Available via license: CC BY
Content may be subject to copyright.
Ecology and Evolution. 2024;14:e11069.
|
1 of 10
https://doi.org/10.1002/ece3.11069
www.ecolevol.org
Received:11December2023
|
Accepted :13Decemb er2023
DOI: 10.1002/ece 3.11069
RESEARCH ARTICLE
Freshwater salinization and the evolved tolerance of
amphibians
Rick Relyea | Brian Mattes | Candace Schermerhorn | Isaac Shepard
This is an op en access arti cle under the ter ms of the CreativeCommonsAttribution License, which permits use, distribution and reproduction in any medium,
provide d the original wor k is properly cited.
©2024TheAut hors.Ecology a nd Evolutio n published by John Wiley & So ns Ltd.
Depar tment of Biological Sciences,
Rensselaer Polytechnic Institute, Troy,
NewYork,USA
Correspondence
Rick Relyea, Dep artment of Biological
Sciences, Rensselaer Polytechnic Institute,
Troy,NewYork12180USA.
Email: relyer@rpi.edu
Funding information
Directorate for Biological Sciences,
Grant /AwardNumber:DEB16-55168;
Rensselaer Polytechnic Institute
Abstract
The increasing salinization of freshwaters is a growing environmental issue as a
resultofmining,agriculture,climatechange,andtheapplicationofde-icingsaltsin
regions that experience ice and snow. Due to narrow osmotic limits, many fresh-
water species are particularly susceptible to salinization, but it is possible that re-
peated exposures over time could favor the evolution of increased salt tolerance.
Using collected nine populations of larval wood frogs (Rana sylvatica) as eggs from
ponds and wetlands with close proximity to roads and spanning a wide gradient
of salt concent rations. In the fir st experiment , we used a time-to-de ath experi-
ment to examine the salt tolerance. In a second experiment, we examined whether
populationdifferencesinsalttolerancewereassociatedwithtrade-offsingrowth,
development, or behavior in the presence of control water or a sublethal salt con-
centration. We found that populations collected from ponds with low and interme-
diatesaltconcentrationsexhibitedsimilartolerance curvesover a96-hexposure.
However, the population from a pond with the highest salt concentration exhibited
a much higher tolerance. We also found population differences in growth, develop-
ment, and activity level among the populations, but these were not associated with
population differences in tolerance. In addition, the sublethal concentration of salt
had no impact on growth and development, but it did cause a reduction in tadpole
activity across the populations. Collectively, these results provide further evidence
that some species of freshwater organisms can evolve tolerance to increasing salin-
ization, although it may only occur under relatively high concentrations and with-
outtrade-offsingrowth,development,orbehavior.
KEYWORDS
amphibian, evolution, microevolution, sensitivity, toxicology
TAXONOMY CLASSIFICATION
Ecotoxicology, Evolutionary ecology
2 of 10
|
RELY E A e t al.
1 | INTRODUCTIO N
The salinization of global freshwaters is a global issue with contribu-
tions from mining, agriculture, saltwater intrusion from sea level rise
and storm surge events linked to climate change, and the application
ofde-icing salts on roads in regions that experience snow and ice
(Dugan et al., 2017; Hintz & Relyea, 2019). Indeed, more than 18
millionmetrictonsofde-icingsaltsareappliedtoroadsintheUnited
States each winter (Jackson & Jobbágy, 2005). Much ofthese de-
icing salts wash into adjacent freshwater habitat s (Corsi et al., 2010;
Dugan et al., 2017 ), which has myriad ecological consequences for
aquatic s pecies, communi ties, and ecosys tems (Hintz & Relyea , 2019;
Schuler et al., 20 19). During the past decade, scientists have taken
an “evolutionary toxicology ” approach to better understand how
freshwater organisms respond to salinization (Bickham, 2011; Brady
et al., 2017). This is especially true as evolution via natural selec-
tion to anthropogenic changes on contemporary timescales in other
conservation contexts has gained greater appreciation (Stockwell
et al., 2003; Kinnison et al., 20 07).Indeed,evolvedtolerancetode-
icing salts has been demonstrated in various taxonomic groups in-
cluding zooplankton, (Coldsnow & Relyea, 2018; Hintz et al., 2019)
and aquatic macrophytes (Bora et al., 2020). However, it remains to
beseenwhethertheabilitytoevolve toleranceto de-icingsaltpol-
lution is common among freshwater organisms.
Given their narrow osmotic range (Boutilier et al., 1992), and sensi-
tivity to chloride (Brad y, 2012; Sanzo & Hecnar, 2006) amphibians have
been of par ticular interest to scientists trying to understand the abilit y
ofo rg anis ms toad ap ttothecontinuings tre ssofd e-ic in gs al tp ollution.
Moreover, as amphibians are in decline globally (Stuart et al., 2004),
understanding their ability to adapt to contaminants such as salts has
important conservation implications (Blaustein & Bancroft, 2007 ).
Some evidence suggests that it is possible for amphibians to evolve
increased tolerance to high salt concentrations where they have likely
livedinbrackishhabitatsforcenturies(Gomez-Mestre&Tejedo,2003;
Hopkins & Brodie, 2015). However, freshwater salinization is a much
morerecentphenomenon,particularlyinregionswherede-icingsalts
havebeen increasinglyappliedtoroads sincethe 1940s. Asaresult,
wetlands closer to roads can contain 20 times more salt than wetlands
far from roads and these high salt concentrations can have substan-
tial negative effects on the survival of amphibian embryos and larvae
(Sanzo & Hecnar, 2006; Karraker et al., 20 08; Brady, 2012; Petranka &
Francis, 2013; Hill & Sadowski, 2016).
One amphibian species that has received growing research atten-
tion is the wood frog (Rana sylvatica), which has a geographic range
that spans much of the eastern continental U.S. and Canada over to
Alaska.Wood frogsare anexcellentstudy speciesbecausethey are
explosive breeders, which results in populations throughout a region
all breeding within the same week. This synchronized breeding behav-
ior allows researchers to directly compare populations that are living
in wetlands that vary widely in salinity. In past studies in Connecticut
(USA), wood frog larvae that were collec ted as eggs from ponds
with high salt concentrations exhibited lower sur vival (Brady, 2013,
2017 ), slower growth and development (Forgione & Brady, 2022), and
reduced larval activity (Brady et al., 2022; Hall et al., 2 017) compared
to conspecific populations collected as eggs in wetlands far from
roads. These studies have suggested that wood frogs are maladapted
to salinization. However, other studies of wood frogs in Vermont
(USA)havefoundthatpopulationscomingfromhigh-saltwetlandsdo
not differ in survival, but have superior locomotor performance, adult
mass, and fecundit y, suggesting adapt ation rather than maladapt ation
(Brady et al., 20 19).Population-leveladaptiontohigh-saltconcentra-
tions has also been observed in a species of salamander in roadside
ponds ( Ambystoma maculatum; Brady, 2012), but contradictory results
werefoundinshort-termlabstudies(Bradyetal.,2017).
It currently remains unclear what conditions seem to favor the
evolution of adaptation versus maladaptation to salinization. Most
studies showing maladaptation to salt have compared wood frog pop-
ulations living in wetlands close to roads to wood frog populations
living in wetlands far from roads (Brady, 2013, 20 17). However, close
proximit ytoroads(e.g.,within50 m,wheresaltconcentrationsareel-
evated; Karraker et al., 2008) may present additional selective forces
(e.g., other pollutants) that may affect how populations are able to
evolveadaptationsormaladaptat ionstosalt.Analternat iveapproach
would be to compare populations collected from wetlands that vary
widely in salt concentration but are all in close proximity to roads.
In addition to examining adapt ations and maladaptations to sa-
linization by examining survival when exposed to lethal concentra-
tions, it is also important to examine life history traits (e.g., growth,
development) and behavior when animals are exposed to sublethal
salt concentrations (Brady et al., 2022). This is particularly interesting
when consideringpotential trade-of fsof evolved tolerance, where
we might hypothesize that increased tolerance comes at the cost of
slower growth, slower development, or impaired behavior. For exam-
ple, wood frog populations can evolve tolerance to pesticides at the
expense of parasite resistance (Billet et al., 2021; Hua et al., 2017 ).
To understand how amphibians may be evolving to salinization,
we conducted two experiments using nine wood frog populations
from wetlands that were all in close proximit y to roads but differed
widely in their salt concentrations. The first experiment tested the
hypothesis that populations living in wetlands with high salt concen-
trations have evolved higher salt tolerance. The second experiment
tested the hypothesis that populations that have evolved the highest
salt tole rance will exp erience tra de-of fs of slower grow th, slower
development, or impaired behavior.
2 | METHODS
2.1 | Animal collection
We collected newly oviposited wood frog egg masses from nine road-
sid epondsinupstateNewYork,US Athatvariedi ntheirchloride con-
centrati ons at the time o f egg collec tion, ran ging from 1 to 744 mg
Cl−/L (Figure 1). Given that salinity ef fect s tend to diminish beyond
50 m(Karraker et al., 2008), all pondswere lessthan 40 m from the
nearest road. During April 6–12, 2022,we gathered tenpartial egg
|
3 of 10
RE LY E A e t al.
masses(i.e.,approximatelyone-fourthofamass)fromeightsitesand
seven partial masses from one site where only seven masses were
laid.Weplacedthepartialeggmassesin350-Loutdoorwadingpools
(one wading pool for each population) containing aged well water at
theRensselaerAquaticLaboratoryinTroy,NY.Allpopulationsbegan
hatchingbetweenApril14and 15,2022.Oncetheyhatched,wefed
the tadpoles rabbit chow ad libitum (Blue Seal Fresh Show Hutch
Deluxe).Weusedhatchlingtadpolesfromtheseeggmassesinatime-
to-deathexperimentfollowedbyagrowthanddevelopmentexperi-
ment.Atleast1 weekpriortoeachexperiment,webroughttadpoles
fromeachpopulationintothelabtoacclimate.Inthelab,theanimal-
rearingroomwasmaintainedat21°Cwithalight:darkcycleof15:9 h.
2.2 | Time- to- death experiment
To assess the relative levels of tolerance of each population to road
salt contamination, we conducted a time-to-death (T TD) experi-
mentwhereweexposedtadpolestocontrolwater(19 mgCl−/L) and
alethalconcentrationof8 g/Lof NaCl.Time-to-deathexperiments
intentionally use high concentrations of a pollutant (often higher
than experienced in nature) to determine whether different groups
in individuals (e.g., populations) differ in their relative tolerance over
ashor ttimeperiod(1–4 days),duringwhichtheanimalsarenotfed
(Brady, 2013; Bridges & Semlitsch, 2000; Semlitsch et al., 2000).
OnceTTDexperimentsfinddifferencesintolerancebasedonshort-
term times to death, researchers can examine environmentally rel-
evant concentrations over longer periods of time to determine if
differences in tolerance affect performance.
For this experiment, we used a fully randomized design that in-
volved the nine populations of wood frogs and the two salt treat-
ments. We placed one tadpole into each of the experimental units,
whichwere59 mLclearplasticcupsfilledwithapproximately45 mL
ofwater.Thelethalsaltdosewascreatedbyaddinglab-gradeNaCl
toaged tapwatertoreach aconcentrationof8 g/L NaCl,basedon
the concentration used by Buss et al. (2021). From each population,
we placed 15 individuals intocups containing water with a lethal
salt concentration and five individuals into cups containing control
water. Thus, there was a total of 180 cups in a completely random-
ized design. The control water was t ap water that we aged at least
24 hto allow any chlorinetoof f-gas.Wealsosetasidetenindivid-
uals from each population to estimate the initial average mass and
developmental stage (Gosner, 196 0) of each population. The initial
individualmasses of the populationsranged from33to 64 mg; de-
velopmental stages ranged from approximately 26 to 27 (Table 1).
Westarted the experiment on 26April and team memberstook
turns initially checking the tadp oles every 4 h for mor tality. Once
mortalitybegan,wecheckedtheexperimentevery2 handrecorded
the number of dead tadpoles from each population. We checked
for deaths by gently blowing water at the tadpoles using a pipette.
Nonresponsive tadpoles were considered dead. If it was unclear
whether a tadpole was still alive, we looked for a heartbeat under a
dissecting microscope. We set aside individuals who were declared
dead and double checked them at the subsequent check time. If we
determined that the tadpoles were still alive, we returned them to
theexperimentalarray.Checkscontinuedforatotalof96 h(30April).
Therewasnomortalityinthesalt-freecontrols.Uponcompletionof
theexperiment,weeuthanizedallindividualsusing2%MS-222.
FIGURE 1 Chlorideconcentrationsateachoftheninesitesatthetimeofwoodfrogeggcollection.Sitesarerepresentedbyunique
three-orfour-lettercodes.
4 of 10
|
RELY E A e t al.
Weused pair wise log-ranktestswitha Bonferroni correction
to look for di fferences in t he Kaplan–Meier sur vival cur ves for
each population. While this analysis tests for differences among
the nine populations, it does not show what may be driving dif-
ferences in survival. Thus, our analysis sought to determine the
direct effect of the chloride concentration of the source pond
and the death rates of each individual. To do so, we conducted a
Cox mixed effects model. Our response variable in this model was
the probability of tadpole survival, regardless of the pond from
which they came. Our fixed ef fect s were the source-pond chlo-
ride concentrations, the mean individual mass of tadpoles in each
population, and their interaction. We included population as a ran-
dom effect in the model to control for other differences among
populations, aside from chloride concentration, that we did not
measure. We calculated hazard ratios for the fixed ef fect s using
the model parameter estimates. To do this, we raised the natural
root to the power of the parameter values. Subtrac ting the hazard
ratio from one gives the percent increase in survival that a param-
eter contributes to the overall survival of individuals in the study.
Parameter significance was determined using Wald Chi-square
analyses.AllanalyseswereconductedinRversion4.1.0usingthe
survival (Therneau, 2021), Survminer (Kassambara, 2021), coxme
(Therneau, 2020), and car packages (Fox & Weisberg, 2019).
2.3 | Growth and development experiment
To determine whether evolved salt tolerance came with any growth
ordevelopmentaltradeoffs,weconductedafollow-up experiment
on additional wood frog tadpoles from the same nine populations.
We used a randomized block design for this experiment with three
spatial blocks, where each block was a different shelf height in the
lab. In each block, experimental units were assigned one of the nine
populations and one of two salt treatments (control water with no
salt added vs. a higher sublethal salt concentration). We replicated
the 18 treatment combinations twice within each block, which pro-
vided six replicates across the entire experiment. Thus, there were 36
experimental units per block and a total of 108 experimental units.
Theexperimentalunitswere5-Lwhiteplastictubscontaining4 L
of aged tap water. On 9 May, we placed six tadpoles into each tub.
AswiththeTTDexperiment,wesetasidetenindividualsfromeach
population at the outset of the experiment to quantify the average
developmental stage (Gosner, 196 0) and mass of the individuals
used in the experiment from each population. These initial masses
ranged fr om 62 to 179 mg with developm ent stages r anging from
approximately25to28(Gosner,1960 ; Table 1).
Weagedalltapwaterforatleast24 htoallowchlorinetooff-gas.
Forthe sublethal salttreatment, we addedlab-grade NaCl in aged
tapwater toreach achlorideconcentration of165.5 ± 0.3 mg Cl−/L
(mean ± SE).Thisvalue representsthemid-rangeofchloride values
in the ponds from which we collected the wood frogs (Figure 1). The
tap water in the no-salt treatment had a background chloride con-
centrationof25.9 ± 0.3 mgCl−/L(mean ± SE).
We fed the tadpoles a ration of Tetra goldfish food (Spectrum
Pet Brands LLC) mixed with water ad libitum (starting at approxi-
mately 16.5 m g per tadpo le per day and in creasing up to ap proxi-
mately 22.2 mgpertadpoleperdayasthetadpolesgrew)overthe
course of the experiment. We conducted water changes approxi-
mately every4 days over thecourse of the experiment whenever
the water appeared murky.
Nine days into the experiment, we quantified tadpole activity
using scan sampling. Two observers made 10 observations each of
every tub over thecourse of 75 min, which produced 20 observa-
tions of each tub. The observers slowly walked by the t anks and
peered in so as not to disturb the t adpoles. For each tub, we re-
corded the number of individuals moving at a given moment. We
conducted the observations prior to feeding, given that satiated
tadpoles substantially reduce their activity to low levels. We used
the mean percentage of individuals moving in each tub as our re-
sponse variable. For the two tubs that had a tadpole not survive, we
adjusted the percent activity to reflect the reduced number of tad-
poles in that tub. Sur vival across the entire experiment was 99.7%.
Theexperimentwasterminatedafter22 days(31May),whenthe
tadpoles approached metamorphosis (i.e., Gosner stage 39). We then
determined the mass and Gosner stage of each tadpole and used the
mean values of each experimental unit as our response variables.
We used a linear mixed effects model to look for differences in
the relative growth rate of the tadpoles from the nine different pop-
ulations exposed to t wo salt treatments. Relative growth rate was
calculated as the difference in the natural log of final individual mass
and initial population mean mass, divided by the duration of the ex-
periment(i.e.,22 days).Ourmodelusedchlorideconcentrationsofthe
source ponds, salt treatment (salt or no salt), and the interaction be-
tween these variables as the fixed effects. We had a series of nested
random effects in this model, including block, population, and tub,
nested in that order. These random effects were in place to account
TAB LE 1 Themean(±SE) initial mass and median Gosner
developmental stage of wood frog tadpoles from each population
wereusedinthetime-to-death(TTD)andgrowth-rate
experiments.
Population
TTD experiment Growth experiment
Mass
(mg)
Gosner
stage Mass (mg)
Gosner
stage
HST 46 ± 3 26 143 ± 8 28
ATL 52 ± 4 27 116 ± 8 27
ALG 60 ± 3 26 101 ± 9 27
DCH 56 ± 2 26 128 ± 9 28
BOB 33 ± 2 26 124 ± 7 27
DBY 45 ± 4 26 112 ± 5 27
EGTP 64 ± 2 27 125 ± 7 27
COL 48 ± 2 26 140 ± 6 28
APHS 43 ± 3 26 120 ± 4 27
Note: Site abbreviations are the same as in Figure 1.
|
5 of 10
RE LY E A e t al.
for additional variation introduced by the blocks, the individual tubs,
and other aspect s of the natal ponds that we did not measure directly.
We looked at the changes in tadpole development using a mixed
effec ts model almost identic al to the one we used for relative grow th
rate. In this instance, the response variable was the change in the
Gosner stage for each individual tadpole. We calculated the change
in the Gosner stage by subtracting the average initial Gosner stage
of a given population from the final, measured Gosner stage of each
individual in the experiment. Our fixed and random ef fects for this
model were identical to the model examining relative growth rate.
Finally, we used a binomial mixed effects model to look for
differences in ac tivit y among the nine populations in the two salt
treatments. In this analysis, our response variable was whether an
individual was moving or not during each of the twent y obser va-
tion periods, hence the binomial model. The fixed and random ef-
fects variables in this model were the same as in the growth and
development analyses. For each of these analyses (relative growth
rate, development, activity), we used Wald II Chi square analyses
to determine the significance of the fixed effects (chloride concen-
tration, salt treatment, or their interaction). We conducted all our
analyses in R version 4.1.0 (R Core Team 2022) using the lme4 (Bates
et al., 2015) and car (Fox & Weisberg, 2 019) packages.
3 | RESULTS
3.1 | TTD experiment
ThewoodfrogsfromtheAPHSpopulationssurvivedabouttwiceas
longinthe lethal-salttreatment comparedtotheothereightpopu-
lations (Figure 2).Indeed,ourlog-ranktest showedthatthis group
was statistically different from all the others (all pairwise p values for
APHS < 0.01).TheAPHSpopulationalsohadthehighestconcentra-
tion of chloride at the time of egg collection (Figure 1). The survival
curves for the other eight the populations did not differ from each
other (all pairwise pvalues > .2). OurCoxmixed effects model, de-
signed to determine the explicit link bet ween chloride concentration
and survival, showed that chloride concentration had a significant,
negative effect on the probabilit y of dying in the TTD experiment
(Table 2). In other words, individuals collected from ponds with
higher chloride concentrations were more likely to survive than in-
dividuals collected from ponds with lower concentrations. However,
this effect was quite weak, with a Hazard Ratio of more than 0.99
meaning that for wood frogs in our study, a single unit increase in
chloride concentration confers less than a 1% increase in likelihood
of survival Indeed, it is likely that this weak but st atistically signifi-
cantrelationshipisdrivenentirelybytheAPHSpopulation.
3.2 | Growth and development experiment
While there were differences in the relative grow th rates of the tad-
poles (Table 3, Figure 3a), these dif ferences were not explained by
the chloride concentration of their natal pond, the salt treatments,
or their interaction. Indeed, none of our model terms were signifi-
cant and all effec t sizes were small.
Similarly, while there were some differences in development
among the nine populations (Table 3, Figure 3b), there were no sig-
nificant effec ts on development for any of the variables we included
in our models.
Tadpole activity was not affected by population, but was af-
fected by the salt treatment (Table 3, Figure 3c). Across all po pu-
lations, the moderate salt concentration made tadpoles less active.
There was no interac tion between population and salt treatment.
FIGURE 2 Thepropor tionofwoodfrogsalivefromeachoftheninepopulationsusedinthetime-to-deathexperimentwhenexposed
to8 g/LofNaCl.Warmercolorsindicatepopulationswhosesourcepondhadhigherchlorideconcentrationsandcoolercolorsindicate
populationswhosesourcepondhadlowerchlorideconcentrations.Theasteriskindic atessignificantlydifferentKaplan–Meyersurvivalas
calculatedusinglog-ranktest s(seetex t).Therewerenodeathsintheno-saltcontrols,sothesurvivalcurvesfortheninecontrolsarenot
shown.
6 of 10
|
RELY E A e t al.
4 | DISCUSSION
The tolerance, grow th, development, and behavior of the wood frog
tadpoles in our experiments were largely consistent between each
of the nine populations with one notable exception. The popula-
tionfromthepond with the highestchlorideconcentration(APHS)
showed a muc h higher toleranc e to de-icing s alts than the oth er
eight, as demonstrated by its higher survival in the lethal salt dose
experiment. This finding suppor ts our hypothesis that wood frogs
from wetlands with higher salt concentrations can evolve higher salt
tolerance.
Prior to this study, there was mixed evidence about the evo-
lutiona ry responses of w ood frogs to de-icing sal ts. Our findin gs
agree with Brady et al. (20 19), who found that wood frog popula-
tions from Vermont exhibited superior locomotor performance
and fitn ess-re lated trait s (e.g., body m ass, fecund ity) if they c ame
from roadside ponds compared to populations from ponds far from
roads. That study found no differences in sur vival between roadside
versus woodland populations. However, it contrasted with studies
of wood frog populations from Connecticut that have shown that
de-icingsaltscanleadtomaladaptation(Brady,2013, 2017). When
studies have investigated the ability for wood frogs to adapt to
roadside habitat s with de-icing salt pollution, they havecompared
roadsidepopulationstothosefarfromroadsthatreceivenode-icing
salt pollution. Other work has shown that proximity to roads better
explains egg density than salinit y (Karraker et al., 2008) suggesting
that for some maladapted traits, aspects of being near a road other
than salinity may be important for determining the fitness of popu-
lations of wood frogs in these habitats. This could potentially explain
why nearly all our populations showed identical responses in their
growth and activity levels. That is, the evolved responses of wood
frogs to de -icing sa lts may also be af fected by fa ctors associ ated
with roadside ponds.
One potential explanation for the pattern in our data is that the
APHS population was only recently established and therefore had
not evolved maladaptation the ways that the other eight popula-
tions had. In some populations of wood frogs, individuals from ponds
far from roads have higher survival in high salt environments than
those from roadside ponds (Brady, 2013, 2017; Brady et al., 2022;
Forgione & Brady, 2022; Hall et al., 2017 ). However, it remains un-
certain how many generations it takes for maladaptation to develop
in wood frog populations living adjacent to roads. Rapid evolution,
including adaptation and maladaptation, still requires multiple gen-
erations for trait change to occur (Hendr y, 2017 ). However, this ex-
planation would assume that maladaptation is the norm across all
populations which isn't the case (Brady et al., 2 019).
An alternative explanation is that the APHS population has
evolvedhighertolerancetode-icingsaltsduetothehigherconcen-
tration of salt in their source pond. Whether the increased toler-
ance we observed is the product of true evolution or phenot ypic
plasticity (e.g., Hua et al., 2015) is an open question. Regardless, the
fact that we only saw increased tolerance in the population coming
from the pond with the highest chloride concentration would sug-
gest that there may be some threshold level of chloride pollution
necessary for wood frogs to respond. Our highest salt concentration
(774 mgCl−/L) is similar to the highest concentrations obser ved in
other studies, yet several of those studies did not detect increased
tolerance in the highest salt populations (Brady & Goedert , 2017;
Brady et al., 2017). However, as noted earlier, a subsequent study in
Vermontdiddetectsuperiorfitness-relatedtraitsinroadsidepopu-
lations living with elevated salt concentrations (Brady et al., 2019).
Alternatively, the higher chloride tolerance of the APHS pop-
ulation could be the result of phenotypic plasticity. While all the
TAB LE 2 TheresultsofaCoxmixed-effec tsmodelexaminingchlorideconcentrationofthesourceponds,tadpolemass,andtheir
interactionontadpolesurvivalinthetime-to-deathexperiment.
Parameter Estimate Variation HR d.f.
X2
p
Pond chloride −9.9 3 × 10 −3 5.97 × 10−3 0.990 18. 41 .004
Mass −1. 94 × 10−2 2.84 × 10−2 0.028 10.04 .834
Chloride × Mass 1.62 × 10 −4 1.27 × 10−4 0.00 01 11.65 .199
Population (random effect) 0.2 59 0. 51 – – – –
Note: The p opulation was included in the model as a random effect. The estimate column shows the calculated parameter values for the fixed effects
and the amount of variation explained by the random effect. The variation column shows the standard errors for the fixed effe cts and the st andard
deviation for the random ef fect of the population. HR is the hazard ratio, calculated by t aking the exponent of the coefficient. Degrees of freedom,
X2
values, and p values were calculated using a Type II Wald
X2
analysis.
TABLE 3 ANOVAresult sofwoodfrogpopulationandsalt
treatment on the relative growth rate, Gosner developmental stage,
and activity of t adpoles.
Response variable
Explanatory
variable d.f. F p
Relative growth
rate
Population (P) 833.65 <.0 01
Salt (S) 1 1.49 .225
P × S 81.33 .238
Gosner
developmental
stage
Population (P) 81.95 .62
Salt (S) 1 0.01 .919
P × S 80.214 .988
Activity Population (P) 81.06 .40
Salt (S) 1 5.00 .03
P × S 80.74 .66
|
7 of 10
RE LY E A e t al.
tadpoles from each population in our experiments underwent the
same exposure to road salts, plastic responses can still be present in
the form of inherited environmental ef fect s, such as maternal effects
(Rossiter, 1996).Aswithfrequencydependentselection,ourexper-
iment was not designed to test for mechanisms such as maternal
effects. However, in the future, combining laboratory and field ex-
periments would be helpful for exploring this possibility further.
Source pond chloride concentrations and our experimental salt
treatments had no effec t on the growth or development of the wood
frog tadpoles. However, the experimental salt treatment did depress
FIGURE 3 The(a)relativegrowthrate,(b)changeinGosnerstage,and(c)activityoftadpolesfortheninewoodfrogpopulationsin
the presence and absence of a sublethal salt concentration. The populations are plotted in reference to the chloride concentration of their
source ponds. Data points represent means ±1 SE.
8 of 10
|
RELY E A e t al.
activity levels regardless of source pond chloride levels. The reduced
activity levels in response to experimental salt treatment s align with
other studies that have found similar results in wood frogs (Hall
et al., 20 17) and other amphibians (Squires et al., 2010), although
such affects can change over ontogeny (Kearney et al., 2014). This
finding further reinforces the idea that current salt exposure may
be more important for determining behavioral responses than the
historic saltiness of the natal pond.
Interestingly,thehigher tolerance tode-icingsaltsin the APHS
population did not lead to differences in growth rate, development,
or activity. This is surprising, as there are good reasons to believe
thatthesetraitsshouldbeeitherhigherorlowerintheAPHSpopu-
lation compared to the others. For example, it would make sense that
havingahighertolerancetode-icingsaltswouldcoincidewithfaster
growth rates and higher activity levels, as shown in other studies
(Hall et al., 2017 ).Alternatively,itwouldnothavebeensurprisingto
see a tradeoff in growth and development rates with increased envi-
ronmental stress (Berven et al., 1979). Indeed, tradeoffs associated
with tolerance to pollutants have been recorded in wood frogs in
other contexts (Hua et al., 2017). However, the growth and activit y
levels of the populations were not explained by the concentration of
chloride in their natal ponds. It's possible that tradeoffs exist along
other trait axes that we did not measure, such as parasite resistance,
which has been shown to decrease with increasing chloride concen-
trations (Milotic et al., 2017). However, given the myriad of ways that
maladaptationtode-icingsaltshasbeenrecordedinwoodfrogs,we
must be careful to design studies that accurately assess whether
trait changes are the result of tradeof fs or direct selection pressures
(Schluter et al., 1991).
Unfortunately, we do not have historic al records of the chloride
concentrations in any of the ponds we used in our experiment, but
it seems likely that similar concentrations have been occurring for
decades.Whiletheconstituentsodiumandchlorideionsofde-icing
salts are typically conserved in waterbodies (Dugan et al., 2017;
Godwin et al., 2003), t hey can fluct uate, especia lly since de-icing
salt application is not always consistent due to variation in weather
patterns (Venäläinen, 2001). While we collected wood frogs from
ponds that had a gradient in chloride concentrations at the time of
collection, it is possible that these concentrations could experience
springtime variation over multiple years.
That the population of wood frogs from the pond with the
highest de-icing salt pollution showed the highest tolerance
was encouraging from a conservation perspective. However,
it is impor tant that we exercise caution when extrapolating
short-term, high-concentration salt exposures to longer-term,
lower-concentrationexperiments innature,becausesometimes
one can observed contradictor y outcomes, as observed in lar-
val Ambystoma salamander populations (Brady, 2012; Brady
et al., 2017). Moving forward, it will be necessary to begin fo-
cusing research effort s on populations of wood frogs from ponds
with very high levels of chloride contamination (i.e., greater than
700 mg/L).Indeed, our mostcontaminated pond had a chloride
level of 774 mg/L whichishigher thanmany of the ponds used
in previous studies (Brady, 2013, 2017). Focusing on highly im-
pacted ponds will not only help determine whether our observa-
tion is par t of a broader pattern, but also provide insight into the
mechanisms that might drive higher tolerance to chloride among
wood frog populations.
4.1 | Conclusions
The results of this study demonstrate that populations of amphib-
ians, and perhaps many other taxa, have the abilit y to evolve in-
crease salt tolerance when exposed to high concentrations of salt
in ponds and wetlands. Thus, we may need to focus our research
efforts on populations of from ponds with ver y high levels of chlo-
ridecontamination(i.e.,greater than 700 mg/L).Ifsuchevolution
of increased salt tolerance is common in aquatic taxa, it sug gests
that entire communities may have the ability to evolve higher
tolerance, which would allow these species to persist while we
work to reduce freshwater salinization, particularly through more
targeted applications of road salts that can reduce salt use by as
muchas50percent(Hint zetal.,2022). Future studies should ex-
amine many other populations of amphibians and associated taxa
and also examine whether the tolerance to salinization is constitu-
tive or if it can be rapidly induced, as has been observed in many
pesticides (e.g., Hua et al., 2015).
AUTHOR CONTRIBUTIONS
Rick Relyea: Conceptualization (equal); data curation (equal); formal
analysis (equal); funding acquisition (equal); investigation (equal);
projec t administrat ion (equal); super vision (equal) ; writing – origi-
naldraft(equal);writing–reviewandediting(equal).Brian Mattes:
Investigation (supporting). Candace Schermerhorn: Investigation
(supporting). Iasaac Shepard: Conceptualization (equal); formal anal-
ysis (equal); investigation (lead); methodolog y (equal); supervision
(equal);writing–originaldraft(equal).
ACKNOWLEDGEMENTS
We thank Nate Fahey and McKenna Conners for their assistance
with the experiment. We also thank the editor and anonymous
reviewers for their helpful feedback. Major funding was provided
by NSF grant D EB 16-55168 to RAR a nd Rensselaer Poly technic
Instituteendowedchairfunds toRAR .Thisresearchwas approved
undertheuniversityIACUCprotocolREL0 01-22.
FUNDING INFORMATION
Major funding was provided by NSF grant DEB 16–55168 and
Rensselaer Polytechnic Institute endowed chair funds.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest .
DATA AVAIL AB ILI T Y STAT EME N T
Alldatawillbepubliclyarchiveduponpublicationacceptance.
|
9 of 10
RE LY E A e t al.
REFERENCES
Bates, D., Mächler, M., Bolker, B., & Walker, S. (2015). Fitting linear
mixed-effectsmodelsusinglme4. Journal of Statistical Software, 67,
1–4 8 . https:// doi. org/ 10. 18637/ jss. v067. i01
Berven, K. A.,Gill, D. E., & Smith-Gill,S .J.(1979).Countergradient se-
lection in the green frog , Rana clamitans. Evolution, 33, 609–623.
h t t p s : / / d o i . o r g / 1 0 . 1 1 1 1 / j . 1 5 5 8 - 5 6 4 6 . 1 9 7 9 . t b 0 4 7 1 4 . x
Bickham, J. W. (2011). The four cornerstones of evolutionary toxicol-
og y. Ecotoxicology, 20, 497–502. htt ps:// doi. org/ 10. 10 07/ s1064
6 - 0 1 1 - 0 6 3 6 - y
Bi lle t ,L . S ., Wu e r thn e r,V.P.,H ua, J. ,Re l ye a ,R . A .,& Ho ver m an, J.T.(2 021).
Population-levelvariation in infectionoutcomesnot influencedby
pesticide exposure in larval wood frogs (Rana sylvatica). Freshwater
Biology, 66,1169–1181.ht tp s:// doi. o rg / 10. 1111 / fwb. 1370 8
Blaustein,A.R.,&Bancroft,B.A.(20 07).Amphibianpopulationdeclines:
Evolutionary considerations. Bioscience, 57, 4 37–44 4. ht tp s : // doi.
o r g / 1 0 . 1 6 4 1 / B 5 7 0 5 1 7
Bor a, L. S.,Th om az,S.M.,&Padi al,A .A .(2 020). Evi denceofr apidevol u-
tion of an invasive poaceae in response to s alinit y. Aquatic Sciences,
82, 76. h t t p s : / / d o i . o r g / 1 0 . 1 0 0 7 / s 0 0 0 2 7 - 0 2 0 - 0 0 7 5 0 - y
Boutilier, R. G., Stiffler, D. F., & Toews, D. P. (1992). Exchange of respira-
tory gases, ions and water in amphibious and aquatic a mphibians. In
M. E. Feder & W. W. Burggren (Eds.), Environmental p hysiology of the
amphibians(pp.81–124).TheUniversityofChicagoPre ss.
Brady, S. P. (2012). Road to evolutio n? Local a daptation to roa d adjacency
in an amphibian (Ambystoma maculatum). Scientific Reports, 2,235.
Brady, S. P. (2013). Microgeographic maladaptive performance and deme
depression in response to roads and r unoff. Pee rJ, 1, e163. h tt ps: //
doi. org/ 10. 7717/ peerj. 163
Brady, S. P. (2017). Environmental exposure does not explain putative
maladaptationin road-adjacentpopulations.Oecologia, 184, 9 31–
942. h t t p s : / / d o i . o r g / 1 0 . 1 0 0 7 / s 0 0 4 4 2 - 0 1 7 - 3 9 1 2 - 6
Brady, S. P., & Goedert, D. (2017). Positive sire effects and adaptive gen-
otype by environment inter action occur despite pattern of local
maladaptation in roadside population of an amphibian. Copeia, 105,
533 –54 2.
Brady, S. P., Goed ert, D., Fr ymus, L . E., Za mora-Ca macho, F. J., Smith ,
P.C.,Zeiss,C.J.,Comas,M.,Abbott, T.A.,Basu,S.P.,DeAndressi,
J. C., Forgione, M. E., Maloney, M. J., Priester, J. L., Senturk, F.,
Szeligowski,R. V.,Tucker,A. S., Zhang, M., & Calsbeek,R. (2022).
Salted roads lead to oedema and reduced locomotor function in
amphibian populations. Freshwater Biology, 67,1150–1161. h t t p s : //
doi. org/ 10. 1111/ fwb. 13907
Brady, S. P., Richardson, J. L ., & Kunz, B. K. (2017). Incorp orating evolu-
tionar y insights to improve ecotoxicology for freshwater species.
Evolutionary Applications, 10, 829–838. ht tp s:// doi. o rg / 10. 1111 /
eva.12507
Brady, S. P., Zamor a-Camacho, F. J., Eriksson, F. A. A., Goedert, D.,
Comas, M., & Calsbeek, R. (2019). Fitter frogs from polluted ponds:
Thecompleximpactsofhuman-alteredenvironments.Evolutionary
Applications, 12,1360–1370.h t t p s : / / d o i . o r g / 1 0 . 1 1 1 1 / e v a . 1 2 7 5 1
Bridges , C. M., & Semlit sch, R . D. (200 0). Variation in pesticide tolerance
of tadpoles among and within species of Ranidae and patterns of
amphibian decline. Conservation Biology, 14, 1490 –1499.
Buss,N.,Swierk,L.,&Hua,J.(2021).Amphibianbreedingphenologyin-
fluences offspring size and response to a common wetland contam-
inant. Frontiers in Zoology, 18, 31. https:// doi. org/ 10. 1186/ s1298
3 - 0 2 1- 0 0 4 1 3 - 0
Coldsnow,K.D.,&Relyea,R .A .(2018). Toxicityofvariousroad-deicing
salts t o Asian clam s (Corbicula fluminea). Environmental Toxicology
and Chemistry, 37,1839–1845.https:// doi. o rg/ 10. 1002/ etc. 4126
Corsi, S. R., Gracz yk, D. J., Geis, S. W., Booth, N. L ., & Richards, K . D.
(2010).Afreshlookatroads alt:Aquatictoxicityandwater-quality
impac ts on local, r egional, and nat ional scales . Environmental Science
& Technology, 44,7376–7382.https:// doi. org/ 10. 1021/ es101 3 33u
Dugan , H. A., Ba rtlet t, S. L ., Burke, S . M., Doub ek, J. P.,K rivak-Tetley,
F. E., Skaff, N. K ., Summers, J. C ., Farrell, K. J., McCullough, I. M.,
Morale s-Williams, A. M., Roberts, D. C., Ouyan g, Z., Scordo, F.,
Hanson, P. C., & Weathers, K. C. (2017). Salting our fre shwater
lakes. Proceedings of the National Academy of Sciences, 114, 4453–
4458.https :// doi . org/ 10. 107 3/ pn as. 1620 2 11114
Forgione, M. E., & Brady, S. P. (2022). Road s alt is more toxic to wood frog
embryos from po lluted ponds. Environmental Pollution, 296,118757.
h t t p s : / / d o i . o r g / 1 0 . 1 0 1 6 / j . e n v p o l . 2 0 2 1 . 1 1 8 7 5 7
Fox, J., & Weisberg, S. (2019). An R companion to applied regression. Sage.
Godwin, K. S., Hafner,S . D., & Buff, M. F. (2003). Long-term trends in
sodium and chloride in the Mohawk River, New York: The ef fect
offif ty years ofroad-salt application.Environmental Pollution, 124,
273–2 81 .h t t p s : / / d o i . o r g / 1 0 . 1 0 1 6 / S 0 2 6 9 - 7 4 9 1 ( 0 2 ) 0 0 4 8 1 - 5
Gomez-Mes tre, I., & Tejed o, M. (200 3). Local ad aptatio n of an anuran
amphibian to osmotically stressful environments. Evolution, 57,
1889–1899.
Gosner, K. L. (1960).A simplifiedtable for staginganuran embryos and
larvae with notes on identification. Herpetologica, 16 ,183–190.
Hall, E. M ., Brady, S. P., Mattheus, N. M., Earley, R. L., Diamond, M., &
Crespi, E. J. (2017). Physiological consequences of exposure to sa-
linized roadside ponds on woo d frog lar vae and adult s. Biological
Conservation, 209,98–106.https:// doi. org/ 10. 1016/j. biocon. 2017.
02. 013
Hendr y, A. P. (2017). Eco- evolutionary dynamics. Princeton University
Press. http:// www. jstor. org/ stable/ j. ctt1q 1xr0c
Hill, A . R., & Sadowski , E. K. (2016). Chlori de concentratio ns in wet-
lands along a rural to urban land use gradient. Wetlands, 36,73–83.
h t t p s : / / d o i . o r g / 1 0 . 1 0 0 7 / s 1 3 1 5 7 - 0 1 5 - 0 7 1 7 - 4
Hintz,W.D.,Fay,L.,&Relyea,R.A.(2022).Roadsalt s,humansafety,and
the rising salinity of our fresh waters. Frontiers in Ecology and the
Environment, 20,22–30.http s:// doi. org / 10. 10 02/ fee. 2433
Hintz, W. D., Jones, D. K., & Relyea, R.A. (2019). Evolved toleranceto
freshwater salinization in zooplankton: Life-history trade-offs,
cross-tolerance and reducing c ascading effects. Philosophical
Transactions of the Royal Society, B: Biological Scie nces, 374,
20180012. https:// doi. org/ 10. 1098/ rstb. 2018. 0 012
Hintz,W.D.,&Relyea,R .A.(2019).Areviewofthespecies,community,
and ecosystem impacts of road salt salinisation in fresh waters.
Freshwater Biology, 64, 1081–1097. http s:// doi. o rg / 10. 1111/ f wb.
13286
Hopkins,G .R ., &Brodie, E. D.(2015). Occurrence ofamphibiansinsa-
linehabitats:Areviewandevolutionaryperspective.Herpetological
Monographs, 29, 1–27. h t t p s : / / d o i . o r g / 1 0 . 1 6 5 5 / H E R P M O N O G R
A P H S - D - 1 4 - 0 0 0 0 6
Hua, J., Jo nes, D. K., Matt es, B. M., Cothr an, R. D., Relyea, R . A., &
Hoverma n, J. T.(2 015). The cont ributio n of phenot ypic pla sticit y
to the evolution of insecticide tolerance in amphibian populations.
Evolutionary Applications, 8,586–596.
Hua, J., Wuerthner, V. P., Jones, D. K ., Mattes, B., Cothran, R. D., Relyea,
R. A. , & Hoverman, J. T. (2017). Evolved p esticide tole rance in-
fluences susceptibilit y to parasites in amphibians. Evolutio nary
Applications, 10,802–812.h t t p s : / / d o i . o r g / 1 0 . 1 1 1 1 / e v a . 1 2 5 0 0
Jackson,R.B., &Jobbágy,E.G. (2005).Fromicyroadstosaltystreams.
Proceedi ngs of the National Academy of Sciences of the United States
of America, 102, 14 487–14488 . https:// doi. org/ 10. 1073/ pnas.
0 5 0 7 3 8 9 1 0 2
Karraker, N. E., Gibbs, J. P., & Vonesh, J. R. (2008). Impac ts of road de-
icing salt on the demographyof vernalpool-breeding amphibians.
Ecological Applications, 18, 724–734. h t t p s : / / d o i . o r g / 1 0 . 1 8 9 0 / 0 7 -
164 4. 1
Kassambara,A.(2021).Sur vminer.
Kearney,B.D.,Pell,R.J.,Byrne,P.G.,&Reina,R .D.(2014).Anuranlar-
val developmental plasticity and sur vival in response to variable
salinity of ecologically relevant timing and magnitude. Journal of
10 of 10
|
RELY E A e t al.
Experimental Zoology Par t A: Ecological Gen etics an d Physiology, 321,
541–549.https:// doi. org/ 10. 1002/ jez. 1887
Kinnison, M. T., & Hairston, N. G. Jr. (2007). Eco-evolutionary biol-
ogy: Contemporary evolution and the dynamic s of persistence.
Functional Ecology, 21,44 4–454.
Milotic, D., Milotic, M., & Koprivnikar, J. (2017). Effects of road salt on
larval amphibian susceptibility to parasitism through behavior and
immunocompetence. Aquatic Toxicology, 189, 42–49. ht t p s : // d o i .
o r g / 1 0 . 1 0 1 6 / j . a q u a t o x . 2 0 1 7 . 0 5 . 0 1 5
Petranka,J.W.,&Francis,R.A.(2013).Effectsofroadsaltsonseasonal
wetlands: Poor prey performance may compromise the growth of
predatory salamanders. Wetlands, 33,707–715.
Rossiter, M. (1996). Incidence and consequences of inherited envi-
ronment al effects. Annua l Review of Ecology and Systematic s, 27,
451– 476.
Sanzo, D.,& Hecnar,S. J. (2006).Effect sof roadde-icingsalt(NaCl) on
larva l wood frogs (Rana sylvatica). Environmental Pollution, 140,247–
256.h t t p s : / / d o i . o r g / 1 0 . 1 0 1 6 / j . e n v p o l . 2 0 0 5 . 0 7 . 0 1 3
Schluter, D., Price, T. D., Rowe, L., & Grant , P. R. (1991). Conflicting selec-
tion pressures andlife history tr ade-offs. Proceedings of the Royal
Society of London. Series B: Biological Sciences, 246 , 11–17.h tt p s : //
doi. org/ 10. 1098/ rspb. 1991. 0118
Schuler,M.S.,Cañedo-Argüelles,M., Hintz,W.D., Dyack,B.,Birk,S.,&
Relyea, R.A .(2019).Regulationsareneededtoprotectfreshwater
ecosystems from salinization. Philosophical Transac tions of the Royal
Society, B: Biological Sciences, 374 , 20180 019. https:// doi. org/ 10.
1098/ rstb. 2018. 0019
Semlit sch,R.D.,Bridges,C.M.,&Welch,A .M.(200 0).Geneticvariation
and a fitness tradeoff in the tolerance of gray treefrog (Hyla versi-
color) tadpoles to the insecticide carbar yl. Oecologia, 125,1 7 9–1 8 5 .
Squires, Z. E., Bailey, P. C. E., Reina, R. D., & Wong, B. B. B. M. (2010).
Compensatory growth in tadpoles af ter transient salinity stress.
Marine and Freshwater Research, 61, 219–222. https:// doi. org/ 10.
1071/ MF09123
Stockwell,C .A .,Hendr y,A.P.,&Kinnison,M.T.(20 03).Contemporar y
evolution meets conservation biology. Trends in Ecology & Evolution,
18,94–101.h t t p s : / / d o i . o r g / 1 0 . 1 0 1 6 / S 0 1 6 9 - 5 3 4 7 ( 0 2 ) 0 0 0 4 4 - 7
Stuart,S.N., Chanson,J. S.,Cox,N.A .,Young,B.E.,Rodrigues,A .S.L.,
Fischma n, D. L., & Waller, R. W. (20 04). Status and t rends of amphib -
ian declines and extinctions worldwide. Science, 306,1783–1786.
Therneau, T. (2020). Coxme: Mixed effects cox models.
Therneau,T.(2021).Apack ageforsurvivalanalysisinR.
Venäläine n, A. (20 01). Estima tion of road s alt use base d on winter air
temperature. Meteorological Applications, 8, 333–338. h t t p s : // d o i .
o r g / 1 0 . 1 0 1 7 / S 1 3 5 0 4 8 2 7 0 1 0 0 3 0 8 5
SUPPORTING INFORMATION
Additional supporting information can be found online in the
Suppor ting Information section at the end of this article.
How to cite this article: Relyea, R., Mattes, B., Schermerhorn,
C., & Shepard, I. (2024). Freshwater salinization and the
evolved tolerance of amphibians. Ecology and Evolution, 14,
e11069. https://doi.org/10.1002/ece3.11069