Phloem Sugar Flux and Jasmonic Acid-Responsive Cell WallInvertase Control Extrafloral Nectar Secretionin Ricinus communis

Cynthia Millán-Cañongo & Domancar Orona-Tamayo &

Martin Heil

Received: 11 March 2014 /Revised: 11 June 2014 /Accepted: 30 June 2014 /Published online: 15 July 2014# Springer Science+Business Media New York 2014

Abstract Plants secrete extrafloral nectar (EFN) that attractspredators. The efficiency of the resulting anti-herbivore de-fense depends on the quantity and spatial distribution of EFN.Thus, according to the optimal defense hypothesis (ODH),plants should secrete EFN on the most valuable organs andwhen herbivore pressure is high. Ricinus communis plantssecreted most EFN on the youngest (i.e., most valuable)leaves and after the simulation of herbivory via the applicationof jasmonic acid (JA). Here, we investigated the physiologicalmechanisms that might produce these seemingly adaptivespatiotemporal patterns. Cell wall invertase (CWIN; EC3.2.1.26) was most active in the hours before peak EFNsecretion, its decrease preceded the decrease in EFN secretion,and CWIN activity was inducible by JA. Thus, CWIN appearsto be a central player in EFN secretion: its activation by JA islikely to cause the induction of EFN secretion after herbivory.Shading individual leaves decreased EFN secretion locally onthese leaves with no effect on CWIN activity in the nectaries,which is likely to be because it decreased the content ofsucrose, the substrate of CWIN, in the phloem. Our resultsdemonstrate how the interplay of two physiological processescan cause ecologically relevant spatiotemporal patterns in aplant defense trait.

Keywords Extrafloral nectary . Jasmonic acid . Plant cellwallinvertase . Indirect defence . Ecological interaction .

Optimal defence theory

Introduction

Many plants respond to herbivore-inflicted damage with thesecretion of extrafloral nectar (EFN) and the release of volatileorganic compounds, both of which attract predators and para-sitoids, thus serving as a strategy of indirect defense againstherbivory (Heil 2008). The optimal defense hypothesis (ODH)aims to explain the allocation of anti-herbivore defenses indifferent parts and ontogenetic stages of plants, and it predictsthat plants should most intensively defend the most valuabletissues, or organs, and enhance defense levels when enemypressure is highest (McKey 1974; 1979; Rhoades 1979;Zangerl and Rutledge 1996). Indeed, studies in several differentplant species have shown that the spatiotemporal patterns ofEFN secretion or herbivore-induced volatile release are consis-tent with the patterns predicted by the ODH. Plants secrete moreEFN and release more volatiles after damage (Bixenmann et al.2011; Escalante-Pérez et al. 2012; Heil 2004; Heil et al. 2001;Wäckers et al. 2001; Wooley et al. 2007), that is, when enemypressure is high. Young leaves show the highest rates of volatilerelease and EFN secretion, and developing fruits, which areparticularly valuable according to the ODH, also can show veryhigh levels of EFN secretion (Holland et al. 2009; Radhika et al.2008; Rostás and Eggert 2008; Stephenson 1982; Wäckers andBonifay 2004; Wooley et al. 2007).

The aim of our study was to investigate possible physio-logical mechanisms that can explain these seemingly adaptivespatiotemporal patterns in EFN secretion. To date, inductionof nectar secretion (both floral nectar and EFN) by the hor-mone jasmonic acid (JA) has been reported for more than tenplant species from six different families (Bruinsma et al. 2008;Escalante-Pérez et al. 2012; Heil 2004; Heil et al. 2001; 2004;Radhika et al. 2008). Furthermore, cell-wall invertase (CWIN)has been shown to play a role in the unloading of sucrose fromthe phloem and in the active secretion process in the floral

CynthiaMillán-Cañongo andDomancar Orona-Tamayo have contributedequally to this paper

C. Millán-Cañongo :D. Orona-Tamayo :M. Heil (*)Departamento de Ingeniería Genética, CINVESTAV-Irapuato, Km.9.6 Libramiento Norte, 36821 Irapuato, Guanajuato, Méxicoe-mail: mheil@ira.cinvestav.mx

J Chem Ecol (2014) 40:760–769DOI 10.1007/s10886-014-0476-3

nectaries of Arabidopsis (Ruhlmann et al. 2010) or theextrafloral nectaries of Acacia cornigera (Orona-Tamayoet al. 2013). Photosynthesis in green nectaries (Lüttge 2013)has been identified as a further source of the carbohydratesthat form the quantitatively dominant components of EFN(González-Teuber and Heil 2009), and floral nectaries fre-quently store starch before secretion starts (Ren et al. 2007).However, starch grains are seldom reported from extrafloralnectaries (Gaffal 2012), and EFN secretion can be limited bylight intensity (Bixenmann et al. 2011). Thus, the availabilityof sucrose (the substrate of CWIN) in the phloem likelyrepresents a putatively limiting factor of EFN secretion (Heil2011). Therefore, we hypothesised that EFN secretion shoulddepend directly on concurrent sucrose flow in the phloem.

Here, we used Ricinus communis L. (the castor oil plant)(Euphorbiaceae; see Fig. 1a) to test the following predictions:(1) EFN secretion rate is highest on the youngest leaves; (2)EFN secretion can be induced by JA application; (3) spatio-temporal patterns in CWIN activity resemble, but precede,those in EFN secretion; (4) CWIN activity can be induced byJA application; and (5) reducing photosynthesis in singleleaves reduces EFN secretion by the nectaries that are associ-ated with these leaves. Ricinus communis plants bear threetypes of extrafloral nectaries: stipular (on the sheath at theinsertion point of the petiole, Fig. 1b); petiolar (Fig. 1c); andon the base of the leaf blade (Fig. 1d). The EFN containssucrose, glucose, and fructose in a 3:1:1 ratio (Baker et al.1978). Earlier studies have speculated that extrafloral nectar-ies contain a ‘metabolic enzyme’ that is involved in sucrosemetabolism (Nichol and Hall 1988). We used R. communis toinvestigate whether three general physiological factors, which

are not restricted to EFN-secreting plants (JA induction afterherbivory, CWIN activity, and sucrose in the phloem) canexplain the complex spatiotemporal patterns that are requiredto secrete EFN according to the predictions of the ODH.

Methods and Materials

Plant Material and Study Site From a wild population aroundCINVESTAV-Irapuato in the state of Guanajuato in CentralMéxico (20.7184°N 101.3296°W), we selected a total of 110R. communis plants that were growing 10–15 m apart and,therefore, likely to represent genetically different individuals.All plants used in this study were between 1.0 and 1.5 m high,grew in full sun, and showed no visible signs of infection bypathogens or damage by herbivores. The experiments wereconducted from November 2012 to May 2013.

Ontogenetic and Temporal Patterns in EFN Secretion Priorto nectar quantification, leaf ontogenetic stages were de-fined following Radhika et al. (2008): leaves 1 and 2(L1 and L2) were the youngest leaves (L1 was stillunfolding, whereas L2 usually was fully unfolded but stillsoft), leaves 3 and 4 (L3 and L4) were fully developed,and the mature leaf 5 (L5) was the oldest non-senescentleaf (inset in Fig. 2). Five out of 10 randomly selectedplants were pre-induced with an aqueous solution of1 mM JA (Heil 2004) 1 day before and then again 2 hrbefore the first EFN quantification. The other five plantswere used as controls and were sprayed with distilledwater. All plants in contact with the experimental plants

a

b c d

Fig. 1 a Ricinus communis plant.Extrafloral nectaries: b stipular; cpetiolar and d leaf blade

J Chem Ecol (2014) 40:760–769 761

were pruned, and a ring of sticky resin (Tangletrap®, TheTanglefoot Company; www.contech-inc.com) was appliedto the base of the shoot to exclude climbing insects.Extrafloral nectaries were washed with distilled waterand allowed to dry 2 hr before the first quantification, inorder to remove all accumulated EFN. After drying, theleaves, including the petioles, were placed inside meshbags, which allowed sunlight to reach the leaf surface forthe entire day of EFN quantification. EFN secretion wasmeasured at 07:00, 09:00, 11:00, 13:00, and 15:00 hr. Weused two nectaries each on the leaf blade, petiole andstipules to quantify the accumulated EFN by measuringits volume using 5 μl microcapillaries (Mikrokapillaren,Hirschmann Laborgeräte, Germany) and measuring theconcentration of soluble solids with a temperature-compensated hand refractometer (Atago), as describedearlier (Heil 2004). Depending on the experimental de-sign, values were pooled for all nectaries of a leaf orcalculated separately for stipular, petiolar, and leaf bladenectaries. In all cases, the leaves were collected after thelast collection of EFN and oven-dried at 60 °C to relateEFN amounts (as soluble solids) to the dry mass of thesecreting leaf.

Light Effects on EFN Secretion We performed two sets ofexperiments in which individual leaves at different positionswere shaded to decrease their photosynthesis. First, in additionto the mesh bags, L1 and L3 on five out of ten randomlyselected plants also were covered with black plastic bags(which were left open at the base to allow gas exchange andto avoid overheating and water condensation), whereas L2and L4 were covered only by the mesh bags. In the secondexperiment, L2 and L4 on five out of ten randomly selectedplants were covered with black plastic bags. These treatmentswere applied to both JA-treated and un-induced control plants,and the secretion of EFN was measured as described above.

Cell Wall Invertase Activity Cell wall invertase (CWIN) ac-tivity was determined in two independent experiments, oneresembling the conditions as in the experiment dedicated todetermine the time course in EFN secretion, the other oneresembling the shading experiment. All experimental condi-tions were as described above. Nectary tissues were collectedfrom five randomly selected plants per treatment (in the firstexperiment pooled over leaf age classes and nectary types butseparately for JA-treated and control plants, in the secondexperiment separately for JA-treated and control plants andseparately for shaded and sun-exposed leaves) and immedi-ately frozen in dry ice. Enzymatic activity was quantified asdescribed by Orona-Tamayo et al. (2013) and Ruhlmann et al.(2010) with some minor modifications. Ground nectary tissue(15 mg) was mixed with 3 mg of polyvinylpyrrolidone (PVP)and then with 500 μl of ice-cold 50 mM HEPES-NaOH(pH 8.0, containing 5 mM MgCl2, 2 mM EDTA, 1 mMMnCl2, and 1 mM CaCl2). Samples were incubated on icefor 10 min and then centrifuged at 10,000g for 20 min at 4 °C.The supernatant was discarded, and the pellet containing thecell walls with associated invertases was washed three timeswith 500 μl of extraction buffer by re-suspending and centri-fugation as described above. Finally, pellets were washed with500 μl of ice-cold 80 mM sodium citrate, pH 4.8, and theinvertase activity was measured as described previously (Heilet al. 2005, 2014) with some modifications. In short, 300 μl of80mM sodium citrate (pH 4.8; room temperature) were addedto the pellets, and the mixture was incubated at 37 °C. Thetubes then were centrifuged at 10,000 xg for 1 min at roomtemperature, and 20 μl of each sample were mixed with190 μl of HK reaction solution (Glucose (HK) Assay KitProduct Code GAHK-20, Sigma-Aldrich). After reachingthe steady state, 100 μl of an aqueous 100 mM solution ofsucrose were added to the samples, and the absorption wasmeasured at 340 nm in a μQuant® Microplate-reader contin-uously every 5 min for 40 min (Heil et al. 2005, 2014).

L1 L2L3

L4

L5

Leaf position

EF

N (

mg

g-1h-1

)

0.50

0.40

0.30

0.20

0.10

01 2 3 4 5

07:00Time of measurement 09:00 11:00 13:00 15:00

Leaf position

ControlJA

1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5

Fig. 2 Time course of nectarsecretion in differently-agedleaves of Ricinus communis.Extrafloral nectar (EFN) wasquantified at 07:00, 09:00, 11:00,13:00, and 15:00 separately forleaves L1–L5. Bars representmeans±SE (N=5) of EFNsecretion [mg soluble solids per gleaf dry mass and hour] of controlplants (grey bars) and jasmonicacid (JA)-treated plants (blackbars). See Table 1 for statisticalanalysis. Inset figure modifiedfrom Radhika et al. (2008)

762 J Chem Ecol (2014) 40:760–769

Table 1 Analysis of the effects of time of secretion, treatment with JA and leaf number on EFN secretion in Ricinus communis plants

Source of variance SS DF MS F Values P Values

Time of secretion 0.461 4 0.115 218.179 < 0.001

Leaf number 0.468 4 0.117 221.192 < 0.001

JA 0.168 1 0.168 317.868 < 0.001

Time of secretion × Leaf number 0.229 16 0.014 27.057 < 0.001

Time of secretion × JA 0.082 4 0.021 38.905 < 0.001

Leaf number × JA 0.060 4 0.015 28.500 < 0.001

Time of secretion × Leaf number × JA 0.033 16 0.002 3.886 < 0.001

SS sum of squares, DF degrees of freedom, MS mean of squares, F F value, P P value

Leaf 1

Leaf 2

Leaf 3

Leaf 4

Leaf 1

Leaf 2

Leaf 3

Leaf 4

EF

N (

mg

g-1h-1

)

0.10

0.30

0

0.20

0.10

0.30

0

0.20

0.10

0.30

0

0.20

0.10

0.30

0

0.20

0.10

0.30

0

0.20

0.10

0.30

0

0.20

0.10

0.30

0

0.20

0.10

0.30

0

0.20

07:00 11:0009:00 13:00 15:00 07:00 11:0009:00 13:00 15:00

Leaf blade Petiolar Stipular

Nectaryposition

Control JA

**

*

*

* **

*

*

**

*

*

*

*

Time of measurement

Fig. 3 Nectar secretion bydifferent extrafloral nectaries ofRicinus communis plants.Extrafloral nectar (EFN) wasquantified at 07:00, 09:00, 11:00,13:00, and 15:00 separately forleaves L1–L5 and for stipularnectaries (red bars), petiolarnectaries (green bars) and leafblade nectaries (red bars). Barsrepresent means±SE (N=5) ofEFN secretion [mg soluble solidsper g leaf dry mass and hour]. Leftcolumn: control plants, rightcolumn: Jasmonic acid (JA)-treated plants. Asterisks indicatesignificant treatment effects(P<0.05) . See Table 1 forstatistical analysis

J Chem Ecol (2014) 40:760–769 763

Statistical Analysis Data were analyzedwith Least SignificantDifference (LSD) post hoc tests after analysis of variance(ANOVA) when comparing more than two conditions, andwith t-tests when comparing two conditions, or treatments,using the Statistical Package for the Social Sciences 17.0(SPSS Inc, Chigago, IL, USA). Before ANOVA, the data weretested for homogeneity of variances. The results of the anal-yses of the effects on EFN production or invertase activity ofthe time of day, JA-treatment, leaf ontogenetic stage, or shad-ing are reported in tables that present: SS (Sum of Squares),DF (Degrees of Freedom), MS (Mean Square), F (F-value),and P (P value). Sample size was N=5 in all experiments.

Results

Ontogenetic Patterns and JA-Responsiveness of EFNSecretion Leaf ontogenetic stage, time of day, and treatmentwith exogenous JA had significant effects on EFN secretionby single leaves of R. communis (Table 1). EFN secretion wasfirst detected at 09:00 and then every 2 hr until 15:00 hr. Thesecretion of EFN peaked shortly before noon (at 11:00 hr).Leaf age significantly affected secretion rates, which werehighest on the youngest leaves (L1) at all times. JA-treatedplants showed the same general patterns at significantly en-hanced total EFN secretion rates (Fig. 2).

EFN Secretion by Different Nectary Types In the controls,stipular nectaries of leaf L1 exhibited significantly highersecretion rates than petiolar and leaf blade nectaries(P<0.05); however, the secretion rates decreased with leafage (Fig. 3; left panel). JA-treated plants showed a similaroverall pattern of EFN secretion to that observed in the con-trols but at elevated rates of secretion (Fig. 3; right panel).

JA-Responsiveness of CWIN Activity We used the above-described patterns in EFN secretion to investigate whetherCWIN activity in extrafloral nectaries might control thekinetics of EFN secretion. We observed high levels ofCWIN activity at 07:00 hr (Fig. 4a), that is, before themain peak of EFN secretion (Fig. 4b). The peak in CWINactivity occurred at 09:00 hr, and its activity decreasedthereafter. Hence, the temporal pattern in CWIN activitypreceded the pattern in EFN secretion by 2 hr. Similarly,we found that CWIN activity was significantly induced bythe JA treatment (Tables 2 and 3, Fig. 4a), in a pattern thattemporally and quantitatively resembled the pattern seenin EFN secretion (notable differences already at 09:00 hr,strongest differences at 11:00 hr, and differences then stillsignificant but quantitatively lower at 13:00 and 15:00 hr;see Fig. 4).

Light Effect on EFN Secretion When individual leaves wereshaded, they exhibited in general lower rates of EFN secretionthan sun-exposed leaves at the same position; however, theintensity of the effect depended on the positions of the leaves.Compared with control plants (Fig. 5a), when L1 and L3leaves were bagged, the reduction in EFN secretion wassignificant in most cases (Fig. 5b), whereas bagging leavesL2 and L4 significantly reduced EFN secretion only in L2(Fig. 5 c). When plants were treated with JA, overall secretionrates increased in accordance with the results of the earlier

007:00 09:00 11:00 13:00

Cel

l wal

l inv

erta

se

(µg

glu

ml-1

min

-1)

6.0

4.0

2.0

8.0

10.0 ControlJA

a

15:00

a a

b

c

a

c,d

e

c

d

Time of measurement

ControlJA

EF

N(m

g g-1

h-1)

1.2

1.0

0.8

0.6

0.4

0.2

007:00 09:00 11:00 13:00 15:00

b

aa,b

dc,d

e

b

d

c

a,b

Time of measurement

Fig. 4 Time course and jasmonic acid (JA)-responsiveness of nectarsecretion and cell-wall invertase (CWIN) activity in Ricinus communis.EFN was quantified at 07:00, 09:00, 11:00, 13:00, and 15:00 and CWINactivity was quantified at the same times, in control plants (grey bars) andJA-treated plants (black bars). a, Cell wall invertase activity, bars repre-sent means±SE (N=5) of sucrose hydrolyzing activity [μg glucosereleased per min per ml]. b, Extrafloral nectar (EFN) secretion, barsrepresent means±SE (N=5) of EFN secretion [mg secreted soluble solidsper g leaf dry mass and hour]. Different letters indicate significantdifferences between all factor combinations (Univariate ANOVA afterPost hoc Tukey test, P<0.05). See Table 1 for the effects of time ofsecretion and JA-treatment on EFN secretion and Table 1 for the effects ofthe same factors on CWIN activity

Table 2 Analysis of the effect of time of secretion and jasmonic acid(JA) treatment on enyzmatic cell-wall invertase (CWIN) activity

Source of variance SS DF MS F Values P Values

Time of secretion 141.356 4 35.339 79.211 < 0.001

JA 28.678 1 28.678 64.280 < 0.001

Time of secretion × JA 10.152 4 2.538 5.689 < 0.001

SS sum of squares, DF degrees of freedom, MS mean of squares,F F value, P P value

764 J Chem Ecol (2014) 40:760–769

experiments (Fig. 5d); however, EFN secretion on shadedleaves was significantly lower compared with JA-treatedleaves at the same positions that had full access to sunlight(Fig. 5 e,f).

Light Effects on CWIN Activity Cell wall invertase activ-ity in the three types of nectaries on leaves thatwere either sun-exposed or shaded from sunlight,exhibited no significant differences among nectary

Table 3 Effect of time of the day, sun-exposed and jasmonic acid (JA) treatment on cell-wall invertase (CWIN) activity from different extrafloralnectaries

Source of variance SS DF MS F Values P Values

Time of sampling 0.003 3 0.001 15.768 < 0.001

Treatment 0.008 1 0.008 122.274 < 0.001

Nectary position 0.000 2 0.000 0.963 0.387

Time of secretion × Treatment 0.000 3 0.000 1.811 0.153

Time of secretion × Nectary position 0.001 6 0.000 2.016 0.075

Treatment × Nectary position 0.000 2 0.000 0.205 0.815

Time of secretion × Treatment × Nectary position 0.000 6 0.000 0.879 0.515

SS sum of squares, DF degrees of freedom, MS mean of squares, F F value, P P value

L5

L1 L2L3

L4

a

L1L2

L3

L4

L1L2

L3

L4

L1 L2L3

L4

L1L2

L3

L4

L1 L2L3

L4

0

0EF

N (

mg

g-1h-1

)

0 0

0.40

0.30

0.20

0.10

0.50

0

0

0.40

0.30

0.20

0.10

0.50

Sun-exposed + JAShaded + JA

b

c

d

e

f

Sun-exposedShaded

07:00 09:00 11:00 15:00 07:00 09:00 11:00 13:00 15:00

Leafposition

0.40

0.30

0.20

0.10

0.50

13:00

0.40

0.30

0.20

0.10

0.50

0.40

0.30

0.20

0.10

0.50

0.40

0.30

0.20

0.10

0.50

*

*

*

*

*

*

1 2 3 4 51 2 3 4 5 1 2 3 4 5 1 2 3 4 51 2 3 4 51 2 3 4 51 2 3 4 5 1 2 3 4 5 1 2 3 4 51 2 3 4 5

L5 L5

L5 L5

L5

* *

**

*

*

*

** ***

Time of measurement

Fig. 5 Nectar secretion in differently-aged leaves of Ricinus communisunder different light conditions. Panels a-c represent values of controlplants (a all leaves sun-exposed; b leaves L1 and L3 shaded; c leaves L2and L4 shaded), panels d-f represent values of jasmonic acid (JA)-treatedplants (c, all leaves sun-exposed; d L1 and L3 shaded; e leaves L2 and L4shaded). Bars represent means±SE (N=5) of extrafloral nectar (EFN)

secretion [mg soluble solids per g leaf dry mass and hour]. Asterisksindicate significant treatment effects (P<0.05 according to t-test) amongvalues obtained at a specific leaf position and a given time in the blockingtreatment (shading of the respective leaf or its neighbours) and the fullysun-exposed control

J Chem Ecol (2014) 40:760–769 765

positions or light conditions (Table 4); however, CWINactivity responded significantly to external application ofJA in all nectaries (Table 4) and peaked at 09:00 hr(Fig. 6 a,b).

Discussion

Plants would benefit from protecting particularly those tissuesand organs from herbivores and pathogens that are mostsensitive to damage and most relevant for their fitness. Thesecretion of EFN fulfils these predictions, because EFN (i)usually contributes to the defense of plants against herbivores(Chamberlain and Holland 2009; Koricheva and Romero

2012; Rosumek et al. 2009), (ii) is commonly secreted athighest rates by nectaries on the youngest leaves or the devel-oping fruits (Radhika et al. 2008; Stephenson 1982; Wäckersand Bonifay 2004), and (iii) its secretion usually is activatedafter herbivory (Escalante-Pérez et al. 2012; Heil 2004; Heilet al. 2001, 2004; Radhika et al. 2008, 2010a, b; Rogers et al.2003; Wäckers et al. 2001; Wooley et al. 2007). Because themain defensive effect of EFN results from the attraction ofants, which then locally defend extrafloral nectaries as reliablefood sources (a behavior that includes the surrounding parts ofthe plant surface, see Heil 2008 for a review), plants mustguide these mutualists by secreting EFN on those organs thatcurrently require defense . Here, we discuss how these com-plex patterns in defense investment can be explained by theinterplay of two physiological factors: a JA-responsive CWINactivity and the availability of sucrose in the phloem.

The spatiotemporal patterns of CWIN activity resem-bled those of EFN secretion in all aspects, and temporallypreceded EFN secretion kinetics by 2 hr, as one wouldexpect for an enzyme that is directly underlying the ex-pression of a phenotypic trait. Hence, we conclude thatCWIN activity represents a central regulatory element inthe secretion of EFN by R. communis. Conclusive evi-dence would require the use of transgenic plants, whichare not available in this species, or of an invertase inhib-itor that specifically blocks the activity of this enzyme.However, our interpretation is consistent with earlier stud-ies in Arabidopsis (for floral nectaries; Ruhlmann et al.2010) and Acacia cornigera (for extrafloral nectaries;Orona-Tamayo et al. 2013), which indicated that CWINis involved in the unloading of sucrose from the phloeminto the nectary (Heil 2011; see also https://www.youtube.com/watch?v=Nd8ryN_7BP8). In the current study, wehave demonstrated that CWIN activity in the nectariescan be induced by exogenous JA, which is likely torepresent the mechanistic explanation of the herbivore-induced secretion of EFN.

CWIN activity was similar across the three nectary typesand remained stable when individual leaves were shaded,

Table 4 Effect of time of sampling, shaded leaves and jasmonic acid (JA) treatments on cell-wall invertase (CWIN) activity from different extrafloralnectaries

Source of variance SS DF MS F Values P Values

Time of secretion 0.001 3 0.000 3.731 0.015

Treatment 0.004 1 0.004 46.771 < 0.001

Nectary position 0.000 2 0.000 1.007 0.370

Time of secretion × Treatment 0.000 3 0.000 0.424 0.736

Time of secretion × Nectary position 0.001 6 0.000 1.362 0.242

Treatment × Nectary position 0.000 2 0.000 1.804 0.172

Time of secretion × Treatment × Nectary position 0.000 6 0.000 0.181 0.981

SS sum of squares, DF degrees of freedom, MS mean of squares, F F value, P P value

Cel

l wal

l inv

erta

se

(µg

glu

ml-1

min

-1)

0

0.06

0.04

0.02

0

Nectary position07:00

LB P S LB P S LB P S LB P S09:00 11:00 13:00

Sun-exposedSun-exposed + JA

a

b ShadedShaded + JA

0.06

0.04

0.02

Time of measurement

Fig. 6 Cell wall invertase activity in extrafloral nectaries does not dependon sunlight. Mean invertase activity±SE (μg glucose released per min perml; N=5) is indicated separately for the different types on nectary (LB:leaf blade; P: petiolar; S: stipular) for control plants (grey bars) and JA-treated plants (black bars). a plants completely sun-exposed; b plants withshaded leaves. See Table 4, respectively, for statistical analysis

766 J Chem Ecol (2014) 40:760–769

whereas EFN secretion rates differed significantly among thesame types of nectaries and dropped after shading. Similarly, astudy in sugar beet (Beta vulgaris) has shown that CWINactivity does not change with changing light conditions(Vassey 1989), whereas EFN secretion in other species alsoresponded to changes in light conditions (Bixenmann et al.2011). Hence, CWIN activity alone cannot explain the ob-served patterns in EFN secretion rates. We argue that sucroseavailability in the phloem represents a second important fac-tor, which explains that shaded leaves exhibited dramaticallyreduced EFN secretion rates. Our shading treatment withblack plastic bags is likely to have affected several physiolog-ical processes and, therefore, the results must be interpretedwith caution. However, it appears reasonable to assume thatthe most drastic effect of this treatment was a strong reductionor even a complete inhibition of photosynthesis in the shadedleaves and, thus, reduced levels of sucrose in the phloem. Thisinterpretation is in line with the observation that the stipularnectaries, which are located close to the phloem in the mainshoot, exhibited higher secretion rates than the petiolar nec-taries or the nectaries at the base of the leaf blade. Followingthe predictions of the ODH, one would expect the nectaries onthe leaf blade (which is normally attacked by herbivores) to bethe most active. For example, in cotton (Gossypium hirsutum),the bracteal nectaries exhibit higher secretion rates than thefoliar nectaries, and attract more defensive ants to thesehighly vulnerable and important (reproductive) organs(Wäckers and Bonifay 2004). The three youngest leavesof R. communis plants produce more than 90 % of the EFN,and their secretion rate increased strongly in response toherbivory (Wäckers et al. 2001) or exogenous JA treatment(Radhika et al. 2010a).

Based on our results, we hypothesize that the seeminglyadaptive patterns in EFN secretion are generated by the inter-play of general physiological factors that are not exclusivelyrelated to EFN. First, EFN is a common indirect defense trait,and its secretion depends on endogenous levels of JA in allplant species that have been investigated so far (Heil 2008;2011). Second, CWIN has been shown in several plants torespond to herbivory with enhanced expression levels and/orenzyme activities (Arnold and Schultz 2002; Kaplan et al.2008; Landgraf et al. 2013; Ohyama et al. 1998) and theseresponses are likely to depend on herbivore-induced levels ofendogenous JA (Arnold et al. 2004; Arnold and Schultz 2002;Bogatek et al. 2002; Schaarschmidt et al. 2007; Zhang et al.1996). CWIN is centrally involved in the re-allocation ofcarbohydrates that enable plants to compensate for herbivory(Appel et al. 2012; Arnold et al. 2004; Castrillon-Arbelaezet al. 2012; Herbers et al. 1996; Kaplan et al. 2008; Landgrafet al. 2013; Roitsch and González 2004; Sturm and Chrispeels1990; Zhang et al. 1996) and, thus, likely to be linked to JA-dependent signaling in all plants. Third, EFN secretion, butnot CWIN activity, is reduced on shaded leaves, which

indicates the importance of sucrose availability in the phloemas a limiting factor. Nevertheless, young leaves usually arestrong sinks for carbohydrates and then quickly shift and actas sources of assimilates, and also young fruits are strongsinks (Lunn and Hatch 1995; Schaffer et al. 1987; Turgeon1989). Hence, the sucrose content in the phloem of defense -requiring organs generally should be high (Claussen et al.1985; Lohaus et al. 1994; Martin et al. 1993; Pharr and Sox1984; Samarakoon and Rauser 1979), independent of thedirection of carbohydrate flow.

We conclude that EFN secretion is likely to depend on aJA-inducible CWIN activity and on the carbohydrate alloca-tion patterns in plants. These patterns usually coincide withthe value and vulnerability of different organs, thus allowingfor EFN-secretion patterns that are optimal for defense . Evenin the above-mentioned example of cotton (Wäckers andBonifay 2004), the enhanced secretion rates on bracteal nec-taries might be a simple physiological consequence of the sinkstatus of flowers and young fruits, rather than a pattern that hasspecifically evolved in adaptation to ecological, defense -related selection pressures. The JA-responsiveness of CWINas a central controller of EFN secretion rates provides aphysiological explanation for the observation that EFN secre-tion is usually induced after herbivory. Future studies shouldaim at a broad comparative approach to test for the generalapplicability of these findings in EFN-producing plants acrossseveral major taxa of plants. According to current phyloge-netic evidence, EFN secretion has evolved several times in-dependently in many clades of plants (Weber and Keeler2013). The creation of a functioning EFN secretion by thesimple assembly of two general (and, thus, pre-existing) traitswould provide an elegant explanation for the short evolution-ary times that plants require to gain extrafloral nectaries.

Acknowledgments We thank Antonio Cisneros for taking the photo-graphs of R. communis, CINVESTAV-Irapuato for providing all fieldfacilities and Caroline Woods as well as three anonymous referees formultiple valuable comments on an earlier version of this manuscript.CONACyT of México is gratefully acknowledged for financial supportto DOT (grant: 191236) and MH (project grants: 129678 and 130656).

References

Appel HM, Arnold TM, Schultz JC (2012) Effects of jasmonic acid,branching and girdling on carbon and nitrogen transport in poplar.New Phytol 195:419–426

Arnold TM, Schultz JC (2002) Induced sink strength as a prerequisite forinduced tannin biosynthesis in developing leaves of Populus.Oecologia 130:585–593

Arnold T, Appel HM, Patel V, Stocum E, Kavalier A, Schultz JC (2004)Carbohydrate translocation determines the phenolic content ofPopulus foliage: a test of the sink–source model of plant defense.New Phytol 164:157–164

Baker DA, Hall JL, Thorpe JR (1978) A study of the extrafloral nectariesof Ricinus communis. New Phytol 81:129–137

J Chem Ecol (2014) 40:760–769 767

Bixenmann RJ, Coley PD, Kursar TA (2011) Is extrafloral nectar pro-duction induced by herbivores or ants in a tropical facultative ant-plant mutualism? Oecologia 165:417–425

Bogatek R, Côme D, Corbineau F, Ranjan R, Lewak S (2002) Jasmonicacid affects dormancy and sugar catabolism in germinating appleembryos. Plant Physiol Biochem 40:167–173

Bruinsma M, Ildema H, van Loon JJA, Dicke M (2008) Differentialeffects of jasmonic acid treatment of Brassica nigra on the attractionof pollinators, parasitoids, and butterflies. Entomol Exp Appl 128:109–116

Castrillon-Arbelaez P, Martinez-Gallardo N, Arnaut H, Tiessen A,Delano-Frier J (2012) Metabolic and enzymatic changes associatedwith carbon mobilization, utilization and replenishment triggered ingrain amaranth (Amaranthus cruentus) in response to partial defoli-ation by mechanical injury or insect herbivory. BMC Plant Biol12:163

Chamberlain SA, Holland JN (2009) Quantitative synthesis of contextdependency in ant-plant protection mutualisms. Ecology 90:2384–2392

Claussen W, Loveys BR, Hawker JS (1985) Comparative investigationson the distribution of sucrose synthase activity and invertase activitywithin growing, mature and old leaves of some C3 and C4 plantspecies. Physiol Plant 65:275–280

Escalante-Pérez M, Jaborsky M, Lautner S, Fromm J, Müller T, DittrichM, Kunert M, Boland W, Hedrich R, Ache P (2012) Poplarextrafloral nectaries: two types, two strategies of indirect defensesagainst herbivores. Plant Physiol 159:1176–1191

Gaffal KP (2012) How common is the ability of extrafloral nectaries toproduce nectar droplets, to secrete nectar during the night and tostore starch? Plant Biol 14:691–695

González-Teuber M, Heil M (2009) Nectar chemistry is tailored for bothattraction of mutualists and protection from exploiters. Plant SignalBehav 4:809–813

Heil M (2004) Induction of two indirect defences benefits Lima bean(Phaseolus lunatus, Fabaceae) in nature. J Ecol 92:527–536

Heil M (2008) Indirect defence via tritrophic interactions. New Phytol178:41–61

Heil M (2011) Nectar: generation, regulation and ecological functions.Trends Plant Sci 16:191–200

Heil M, Koch T, Hilpert A, Fiala B, Boland W, Linsenmair KE (2001)Extrafloral nectar production of the ant-associated plant,Macarangatanarius, is an induced, indirect, defensive response elicited byjasmonic acid. Proc Natl Acad Sci U S A 98:1083–1088

Heil M, Greiner S, Meimberg H, Krüger R, Noyer J-L, Heubl G,Linsenmair KE, Boland W (2004) Evolutionary change from in-duced to constitutive expression of an indirect plant resistance.Nature 430:205–208

Heil M, Büchler R, Boland W (2005) Quantification of invertase activityin ants under field conditions. J Chem Ecol 30:431–437

Heil M, Barajas-Barron A, Orona-Tamayo D, Wielsch N, Svatos A(2014) Partner manipulation stabilises a horizontally transmittedmutualism. Ecol Lett 17:185–192

Herbers K, Meuwly P, Frommer WB, Metraux J-P, Sonnewald U (1996)Systemic acquired resistance mediated by the ectopic expression ofinvertase: possible hexose sensing in the secretory pathway. PlantCell 8:793–803

Holland JN, Chamberlain SA, Horn KC (2009) Optimal defence theorypredicts investment in extrafloral nectar resources in an ant-plantmutualism. J Ecol 97:89–96

Kaplan I, Halitschke R, Kessler A, Rehill BJ, Sardanelli S, Denno RF(2008) Physiological integration of roots and shoots in plant defensestrategies links above-and belowground herbivory. Ecol Lett 11:841–851

Koricheva J, Romero GQ (2012) You get what you pay for: reward-specific trade-offs among direct and ant-mediated defences in plants.Biol Lett 8:628–630

Landgraf R, Schaarschmidt S, Hause B (2013) Repeated leaf woundingalters the colonization of Medicago truncatula roots by beneficialand pathogenic microorganisms. Plant Cell Environ 35:1344–1357

Lohaus G, Burba M, Heldt HW (1994) Comparison of the contents ofsucrose and amino acids in the leaves, phloem sap and taproots ofhigh and low sugar-producing hybrids of sugar beet (Beta vulgarisL.). J Exp Bot 45:1097–1101

Lunn JE, Hatch MD (1995) Primary partitioning and storage of photosyn-thate in sucrose and starch in leaves of C4 plants. Planta 197:385–391

Lüttge U (2013) Green nectaries: the role of photosynthesis in secretion. JLinn Soc Bot 173:1–11

Martin T, Frommer WB, Salanoubat M, Willmitzer L (1993) Expressionof an Arabidopsis sucrose synthase gene indicates a role inmetabolization of sucrose both during phloem loading and in sinkorgans. Plant J 4:367–377

McKey D (1974) Adaptive patterns in alkaloid physiology. Am Nat 108:305–320

McKey D (1979) The distribution of secondary compoundswithin plants.In: Rosenthal GA, Janzen DH (eds) Herbivores: their interactionswith secondary plant metabolites. Academic Press, New York, pp55–133

Nichol P, Hall JL (1988) Characteristics of nectar secretion by theextrafloral nectaries of Ricinus communis. J Exp Bot 39:573–586

Ohyama A, Nishimura S, Hirai M (1998) Cloning of cDNA for a cellwall-bound acid invertase from tomato (Lycopersicon esculentum)and expression of soluble and cell wall-bound invertases in plantsand wounded leaves of L. esculentum and L. peruvianum. GenesGen Syst 73:149–157

Orona-Tamayo D, Wielsch N, Escalante-Pérez M, Svatos A, Molina-Torres J, Muck A, Ramirez-Chávez E, Ádame-Alvarez R-M, HeilM (2013) Short-term proteomic dynamics reveal metabolic factoryfor active extrafloral nectar secretion by Acacia cornigera ant-plants. Plant J 73:546–554

Pharr DM, Sox HN (1984) Changes in carbohydrate and enzyme levelsduring the sink to source transition of leaves ofCucumis sativusL., astachyose translocator. Plant Sci Lett 35:187–193

Radhika V, Kost C, Bartram S, Heil M, Boland W (2008) Testing theoptimal defence hypothesis for two indirect defences: extrafloralnectar and volatile organic compounds. Planta 228:449–457

Radhika V, Kost C, Boland W, Heil M (2010a) The role of jasmonates infloral nectar secretion. PLoS One 5:e9265

Radhika V, Kost C, Mithöfer A, Boland W (2010b) Regulation ofextrafloral nectar secretion by jasmonates in lima bean is lightdependent. Proc Natl Acad Sci U S A 107:17228–17233

Ren G, Healy RA, Klyne AM, Horner HT, James MG, Thornburg RW(2007) Transient starch metabolism in ornamental tobacco floral nec-taries regulates nectar composition and release. Plant Sci 173:277–290

Rhoades DF (1979) Evolution of plant chemical defense against herbi-vores. In: Rosenthal GA, Janzen DH (eds) Herbivores: their inter-action with secondary plant metabolites. Academic Press, NewYork, pp 4–53

Rogers WE, Siemann E, Lankau RA (2003) Damage induced productionof extrafloral nectaries in native and invasive seedlings of Chinesetallow tree (Sapium sebiferum). Am Midl Nat 149:413–417

Roitsch T, González M-C (2004) Function and regulation of plant inver-tases: sweet sensations. Trends Plant Sci 9:606–613

Rostás M, Eggert K (2008) Ontogenetic and spatio-temporal patterns ofinduced volatiles in Glycine max in the light of the optimal defencehypothesis. Chemoecology 18:29–38

Rosumek FB, Silveira FAO, Neves FD, Barbosa NPD, Diniz L, Oki Y,Pezzini F, Fernandes GW, Cornelissen T (2009) Ants on plants: ameta-analysis of the role of ants as plant biotic defenses. Oecologia160:537–549

Ruhlmann JM, Kram BW, Carter CJ (2010) CELLWALL INVERTASE4 is required for nectar production in Arabidopsis. J Exp Bot 61:395–404

768 J Chem Ecol (2014) 40:760–769

Samarakoon AB, Rauser WE (1979) Carbohydrate levels andphotoassimilate export from leaves of Phaseolus vulgaris ex-posed to excess cobalt, nickel, and zinc. Plant Physiol 63:1165–1169

Schaarschmidt S, González M-C, Roitsch T, Strack D, Sonnewald U,Hause B (2007) Regulation of arbuscular mycorrhization by carbon.The symbiotic interaction cannot be improved by increased carbonavailability accomplished by root-specifically enhanced invertaseactivity. Plant Physiol 143:1827–1840

Schaffer AA, Sagee O, Goldschmidt EE, Goren R (1987) Invertaseand sucrose synthase activity, carbohydrate status and endoge-nous IAA levels during citrus leaf development. Physiol Plant69:151–155

Stephenson AG (1982) The role of the extrafloral nectaries of Catalpaspeciosa in limiting herbivory and increasing fruit production.Ecology 63:663–669

Sturm A, Chrispeels MJ (1990) cDNA cloning of carrot extracellularbeta-fructosidase and its expression in response to wounding andbacterial infection. Plant Cell 2:1107–1119

Turgeon R (1989) The sink-source transition in leaves. Annu Rev PlantBiol 40:119–138

Vassey TL (1989) Light/dark profiles of sucrose phosphate synthase,sucrose synthase, and acid invertase in leaves of sugar beets. PlantPhysiol 89:347–351

Wäckers FL, Bonifay C (2004) How to be sweet? Extrafloral nectarallocation by Gossypium hirsutum fits optimal defense theory pre-dictions. Ecology 85:1512–1518

Wäckers FL, Zuber D, Wunderlin R, Keller F (2001) The effect ofherbivory on temporal and spatial dynamics of foliar nectar produc-tion in cotton and castor. Ann Bot 87:365–370

Weber MG, Keeler KH (2013) The phylogenetic distribution ofextrafloral nectaries in plants. Ann Bot 111:1251–1261

Wooley SC, Donaldson JR, Gusse AC, Lindroth RL, Stevens MT (2007)Extrafloral nectaries in aspen (Populus tremuloides): heritable ge-netic variation and herbivore-induced expression. Ann Bot 100:1337–1346

Zangerl AR, Rutledge CE (1996) The probability of attack and patterns ofconstitutive and induced defense: a test of optimal defense theory.Am Nat 147:599–608

Zhang L, Cohn NS, Mitchell JP (1996) Induction of a pea cell-wallinvertase gene by wounding and its localized expression in phloem.Plant Physiol 112:1111–1117

J Chem Ecol (2014) 40:760–769 769