LC

RDa

b

a

ARRA

KCPLRBF

1

alpTttcrlltttr

h0

Scientia Horticulturae 218 (2017) 177–186

Contents lists available at ScienceDirect

Scientia Horticulturae

journa l h om epage: www.elsev ier .com/ locate /sc ihor t i

ight quality regulates plant architecture in different genotypes ofhrysanthemum morifolium Ramat

obrecht Diercka,b, Emmy Dhooghea, Johan Van Huylenbroecka,ominique Van Der Straetenb, Ellen De Keysera,∗

Institute for Agricultural and Fisheries Research (ILVO) Caritasstraat 39, Melle, BelgiumLaboratory of Functional Plant Biology, Department of Physiology, Ghent University, K.L. Ledeganckstraat 35, Ghent, Belgium

r t i c l e i n f o

rticle history:eceived 28 October 2016eceived in revised form 7 February 2017ccepted 9 February 2017

eywords:hrysanthemum morifolium ramatlant architectureED lighted

a b s t r a c t

Specific light spectra play an important role in plant photomorphogenic responses. Plants are sensitive tolight ranging from UV (280–400 nm) to far-red light (700–800 nm). The shoot architecture or plant shapeis an important quality trait in ornamental plants and can be altered under specific light spectra. The shadeavoidance syndrome is a well-documented response of plants to canopy shading and low R:FR conditions,characterized by shoot elongation and inhibited branching. Treatments with LED light combinations toobtain different spectral compositions were tested on rooted cuttings of 3 chrysanthemum genotypes(a pot chrysanthemum, a cut flower and a disbud chrysanthemum genotype) to assess the effect onshoot architecture. Red light treatment generally showed increased bud outgrowth and increased averagebud length while blue + far-red light treatment resulted in decreased bud outgrowth and bud length.

luear-red

Some effects were genotype dependent, such as plant height, which increased under blue + far-red lighttreatments compared to red light treatment, only for the pot chrysanthemum genotype.

Treatment with blue + far-red light in 25 decapitated cuttings showed a strong elongation of the top-most axillary bud and inhibition of underlying buds for the pot chrysanthemum and cut flower genotypes.This effect also persisted in greenhouse conditions.

© 2017 Elsevier B.V. All rights reserved.

. Introduction

The morphological response to light quality in plants is part of strategy to adapt to a changing light environment. This is regu-ated by the interaction between plant photoreceptors sensitive toarticular wavelengths and their downstream signaling pathways.he most striking effect of light composition on shoot architec-ure is the shade avoidance syndrome. This phenomenon describeshe elongated growth of plants, growing in high density, to escapeanopy shading (Pierik and de Wit, 2014). The shade avoidanceesponse includes increased internode elongation, inhibited axil-ary bud outgrowth, petiole elongation and upward bending ofeaves (hyponasty). This response to shading is caused by changes inhe red (R) to far-red (FR) light ratio, resulting in a FR enhancement

hat is perceived by the plant photoreceptors. In a dense canopy,he surrounding foliage absorbs red light, while much more far-ed light reaches the lower canopy. The R:FR ratio is calculated by

∗ Corresponding author.E-mail address: ellen.dekeyser@ilvo.vlaanderen.be (E. De Keyser).

ttp://dx.doi.org/10.1016/j.scienta.2017.02.016304-4238/© 2017 Elsevier B.V. All rights reserved.

dividing the photon irradiance between 655 and 665 nm (R) withthe photon irradiance between 725 and 735 (FR). This R:FR ratioranges from 1.2 in daylight to 0.1 under canopy shading (Franklin,2008). Blue (B) and UV light are also known to be involved in plantphotomorphogenic responses and are perceived by a number ofphotoreceptors, including phytochromes, cryptochromes, and pho-totropins. Blue light and UV light generally show reduced plantheight and effects on shoot branching are species dependent (forreview see Huché-Thélier et al., 2016). Stem elongation in chrysan-themum is also inhibited by blue light treatment (Shimizu et al.,2006).

The shade avoidance response includes involvement of cryp-tochrome and phytochrome A (Casal, 2013) but the phytochromeB (PHYB) photoreceptor predominantly regulates the response tored and far-red light (Pierik and de Wit, 2014). Phytochrome B ispresent in two forms: the inactive Pr and the active Pfr form. ThePr form absorbs red light and gets converted to the Pfr form. ThePfr form absorbs far-red light, which reverts the receptor to the Pr

form. In this way a changing red to far-red ratio results in a chang-ing equilibrium between phytochrome B in the inactive Pr or theactive Pfr form. This relationship is quantified by the phytochrome

1 rticul

p�U�era1ao

hatiaoawpRiamisipA2baniwe

tuioi2R

silh(2h

rwsmtwiuptb

r

78 R. Dierck et al. / Scientia Ho

hotostationary state (PSS) or photoequilibrium, quantified by the-value as the ratio of the Pfr form to the total phytochrome B.sing spectroradiometric data and phytochrome absorbance, the-value can be estimated (Sager et al., 1988). This value is consid-red to be more closely correlated to plant responses than the R:FRatio because the phytochrome photo-conversion involves inter-cting wavelengths between 350 and 850 nm (Kelly and Lagarias,985; Sager et al., 1988). Daylight conditions have a �-value ofround 0.81 while single source red and far-red light have �-valuesf 0.89 and 0.2, respectively.

A low red to far-red ratio, characterized by a low �-value,as been linked to an increased biosynthesis of auxin, associ-ted with the shade avoidance response (Halliday et al., 2009). Inhis response, auxin is responsible for shoot elongation and alsonfluences the axillary bud outgrowth through the mechanism ofuxin in apical dominance, where growth of the shoot apex inhibitutgrowth of underlying axillary buds (Cline, 1991). This inter-ction of light and auxin has been demonstrated in Arabidopsis,here auxin responsive genes were found to be upregulated in

hyB mutants, which have a reduced branching phenotype (Krishnaeddy and Finlayson, 2014). This indicates a promotion of branch-

ng through repression of auxin signaling by PHYB activation with high R:FR ratio. Inactivation of PHYB by a low R:FR (or in the phyButant) upregulates auxin responsive genes and inhibits branch-

ng. Phytochrome interacting factors PIF4 and PIF5 have beenhown to be involved in this interaction of PHYB and auxin signal-ng (Hornitschek et al., 2012). Furthermore, also auxin biosynthesislays a role, as increased accumulation of IAA has been shown inrabidopsis seedlings under low R:FR conditions (Keuskamp et al.,010). A low R:FR ratio has also been shown in Sorghum and Ara-idopsis to promote the expression of the BRC1 transcription factor,ssociated with inhibition of bud outgrowth, through PHYB sig-aling (Kebrom et al., 2006; Finlayson et al., 2010). In addition,

nvolvement of strigolactones has also been shown in Sorghum,here the strigolactone signaling gene MAX2 was required for the

ffect of PHYB on shoot branching (Kebrom et al., 2006, 2010).The effect of light quality could be applied to control shoot archi-

ecture and the outgrowth of axillary buds. Some studies have madese of spectral filters with a CuSO4 solution or a far-red absorb-

ng dye, to block out far-red light and increase the compactnessf plants. This was reported to result in reduced plant height andnternode length in chrysanthemum (Rajapakse et al., 1992; Li et al.,003; Khattak et al., 2004) as well as in other species (reviewed byajapakse et al., 1999).

Another way to adjust the R:FR ratio is to use R and FR lightources. In many plants, using FR light (or low R:FR) results inncreased plant height and decreased bud outgrowth and budength (Demotes-Mainard et al., 2016). R and FR LED light sourcesave been examined in chrysanthemum for flower bud inductionSingh, 2013; Jeong et al., 2014), rooting and biomass (Kurilcik et al.,008; Christiaens et al., 2015; Hong et al., 2015) as well as planteight (Lund et al., 2007).

In cut flower chrysanthemum, the final shape of the plant isequired to be elongated and unbranched, except for some branchesith flowers near the apex. In the production of cut flowers with a

ingle top flower (disbud types), axillary buds have to be removedanually. In general though, the early production phase of these

ypes still requires a compact growth for better shoot quality, forhich growth retardants are used. Only rarely, if plants would reach

nsufficient length through natural growth, gibberellic acid (GA) issed to stimulate elongation. In that case, a light treatment couldresent a non-chemical alternative to control growth. The produc-

ion of pot and garden chrysanthemums requires a highly branched,ushy plant habit.

In this study, we tested different LED combinations to find a lightecipe to stimulate axillary bud outgrowth and a bushy architecture

turae 218 (2017) 177–186

for pot and garden chrysanthemums, as well as a light treatmentthat inhibits outgrowth of axillary buds for disbud type chrysanthe-mums. Hereto, we investigated the effect of different combinationsof red, blue and far-red LED light, compared to fluorescent light onshoot architecture (plant height and bud outgrowth) in chrysan-themum. Treatments with far-red LED light were always given incombination with other light spectra (in this case blue), since far-red light does not contribute to the photosynthetic active radiation(McCree, 1971).

2. Materials and methods

2.1. Plant material

Unrooted cuttings of different chrysanthemum genotypes wereobtained from companies in Belgium (Gediflora for genotype C9;Dataflor for genotype C13) and the Netherlands (Dekker Chrysan-ten for genotype C17). Genotype C9 is a pot chrysanthemum type,characterized by a short growth habit and spontaneous branch-ing. C13 is a special type of disbud chrysanthemum from whichunwanted lateral buds are being removed during growth in orderto obtain large flowers. C17 is a cut flower chrysanthemum withan elongated growth habit and strong apical dominance. Cuttingswere rooted in a standard greenhouse at 20 ◦C under long day light(16 h High Pressure Sodium (HPS) 100 �mol m−2 s−1) conditions for3 weeks before start of the experiment in the LED growth cham-ber. In all experiments, rooted cuttings were randomly placed in a56-cell tray for each treatment.

2.2. Experimental set-up

Light conditions in the growth chamber (MAIS AUTOMATISER-ING NV, St. Katelijne Waver Belgium) were set by adjusting thefluence rate of red, far-red and blue light (Koninklijke Philips N.V.,Amsterdam, Green Power LED research modules) with a total flu-ence rate of 60 �mol m−2 s−1 PAR light. Fluorescent light (FL: coolwhite Philips Master TLD 36W/840) of 60 �mol m−2 s−1 served asa control condition in a separate growth chamber for the secondand third experiment. The light spectra of the blue, red, far-red andfluorescent light sources is found in Supplementary Fig. 1. In thefirst experiment, fluence rates of 30 and 90 �mol m−2 s−1 were alsotested. The temperature was set at a constant temperature of 20 ◦Cwith a relative humidity of 70% and 19 h photoperiod. The spectraldistribution of light intensity between 200 and 900 nm was mea-sured with a JAZ spectrophotometer (Ocean Optics Inc., Dunedin,FL, USA) and converted with Spectrasuite (Ocean Optics) and Excelto �mol m−2 s−1 values. Fluence rate of FR light is also expressedas �mol m−2 s−1 but does not contribute to the total PAR (totalfluence rate between 300 and 700 nm). The �-value was calcu-lated using phytochrome absorbance data from Sager et al. (1988).Rooted cuttings were kept in the growth chamber for 6 weeks.

1. In the first experiment, 25 rooted cuttings of cut flower chrysan-themum genotype C17 were tested per light treatment. Ten ofthese cuttings were decapitated at the start of the experimentand 15 were left intact. The used light qualities and fluence ratesare in Table 1. The treatments with R and BFR represented anextremely high �-value of 0.89 and an extremely low �-value of0.2 respectively. These treatments were also given at a fluence

rate of 30 �mol m−2 s−1 and 90 �mol m−2 s−1. Treatment with60 �mol m−2 s−1 FL was used as a control. 60 �mol m−2 s−1 bluelight and other combinations of blue and red light were addedto include intermediate �-values for 60 �mol m−2 s−1.

R. Dierck et al. / Scientia Horticul

Table 1Light conditions for rooted cuttings of Chrysanthemum morifolium Ramat. genotypeC17. The first column encodes the different light treatments according to the fluencerate of red (R), far-red (FR), blue (B) or fluorescent light (FL) in the recipe. The R,FR and B columns present the fluence rates provided by the red, far-red and blueLEDs respectively for each treatment. The FL column presents the fluence rate ofthe treatment with fluorescent light. The PAR column shows the fluence rate forthe photosynthetic active radiation that does not include FR light. The last columnpresents the estimated �-value.

Fluence rate (�mol m−2 s−1)

Code R FR B FL PAR �-value

60FL 0 0 0 60 60 0.8460B 78FR 0 78 60 0 60 0.2060R 60 0 0 0 60 0.8960B 0 0 60 0 60 0.5130R 30B 30 0 30 0 60 0.8410R 50B 10 0 50 0 60 0.8050R 10B 50 0 10 0 60 0.8890B 78FR 0 78 90 0 90 0.2090R 90 0 0 0 90 0.8930B 50FR 0 50 30 0 30 0.1830R 0 0 0 0 30 0.89

Table 2Light conditions for rooted cuttings of Chrysanthemum morifolium Ramat. genotypeC9, C13 and C17. The first column encodes the different light treatments accordingto the fluence rate of red (R), far-red (FR) or blue (B) light in the recipe. The R,FR and B columns present the fluence rates provided by the red, far-red and blueLEDs respectively for each treatment. The PAR column shows the fluence rate forthe photosynthetic active radiation that does not include FR light. The last columnpresents the estimated �-value.

Code Fluence rate (�mol m−2 s−1)

R FR B PAR �-value

60R 60 0 0 60 0.8944R16B16FR 44 16 16 60 0.8160B60FR 0 60 60 60 0.20

Table 3Light conditions for rooted cuttings of Chrysanthemum morifolium Ramat. genotypeC9, C13 and C17. The first column encodes the different light treatments accordingto the fluence rate of red (R), far-red (FR), blue (B) or fluorescent light in the recipe.The R, FR and B columns present the fluence rates provided by the red, far-red andblue LEDs respectively for each treatment. The FL column presents the fluence rateof the treatment with fluorescent light. The PAR column shows the fluence rate forthe photosynthetic active radiation that does not include FR light. The last columnpresents the estimated �-value.

Code Fluence rate (�mol m−2 s−1)

R FR B FL PAR �-value

60FL 0 0 0 60 60 0.8460R 60 0 0 0 60 0.89

2

(fwnbaa

sm

−2 −1

57R3B 57 0 3 0 60 0.8960B60FR 60 60 0 0 60 0.19

. In the second experiment, rooted cuttings of cultivars C9, C13and C17 were grown under different light conditions listed inTable 2 and Table 3.

A first part of the experiment consisted of 3 light treatmentsTable 2): the two treatments with extreme �-values were selectedrom the previous experiment and the treatment with 44R16FR16Bas chosen to more closely resemble the R:FR ratio and �-value ofatural light. Cuttings were treated for 6 weeks in the growth cham-er and half were decapitated, while the other half was left intactt the moment of insertion in the growth chamber. Both for intact

nd decapitated plants 15 cuttings were measured per treatment.

The second part of the experiment used rooted cuttings of theame genotypes (C9, C13 and C17) and consisted of 4 light treat-ents (Table 3). The 57R3B treatment was chosen to represent

turae 218 (2017) 177–186 179

the 95% red light and 5% blue light of HPS lighting. The FL treat-ment was used as a control and the 2 treatments with extreme�-values of experiment 1 were included. Treatments were given foreither 6 weeks in the growth chamber or for 3 weeks in the growthchamber, followed by 3 weeks in the greenhouse. Per treatment 10decapitated cuttings were measured.

In all experiments, axillary bud length and the number of nodeswere measured after 6 weeks under light treatment. The percentageof outgrown axillary buds was estimated by the ratio of the numberof buds larger than 0.5 cm to the total number of nodes. On intactplants, plant height and internode length were measured as well.Internode length is presented as the ratio of plant height to thenumber of internodes.

2.3. Statistics

To identify significant differences in light recipes, ANOVA withLevene’s and Shapiro-Wilk tests for assumptions and Tukey posthoc analyses were performed with Statistica 64, StatSoft, Inc.(2014).

3. Results

For the intact C17 plants of the first experiment, the percentageof bud outgrowth was low throughout all light treatments withonly a noticeable difference in the 90R treatment (Fig. 1A). Thedecapitated plants showed a higher percentage of bud outgrowth,ranging from 20 to 50% (Fig. 1B). In the treatments with a fluencerate of 60 �mol m−2 s−1 the plants under 60R and 30R30B, showeda significantly higher percentage of outgrown buds than the 60BFRtreatment. The 60 B and the 60FL treatments showed similar per-centages of bud outgrowth that were somewhat smaller than theother treatments. In the 30 �mol m−2 s−1 fluence rate treatments,the 30R showed a higher percentage of bud outgrowth than 30BFR,like in the 60 �mol m−2 s−1 treatments, but in the 90 �mol m−2 s−1

treatments this was not the case and the bud outgrowth percentagewas similar in 90R and 90BFR.

The bud length of intact plants was low and did not differ muchbetween the different treatments (Fig. 1C), however, treatmentswith BFR and B light showed significantly lower bud lengths com-pared to the other treatments for all 3 intensities that were tested.The decapitated plants under a fluence rate of 60 �mol m−2 s−1,showed the highest axillary bud length in the 60R treatmentand the lowest in the 60BFR treatment (Fig. 1D). In the 30 and90 �mol m−2 s−1 treatments, the bud length was also higher in theR than in the BFR treatment, but the difference between R and BFRwas only significant at 60 and 90 �mol m−2 s−1· The other treat-ments at 60 �mol m−2 s−1 showed similar bud lengths. The budlength of decapitated cuttings was generally higher at the positionsthat were closest to the shoot apex and especially for the treatmentwith 60R. This is shown by the greater bud lengths of the first fivepositions under the apex compared to the bud length of the rest ofthe nodal positions (Supplementary Fig. 2).

The intact plants showed a difference in plant height betweenthe light intensities, with a decreasing plant height at lower lightintensities for both the R and BFR treatments. For the differentlight qualities at 60 �mol m−2 s−1, 60R and 60BFR showed similarplant heights that were significantly higher compared to the othertreatments (Fig. 1E).

Internode length was higher in the BFR than in the R light condi-tion, which was significant at the 30, and 90 �mol m−2 s−1 fluence

rates (Fig. 1F). In the 60 �mol m s fluence rate, the 60R, 60BFRand 60 B treatments showed similar internode lengths that weresignificantly higher than the other treatments at this intensity.While plant height did not differ much between the R and BFR treat-

180 R. Dierck et al. / Scientia Horticulturae 218 (2017) 177–186

Fig. 1. Shoot architecture measurements in intact and decapitated C17 Chrysanthemum morifolium Ramat. cuttings under different light treatments. A) Percentage of budoutgrowth of intact plants (n = 15 ± SE). B) Percentage of bud outgrowth of decapitated plants (n = 10 ± SE). C) Bud length of intact plants (n = 15 ± SE). D) Bud length decapitatedplants (n = 10 ± SE). E) Plant height of intact plants (n = 15 ± SE). F) Internode length of intact plants (n = 15 ± SE). Letters indicate significant differences at a p-value <0.05with ANOVA and Tukey HSD post-hoc.

rticul

mswbft3a

eploodiiat

lbabtt

lioalht

tgoPtaiiswtogboag(

w(ttIaCohs

u

R. Dierck et al. / Scientia Ho

ents for the three intensities that were tested, internode lengthhowed a trend with greater internode length in the treatmentith BFR compared to the R treatments. The lack of correlation

etween plant height and internode length is explained by the dif-erence in the number of nodes, which is consistently higher in the Rreatments than in the BFR treatments (30.4 ± 0.98 vs. 25.2 ± 0.82;2.2 ± 0.69 vs 28 ± 0.69 and 36.4 ± 1.10 vs. 29.4 ± 0.63 for 30, 60nd 90 �mol m−2 s−1 respectively).

For the recurring treatments of R and BFR in the different flu-nce rates of 30, 60 and 90 �mol m−2 s−1, a two-way ANOVA waserformed to test for interaction between the intensity and the

ight quality (Supplementary Fig. 3). Concerning the percentagef bud outgrowth in intact plants there was no significant effectf intensity, light quality and no significant interaction effect. Inecapitated plants, there was a significant effect of the light qual-

ty treatments but not of light intensity. The interaction betweenntensity and light quality was significant, with a higher percent-ge of bud outgrowth for the R treatment compared to the BFRreatment at 30 and 60 �mol m−2 s−1 but not at 90 �mol m−2 s−1.

Bud length in intact plants showed a significant effect for theight quality treatments but not for light intensity or the interactionetween intensity and light quality. In decapitated plants, there was

significant effect on bud length for light intensity and light qualityut not for the interaction between quality and intensity since therend for 30, 60 and 90 �mol m−2 s−1 was a higher bud length forhe R treatment compared to the BFR treatment.

Plant height of the intact cuttings showed a significant effect foright intensity but not for light quality or the interaction betweenntensity and light quality. Internode length of the intact plantsnly showed a significant effect for the light quality treatmentsnd not for light intensity or the interaction between intensity andight quality since the trend for 30, 60 and 90 �mol m−2 s−1 was aigher internode length for the treatment with BFR compared tohe R treatment.

In the second experiment, 3 genotypes of chrysanthemum wereested including the cut flower genotype C17, a pot chrysanthemumenotype C9 and a cut flower disbud type C13. Based on the resultsf the first experiment, fluence rate was set at 60 �mol m−2 s−1

AR and the number of light treatments were reduced to includehe two extreme �-values (0.89 for 60R and 0.20 for 60B60FR). In

first part of the experiment, a treatment with 44R16FR16 B wasncluded to represent the light conditions of natural light. In thentact cuttings, there was no remarkable bud outgrowth (data nothown). In the decapitated plants, the percentage of bud outgrowthas highest in the C9 genotype under R light (Fig. 2A). In all geno-

ypes, the BFR condition resulted in the lowest percentage of budutgrowth, while the R light condition gave the highest bud out-rowth percentage. This difference was significant in all genotypesut was more pronounced in C9 and C13 than in C17, where the budutgrowth percentage in R was only slightly higher than in RBFRnd BFR. Bud length in intact plants was generally small in the 3enotypes and consistently lower than 0.5 cm in all light conditionsdata not shown).

In the decapitated plants, axillary bud length in the C9 genotypeas significantly higher under R light compared to RBFR and BFR

Fig. 2B). For C17, bud length in R and RBFR was significantly higherhan in BFR conditions. In contrast, the bud length of C13 in the BFRreatment was significantly higher than in BRFR and BFR conditions.n intact plants of C9, plant height differed significantly betweenll light conditions and was higher in BFR than in R and BRFR. In13 and C17, plant height in R was higher than in BFR, which wasnly significant in C17 (Fig. 2C). Internode length was significantly

igher in the BFR than in R and BRFR for the C9 but did not showignificant differences for the C13 and C17 genotypes (Fig. 2D).

In the second part of experiment 2, only decapitated plants weresed, based on the strong apical dominance observed in all geno-

turae 218 (2017) 177–186 181

types. Light treatments included the same two extreme �-values,a control with fluorescent light and a treatment with 95% red lightand 5% blue light (57R3B) to represent the light conditions of aHPS light treatment. Furthermore, half of the cuttings received thistreatment for 6 weeks (w6), whereas the other half of the cut-tings was transferred to greenhouse conditions after 3 weeks oftreatment in the growth chamber (w3). The percentage of bud out-growth was generally lower in the BFR treatments in all genotypesand both in cuttings that were in the growth chamber for 6 weeksand those that were transferred to greenhouse conditions after 3weeks (Fig. 3A). The BFR effect on bud outgrowth was significantin the C17 and C13 genotypes for both w6 and w3 and for the C9genotype with w3. This effect was most pronounced in C17 (Fig. 3A).There was little difference in bud outgrowth between the 60FL, 60Rand 57R3B conditions for all genotypes except for C17. The cuttingsof C17 that spent 6 weeks in the growth chamber had a higher per-centage of bud outgrowth in the 57R3B condition compared to the60FL treatment.

Bud length was significantly lower in BFR conditions than in theR treatment for the C13 genotype with w6 and w3, and for C17 withw6. The bud length of the R treatment was higher than the FL and57R3 B conditions for w6 and w3 (Fig. 3B). Genotype C9 showedsimilar bud lengths for all light conditions in the cuttings that weretransferred to greenhouse conditions after 3 weeks. The cuttingsthat spent 6 weeks in the growth chamber showed a significantlylower bud length under fluorescent light compared to the othertreatments.

At the end of the experiment, the plant phenotypes (Fig. 4) visu-ally showed minimal differences between the cuttings that hadspent 6 weeks in the growth chamber versus the cuttings that werein the growth chamber for 3 weeks and under greenhouse condi-tions for 3 weeks. From the images, it is noticeable that the C17 andC9 genotype have one axillary shoot at the top that is highly elon-gated in the 60B60FR treatment for both w6 and w3. This elongationis not seen in the C13 genotype. In the first part of experiment 2,bud length for C13 was higher under BFR treatment compared to R,while in the second part of experiment 2, C13 bud length was higherunder R treatment compared to BFR. This resulted from a pheno-type of C13 in the first part with an elongated top axillary shootunder BFR treatment (average shoot length from base to apex of:0.75; 1.68; 1.98; 2.72 and 3.3 cm respectively), similar to the phe-notypes of C17 and C9 (Fig. 4). In the second part of the experiment,genotype C13 showed an opposite phenotype under BFR treatmentwith longer axillary shoots at the base of the cuttings and shorterat the apex (average shoot length from base to apex of 2.94; 2.9;1.55; 0.77; 0.73; 0.8 cm and 4.58; 2.76; 1.06; 0.4 for w6 and w3respectively).

4. Discussion

Treatment of rooted cuttings with different light spectraresulted in observable differences in shoot architecture and budoutgrowth in the different chrysanthemum genotypes under study.Results from the first experiment with the cut flower chrysanthe-mum C17 revealed a strong apical dominance in this genotype,evidenced by the low percentage of bud outgrowth and low budlength throughout all of the light treatments in the intact plants.In spite of this strong apical dominance there was still an observ-able trend with cuttings under red light (with a high �-value of0.89), showing larger bud length and a higher percentage of budoutgrowth than cuttings under BFR (with a low �-value of 0.2) for

all light intensities that were tested. Remarkably, plant height in theintact plants did not differ much between the two treatments withextreme �-values. This result would seem opposed to an increasedshoot elongation that is expected with a low �-value in the shade

182 R. Dierck et al. / Scientia Horticulturae 218 (2017) 177–186

F ) and

B lantsl alue <

aotssafoacatthiPgua

ig. 2. Shoot architecture measurements in C9 (pot chrysanthemum), C13 (disbudFR light treatments (n = 15 ± SE). A) Percentage of bud outgrowth of decapitated p

ength of intact plants. Letters indicate significant differences per genotype at a p-v

voidance response (Smith and Whitelam, 1997). However, intern-de length was higher for the BFR treatment compared to the Rreatment for all the light intensities that were tested. This resulttill shows that the treatment with far-red light causes an increasedhoot elongation. The decapitated plants that were released frompical dominance did show axillary bud outgrowth. Noticeable dif-erences between light treatments were the high percentage of budutgrowth and bud length in the R treatment compared to the BFRnd B treatments. Remarkably, the control treatment with fluores-ent light showed a low bud outgrowth percentage and bud lengths well, similar to the aforementioned treatments. The C17 geno-ype was also included in the further experiments. This showedhat the percentage of bud outgrowth and bud length were alsoigher in the R treatment compared to the BFR treatment for decap-

tated cuttings but the difference was stronger in experiment 1.

lants in the second experiment were also shorter and differentreenhouse conditions for the cuttings in both experiments couldnderlie this contrast. The visual evaluation of the plant phenotypeslso revealed that under the BFR treatment, there was one axillary

C17 (cut flower) Chrysanthemum morifolium Ramat. genotypes under R, BRFR and. B) Bud length of decapitated plants. C) Plant height of intact plants. D) Internode0.05 with ANOVA and Tukey HSD post-hoc.

bud at the top of the decapitated cutting that grew out stronglyand reached a high length, while underlying axillary shoots or budswere short or inhibited from growing out. This was also observedwith genotype C9 under BFR treatment.

A likely explanation for this observation is that the release ofapical dominance induces bud outgrowth in the axillary buds clos-est to the decapitated stem surface. The low R:FR condition wouldthen increase elongation of this outgrowing bud and correlativeinhibition (Snow, 1940) between buds would inhibit underlyingbuds from growing out. A low R:FR ratio is also known to increaseauxin levels and the expression of auxin responsive genes (Hallidayet al., 2009), which would strengthen the inhibition of bud out-growth by auxin. However, this is not evident when looking at thebud lengths of decapitated cuttings in the first experiment, where60R showed greater subapical bud length than the BFR treatment

−2 −1

at 60 �mol m s . Strangely, plant height in the intact plants ofthe first part of the second experiment was higher for R than for FR,while internode length did not differ between the treatments. Thisresult for C17 does not match the higher internode length under far-

R. Dierck et al. / Scientia Horticulturae 218 (2017) 177–186 183

Fig. 3. Shoot architecture measurements in decapitated C9 (pot chrysanthemum), C13 (disbud) and C17 (cut flower) Chrysanthemum morifolium Ramat. genotypes underd is thew cant dp

rteIsd(cocdtonattem

bPwaaplfiu2c

oIhtb

ifferent light treatments (60FL, 60R, 57R3B and 60B60FR) (n = 10 ± SE). At the X-axeeks (w3). A) Percentage of bud outgrowth. B) Bud length. Letters indicate signifiost-hoc.

ed treatment that was seen in the first experiment with C17. Thisype of chrysanthemum is grown for the production of cut flow-rs that contain several flowers on axillary shoots near the apex.n the production cycle, the initiation of flowering is induced byhort day conditions or occurs spontaneously under long day con-itions after the formation of a certain amount of vegetative leafsLDL), resulting in the shoot apex transforming into an inflores-ence meristem, followed by release from apical dominance andutgrowth of subapical axillary buds. Under subsequent short dayonditions floral initiation also occurs in the axillary shoots, pro-ucing a spray type flower. Considering this production cycle andhe strong apical dominance during vegetative growth, a controlf bud outgrowth by application of different light conditions isot immediately useful. However, during the early growth stage

compact growth is still desired, which was achieved under BFRreatment in the second experiment. Nevertheless, since the plantshat were tested showed a variation in plant height between thexperiments it would be necessary to perform additional experi-ents to achieve a consistent result.The pot type chrysanthemum C9 showed a higher percentage of

ud outgrowth in the R treatment compared to the BFR treatment.lant height and internode length of intact plants for genotype C9ere higher in the BFR treatment than in the R treatment. C9 is

pot chrysanthemum genotype that requires a compact growthnd highly branched phenotype. Since results showed the mostositive effects on the percentage of bud outgrowth and on bud

ength under the 60R treatment, this treatment would be advisableor applications during the early growth stage. A precaution heres that there are indications of deformations in leaf morphologynder single-source red light (Nhut and Nam, 2010; Van Ieperen,012), however such effects were not seen in our experiments withhrysanthemum.

Disbud genotype C13 also displayed a higher percentage of budutgrowth in the R treatment compared to the BFR treatment.

nternode length for C13 was higher in the BFR treatment but planteight was higher in the R treatment. C13 is a disbud type chrysan-hemum and with this genotype, axillary branching is unwantedecause one single apical flower is desired and axillary shoots are

genotype and the period in the growth chamber is indicated (6 weeks (w6) and 3ifferences per genotype per week at a p-value <0.05 with ANOVA and Tukey HSD

removed for this purpose. Treatments with FR showed a reducedpercentage of bud outgrowth in this genotype, however an assess-ment of flowering development and quality would be needed todetermine the usefulness of FR application in the production. Forinstance, a high level of red light is known to have an inhibitoryeffect on flower bud initiation in chrysanthemum (Jeong et al.,2014). It must be noted that although three genotypes were cho-sen that each represent a different type of chrysanthemum, this lownumber of genotypes limits the prediction we can make for thesetypes in general. Considering also the variation between the geno-types, in future research it would be necessary to include a largerrepresentation of genotypes.

In general, for the intact rooted cuttings of the chrysanthemumgenotypes that were studied, there was a low percentage of budoutgrowth. This indicates a strong apical dominance in these plants,where growth of the shoot apex inhibits axillary bud outgrowth(Cline, 1991). A consistent trend was also seen in most experi-ments, where R conditions resulted in higher percentage of budoutgrowth and bud lengths than BFR conditions (except for C13 inthe second part of the last experiment). These results are consis-tent with a reduced branching under low R:FR light conditions asa part of the shade avoidance syndrome and are in line with simi-lar observations in a variety of other plant species (summarised inDemotes-Mainard et al., 2016). Low R:FR ratios in chrysanthemumhave been reported to induce an increased stem length (Khattakand Pearson, 2006). This was similarly observed in the C9 geno-type but was not seen for C17 and C13. A possible explanation forthis difference could be that the low R:FR ratio and low �-value inour study was obtained by combining blue with far-red light andblue light is known to decrease plant height (Huché-Thélier et al.,2016). This has been shown in vitro chrysanthemum plants thathad a decreased plant height under 1:1 blue and far-red conditionscompared with fluorescent light, whereas a 1:1 red and far-redtreatment increased plant height (Kim et al., 2004). In this study,

the focus was to obtain extreme �-values by combining blue andfar-red light to study bud outgrowth and plant height as a partof the shade avoidance response mediated through phytochromeB. In further experiments, it would be interesting to separately

184 R. Dierck et al. / Scientia Horticulturae 218 (2017) 177–186

F of gen( owth

lctrnte

ig. 4. Plant phenotypes of decapitated Chrysanthemum morifolium Ramat. cuttings

60FL; 60R; 57R3B; and 60B60FR) in the growth chamber or after 3 weeks in the gr

ook at R + FR treatments and B light treatments mediated throughryptochromes. Although no impact on shoot branching in cryp-ochrome loss of function mutants of several species has beeneported, overexpression of CRY1 and CRY2 induced branched phe-

otypes in tomato but not in Arabidopsis (Leduc et al., 2014). Alsoesting this for chrysanthemum would offer new insights in theffect of blue light in the regulation of bud outgrowth.

otype C17, C13 and C9 after 6 weeks (w6) of growth under different light treatmentschamber and a consecutive 3 weeks in the greenhouse (w3).

During the early production of both pot, cut flower and disbudtype chrysanthemums, a compact plant architecture is preferredby growers. During this stage, the rooting process of cuttings isalso important. In a study on the rooting of chrysanthemum cut-

tings, it was shown that the best results for rooting were with 100%red light treatment or 100% blue light treatment (Christiaens et al.,2015). Since the effects on shoot branching and plant habit under

rticul

RtbdtoiewtCspnw

trhbmpwag

5

lelcleltimwiga

A

t(tac

A

i0

R

C

C

R. Dierck et al. / Scientia Ho

light were better than the blue light treatments for pot chrysan-hemum C9, a treatment with 100% red light would be optimal foroth shoot architecture and rooting of chrysanthemum cuttingsuring the early production stage. However, in a follow-up study onhe rooting of chrysanthemum (Christiaens, 2015), cuttings showedptimal rooting under conditions with a combination of R,B and anncreased amount of FR light. Since our results showed negativeffects of FR light on bud outgrowth, a separate initial treatmentith FR light could be given to induce rooting, after which R light

reatments can be used to stimulate branching. For disbud type13 and cut flower genotype C17, treatments with B and FR lighthowed a shorter plant height and decreased bud outgrowth com-ared to the R light treatments. Since axillary bud outgrowth isot wanted for the production of these genotypes, a BFR treatmentould be optimal for a compact, unbranched shape.

The last experiment, where half of the cuttings were transferredo greenhouse conditions after 3 weeks in the growth chamber,evealed little differences between the phenotypes of plants thatad spent 6 weeks in the growth chamber and plants that hadeen transferred. This indicates that a short initial period of treat-ent with different light conditions results in a growth habit that

ersist throughout later stages under greenhouse conditions. Thisas most evident with the cuttings under BFR light, where the top

xillary shoot still dominated underlying shoots after 3 weeks ofreenhouse conditions.

. Conclusions

LED light treatment of chrysanthemum cuttings with differentight qualities showed effects on shoot architecture with a gen-ral trend of increased bud outgrowth and increased average budength under red light treatment with an extreme �-value of 0.89ompared to decreased bud outgrowth and bud length under BFRight treatment with an extreme �-value of 0.2. There was a strongffect of apical dominance in all but one of the genotypes, causingittle effect of light treatments on shoot branching in intact cut-ings. However, some effects were genotype dependent, showingncreased plant height under BFR treatments compared to R treat-

ent for C9 but not for C13 or C17. The use of different light qualitiesith LED light offers promising results to control plant architecture

n chrysanthemum. Furthermore, the possibility of multi-layeredrowth systems strengthens the use of LED light as an interestinglternative to traditional lighting.

cknowledgements

This research was funded by the Institute for the Promo-ion of Innovation through Science and Technology in FlandersIWT-Vlaanderen Grant No. 110771). The authors also wish tohank Magali Losschaert for assistance with the measurements andcknowledge Roger Dobbelaere and Frederik Delbeke for takingare of the plants in the greenhouse.

ppendix A. Supplementary data

Supplementary data associated with this article can be found,n the online version, at http://dx.doi.org/10.1016/j.scienta.2017.2.016.

eferences

asal, J.J., 2013. Photoreceptor signaling networks in plant responses to shade.Annu. Rev. Plant Biol. 64, 403–427, http://dx.doi.org/10.1146/annurev-arplant-050312-120221.

hristiaens, A., Van Labeke, M.C., Gobin, B., Van Huylenbroeck, J., 2015. Rooting ofornamental cuttings affected by spectral light quality. In: Acta Horti‘1culturae.

turae 218 (2017) 177–186 185

International Society for Horticultural Science (ISHS), Leuven, Belgium, pp.219–224, http://dx.doi.org/10.17660/actahortic.2015.1104.33.

Christiaens, A., 2015. Verrood licht stimuleert beworteling van stekken. Acta Hort.(ISHS) 1104, 219–224.

Cline, M.G., 1991. Apical dominance. Bot. Rev. 57, 318–358, http://dx.doi.org/10.1007/BF02858771.

Demotes-Mainard, S., Péron, T., Corot, A., Bertheloot, J., Le Gourrierec, J.,Pelleschi-Travier, S., Crespel, L., Morel, P., Huché-Thélier, L., Boumaza, R., Vian,A., Guérin, V., Leduc, N., Sakr, S., 2016. Plant responses to red and far-red lights,applications in horticulture. Environ. Exp. Bot. 121, 4–21, http://dx.doi.org/10.1016/j.envexpbot.2015.05.010.

Finlayson, S.A., Krishnareddy, S.R., Kebrom, T.H., Casal, J.J., 2010. Phytochromeregulation of branching in Arabidopsis. Plant Physiol. 152, 1914–1927, http://dx.doi.org/10.1104/pp.109.148833.

Franklin, K.A., 2008. Shade avoidance. New Phytol. 179, 930–944, http://dx.doi.org/10.1111/j.1469-8137.2008.02507.x.

Halliday, K.J., Martinez-Garcia, J.F., Josse, E.-M., 2009. Integration of light and auxinsignaling. Cold Spring Harb. Perspect. Biol. 1, a001586, http://dx.doi.org/10.1101/cshperspect.a001586.

Hong, Y., Huang, H., Dai, S., 2015. An in vivo study of the best light emitting diode(LED) systems for cut chrysanthemums. Open Life Sci. 10, 310–321, http://dx.doi.org/10.1515/biol-2015-0031.

Hornitschek, P., Kohnen, M.V., Lorrain, S., Rougemont, J., Ljung, K., López-Vidriero,I., Franco-Zorrilla, J.M., Solano, R., Trevisan, M., Pradervand, S., Xenarios, I.,Fankhauser, C., 2012. Phytochrome interacting factors 4 and 5 control seedlinggrowth in changing light conditions by directly controlling auxin signaling.Plant J. 71, 699–711, http://dx.doi.org/10.1111/j.1365-313X.2012.05033.x.

Huché-Thélier, L., Crespel, L., Gourrierec, J., Le Morel, P., Sakr, S., Leduc, N., 2016.Light signaling and plant responses to blue and UV radiations-perspectives forapplications in horticulture. Environ. Exp. Bot. 121, 22–38, http://dx.doi.org/10.1016/j.envexpbot.2015.06.009.

Jeong, S.W., Hogewoning, S.W., van Ieperen, W., 2014. Responses of supplementalblue light on flowering and stem extension growth of cut chrysanthemum. Sci.Hortic. (Amsterdam) 165, 69–74, http://dx.doi.org/10.1016/j.scienta.2013.11.006.

Kebrom, T.H., Burson, B.L., Finlayson, S.A., 2006. Phytochrome B represses teosintebranched1 expression and induces sorghum axillary bud outgrowth inresponse to light signals 1. Plant Physiol. 140, 1109–1117, http://dx.doi.org/10.1104/pp.105.074856.1.

Kebrom, T.H., Brutnell, T.P., Finlayson, S.A., 2010. Suppression of sorghum axillarybud outgrowth by shade, phyB and defoliation signalling pathways. Plant CellEnviron. 33, 48–58, http://dx.doi.org/10.1111/j.1365-3040.2009.02050.x.

Kelly, J.M., Lagarias, J.C., 1985. Photochemistry of 124-kilodalton Avenaphytochrome under constant illumination in vitro. Biochemistry 24,6003–6010, http://dx.doi.org/10.1021/bi00342a047.

Keuskamp, D.H., Pollmann, S., Voesenek, L.A.C.J., Peeters, A.J.M., Pierik, R., 2010.Auxin transport through PIN-FORMED 3 (PI) controls shade avoidance andfitness during competition. Pnas 107 (52), 22740–22744, http://dx.doi.org/10.1073/pnas.1013457108/-/DCSupplemental.

Khattak, A.M., Pearson, S., 2006. Spectral filters and temperature effects on thegrowth and development of chrysanthemums under low light integral. PlantGrowth Regul. 49, 61–68, http://dx.doi.org/10.1007/s10725-006-0020-8.

Khattak, A.M., Pearson, S., Johnson, C.B., 2004. The effects of far red spectral filtersand plant density on the growth and development of chrysanthemums. Sci.Hortic. (Amsterdam) 102, 335–341, http://dx.doi.org/10.1016/j.scienta.2004.05.001.

Kim, S.-J., Hahn, E.-J., Heo, J.-W., Paek, K.-Y., 2004. Effects of LEDs on netphotosynthetic rate, growth and leaf stomata of chrysanthemum plantlets invitro. Sci. Hortic. (Amsterdam) 101, 143–151, http://dx.doi.org/10.1016/j.scienta.2003.10.003.

Krishna Reddy, S., Finlayson, S.A., 2014. Phytochrome B promotes branching inArabidopsis by suppressing auxin signaling. Plant Physiol. 164, 1542–1550,http://dx.doi.org/10.1104/pp.113.234021.

Kurilcik, A., Miklusyte-Canova, R., Dapkuniene, S., Zilinskaite, S., Kurilcik, G.,Tamulaitis, G., Duchovskis, P., Zukauskas, A., 2008. In vitro culture ofChrysanthemum plantlets using light-emitting diodes. Open Life Sci. 3,161–167, http://dx.doi.org/10.2478/s11535-008-0006-9.

Leduc, N., Roman, H., Barbier, F., Péron, T., Huché-Thélier, L., Lothier, J.,Demotes-Mainard, S., Sakr, S., 2014. Light signaling in bud outgrowth andbranching in plants. Plants 3, 223–250, http://dx.doi.org/10.3390/plants3020223.

Li, S., Rajapakse, N.C., Young, R.E., 2003. Far-red light absorbing photoselectiveplastic films affect growth and flowering of chrysanthemum cultivars far-redlight absorbing photoselective plastic films chrysanthemum cultivars.HortScience 38, 284–287.

Lund, J.B., Blom, T.J., Aaslyng, J.M., 2007. End-of-day lighting with differentred/far-red ratios using light-emitting diodes affects plant growth ofChrysanthemum × morifolium Ramat. Coral Charm. HortSci. 42, 1609–1611.

McCree, K.J., 1971. The action spectrum, absorptance and quantum yield ofphotosynthesis in crop plants. Agric. Meteorol. 9, 191–216, http://dx.doi.org/10.1016/0002-1571(71)90022-7.

Nhut, D.T., Nam, N.B., 2010. Light-Emitting Diodes (LEDs): an Artificial LightingSource for Biological Studies, In: Van Toi, V., Khoa, T.Q.D. (Eds.), The ThirdInternational Conference on the Development of Biomedical Engineering inVietnam: BME2010, 11–14 January, 2010, Ho Chi Minh City, Vietnam. Springer

1 rticul

P

R

R

S

86 R. Dierck et al. / Scientia Ho

Berlin Heidelberg, Berlin, Heidelberg, pp. 134–139.10.1007/978-3-642-12020-6 33.

ierik, R., de Wit, M., 2014. Shade avoidance: phytochrome signalling and otheraboveground neighbour detection cues. J. Exp. Bot. 65, 2815–2824, http://dx.doi.org/10.1093/jxb/ert389.

ajapakse, N.C., Pollock, R.K., McMahon, M.J., Kelly, J.W., Young, R.E., 1992.Interpretation of light quality measurements and plant response in spectralfilter research. HortScience 27, 1208–1211.

ajapakse, N.C., Young, R.E., McMahon, M.J., Oi, R., 1999. Plant height control by

photoselective filters: current status and future prospects. HortTechnology 9,618–624.

ager, J.C., Smith, W.O., Edwards, K.L., 1988. Photosynthetic efficiency andphytochrome photoequilibria determination using spectral data. Am. Soc.Agric. Eng. 31, 1882–1889.

turae 218 (2017) 177–186

Shimizu, H., Ma, Z., Tazawa, S., Douzono, M., Runkle, E.S., Heins, R.D., 2006. Bluelight inhibits stem elongation of chrysanthemum. In: Acta Horticulturae.International Society for Horticultural Science (ISHS), Leuven, Belgium, pp.363–368, http://dx.doi.org/10.17660/ActaHortic.2006.711.50.

Singh, M.C., 2013. Effect of LEDs on Flower Bud Induction in Chrysanthemum 2,2823.

Smith, H., Whitelam, G.C., 1997. The shade avoidance syndrome: multipleresponses mediated by multiple phytochromes. Plant Cell Environ. 20,840–844, http://dx.doi.org/10.1046/j.1365-3040.1997. d01-104.x.

Snow, R., 1940. A hormone for correlative inhibition. New Phytol. 39, 177–184,http://dx.doi.org/10.1111/j.1469-8137.1940.tb07131.x.

Van Ieperen, W., 2012. Plant morphological and development responses to lightquality in a horticultural context. Wageningen, Netherlands In: Proc. VII Int.Symp. Light Hortic. Syst., 956, pp. 131–139.