):
Wiley-Blackwell.