Research Report

Fern Stomatal Responses to ABA and CO2 Depend onSpecies and Growth Conditions1[OPEN]

Hanna Hõrak, Hannes Kollist, and Ebe Merilo*

Plant Signal Research Group, University of Tartu, Institute of Technology, Tartu 50411, Estonia

ORCID IDs: 0000-0002-6392-859X (H.H.); 0000-0002-6895-3583 (H.K.); 0000-0003-1767-6413 (E.M.).

Changing atmospheric CO2 levels, climate, and air humidity affect plant gas exchange that is controlled by stomata, small poreson plant leaves and stems formed by guard cells. Evolution has shaped the morphology and regulatory mechanisms governingstomatal movements to correspond to the needs of various land plant groups over the past 400 million years. Stomata close inresponse to the plant hormone abscisic acid (ABA), elevated CO2 concentration, and reduced air humidity. Whether the activeregulatory mechanisms that control stomatal closure in response to these stimuli are present already in mosses, the oldest plantgroup with stomata, or were acquired more recently in angiosperms remains controversial. It has been suggested that thestomata of the basal vascular plants, such as ferns and lycophytes, close solely hydropassively. On the other hand, activestomatal closure in response to ABA and CO2 was found in several moss, lycophyte, and fern species. Here, we show thatthe stomata of two temperate fern species respond to ABA and CO2 and that an active mechanism of stomatal regulation inresponse to reduced air humidity is present in some ferns. Importantly, fern stomatal responses depend on growth conditions.The data indicate that the stomatal behavior of ferns is more complex than anticipated before, and active stomatal regulation ispresent in some ferns and has possibly been lost in others. Further analysis that takes into account fern species, life history,evolutionary age, and growth conditions is required to gain insight into the evolution of land plant stomatal responses.

Stomatal pores, formed by guard cells on plant leavesand stems, mediate CO2 uptake for photosynthesis andwater loss via transpiration. Adequate regulation of thestomatal aperture in response to changing environ-mental conditions is essential for thriving plant growth.In angiosperms, whose stomatal regulation has beenstudied the most, stomata close in response to abscisicacid (ABA), elevated CO2 concentration, reduced airhumidity, darkness, and air pollutants, whereas theyopen in response to light, increased air humidity, anddecreased CO2 concentration.

In recent years, the evolution of plant stomatal sig-naling pathways has become a subject of intensivestudy and passionate debate. The central role of ABA inangiosperm stomatal responses has been known for along time (Cutler et al., 2010). More recently, experi-ments assessing stomatal responses to ABA have beenconducted with mosses and basal vascular plants suchas ferns and lycophytes. Lack of stomatal closure in

response to ABA treatment in several fern and lyco-phyte species led to the hypothesis that these plantgroups use only hydropassivemechanisms for stomatalregulation (Brodribb and McAdam, 2011). Moreover,high ABA levels induced in response to drought didnot inhibit the opening of fern and lycophyte stomataupon rehydration, suggesting that endogenous ABAdid not control stomatal responses in these plant spe-cies (McAdam and Brodribb, 2012).

On the other hand, stomata in the epidermal strips ofthe lycophyteSelaginella uncinata and in the sporophytes ofthe mosses Physcomitrella patens and Funaria hygrometricaclosed in response to ABA in a concentration-dependentmanner (Chater et al., 2011; Ruszala et al., 2011), indicat-ing a conserved role for ABA in the stomatal responsesof plants. Recently, dose-dependent ABA-induced sto-matal closure also was shown to be present in the fernsPolystichum proliferum and Nephrolepis exaltata (Cai et al.,2017). Several genes encoding proteins involved in ABAsignal transduction are expressed in the stomata-bearingsporophyte of themossP. patens (O’Donoghue et al., 2013)and in the epidermal fraction of the fern P. proliferum(Cai et al., 2017). Furthermore, the P. patens and Selaginellamoellendorffii homologs of OPEN STOMATA1 (OST1), aSnRK-type kinase that participates in ABA-induced sto-matal closure via phosphorylation of the central guardcell anion channel SLOW ANION CHANNEL1 (SLAC1)inArabidopsis (Arabidopsis thaliana; Geiger et al., 2009; Leeet al., 2009; Vahisalu et al., 2010), complemented the ABAinsensitivity of stomatal closure in the Arabidopsis ost1mutant (Chater et al., 2011; Ruszala et al., 2011). TheP. patens deletion mutant that lacked OST1 had impairedABA-induced stomatal closure (Chater et al., 2011), and

1 This work was supported by the Estonian Ministry of Scienceand Education (grant no. PUT1133 to E.M. and grant no. IUT2-21to H.K.) and by the European Regional Development Fund (Centerof Excellence in Molecular Cell Engineering).

* Address correspondence to ebe.merilo@ut.ee.The author responsible for distribution of materials integral to the

findings presented in this article in accordance with the policy de-scribed in the Instructions for Authors (www.plantphysiol.org) is:Ebe Merilo (ebe.merilo@ut.ee).

E.M. and H.H. conceived and designed the study; E.M. performedmost of the experiments; E.M. analyzed the data; H.H. wrote thearticle with E.M. and H.K.

[OPEN] Articles can be viewed without a subscription.www.plantphysiol.org/cgi/doi/10.1104/pp.17.00120

672 Plant Physiology�, June 2017, Vol. 174, pp. 672–679, www.plantphysiol.org � 2017 American Society of Plant Biologists. All Rights Reserved. www.plantphysiol.orgon March 12, 2020 - Published by Downloaded from

Copyright © 2017 American Society of Plant Biologists. All rights reserved.

P. patens OST1 and several other OST1-like SnRKs fromdifferent nonvascular plants could activate ArabidopsisSLAC1 in oocytes (Lind et al., 2015). These data indicatethat the core stomatal ABA-signaling pathway isconserved in plants. Recently, a homolog of OST1 wasshown to regulate sex determination in the fern Cera-topteris richardii (McAdam et al., 2016), suggesting thatat least some of the components of the ABA-signalingpathway that control stomatal responses in angio-sperms could have different or additional functions inferns. Moreover, the ABA signal transduction path-ways of ferns and angiosperms may be at least partlydifferent; thus, the analysis of fern homologs ofcomponents involved in ABA-response pathways ofangiosperms is not sufficient to fully understand themechanisms of ABA responsiveness in ferns.Plant stomata open or close in response to subambient

or above-ambient CO2 concentration, respectively. Thisis the basic mechanism through which the supply ofphotosynthetic CO2 is regulated. The question of stomatalresponsiveness to CO2 in plant groups older than angio-sperms remains controversial. Gas-exchange analysisof several fern and lycophyte species indicated thatthe stomata of these plants open in response to sub-ambient CO2 concentrations but do not close at above-ambient CO2 levels (Brodribb et al., 2009; Brodribb andMcAdam, 2013). Similarly, stomata in the mossesP. patens and F. hygrometrica had strong CO2 responsesin the range of 0 to 400 ppm CO2 but did not closemarkedly at above-ambient CO2 levels (Chater et al.,2011). However, stomata of the lycophyte S. uncinataresponded to both subambient and above-ambient CO2concentrations (Ruszala et al., 2011), as did the sto-mata of the fern Phyllitis scolopendrium (Mansfield andWillmer, 1969). Recently, several fern species were shownto close their stomata in response to CO2 elevation fromthe ambient 400 to 800 ppm, similar to the angiospermArabidopsis (Franks and Britton-Harper, 2016). It washypothesized that different growth conditions andspecies-specific behaviors may account for the variedresults obtained in the analysis of stomatal responses ofmosses, ferns, and lycophytes (Roelfsema and Hedrich,2016). For example, in some fern species, stomata closedin response to CO2 elevation from 0 to 400 ppm, in othersthey did not respond or even opened, whereas in somespecies the response depended on growth light intensity(Creese et al., 2014). Thus, a consensus on whether thereis a universal mechanism that governs stomatal re-sponses to CO2 and ABA in all land plants has not beenreached to date, and further experiments assessing theeffect of growth conditions and species are required.Importantly, the above-mentioned studies paid little at-tention to the fact that fern stomata are morphologicallydifferent from those of angiosperms: they lack subsidi-ary cells, and there is little mechanical interaction be-tween guard cells and epidermal cells during guard cellswelling (i.e. diminished mechanical advantage of epi-dermal cells over guard cells; Franks and Farquhar,2007). Large kidney-shaped stomata of ferns that growin the forest understory characterized by deep shade and

more humid air have been suggested to have slowdynamics of stomatal movements (Hetherington andWoodward, 2003). These differences in guard cell mor-phological and mechanical properties undoubtedlytranslate into differences in stomatal functioning.

Here, we provide evidence of growth condition- andspecies-dependent stomatal responsiveness to ABAand CO2 in three temperate fern species. The resultsindicate that, while the stomata of some ferns can re-spond to ABA and CO2, the kinetics and, thus, the un-derlying mechanisms of stomatal regulation differ, atleast partly, between ferns and angiosperms. Further-more, the stomatal response to decreased air humidityis slowed down at low CO2 concentration in Athyriumfilix-femina, indicating that fern stomatal responses arenot purely hydropassive. Thus, species and growthconditions affect fern stomatal responsiveness andshould be taken into account when drawing majorconclusions about the evolution of the mechanismsthat control stomatal regulation in ferns.

RESULTS AND DISCUSSION

The Stomatal Response to ABA Is Species Specific andDepends on Growth Conditions in Ferns

As the presence of active stomatal responses in fernshas remained controversial to date, we studied whethertheir stomatal responses to ABA and CO2 could beinfluenced by growth conditions. Three temperate fernspecies (A. filix-femina, Dryopteris carthusiana, and Dry-opteris filix-mas) were collected from nature and trans-ferred to the laboratory, where they were grown in twodifferent growth conditions: in a growth cabinet char-acterized by constant relative air humidity of 70% bothday and night (from here on referred to as the growthcabinet) or in a growth room with lower and variablerelative air humidity (RH; ;30%–40% during the dayand;40%–65% at night; from here on referred to as thegrowth room). The light regimewas 16/8 h and 12/12 hlight/dark in the growth room and the growth cabinet,respectively. After the ferns had developed new leavesindoors, gas-exchange analysis was carried out with acustom-built device that enabled parallel recording ofstomatal conductance in four plants (see “Materials andMethods”; one measurement cuvette is shown in Fig.1A). Similar experiments were carried out with theangiosperm Arabidopsis (Columbia-0) grown in iden-tical conditions and using another custom-built devicethat enabled parallel recording of stomatal conductancein eight plants (see “Materials andMethods”). Stomatalresponses of the ferns and Arabidopsis to 10 mM ABAsprayed on leaves or changes in CO2 concentrationwere measured.

The stomatal conductance of the angiosperm Arabi-dopsis decreased strongly in response to 10 mM ABAapplication in plants grown both in the growth cabinetas well as in the growth room (Fig. 1B), pointing toprominent ABA responsiveness of Arabidopsis stomatairrespective of growth conditions. However, the growth

Plant Physiol. Vol. 174, 2017 673

Growth Conditions Affect Fern Stomatal Responses

www.plantphysiol.orgon March 12, 2020 - Published by Downloaded from Copyright © 2017 American Society of Plant Biologists. All rights reserved.

conditions affected stomatal responsiveness in ferns.When grown in the growth cabinet at 70% RH, thestomatal conductance of A. filix-femina and D. filix-masdecreased in response to spraying with ABA, whereasD. carthusiana did not respond to ABA treatment (Fig.1C). None of the tested fern species responded to ABAwhen grown in the growth room at lower RH (Fig. 1C).Thus, fern species differ in their ability to close stomatain response to ABA, and stomatal responsiveness isaffected by growth conditions.

Air humidity has been shown to affect stomatalmorphology and responsiveness in various angiospermspecies. High relative air humidity during growth hasbeen associated with longer and more open stomata(Torre et al., 2003; Nejad and van Meeteren, 2005;Fanourakis et al., 2011; Aliniaeifard et al., 2014), lowerlevels of ABA (Zeevaart, 1974; Nejad and vanMeeteren,2007; Okamoto et al., 2009; Aliniaeifard et al., 2014), anddecreased stomatal responsiveness to ABA (Pospíšilová,1996; Fanourakis et al., 2013; Pantin et al., 2013b; Arveet al., 2014). Exposure to low RH can enhance stomatalABA responsiveness in young leaves that are constantlyin a microenvironment with high RH (Pantin et al.,2013b). This indicates that reduced RH is required toprime stomatal ABA responses, potentially via increasedABA levels, as ABA application can restore stomatalABA responsiveness (Pantin et al., 2013b) and results innormal stomatal development (Fanourakis et al., 2011) inplants grown at high RH. However, in these reports, theRH that led to reduced ABA sensitivity was very high:90% or more. In our experiment, lower growth RH(30%–65%) led to the loss of stomatal ABA respon-siveness in two ferns compared with growing them at

70% RH (Fig. 1C). This could be explained by in-creased ABA levels induced by low air humidity, asobserved in other species (Bauer et al., 2013; McAdamand Brodribb, 2015), which could potentially lead topartial ABA-induced closure of the stomata of fernsgrown in the growth room and no further effect ofABA treatment on stomatal conductance. If the ABAresponse is already activated in ferns grown at lowerRH, these plants would be expected to have lowerstomatal conductance. However, as the magnitude ofABA-induced stomatal closure was modest even inthe ferns grown at 70% RH, such a difference in sto-matal conductance could remain undetected. It hasbeen shown that the ABA levels of ferns do not in-crease in response to short-term low-humidity treat-ment (McAdam and Brodribb, 2015), but ABA levelsin ferns subjected to long-term growth at low RHremain to be characterized.

The guard cell ABA response appears to be present insome fern species and not in others (Fig. 1C; BrodribbandMcAdam, 2011; Cai et al., 2017), suggesting that theABA responsiveness may have been lost in some line-ages of ferns. In addition to the direct effect on stomata,ABA reduces leaf hydraulic conductance in angio-sperms (Shatil-Cohen et al., 2011; Pantin et al., 2013a).Thus, the ABA-induced decrease in stomatal conduc-tance inA. filix-femina andD. filix-mas grown at 70% RHcould have occurred due to changes in leaf hydraulicsand not due to the direct effect of ABA on guard cells.Different procedures for ABA application could explainthe varying ABA responsiveness of ferns and lycophytesfound in experiments. It is possible that ABA applied viathe transpiration stream is incapable of inducing stomatal

Figure 1. Stomatal responses of ferns and Arabidopsis grown in two different conditions to foliar ABA spraying (10 mM). A,D. filix-mas in the gas-exchange measurement chamber. B, Mean6 SE stomatal conductance before and 47 min after treatmentwith ABA in Arabidopsis grown in the growth room and in the growth cabinet (n = 7). C, Mean 6 SE stomatal conductancebefore and 47 min after treatment with ABA in the ferns A. filix-femina, D. filix-mas, and D. carthusiana grown in differentconditions (n = 7 or 9 for A. filix-femina, n = 4 or 7 forD. filix-mas, and n = 5 or 4 forD. carthusiana, grown in the growth roomor the growth cabinet, respectively). Statistically significant differences are denoted with asterisks (repeated-measures ANOVAwith Tukey’s posthoc test).

674 Plant Physiol. Vol. 174, 2017

Hõrak et al.

www.plantphysiol.orgon March 12, 2020 - Published by Downloaded from Copyright © 2017 American Society of Plant Biologists. All rights reserved.

closure in ferns and lycophytes, whereas ABA applied toepidermal peels or by leaf spraying enters guard cells di-rectly and, thus, is more efficient in promoting stomatalclosure (Fig. 1C; Brodribb and McAdam, 2011; Ruszalaet al., 2011; Cai et al., 2017).Although the stomata of two fern species closed in

response to ABA, the extent of the closure was smallcompared with the angiosperm Arabidopsis (Fig. 1, Band C). This indicates that, while several ferns are ca-pable of ABA perception and response, the kinetics andunderlying molecular mechanisms of this responsecould be different in ferns and in angiosperms. Thecomponents of the core stomatal ABA signaling path-way have been shown to be present in mosses, lyco-phytes, ferns, and angiosperms (Hanada et al., 2011; Caiet al., 2017), indicating an early origin of ABA signaltransduction. Nevertheless, it is possible that not all ofthese proteins have strictly stomata-related functions inferns, as was shown recently for a homolog of the OST1kinase that participates in sex determination in a fern(McAdam et al., 2016). However, other kinases besidesOST1 can activate anion channels toward stomatalclosure, and although OST1 is the most importantamong them in Arabidopsis (Kollist et al., 2014), ABA-induced signaling pathways do not have to be identicalin ferns and angiosperms. Thus, it is possible that theclassical stomatal ABA-response pathway functionsonly partially in ferns and is not capable of inducingstomatal closure in response to ABA treatment in allfern species. A wider study of stomatal ABA respon-siveness across the evolutionary tree of ferns wouldbe required to understand how guard cell ABA re-sponsiveness has evolved. For example, an analysis ofdifferent fern classes showed that, whereas stomatalopening in response to blue light is present in lyco-phytes and most classes of ferns, it has been lost in thefern group Polypodiopsida (Doi et al., 2015). Thus, it isconceivable that the stomatal ABA responsiveness hasbeen lost in some branches of the evolutionary tree offerns, and systematic analysis of stomatal responses indifferent taxa is required to bring further insight to thisquestion.

Stomatal Responses to Subambient and Above-AmbientCO2 Concentrations Differ and Depend on GrowthConditions in Ferns

Next, we studied stomatal responses to subambientand above-ambient CO2 concentrations in Arabidopsisand the three fern species. In Arabidopsis, strong andsteady stomatal opening occurred in response to thereduction of CO2 concentration from 400 to 100 ppm,and stomata closed quickly when CO2 was returned to400 ppm (Fig. 2, A and B; Supplemental Table S1). Faststomatal closure also occurred in response to furtherelevation of CO2 concentration from 400 to 800 ppm(Fig. 2, A and B; Supplemental Table S1). Stomatal re-sponsiveness to changes in CO2 levels did not dependmarkedly on growth conditions in Arabidopsis.

Fast stomatal opening in response to a reduction ofCO2 concentration from 400 to 100 ppm occurred in allferns irrespective of growth conditions (Fig. 2, C and D;Supplemental Tables S2 and S3). Elevation of CO2 from100 ppm back to 400 ppm and then further to 800 ppmcaused stomatal closure in nearly all tested fern speciesand conditions (Fig. 2, C andD; Supplemental Tables S2and S3). There was one exception: D. filix-mas grown inthe growth room did not respond significantly to above-ambient levels of CO2, whereas its stomatal opening inresponse to a change from 400 to 100 ppm and closureinduced by CO2 elevation from 100 to 400 ppm was oneof the largest (Fig. 2, C and D; Supplemental Tables S2and S3). Previously, high RHhas been shown to enhancethe response to CO2 elevation in Vicia faba (Talbott et al.,2003). In accordancewith this, stomatal responses toCO2elevation were reduced at least partially in ferns grownat lower RH. In addition toD. filix-mas, the lowRHof thegrowth roomalso affected CO2 responsiveness inA. filix-femina, whose final stomatal conductance achieved afterthe transition from 100 to 400 ppm was significantlyhigher than the pretreatment value at 400 ppm, indi-cating a slowed closure response (Fig. 2D; SupplementalTable S2). Thus, as in experiments with ABA, growth inthe growth room with lower RH tended to reduce thestomatal sensitivity in D. filix-mas and A. filix-femina. Asstomatal responses to CO2 are tightly coupled with ABAresponses (Xue et al., 2011; Merilo et al., 2013; Chateret al., 2015), the analogous dampening of responsesto ABA and elevated CO2 found in A. filix-femina andD. filix-mas when grown at low RH suggests that theunderlying mechanisms responsible for the effect ofgrowth conditions could be similar and potentiallyrelated to ABA levels.

All the fern species studied here had the ability toopen stomata due to CO2withdrawal and close stomatadue to CO2 enrichment, both in the subambient andabove-ambient range of CO2 levels. However, theirstomatal opening in response to lowCO2wasmarkedlyfaster than stomatal closure in response to elevated CO2(Fig. 2C). Thus, the previously documented presenceand absence of stomatal responses to subambientand above-ambient CO2 concentration, respectively(Brodribb et al., 2009; Brodribb and McAdam, 2013),could be explained by the relatively slow response toelevated CO2 in ferns. The closure kinetics of the ele-vated CO2 response was clearly different betweenthe studied ferns and the angiosperm Arabidopsis;the latter showed a fast drop in stomatal conductancein response to the transition from 400 to 800 ppm com-pared with the steady decrease in ferns, whereas, in ferns,the opening under subambient CO2 levelswas faster thanin Arabidopsis (Fig. 2, A and C). A similar difference inthe kinetics of high CO2-induced stomatal closure inferns and Arabidopsis has been found before (Franksand Britton-Harper, 2016). These differences in the ki-netics of CO2 responses between ferns and angiospermsmay result from their different stomatal morphologyand related mechanical properties. Ferns have lowerstomatal conductance and a diminished mechanical

Plant Physiol. Vol. 174, 2017 675

Growth Conditions Affect Fern Stomatal Responses

www.plantphysiol.orgon March 12, 2020 - Published by Downloaded from Copyright © 2017 American Society of Plant Biologists. All rights reserved.

advantage of epidermal cells over guard cells (Franksand Farquhar, 2007) that could result in faster openingof their stomata. However, ferns also show reducedrates of change in stomatal aperture compared withangiosperms due to less lateral movement of guardcells and reduced shuttling of osmotica between guard

and epidermal cells, which could explain their slowerclosure responses (Franks and Farquhar, 2007).

Interestingly, the pretreatment stomatal conductanceof Arabidopsis plants grown at lower humidity in thegrowth room was higher than that in the plants grownin the growth cabinet in experiments with both ABA

Figure 2. Stomatal responses of ferns and Arabidopsis grown in different conditions to changes in CO2 concentration. A, Time-resolved stomatal conductance of the angiosperm Arabidopsis grown in different conditions. At time point 0 min, CO2 wasdecreased from 400 to 100 ppm; at time point 120 min, it was elevated back to 400 ppm; and at time point 240 min, it wasincreased further to 800 ppm.Mean6 SE is shown (n = 8). B, Last stomatal conductance values fromA at each CO2 concentration.Mean6 SE is shown. C, Time-resolved stomatal conductance of the fernsA. filix-femina,D. filix-mas, andD. carthusiana grown indifferent conditions. CO2 treatments were applied as described above for Arabidopsis. Mean6 SE is shown (n = 7 or 6 for A. filix-femina, n= 4 or 8 forD. filix-mas, and n= 6 or 6 forD. carthusiana, grown in the growth roomor the growth cabinet, respectively).D, Last stomatal conductance values at each CO2 concentration from C. Mean 6 SE is shown.

676 Plant Physiol. Vol. 174, 2017

Hõrak et al.

www.plantphysiol.orgon March 12, 2020 - Published by Downloaded from Copyright © 2017 American Society of Plant Biologists. All rights reserved.

and CO2 (Figs. 1B and 2, A and B; Students t test, P ,0.05). This seems counterintuitive, as previous studieshave shown higher stomatal conductance or widerstomatal aperture at higher RH in several angiospermspecies (Torre et al., 2003; Nejad and van Meeteren,2005; Fanourakis et al., 2011; Aliniaeifard et al., 2014).However, here, the measurements of stomatal behaviorwere carried out at ;70% RH, which was higher thanthe growth room RH of ;40%. Thus, it is possible that,in Arabidopsis, stomatal opening occurredwhen plantswere transferred from the growth room to the mea-surement chamber. This fast opening response was notpresent in any of the tested fern species, where thestomatal conductance of plants grown in the growthroomwas similar to that in the growth cabinet (Figs. 1Band 2, C and D; two-way ANOVA with Tukey’s post-hoc test where appropriate; Supplemental Tables S4and S5). These data suggest a greater flexibility andfaster stomatal responsiveness to changes in vaporpressure deficit (VPD) in angiosperms. This could beexpected, as the responses to changes in VPD are linkedwith the biosynthesis and presence of ABA (Bauer et al.,2013; McAdam and Brodribb, 2015) as well as ABAsignaling (Merilo et al., 2013), and stomatal ABA re-sponses in angiosperms are faster and stronger than inferns (Brodribb and McAdam, 2011; Fig. 1, B and C).

Dampening of the Response to High VPD at Low CO2Concentration Suggests the Presence of Active StomatalRegulation in Ferns

The stomatal responses of ferns have been proposedto be regulated solely hydropassively by leaf waterstatus (Brodribb and McAdam, 2011). If fern stomataindeed act as purely passive valves, itwould be expected

that the stomatal response to increased VPD would befaster in plants with higher stomatal conductance, as thiswould result in higher transpiration, greater loss of wa-ter, and more negative leaf water potential. To test this,we analyzed stomatal responsiveness to reduction in RHcorresponding to higher VPD in either low (50 ppm) orambient (400 ppm) CO2 concentration in ferns grown inthe growth room and the growth cabinet. As the stomataof A. filix-femina responded to ABA and those of D.carthusiana did not, we chose these two species for theexperiments. As expected, the stomatal conductancewas higher in low CO2 concentration in both species,albeit the difference was greater for A. filix-femina (Fig.3A; Supplemental Fig. S1A).

Stomata ofA. filix-femina closed in response to a rapidincrease in VPD from 0.7 to 1 kPa to 2 to 2.2 kPa in bothlow and ambient CO2 concentrations (Fig. 3). However,the response to high VPD was significantly slower in lowthan in ambient CO2 (Fig. 3C). This was true for fernsgrown both in the growth room as well as in the growthcabinet, indicating the robustness of the response. Thisfinding is not in accordance with the model of passiveVPD-dependent stomatal regulation in ferns, since sto-matal conductance was roughly 2 times higher at lowCO2 and this should have caused a faster hydro-passive response to VPD. This experiment indicatesthat there is more than just a passive mechanism thatregulates the stomatal response to a change in VPDin this fern species. As ABA and the components of itssignaling pathway mediate stomatal responses tochanges in VPD in angiosperms (Bauer et al., 2013;Merilo et al., 2013; McAdam and Brodribb, 2015) and A.filix-femina is a fern with ABA-responsive stomata (Fig.1C), ABA is an obvious candidate as a mediator of sto-matal response to high VPD in A. filix-femina. Stomatal

Figure 3. Stomatal response of A. filix-femina to reduced air humidity at different CO2 concentrations. A, Time-resolved stomatalconductance ofA. filix-femina. At time point 0min, VPDwas increased from 0.7 to 1 kPa to 2 to 2.2 kPa. Mean6 SE is shown (n = 7 or8 for the growth roomor the growth cabinet, respectively). B, Stomatal response in relative units, calculated basedon thedata presentedin A. C, Stomatal half-response times, calculated based on the data presented in A. Different letters denote statistically significantdifferences (two-way ANOVAwith Tukey’s posthoc test; the effect of CO2 concentration was statistically significant).

Plant Physiol. Vol. 174, 2017 677

Growth Conditions Affect Fern Stomatal Responses

www.plantphysiol.orgon March 12, 2020 - Published by Downloaded from Copyright © 2017 American Society of Plant Biologists. All rights reserved.

closure in response to increased VPD appeared to bedelayed in low CO2 concentration also in D. carthusiana(Supplemental Fig. S1) grown in both the growth roomand the growth cabinet, but there was no statisticallysignificant difference between stomatal half-responsetimes to increased VPD at low and ambient CO2 con-centrations. Thus, it is possible that, in D. carthusiana,hydropassive stomatal control ismore important in VPDresponses.

Interestingly, despite the low stomatal conductance offerns, their net CO2 assimilation rate ranged between 4.1and 4.5 mmol m22 s21 compared with 5.5 mmol m22 s21

in Arabidopsis. This resulted in twice as high instanta-neous water use efficiency of ferns in comparison withArabidopsis. As ferns naturally grow in the shadedconditions of the forest understory, where CO2 con-centration is higher than in upper layers (Bazzaz andWilliams, 1991), it is possible that the low stomatalconductance of ferns is an adaptive trait associated withgrowth habitat. Adaptation to growth in shade and in-creased CO2 concentration also could explain the strongstomatal sensitivity to CO2 withdrawal in ferns.

CONCLUSION

The results presented here indicate that, while thestomatal responses of some ferns may be hydropassive,as described before (Brodribb and McAdam, 2011),other fern species have active mechanisms for stomatalregulation as well. Large-scale studies of stomatal re-sponsiveness in ferns from different families and evo-lutionary agewould be necessary to gain further insightinto the evolution of passive and active stomatal controlmechanisms in ferns. Moreover, as shown here, growthconditions can have strong effects on stomatal respon-siveness in ferns; for example, responses to ABA andCO2wereweakened in ferns grown at RH of 30% to 40%compared with 70% RH, while fast stomatal responseswere maintained in the angiosperm Arabidopsis grownin both conditions. Whether contrasting patterns of sto-matal ABA sensitivity detected for ferns in this study areaccompanied by differences inABA synthesis in variableconditions remains to be studied in the future.

MATERIALS AND METHODS

Plant Material and Growth Conditions

The ferns Athyrium filix-femina, Dryopteris carthusiana, and Dryopteris filix-maswere used in the experiments. Specimens of D. filix-mas were collected from wildgardens in Kooraste and Näräpää in Põlvamaa, southern Estonia, in August 2016.Specimens of D. carthusianawere collected from a clear-cut forest area in Kooraste,and specimens of A. filix-femina were collected from a forest in Näräpää betweenAugust and October 2016. All the ferns were transported to the laboratory, trans-planted into a 4:2:3 peat:vermiculite:water mixture, and grown either in a growthroom at 16/8 h light/dark, low RH (30%–40% during the day and 40%–65% atnight), and 100 to 150mmolm22 s21 light or in a growth cabinet (MCA1600; SnijdersScientific) at 12/12 h light/dark, 70% RH, and 100 mmol m22 s21 light. Thesegrowth conditions are referred to in this article as the growth room and the growthcabinet, respectively. Only leaves developed indoors in the respective conditionswere used for gas-exchange measurements. Arabidopsis (Arabidopsis thaliana)

Columbia-0 plants used for comparison were grown in identical conditions andwere 23 to 27 d old during the experiments.

Gas-Exchange Experiments

The stomatal conductance of ferns was recorded with a thermostated four-chamber, custom-built,flow-throughgas-exchangedevice.Themainbodyof thesystem consists of four thermostated gas-exchange cuvettes formed by twoglasscylinders (i.d. 10.6 cm, height 15.6 cm) and a thermostated water jacket betweenthem.The cuvettes areplacedona stand composedof twowell-fitted glassplatesthat form a bottom for the gas-exchange cuvette. One of these plates containsperforations for the plant stem and is removable. The other glass plate containsgas input and output ports, a temperature sensor, and a fan to guarantee highturbulence and uniform gas mixing. The fan ensured that the boundary layerconductance was high (11–12 cm s21) and, further, that heat exchange betweenleaves and the chamber air was very good, reducing the effect of transpirationon leaf temperature. Modeling gum was used to ensure an air-tight separationof plant shoots within the cuvette from roots in the soil. The chambers arehermetically sealed and operate under slight overpressure of a few millibars toavoid uncontrolled intake of ambient air. Air flow rate through the chamberwas 2.5 Lmin21. Ambient air passing through a large buffer volume of 25 Lwasused. The air temperature inside the chambers was measured continuouslywith negative temperature coefficient thermistors (model -001; RTI Electronics)and was between 24°C and 25°C. Leaf temperature was determined from leafenergy balance based on absorbed light and transpiration. All tubing andconnections were made of Teflon and stainless steel. For illumination, four 50-Whalogen lamps provided photosynthetic photon flux density of 500mmolm22 s21

to each cuvette. Concentrations of CO2 and water vapor in the reference channel(i.e. air entering the measuring cuvette) and measurement channel (air comingout from the cuvette) were measured with an infrared gas analyzer (Li-7000;Li-Cor), and stomatal conductance and net assimilation rate were calculatedwith a custom-written program as described by Kollist et al. (2007).

The stomatal conductance of the plants was measured at ;70% RH,500 mmol m22 s21 light, and 24°C to 25°C air temperature. When stomatal con-ductance had stabilized, plants were treated with ABA, change in CO2 concen-tration, or change inVPD.ForABA treatment, cuvette coverswere removed, 10mM

ABAwith 0.012% Silwet L-77 (Duchefa) and 0.05% ethanol was sprayed on leaves,cuvette covers were returned, and measurement of stomatal conductance contin-ued. In experiments with CO2, the CO2 concentration in the chamber was reducedfrom 400 to 100 ppm for 2 h, thereafter returned to 400 ppm for 2 h, and then el-evated to 800 ppm for 2 h. In experimentswith high VPD, stomatal conductance ofthe plants was allowed to stabilize in the measurement cuvette with either 50 or400ppmCO2. Thereafter, the air humidity in the chamberwas lowered from;70%to ;30% to 40%, yielding a change in VPD from 0.7 to 1 kPa to 2 to 2.2 kPa.

The stomatal conductance of Arabidopsis was recorded with an eight-chamber, custom-built, temperature-controlled gas-exchange device analogousto the one described before (Kollist et al., 2007). The stomatal conductance ofplants was measured at;70% RH, 150 mmol m22 s21 light, and 24°C to 25°C airtemperature. When stomatal conductance had stabilized, plants were treatedsimilar to ferns with either ABA spraying or changes in CO2 concentration.

Data Analysis and Statistics

In the ABA and CO2 experiments, stomatal conductance values before andafter application of these stimuli were compared by repeated-measures ANOVAwith Tukey’s posthoc test. In experiments with reduced air humidity, stomatalhalf-response times were calculated by scaling the whole 42-min stomatal re-sponse to a range from 0% to 100% and by calculating the time when 50% sto-matal closure was achieved. Half-response times were compared by two-wayANOVAwith Tukey’s posthoc test. All statistical analyses were carried out withStatistica (StatSoft). All effects were considered significant at P , 0.05.

Supplemental Data

The following supplemental materials are available.

Supplemental Figure S1. Stomatal response of D. carthusiana to reducedair humidity in different CO2 concentrations.

Supplemental Table S1. Repeated-measures ANOVA with Tukey’s HSDposthoc test for Figure 2B.

Supplemental Table S2. Repeated-measures ANOVA with Tukey’s HSDposthoc test for Figure 2D, plants grown in the growth room.

678 Plant Physiol. Vol. 174, 2017

Hõrak et al.

www.plantphysiol.orgon March 12, 2020 - Published by Downloaded from Copyright © 2017 American Society of Plant Biologists. All rights reserved.

Supplemental Table S3. Repeated-measures ANOVA with Tukey’s HSDposthoc test for Figure 2D, plants grown in the growth cabinet.

Supplemental Table S4. Two-way ANOVA for pretreatment stomatal con-ductance in Figure 1C.

Supplemental Table S5. Two-way ANOVA with Tukey’s HSD posthoctest for pretreatment stomatal conductance in Figure 2D.

ACKNOWLEDGMENTS

We thank Toomas Kukk for help with fern species determination.

Received February 1, 2017; accepted March 27, 2017; published March 28, 2017.

LITERATURE CITED

Aliniaeifard S, Malcolm Matamoros P, van Meeteren U (2014) Stomatalmalfunctioning under low VPD conditions: induced by alterations instomatal morphology and leaf anatomy or in the ABA signaling? PhysiolPlant 152: 688–699

Arve LE, Carvalho DR, Olsen JE, Torre S (2014) ABA induces H2O2 pro-duction in guard cells, but does not close the stomata on Vicia faba leavesdeveloped at high air humidity. Plant Signal Behav 9: e29192

Bauer H, Ache P, Lautner S, Fromm J, Hartung W, Al-Rasheid KAS,Sonnewald S, Sonnewald U, Kneitz S, Lachmann N, et al (2013) Thestomatal response to reduced relative humidity requires guard cell-autonomous ABA synthesis. Curr Biol 23: 53–57

Bazzaz FA, Williams WE (1991) Atmospheric CO2 concentrations within amixed forest: implications for seedling growth. Ecology 72: 12–16

Brodribb TJ, McAdam SAM (2011) Passive origins of stomatal control invascular plants. Science 331: 582–585

Brodribb TJ, McAdam SAM (2013) Unique responsiveness of angiospermstomata to elevated CO2 explained by calcium signalling. PLoS ONE 8:e82057

Brodribb TJ, McAdam SAM, Jordan GJ, Feild TS (2009) Evolution ofstomatal responsiveness to CO2 and optimization of water-use efficiencyamong land plants. New Phytol 183: 839–847

Cai S, Chen G, Wang Y, Huang Y, Marchant B, Wang Y, Yang Q, Dai F,Hills A, Franks PJ, et al (2017) Evolutionary conservation of ABA sig-naling for stomatal closure in ferns. Plant Physiol 174: 732–747

Chater C, Kamisugi Y, Movahedi M, Fleming A, Cuming AC, Gray JE, BeerlingDJ (2011) Regulatory mechanism controlling stomatal behavior conservedacross 400 million years of land plant evolution. Curr Biol 21: 1025–1029

Chater C, Peng K, Movahedi M, Dunn JA, Walker HJ, Liang YK, McLachlanDH, Casson S, Isner JC, Wilson I, et al (2015) Elevated CO2-induced responsesin stomata require ABA and ABA signaling. Curr Biol 25: 2709–2716

Creese C, Oberbauer S, Rundel P, Sack L (2014) Are fern stomatal re-sponses to different stimuli coordinated? Testing responses to light,vapor pressure deficit, and CO2 for diverse species grown under con-trasting irradiances. New Phytol 204: 92–104

Cutler SR, Rodriguez PL, Finkelstein RR, Abrams SR (2010) Abscisic acid:emergence of a core signaling network. Annu Rev Plant Biol 61: 651–679

Doi M, Kitagawa Y, Shimazaki K (2015) Stomatal blue light response ispresent in early vascular plants. Plant Physiol 169: 1205–1213

Fanourakis D, Carvalho SMP, Almeida DPF, Heuvelink E (2011) Avoid-ing high relative air humidity during critical stages of leaf ontogeny isdecisive for stomatal functioning. Physiol Plant 142: 274–286

Fanourakis D, Heuvelink E, Carvalho SMP (2013) A comprehensive analysis ofthe physiological and anatomical components involved in higher water lossrates after leaf development at high humidity. J Plant Physiol 170: 890–898

Franks PJ, Britton-Harper ZJ (2016) No evidence of general CO2 insensi-tivity in ferns: one stomatal control mechanism for all land plants? NewPhytol 211: 819–827

Franks PJ, Farquhar GD (2007) The mechanical diversity of stomata and itssignificance in gas-exchange control. Plant Physiol 143: 78–87

Geiger D, Scherzer S, Mumm P, Stange A, Marten I, Bauer H, Ache P,Matschi S, Liese A, Al-Rasheid KAS, et al (2009) Activity of guard cellanion channel SLAC1 is controlled by drought-stress signaling kinase-phosphatase pair. Proc Natl Acad Sci USA 106: 21425–21430

Hanada K, Hase T, Toyoda T, Shinozaki K, Okamoto M (2011) Origin andevolution of genes related to ABA metabolism and its signaling path-ways. J Plant Res 124: 455–465

Hetherington AM, Woodward FI (2003) The role of stomata in sensing anddriving environmental change. Nature 424: 901–908

Kollist H, Nuhkat M, Roelfsema MRG (2014) Closing gaps: linking ele-ments that control stomatal movement. New Phytol 203: 44–62

Kollist T, Moldau H, Rasulov B, Oja V, Rämma H, Hüve K, Jaspers P,Kangasjärvi J, Kollist H (2007) A novel device detects a rapid ozone-induced transient stomatal closure in intact Arabidopsis and its absencein abi2 mutant. Physiol Plant 129: 796–803

Lee SC, Lan W, Buchanan BB, Luan S (2009) A protein kinase-phosphatasepair interacts with an ion channel to regulate ABA signaling in plantguard cells. Proc Natl Acad Sci USA 106: 21419–21424

Lind C, Dreyer I, López-Sanjurjo EJ, von Meyer K, Ishizaki K, Kohchi T,Lang D, Zhao Y, Kreuzer I, Al-Rasheid KAS, et al (2015) Stomatalguard cells co-opted an ancient ABA-dependent desiccation survivalsystem to regulate stomatal closure. Curr Biol 25: 928–935

Mansfield TA, Willmer CM (1969) Stomatal responses to light and carbondioxide in the hart’s-tongue fern, Phyllitis scolopendrium Newm. NewPhytol 68: 63–66

McAdam SAM, Brodribb TJ (2012) Fern and lycophyte guard cells do notrespond to endogenous abscisic acid. Plant Cell 24: 1510–1521

McAdam SAM, Brodribb TJ (2015) The evolution of mechanisms driving thestomatal response to vapor pressure deficit. Plant Physiol 167: 833–843

McAdam SAM, Brodribb TJ, Banks JA, Hedrich R, Atallah NM, Cai C,Geringer MA, Lind C, Nichols DS, Stachowski K, et al (2016) Abscisicacid controlled sex before transpiration in vascular plants. Proc NatlAcad Sci USA 113: 12862–12867

Merilo E, Laanemets K, Hu H, Xue S, Jakobson L, Tulva I, Gonzalez-Guzman M, Rodriguez PL, Schroeder JI, Brosche M, et al (2013) PYR/RCAR receptors contribute to ozone-, reduced air humidity-, darkness-and CO2-induced stomatal regulation. Plant Physiol 162: 1652–1668

Nejad AR, van Meeteren U (2005) Stomatal response characteristics of Trades-cantia virginiana grown at high relative air humidity. Physiol Plant 125: 324–332

Nejad AR, van Meeteren U (2007) The role of abscisic acid in disturbedstomatal response characteristics of Tradescantia virginiana duringgrowth at high relative air humidity. J Exp Bot 58: 627–636

O’Donoghue MT, Chater C, Wallace S, Gray JE, Beerling DJ, Fleming AJ(2013) Genome-wide transcriptomic analysis of the sporophyte of themoss Physcomitrella patens. J Exp Bot 64: 3567–3581

Okamoto M, Tanaka Y, Abrams SR, Kamiya Y, Seki M, Nambara E (2009)High humidity induces abscisic acid 89-hydroxylase in stomata andvasculature to regulate local and systemic abscisic acid responses inArabidopsis. Plant Physiol 149: 825–834

Pantin F, Monnet F, Jannaud D, Costa JM, Renaud J, Muller B, Simonneau T,Genty B (2013a) The dual effect of abscisic acid on stomata. New Phytol197: 65–72

Pantin F, Renaud J, Barbier F, Vavasseur A, Le Thiec D, Rose C, Bariac T,Casson S, McLachlan DH, Hetherington AM, et al (2013b) Develop-mental priming of stomatal sensitivity to abscisic acid by leaf microcli-mate. Curr Biol 23: 1805–1811

Pospíšilová J (1996) Effect of air humidity on the development of functionalstomatal apparatus. Biol Plant 38: 197

Roelfsema MRG, Hedrich R (2016) Do stomata of evolutionary distant speciesdiffer in sensitivity to environmental signals? New Phytol 211: 767–770

Ruszala EM, Beerling DJ, Franks PJ, Chater C, Casson SA, Gray JE,Hetherington AM (2011) Land plants acquired active stomatal controlearly in their evolutionary history. Curr Biol 21: 1030–1035

Shatil-Cohen A, Attia Z, Moshelion M (2011) Bundle-sheath cell regula-tion of xylem-mesophyll water transport via aquaporins under droughtstress: a target of xylem-borne ABA? Plant J 67: 72–80

Talbott LD, Rahveh E, Zeiger E (2003) Relative humidity is a key factor inthe acclimation of the stomatal response to CO2. J Exp Bot 54: 2141–2147

Torre S, Fjeld T, Gislerød HR, Moe R (2003) Leaf anatomy and stomatalmorphology of greenhouse roses grown at moderate or high air hu-midity. J Am Soc Hortic Sci 128: 598–602

Vahisalu T, Puzõrjova I, Brosché M, Valk E, Lepiku M, Moldau H, PechterP, Wang YS, Lindgren O, Salojärvi J, et al (2010) Ozone-triggered rapidstomatal response involves the production of reactive oxygen species, andis controlled by SLAC1 and OST1. Plant J 62: 442–453

Xue S, Hu H, Ries A, Merilo E, Kollist H, Schroeder JI (2011) Central functionsof bicarbonate in S-type anion channel activation and OST1 protein kinase inCO2 signal transduction in guard cell. EMBO J 30: 1645–1658

Zeevaart JAD (1974) Levels of (+/2) abscisic acid and xanthoxin in spinachunder different environmental conditions. Plant Physiol 53: 644–648

Plant Physiol. Vol. 174, 2017 679

Growth Conditions Affect Fern Stomatal Responses

www.plantphysiol.orgon March 12, 2020 - Published by Downloaded from Copyright © 2017 American Society of Plant Biologists. All rights reserved.