ZOU Jie,ZHOU Cheng-bo,XU Hong,CHENG Rui-feng,YANG Qi-chang,LI Tao

1 Institute of Environment and Sustainable Development in Agriculture,Chinese Academy of Agricultural Sciences,Beijing 100081,P.R.China

2 Sonneteck Technology Co.,Ltd.,Xiamen 361000,P.R.China

Abstract Light-emitting diodes (LEDs) have been widely applied in the controlled environment agriculture,which are characterized by relatively narrow-band spectra and energetical efficiency.Most recently,the spectrum of Sunlike LEDs has been engineered and it closely resembles solar spectrum in the range of photosynthetic active radiation (PAR,400–700 nm).To investigate how plant growth responses to the spectrum of Sunlike LEDs,cucumber and lettuce plants were cultivated and their responses were compared with the conventional white LEDs as well as composite of red and blue LEDs (RB,R/B ratio was 9:1).We observed that although Sunlike LEDs resulted in a longer stem in cucumber,dry weight and leaf area were similar as those under RB LEDs,and significantly higher than those under white LEDs.Moreover,cucumber leaves grown under Sunlike and white LEDs showed higher photosynthetic capacity than those grown under RB LEDs.For lettuce,plants grown under Sunlike LEDs showed larger leaf area and higher dry weight than the other two treatments.However,the leaf photosynthetic capacity of lettuce grown under Sunlike LEDs was the lowest.In this context,the spectrum induced plant functions are species-dependent.Furthermore,the three types of LEDs show distinct light spectra and they are different in many aspects.Therefore,it is difficult to attribute the different plant responses to certain specific light spectra.We conclude that plants grown under Sunlike LEDs exhibit larger leaf area,which may be due to some specific spectrum distributions (such as more far-red radiation),and consequently are favorable for light interception and therefore result in greater production.

Keywords:cucumber,leaf photosynthesis,lettuce,plant morphology,Sunlike LEDs

1.lntroduction

Rising global food demands and the development of new technology provide challenges and opportunities for sustainable agriculture.Controlled environment agriculture(CEA) has attracted a great deal of attention,due to the fact that an increasing number of environmental factors can be actively controlled,which may increase desired production and quality.Light-emitting diodes (LEDs) are characterized by relatively narrow-band spectra and energetical efficiency,which are considered as suitable light sources for plant growth.Recently,applying LEDs in CEA is constantly expanding (Bantiset al.2018),and the light spectra for plant growth have primarily been selected based on the quantum yield of photosynthesis as measured by McCree(1972).A common feature of these LED light sources is that their spectra mainly concentrated on red and blue spectra that are different from the spectrum of natural daylight (Linet al.2013;Terfaet al.2013;Hernández and Kubota 2016;Kaiseret al.2019).However,plant growth is not merely determined by net photosynthetic rate that was driven mainly by red and blue spectra.Plants have evolved under natural broadband spectra.Therefore,apart from red and blue spectra which are the most efficient spectra for individual leaf photosynthesis,other specific parts of the spectrum also play functional roles for mediating plant growth.It is well known that far-red light can regulate plant photomorphogenesis (Smith 1982;Franklin and Whitelam 2005;Fankhauser and Batschauer 2016).Plants often show a shade-avoidance syndrome(SAS) in response to a higher fraction of far-red light,which is characterized by increasing elongation of stems and petioles,increasing in leaf expansion,and elevating leaf angles (Franklin 2008).These properties could facilitate canopy light interception and consequently plant growth(Zhanget al.2019).Green light can penetrate deeper into the crop,and consequently increase crop photosynthesis and light utilization in lower leaf layers (Johkanet al.2012).Kimet al.(2004) showed that increasing green light fraction from zero to 24% resulted in～40% higher biomass production in lettuce.Hogewoninget al.(2010a)reported that cucumber plants grown under artificial solar spectrum resulted in a higher biomass production and a more natural appearance,in comparison with plants grown under fluorescent tubes and high-pressure sodium lamps of which spectra does not resemble that of natural daylight.In this context,developing LEDs with a composite light spectrum that resembles the solar spectrum may stimulate plant production in CEA.

Production management in CEA often requires labor work and people are unavoidably exposed to the monochromatic red and blue light or other unnatural light spectra,which may induce symptoms of discomfort.For instance,blue light exposure may result in a sharp decline or even total loss of human visual acuity (Behar-Cohenet al.2011).In addition,blue light inhibits melatonin secretion,which affects sleep,regulates jet lag and has a significant impact on biologic rhythm (Wrightet al.2001).Therefore,it is significant to develop a light source that not only stimulates plant growth,but also minimizes the harmfulness to workers in the CEA.

The research on LEDs has yielded technologies that make LED modules more energetically efficient and versatile as lighting systems (Morrow 2008;Rehmanet al.2017).Most recently,the spectrum of Sunlike LEDs has been engineered and it closely resembles solar spectrum in the range of photosynthetic active radiation (PAR,400–700 nm)(Fig.1).This is essential for minimizing the impact of monochromatic light spectrum on biologic rhythm for human beings.The current study aims at investigating how plant growth responses to the spectrum of Sunlike LEDs,they were compared with the widely-used white LEDs as well as composite of red and blue LEDs.Lettuce (Lactuca sativaL.cv.Tiberius) and cucumber (Cucumis sativus cv.Zhongnong 16) plants were cultivated under a fully controlled environment.Plant growth and morphological properties,as well as leaf photosynthetic properties,were investigated.

2.Materials and methods

2.1.Plant materials and growth conditions

Fig.1 Relative photon flux density of Sunlike LEDs and solar radiation measured at noon (12:00–13:00) of clear day in summer in Beijing (A) and of white LEDs and mixture of red and blue LEDs (RB) (B).

Cucumber seeds (Cucumis sativuscv.Zhongnong 16,Zhongshu Seed Group,China) were germinated in rockwool plugs (Grodan,Roermond,the Netherlands),and transferred to rockwool cubes (10 cm×10 cm×7 cm;Grodan) upon unfolding of the second leaf (1 week after sowing).Seedlings were grown in an environmentallycontrolled growth chamber,where three cultivation frames with 150 cm×70 cm×60 cm (length×width×height) each were installed,each frame with a different light treatment.In the growth chamber,CO2concentration was close to ambient,day/night temperature was 23/21°C,and relative humidity was～65%.

Lettuce (Lactuca sativaL.cv.Tiberius,Rijk Zwaan,The Netherlands) seeds were sown in plastic germination trays(33 cm×26 cm×4 cm) filled with sponge cube blocks (3 cm×3 cm×3 cm) and water.Upon unfolding of the second leaf(2 weeks after sowing),the seedlings were transferred to the same growth chamber as the cucumber.Lettuce plants were cultivated in a customized hydroponic system with density of 33 plants per m2.Hoagland solution was applied for plant cultivation.During the experiment,cucumber and lettuce plants were grown in succession.

Three types of LED light tubes,Sunlike LEDs(Sonneteck Technology Co.,Ltd.,Xiamen,China),white LEDs and mixture of red and blue LEDs (RB,with red:blue ratio 9:1) (iGrowLite Co.,Ltd.,Guangzhou,China),were mounted in each of the cultivation frames,60 cm above the growth plate.Light spectra of the three types of LEDs as well as the solar light spectra were presented in Fig.1,which were measured using spectroradiometer (Avaspec-2048CL,Avantes,Apeldoorn,The Netherlands).The photon flux density (PFD) of the three treatments measured at the 10 cm above the canopy of lettuce was～210 μmol m–2s–1,and was～230 μmol m–2s–1for cucumber.The photostationary state of phytochrome (PSS) was calculated according to Sageret al.(1988),and were 0.81,0.84,0.88 in the Sunlike,white,and RB LED treatments,respectively (Table 1).The photoperiod was 16 h continuing from 06:00 to 22:00.The PFD was monitored daily by a spectroradiometer(Avaspec-2048CL,Avantes,Apeldoorn,The Netherlands).To avoid interaction,light reflection curtains (OMS film Co.,Ltd.,Xiamen,China) were fixed between treatments.

2.2.Growth and morphology analysis

For growth and morphology analysis,destructive measurements were conducted 15 days after treatment.Six plants from each treatment were randomly selected,and fresh and dry weights of plant organs (leaves,roots and stems) were determined.Plant organs were dried for at least 48 h at 80°C in a ventilated oven (DHG-9625A,YIHENG,Shanghai,China).Leaf area was measured with a leaf area meter (LI-3100C,Li-Cor Inc.,Lincoln,USA).Specific leaf area (SLA) was calculated by dividing the leaf area by leaf dry weight.Stem length of cucumber was measured with a vernier caliper (SANTO,Shanghai,China).

2.3.Leaf photosynthesis measurements

Leaf photosynthesis measurements were taken 1–2 days before destructive measurements,which were performed on the upper-most fully expanded leaves,between 8:30 and 15:30.The measurements were performed with the LI-6400XT photosynthesis system (Li-Cor Biosciences,Lincoln,Nebraska,USA) with the leaf chamber fluorometer(Li-Cor Part No.6400-40,2 cm2).The adaxial side of the leaf was illuminated by a mixture of red (90%) and blue (10%)LEDs in the fluorometer,with peak intensities of 635 and 465 nm,respectively.The starting photosynthetic photon flux density (PPFD) was 250 μmol m–2s–1,followed by 1 500 μmol m–2s–1.At each PPFD,the measurements were taken when the photosynthetic rate (Pn) reached a steady state (after about 5–10 min).During the measurement,Pn,stomatal conductance (Gs),fluorescence yield under actinic light (Fs),maximal (Fm´) and minimal (Fo´) fluorescence yield of a light-adapted leaf were also measured at each PPFD,which were used to calculate PSII operating efficiency(ФPSII).In the leaf chamber,CO2partial pressure was 400 μmol mol–1,leaf temperature was 23°C,leaf-to-air vapor pressure deficit (VPDleaf–air) was maintained between 0.7–1.0 kPa,and the flow rate of air through the system was 500 μmol s–1.These measurements were taken on six plants in each treatment.

2.4.Leaf biochemical component measurements

One day before destructive measurements,the uppermost fully expanded leaves were collected in vials,flash-frozen in liquid nitrogen and transferred to a freezer(–80°C) for storage.These samples were used for measuring chlorophyll contents.Leaf pigments were extracted with 95%ethanol and the absorbance was measured at 665,649and 470 nm using a UV-Vis spectrophotometer (UV-1800,Shimadzu,Japan).The chlorophyll concentrations were calculated using equations derived by Wellburn (1994).Dry leaf samples from destructive measurements were collected and powdered using a grinder for determining leaf total nitrogen with an elemental analyser (vario PYRO cube,Isoprime,UK).

Table 1 The fraction of spectrum distribution and photostationary state of phytochrome (PSS) of three types of LEDs

2.5.Statistics

Treatment effects on the measured plant parameters were evaluated by one-way ANOVA using SPSS 20 (SPSS Inc.,Chicago,IL).Six plants from each treatment were randomly selected.Data were firstly tested for normality (Shapiro–Wilk test) and homogeneity of variances (Levene’s test).Subsequently,fisher’s protected least significant differences(LSD) at 95% confidence were determined.

3.Results

3.1.Plant morphology,dry weight and partitioning

Plants grown under white,RB,and Sunlike LEDs showed distinct visual appearance (Fig.2).For both cucumber and lettuce,the plants grown under Sunlike LEDs with higher visual appearance developed obviously faster than those grown under white and RB LEDs.Cucumber plants grown under RB LEDs had a slightly bigger appearance than those grown under white LEDs,while lettuce plants showed similar visual appearance under white and RB LEDs.

The differences in plant morphology,fresh and dry weights are shown quantitatively in Table 2.For cucumber,plants grown under RB and Sunlike LEDs had similar shoot dry weight and leaf area,and they were significantly higher than those grown under white LEDs.In accordance with the differences in visual appearance,stem length of cucumber plants was 46 and 21% higher under Sunlike LEDs compared with the plants grown under white and RB LEDs,respectively.Furthermore,plants grown under Sunlike LEDs showed significantly lower dry weight partitioning to leaf and higher dry weight partitioning to stem,compared with plants grown under white and RB LEDs.SLA of cucumber plants was not affected by the different light spectra.

For lettuce,plants grown under white and RB LEDs had similar total dry weight and leaf area,and fresh weight was 24 and 20% lower than those grown under Sunlike LEDs,respectively.SLA was the highest in plants grown under Sunlike LEDs and the lowest in plants grown under RB LEDs.Dry weight partitioning was not affected by the light treatments.

Fig.2 Cucumber (upper) and lettuce (lower) plants grown under white,mixture of red and blue (RB),and sunlike LEDs 15 days after transplant.Scale bar=10 cm.

Table 2 Plant fresh (FW) and dry weights (DW),leaf area,stem length,specific leaf area (SLA),and dry weight partitioning(DWP) to different plant organs under white,red and blue (RB),and Sunlike LEDs

3.2.Leaf photosynthetic properties

Cucumber and lettuce plants showed distinct leaf photosynthetic properties in response to the tested light spectra (Fig.3).Specifically,at both low and high PPFDs,cucumber leaves grown under RB LEDs had significantly lowerPnin comparison with leaves grown under white and Sunlike LEDs (Fig.3-A),ФPSIIshowed the same response asPnat high PPFD,while at low PPFD,significant differences were observed only between RB and white LED treatments (Fig.3-C).ForGs,significant difference was only occurred at high PPFD (Fig.3-E).

For lettuce,Pnof leaves grown under Sunlike LEDs was 13 and 16% lower than that under white LEDs and RB LEDs,respectively,at high PPFD,while it was similar between treatments at low PPFD (Fig.3-B),and the difference inФPSIIwas negligible (Fig.3-D).In accordance withPn,Gsof lettuce leaves grown under Sunlike LEDs was also apparently lower than the other treatments (Fig.3-F).

Fig.3 Net leaf photosynthetic rate (Pn),PSII operating efficiency (ФPSII),and stomatal conductance (Gs) of cucumber and lettuce plants grown under white,mixture of red and blue (RB),and Sunlike LEDs in response to low (250 μmol m–2 s–1) and high (1 500 μmol m–2 s–1) photosynthetic photon flux density (PPFD).Error bars show SE (n=6).Different letters show statistically significant differences (P<0.05)

3.3.Leaf biochemical components

For cucumber,leaves grown under Sunlike and RB LEDs had significantly lower total nitrogen content and chlorophyll content than the treatment with white LEDs,and Chla/bratio was not affected by the treatment (Table 3).For lettuce,leaves grown under RB LEDs had significantly higher total nitrogen content than the other treatments,while treatment had no effect on chlorophyll contents and Chla/bratio.

4.Discussion

Exploring appropriate light recipe is one of the constant topics in the controlled environment agriculture.This study compared the growth of cucumber and lettuce plantsunder three types of LEDs which had different light spectra(Fig.1).It is conspicuous that plants grown under Sunlike LEDs showed greater plant size and biomass accumulation compared with compared with the plants grown under white LEDs.In comparison with responses to RB LEDs,however,such stimulating effects occurred in lettuce but not in cucumber.This indicates that spectrum responses of plant growth are species-dependent.Plant biomass accumulation to a great extent depends on the canopy light interception which correlates with the shoot architecture (Hogewoninget al.2010a;Sarlikiotiet al.2011).Plants grown under Sunlike LEDs showed an open canopy with larger leaf area(Fig.2),which is favorable for light interception and plant photosynthesis,consequently resulted in more biomass accumulation.

Table 3 Leaf chemical components of plants grown under white,mixture of red and blue (RB),and Sunlike LEDs

Plant morphogenesis is highly sensitive to its growth light spectrum (Hogewoninget al.2010a).The three types of LEDs used in this study showed distinct light spectra and they are different in many aspects (Table 1).Therefore,it is difficult to attribute the different plant morphological responses to certain specific light spectra.It is well recognized that blue and far-red light strongly mediates plant morphology (Demotes-Mainardet al.2016;Hernández and Kubota 2016;Park and Runkle 2017).Among the three types of LEDs,both Sunlike and white LEDs emit continuous light spectra,and the fraction of far-red radiation in the Sunlike LEDs was obviously higher than the white LEDs(Table 1),while far-red and green radiation was absent in the RB LEDs.Furthermore,blue radiation in the RB LEDs was almost three times lower than the others,these discrepancies could cause different responses in plant growth and development.

Many studies showed that higher fraction of far-red light resulted in taller plants with larger leaves,and more biomass partitioning to stems at the expense of partitioning to leaves (Franklin 2008;Demotes-Mainardet al.2016).This phenomenon was also observed in the plants grown under Sunlike LEDs which emit higher fraction of far-red light than the white LEDs (Table 2).The far-red light stimulated leaf expansion is particularly important for the young plants which often have limited effective leaf area to capture radiation for photosynthesis.This might be an important explanation for the higher biomass production under Sunlike LEDs than white LEDs.Another possible reason could be the differences in the fraction of green and red light.Specifically,Sunlike LEDs emit～10% lower green light and～10% higher red light than white LEDs,and this may result in a significantly higher light use efficiency in Sunlike LEDs than white LEDs,because red light is the most efficient light spectrum for leaf photosynthesis,while green light is much less efficient as it is only absorbed weakly by chlorophylls and pigments (McCree 1972;Hogewoninget al.2012).

In comparison with RB LEDs,the stimulating effect of Sunlike LEDs on plant growth occurred in lettuce,but not in cucumber (Table 2).This is possibly because the spectrum induced plant functions are species-dependent.Previous study reported that blue light is qualitatively required for normal leaf photosynthetic functioning and quantitatively mediates leaf responses (Hogewoninget al.2010b;Kaiseret al.2019).Hernández and Kubota (2016) showed that increasing the fraction of blue light retards stem elongation and leaf expansion in cucumber.Similar results were found that blue light with increasing intensity controls hypocotyl extension in vegetable seedlings (Kwacket al.2015).Therefore,the shorter stem length and smaller leaf area in the white LEDs than RB LEDs can be attributed to the greater fraction of blue light in white LEDs.Although Sunlike LEDs also have higher fraction of blue light than RB LEDs,the stem of cucumber in the Sunlike LEDs was longer than in RB LEDs (Table 2),such inconsistent with general knowledge could be that the extent of blue light effect on stem retardation was less than that of far-red light effect on stem elongation.For biomass production of cucumber,Sunlike LEDs do not have advantageous in comparison with RB LEDs (Table 2),because the latter emit about three times more red radiation (Table 1) that is more favorable for leaf photosynthesis.For lettuce,however,the remarkably higher fraction of red light in the RB LEDs did not result in higher biomass production compared with the others,this may partly be explained by the green light effect in the white LEDs and Sunlike LEDs,as previous study reported that adding green light to red and blue light stimulates lettuce growth,because green light can penetrate the plant canopy and potentially increase plant growth by increasing photosynthesis from the lower canopy leaves(Johkanet al.2012).Interestingly,Hernández and Kubota(2016) showed that addition of green light did not have any effect on the growth and morphology of cucumber,and they concluded that the effect of green light on plant growth is species-specific.In this context,specific explanation of different responses between cucumber and lettuce to the three types of LEDs is complicated.

Leaf physiological properties highly correlate with the spectrum of growth radiation.A higher fraction of blue light is generally associated with the development of‘sun-type’ leaves,which are characterized with a high leaf photosynthetic capacity (Hogewoninget al.2010b).In treatments of Sunlike and white LEDs,therefore,the greater fraction of blue light explains the higher photosynthetic capacity of cucumber leaves compared with plants grown under RB LEDs (Fig.3),which correlates with the responses of leaf total nitrogen content (Table 3).For lettuce,however,leaf photosynthetic capacity of plants grown under Sunlike LEDs was significantly lower than the other treatments,and it was similar between white and RB LEDs (Fig.3).This phenomenon most likely resulted from the interaction between blue and far-red light in the Sunlike and white LEDs,because our earlier study showed that lettuce is very sensitive to far-red light,and adding far-red light could downregulate leaf photosynthetic capacity that correlates with a lower stomata conductance and leaf total nitrogen content (Zouet al.2019).It is not surprising that the lower leaf photosynthetic capacity (Fig.3) does not match with the higher biomass production in lettuce (Table 2),because the leaf photosynthetic capacity reflects the photosynthetic potential at leaf level,which was measured at 1 500 μmol m–2s–1PPFD that did not exist in the growth environment.This further illustrated that the higher biomass production was attributed to the higher light interception induced by the plant architecture adaptation.

5.Conclusion

It can be concluded that plants grown under Sunlike LEDs exhibit an open plant canopy,which is favorable for light interception and consequently results in greater production.However,the spectrum-induced plant functions are speciesdependent and specific explanation on the mechanisms underlying the wavelength dependency of the responses is complicated.Our results suggested that Sunlike LEDs can be considered as a promising light source for controlled environment agriculture,particularly for very specialized applications,such as in space or laboratory.

Acknowledgements

This work was financially supported by the National Key Research and Development Program of China(2017YFB0403902),the National Natural Science Foundation of China (31872955) and the Central Publicinterest Scientific Institution Basal Research Fund,China(BSRF201911).