Quantifying the Effects of Light Quantity and Quality on Culinary Herb Physiology Alexander Gaston Litvin-Zabal Iowa State University

Iowa State University Capstones, Theses and Graduate Theses and Dissertations Dissertations

2019 Quantifying the effects of light quantity and quality on culinary herb physiology Alexander Gaston Litvin-Zabal Iowa State University

Follow this and additional works at: https://lib.dr.iastate.edu/etd Part of the Agriculture Commons, and the Horticulture Commons

Recommended Citation Litvin-Zabal, Alexander Gaston, "Quantifying the effects of light quantity and quality on culinary herb physiology" (2019). Graduate Theses and Dissertations. 17042. https://lib.dr.iastate.edu/etd/17042

This Dissertation is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Graduate Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. Quantifying the effects of light quantity and quality on culinary herb physiology

Alexander Gaston Litvin-Zabal

A dissertation submitted to the graduate faculty

in partial fulfillment of the requirements for the degree of

DOCTOR OF PHILOSOPHY

Major: Horticulture

Program of Study Committee: Christopher J. Currey, Major Professor William R. Graves Ajay Nair Roberto G. Lopez Lester A. Wilson

The student author, whose presentation of the scholarship herein was approved by the program of study committee, is solely responsible for the content of this dissertation. The Graduate College will ensure this dissertation is globally accessible and will not permit alterations after a degree is conferred.

Iowa State University

Ames, Iowa

2019

DEDICATION

I dedicate this work to my family, from whom my long academic endeavors have separated us by time and space. To my father, Dr. Miguel E. Litvin, who pushed me to go farther, and inspired the technological aspects of my work. To my mother, Anamaria

Zabal Litvin, who bestowed upon me my penchants for agricultural science. To my brothers, Arturo and Gustavo Litvin, who kept me grounded, and inspired me every day with their own amazing accomplishments. To my aunt Maria Elena Zabal, who provided me support and perspective from her own work in academia. Additionally, I dedicate this to my love, Krisa Ann Lewis, who stood by me for reasons unknown, while I worked through the hardest of days. Finally, to Shelby Litvin, who has known no life other than my academic days, and who celebrated the end of each day of work with play. Each one of you imbued me with your love, and from it, the strength to grow, thrive, and blossom as a plant… scientist. iii

TABLE OF CONTENTS

Page

ACKNOWLEDGMENTS ...... v

ABSTRACT ...... vi

CHAPTER 1: INTRODUCTION ...... 1

CHAPTER 2: INCREASING DAILY LIGHT INTEGRALS ENHANCES GROWTH OF EIGHT CULINARY HERBS GROWN HYDROPONICALLY ...... 24 Abstract ...... 24 Materials and methods ...... 26 Plant materials and culture ...... 26 Daily light integral treatments ...... 28 Data collection and calculation ...... 29 Experimental design and statistical analyses...... 29 Results ...... 30 Discussion ...... 31 Conclusion ...... 35 References ...... 35

CHAPTER 3: EFFECTS OF SUPPLEMENTAL LIGHT SOURCE ON BASIL, DILL, AND PARSLEY GROWTH, MORPHOLOGY, AROMA, AND FLAVOR ...... 42 Abstract ...... 42 Materials and methods ...... 47 Plant materials and propagation ...... 47 Hydroponic culture and greenhouse environment...... 47 Supplemental lighting treatments ...... 49 Plant growth data collection and calculation ...... 49 Aroma and flavor analyses ...... 51 Experimental design and statistical analyses...... 52 Results ...... 52 Basil ...... 52 Dill ...... 54 Parsley ...... 54 Discussion ...... 55 Conclusion ...... 61 References ...... 61

CHAPTER 4: BLUE LIGHT FRACTION EFFECTS GROWTH, MORPHOLOGY, BIOMASS PARTITIONING, METABOLISM, AND FLAVONOID ACCUMULATION OF BASIL ...... 79 Abstract ...... 79 Materials and methods ...... 83 Plant materials and culture ...... 83 iv

Hydroponic culture and controlled environment conditions ...... 84 Sole-source lighting treatments ...... 85 Plant growth data collection and calculation ...... 85 Chlorophyll content and phenolic concentration analysis ...... 87 Data calculated ...... 88 Experimental design and statistical analyses...... 89 Results ...... 89 Morphological effects...... 89 Gas exchange ...... 90 Nutrient analysis and flavonoids ...... 91 Discussion ...... 93 Conclusion ...... 100 References ...... 101

CHAPTER 5: GENERAL CONCLUSIONS...... 120 Chapter 2: Modeling of culinary herbs to daily light integral ...... 120 Chapter 3: Supplemental light source spectra ...... 121 Chapter 4: Blue light fraction effect on basil plant functions ...... 122 Future suggestions ...... 124 References ...... 125 v

ACKNOWLEDGMENTS

I would like to thank my advisor, Dr. Christopher J. Currey, for the years of service and contribution to this dissertation, and that of my committee members Drs.

William R. Graves, Ajay Nair, Roberto G. Lopez, and Lester A. Wilson whom have generously offered their help during the course of my studies. Furthermore, all great research is done in collaboration of the help of others, and the greenhouse manager Peter

Lawlor is someone who plays a crucial role in securing the success of any research project in the Horticulture department.

I would also like to thank my undergraduate helpers Brianna Vest and Erica

Schlichte for their assist in conducting my research. Also, I owe a great deal of gratitude and thanks to the support of my fellow graduate students, without whom, these past 4 years would have been a bleak process.

Finally, I would like to acknowledge my family, to whom this work is dedicated, whose support in all matters possible have continuously provided me with the will and strength to persevere.

ABSTRACT

Lighting influences plant growth and development on a quantity and quality basis, promoting the production of both primary and secondary metabolites. While the amount of light necessary for culinary herb production is species-specific, with growth curves for increasing daily light integrals (DLI), the specific spectra of incidental light absorbed plays a role in photosynthetic efficiency, biomass partitioning, and secondary metabolites related health and nutrition. The two primary objectives of this research were to 1)

Quantify morphological growth of culinary herbs under increasingly higher DLI, and 2)

Identify the morphological, physiological, and phytochemical responses of selected herbs to the spectral composition of light sources and blue light fraction from either supplemental or sole-source lighting. Biomass increased with increasing DLI, and the relationship was either linear or quadratic, depending upon saturating DLI. Additionally, the proportions of light spectra affected plant height, biomass, gas exchange, photosynthetic efficiency, and phenolic accumulation by alteration of blue light fraction, or use of broad-spectrum lighting. By identifying responses of plants to light quantity and quality, the goal of this research was to 1) Provide information on light quantity optimization in food crop production, and 2) Improve the quality of food crops produced by emphasizing the promotion of plant photoprotective compounds that increase both production efficiency and food nutrition.

CHAPTER 1: INTRODUCTION

Culinary herbs

Culinary herbs are selected for what they can offer to dishes, adding aroma and flavor to foods that would otherwise be lacking without spice. They are popular due to their enhancement of flavor and aroma of food (Brown, 1991; Pripdeevech et al., 2010; Simon et al., 1999), historic cultural value (Cook and Samman, 1996; Justesen and Knuthsen, 2001; Paton, 1992), and ornamental appeal (Morales and Simon, 1996). Their appeal is due in part to pigmentation, aromatic traits, and flavor characteristics (Cook and Samman, 1996; Justesen and Knuthsen,

2001; Stefova et al., 2003). Most common culinary herbs were initially characterized, used, and cultivated within the Eurasian and African regions (van Wyk, 2014). Herbs such as basil

(Ocimum basilicum), sage (Salvia officinalis), mint (Mentha sp.), thyme (Thymus vulgaris), oregano (Origanum vulgare), rosemary (Rosmarinus officinalis), and lavender (Lavandula officionalis) are related and classified from the same family (Lamiaceae). Similarly, Apiaceae is another common family for culinary herbs including parsley (Petroselinum crispum), cilantro

(Coriandrum sativum), dill (Anethum graveolens), and celery (Apium graveolens), among others.

Among these culinary herbs, basil is a robust leafy annual originating from tropical Asia, and popular among growers and consumers alike (van Wyk, 2014). Purple or opal basil is unique for its high anthocyanin content, which contributing to its colorful foliage and nutritional value.

Much of the characteristic aroma and flavor profile of its essential oils and phenolic compounds are attributed to linalool, eugenol, estragole, and 1,8-cineole (Pripdeevech et al., 2010; Simon et al., 1999; Telci et al., 2006). Of the essential oils, linalool and 1,8-cineole are most associated with pungent aromas (Pripdeevech et al., 2010). Dill (Anethum graveolens), an annual or biennial herb, is popular as a dried spice used as a garnish (Kiple and Ornelas, 2000). Originating 2 from southwestern Asia and Europe, dill is reportedly helpful for digestive ailments (Kiple and

Ornelas, 2000). Notable compounds include isorhamnetin, kaempferol, dill ether, dillapiole, (-)-

α-phellandrene, (+)-(4S)-Carvone, and (+)-limonene (Blank and Grosch, 1991; Radulescu et al.,

2010; Reichert et al., 1998; USDA, 2011). Parsley (Petroselinum crispum), with large, flat palmately compound leaves, is a garnishing herb and often used as a spice (Farrel, 1999). It originates from the areas around the Mediterranean, southern Europe, and western Asia

(Harborne and Baxter, 2001). Parsley contains high apigenin, apiole, myristicin, 1,3,8-p- menthatiene, and tetramethoxyallylbenzene content (Harborne and Baxter, 2001; USDA, 2011).

Aroma and flavor accentuation in culinary herbs is considered desirable for consumers (Cook and Samman, 1996; van Wyk, 2014) and are thus a central focus for improvement in production.

Controlled environment agriculture

Food crops produced under protected culture are an important sector in food production that has grown over the decades (USDA, 1998, 2014). Horticultural food production is worth just over $1.4 billion, with food crops grown under protected culture, such as in greenhouses, valued at approximately $800 million in 2012 (USDA, 2014), a 260% increase since 1997 (USDA,

1998). Protected culture allows for more flexibility in location, mitigating seasonal variations and availability of arable land. This is supported by the 44% increase in crop production in the sector from the previous census reports (2009) up to the most recent (2014) of 5.2 million cwt, suggesting increased investment in horticultural operations in recent years (USDA, 2009, 2014).

Specifically, herb production has attracted significant investment in horticulture with a recent value over $96 million (USDA, 2014).

These production statistics reflect the growing demand for food within protected climates. Part of the increase in this sector’s food production is attributed to consumer demand for increased produce quality, access to produce in urbanized or food deserts (Brown and Miller, 3

2008; Tixier and de Bon, 2006), and disruption to supply chains from increasingly volatile weather (Amann et al., 2011; Despommier, 2011). Small-scale CEA operations embedded within urban areas have contributed to the increase in food production (Brown and Miller, 2008; Tixier and de Bon, 2006). Ultimately, CEA provides the benefit of facilitating continued year-round production for growers, while maintaining a reliable marketplace for consumers.

Herb cultivation in greenhouses and other controlled environment agriculture (CEA) facilities is dependent on several factors, including production systems (Walters and Currey,

2015), mineral nutrition (De Pascale et al., 2006), air temperature (Chang et al., 2005), CO2 concentrations (Zobayed and Saxena, 2004), and light intensity and duration (Beaman et al.,

2009; Chang et al., 2008). All of these factors allow for controllable and predictable climates year-round, maintaining consistent production and quality regardless of season or location

(Brown and Miller, 2008).

Hydroponic production

Hydroponic production emerged as a viable platform for regulating mineral nutrient concentrations while eliminating the need for soils (Jensen, 1999; Jensen, 2002). Familiar hydroponics systems include water-culture systems such deep-flow technique (DFT) and nutrient-film technique (NFT). Deep-flow technique hydroponics, also known as a raceway system, deep-water culture, raft-culture, and others, are composed of a large reservoir filled with nutrient solution and raft. Plants are placed into holes in the raft and kept suspended over or floating on nutrient solution while roots grow directly into the nutrient (Jensen, 2002). These

DFT systems, due to their simplicity, have fewer parts that can breakdown or fail, adding to its reliability. Conversely, the large, shared root zone increases the potential for spreading diseases quickly. Additionally, DFT nutrient solutions require aeration to prevent root hypoxia. For NFT 4 systems, nutrient solution is continuously provided to plants through sloped troughs, collected in a reservoir, pH and electrical conductivity (EC) adjusted, and recirculated. Due to the amount of parts and equipment used, NFT systems require more maintenance than DFT systems.

Air temperature

Temperature management in CEA is an important factor in stabilizing environmental conditions conducive to plant growth and development. Temperature influences the rate of enzyme activity in plants (Berry and Björkman, 1980) and is a primary determinant of plant growth and development (Ritchie and NeSmith, 1991). While temperature affects vegetative growth and development (Karlsson and Heins, 1992), temperature also influences nutrient uptake

(Turner and Lahav, 1985), gas exchange (Crafts-Brandner and Salvucci, 2000), and metabolism

(Airaki et al., 2012; Wolf et al., 1991). Low temperatures can diminish chlorophyll content (Koç et al., 2010), reduce enzyme activity (Berry and Björkman, 1980), and diminish developmental growth compared to plants grown at warmer temperatures, although supra-optimal temperature may also result diminished development and, ultimately, death (Lipiec et al., 2013). As temperatures increase above species-specific optimal temperatures, ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBisCO) deactivation increases, resulting in reduced photosynthesis

(Pn), and increased photorespiration (Crafts-Brandner and Salvucci, 2000). Under supra-optimal temperatures, reductions to Pn can be mitigated through control of other environmental parameters, such as increasing carbon dioxide (CO2) concentrations (Busch and Sage, 2017).

These temperature responses and requirements can change across species, making a temperature control an important factor within CEA for stabilizing environmental conditions.

Carbon dioxide

Carbon dioxide is taken up by plants through gas exchange, and is a primary component of the overall processes for photosynthesis (Pn) through fixation by RuBisCO (Crafts-Brandner and Salvucci, 2000). Because of poor affinity to CO2 by RuBisCO (Busch and Sage, 2017), binding of Ribulose-1,5-bisphosphate (RuBP) to O2 leads to photorespiration, and reduces energy output by Pn (Herrick and Thomas, 1998). Although C4 plants can transport CO2 to the bundle sheath, thereby reducing photorespiration by concentrating CO2 (Huber and Edward,

1975), carbon fixation in C3 plants occurs upon uptake through stomata and is thus subject to atmospheric and intracellular CO2 concentrations. Busch and Sage (2017) reported raising CO2 concentrations increased net Pn, rate of carboxylation, and overall efficiency of Pn, particularly under increasing temperatures. Fixation of CO2 is important as Pn, and subsequent energy produced from RuBisCO, greatly affect plant growth and development. Wheeler et al. (1991)

–1 reported that increasing CO2 concentrations from 350 to 1000 µmol∙mol increased dry mass accumulation of potato (Solanum tuberosum L.) leaves, and tubers. Furthermore, increasing CO2 concentrations can increase the light saturation point in some species (Herrick and Thomas,

1998; Zimmerman et al., 1997), thereby increasing overall photosynthetic capacity. With respect to photosynthesis, CO2 can thus be classified as regulatory parameter for lighting, influencing the efficiency and Pn capacity. Through the careful control of CO2 concentrations within controlled environments, studies are thus better able to isolate lighting effects on Pn, adding to the precision of CEA.

Light

Light intercepted by a leaf may be absorbed, reflected, or transmitted through the canopy.

Absorbed light is used for photochemical reactions (Pn), non-photochemical quenching through 6 dissipation by the xanthophyll cycle, or is re-emitted by fluorescence (Chen et al., 2014; Genty et al., 1989; Ross, 1967). For Pn, plant responses are primarily associated with the amount of light

(quantity), the spectra of that light (quality), and the duration of each lighting period

(photoperiod). Light quantity is the amount of photons for Pn, non-photochemical quenching, and fluorescence, resulting in growth and development as well as stress response (Chen et al.,

2014; Darko et al., 2014; Genty et al., 1989; Ross, 1967). Instantaneous light intensity is the quantum measurement of radiation presented as micromoles of photons per square meter every second (μmol∙m–2∙s–1). The aggregated instantaneous light quantity over the course of a day is the daily light integral (DLI), which strongly affects crop yield (Faust and Logan, 2018). Herbs cultivated under ambient light are subject to variations in light intensity and duration of the region, season, and weather (Faust and Logan, 2018; Seginer et al., 2006), which then affect biomass yield (Albright et al., 2000). Ambient light transmission into a greenhouse is approximately 35-75% due to greenhouse glazing and superstructure (Fisher and Donnelly,

2001), further reducing DLI inside greenhouses. Thus, supplemental lighting is used to increase photosynthetic light in greenhouses.

Light quality refers to the specific wavelengths of radiation. Wavelengths from approximately 400 to 700 nm provide lighting for plants, known also as photosynthetically active radiation (PAR). Models for the action spectra of light absorbed by plants are reported by Hoover

(1937), McCree (1972), and Inada (1976), illustrating changes in Pn across wavelengths (Sager et al., 1988). The relative quantum efficiency (RQE) of light is the absorption efficiency of incidental wavelengths across a plant’s action spectrum (Sager et al., 1988). Massa et al. (2008) reported increased production of secondary metabolites or biomass at wavelengths of 430 to 450 7 nm (blue light) or 640 to 660 nm (red light), respectively, corresponding to peak absorption ranges of chlorophylls a and b.

While light quantity and quality influence the rate and efficiency of Pn, photoperiod refers to day length. This adjusts a plant’s circadian rhythm, responses to seasonal changes, and initiates phases in its lifecycle transitioning into processes such as dormancy and flowering. The primary sensing mechanism for photoperiod is phytochrome (Taiz and Zeiger, 2010), specifically phytochromes red (Pr) and far-red (Pfr) (Schafer and Nagy, 2006; Taiz and Zeiger, 2010). In the absence of phytochrome sensing, cryptochromes also play a role in regulating the circadian rhythm in plants (Thomas, 2006). Blue light sensing by plants has been shown by Millar et al.

(1995) to regulate photoperiod responses in plants when phytochrome synthesis was knocked out, with the reporter gene construct luciferase used to monitor mutation effect of the absence of

Pr and Pfr. While manipulating photoperiod is an essential factor in producing numerous ornamental crops in controlled environments to control flowering, it is not intensively managed for CEA food crop production. However, light quantity and quality are essential aspects of CEA lighting for food crop production.

Electric light

Electric light can increase the DLI for greenhouse crops under light-limited conditions

(supplemental lighting), or acts as the sole light source for indoor crop production (sole-source lighting). With supplemental lighting, growers increase yields from increasing plant growth

(Currey et al., 2012) or shortened crop production times allowing for additional growth cycles

(Fisher and Donnelly, 2001; Styer and Koranski, 1997).

Traditionally, gas-filled, broad-spectrum lamps, such as high-pressure sodium (HPS) or metal halide, have been used as supplemental lighting because they provide broad-spectrum 8 radiation across the full range of PAR; however, electrical inefficiencies resulting from thermal radiation increases operating costs (Hogewoning et al. 2007; Trouwborst et al., 2010) and prevent lamp placement in close proximity to plants (Gomez et al., 2013).

Efficient light sources that last longer and focus output in those spectra promoting growth offer possible reductions in costs for equipment and operation (Li and Kubota, 2009; Morrow,

2008). While many lamps exist today for use as supplemental lighting, light-emitting diode

(LED) lamps have emerged as advantageous due to low thermal output, making them easy to position in small enclosures, while providing specific wavelength most used in Pn (Randall and

Lopez, 2015). Illumination by LEDs is discrete, with specific wavelengths providing spectra that can more precisely target plant photoreceptors than HPS or metal halide lamps. The narrow- spectra lighting provided by modern high-intensity LEDs can be in blue (400–500 nm), green

(500–600 nm), red (600–700 nm), ultra-violet (UV; 320–400 nm), and far-red (700–800 nm) wavelengths, with specific wavelengths within each color available on the market (Stutte, 2009.

Though preliminary research into LED technology was initially done in the late 1980’s when high-intensity LEDs sufficient for Pn were first produced (Morrow, 2008), technology was still primitive due to poor output of blue light sources; successful LED technology was not reported until the 1990’s (Miyashita et al., 1995). Since then, advancements in technology have considerably improved the spectral quality, providing better control over wavelength design of the LEDs, light intensity produced, and energy efficiency of the lamps (Massa et al., 2008;

Morrow, 2008). Modern LED lighting efficiently provides precise blue, red, and far-red light, with other spectra for UV and green light improving (Massa et al., 2008).

While UV light can be damaging to plants, small quantities can trigger a stress response that promotes phenolic compound synthesis (Li and Kubota, 2009; Winkel-Shirley, 2002). 9

Phenolic compound synthesis helps reduce the risk of high-energy radiation such as UV light from damaging DNA (Li and Kubota, 2009; Winkel-Shirley, 2002). Only UV-A/B and blue light have been associated with positive responses in plants. For instance, Warren et al. (2003) reported increases of both quercetin and kaempferol in black cottonwood (Populus balsamifera ssp. trichocarpa) in response to UV-B radiation. Many phenolic compounds, comprising many aromas and flavors in culinary herbs, are specifically promoted through light stress and increase with the quantity or duration of UV or blue light (Blande et al., 2014; Dixon and Paiva, 1995;

Kopsell et al., 2015; Taulavuori et al., 2016). Plants treated with UV are generally given very low quantities of just a few micromoles, and used as a finishing treatment to trigger phenolic accumulation prior to harvest (Goto et al., 2016). For photoprotective mediation, in addition to the effects of ultra-violet (UV) light, blue light is also involved in stress responses related to phenolic biosynthesis and light seeking/avoidance growth (Caldwell and Britz, 2006; Kopsell and Sams, 2013; Randall and Lopez, 2015; Taulavuori et al., 2016; Li and Kubota, 2009).

Blue light is involved in several essential plant functions. Notably, blue light plays a central role in initial stomatal opening in the morning (Baroli et al., 2008; Takemiya, et al., 2013;

Wu et al., 2007), plant height suppression (Hernández and Kubota, 2012), circadian rhythm

(Thomas, 2006), and promotion of photoprotective pigmentation (Dixon and Paiva, 1995;

Taulavuori et al., 2016). In the morning, gas exchange is influenced by blue light photoreceptors, such as cytochromes and phototropins, aiding stomatal conductance (gs) by playing a role in opening guard cells to initiate gas exchange (Briggs and Huala, 1999; Humble and Hsiao, 1970).

Stomatal regulation influences gas exchange and water loss through transpiration (Xu and Zhou,

2008). Due to the effect on stomatal opening, blue light may have a role in increased gs compared to other wavelengths, potentially affecting water use efficiency. Transpiration is 10

dependent on available water for uptake, and a water deficiency can reduce Pn by limiting gs

(Farquhar et al., 1982), thereby reducing root hydraulics and nutrient uptake. Ultimately, several factors affect stomatal regulation and play an important role in the growth of plant tissue by changes in water availability, light quality, and hormone levels. The interaction of these components is still not thoroughly understood and further research on light quality affecting water use and hormone activity is needed.

Blue light signaling is important for triggering photoprotective responses in plants to high energy radiation. As reported by Li and Kubota (2009), increasing blue light increases the leafy green nutritional value. In fact, blue light promotion of tissue pigmentation and nutrition

(Kopsell and Sams, 2013) can be a key reason for its inclusion in lighting. Furthermore, blue light is involved in stress responses related to secondary metabolite biosynthesis for photoprotective pigmentation, and photomorphogenic responses for light-seeking or -avoidance morphogenesis (Caldwell and Britz, 2006; Kopsell and Sams, 2013; Taulavuori et al., 2016).

These flavonoids are the same compounds of interest in culinary herbs. However, too much blue light can lead to excessive gas exchange, leading to water loss in favor of driving CO2 fixation through Pn (Xu and Zhou, 2008), reduction in size and mass (Randall and Lopez, 2015;

Taulavuori et al., 2016), and higher costs associated with the energy inputs for blue LEDs

(Currey and Lopez, 2013; Massa et al., 2006).

Green light use in horticulture is increasing, as recent reports suggest its application can increase the fresh biomass for some plants (Hernández and Kubota, 2016; Zhang et al., 2011).

Compared to the visual appearance of leaves under dichromatic blue and red light, green light is added in some LED lamps to create the appearance of white light, allowing for better visual inspection (Kim et al., 2004). Although green light contributes to overall Pn, the effect of green 11 light is also deleterious to blue light-induced photomorphogenesis (Bugbee, 2016; McCree,

1972). Kim et al. (2004) reported using up to 24% green light increases plant growth, but green light above 50% reduced final growth. Among other receptors, green light radiation acts upon the same cryptochrome sensory as blue light (Bouly et al., 2007). This, in part, may have a role in green light’s antagonistic action on anthocyanin accumulation to blue light radiation (Carvalho and Folta, 2016). For instance, Dougher and Bugbee (2001) report light between 580 and 600 nm was deleterious for lettuce (Lactuca sativa) dry mass. Blue light antagonism responses in plants by green light is dose-dependent at a 1:2 blue:green ratio for green light impediment of blue light responses (Frechilla et al., 2000). Because of this, green light use may diminish or offset plant responses to blue light.

Chlorophyll absorbs red light at a peak much higher than its absorption of blue light, influencing growth and photomorphogenic responses in plants as a fundamental spectral component for photosynthesis and light sensing. Increasing red light promotes leaf expansion leaves and biomass accumulation (Hernández and Kubota, 2012; Shacklock et al., 1992) and regulate circadian rhythms (Franklin and Whitelam, 2005; Lorrain et al., 2008; Sharrock, 2008).

In fact, much of a plant’s perception of light and diurnal cycles is related to red and far-red light perception by phytochromes Pr and Pfr (Schmitt and Wulff, 1993; Sharrock, 2008). As Pr is stimulated by red light, the phytochrome chromophore undergoes a conformation change to convert into Pfr. Similarly, Pfr is converted into Pr when red light is applied. By this mechanism, plants perceive changes in light quality (Franklin & Whitelam, 2005; Lorrain et al., 2008), emphasizing the important roles of red light on plant responses.

Far-red light is often associated with red light because of phytochrome action. Far-red light increases proportionally to red light at the end of the day. Furthermore, while both red and 12

far-red light contribute to Pn, use of both spectra increase Pn higher than the sum of their individual contributions, known as the Emerson effect (Emerson and Rabinowitch, 1960).

Ultimately, plant responses to light quality influence a substantial part of growth and development, and the control of specific wavelengths in CEA have become a central focus for photobiology research in horticulture.

Aroma and flavor synthesis

Phenolics and essential oils are the compounds comprising many of the flavors and aromas in foods, including culinary herbs. These metabolites are dietary nutrients for human health (USDA, 2011; Cook and Samman, 1996), promoting a range of benefits that include control of blood pressure, cholesterol, and a reduced risk of certain cancers (Hollman, 2001;

Huang et al., 2009; Pandey and Rizvi, 2009). Many phenolic compounds are found naturally in herbs, though specific compounds and their concentrations are species-specific (Taulavuori et al.,

2016). Flavonoid synthesis through the phenylpropanoid pathway begins with phenylalanine

(Ryan et al., 2002). Intermediaries of the pathway are synthesized by several enzymes along the way and depend on the specific branched pathway the final compound (Shirley, 1996), resulting in the production of chalcones, flavonols, flavones, anthocynanis, and tannins. Chalcone synthase produces chalcone, and subsequent steps through chalcone isomerase, flavonone 3- hydroxylase, flavonoid 3’5’-hydroxylase, and flavonol synthase result in the production of key flavonoids kaempferol, quercetin, and myricetin (Winkel-Shirley, 2001). Although flavonoids are generally classified as photoprotective pigments (Agati et al., 2013; Winkel-Shirley, 2002), specific compounds are synthesized in response to stress signals that play a role in mitigating drought stress (Ma et al., 2014), pathogenic presence (Sherwood et al., 1970), mechanical damage, low temperature (Dixon and Paiva, 1995), air quality (Bartwal et al., 2013), and even 13 pollinator attraction (Winkel-Shirley, 2002). Although these responses are diverse, many of these pathways interact with light signaling.

Essential oil synthesis through light-mediated stresses can be promoted through several more pathways than that of flavonoids. Isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP) are produced via the mevalonic acid (MVA) pathway or the 2-C- methylerythritol-4-phosphate (MEP) pathway and are precursors to essential oil synthesis

(Rehman et al., 2016). The final essential oils are synthesized by a variety of terpene synthases

(TPS) depending on the initial promoting signal and ending product. Isoprenoid synthesis pathway steps are split among the cytosol, peroxisomes, plastids, and mitochondria (Rehman et al., 2016), with MEP associated with synthesis steps in chloroplasts, and MVA in the cytosol.

Ultimately, these pathways produce geranyl diphosphate, farnesyl diphosphate, or geranyl geranyl diphosphate that form mono- or sesquiterpenes that result in essential oils, di terpenes, or alternatively are used in gibberellin biosynthesis (Hedden and Proebsting, 1999). Promotion of essential oil synthesis is mediated, in part, by light intensity and quality, as seen in basil (Chang et al., 2008; Fernandes et al., 2013), but also by air quality (Blande et al., 2014), and temperature

(Hälvä et al., 1993), and through natural variation of compounds and respective concentrations across species (Rehman et al., 2016). While essential oils corresponding to desired aromas in plants are multi-faceted in their regulation, all depend to some degree on Pn function and gas exchange.

Within CEA, several factors can affect plant growth and development, yet all appear to be secondary to the metabolism mediated by Pn. Growth is primarily mediated by light quantity, and while intensity and photoperiod are components of this, the overall DLI is the main factor for growth. Additionally, the efficiency of gas exchange and synthesis of photoprotective pigments 14 are largely influenced by light quality. Light spectra mediate cell expansion, circadian rhythm, and the promotion of essential oils and phenolics. In particular, blue light influences stomatal conductance, PSII efficiency, CO2 use efficiency, and secondary metabolite synthesis. Because of this, investigation of culinary herb requirements for growth, development, and metabolism in response to light quantity and quality are required for the improvement of food crop production in CEA.

References

Agati, G., C. Brunetti, M. Di Ferdinando, F. Ferrini, S. Pollastri, and M. Tattini. 2013. Functional roles of flavonoids in photoprotection: new evidence, lessons from the past. Plant Physiol. Biochem. 72:35–45.

Airaki, M., M. Leterrier, R.M. Mateos, R. Valderrama, M. Chaki, J.B. Barroso, L.A. Del Rio, J.M. Palma, and F.J. Corpas. 2012. Metabolism of reactive oxygen species and reactive nitrogen species in pepper (Capsicum annum L.) plants under low temperature stress. Plant Cell Environ. 35:281–295.

Albright, L.D., A.J. Both, and A.J. Chiu. 2000. Controlling greenhouse light to a consistent daily integral. Amer. Soc. Agr. Biol. Eng. 43:421–431.

Amann, M., I. Bertok, J. Borken-Kleefeld, J. Cofala, C. Heyes, L. Höglund-Isaksson, Z. Klimont, B. Nguyen, M. Posch, P. Rafaj, R. Sandler, W. Schöpp, F. Wagner, and W. Winiwarter. 2011. Cost-effective control of air quality and greenhouse gases in Europe: modeling and policy applications. Environ. Modelling Software 26:1489–1501.

Baroli, I., G.D. Price, M.R. Badger, and S. von Caemmerer. 2008. The contribution of photosynthesis to the red light response of stomatal conductance. Plant Physiol. 146:737–747.

Bartwal, A., R. Mall, and P. Lohani. 2013. Role of secondary metabolites and brassinosteroids in plant defense against environmental stresses. J. Plant Growth Regulat. 32:216–232.

Beaman, A.R., R.J. Gladon, and J.A. Schrader. 2009. Sweet basil requires an irradiance of 500 µmol∙m–2∙s−1 for greatest edible biomass production. HortScience 44:64−67.

Berry, J. and O. Björkman. 1980. Photosynthetic response and adaptation to temperature in higher plants. Ann. Rev. Plant Physiol. 31:491–543.

Blande, J.D., J.K Holopainen, and Ü. Niinemets. 2014. Plant volatiles in polluted atmospheres: stress responses and signal degradation. Plant Cell Environ. 37:1892–1904.

Blank, I. and W. Grosch. 1991. Evaluation of potent ordorants in dill seed and dill herb (Anethum graveolens L.) by aroma extract dilution analysis. J. Food Sci. 56:63–67. 15

Bouly, J.P., E. Schleicher, M. Dionisio-Sese, F. Vandenbussche, D. van Der Straeten, N. Bakrim, S. Meier, A. Batschauer, P. Galland, R. Bitti, and M. Ahmad. 2007. Cryptochrome blue light photoreceptors are activated through interconversion of flavin redox states. J. Biol. Chem. 282:9383–9391.

Briggs, W.R. 2002. Phototropins 1 and 2: versatile plant blue-light receptors. Trends Plant Sci. 7:204–210.

Briggs, W.R. and E. Huala. 1999. Blue-light photoreceptors in higher plants. Ann. Rev. Cell Dev. Biol. 15:33–62.

Brown, C. and S. Miller. 2008. The impact of local markets: a review of research on farmer markets and community supported agriculture (CSA). Am. J. Agr. Econ. 90:1298–1302.

Brown, S. 1991. Culinary herb use in southern California restaurants. California Agr. 45:4−6.

Bugbee, B. 2016. Toward an optimal spectral quality for plant growth and development: the importance of radiation capture. Acta Hort. 1134:1–12.

Busch, F.A. and R.F. Sage. 2016. The sensitivity of photosynthesis to O2 and CO2 concentration identifies strong Rubisco control above the thermal optimum. New Phytol. 213:1036–1051.

Caldwell, C.R. and S.J. Britz. 2006. Effects of supplemental ultraviolet radiation on the carotenoid and chlorophyll composition of greenhouse-grown leaf lettuce (Lactuca sativa L.) cultivars. J. Food Composition Anal. 19:637–644.

Carvalho, S.D. and K. Folta. 2016. Green light control of anthocyanin production in microgreens. Acta Hort. 1134:13–18.

Chang, X., P.G. Alderson, and C.J. Wright. 2005. Effect of temperature integration on the growth and volatile oil content of basil (Ocimum basilicum L.). J. Hort. Sci. Biotechnol. 80:583−598.

Chang, X., P.G. Alderson, and C.J. Wright. 2008. Solar irradiance level alters the growth of basil (Ocimum basilicum L.) and its content of volatile oils. Environ. Expt. Bot. 63:216−223.

Chen, C.C., M.Y Huang, K.H. Lin, S.L. Wong, W.D. Huang, and C.M. Yang. 2014. Effects of light quality on the growth, development and metabolism of rice seedlings (Oryza sativa L.). Res. J. Biotech. 9:15–24.

Cook, N.C. and S. Samman.1996. Flavonoids- chemistry, metabolism, cardioprotective effects, and dietary sources. J. Nutr. BioChem. 7:66–76.

Crafts-Brandner, S.J. and M.E. Salvucci. 2000. Rubisco activase constrains the photosynthetic potential of leaves at high temperature and CO2. PNAS 97:13430–13435.

Currey, C.J. and R.G. Lopez. 2013. Cuttings of Impatiens, Pelargonium, and Petunia propagated under light-emitting diodes and high-pressure sodium lamps have comparable growth, morphology, gas exchange, and post-transplant performance. HortScience 48:428–434. 16

Currey, C.J., V.A. Hutchinson, and R.G. Lopez. 2012. Growth, morphology, and quality of rooted cuttings of several herbaceous annual bedding plants are influenced by photosynthetic daily light integral during root development. HortScience 47:25–30.

Darko, E., P. Heydarizadeh, B. Schoefs, M.R. Sabzalian. 2014. Photosynthesis under artificial light: the shift in primary and secondary metabolism. Philos. Trans. R. Soc. London B. Biol. Sci. 369:1–7. doi: 10.1098/rstb.2013.0243.

De Pascale, S., A. Maggio, F. Orsini, and G. Barbieri. 2006. Nutrients influence on ready to eat sweet basil quality. Acta Hort. 718:523−530.

Despommier, D. 2011. The vertical farm: Controlled environment agriculture carried out in tall buildings would create greater food safety and security for large urban populations. J. Verbraucherschutz Lebensmittelsicherheit 6:233–236.

Dixon, R.A. and N.L. Paiva. 1995. Stress-induced phenylpropanoid metabolism. Amer. Soc. Plant Physiol. 7:1085–1097.

Dou, H., G. Niu, M. Gu, J.G. Masabni. 2018. Responses of sweet basil to different daily light integrals in photosynthesis, morphology, yield, and nutritional quality. HortScience 53:496–503.

Dougher, T.A.O. and B. Bugbee. 2001. Differences in the response of wheat, soybean and lettuce to reduced blue radiation. Photochemistry Photobiology 73:199–207.

Emerson, R. and E. Rabinowitch. 1960. Red drop and role of auxiliary pigments in photosynthesis. Plant Physiol. 35:447–485.

Farquhar, G.D. and T.D. Sharkey. 1982. Stomatal conductance and photosynthesis. Ann. Rev. Plant Physiol. 33:317–345.

Farrel, K.T. 1999. Spices, condiments and seasonings. Aspen Publishers, Gaithersburg, MD.

Faust, J.E. and J. Logan. 2018. Daily light integral: A research review and high-resolution maps of the United States. HortScience 53:1250–1257.

Fernandes, V.F., L.B. de Almeida, E.V.R. da S. Feijo, D. da C. Silva, R.A. de Oliveira, M.S. Mielke, and L.C. do B. Costa. 2013. Light intensity on growth, leaf micromorphology and essential oil production of Ocimum gratissimum. Braz. J. Pharmacogn. 23:419–424.

Fisher, P. and C. Donnelly. 2001. Evaluating supplemental light for your greenhouse. Uni. NH Coop Ext. 1–6.

Franklin, K.A. and G.C. Whitelam. 2005. Phytochromes and shade-avoidance responses in plants. Ann. Bot. 96:169–175.

Frechilla, S., L.D. Talbott, R.A. Bogomolni, and E. Zeiger. 2000. Reversal of blue light- stimulated stomatal opening by green light. Plant Cell Physiol. 41:171–176. 17

Genty, B., J.M. Briantais, and N.R. Baker. 1989. The relationship between the quantum yield of photosynthetic electron transport and quenching of chlorophyll fluorescence. Biochimica et Biophysica Acta 990:87–92.

Gomez, C. and C.A. Mitchell. 2016. Physiological and productivity response of high-wire tomato as affected by supplemental light source and distribution within the canopy. J. Amer. Soc. Hort. Sci. 141:196–208.

Gomez, C., R.C. Morrow, C.M. Bourget, G.D. Massa, and C.A. Mitchell. 2013. Comparison of intracanopy light-emitting diode towers and overhead high-pressure sodium lamps for supplemental lighting of greenhouse-grown tomatoes. HortTechnology 23:93–98.

Goto, E., K. Hayashi, S. Furuyama, S. Hikosaka, and Y. Ishigami. 2016. Effect of UV light on phytochemical accumulation and expression of anthocyanin biosynthesis genes in red leaf lettuce. Acta Hort. 1134:179–186.

Hälva, S., L.E. Craker, J.E. Simon, and D.J. Charles. 1993. Growth and essential oil in dill, Anethum graveolens L., in response to temperature and photoperiod. J. Herbs Spices Medicinal Plants. 1:47–56.

Harborne, J.B. and H. Baxter. 2001. Chemical dictionary of economic plants. Wiley and Sons, New York, NY.

Hedden, P. and W.M. Proebsting. 1999. Genetic analysis of gibberellin biosynthesis. Plant Phyiol. 119:365–370.

Hernández, R. and C. Kubota. 2012. Tomato seedling growth and morphological responses to supplemental led lighting red:blue ratios under varied daily solar light integrals. Acta Hort. 956:187–194.

Hernández, R. and C. Kubota. 2016. Physiological responses of cucumber seedlings under different blue and red photon flux ratios using LEDs. Environ. Expt. Bot. 121:66–74.

Herrick, J.D. and R.B. Thomas. 1999. Effects of CO2 enrichment on the photosynthetic light response of sun and shade leaves of canopy sweetgum trees (Liquidambar styraciflua) in a forest ecosystem. Tree Physiol. 19:779–786.

Hogewoning, S.W., G. Trouwborst, G.J. Engbers, J. Harbinson, W. van Ieperen, J. Ruijsch, J., O. van Kooten, A.H.C.M. Schapendonk, and C.S. Pot. 2007. Plant physiological acclimation to irradiation by light-emitting diodes (LEDs). Acta Hort. 761:183–191.

Hollman, P.C.H. 2001. Evidence for health benefits of plant phenols: local or systemic effects? J Sci. Food Agric. 81:842–852.

Hoover, W.H. 1937. The dependence of carbon dioxide assimilation in a higher plant on wavelength of radiation. Smithsonian Misc. Collections 95:1–13. 18

Huang, W.Y., Y.Z. Cai, and Y. Zhang. 2009. Natural phenolic compounds from medicinal herbs and dietary plant: potential use for cancer prevention. Nutrition Cancer. 62:1–20 DOI: 10.1080/01635580903191585.

Huber, S.C. and G.E. Edwards. 1975. C4 photosynthesis: light-dependent CO2 fixation by mesophyll cells, protoplasts, and protoplast extracts of Digitaria sanguinalis. Plant Physiol. 55:835–844.

Humble, G.D. and T.C. Hsiao. 1970. Light-dependent influx and efflux of potassium of guard cells during stomatal opening and closing. Plant Physiol. 46:483–487.

Inada, K. 1976. Action spectra for photosynthesis in higher plants. Plant Cell Physiol. 17:355– 365.

Jensen, M.H. 1999. Hydroponics worldwide. Acta Hort. 481:719–730

Jensen, M.H. 2002. Deep flow hydroponics–past, present and future. Proc. Nat. Agr. Plastics Congress 30:40–46.

Justesen, U. and P. Knuthsen. 2001. Composition of flavonoids in fresh herbs and calculation of flavonoid intake by use of herbs in traditional Danish dishes. Food Chem. 73:245–250.

Karlsson, M.G. and R.D. Heins. 1992. Chrysanthemum dry matter partitioning patterns along irradiance and temperature gradients. Can. J. Plant Sci. 72:307–316.

Kim, H.H., G.D. Goins, R.M. Wheeler, and J.C. Sager. 2004. Green-light supplementation for enhanced lettuce growth under red- and blue-light-emitting diodes. HortScience 39:1617–1622.

Kiple, K.F. and K.C. Ornelas. 2000. The Cambridge world history of food. Cambridge Univ. Press, Cambridge, UK.

Koç, E., C. İşlek, and A.S. Üstün. 2010. Effect of cold on protein, proline, phenolic compounds and chlorophyll content of two pepper (Capsicum annum L.) varieties. G.U. J. Sci. 23:1–6.

Kopsell, D.A. and C.E. Sams. 2013. Increases in shoot tissue pigments, glucosinolates, and mineral elements in sprouting broccoli after exposure to short-duration blue light from light emitting diodes. J. Amer. Soc. Hort. Sci. 138:31–37.

Kopsell, D.A., C.E. Sams, and R.C. Morrow. 2015. Blue wavelengths from LED lighting increase nutritionally important metabolites in specialty crops. HortScience 50:1285–1288.

Li, Q. and C. Kubota. 2009. Effects of supplemental light quality on growth and phytochemicals of baby leaf lettuce. Environ. Expt. Bot. 67:59–64.

Lipiec, J., C. Dourssan, A. Nosalewicz, and K. Kondracka. 2013. Effect of drought and heat stresses on plant growth and yield: a review. Int. Agrophysics 27:463–477. 19

Lorrain, S., T. Allen, P.D. Duek, G.C. Whitelam. 2008. Phytochrome-mediated inhibition of shade avoidance involves degradation of growth-promoting bHLH transcription factors. Plant J. 53:312–323.

Ma, D., D. Sun, C. Wang, Y. Li, and T. Guo. 2014. Expression of flavonoid biosynthesis genes and accumulation of flavonoid in wheat leaves in response to drought stress. Plant Physiol. Biochem. 80:60–66.

Massa, G.D., H. Kim, R.M. Wheeler, and C.A. Mitchell. 2008. Plant productivity in response to LED lighting. HortScience 43:1951–1956.

Massa, G.D., J.C. Emmerich, R.C. Morrow, C.M. Bourget, C.A. Mitchell. 2006. Plant-growth lighting for space life support: a review. Gravitational Space Biol. 19:19–29.

McCree, K.J. 1972. The action spectrum absorptance and quantum yield of photosynthesis in crop plants. Agr. Meteorol. 9:191–216.

Millar, A.J., M. Straume, J. Chory, N. Chua, and S.A. Kay. 1995. The regulation of circadian period by phototransduction pathways in Arabidopsis. Sci. 267:1163–1166.

Miyashita, Y., Y. Kitaya, C. Kubota, T. Kozai, and T. Kimura. 1995. Effects of red and far-red light on the growth and morphology of potato plantlets in-vitro: using light emitting diodes as a light source for micropropagation. Acta Hort. 393:189–194.

Moe, R., S.O. Grimstad, and H.R. Gislerod. 2006. The use of artificial light in year round production of greenhouse crops in Norway. Acta Hort. 711:35–42.

Morales, M.R. and J.E. Simon. 1996. New basil selections with compact inflorescences for the ornamental market, p. 543−546. In: J. Janick (ed). Progress in new crops. ASHS Press, Arlington, VA.

Morrow, R. 2008. LED lighting in horticulture. HortScience 43:1947–50.

Pandey, K.B. and S.I. Rizvi. 2009. Plant polyphenols as dietary antioxidants in human health and disease. Oxidative Medicine Cellular Longevity 2:270–278.

Paton, A. 1992. A synopsis of Ocimum L. (Labiatae) in Africa. Kew Bul. 47:403−435.

Pripdeevech, P., W. Chumpolsri, P. Suttiarporn, and S. Wongpornchai. 2010. The chemical composition and antioxidant activities of basil from Thailand using retention indices and comprehensive two-dimensional gas chromatography. J. Serbian Chem. Soc. 75:1503–1513.

Radulescu, V., M.L. Popescu, and D.C. Ilies. 2010. Chemical composition of the volatile oil from different plant parts of Anethum graveolens L. (Umbelliferae) cultivated in Romania. Farmacia 58:594–600. 20

Randall, W.C. and R.G. Lopez. 2015. Comparison of bedding plant seedlings grown under sole- source light-emitting diodes (LEDs) and greenhouse supplemental lighting from LEDs and high- pressure sodium lamps. HortScience 50:705–713.

Rehman, R., M.A. Hanif, Z. Mushtaq, and A.M. Al-Sadi. 2016. Biosynthesis of essential oils in aromatic plants: a review. Food Rev. Int. 32:117–160.

Reichert, S., M. Wust, T. Beck, and A. Mosandl. 1998. Stereoisomeric flavor compounds LXXXI: Dill ether and its cis-stereoisomers: Synthesis and enantioselective analysis. J High Resolution Chromatography 21:185–188.

Ritchie, J.T. and D.S. NeSmith. 1991. Temperature and crop development, p. 5–25. In: Hanks, J. and J.T. Ritchie (eds.), Modeling plant and soil systems. Amer. Soc. Agron. Madison, WI.

Ross, R.T. 1967. Some thermodynamics of photochemical systems. J. Chem. Physics. 46:4590– 4593.

Ryan, K.G., K.R. Markham, S.J. Bloor, J.M. Bradley, K.A. Mitchell, and B.R. Jordan. 1998. UVB radiation induced increase in quercetin:kaempferol ratio in wild-type and transgenic lines of Petunia. Photochemistry Photobiology 68:323–330.

Sager, J.C., W.O. Smith, J.L. Edwards, and K.L. Cyr. 1988. Photosynthetic efficiency and phytochrome photoequilibria determination using spectral data. Amer. Soc. Agr. Eng. 31:1882– 1889.

Schäfer, E. and F. Nagy. 2006. Photomorphogenesis in plants and bacteria: function and signal transduction mechanisms, 3rd ed. Springer Science and Business Media, Dordrecht, The Netherlands.

Schmitt, J. and R.D. Wulff. 1993. Light spectral quality, phytochrome and plant competition. Trends Eco. Evo. 8:47–51.

Seginer, I., L.D. Albright, and I. Ioslovich. 2006. Improved strategy for a constant daily light integral in greenhouses. Biosystems Eng. 93:69–80.

Shacklock, P.S., N.D. Read, and A.J. Trewavas. 1992. Cytosolic free calcium mediates red light- induced photomorphogensis Nature 358:753–755.

Sharrock, R.A. 2008. The phytochrome red/far-red photoreceptor superfamily. Genome Biol. 9:230

Sherwood, R.T., A.F. Olah, W.H. Oleson, and E.E. Jones. 1970. Effect of disease and injury on accumulation of a flavonoid estrogen, coumestrol, in alfalfa. Phytopathology 60:684–688.

Shirley, B. 1996. Flavonoid biosynthesis: ‘new’ functions for an ‘old’ pathway. Trends Plant Sci. 1:377–382. 21

Simon, J.E., M.R. Morales, W.B. Phippen, R.F. Viera, and Z. Hao. 1999. Basil: a source of aroma compounds and a popular culinary and ornamental herb, p. 499–505. In: J. Janick (ed.). Perspectives on New Crops and New Uses. ASHS Press, Alexandria, VA.

Stefova, M., T. Stafilov, S. Kulevanova. 2003. HPLC analysis of flavonoids, p. 183–195. In: J. Cazes (ed.). Encyclopedia of Chromatography. Marcel Dekker, New York, NY.

Stutte, G.W. 2009. Light-emitting diodes for manipulating the phytochrome apparatus. HortScience 44:231–234.

Styer, R.C. and D.S. Koranski. 1997. Plug and transplant production: a grower’s guide. Ball Publishing, Batavia, IL.

Taiz, L. and E. Zeiger. 2010. Plant physiology, 5th ed. Sinauer Associates. Sunderland, MA.

Takemiya, A., N. Sugiyama, H. Fujimoto, T. Tsutsumi, S. Yamauchi, A. Hiyama, Y. Tada, J.M. Christie, and K. Shimazaki. 2013. Phosphorylation of BLUS1 kinase by phototropins is a primary step in stomatal opening. Nature 4:2094.

Taulavuori, K., V. Hyoky, J. Oksanen, E. Taulavuori, and R. Julkunen-Tiitto. 2016. Species- specific differences in synthesis of flavonoids and phenolic acids under increasing periods of enhanced blue light. Environ. Exp. Bot. 121:145–150.

Telci, I., E. Bayram, G. Yilmaz, and B. Avci. 2006. Variability in essential oil composition of turkish basils (Ocimum basilicum L.). Biochem. Sys. Ecol. 34:489–497.

Thomas, B. 2006. Light signals and flowering. J. Expt. Bot. 57:3387–3393.

Tixier, P. and H. de Bon. 2006. Urban horticulture. Cities farming for the future, urban agriculture for green and productive cities. RUAF Foundation, IDRC and IIRR, Silang, The Philippines.

Trouwborst, G., J. Oosterkamp, S.W. Hogewoning, J. Harbinson, and W. van Ieperen. 2010. The responses of light interception, photosynthesis and fruit yield of cucumber to LED-lighting within canopy. Physiol. Plant. 138:289–300.

Turner, D.W. and E. Lahav. 1985. Temperature influences nutrient absorption and uptake rates of bananas grown in controlled environments. Scientia Hort. 26:311–322.

U.S. Dept. Agr. 1998. Census of horticultural specialties. Census of Agriculture. Accessed April 6 2019. http://usda.mannlib.cornell.edu/usda/AgCensusImages/1997/03/04/1673/Table-01.pdf/

U.S. Dept. Agr. 2009. Census of horticultural specialties. Census of Agriculture. Accessed April 6 2019. https://www.nass.usda.gov/Publications/AgCensus/2007/Online_Highlights/Census_of_Horticult ure_Specialties/HORTIC.pdf/ 22

U.S. Dept. Agr. 2011. USDA database for the flavonoid content of selected foods. Release 3. Nutrient Data Laboratory.

Walters, K.J. and C.J. Currey. 2015. Hydroponic greenhouse basil production: Comparing systems and cultivars. HortTechnology 25:645–650.

Warren, J.M., J.H. Bassman, J.K. Fellman, D.S. Mattinson, and S. Eigenbrode. 2003. Ultraviolet- B radiation alters phenolic salicylate and flavonoid composition of Populus trichocarpa leaves. Tree Physiol. 23:527–535.

Wheeler, R.M., T.W. Tibbitts, and A.H Fitzpatrick. 1991. Carbon dioxide effects on potato growth under different photoperiods and irradiance. Crop Sci. 31:1209–1213.

Winkel-Shirley, B. 2001. Flavonoid biosynthesis. A colorful model for genetics, biochemistry, cell biology, and biotechnology. Plant Physiol. 126:485–493.

Winkel-Shirley, B. 2002. Biosynthesis of flavonoids and effects of stress. Curr. Opin. Plant Biol. 5:218–223.

Wittwer, S.H. and N. Castilla. 1995. Protected cultivation of horticultural crops worldwide. HortTechnology 5:6–23.

Wolf S., A. Marani, and J. Rudich. 1991. Effect of temperature on carbohydrate metabolism in potato plants. J. Exp. Bot. 42:619–625.

Wu, M.C., C.Y. Hou, C.M. Jiang, Y.T. Wang, C.Y. Wang, H.H. Chen, H.M. Chang. 2007. A novel approach of LED light radiation improves the antioxidant activity of pea seedlings. Food Chem. 101:1753–1758.

Xu, Z. and G. Zhou. 2008. Responses of leaf stomatal density to water status and its relationship with photosynthesis in a grass. J. Expt. Bot. 59:3317–3325.

Zhang, T., S.A. Maruhnich, and K.M. Folta. 2011. Green light induces shade avoidance symptoms. Plant Physiol. 157:1528–1536.

Zimmerman, R.C., D.G. Kohrs, D.L. Steller, and R.S. Alberte. 1997. Impacts of CO2 enrichment on productivity and light requirement of Eelgrass. Plant Physiol. 115:599–607. 23

Zobayed, S. and P.K. Saxena. 2004. Production of St. John’s wort plants under controlled environment for maximizing biomass and secondary metabolites. In Vitro Cell. Dev. Biol. Plant 40:108–114.

CHAPTER 2: INCREASING DAILY LIGHT INTEGRALS ENHANCES GROWTH OF

EIGHT CULINARY HERBS GROWN HYDROPONICALLY

A paper prepared for submission to HortScience Alexander G. Litvin and Christopher J. Currey

Abstract

Species-specific responses to daily light integral (DLI) influence light management for greenhouse crop production. Our objective was to quantify the growth of eight species of culinary herbs in response to DLIs common to greenhouse environments. Seedlings of sweet basil (Ocimum basilicum ‘Nufar’), parsley (Petroselinum crispum ‘Giant of Italy’), dill (Anethum graveolens ‘Bouquet’), mint (Mentha sp.), oregano (Origanum vulgare), sage (Salvia officinalis), cilantro (Coriandrum sativum ‘Santo’), and thyme (Thymus vulgaris) were placed in nutrient- film technique hydroponic systems in a glass-glazed greenhouse. Plants were grown under supplemental lighting from high-pressure sodium lamps and placed under no shade or shade cloth of 33% or 69%. The experiment was repeated five times, resulting in DLIs ranging from 2 to 20 mol∙m–2∙d–1. Data collected or calculated were growth index (GI), shoot fresh (SFM), dry mass (SDM), and water content (w/w). Growth index in response to increasing light was curvilinear for all species except basil as DLI increased from 2 to 20 mol∙m–2∙d–1. As DLI increased, SFM increased linearly for basil, cilantro, dill, thyme, and oregano, and quadratically for parsley, mint, and sage. The magnitude of increase with respect to photosynthetically active radiation (PAR) productivity for SFM was highest for dill (9.9 g∙mol–1), and lowest for oregano

(0.4 g∙mol–1) within the linear range of this study, was noted by an increase of 46.0 and 1.7 g as

DLI increased from 8 to 12 mol∙m–2∙d–1 for dill and oregano respectively. The SFM in response to increasing DLI was curvilinear for parsley, mint, and sage within the study’s parameters, with max accumulation of mass at DLIs of 19.0, 17.1, and 14.8 mol∙m–2∙d–1 respectively, after which 25 further increases in DLI resulted in less SFM. Yet SDM for parsley and mint were linear up to 20 mol∙m–2∙d–1, while sage increased curivilinearly with an optimal max DLI of 14.8 mol∙m–2∙d–1.

Understanding the relationship of species-specific responses to DLI for may provide growers essential lighting data to enhance production by efficiently implementing light management practices.

Culinary herbs are popular due to their enhancement of flavor and aroma of food (Brown,

1991; Pripdeevech et al., 2010; Simon et al., 1999), historic cultural value (Cook and Samman,

1996; Justesen and Knuthsen, 2001; Paton, 1992), and ornamental appeal (Morales and Simon,

1996). Culinary herbs are commonly grown as field crops, but are increasingly grown under protected culture (USDA, 1998, 2014). The economic value in 2012 of food crops grown under protected culture, such as in greenhouses, was reported at approximately $800 million (USDA,

2014), having increased by 260% since 1997 (USDA, 1998). More specifically, the sector for culinary herbs produced in protected culture increased by 570% in value from 1997 to 2012

(USDA, 1998, 2014).

When growing culinary herbs in greenhouses and other controlled environment agriculture (CEA) facilities, production systems (Walters and Currey, 2015), mineral nutrition

(De Pascale et al., 2006), air temperature (Chang et al., 2005), and light intensity and duration

(Beaman et al., 2009; Chang et al., 2008) all influence yield. The amount of light in particular is a critical determinate of plant growth, but due to infrastructure and glazing, can be greatly reduced within a greenhouse (Hanan, 1998; Walker and Slack, 1970). The quantity of light or photosynthetically active radiation (PAR) received by a plant strongly affects yield (Beaman et al., 2009; Dou et al., 2018; Faust et al., 2005). For example, yield of culinary herbs will vary 26 greatly depending on seasonal changes (Albright et al., 2000), and effective light transmission through glazing (Walker and Slack, 1970) if the grower only relies on ambient light.

There are conflicting reports and limited data quantifying the effect of PAR on culinary herb growth. Morgan (2001) reported many herbs have low light requirements. However, other studies report a high DLI of 18 mol∙m–2∙d–1 for cilantro, parsley, and dill improved growth compared to low DLI (Currey et al., 2017), and increasing DLIs up to 28 mol∙m–2∙d–1 increases fresh mass of basil (Ocimum basilicum L.) (Beaman et al., 2009; Walters and Currey, 2018). For growers wanting to invest in supplemental lighting to offset low DLI, understanding how yield changes in response to DLI is essential. As DLI increases, light intensity can become supra- optimal for growth as whole plant photosynthesis reaches saturation, resulting in diminishing rates of increase in yield per incremental increase in light (Currey et al., 2012; Beaman et al.,

2009; Faust et al., 2005). Inevitably, this reduces the efficacy of supplemental lighting, and can result in higher production costs per unit due to the reduced effect of additional light and associated electrical use.

CEA producers of culinary herbs can benefit from optimizing supplemental lighting practices to improve yields and/or cost efficiency. Therefore, the objective of our research was to quantify the growth of culinary herbs, grown hydroponically in a greenhouse, in response to

DLI, and model yield of size and both shoot fresh and dry mass (SFM, SDM) in response to DLI.

We hypothesize that yield will increase for all species as DLI increases from 2 to 20 mol∙m–2∙d–1, though the magnitude of effect and optimum DLI will vary with species.

Materials and methods

Plant materials and culture

Seeds of basil, parsley, dill, thyme, mint, oregano, sage, and cilantro were obtained from a commercial supplier (Johnny’s Selected Seeds; Winslow, MA). Three (basil, sage, and 27 cilantro) or four (parsley, dill, mint, oregano, and thyme) seeds were sown into cubes of 162-cell phenolic foam propagation cubes (Oasis Horticubes XL; Smithers-Oasis, Kent, OH), and placed in an environmental growth chamber (E-41; Percival Scientific, Perry, IA). During seed germination and seedling growth, air average daily temperature (ADT) was maintained at a constant 23.0 ± 0.4 °C. A combination of fluorescent and incandescent lights provided 197 ± 4

μmol∙m–2∙s–1 from 0600 HR to 2200 HR. Light was measured with a quantum light sensor (SQ-

222; Apogee Instruments Inc., Logan, UT) along with air temperature by a thermocouple

(TMC1-HD; Onset, Bourne, MA) in a naturally aspirated solar shield (RS3; Onset, Bourne, MA) every 15 s in a growth chamber and recorded by a data logger (HOBO U12 PPFD/Temp; Onset) every 15 min. Seedlings were irrigated daily with deionized water only until radicle emergence, after which by deionized water supplemented with 100 mg∙L–1 nitrogen (N) supplied from a complete, water-soluble fertilizer (Jack’s Hydro FeED 16N–1.8P–14.3K; JR Peters, Allentown,

PA).

Each cube was thinned to one (basil), two (cilantro, mint, and sage), or three (dill, oregano, parsley, and thyme) seedlings per cube to simulate commercial practices, and four cubes of each species were transplanted into one of nine NFT troughs by inserting into 3.5-cm wide holes at 20-cm spacing in the plastic trough (Botanicare, Chandler, AZ) measuring 10 cm wide, 5 cm tall, and 200 cm long (GT50–612; FarmTek, Dyersville, IA) on a 3% slope. A 150-L reservoir (Premium Reservoir; Botanicare) was filled with deionized water and initially amended

–1 with 233 mg·L of MgSO4 and water-soluble fertilizer (Hydro FeED; JR Peters) to an electrical conductivity (EC) of 1.5 dS∙m–1, and pH increased to 6.0 using potassium bicarbonate (JR

Peters). Cubes were placed directly on the nutrient solution, with a flow of approximately 1 L∙m–

1 (Aqua 33-W, Active Aqua; Hydrofarm, Petaluma, CA). One aquatic pump (Aqua 33-W, Active 28

Aqua) circulated nutrient solution through a flex hose (1.9 cm ID, Active Aqua) to a heater/chiller regulator (SeaChill TR-10; TECO, Terrell, TX) to maintain medium at 22.5 ± 0.5

°C. Nutrient solution of each NFT system was monitored with a pH/EC meter (HI 9813-6;

Hanna Instruments, Woonsocket, RI) and corrected with amendments of deionized water for solution volume, concentrated fertilizer solution (Hydro FeED; JR Peters) for maintaining EC at

1.5 mS·cm–1, and through additions of phosphoric and citric acid (pH Down; General

Hydroponics, Sebastopol, CA) or potassium bicarbonate (JR Peters) for a pH of 6.0.

Daily light integral treatments

Three DLI levels were created in the greenhouse by using a combination of shade cloth and supplemental lighting provided by 1000-W high-pressure sodium (HPS) lamps (PL 3000;

P.L. Light Systems, Beamsville, ON, Canada). Photosynthetic photon flux density (PPFD) was reduced with shade cloths individually placed over each NFT system at three treatment levels, each replicated in triplicate within each experimental run. Treatments consisted of no shade cloth, 33% light reduction shade cloth (Harmony 3315 O FR; Svensson, Charlotte, NC), or 69% light reduction shade cloth (Harmony 6920 O FR; Svensson). Shade cloth was draped over a

PVC pipe infrastructure constructed onto the frame of each NFT system, including that of the unshaded treatment, at 100 cm above trough level, and cloth was draped to ensure full coverage.

The greenhouse was located in Ames, IA, at 42° N. The structure comprised a metal infrastructure with clear glass glazing. The greenhouse was supplemented with approximately

180 μmol∙m–2∙s–1 each day from 0600 to 1100 HR, and from 1400 to 2200 HR, by 1000-W HPS lamps (P.L. Light Systems). Greenhouse conditions were controlled with radiant hot-water heating and fog cooling, adjusted by automated environmental controls (Titan; ARGUS Control

Systems, Surrey, B.C., Canada), for ADT set points of 24 °C during the day, and 20 °C at night. 29

Actual ADT air was 24.2 ± 2.3 °C during the day, and 20.0 ± 1.8 °C at night as monitored by a thermocouple (41342; R.M. Young Company, Traverse City, MI) placed under the shade cloth of each of the nine NFT systems. A quantum light sensor (LI-190R; LICOR, Lincoln, NE) measured light intensity, and it was logged hourly and daily by a data logger (CR1000; Campbell

Scientific, Logan, UT). Mean DLI varied across experimental runs in conjunction with seasonal fluctuations in light, resulting in DLIs from 2 to 20 mol∙m–2∙d–1 within the study.

Data collection and calculation

Twenty one (basil, cilantro, dill, oregano, sage, and thyme) or 28 d (mint and parsley) after transplanting into NFT systems, growth index (GI) was calculated for each species. Two widths across the plant canopy were measured perpendicular to one another, and their mean was then averaged with the height from the surface of the substrate to the top of their respective canopies. Shoots were severed at the substrate and shoot fresh mass (SFM) was recorded. Shoots were placed in paper bags and then dried in a forced-air oven at 67 °C for 3 d; shoot dry mass

(SDM) was recorded immediately upon removal from the drying oven. Water content was calculated as the difference between SFM and SDM of a given sample.

Experimental design and statistical analyses

The design was a randomized complete block design for each species. There were three replications (individual NFT systems), comprising of four subsamples (cubes) of each species, for each level of shading per experimental run, and the experiment was repeated five times over a year. All data were subjected to linear or quadratic regression analysis in SigmaPlot (SigmaPlot

11.0; Systat Software, San Jose, CA).

Results

Growth index of plants in response to DLI was curvilinear, increasing to species-specific

DLI maxima between 11.8 (sage) and 17.7 mol·m–2·d–1 (oregano) before decreasing as DLI further increased to 20.0 mol·m–2·d–1 (Fig. 1 and 2). However, basil GI increased linearly with

DLI (Fig. 1) from 16.9 at 2.1 mol·m–2·d–1 to 33.4 at 20.0 mol·m–2·d–1.

Shoot fresh mass increased linearly as DLI increased to 20 mol·m–2·d–1, with a max SFM of 141.7, 94.8, 185.6, 20.9, and 8.1 g per cube for basil, cilantro, dill, thyme, and oregano respectively. The total increase in SFM as DLI increased from 2 to 20 mol·m–2·d–1, were 136.1,

85.8, 175.1, 19.2, and 7.1 g for basil, cilantro, dill, thyme, and oregano, respectively. Parsley, mint, and sage SFM increased curvilinearly a maximum SFM at 80.5 g (16.3 mol·m–2·d–1), 52.6 g (14.9 mol·m–2·d–1), and 43.0 g (15.9 mol·m–2·d–1) respectively. After these maxima, SFM decreased for a total increase from 2 to 20 mol·m–2·d–1 DLI of 71.2, 61.9, and 37.4 g for parsley, mint, and sage respectively (Fig. 1 and 2).

As DLI increased to 20 mol·m–2·d–1, SDM increased linearly for basil, cilantro, parsley, dill, mint, thyme, and oregano by 11.1, 10.8, 10.1, 16.7, 16.4, 3.2, and 1.2 g respectively (Fig. 1 and 2). However, SDM for sage was curvilinear in response to increasing DLI, increasing by 4.8 g from 2.1 to 15.9 mol·m–2·d–1, and decreased thereafter.

Concentration of plant water content decreased curvilinearly as DLI increased. However, the magnitude of change varied with species. As DLI increased to 20 mol·m–2·d–1, water content decreased by a total of 2.7, 5.6, 5.3, 2.6, 5.2, 6.1, 7.8, and 4.9 % relative to SFM for basil, cilantro, parsley, dill, mint, sage, thyme, and oregano respectively (Fig. 1 and 2).

Discussion

The results of our research support previous reports that increasing DLI generally enhances growth of culinary herbs. However, the current study included eight different culinary herb species, and were grown under a wide range of DLIs. Therefore, we were able to model responses to DLI and provide a more comprehensive assessment on the effects of DLI on herb growth.

While increasing DLI up to 20 mol·m–2·d–1 increased GI for basil, the extent and magnitude of the promoting effect of DLI on GI was different for the remaining species. This variation among different species agrees with Faust et al. (2005), who reported the effects of increasing DLI on height and compactness varied for different annual bedding plant species.

Additionally, Mortensen (2004) reported that some ornamental floriculture species become more compact under high light, as evident in this study in curvilinear GI responses, though mass did not necessarily follow similar trends. The dimensions of plants is much more important for ornamental plants grown in containers, where aesthetic appearance is paramount; however, for fresh-cut culinary herbs, there is little implication on shoot dimensions on the ultimate marketability or profitability since shoots are packed in plastic clamshells and sold based on mass.

Increasing DLI enhanced SFM and SDM, but the extent and limits of increase in which growth is linear with increasing DLI is species-specific. For example, the relationship between

DLI and SFM was linear for basil, cilantro, dill, thyme, and oregano between 2 to 20 mol∙m–2∙d–

1, while for parsley, mint, and sage growth increased as DLI increased up to 17.1, 19.0, and 14.8 mol∙m–2∙d–1, after which SFM decreased (Table 1). Walters and Currey (2018) noted that basil grown at 15 mol·m–2·d–1 had two- to three-fold greater SFM compared to plants grown under 7 mol·m–2·d–1 depending on cultivar, and Dou et al. (2018) noted that SFM increases were linear 32 between 9.3 and 17.8 mol·m–2·d–1 for basil. Although grown at a lower range of DLIs than the used in this study, Mapes and Xu (2014) reported linear increases in shoot mass as DLI increased from 3.1 to 9.2 mol∙m–2∙d–1, consistent with the linear responses observed in this study (Fig. 2,

Table 1). Similarly, SFM of cilantro, dill, and flat-leaf parsley increased by 21.0, 17.1, and 13.3 g, respectively, when grown under a high DLI (17.9–19.3 mol·m–2·d–1) compared to plants grown under a low DLI (5.5–7.5 mol∙m–2∙d–1; Currey et al. 2017). Our results agree with previous reports that yield of culinary herbs is enhanced with increasing light; however, our results provide a more comprehensive quantification of how DLI affects mass for a wider variety of culinary herbs.

Classifying plant responses to DLI enables producers to accommodate species-specific requirements for photosynthetic light. Moe (1994) and Faust (2011) suggest grouping species based on DLI requirements as low (5 to 10 mol·m–2·d–1), medium (10 to 20 mol·m–2·d–1), high

(20 to 30 mol·m–2·d–1), and very high light (>30 mol·m–2·d–1) plants. Fresh mass is the primary interest of a fresh-cut hydroponic herb producer, and so herb species should be classified based on the impact of DLI on fresh mass. Mint, sage, and parsley could be classified as medium-light plants based on maximal SFM production at DLIs of 17.1, 14.8, and 19.0 mol·m–2·d–1, respectively, in the present study. Alternatively, basil, cilantro, dill, thyme, and oregano may be

–2 –1 classified as high or very high-light plants, as their optimal DLIs are above 20 mol·m ·d , the upper range of DLIs used in this study. Beaman et al. (2009) reported SFM and SDM of sweet basil increases as light intensity increases from 300 to 500 µmol·m–2·s–1 (16-h photoperiod; DLI of 17.3 to 28.8 mol·m–2·d–1), but decreases as light further increases to 600 µmol·m–2·s–1 (35.6 mol·m–2·d–1), supporting classification of basil as high light species. While further studies growing those species where growth was not saturated at DLIs >20 mol·m–2·d–1 would be useful 33 for more precise classification, DLIs of that magnitude are seldom used for greenhouse production, whether from a cost-prohibitive amount of supplemental lighting or excessive thermal radiation from ambient sunlight.

As DLI increased, allocation of growth and plant physiology may change to support larger structures (Faust et al., 2005). At increasingly higher DLIs, the concentration of water in

SFM declined (Fig. 1 and 2), although overall water content increased. Linear models for SDM in response to DLI fit more closely than those for SFM for each species respectively. We postulate this is due, in part, to diminished water content of SFM as DLI increases. Faust et al.

(2005) suggests increases in proportional SDM to SFM at higher DLI could be attributed in part to toning, and reported additional structural components and carbohydrates produced in plant stem tissue may play a role in that proportional increase of SDM (Faust et al. 2005; Haque et al.,

2015).

Increasing DLI ameliorates the negative effects of low light on growth and development of culinary herbs. However, supplemental lighting in greenhouses is a large expenditure in capital investments and operating costs, and inefficient lighting may reduce a grower’s profitability. One strategy towards efficient light management in a greenhouse is grouping species by similar light requirements. For species with quadratic relationships of SFM in response to DLI, the limited capacity of whole plant photosynthesis under increasing light intensities (Garland et al., 2012; Marshall and Biscoe, 1980; Oh et al., 2009) highlight potential losses of production output if using excessive supplemental lighting. Lighting for these species should then be restricted to DLIs at or below the vertex maximum from yield (Table 1).

Commercial producers actively manage light to enhance profitability through increased yields. The models describing mass in response to DLI (Fig. 1 and 2) highlight the diversity in 34 the magnitude and extent of increase in mass in response to DLI across species. All herbs in this study would benefit from increasing light up to species-specific maxima, and supplemental light should not be provided after these maxima. Additionally, the magnitude of increase in SFM varies among species. The linear models presented herein may be used to estimate the effect of

DLI on yield. For example, increasing the DLI from 8 to 12 mol·m–2·d–1 would increase the yield of basil, cilantro, and sage by 29.2, 17.9, and 18.2 g, respectively. While the predictive models we present for SFM are useful, the PAR productivity of culinary herbs may provide a simpler approach to determining efficient supplemental light. The PAR productivity is a measurement of the increase in harvestable yield on a per-mole PAR basis (g·mol–1 PAR; Kubota et al., 2016). In our study we calculated the PAR productivity of oregano, thyme, sage, cilantro, mint, parsley, basil, and dill to be 0.4, 0.9, 2.4, 3.9, 4.3, 5.2, 6.3, and 9.9 g∙mol–1, respectively. For instance, while thyme SFM increases by 52% as DLI increases from 8 to 12 mol∙m–2∙d–1, the actual increase in SFM is much lower compared to the promotion in SFM for herbs with greater PAR productivity in response to the same increase in light. Those species with greater PAR productivity had overall higher yields and incremental increases from additional lighting, emphasizing the differences in crop productivity across species, and should be considered for any commercial producers making supplemental lighting decision with the aim of maximizing for mass production per unit area of greenhouse space.

For greenhouses relying on ambient light alone, yields may diminish when the DLI becomes limiting during late fall, winter, and early spring (Faust and Logan, 2018). While our data may be used to predict reductions in yield with diminishing DLIs, it may also be used to adjust production times to meet target yields. However, yield reductions on a per-area basis due to lower mass for individual plants may be compensated by increasing the planting density 35

(Walters, 2015). For example, sweet basil grown at a density of 100 plants per m2 under ~7 mol∙m–2∙d–1 produces similar yields per unit area to plants grown at 44 plants per m2 under ~15 mol∙m–2∙d–1 (Walters, 2015). The relationship between mass and DLI varies across species and highlights the importance of grouping crop selections, and the extent to which light can be a limiting factor for efficiently maximizing yields in low DLI regions or seasons. Ultimately, under low DLI conditions, growers must increase their planting density to close canopy space, or increase lighting for the crops.

Conclusion

Our understanding of how culinary herbs respond under a range of sub- to supra-optimal lighting conditions highlights the diversity of species often grown under identical conditions, as lighting requirements can be very species-specific. The models generated herein can help growers maximize size and/or biomass production by changing lighting practices, but also may aid in better grouping of species by lighting requirements to minimize inefficiencies. Although other factors such as temperature, carbon dioxide, and humidity play significant roles in plant growth and development, this study’s focus on DLI sought to replicate conditions expected for commercial practices. While further studies are required to describe how changes in other environmental and atmospheric conditions may interact with DLI, growers are encouraged to conduct trials specific to their facilities.

References

Albright, L.D., A.J. Both, and A.J. Chiu. 2000. Controlling greenhouse light to a consistent daily integral. Amer. Soc. Agr. Bio. Eng. 43:421–431.

Beaman, A.R., R.J. Gladon, and J.A. Schrader. 2009. Sweet basil requires an irradiance of 500 µmol∙m−2∙s−1 for greatest edible biomass production. HortScience 44:64−67.

Brown, S. 1991. Culinary herb use in southern California restaurants. CA Agr. 45:4−6. 36

Buwalda, J.G., J.S Meekings, G.S. and Smith. 1991. Seasonal changes in photosynthetic capacity of leaves of kiwifruit (Actinidia deliciosa) vines. Physiol. Plantarum 83:93–98.

Chang, X., P.G. Alderson, and C.J. Wright. 2005. Effect of temperature integration on the growth and volatile oil content of basil (Ocimum basilicum L.). J. Hort. Sci. Biotechnol. 80:583−598.

Chang, X., P.G. Alderson, and C.J. Wright. 2008. Solar irradiance level alters the growth of basil (Ocimum basilicum L.) and its content of volatile oils. Environ. Expt. Bot. 63:216−223.

Cook, N.C. and S. Samman.1996. Flavonoids- chemistry, metabolism, cardioprotective effects, and dietary sources. Nutr. BioChem. 7:66–76.

Currey, C.J., K.J. Walters, and N.J. Flax. 2017. Hydroponic cilantro, dill, and parsley growth and tissue nutrient concentration as affected by electrical conductivity or daily light integral. HortScience 52:S242.

De Pascale, S., A. Maggio, F. Orsini, and G. Barbieri. 2006. Nutrients influence on ready to eat sweet basil quality. Acta Hort. 718:523−530.

Dou, H., G. Niu, M. Gu, and J.G. Masabni. 2018. Responses of sweet basil to different daily light integrals in photosynthesis, morphology, yield, and nutritional quality. HortScience 53:496– 503.

Faust, J.E. 2011. Light, p. 83−94. In: J. Nau (ed.). Ball Redbook, Vol 2. Crop production. 18th ed. Ball Publishing, West Chicago, IL.

Faust, J.E., V. Holcombe, N.C. Rajapakse, and D.R. Layne. 2005. Effect of daily light integral on bedding plant growth and flowering. HortScience 40:645–649.

Faust, J.E. and J. Logan. 2018. Daily light integral: A research review and high-resolution maps of the United States. HortScience 53:1250–1257.

Garland, K.F., S.E. Burnett, M.E. Day, and M.W. van Iersel. 2012. Influence of substrate water content and daily light integral on photosynthesis, water use efficiency, and morphology of Heuchera americana. J. Amer. Soc. Hort. Sci. 137:57–67.

Hanan, J. 1998. Greenhouses: Advanced technology for protected horticulture. CRC Press, Boca Raton, FL.

Haque, M.S., K.H. Kjaer, E. Rosenqvist, and C.O. Ottosen. 2015. Continous light increases growth, daily carbon gain, antioxidants, and alters carbohydrate metabolism in a cultivated and wild tomato species. Front. Plant Sci. 6:522 doi: 10.3389/fpls.2015.00522. 37

Justesen, U. and P. Knuthsen. 2001. Composition of flavonoids in fresh herbs and calculation of flavonoid intake by use of herbs in traditional Danish dishes. Food Chem. 73:245–250.

Kubota, C., M. Kroggel, A.J. Both, J.F. Burr, and M. Whalen. 2016. Does supplemental lighting make sense for my crop? Acta Hort. 1134:403-412.

Marshall, B. and P.V. Biscoe. 1980. A model for C3 leaves describing the dependence of net photosynthesis on irradiance. J. Exp. Bot. 31:29–39.

Mapes, C. and Y. Xu. 2014. Photosynthesis, vegetative habit and culinary properties of sage (Salvia officinalis) in response to low-light conditions. Can. J. Plant Sci. 94:881–889.

Moe, R. 1994. Vekstfysiologiske aspekter med hoege belysningsstyrker. Gartneryrket 84:8–13.

Morales, M.R. and J.E. Simon. 1996. New basil selections with compact inflorescences for the ornamental market, p. 543−546. In: J. Janick (ed.). Progress in new crops. ASHS Press, Arlington, VA.

Morgan, L. 2001. Fresh culinary herb production: A technical guide to the hydroponic and organic production of commercial fresh gourmet herb crops. Suntec (NZ) Ltd., Tokomaru, New Zealand.

Mortensen, L.M. 2004. Growth and light utilization of pot plants at variable day-to-day irradiances. Europ. J. Hort. Sci. 69:89–95.

Oh, W., I.H. Cheon, K.S. Kim, and E.S. Runkle. 2009. Photosynthetic daily light integral influences flowering time and crop characteristics of Cyclamen persicum. HortScience 44:341– 344.

Paton, A. 1992. A synopsis of Ocimum L. (Labiatae) in Africa. Kew Bul. 47:403−435.

Simon, J.E., M.R. Morales, W.B. Phippen, R.F. Vieira, and Z. Hao. 1999. Basil: A source of aroma compounds and a popular culinary and ornamental herb, p. 449−505. In: 17 J. Janick (ed.). Perspectives on new crops and new uses. ASHS Press, Arlington, VA.

U.S. Dept. Agr. 1998. Census of horticultural specialties. Census of Agriculture. Accessed April 6 2019. http://usda.mannlib.cornell.edu/usda/AgCensusImages/1997/03/04/1673/Table-01.pdf/

Walker, J.N. and D.C. Slack. 1970. Properties of greenhouse covering materials. Transactions of the ASAE 13:682–684.

Walters, K.J. 2015. Quantifying the effects of hydroponic systems, nutrient solution, and air temperature on growth and development of basil (Ocimum L.) species. MS Thesis Iowa State Univ.

Walters, K.J. and C.J. Currey. 2015. Hydroponic greenhouse basil production: Comparing systems and cultivars. HortTechnology 25:645–650.

Walters, K.J. and C.J. Currey. 2018. Effects of nutrient solution concentration and daily light integral on growth and nutrient concentration of several basil species in hydroponic production. HortScience 53:1319–1325. 39

Table 1. Linear regression models for shoot fresh mass of parsley, mint, and sage. The linear range was identified first calculating the optimal daily light integral (DLI) by solving quadratic regression equations for the vertex, then excluding those data from supra-optimal DLIs.

Optimal DLI

Species (mol·m–2·d–1) Linear model R2

Parsley 19.0 y = 10.14 + 4.55 x 0.81

Mint 17.1 y = -0.02 + 3.69 x 0.79

Sage 14.8 y = 3.36 + 2.81 x 0.68

Fig. 1. Growth index, shoot fresh mass (g), shoot dry mass (g), and water content (w/w) for basil, cilantro, parsley, and dill across mean daily light integral from 2 to 20 mol·m–2·d–1. Data were collected 21 d after transplanting for basil, cilantro, and dill and 28 d after transplanting for parsley. Equations and R2 correspond to respective best-fit curves of data to increasing DLI. Symbols in grey were for plants grown at supra-optimal DLIs and were excluded when determining the linear range of SFM in response to DLI.

Fig. 2. Growth index, shoot fresh mass, shoot dry mass, and water content for mint, sage, thyme, and oregano across mean daily light integral from 2 to 20 mol·m–2·d–1. Data were collected 21 d after transplanting for sage, thyme, and oregano and 28 d after transplanting for mint. Equations and R2 correspond to respective best-fit curves of data to increasing DLI. Symbols in grey were for plants grown at supra-optimal DLIs and were excluded when determining the linear range of SFM and SDM in response to DLI. 42

CHAPTER 3: EFFECTS OF SUPPLEMENTAL LIGHT SOURCE ON BASIL, DILL,

AND PARSLEY GROWTH, MORPHOLOGY, AROMA, AND FLAVOR

A paper prepared for submission to The Journal of American Society for Horticultural Science Alexander G. Litvin, Christopher J. Currey, and Lester A. Wilson

Abstract

Although traditional light sources such as high-pressure sodium (HPS), have been standard in supplemental lighting, narrow-spectra light-emitting diodes (LEDs) offer potential benefits for enhancing secondary metabolites such as flavonoids in culinary herbs. Our objectives were to quantify the effect of supplemental light source and spectra on growth, gas exchange, and aroma and flavor of culinary herbs. Basil (Ocimum basilicum L. ‘Nufar’), dill

(Anethum graveolens L. ‘Fernleaf’), and parsley (Petroselinum crispum L. ‘Giant of Italy’) were transplanted into hydroponic systems in a glass-glazed greenhouse. Plants were provided with a supplemental photosynthetic photon flux density (PPFD) of 100 μmol∙m–2∙s–1 from high- pressure sodium (HPS) lamp, or light-emitting diodes (LEDs) with low blue (B) to red (R) light ratio of 7:93 (Low Blue–LB) or high B:R at 30:70 (High Blue–HB). Compared to plants grown under HPS lamps, basil grown under LB and HB LED lighting were shorter, while only HB- grown parsley were shorter; height of dill was unaffected by light source. Basil and parsley shoot fresh mass (SFM) were lower for HB-treated plants compared to HPS, though dill was unaffected by supplemental light source. Shoot dry mass (SDM) of basil, dill, and parsley were unaffected by light source. Both LED treatments increased photosynthesis (Pn) for basil and parsley compared to HPS-grown plants. Stomatal conductance (gs) was higher for LB and HB in basil compared to HPS in the morning and evening, but only HB-treated parsley was higher than

HPS lighting in initially in morning. Basil grown under LB, and parsley under both LEDs had a 43

lower chlorophyll fluorescence (Fv/Fm) than those under HPS by the evening, but all three species had more chlorophyll b under LB light than HPS. Essential oil and phenolic accumulation was influenced by supplemental light treatment and responses varied among species. Lighting from LEDs resulted in a two-fold increase in orientin and myristicin for basil and dill, respectively, while HB increased dillapiole concentration by 89% compared to HPS- grown dill. Notably, quercetin concentration was 2.8-times higher in dill grown under HB compared to HPS. Myrecene increased in all three species under either one (basil–HB; dill–LB) or both (parsley) LED lights compared to HPS. The increased content of aromatic and flavor compounds demonstrate the potential of supplemental lighting systems using specific wavelengths to add value, but come with the caveat of understanding the additional stress imparted onto the photosynthetic mechanisms and the subsequent effect on biomass accumulation. However, any minor yield reduction may be offset by the diminished energy requirements for LED lights.

Culinary herbs are used globally as ingredients in cuisine and as therapeutic components in medications (USDA, 2011: Cook and Samman, 1996; van Wyk, 2014). Herbs such as basil, dill, and parsley are popular based primarily for their aroma and flavor attribute, contributing to their nutritional value (Brown, 1991; Pripdeevech et al., 2010; Simon et al., 1999), historic cultural value (Cook and Samman, 1996; Farrell, 1999; Kiple and Ornelas, 2000; Justesen and

Knuthsen, 2001; Paton, 1992), and ornamental appeal (Morales and Simon, 1996).

Herb production under protected climates, such as greenhouses, has increased by 44% in the United States between 2009 and 2014 (USDA, 2009, 2014) due to year-round availability and crop quality by controlling environmental parameters. Growing culinary herbs in controlled 44 environment agriculture (CEA) facilities can enhance yield and quality of herbs through appropriate cultivar and production system selection (Walters and Currey, 2015) and managing mineral nutrition (De Pascale et al., 2006), air temperature (Chang et al., 2005), and light intensity and duration (Beaman et al., 2009; Chang et al., 2008), allowing for predictable and controllable environments year-round, regardless of season or location (Brown and Miller, 2008;

Jensen, 1999; Moe et al., 2006; Tixier and de Bon, 2006; Wittwer and Castilla 1995).

Traditionally, gas-filled, broad-spectrum lamps, such as high-pressure sodium (HPS) or metal halide, have been used as supplemental lighting, but inefficiencies resulting in thermal radiation increases operating costs (Hogewoning et al. 2007; Trouwborst et al., 2010) and prevents close proximity of lamps to plants (Gomez et al., 2013). Using LEDs, with their long luminous lifespan and narrowband wavelengths, can increase plant quality and offer possible long-term reductions in operating costs (Li and Kubota, 2009; Morrow, 2008; Randall and

Lopez, 2015). Modern high-intensity LED lighting provides narrow-spectra light including B, R, and far-red (FR) wavelengths, among others, offering new efficiencies in supplemental and sole- source lighting (Massa et al., 2008; Randall and Lopez, 2015). For example, the mass per electrical use (g·kWh–1) under LEDs have been reported as much as 4.7 times that of cool fluorescent lighting for dry mass of basil (Piovene et al., 2015), and intracanopy LEDs providing up to 4 times the FW g·kWh–1 of tomatoes (Solanum lycopersicum × S. hbrochaites) than under

HPS. However, Hernández and Kubota (2015) reported cucumber plants (Cucumis sativus) had

3.0 and 3.3 g·kWh–1 under B and R LEDs compared to the 3.5 g·kWh–1 under HPS lighting, but noted that higher efficiency LED lamps currently on the market should achieve up to 4 g·kWh–1.

Specific wavelengths of light common in LEDs, such as B and R, play unique roles in plant responses, and the commercial producers must decide what proportions of these 45 wavelengths to use. For example, R light can mediate photomorphogenesis of leaves and contribute significantly to biomass accumulation (Hernández and Kubota, 2012; Shacklock et al.,

1992), and to phytochrome regulation of circadian rhythms and shade avoidance responses in conjunction with FR light (Franklin and Whitelam, 2005; Lorrain et al., 2008; Sharrock, 2008).

Additionally, R light has shown to increase phenolic concentrations in plants compared to white light, but not to the same magnitude as B light (Li and Kubota, 2009). Blue light plays a central role in initial stomatal opening in the morning (Baroli et al., 2008; Takemiya, et al., 2013; Wu et al., 2007), control of plant height (Hernández and Kubota, 2012), circadian rhythm (Taiz and

Zeiger, 2010), and promotion of photoprotective pigmentation (Taulavuori et al., 2016).

However, too much B light can lead to excessive gas exchange, leading to loss of water in favor of driving CO2 fixation through photosynthesis (Xu and Zhou, 2008), reduction in size and mass

(Randall and Lopez 2015; Taulavuori et al., 2016), and higher costs associated with the energy inputs for B LEDs (Currey and Lopez, 2013; Massa et al., 2006). For photoprotective mediation, in addition to the effects of ultra-violet (UV) light, B light is also involved in stress responses related to phenolic biosynthesis and light seeking/avoidance growth (Caldwell and Britz, 2006;

Kopsell and Sams, 2013; Randall and Lopez, 2015; Taulavuori et al., 2016; Li and Kubota,

2009). Although B light can stimulate phenolic biosynthesis through the phenylpropanoid pathway (Taulavuori et al., 2016), increases are not linear as B light increases, and different B:R light ratios are required for optimum phenolic accumulation for different species (Craver et al.,

2017). Additionally, supplemental lighting using B light can increase the nutritional value of leafy greens as compared to white light (Li and Kubota, 2009). Taken together, the use of B light used for supplemental lighting should be carefully considered.

Some secondary metabolites, essential oils, and phenolic compounds comprising many of the aromas and flavors in culinary herbs are specifically promoted through light stress, and increase with the quantity or duration of UV or B light (Blande et al., 2014; Dixon and Paiva,

1995; Kopsell et al., 2015; Taulavuori et al., 2016). These metabolites are dietary nutrients for human health (USDA, 2011: Cook and Samman, 1996), promoting a range of benefits including control of blood pressure, cholesterol, and reduced risk of certain cancers (Hollman, 2001;

Huang et al., 2009; Pandey and Rizvi, 2009). A variety of aromatic and flavor compounds can be found naturally in moderate concentrations depending on species (Rehman et al., 2016), and many, such as myrecene, quercetin, and others, can be further promoted through stress-specific signaling (Dixon and Paiva, 1995; Ma et al., 2014; Petrussa et al., 2013; Winkel-Shirley, 2002), temperature (Blande et al., 2014), light intensity (Blande et al., 2014; Rehman et al., 2016), and other biotic and abiotic stresses.

Although traditional light sources such as HPS are widely used supplemental sources, the targeted effect of narrow-spectra LEDs offer new potential for enhancing growth, aroma, and flavor. Because of the role of light quality on photosynthesis and the isoprenoid and phenlypropanoid pathways for secondary metabolites such as essential oils and flavonoids, our objectives were to quantify the effect of supplemental light source and spectra on growth, gas exchange, and aroma and flavor of culinary herbs grown hydroponically in a greenhouse. We hypothesize the narrow-spectra of LEDs will provide an advantage compared to HPS lighting by increasing photochemical quenching, potentially increasing growth, and phenolic accumulation.

Under equal light intensity, narrow-spectra lighting with greater proportions of B light will produce greater quantities aromatic and flavor compounds, albeit in more compact plants, than

HPS. Additionally, LEDs with a lower proportion of B light will provide a more efficient light 47 source for photosynthesis per incident light, and higher accumulation of aromatic and flavor compounds in comparison to HPS lighting, with minimal yield penalty.

Materials and methods

Plant materials and propagation

Seeds of basil (Ocimum basilicum L. ‘Nufar’), dill (Anethum graveolens L. ‘Fernleaf’), and parsley (Petroselinum crispum L. ‘Giant of Italy’) supplied by a commercial seed company

(Johnny’s Seeds, Winslow, ME) were sown (two seeds per cell) into 276-cell phenolic foam propagation cubes (Oasis Horticubes XL; Smithers-Oasis, Kent, OH) initially saturated with deionized water and allowed to drain, placed in an environmental growth chamber (E-41;

Percival Scientific, Perry, IA), and irrigated with deionized water. During seed germination and seedling growth, air average daily temperature (ADT) was 23 ± 1.3 °C with a 24-h set point of

23 °C. A combination of fluorescent and incandescent lights provided 443 ± 17 μmol∙m–2∙s–1 from 0600 to 2200 HR throughout propagation. Light was measured with a quantum light sensor

(SQ-222; Apogee Instruments Inc., Logan, UT) and air temperature with a thermocouple

(TMC1-HD; Onset, Bourne, MA) in a naturally aspirated solar shield (RS3; Onset) every 15 s in a growth chamber and averages were recorded by a data logger (HOBO U12 PPFD/Temp;

Onset, Bourne, MA) every 15 min. Seedlings were irrigated daily with deionized water supplemented with 100 mg∙L–1 nitrogen (N) supplied from a complete, balanced, water-soluble fertilizer (Jack’s Hydro Feed 16N–1.8P–14.3K; JR Peters, Allentown, PA). Seedlings were grown for 2 (basil) or 3 (parsley and dill) weeks prior to transplanting to greenhouse.

Hydroponic culture and greenhouse environment

Each cube was thinned to a single seedling per cell and individually transplanted into one of nine deep-flow technique (DFT) hydroponic units. Supporting water media was contained by a plastic open tank measuring 15 cm H × 91 cm W × 182 cm L, with a 227 L capacity (Active 48

Aqua; Hydrofarm, Petaluma, CA). Initial nutrient solutions were comprised of deionized water

–1 supplemented with 53 mg of MgSO4 and brought an electrical conductivity (EC) of 1.5 dS∙m by water-soluble fertilizer (Hydro FeED; JR Peters). Twenty four plants of each species were placed into 6-cm diameter black net pots (FarmTek, Dyersville, IA) inserted into 3.5-cm holes spaced 15 cm apart on center on an extruded polystyrene foam raft (Scoreboard; Dow, Midland,

MI) placed directly on the nutrient solution such that the foam cubes were in direct contact with nutrient solution. Air was supplied to six 15 cm-long air stones (Active Aqua) in each DFT system by a 110-L air pump (Active Aqua; Hydrofarm). Another pump (Aqua 30-W, Active

Aqua) circulated water through water heater/chillers (SeaChill TR-10; TECO, Terrell, TX) to maintain the nutrient solution at 22.5 °C. Nutrient solutions of each DFT system were monitored daily with a pH/EC meter (HI 9813-6; Hanna Instruments, Woonsocket, RI) and adjusted using phosphoric and citric acid (pH Down; General Hydroponics, Sebastopol) or potassium bicarbonate (JR Peters) to maintain a target pH of 6.0, while EC was maintained at 1.5 using a concentrated stock solution of water-soluble fertilizer (Hydro FeED; JR Peters).

All systems were placed in a glass-glazed greenhouse located in Ames, IA (42° N). Day and night temperature set points were 24 and 20 °C respectively, and were maintained with radiant hot water heating and fog cooling and controlled using an environment computer controller (Titan; ARGUS Control Systems, Surrey, BC, Canada). Automated greenhouse controls (Titan; ARGUS) provided adjustable shading using an aluminized shade cloth with 41% diffused light transmission (XLS 15 Revolux; Ludvig Svensson, Kinna, Sweden) on motorized rollers, resulting in a mean daily photosynthetic photon flux density (PPFD) of 8.3 ± 2.1 mol∙m–

2∙d–1.

Supplemental lighting treatments

Ambient greenhouse PPFD was supplemented with 100 μmol∙m–2∙s–1 by lighting treatments between 0600 and 2200 HR. Each DFT system was placed underneath a 400-W high- pressure sodium (HPS) lamp (PL 3000; P.L. Light Systems, Beamsville, ON, Canada), or LEDs with low B:R (B – 450 nm peak λ; R – 670 nm peak λ) light ratio of 7:93 (DR/B –Low Blue–LB,

GreenPower LED Toplighting; Philips, Eindhoven, Netherlands) or high B:R at 30:70 (DR/B-

High Blue–HB GreenPower LED Toplight; Philips; Fig. 1). Mapping of light distribution across the growing area was done by recording light intensity every 10 cm across the surface of the raft using a spectroradiometer (PS-100; Apogee Instruments, Logan, UT), adjusting lamp height, and wrapping lights with aluminum screening until the PPFD across treatment areas reached 102 ± 2

μmol∙m–2∙s–1. Individual light treatments were isolated from one another by a 6 mm-thick black and white plastic sheeting (Hydrofarm) folded over to create partitions extending from above lamp height to below growing media.

Thermistors (CS215; Campbell Scientific, Logan UT) inside an aspirated radiation shield

(TS100; Apogee Instruments) and quantum sensors (LI-190R; LICOR, Lincoln, NE) placed in the middle of each DFT system were connected to a data logger (CR1000; Campbell Scientific), and measured temperature and PPFD, respectively, every 30 s. Data was averaged and recorded by the data logger hourly, with daily averages automatically calculated and logged.

Plant growth data collection and calculation

Photosynthesis (Pn), stomatal conductance (gs), and transpiration (E) and were measured by an infrared gas analyzer (LX-6400XT; LI-COR Biosciences, Lincoln, NE) with a 6-cm2 clear leaf chamber 20 d (basil) or 27 d (dill and parsley) after transplanting seedlings. Reference CO2 concentration in the gas analyzer chamber was 500 μmol∙mol–1 and water vapor was maintained 50 at 8 mmol. Leaf chamber temperature was aspirated to greenhouse ambient conditions as previously described. Chlorophyll fluorescence was sampled and averaged from two plants within each repetition; measured on a fully expanded leaf approximately three nodes below the apical meristem using a portable flurometer (Handy PEA; Hansatech, Pentney, Norfolk, UK) for dark-adapted measurements. Leaves were dark adapted by placing a clip with shutter

(Hansatech) for 15 min before securing the flurometer to the clip and opening the shutter.

Flurometer measurements were done by exciting tissue for one second with increasing light up to

3512 μmol∙m–2∙s–1 to saturate photosystem II (PSII) and data were expressed as the change in fluorescence to max fluorescence (Fv/Fm). To measure the effect of day length across treatments, gas exchange and chlorophyll fluorescence were measured in the morning (between 0600 and

0800 HR), and end of day (between 2000 and 2200 HR).

Growth measurements and destructive harvests were performed 21 d (basil) or 28 d (dill and parsley) after initiating treatments. Heights were recorded from the length from the surface of the foam board to the apical meristem. Plants were severed at the substrate surface and shoot fresh mass (SFM) was immediately recorded. Fresh tissue was then tripled-rinsed in deionized water after weighing and placed in a paper bag in a forced-air oven at 67°C for 3 d, after which shoot dry mass (SDM) was recorded.

Three additional plants harvested from each treatment repetition were used for chlorophyll content, aroma, and phenolic concentration analysis. Preparation of samples for chlorophyll analysis was done by cutting ≈150 mg of leaf tissue into 1.5 mL centrifuge tubes and flash freezing in liquid nitrogen. Samples were then ground into a powder within their respective centrifuge tubes and aliquots of 1.5 mL of reagent-grade ethanol were added before storing at 4

°C until further analysis. Samples were spun down for 2 min at 5000 g in a centrifuge (5415 C; 51

Eppendorf, Hamburg, Germany) before transferring 750 µL of supernatant into a new tube and adding an additional 750 µL of ethanol. Chlorophyll content was quantified by spectrophotometer (Genesys 20 Visible Spectrophotometer; Spectronic Instruments Inc.,

Rochester, NY) at 665 and 649 λ for chlorophyll a and b, as described by Ritchie (2006).

Aroma and flavor analyses

Approximately 2 to 3 g of leaf tissue from up to three plants within each rep were placed into glass sampling jars for 30 min, and were then analyzed for aroma compounds by a gas chromatograph (Model 3700; Varian, Palo Alto, CA) with integrator (3390A Integrator; Hewlett

Packard, Palo Alto, CA) using a 30 m column (DB5; J&W Scientific, Folsom, CA) following method by Wilson et al. (1992), set to an initial temperature of 30°C for 2 min, followed by single-step ramping of temperature at 7°C per min until 200°C was reached, then held for 5 min.

Peaks were identified using standards for predominate aroma compounds common to each species (Justesen and Knuthsen, 2001; Tucker and DeBaggio, 2009; USDA, 2011; van Wyk,

2014). Phenolic concentration of key flavor compounds were prepared by extracting analytes from plant material into a methanol solvent solution for sample preparation as described by

Khoddami et al. (2013). In order to reduce particle size for better extraction, 1 to 2 g of plant material were cut from mature leaves, weighed and inserted into a 15 mL centrifuge tube with a milling ball (10 mm PM 100 planetary mill grinding ball; Retsch Technology GmbH, Haan,

Germany) and flash-frozen in liquid nitrogen. Frozen samples were then shaken and vortexed until tissue was ground into a fine powder. Aliquots of 7.5 mL of methanol solvent were added to each centrifuge tube and then vortexed, left at room temperature prior to filtration through filter paper (Whatman No. 2 filter paper; GE Healthcare UK Limited, Amersham Place Little

Chalfont, UK), and transferred into a 2-mL auto sampler vial. 52

Samples were analyzed by quadruple time-of-flight high-performance liquid chromatography (6450 QTOF HPLC; Agilent, Santa Clara, CA) with an LC column (XDB C18,

4.6*150 mm, 1.8 μ; Agilent) at a flow rate of 700 μL∙min–1 and temperature maintained at 30 °C.

Samples were injected at a volume of 5 μL with a solvent gradient of 95% solvent A (1:1

H20/MeOH with 1% formic acid) to 5% solvent B (ACN with 0.1% formic acid) held for 5 min and then changing to 5% solvent A to 95% solvent B over 10 min and held for 3 min. Analytes were read at a UV wavelength of 280 nm.

Experimental design and statistical analyses

The experiment was designed as a randomized complete block design for each species.

There were three replications (individual DFT systems) for each supplemental light treatment, with ten subsamples (individual plants) averaged in each replicate for growth measurements, and three subsamples averaged for physiological measurements. Replications were blocked across three runs over time. Data was analyzed using two-way ANOVA (α = 0.05) in SigmaPlot

(SigmaPlot 11.0; Systat Software, San Jose, CA). Tukey honestly significant difference (HSD) was subsequently used for mean separation.

Results

Basil

Light source affected morphology differently, with one or both LED lighting treatments resulting in shorter plants with less SFM compared to plants grown under HPS lamps. For example, basil height grown under both LEDs were 2.5 cm shorter compared to those under HPS lighting (21.3 cm). While HB-grown basil had SFM of 20.1 g, 3.0 g less than those under HPS,

SDM of basil was similar across all light source treatments (Fig. 2). 53

Over the course of the day, gas exchange was higher in either one or both LED-grown basil compared to HPS-grown. In the morning Pn was 19% and 21% higher in HB- and LB- grown basil compared to HPS, respectively (Fig. 3), though by evening only HB-treated plants had higher Pn than HPS-grown plants (37%), and had the least reduction in Pn over the course of the day (9%) compared HPS (21%) (Fig. 3). Similarly, gs for basil in the evening was 31% higher for plants grown under HB compared to HPS. While E was successively higher across treatments in the morning for LB- and HB-grown basil compared to HPS, only basil grown under

HB had higher E (33% increase) than those under HPS lighting in the evening.

The efficiency of PSII, denoted by Fv/Fm, was similar for all treatments in the morning

(data not shown). By the evening, Fv/Fm declined slightly with LB-grown basil having a lower

Fv/Fm (0.80) compared to those grown under HPS lighting (0.82; Fig. 5). Although chlorophyll a had similar concentrations across lighting treatments (3.93 µg·mg–1), chlorophyll b of basil grown under LB (3.60 µg·mg–1) increased by 30% and 42% compared to those grown under HB and HPS, respectively (Fig. 6).

In addition to morphology and gas exchange, secondary metabolites of essential oils and phenolics accumulation for basil were affected by supplemental light, and tended to increase under LED lighting. For example, myrecene was 134% higher in HB-grown plants compared to

HPS (Fig. 10). Similarly, both LED-grown basil had higher concentrations myricetin and orientin, but were lower in kaempferol, compared to those grown under HPS. Only HB-grown plants had higher concentrations of isorhmnetin compared to HPS-grown (Fig. 7). Myricetin exhibited a 2-fold increase in concentration in basil grown under either LED light source compared to HPS lighting respectively.

Dill

Height, SFM, and SDM of dill were 24.2 cm, 12.0 g, and 1.22 g respectively, and were unaffected by supplemental light source (Fig. 2). Chlorophyll a concentrations were similar across treatments. However, concentrations of chlorophyll b were higher in LB treated plants compared to either HB or HPS by 28% and 26%, respectively (Fig. 6).

Essential oil accumulation in dill were more affected by LB lighting than any other light source, and although some trends for higher essential oil content was observed under LED lighting compared to HPS (Fig. 10), LB-grown dill had 322% more myrecene and 285% more cineole than under HPS lighting. Flavonoid concentrations were generally higher in LED-grown dill, with a trend of higher phenolic content with increasing B light content, except for kaempferol, which was 84% higher in LB treated than HPS. Flavonoid concentrations of quercetin, myricetin, myristicin, and dillapiole were 177%, 98%, 128%, and 89% higher in dill grown under HB lighting compared to HPS lighting, respectively (Fig. 8). Notably, quercetin concentration was 2.8-times higher in plants grown under HB compared to HPS.

Parsley

Light source and quality affected parsley growth as both LB- and HB-grown plants were shorter than HPS-grown, but only HB-grown basil and parsley had less SFM than HPS. For instance, height of parsley under LED lights were 2.7 cm shorter compared to those under HPS

(22.0 cm; Fig. 2). Similarly, SFM of HB-grown parsley were 2.4 g less than HPS (13.2 g). The

SDM of parsley was similar across lighting treatments.

Morning Pn was higher for parsley under LB and HB LEDs compared to HPS-grown plants by 29% and 30%, respectively (Fig. 4), and remained higher into the evening. While in the morning the gs of parsley was 3% higher under HB light than LB, gs were 37% higher under HB than HPS. By evening both HB and LB-grown plants had 19% and 20% more gs than HPS- 55 grown plants, respectively. Transpiration was highest in HB-grown parsley in the morning with

4% and 26% higher E than LB or HPS respectively, though by evening both LED-grown parsley had higher E than plants under HPS. In contrast, Fv/Fm of parsley was 0.84 across treatments in the morning (data not shown), though LB- and HB-grown plants decreased to 0.825 by evening, and were slightly lower than HPS (0.834) (Fig. 5). Although chlorophyll a content of parsley

(5.77 µg·mg–1) was similar across lighting treatments, chlorophyll b in parsley was highest in

LB-grown plants, 22% greater than HPS-grown plants.

Parsley grown under LED also had higher concentrations of essentials oils, with 343% and 248% higher myrecene accumulation compared to HPS-grown plants for LB- or HB-grown parsley, respectively (Fig. 10). Additionally, while there were trends of greater flavonoid concentrations LED-grown parsley compared to HPS-grown plants, only isohamnetin was significantly affected by light source, with a 90% increase in concentration for HB-grown plants compared to HPS-grown plants (Fig. 9).

Discussion

Supplemental lighting from HPS lamps is ubiquitous in greenhouse crop production.

However, the results of our research, with respect to plant growth, development, and quality, demonstrate the potential benefits of using narrow-spectra LED lighting for efficiently producing crops and adding value compared to traditional broad-spectrum, high-intensity light sources.

Although basil (LB- and HB-grown) and parsley (HB-grown) were shorter and both species under HB had less SFM compared to plants grown under HPS light, they are minor with respect to implications for commercial CEA crop production. Furthermore, for many of the other parameters analyzed, such as SDM, gas exchange, and aroma and flavor compounds, plants 56 grown under LED lighting had similar or improved characteristics to those grown under traditional HPS lighting.

Supplemental light source affected plant height differently among species; basil (LB and

HB) and parsley (HB) were shorter than those grown under HPS lighting, while dill was unaffected by light source. Parsley grown under HB were also shorter than under HPS light.

These results align well with other reports of plant height suppression under LED lighting compared to HPS. Wheeler et al. (1991) reported that soybean (Glycine max) stem length was suppressed when grown under HPS lighting when additional B light was added, and Hernández and Kubota (2016) reported accumulating degrees of height suppression for cucumbers grown under increasing amount of B light. While the magnitude of height suppression varies with species, LEDs produce more compact radishes (Raphanus sativus; Cope et al., 2014), petunias

(Petunia × hybrida; Currey and Lopez, 2013), cannabis (Cannabis sativa; Lalge et al., 2017), and sweet basil compared to plants grown under broad-spectrum lighting (Fraszczak et al.,

2014). Stem elongation is associated with several mechanisms including etiolation and shade avoidance. Cell elongation of stem tissue results in internode elongation, ultimately contributing to overall stem length (Litvin et al., 2016), and this elongation can be reduced under high light intensity (Bell and Galloway, 2008; Huber et al., 1998). Additionally, shorter stem lengths occurs when R:FR is high (Schmitt and Wulff, 1993), similar to this study wherein the LED lamps do not emit FR light, while HPS lamp do. Both basil and parsley were shorter and had less

SFM under one or both of the LEDs compared to those under HPS lighting, but the number of nodes (data not shown) were similar across lighting treatments. This suggests minimal, if any, advantage for additional structural growth development of HPS-grown plants. While increased stem elongation may result in a larger plant, this does not necessarily mean increased SDM, and 57 in fact is undesirable for some, such as with producers of containerized ornamental plants

(Randall and Lopez, 2015).

As Pn increases, generally so does gs and E, and were all at least similar or greater for

LED-treated basil and parsley at time of the day compared to HPS (Fig. 2 and 3), despite less

SFM under HB than HPS light. Within each species, plants had similar SDM across treatments, and the differences in SFM can be attributed to varying water content. Gas exchange can increase with B light; with basil, Pn under HB and LB LEDs up to 39% and 35% higher, and gs

21% and 14% higher over the course of the day compared to plants under HPS, respectively.

With up to 95% of water uptake transported directly for gas exchange (McElrone et al., 2013), and available internal water content driving turgor pressure cell expansion (Cosgrove, 2000) for internode elongation under light (Huber et al., 2014), the increased gas exchange, measured particularly under HB light, may have contributed to shorter height and less SFM under HB light for basil and parsley compared to HPS. Yet differences in species responses to light sources vary

(van Iersel and Gianino, 2017), as seen in this study where gs and E were increasing greater under LB and HB LEDs than under HPS for basil than with parsley (Fig. 2 and 3). In the morning, gas exchange is influenced by B light photoreceptors, such as cytochromes and phototropins, thought to aid gs by playing a role in promoting guard cell opening to initiate gas exchange in the morning (Briggs and Huala, 1999; Humble and Hsiao, 1970; Taiz and Zeiger,

2010). While Pn occurs across PAR spectra (McCree, 1972; Inada, 1976; Bugbee, 2016), the role of specific light harvesting complexes underscores the effect of light quality interactions on gas exchange. Thus, while Pn was generally higher in both LED treatments for basil and parsley, gs and E were highest for basil in the morning under HB, followed by LB, and then HPS (Fig. 2), while parsley grown under LB was similar to HPS-grown plants (Fig. 3). These results are 58

similar to reports on B light influence on gs (Hogewoning et al., 2010) and differences among species to B light sensitivity (Dougher and Bugbee, 1998; Reymond et al., 1992). In the evening, the LEDs with greater B light sustained higher Pn, gs, and E and compared to basil grown under

LB or HPS light. Parsley were less sensitive to B light, and had generally higher gas exchange under either LED light source compared to HPS.

The relationship of Pn to Fv/Fm, and plants’ ability to use light for photochemical processes depends on the quantum yield efficiency of PSII (Genty et al., 1989), with CO2 assimilation related intrinsically to PSII efficiency. As Pn increases, or is maintained high, Fv/Fm can decrease (Genty et al., 1989; Zhen and van Iersel, 2017), especially over the course of the day (van Iersel et al., 2016). Although Fv/Fm of LB-grown basil and parsley grown under both

LEDs were lower than HPS by evening (Fig. 4) the effect was minimal, with plants grown under

LED lamps typically having similar or greater gas exchange than HPS. Only plants grown under

HPS lighting did not experience a drop in Fv/Fm from morning to evening, potentially suggesting underutilization of light in HPS-grown plants. Chlorophyll content can increase inversely to

Fv/Fm levels under increasing B light (Litvin, 2019). However, although chlorophyll a remained largely unchanged across treatments (Fig. 5), chlorophyll b was higher under LB light for basil and dill compared to both HB and HPS lighting, and parsley under LB compared to HPS.

Chlorophyll b can increase when there is under-utilization of light as with shade (Dai et al.,

2009), or to maintain gas exchange under decreasing Fv/Fm (Genty et al., 1989). We postulate that the increase in chlorophyll b under LB compared to HPS lighting may be due in part to lower Fv/Fm values, indicating less efficient photochemical quenching of incident light, resulting in either the promotion of additional chlorophyll b, or reduced destruction of existing chlorophyll b; however, further studies are needed to determine the effect of chlorophyll biosynthesis and 59 catabolism. Additionally, some light-mediated stresses induced may increase value-added attributes of herbs.

Accentuation of aroma and flavor attributes of culinary herbs are considered desirable for consumers (Cook and Samman, 1996; van Wyk, 2014) and practices that increase their accumulation, thereby adding value to products would thus benefit growers. In general, accumulation of phenolic compounds varied across species in their response to light source, with basil and parsley having increased isorhamnetin under HB. However, basil had lower kaempferol under either LED light compared to HPS (Fig. 6), while there was a more kaempferol for LB- grown dill compared to HPS (Fig. 7); kaempferol content of parsley was unaffected by light sources. (Fig. 8). Myricetin and orientin, two major compounds found in basil, were higher under

LED lighting compared to HPS, but were similar for vicenin. In addition to variation among species in responses to lighting, we found varying trends among compounds within a species across light treatments. This variation highlights the importance of careful light selection based on the species produced and objectives for providing supplemental.

Secondary metabolites responsible for aromatic and flavor qualities of plants are generally part of stress mitigation mechanisms designed to protect plants from abiotic and biotic stress conditions (Petrussa et al., 2013; Winkel-Shirley, 2002). Flavonoid compounds such as kaempferol, quercetin, and others are desirable phenolics for their benefits to human health

(Hollman, 2001; Huang et al., 2009; Pandey & Rizvi, 2009), and are promoted in plants via the phenylpropanoid pathway (Petrussa et al., 2013; Stahlhut et al., 2015). Although several secondary metabolites of essential oils and phenolics may aide in plant stress mitigation (Dixon and Paiva, 1995; Rehman et al., 2016; Winkel-Shirley, 2002), compounds can be explicit to specific stress signals (Dixon and Paiva, 1995; Ma et al., 2014). Likewise, the phenylpropanoid 60 pathway can synthesize kaempferol as a final product or conversely, synthesize quercetin, depending on the signal triggering the response (Ryan et al., 2002; Winkel-Shirley, 2002), yet the increase in both compounds as seen in dill under increasing B light was unexpected, though

Warren et al. (2003) did report increases of both in response to UV-B radiation, and Ryan et al.

(1998) noted that for different species quercetin:kaempferol ratios may not be dissimilar even if total content increased under light stress. Similarly to basil, some major compounds found in dill did increase under HB alone (dillapiole) or both LED light sources (myristicin), agreeing with reports on B light-mediated promotion of phenolics (Dixon and Paiva, 1995). Although parsley exhibited some tendencies for higher concentrations under HB, only isorhamnetin was higher in

HB-grown plants compared to HPS. Previous reports on the sensitivity of parsley to light source for promotion of flavor compounds is currently lacking, but Bugbee (2016) reported the interaction of light quality and light intensity may dictate the magnitude of plant responses. We postulate that light quality may in fact affect parsley flavonoid accumulation, but at higher light intensities than those used in the present experiment.

Accumulation of essential oils in plants can be are mediated, in part, by light intensity and quality, as seen in basil (Chang et al., 2008; Fernandes et al., 2013; Hammock, 2018), by air quality (Blande et al., 2014), and natural variation across species (Rehman et al., 2016).

Compared to plants grown under HPS, myrecene was higher in LB for dill, HB for basil, or both

LED treatments for parsley, while cineole highest under LB for dill (Fig. 9). However, headspace analysis for relative essential oil content did not differ for linalool, limonene, or carvone within this study. Although there appeared to be trends of increased essential oil accumulation in response to LEDs, or the proportion of B light among LED light sources, variance in headspace data computed may have masked some possible treatment effects. 61

Conclusion

The use of narrowband LED light sources may slightly suppress the final height and yields compared to HPS lighting, but the advantages for increased light use efficiency of incident radiation, and accumulation of aroma and flavor in herbs highlights the benefits for their use in commercial production to add value by increasing product quality. Our study compared the effects supplemental lighting from broad- and narrow-spectra light sources on growth, morphology, gas exchange, and secondary metabolite concentrations of culinary herbs. We believe much of the effects reported herein are related to the amount of B light from the different light sources. The spectral distribution of PAR may be broadly characterized for color (B: 400–

500 nm; G: 500–600 nm; R 600–700 nm, Cope et al., 2014; Runkle, 2007). Based on this categorization, the proportion B light in our HPS lights is 6%. However, classifying the spectral distribution of our supplemental light sources by the peak wavelengths for absorption across the

PAR action spectrum for plants, and of the LEDs used (B: 450 ± 20 nm; R: 660 ± 20 nm), similar to foci of B and R light with bandwidth ranges reported by Chen et al. (2016), and encompassing the range of other reported spectral peaks (Inada, 1976; Kopsell et al., 2015; Mickens et al.,

2018; Morrow, 2008; Randal and Lopez, 2015), the B light in HPS lamps is 2.7% B light. As a result, we see a dose-response effect of supplemental B light on gas exchange, growth and morphology, and secondary metabolic functions. Further studies exploring the specific role of B light and culinary herbs would be useful to more clearly define the role of this light on culinary herb physiology.

References

Baroli, I., G.D. Price, M.R. Badger, and S. von Caemmerer. 2008. The contribution of photosynthesis to the red light response of stomatal conductance. Plant Physiol. 146:737–747. 62

Beaman, A.R., R.J. Gladon, and J.A. Schrader. 2009. Sweet basil requires an irradiance of 500 µmol∙m−2∙s−1 for greatest edible biomass production. HortScience 44:64−67.

Bell. D.L. and L.F. Galloway. 2008. Population differentiation for plasticity to light in an annual herb: adaptation and cost. Amer. J. Bot. 95:59–65.

Blande, J.D., J.K Holopainen, and Ü. Niinemets. 2014. Plant volatiles in polluted atmospheres: stress responses and signal degradation. Plant Cell Environ. 37:1892–1904.

Briggs, W.R. and E. Huala. 1999. Blue-light photoreceptors in higher plants. Ann. Rev. Cell Dev. Biol. 15:33–62.

Brown, C. and S. Miller. 2008. The impact of local markets: a review of research on farmer markets and community supported agriculture (CSA). Am. J. Agr. Econ. 90:1298–1302.

Brown, S. 1991. Culinary herb use in southern California restaurants. CA Agric. 45:4−6.

Bugbee, B. 2016. Toward an optimal spectral quality for plant growth and development: the importance of radiation capture. Acta Hort. 1134:1–12.

Chang, X., P.G. Alderson, and C.J. Wright. 2005. Effect of temperature integration on the growth and volatile oil content of basil (Ocimum basilicum L.). J. Hort. Sci. Biotechnol. 80:583−598.

Chang, X., P.G. Alderson, and C.J. Wright. 2008. Solar irradiance level alters the growth of basil (Ocimum basilicum L.) and its content of volatile oils. Environ. Expt. Bot. 63:216–223.

Chen, X., X. Xue, W. Guo, L. Wang, X. Qiao. Growth and nutritional properties of lettuce affected by mixed irradiation of white and supplemental light provided by light-emitting diode. Scientia Hort. 200:111–118.

Cook, N.C. and S. Samman.1996. Flavonoids- chemistry, metabolism, cardioprotective effects, and dietary sources. Nutr. BioChem. 7:66–76.

Cope, K.R., M. Chase, and B. Bugbee. 2014. Photobiological interactions of blue light and photosynthetic photo flux: effects of monochromatic and broad-spectrum light sources. Photochemistry Photobiology 90:574–584.

Cosgrove, D.J. 2000. Loosening of plant cell walls by expansins. Nature 407:321–326.

Craver, J.K., J.R. Gerovac, R.G. Lopez, and D.A. Kopsell. 2017. Light intensity and light quality from sole-source light emitting diodes impact phytochemical concentrations within Brassica microgreens. J. Amer. Soc. Hort. Sci. 142:3–12. 63

Dai, Y., Z. Shen, Y. Liu, L. Wang, D. Hannaway, and H. Lu. 2009. Effects of shade treatments on the photosynthetic capacity, chlorophyll fluorescence, and chlorophyll content of Tetrastigma hemsleyanum Diels et Gilg. Environ. Expt. Bot. 65:177–182.

De Pascale, S., A. Maggio, F. Orsini, and G. Barbieri. 2006. Nutrients influence on ready to eat sweet basil quality. Acta Hort. 718:523−530.

Dixon, R.A. and N.L. Paiva. 1995. Stress-induced phenylpropanoid metabolism. Amer. Soc. Plant Phsiol. 7:1085–1097.

Dougher, T.A. and B.G. Bugbee. 1998. Is blue light good or bad for plants? Life Support Biosphere Sci. 5:129–136.

Farrell, K.T. 1999. Spices, condiments and seasonings. Aspen Publishers, Gaithersburg, MD.

Franklin, K.A. and G.C. Whitelam. 2005. Phytochromes and shade-avoidance response in plants. Ann. Bot. 96:169–175.

Fraszczak, B., A. Golcz, R. Zawirska-Wojtasiak, and B. Janowska. 2014. Growth rate of sweet basil and lemon balm plants grown under fluorescent lamps and LED modules. Acta Sci. Pol. Hortorum Cultus, 13:3–13.

Hammock, H.A. 2018. The impact of blue and red LED lighting on biomass accumulation, flavor volatile production, and nutrient uptake in hydroponically grown Genovese basil. MS Thesis, Univ. of TN, Knoxville. 64

Hernández, R. and C. Kubota. 2012. Tomato seedling growth and morphological responses to supplemental led lighting red:blue ratios under varied daily solar light integrals. Acta Hort. 956:187–194.

Hernández, R. and C. Kubota. 2015. Physiological, morphological, and energy-use efficiency comparisons of LED and HPS supplemental lighting for cucumber transplant production. HortScience. 50:351–357.

Hernández, R. and C. Kubota. 2016. Physiological responses of cucumber seedlings under different blue and red photon flux ratios using LEDs. Environ. Expt. Bot. 121:66–74.

Hogewoning, S.W., G. Trouwborst, H. Maljaars, W. van Ieperen, and J. Harbinson. 2010. Blue light dose-responses of leaf photosynthesis, morphology, and chemical composition of Cucumis sativus grown under different combination of red and blue light. J. Expt. Bot. 61:3107–3117.

Hollman, P.C.H. 2001. Evidence for health benefits of plant phenols: local or systemic effects? J. Sci. Food Agric. 81:842–852.

Huang, W.Y., Y.Z. Cai, and Y. Zhang. 2009. Natural phenolic compounds from medicinal herbs and dietary plant: potential use for cancer prevention. Nutr. Cancer. 62:1–20.

Huber, H., A. Fijan, and H.J. During. 1998. A comparative study of spacer plasticity in erect and stoloniferous herbs. Oikos 81:576–586.

Huber, H., J. De Brouwer, E. von Wettberg, H. During, and N. Anten. 2014. More cells, bigger cells or simply reorganization? Alternative mechanisms leading to changed internode architecture under contrasting stress regimes. New Phytol. 20:193–204.

Humble, G.D. and T.C. Hsiao. 1970. Light-dependent influx and efflux of potassium of guard cells during stomatal opening and closing. Plant Physio. 46:483–487.

Inada, K. 1976. Action spectra for photosynthesis in higher plants. Plant Cell Physiol. 17:355– 365.

Jensen, M.H. 1999. Hydroponics worldwide. Acta Hort. 481:719–730.

Justesen, U. and P. Knuthsen. 2001. Composition of flavonoids in fresh herbs and calculation of flavonoid intake by use of herbs in traditional danish dishes. Food Chem. 73:245–250.

Khoddami, A., M.A. Wilkes, and T.H. Roberts. 2013. Techniques for analysis of plant phenolic compounds. Mol. 18:2328–2375.

Kiple, K.F. and K.C. Ornelas. 2000. The Cambridge world history of food. Eds. Cambridge Uni. Press, Cambridge. 65

Kopsell, D.A., C.E. Sams, and R.C. Morrow. 2015. Blue wavelengths from LED lighting increase nutritionally important metabolites in specialty crops. HortScience 50:1285–1288.

Lalge, A., P. Cerny, V. Trojan, and T. Vyhnanek. 2017. The effects of red, blue and white light on the growth and development of Cannabis sativa L. In XXIV Conf. Mendel Net 8:646–651.

Li, Q. and C. Kubota. 2009. Effects of supplemental light quality on growth and phytochemicals of baby leaf lettuce. Environ. Expt. Bot. 67:59–64.

Litvin, A.G. 2019. Light intensity and quality affect growth, morphology, gas exchange, and secondary metabolite accumulation in culinary herbs. Ph.D. Diss., Iowa State Univ., Ames.

Litvin, A.G., M.W. van Iersel, and A. Malladi. 2016. Drought stress reduces stem elongation and alters gibberellin-related gene expression during vegetative growth of tomato. J. Amer. Soc. Hort. Sci. 141:591–597.

Lorrain, S., T. Allen, P.D. Duek, and G.C. Whitelam. 2008. Phytochrome-mediated inhibition of shade avoidance involves degradation of growth-promoting bHLH transcription factors. Plant J. 53:312–323.

Massa, G.D., H. Kim, R.M. Wheeler, and C.A. Mitchell. 2008. Plant productivity in response to LED lighting. HortScience. 43:1951–1956.

Massa, G.D., J.C. Emmerich, R.C. Morrow, C.M. Bourget, and C.A. Mitchell. 2006. Plant- growth lighting for space life support: a review. Gravitational Space Biol. 19:19–29.

McCree, K.J. 1972. Test of current definitions of photosynthetically active radiation against leaf photosynthesis data. Agr. Meteorol. 10:443–453.

McElrone, A.J., B. Choat, G.A. Gambetta, and C.R. Brodersen. 2013. Water uptake and transport in vascular plants. Nature Educ. Knowledge 4:6.

Mickens, M.A., E.J. Skoog, L.E. Reese, P.L. Barnwell, L.E. Spencer, G.D. Massa, and R.M. Wheeler. 2018. A strategic approach for investigating light recipes for ‘Outredgeous’ red romaine lettuce using white and monochromatic LEDs. Life Sci. Space Res. 19:53–62

Moe, R., S.O. Grimstad, and H.R. Gislerod. 2006. The use of artificial light in year round production of greenhouse crops in Norway. Acta Hort. 711:35–42.

Morales, M.R. and J.E. Simon. 1996. New basil selections with compact inflorescences for the ornamental market, p. 543−546. In: J. Janick (ed). Prog. New Crops. ASHS Press, Arlington, VA.

Morrow, R. 2008. LED lighting in horticulture. HortScience. 43.7. 2008; 1947–50.

Pandey, K.B. and S.I. Rizvi. 2009. Plant polyphenols as dietary antioxidants in human health and disease. Oxidative Medicine Cellular Longevity 2:270–278.

Paton, A. 1992. A synopsis of Ocimum L. (Labiatae) in Africa. Kew Bul. 47:403–435.

Petrussa, E., E. Braidot, M. Zancani, C. Peresson, A. Bertolini, S. Patui, and A. Vianello. 2013. Plant flavonoids– biosynthesis, transport and involvement in stress responses. Int. J. Mol. Sci. 14:14950–14973.

Piovene, C., F. Orsini, S. Bosi, R. Sanoubar, V. Bregola, G. Dinelli, and G. Gianquinto. 2015. Optimal red:blue ratio in led lighting for nutraceutical indoor horticulture. Scientia Hort. 193:202–208.

Rehman, R., M.A. Hanif, Z. Mushtaq, and A.M. Al-Sadi. 2016. Biosynthesis of essential oils in aromatic plants: a review. Food Rev. Int. 32:117–160.

Reymond, P., T.W. Short, and W.R. Briggs. 1992. Blue light activates a specific protein kinase in higher plants. Plant Physiol. 100:655–661.

Ritchie, R.J. 2006. Consistent sets of spectrophotometric chlorophyll equations for acetone, methanol, and ethanol solvents. Photo. Res. 89:24–41.

Runkle, E.S. 2007. Light quality defined. GPN. 17:82.

Schmitt, J. and R.D. Wulff. 1993. Light spectral quality, phytochrome and plant competition. Trends in Ecol. Evo. 8:47–51.

Shacklock, P.S., N.D. Read, and A.J. Trewavas. 1992. Cytosolic free calcium mediates red light- induced photomorphogensis. Nature. 358:753–755. 67

Sharrock, R.A. 2008. The phytochrome red/far-red photoreceptor superfamily. Genome Biol. 9:230

Simon, J.E., M.R. Morales, W.B. Phippen, R.F. Viera, and Z. Hao. 1999. Basil: a source of aroma compounds and a popular culinary and ornamental herb. In Janick, J. Eds., Perspectives on new crops and new uses. ASHS Press, Alexandria, VA.

Stahlhut, S.G. S. Siedler, S. Malla, S.J. Harrison, J. Maury, A.R. Neves, and J. Forster. 2015. Assembly of a novel biosynthetic pathway for production of the plant flavonoid fisetin in Escherichia coli. Metabolic Eng. 31:84–93.

Taiz, L. and E. Zeiger. 2010. Plant physiology, 5th ed. Sinauer Associates. Sunderland, MA.

Tixier, P. and H. de Bon. 2006. Urban horticulture. Cities Farming for the Future, Urban Agriculture for Green and Productive Cities. RUAF Foundation, IDRC and IIRR, Silang, The Philippines.

Tucker, A.O. and T. DeBaggio. 2009. The encyclopedia of herbs: a comprehensive reference to herbs of flavor and fragrance. Eds. 2 Timber Press. Portland, OR.

U.S. Dept. Agr. 2011. USDA database for the flavonoid content of selected foods. Release 3. Nutrient Data Laboratory. Accessed April 6 2019.

U.S. Dept. Agr. 2014. Census of horticultural specialties. Census of Agriculture. Accessed April 6 2019. https://www.nass.usda.gov/Publications/AgCensus/2012/Online_Resources/Census_of_Horticult ure_Specialties/hortic_1_001_001.pdf/ 68 van Iersel, M.W. and G. Gianino. 2017. An adaptive control approach for light-emitting diode lights can reduce the energy costs of supplemental lighting in greenhouses. HortScience 52:72– 77. van Iersel, M.W., G. Weaver, M.T. Martin, R.S. Ferrarezi, E. Mattos, and M. Haidekker. 2016. A chlorophyll fluorescence-based biofeedback system to control photosynthetic lighting in controlled environment agriculture. J. Amer. Soc. Hort. Sci. 141:169–176. van Wyk, B.E. 2014. Culinary herbs and spices of the world. Briza Publications. Pretoria, South Africa.

Walters, K.J. and C.J. Currey. 2015. Hydroponic greenhouse basil production: Comparing systems and cultivars. HortTechnology 25:645–650.

Wheeler, R.M., C.L., Mackowiak, and J.C. Sager. 1991. Soybean stem growth under high- pressure sodium with supplemental blue lighting. Agron. J. 83:903–906.

Wilson, L.A., N.P. Senechai, and M.P. Widrlechner. 1992. Headspace analysis of the volatile oils of Agastache. J. Agric. Food Chem. 40:1362–1366

Winkel-Shirley, B. 2002. Biosynthesis of flavonoids and effects of stress. Current Opinion Plant Biol. 5:218–223.

Wittwer, S.H. and N. Castilla. 1995. Protected cultivation of horticultural crops worldwide. HortTechnology 5:6–23.

Wu, M.C., C.Y. Hou, C.M. Jiang, Y.T. Wang, C.Y. Wang, H.H. Chen, and H.M. Chang. 2007. A novel approach of led light radiation improves the antioxidant activity of pea seedlings. Food Chem. 101:1753–1758.

Xu, Z. and G. Zhou. 2008. Responses of leaf stomatal density to water status and its relationship with photosynthesis in a grass. J. Expt. Bot. 59:3317–3325.

Zhen, S. and M.W. van Iersel. 2017. Photochemical acclimation of three contrasting species to different light levels: Implications for optimizing supplemental lighting. J. Amer. Soc. Hort. Sci. 142:346–354. 69

Fig. 1. Light spectra of supplemental light provided from high-pressure sodium (HPS) lamps or light-emitting diodes with a low blue (B) to red (R) ratio (LB; 7:93 B:R), or high B:R ratio (HB; 30:70 B:R). 70

Fig. 2. Height, shoot fresh mass (SFM), and shoot dry mass (SDM) of basil (Ocimum basilicum), dill (Anethum graveolens), and parsley (Petroselinum crispum) grown in deep-flow technique hydroponic systems under ambient light supplemented with 100 µmol·m–2·s–1 of supplemental light from 0600 to 2200 HR provided from high-pressure sodium (HPS) lamps or light-emitting diodes (LED) with a low blue (B) to red (R) ratio (LB; 7:93 B:R), or high B:R ratio (HB; 30:70 B:R). Data were collected 21 d (basil) or 28 d (dill and parsley) after transplanting and initiating treatments. Each bar represents the mean of 9 replications. Letters indicate significant differences across supplemental light sources within species using Tukey’s honestly significant difference (HSD; P≤0.05).

Fig. 3. Photosynthesis (Pn), stomatal conductance (gs), and transpiration (E) for basil (Ocimum basilicum) grown in hydroponic systems under ambient light supplemented with 100 µmol·m– 2·s–1 of supplemental light from 0600 to 2200 HR provided from high-pressure sodium (HPS) lamps or light-emitting diodes (LED) with a low blue (B) to red (R) ratio (LB; 7:93 B:R), or high B:R ratio (HB; 30:70 B:R). Measurements presented were taken during the morning (0600–0800 HR) and evening (2000–2200 HR) 20 d after initiating treatments. Each bar represents the mean of 9 replications. Letters indicate significant differences across supplemental light sources within species using Tukey’s honestly significant difference (HSD; P≤0.05). 72

Fig. 4. Photosynthesis (Pn), stomatal conductance (gs), and transpiration (E) for parsley (Petroselinum crispum) grown in hydroponic systems under ambient light supplemented with 100 µmol·m–2·s–1 of supplemental light from 0600 to 2200 HR provided from high-pressure sodium (HPS) lamps or light-emitting diodes (LED) with a low blue (B) to red (R) ratio (LB; 7:93 B:R), or high B:R ratio (HB; 30:70 B:R). Measurements presented were taken during the morning (0600–0800 HR) and evening (2000–2200 HR) 27 d after initiating treatments. Each bar represents the mean of 9 replications. Letters indicate significant differences across supplemental light sources within species using Tukey’s honestly significant difference (HSD; P≤0.05). 73

Fig. 5. Chlorophyll fluorescence (Fv/Fm) at evening (2000 – 2200 HR) after 20 d for basil (Ocimum basilicum) and 27 d for parsley (Petroselinum crispum). Plants were grown in hydroponic systems under ambient light supplemented with 100 µmol·m–2·s–1 of supplemental light from 0600 to 2200 HR provided from high-pressure sodium (HPS) lamps or light-emitting diodes (LED) with a low blue (B) to red (R) ratio (LB; 7:93 B:R), or high B:R ratio (HB; 30:70 B:R). Data were collected in the evening (2000–2200 HR) 20 d (basil) or 27 d (parsley) after transplanting and initiating treatments. Each bar represents the mean of 9 replications. Letters indicate significant differences across supplemental light sources within species using Tukey’s honestly significant difference (HSD; P≤0.05).

Fig. 6. Chlorophyll a and b content of basil (Ocimum basilicum), dill (Anethum graveolens), and parsley (Petroselinum crispum)grown in deep-flow technique hydroponic systems under ambient light supplemented with 100 µmol·m–2·s–1 of supplemental light from 0600 to 2200 HR provided from high-pressure sodium (HPS) lamps or light-emitting diodes (LED) with a low blue (B) to red (R) ratio (LB; 7:93 B:R), or high B:R ratio (HB; 30:70 B:R). Data were collected 21 d (basil) or 28 d (dill and parsley) after transplanting and initiating treatments. Each bar represents the mean of 9 replications. Letters indicate significant differences across supplemental light sources within species using Tukey’s honestly significant difference (HSD; P≤0.05).

Fig. 7. Relative content of isorhmanetin, kaempferol, quercetin, myricetin, orientin, and vicenin accumulated in basil (Ocimum basilicum). Plants were grown in deep-flow technique hydroponic systems under ambient light supplemented with 100 µmol·m–2·s–1 of supplemental light from 0600 to 2200 HR provided from high-pressure sodium (HPS) lamps or light-emitting diodes (LED) with a low blue (B) to red (R) ratio (LB; 7:93 B:R), or high B:R ratio (HB; 30:70 B:R). Data were collected 21 d after transplanting and initiating treatments, and presented as normalized relative counts to each other. Each bar represents the mean of 9 replications. Letters indicate significant differences across supplemental light sources within species using Tukey’s honestly significant difference (HSD; P≤0.05). 76

Fig. 8. Relative content of isorhmanetin, kaempferol, quercetin, myricetin, myristicin, and dillapiole accumulated in dill (Anethum graveolens). Plants were grown in deep-flow technique hydroponic systems under ambient light supplemented with 100 µmol·m–2·s–1 of supplemental light from 0600 to 2200 HR provided from high-pressure sodium (HPS) lamps or light-emitting diodes (LED) with a low blue (B) to red (R) ratio (LB; 7:93 B:R), or high B:R ratio (HB; 30:70 B:R). Data were collected 28 d after transplanting and initiating treatments, and presented as normalized relative counts to each other. Each bar represents the mean of 9 replications. Letters indicate significant differences across supplemental light sources within species using Tukey’s honestly significant difference (HSD; P≤0.05). 77

Fig. 9. Relative content of isorhmanetin, kaempferol, quercetin, apigenin, apiole, and malyopiniin accumulated in parsley (Petroselinum crispum). Plants were grown in deep-flow technique hydroponic systems under ambient light supplemented with 100 µmol·m–2·s–1 of supplemental light from 0600 to 2200 HR provided from high-pressure sodium (HPS) lamps or light-emitting diodes (LED) with a low blue (B) to red (R) ratio (LB; 7:93 B:R), or high B:R ratio (HB; 30:70 B:R). Data were collected 28 d after transplanting and initiating treatments, and presented as normalized relative counts to each other. Each bar represents the mean of 9 replications. Letters indicate significant differences across supplemental light sources within species using Tukey’s honestly significant difference (HSD; P≤0.05).

Fig. 10. Relative content of myrecene, cineole, linalool, limonene, and carvone essential oil accumulation for basil (Ocimum basilicum), dill (Anethum graveolens), and parsley (Petroselinum crispum). Essential oil content was measured by headspace gas chromatography of volatized aromatics, and data is presented as normalized relative counts. Plants were grown in deep-flow technique hydroponic systems under ambient light supplemented with 100 µmol·m– 2·s–1 of supplemental light from 0600 to 2200 HR provided from high-pressure sodium (HPS) lamps or light-emitting diodes (LED) with a low blue (B) to red (R) ratio (LB; 7:93 B:R), or high B:R ratio (HB; 30:70 B:R). Data were collected 21 d (basil) or 28 d (dill and parsley) after transplanting and initiating treatments, and presented as normalized relative counts to each other. Each bar represents the mean of 9 replications. Letters indicate significant differences across supplemental light sources within species using Tukey’s honestly significant difference (HSD; P≤0.05). 79

CHAPTER 4: BLUE LIGHT FRACTION EFFECTS GROWTH, MORPHOLOGY,

BIOMASS PARTITIONING, METABOLISM, AND FLAVONOID ACCUMULATION

OF BASIL

A paper prepared for submission to The Journal of Experimental Botany Alexander G. Litvin and Christopher J. Currey

Abstract

Light optimization in sole-source lighting using light-emitting diodes (LEDs) offers opportunities for targeting photoreceptors of key plant mechanisms, improving plant quality.

Because of the importance of blue (B) light on plant growth and development, and metabolism, our objectives were to quantify the effect of increasing B light fraction on morphological, growth, and physiological metabolic responses, as well as comparing plants under dichromatic light to those grown under white (W) light. Basil (Ocimum basilicum L. ‘Nufar’) were grown hydroponically within seven growth chambers. Three LED light bars provided a photosynthetic photon flux density (PPFD) of 300 μmol∙m–2∙s–1, calibrated to one of seven discrete lighting treatments corresponding to six different B:R photon flux ratios (0:100, 20:80, 40:60, 60:40,

80:20, and 100:0), or from W LED bars. Internode length, plant height, total and individual leaf area (LA), leaf fresh mass (LFM), total fresh mass (TFM), leaf dry mass (LDM), root dry mass

(RDM), and total dry mass (TDM) decreased with increasing B light fraction, although plants grown under 100% B light became hyper-elongated. In contrast, photosynthesis (Pn), stomatal conductance (gs), transpiration (E), and intracellular CO2 (Ci) increased with B light fraction, though Pn increase was curvilinear, decreasing above 60% B light. While chlorophyll fluorescence (Fv/Fm) increased linearly with increasing B light, chlorophyll b decreased. Nutrient mineral accumulation increased quadratically with B light fraction, decreasing in concentration 80 after nutrient-specific maxima. Growth for basil under W light were similar to low fractions of B light for LDM, SDM, RDM, and respective mass ratios, both individual and total LA, Pn and E.

Furthermore, mineral content decreased for many macro- and micronutrients with increasing B light fraction, with W-grown plants similar to low B light fractions for P, Ca, Mg, S, Fe, Zn, and

Cu total content. For all flavonoids analyzed (estragole, myricetin, orientin, vicenin, kaempferol, and quercetin), concentrations in fresh leaf tissue increased linearly with increasing B light fraction, with concentrations of estragole, vicenin, and kaempferol, for basil under W light similar to basil grown under low B light fractions, and orientin, myricetin, and quercetin similar to higher B light fractions. While high proportions of B light can negatively affect growth and morphology, our study highlights the importance of B light proportionality on flavonoid accumulation and gas exchange for optimizing spectra to target photoreceptors and the comparison to basil grown under W light.

Culinary herbs are nutritious (Brown, 1991; Pripdeevech et al., 2010; Simon et al., 1999) and historically medicinal plants (USDA, 2011: Cook and Samman, 1996; van Wyk, 2014).

Health benefits of herbs for dietary nutrition have increased their popularity, with interest in increasing these traits. Aromatic and flavor compounds found naturally in moderate concentrations in plants (Rehman et al., 2016) promote a range of benefits, including: control of blood pressure, cholesterol, and reduced risk of certain cancers (Hollman, 2001; Huang et al.,

2009; Pandey and Rizvi, 2009). These compounds are produced as secondary metabolites, comprise many of the flavors found in culinary herbs, and are dietary nutrients for human health

(USDA, 2011: Cook and Samman, 1996). While culinary herbs are commonly grown as field crops, they are also susceptible to damage from harsh weather (Craufurd and Wheeler, 2009), 81 and are increasingly grown under protected culture (USDA, 1998, 2014) to improve productivity and quality traits within controlled environments (CEs).

Growing culinary herbs in CE) facilities enhances yield and quality of herbs by controlling mineral nutrition (De Pascale et al., 2006), air temperature (Chang et al., 2005), CO2 concentrations (Zobayed and Saxena, 2004), and light intensity and duration (Beaman et al.,

2009; Chang et al., 2008), allowing for controllable and predictable climates year-round, maintaining consistent production and quality regardless of season or location (Brown and

Miller, 2008; Jensen, 1999; Moe et al., 2006; Tixier and de Bon, 2006; Wittwer and Castilla

1995). Lighting significantly affect yield (Beaman et al., 2009; Litvin, 2019), and nutritional value (Li and Kubota, 2009; Litvin, 2019) of herbaceous crops, with some flavonoids specifically promoted through light stress (Dixon and Paiva, 1995; Taulavuori et al., 2016).

Electric lighting is common for CEA production, whether as a supplemental light source in greenhouses or as the sole light source in indoor facilities. However, due to the high energy cost from inefficiencies of conventional lighting (Hogewoning et al. 2007; Trouwborst et al., 2010), light use efficiency is important. Light-emitting diodes (LEDs) illuminate discretely within the range of photosynthetically active radiation (PAR) at selected wavelengths, providing a spectrum that can precisely target plant photoreceptors. Modern high-intensity LEDs provide narrow- spectra lighting encompassing blue (B: 400–500 nm), green (G: 500–600 nm), red (R: 600–700 nm) and far-red (FR: 700–780 nm) wavelengths, with many specific wavelengths for each color on the market (Stutte, 2009), adding new efficiencies to sole-source lighting (Massa et al., 2008).

Specific wavelengths of light each play unique roles in plant responses. Chlorophyll absorbs R light at a peak higher than its absorption of B or G light, yielding a higher relative quantum efficiency (RQE; Sager et al., 1988). Plant perception of light and diurnal cycles is 82 related to R and FR light perception by respective phytochromes (Sharrock, 2008). While R light excites photoreceptors at a lower energy, it has a role in photomorphogenesis (Parks et al., 2001), contributes substantially to overall biomass accumulation (Hernández and Kubota, 2012;

Shacklock et al., 1992), and phytochrome regulation of circadian rhythms and shade-avoidance responses in conjunction with FR light phytochromes (Franklin and Whitelam, 2005; Lorrain et al., 2008; Sharrock, 2008). Additionally, R light increases phenolic concentrations in plants compared to white light, though not to the same magnitude as B light (Li and Kubota, 2009).

White light from LEDs generally include B, G, and R diodes to produce the effect, though G LEDs are often created by applying a phosphor coating to B LEDs (Mickens and

Assefa, 2014). Interest in G light is increasing as recent reports suggest fresh mass accumulation may be promoted for some plants (Mickens et al., 2018; Zhang et al., 2011). In contrast, G light also negatively affects mass of plants, though the effect is dependent on the precise wavelengths

(Wang and Folta, 2013). For example, G light at 540 (Kasajima et al., 2008, 2009) and 563 nm

(Kasajima et al., 2008) delay flowering in Arabidopsis, with G light peaks at approximately 550 nm producing the maximum B light reversal. Antagonism of B light responses in plants by G light is dose-dependent at a 1:2 B:G ratio for G light impediment of B (Frechilla et al., 2000).

Because of this, use of G light may diminish or offset plant responses to B light.

Blue light is involved in several essential plant functions. Notably, B light play a central role in initial stomatal opening in the morning (Baroli et al., 2008; Takemiya, et al., 2013; Wu et al., 2007), suppression of plant height (Hernández and Kubota, 2012), circadian rhythm (Taiz and Zeiger, 2010), and promotion of photoprotective pigmentation (Dixon and Paiva, 1995;

Taulavuori et al., 2016). Additionally, increasing amount of B light is reported to increase the nutritional value of leafy greens (Li and Kubota, 2009). In fact, B light promotes tissue 83 pigmentation and phytonutrients (Kopsell and Sams, 2013) through stress response signaling.

The B light involvement in stress responses promotes biosynthesis of secondary metabolites for photoprotective pigmentation, and photomorphogenic responses for light seeking or avoidance growth (Caldwell and Britz, 2006; Kopsell and Sams, 2013; Taulavuori et al., 2016). These flavonoids are the same compounds of interest in culinary herbs.

Because of the role of B light on morphological development, and promotion of nutritionally vital metabolites, our objectives were to evaluate basil for the effects on morphological growth and physiological metabolic responses to increasing B:R light photon flux ratios, and whether W LEDs provide comparable results. We hypothesize that increasing B light fraction will increase gas exchange and flavonoid accumulation while decreasing plant size and yield. Additionally, although W LED provide a fuller spectrum of PAR light, we hypothesize that any increase in morphological development of basil grown under W light will be met with photosynthetic inefficiencies, and a strong reduction in B-light-mediated responses resulting from G light usage.

Materials and methods

Plant materials and culture

Seeds of basil (Ocimum basilicum L. ‘Nufar’), supplied by a commercial seed company

(Johnny’s Seeds, Winslow, MA) were sown (two seeds per cell) into 162-cell phenolic foam propagation cubes (Oasis Horticubes XL; Smithers-Oasis, Kent, OH) initially saturated with deionized water and allowed to drain, placed in an environmental growth chamber (PGC 10;

Percival Scientific, Perry, IA), and irrigated daily with deionized water supplemented with 100 mg∙L–1 nitrogen (N) supplied from a complete, balanced, water-soluble fertilizer (Jack’s Hydro

Feed 16N–1.8P–14.3K; JR Peters, Allentown, PA). During seed germination and seedling growth, air temperature was a constant 23.8 ± 0.7 °C and a combination of fluorescent and 84 incandescent lights provided 450 ± 12.3 μmol∙m–2∙s–1 from 0600 to 2200 HR throughout propagation. Light was measured with a quantum light sensor (SQ-222; Apogee Instruments

Inc., Logan, UT), and air temperature with a thermocouple (TMC1-HD; Onset, Bourne, MA) in a naturally aspirated solar shield (RS3; Onset), every 15 s in a growth chamber with averages recorded by a data logger (HOBO U12 PPFD/Temp; Onset, Bourne, MA) every 15 min.

Hydroponic culture and controlled environment conditions

Seedlings were thinned to one seedling per cube and transplanted into one of seven deep- flow technique (DFT) hydroponic units 14 d after sowing. The DFT systems were constructed using plastic storage bins (Hefty; Reynold Consumer Products, Lake Forest, IL) measuring 17 cm H × 43 cm W × 91 cm L, filled with 45 L of nutrient solution. Initial nutrient solutions were comprised of deionized water supplemented with 10.5 g of MgSO4, brought to an electrical conductivity (EC) of 1.5 dS∙m–1 with a water-soluble fertilizer (Jack’s Hydro FeED 16N–1.8P–

14.3K; JR Peters), and adjusted to a pH of 6.0 using potassium bicarbonate (JR Peters). Nutrient solutions of each DFT system were measured twice daily with a pH/EC meter (HI 9813-6;

Hanna Instruments, Woonsocket, RI) and pH and EC were adjusted using potassium bicarbonate

(JR Peters) to maintain a target pH of 6.0, while EC was maintained at 1.5 using a concentrated stock solution of water-soluble fertilizer (Hydro FeED; JR Peters). Fifteen basil seedlings were placed into 6-cm black net pots spaced 15 cm apart on center, cut into the lid of the DFT, and placed directly in contact with nutrient solution. The nutrient solutions were continuously aerated using 15-cm air stones (Active Aqua Air Stone; Hydrofarm) in each DFT connected to a 8

L·min–1 air pump (Active Aqua Air Pump 8L; Hydrofarm).

Sole-source lighting treatments

Each DFT system was placed into one of seven individual growth chambers (PGC 10 or

LT 105; Percival Scientific) with one of seven discrete lighting treatments (Table 1). From 0600 to 2200 HR, 300 µmol∙m–2∙s–1 from three red (R – 655 nm peak λ), blue (B – 450 nm peak λ), or a combination of R and B LED bars (RAY 44 ‘Blue’ and ‘Red’; Fluence Bioengineering, Austin,

TX) varying in six different B:R ratios (0:100, 20:80, 40:60, 60:40, 80:20, and 100:0), or from three broad-spectrum white (W) LED bars (RAY 44 ‘Indoor Spectrum’; Fluence Bioengineering;

Fig. 1).

Spectral quality and quantity distribution across growing media was mapped at a 10 cm spacing interval using a spectroradiometer (PS-100; Apogee), and used to adjust individual LED bar intensities until mean PPFD across treatment areas reached 300.0 ± 0.6 μmol∙m–2∙s–1 at approximately 10 cm above the surface of the DFT lids. Air temperature and light intensity were measured every 15 s by a naturally aspirated thermocouple (TMC1-HD; Onset Computer Corp.) enclosed in a radiation shield (RS3; Onset Computer Corp.) and quantum sensor (SQ-222;

Apogee) and averages were recorded every 15 minutes by a data logger (HOBO U12

PPFD/Temp; Onset). Air temperature was 23.7 ± 0.4 °C during the day, and 20.3 ± 0.4 °C at night, with temperature set points of 24 and 20 °C for day and night respectively.

Plant growth data collection and calculation

Photosynthesis (Pn), stomatal conductance (gs), transpiration (E), and intracellular CO2

(Ci) were measured by an infrared gas analyzer (LX-6400XT; LI-COR Biosciences, Lincoln,

2 NE) with a 6-cm clear leaf chamber 20 d after transplanting. Reference CO2 concentration in the gas analyzer chamber was 400 μmol∙mol–1, and water vapor was maintained at 8 mmol. Leaf chamber temperature was aspirated to growth chamber environmental conditions as previously 86 described. Chlorophyll fluorescence was measured on fully expanded leaves three nodes below the apical meristem using a portable flurometer (Handy PEA; Hansatech, Pentney, Norfolk, UK) for dark adapted measurements. Clips with shutters (Hansatech) were placed on the leaves of two separate plants (sub-replicates) within each treatment, dark-adapted for 15 min before securing the flurometer to the clip and opening the shutter, and the average was recorded within each treatment. Flurometer measurements were done by exciting tissue for 1 s with increasing light up to 3512 μmol∙m–2∙s–1 to saturate photosystem II (PSII) and data were expressed as the change in fluorescence to max fluorescence (Fv/Fm). To measure the overall effect of time (day length) across treatments, chlorophyll fluorescence was measured at 14 d, and both gas exchange and chlorophyll fluorescence were measured at 20 d, in the morning (between 0600 and 0800 HR), midday (between 1200 and 1400 HR), and end of the day (between 2000 and 2200 HR), and combined for analysis.

Final growth and destructive measurements were preformed 21 d after transplanting from of five samples (individual plants) from each treatment. Heights were recorded from the surface of the DFT system to the apical meristem, and internode lengths were measured from the first basal node to the second, along with the number of nodes formed. Roots were severed at the surface of the substrate cube and leaves were separated from remaining shoot tissue, and stem fresh mass (SFM) and leaf fresh mass (LFM) were then weighed and recorded. Leaves attached directly to the stem were scanned (X9-310; Epson, Suwa, Nagano, Japan), with leaf number and leaf area subsequently measured by image analysis (ImageJ; U.S. National Institute of Health,

Bethesda, MD). Leaves were then tripled rinsed in deionized water before leaves, stems, and roots were placed separately in paper bags and dried in a forced air oven at 67 °C for 3 d and weighed for dry mass. After recording, LDM replicates were submitted to a commercial 87 laboratory (AgSource Harris Laboratories, Lincoln, NE) for analysis of nutrient concentrations

(N, P, K, Ca, Mg, S, Zn, Mn, Fe, B, and Cu).

Chlorophyll content and phenolic concentration analysis

Three additional plants harvested from each treatment were used for chlorophyll content and phenolic concentration analysis. Preparation of samples for chlorophyll analysis was done by cutting ≈150 mg of leaf tissue into 1.5 mL centrifuge tubes and flash freezing in liquid nitrogen.

Samples were then ground into a powder within their respective centrifuge tubes and aliquots of

1.5 mL of reagent grade ethanol were added before storing at 4 °C until further analysis. Samples were spun down for 2 min at 5000 g in a centrifuge (5415 C; Eppendorf, Hamburg, Germany) before transferring 750 µL of supernatant into a new tube and adding an additional 750 µL of ethanol. Chlorophyll content was quantified by spectrophotometer (Epoch 2; BioTeck,

Winooski, VT) at 665 and 649 nm for chlorophyll a and b, as described by Ritchie (2006).

Samples for analyses of relative flavonoid concentrations were collected combing leaves two nodes below the growing tip of apical meristem from three sub-replicates in each treatment.

Samples were prepared by extracting analytes from plant material into a methanol solvent solution as described by Khoddami et al. (2013). In order to reduce particle size for better extraction, 1–2 g of plant material was cut from mature leaves, weighed and inserted into a 15 mL centrifuge tube with a milling ball (10 mm PM 100 planetary mill grinding ball; Retsch

Technology GmbH, Haan, Germany) and flash frozen in liquid nitrogen. Frozen samples were then shaken and vortexed until tissue was ground into a fine powder. Aliquots of 6 mL of methanol solvent were added to each centrifuge tube, and vortexed at room temperature prior to filtration through filter paper (Whatman No. 2 filter paper; GE Healthcare UK Limited,

Amersham Place Little Chalfont, UK), and transferred into a 2 mL auto sampler vial. Samples 88 were analyzed by quadruple and time-of-flight high-performance liquid chromatography mass spectroscopy (6450 QTOF HPLC; Agilent, Santa Clara, CA) with an LC column (XDB C18, 4.6

× 150 mm, 1.8 μ; Agilent) at a flow rate of 700 μL∙min-–1 and temperature maintained at 30 °C.

Samples were injected at a volume of 5 μL with a solvent gradient of 95% solvent A (1:1

H20/MeOH with 1% Formic acid) to 5% solvent B (ACN with 0.1% Formic acid) held for 5 min and then changing to 5% solvent A to 95% solvent B over 10 min and held for 3 min. Analytes were read at a UV wavelength of 280 nm.

Data calculated

Growth data calculated included total fresh mass (TFM; TFM = SFM + LFM) and dry mass (TDM; TDM = SDM + LDM + RDM). Biomass allocation for leaves, stems, and roots were then calculated for dry mass, including leaf mass ratio (LMR; LMR = LDM/TDM), stem mass ratios (SMR; SMR = SDM/TDM), and root mass ratios (RMR; RMR = RDM/TDM) derived from LFM, leaf dry mass (LDM), shoot dry mass (SDM), and root dry mass (RDM).

Total leaf area (total LA) was used to calculate individual LA (individual LA = total LA / no. of leaves). Additional calculations of biomass ratios were calculated for leaf area ratio (LAR; LAR

= Total leaf area/TDM) and specific leaf area (SLA; SLA = LA). Relative flavonoid content and nutrient content was calculated, in accordance with the tissue type used for initial analyses, from flavonoid concentration of each compound (relative flavonoid content = relative flavonoid concentration × LFM), and nutrient concentration of each element (nutrient content = nutrient concentration × LDM). The ratio of Pfr to total phytochrome content (Ptotal; Pfr/Ptotal), known as the photostationary state (PSS; Sagar et al., 1988), was calculated as an indicator of expected photomorphogenesis due to phytochrome activity, and was derived from spectral data (Sagar et al., 1988). 89

Experimental design and statistical analyses

The experiment was designed as a randomized complete block design without replication.

There was one replication (individual DFT systems) for each light treatment per block

(experimental run), and the experiment was repeated four additional times for a total of five blocks. Lighting treatments were re-randomized across growth chambers and recalibrated between each experimental run. Regression analyses (SigmaPlot 11.0; Systat Software, San Jose,

CA) were performed for all data across photon flux ratios of B:R (% B light), and Holm-Sidak post-hoc test was performed comparing all data from plants grown under B light treatments to those grown under W light. Interactions among response variables were used to create predictive models for yield and phenolic concentrations based on photosynthetic conditions at a given % B light level.

Results

Morphological effects

Basil plants grown under 80B light were 5.1 cm shorter than plants grown under 0B, with height inversely proportional to B light content up to 80B at 21 d (Fig. 1). Plants grown under

0% to 80% B light were similar in height to W-grown basil (17.6 cm); however, plants grown under 100B light were 28.2 cm tall, 11.3 cm taller than to the mean height of all other plants.

Internode elongation was reduced as B light increased up to 80% B light (Fig. 2). Internode length of plants grown under 20% or 40% B light were comparable to those grown under W light, while basil under 0B and 100B were longer, and 60B and 80B shorter. Node number was similar across all treatments (Fig. 2).

Although LFM decreased linearly with increasing B light (Fig. 2), TFM was generally similar across treatments. Plants grown under 0B or 20B had the highest accumulated LFM at

24.4 and 24.6 g respectively, decreasing linearly up to 100% B light (15.5 g), the point at which 90

LFM was 33% less than W-grown basil. Similar to stem elongation, SFM decreased as B light increased from 0 to 80%, but then increased under 100B (22.1 g).

The TDM decreased from 3.5 to 2.7 g as B light increased from 0 to 100% B light (Fig.

2), and plants under ≥60% B light had less TDM than W-grown plants. Similarly, basil LDM decreased by 37% as B light increased from 0% to 100% (Fig. 3); plants grown under ≤20% B light similar in LDM compared to W-light. While there were no trends observed for SDM as B light increased, basil grown under 80B light had 24% less than W-grown basil. Alternatively,

RDM decreased with increasing B light, with plants grown under 40% to 100% B light 17% to

39% less accumulated RDM than under W light.

The LMR decreased from 56% to 44% as B light increased to 100%, with 100B-grown basil smaller than W-grown. In contrast, SMR increased from 35% to 48% curvilinearly with increasing B light, with 100B-grown basil proportionally 34% more than W-grown (Fig. 3). The curvilinear decrease in RMR as B light increased beyond 20% (Fig. 3) was 8% at 100B, and less than plants grown under W light. Individual LA decreased linearly by 38% and total LA by 29% as B light increased from 0% to 100%. Plants grown under ≥60% B light had smaller individual

LA than W-grown, while plants grown under ≥80% had smaller total LA. While no trends were observed for SLA or LAR in response to increasing B light, the LAR of 100B-grown plants were

167.3 cm2·g–1, 12% smaller than W-grown basil.

Gas exchange

The Pn increased with increasing B light to a maximum at 60% B light, decreasing thereafter (Fig. 5). Plants grown under 20% to 80% B light had greater Pn across the course of the day than W-grown plants. Both gs and E increased linearly by 40% and 22%, respectively, as

B light increased from 0 to 100% (Fig. 5), While there were no differences between plants grown 91

under B light compared to W light for gs, plants grown under ≥60% B light had higher E compared to plants grown under W light. The Ci also increased linearly by 7% increasing B light up to 100% B light (Fig. 5), with plants grown under W light comparable to those grown under all proportions of B light.

Overall Fv/Fm from morning to evening increased linearly by 13% as B light increased

(Fig. 6). The Fv/Fm for plants grown under ≤20% B light were between 2% and 7% lower than

W-grown plants, whereas plants grown under ≥60% B light were up to 5% higher. Chlorophyll a was unaffected by B light, whereas chlorophyll b decreased by 37% as B light increased from

0% to 100%. Total chlorophyll responded similarly, as basil grown under 100B had 27% less chlorophyll a+b than under 0B. Chlorophyll a, b, or combined (a+b) of W-grown plants were similar in quantity compared to all B light treatments.

Nutrient analysis and flavonoids

The concentration of macronutrients in foliar tissue increased quadratically with increasing B light up to a maximum, before decreasing with additional B light (Fig. 8).

Concentrations of N were highest in 80B-grown plants with 5.5% N, with 17% and 9% more N than 0B- and W-grown basil. Concentrations of P and Mg were highest in 60B-grown plants at

1.3% P and 0.6% Mg, though similar to W-grown plants. Plants grown with ≤60% B light were highest in Ca, with concentrations decreasing thereafter to 10% at 100B; 25% lower in concentration than W-grown plants. Concentrations of micronutrients varied across treatments, exhibiting quadratic relationships to B light for Zn and Cu, or a linear decrease (Fe) or increase

(elemental B) to increasing B light (Fig. 9). Concentrations of Zn were highest in basil grown between 20% (99.0 mg∙L–1) and 60% (104.2 mg∙L–1) B light, decreasing thereafter; 100B-grown plants had 27% less than W-grown basil. Concentrations of Cu in 0B-grown plants (10.6 mg∙L–1) 92 were 27% lower than W-grown basil, and exhibited a curvilinear increase up to 16.2 mg∙L–1 as B light increased up to 40%, and decreased thereafter. Nutrient content was either unaffected by B light or decreased linearly with increasing B light fraction. Nutrient content decreased linearly by

19, 38, 32, 49, and 34% for N, P, K, Ca, and S respectively as B light increased to 100%.

Nutrient content in Mg decreased from 90 to 53 mg, and was curvilinear with respect to increasing B light. The Fe content decreased linearly from 21.7 to 13.1 ng. Both Zn and Mn decreased curvilinearly with increasing B light by as much as 45% and 37%, respectively. Plants grown under 80B were lower in P and B than W light, and plants grown under ≥80% B light were lower also lower in N, K, Ca, S, Mg, Zn, Mn, and Cu. The Al content of plants grown under B light treatments were all similar to W-grown.

The relative concentration of all flavonoids analyzed in this study increased linearly with increasing B light (Figs. 10 and 11). For instance, the estragole concentration increased by 43% as B light increased up to 100%, with plants under 100B 44% higher in concentration than W- grown (Fig. 10). Concentrations of vicenin increased by 65% as B light increased, with plants grown under ≥60% B light having between 42 to 49% higher concentrations than W-grown. In contrast, while orientin and myricetin increased by 67 and 59% from 0 to 100% B light respectively, plants grown under ≤20% B light were up to 32 and 31% lower in concentration respectively, than plants grown under W light. Kaempferol in basil increased by 34% as B light increased, with plants grown under ≥80% B light as much as 39% higher in concentration than

W-grown. Although quercetin increased by 32% with increasing B light, plants grown under different B light fractions were all similar to W-grown concentrations. Despite B light effects on flavonoid concentrations, no trends or differences were seen for the interaction of flavonoid 93 concentration and LFM, suggesting an overall similar flavonoid content regardless of light source.

Discussion

Increasing B light fraction inversely affects plant size and chlorophyll content, yet it also increases PSII efficiency (Fv/Fm), gas exchange, and nutrient and flavonoid concentrations.

Different light spectra impart specific benefits for plant growth and morphology. The quantitative amount of a light can be cumulative, amplifying responses with increasing content.

In this study the extreme ends of B light fraction are associated with either the maximum or minimum responses, while W light or moderate proportions of B light appear to balance the highs and lows of extremes proportions.

Basil height was increasingly suppressed under increasing B light fraction, up to 80% B light and this effect was predominately due to B light’s effect on internode elongation (Fig. 2).

Fraszczak et al. (2014) reported that basil grown under LEDs (16.3% B light) were 9% shorter than basil under fluorescent light (7.5% B light) after 28 d. Wheeler et al. (1991) described height suppression in soybeans (Glycine max) when HPS light was supplemented with additional

B light. The effect of B light on suppressing basil in this study is quantitative up to 80%. Similar height suppression with increasing B light fraction is also reported for cucumber (Cucumis sativus; Hernández and Kubota, 2016; Snowden et al., 2016), tomato (Solanum lycopersicum;

Snowden et al., 2016), radish (Raphanus sativus; Cope et al., 2014), and petunias (Petunia × hybrida; Currey and Lopez, 2013). While increasing B light reduce stem elongation, the use of

100% B light, resulted in hyper-elongation and highlights the multiple mechanisms involved for plants for growth. Hernández and Kubota (2016) reported that although cucumber stem length was shorter with increasing B light fraction, at 100% B light cucumber plants were hyper- elongated, and is purportedly due to R and FR phytochrome activity, rather than B light content 94

(Sager et al., 1988). Phytochromes Pr and Pfr, associated with shade avoidance, control stem elongation (Schmitt and Wulff, 1993), with PSS used to estimate the photomorphogenic effect of a light source (Sager et al., 1988). The phytochrome ratio for PSS of W light, which contained both R and FR light, was 0.86, and was similar to most lighting treatments except 100B (0.48;

Table 1). While R light converts Pr into Pfr, Pfr is reverted by either FR light or the absence of R light, such as in darkness (Taiz and Zeiger, 2010), or when illuminating with 100% B light as in this study. Because of the low PSS corresponding to plants under 100B, we postulate the spectra may induce a phytochrome shade-avoidance response, thereby promoting stem elongation. Plants grown under 20B and 40B had similar internode lengths as W-grown, though total heights were similar for all treatments, with exception to 100B. Taken together, while B light affects plant height, use of additional spectra are needed to maintain and balanced plant growth and function.

While B light fraction interactions with other spectra can categorically affect height, B light effects on biomass accumulation is continuous and decrease linearly from 0% to 100% B light. Literature shows increasing B light reduces plant size (Randall and Lopez, 2015) and biomass (Mickens et al., 2018). In fact, Dougher and Bugbee (2001) reported reductions in

LDM, SDM, RDM, and TDM of wheat (Triticum aestivum) and soybean at increasingly higher

B light fraction, and Snowden et al. (2016) similarly reported decreasing dry mass for tomato and cucumber plants. In this study, both fresh (LFM and TFM) and dry biomass (TDM, LDM, and RDM) decreased with increasing B light, largely agreeing with findings from previous reports. Furthermore, decreasing LMR as B light fraction increased, as seen in this study for basil, appears to be supported by reports of radish LMR also decreasing with B light content

(Cope and Bugbee, 2013). Yet reports for LMR of wheat and soybean (Cope and Bugbee, 2013;

Dougher and Bugbee, 2001) describe increases in LMR with increasing B light fraction, and 95

SMR flat or decreasing, though these studies only evaluated growth under ≤30% B light. The negative impact of B light in LMR and SMR up to 40% B light was negligible in our study.

Greatest changes in LMR, SMR, and RMR occurred as B light increased from 80% to 100%.

Although both LDM and RDM decreased with increasing B light fraction, SDM was unaffected by treatment, and the decreasing LMR and RMR are further supported by the proportionally more growth allocated to stems, resulting in an increasing SMR despite similar SDM. This is further amplified in this study by B light fractions above 80% B light, where hyper-elongation of plants under 100% B light, resulting in the higher SMR, and lower LMR and RMR than W- grown basil, were observed.

Total LA, like mass, decreases linearly as B light fraction increases. Photomorphogenic for reduced leaf expansion under increasing B light are reported for radish, soybean, and for cucumber (Cope and Bugbee, 2013; Snowden et al., 2016). Here, both individual and total LA decreased as B light fraction increased to 100%. Further analysis into the effect of B light on leaf morphology is done by evaluating the ratios of LA and mass. Samuolienė et al. (2010) reported strawberry (Fragaria × ananassa Duch.) had 19% higher SLA when B light (455 nm) was added to R (640 nm) LED light source, though LAR decreased slightly. Because both LA and mass decreased concurrently with increasing B light, ratios of LAR and SLA remained largely unaffected across B light fraction, and were similar to W-grown basil. This may be due to their respective factors of mass and LA decreasing in congruence with one another as B light increased, though minimal changes in SLA and LAR in response to increased B light were reported for hickory wattle (Acacia implexa; Forster and Bonser, 2009) and rice (Oryza sativa L.;

Ohashi-Kaneko et al., 2006), in agreement with this study’s findings. Although basil grown under increasingly higher proportions of B light decreased in LA, gas exchange increased. 96

Apart from light intensity, gas exchange is intrinsically affected by light quality. While

Pn occurs across PAR (McCree, 1972; Inada, 1976; Bugbee, 2016), the RQE of light varies across the spectrum (Sager et al., 1988). Initial gas exchange is influenced by B light photoreceptors in the morning, such as cytochromes and phototropins, which aid gs by playing a role in promoting guard cell opening and initiating gas exchange (Briggs and Huala, 1999;

Christie, 2007; Humble and Hsiao, 1970). Gas exchange is thus dependent on light quality, and increasingly higher B light fractions result in cumulatively higher gs, E, and Ci. However, G light can be antagonistic to B light effects on gas exchange (Frechilla et al., 2000; Folta and

Maruhnich, 2007; Wang and Folta, 2013). Although W light had approximately the same B light content as the 20B treatment, it also had approximately 39% G light. Agreeing with previous reports (Frechilla et al., 2000), Pn of basil under W light was less than under 20B, and more similar to 0B. Additionally, B light alone is insufficient for promoting Pn, as seen in this study.

The curvilinear response of increasing Pn with increasing B light up to a maximum at 60B suggests either a possible detrimental effect on Pn in response to B light beyond this proportion, or a fundamental requirement for more than one spectra to further promoter Pn. Snowdent et al.

(2016) notes Pn of cucumber increases with increasing B light and, similarly, Hogewoning et al.

(2010) reports gs also increases with B light; these reports are similar to our results. Interestingly,

Ci increases with B light. Assmann (1988) reported pulses of B light not only increases gs, but also Ci, while R light pulses reversed Ci accumulation, in full agreeance with the increase seen from 0B (100 % R) to 100B (0% R). Basil grown at 20% to 80% B light, and 60% to 100% B light had higher Pn and E, respectively, than under W light, underlining potential disadvantages of W light for gas exchange. Higher Pn is dependent on the efficiency of plants for 97 photochemical processing of incidental light, and thus is associated with the efficiency of PSII

(Genty et al., 1989).

Photosynthesis is intrinsically associated with Fv/Fm, with Fv/Fm affecting the function of

Pn (Genty et al., 1989). Likewise, higher Fv/Fm affects chlorophyll content as a function of the total Pn capacity; higher Fv/Fm suggests less chlorophyll is needed to maintain Pn (Genty et al.,

1989). Chlorophyll a content of B light-treated basil were all similar to those under W light, and insensitive to B light fraction. However, chlorophyll b and total chlorophyll (a+b) was lower in basil grown in successively higher B light fractions. Despite these trends, chlorophyll content of plants grown under 0% to 100% B light were similar to those under W light. Chlorophyll a content is less responsive to changes in spectra, yet when Fv/Fm decreases, additional chlorophyll b may be synthesized (Dai et al., 2009), reducing over-excitation of chlorophyll a. We postulate additional chlorophyll b was not accumulated due to the higher Fv/Fm measured as B light increased and higher sustained gas exchange. In all, the efficiency of these light reactions for gas exchange facilitate the hydraulic conductivity necessary for nutrient uptake (Chapin, 1991), with increased gas exchange potentially increasing uptake.

Nutrient concentrations of most macronutrients increased in a curvilinear fashion, following the general trend seen with Pn. If gas exchange increases nutrient uptake (Chapin,

1991), then the concept of B light acting as a plant growth regulator (Folta and Childers, 2008) can apply for controlling nutrient concentration. Kopsell and Sams (2013) reported that use of B

(470 nm) light increased concentrations of all macro- and micronutrients in broccoli microgreens

(Brassica oleacea var. italica). Conversely, Gerovac et al. (2016) reported that a light source with 74% R, 18% G, and 8% B light increased N, P, K, S, Ca, and Mg in Brassica microgreens compared to a 87:13 R:B LED light source. Some discrepancies may be due to the exact 98 wavelengths used (Cope et al., 2014; Folta and Maruhnich, 2007) or due to a minimum PPFD of

B light needed to elicit responses (Bugbee, 2016; Wheeler et al., 1991). While micronutrients were expected to follow similar trends as macronutrients (Kopsell and Sams, 2013), curvilinear responses were seen only in Zn and Cu (Fig. 9). While elemental B increases with B light fraction, Fe concentrations decrease. However, when the mass of the plant is taken into consideration, nutrient contents generally decreases with increasing B light for most macro- and micronutrients (Figs. 9 and 10).

Increasing phenolic concentrations with increasing B light fraction agree with reports on

B light-mediated promotion of phenolics (Dixon and Paiva, 1995). However, this study is the first to characterize the effect of B light on growth, morphology, and mineral uptake of basil in conjunction with gas exchange, and accumulation of selected flavonoids. The phenolic synthesis promoted by biotic and abiotic stresses vary across different flavonoids and are specific to the type of stress (Dixon and Paiva, 1995; Ma et al., 2014). Flavonoids such as kaempferol, quercetin, and other phenolics benefit human health (Hollman, 2001; Huang et al., 2009; Pandey

& Rizvi, 2009), and their promotion in plants via the phenylpropanoid pathway (Petrussa et al.,

2013; Stahlhut et al., 2015). Because of the linear increase in flavonoid concentrations as B light increases, we can classify the accumulation of estragole, myricetin, orientin, vicenin, kaempferol, and quercetin in basil as a light-mediated stress responses promoted by B light.

Although all flavonoids analyzed in this study increased under B light, the concentrations of specific flavonoids of B light-treated basil varied with respect to W-grown basil. For instance, basil grown under 100B had higher estragole and vicenin, compared to W-grown, with 60B- and

80B-grown plants also higher in vicenin, and 80B higher in kaempferol. Conversely, plants grown under ≤20B had less orientin and myricetin than W-grown basil. Secondary metabolites 99 responsible for flavor qualities of plants are generally part of stress mitigation mechanisms, protecting plants from abiotic and biotic stress conditions (Petrussa et al., 2013; Winkel-Shirley,

2002). Despite the increases in phenolic concentrations, when the concentrations were extrapolated to leaf content using LFM, the flavonoid content was similar for plants irrespective of treatments.

Although basil grown under increasingly higher B light fractions may reduce mass, moderate use may have benefits for efficient use of spectra with respect to plant size, gas exchange, nutrient uptake, and flavonoid concentrations. Use of B light improves Fv/Fm and gas exchange through photoreceptors of phototropins and cryptochromes involved in stomatal conductance and stress signaling. Because small changes in wavelengths can affect the response and magnitude of lighting effects, there is increasing focus on specific peak wavelength output and absorbance, and it has become necessary to define more precisely these spectra within photobiology and horticulture. While B, G, and R and generally described with ranges spanning

100 nm each, specific wavelengths within each color have varying magnitudes and effects on plant responses (Folta and Maruhnich, 2007). For example, cryptochrome and phototropin action spectrum has a high absorption between 430–470 nm (Briggs and Christie, 2002; Christie et al.,

2014; Frechilla et al., 2000), and are related to gas exchange (Christie, 2007), stem elongation

(Ahmad et al., 2002), and mass. Additionally, Kim et al. (2004) reported up to 24% G light can increase growth. However, G light antagonism of B light can reduce the overall effect of photon flux on growth and metabolism. Dougher and Bugbee (2001) reported light at 580–600 nm reduced the dry mass of lettuce (Lactuca sativa), while Mickens et al. (2018) reported greater fresh mass if the G light was at 520 nm, than 582 nm for lettuce. Additionally, Antagonism of B light responses on stomatal opening by G light is dose-dependent at a 1:2 B:G ratio for G light 100

(peak 540 nm) reversal of B (Frechilla et al., 2000; Talbot et al., 2002). Within this study, W light was 21% B (62 μmol∙m–2∙s–1) with a B:G ratio of approximately 1:2 (both broadly defined and when peak wavelengths ± 20 nm were compared). Furthermore, while the interaction of different wavelengths, as with G light added to B, reducing B light responses (Frechilla et al.,

2000; Kim et al., 2004), other spectra such as R and FR light, may work in tandem to enhance growth (Emerson and Rabinowitch, 1960). For instance, the addition of FR light to a R light source can increase Pn greater than the sum of the two wavelengths’ effect on photosynthesis separately (Emerson and Rabinowitch, 1960). While the use of W light in this study provided plants with need FR to aid in growth, the quantity and quality of the G light spectrum may detrimentally affect basil growth more than its contribution.

Further study of W light use within the context of narrowed spectra definitions is needed.

Furthermore, although G light may act in antagonism of B, it may help improve deficiencies caused by higher B light fractions. Basil grown under two spectra (B and R) were not necessarily disadvantaged in growth and development compared to plants grown under the fuller action spectrum provided from W light. Instead, W-grown basil were similar to results for plants grown somewhere between 0% and 40% B light. The growth and metabolism of basil across these lighting spectra highlight the importance of balancing light quality with production targets; maximizing flavonoid concentration and metabolism while maintaining desirable morphology through careful regulation of light quality.

Conclusion

Use of B light can beneficially increase Fv/Fm of plants, increasing efficiency of photochemical processes. Although Fv/Fm, gs, and E increase linearly with increasing B light up to 100%, maximum Pn is achieved when there is a balance between B and R light. Use of multispectral lighting may help increase growth by acting on different photoreceptors. While B 101 light directly affects stomatal conductance, increasing gas exchange, R light is needed to provide a PSS that minimizes phytochrome-mediated elongation. Furthermore, while use of FR light may be beneficial to plant growth, G light is deleterious to B light contribution to plant functions, but the magnitude of the effect is also dependent on the specific wavelengths of G light. Taken together, B, R, and FR light are efficient contributors to plant health and function, but broad application of G light may negate some of these affects. However, the B light antagonistic qualities of specific G light wavelengths requires attention. Because of this, further research into the use of W light with focus on less antagonistic G wavelengths is merited.

References

Ahmad, M., N. Grancher, M. Heil, R.C. Black, B. Giovani, P. Galland, and D. Lardemer. 2002. Action spectrum for cryptochrome-dependent hypocotyl growth inhibition in Arabidopsis. Plant Physiol. 129:774–785.

Assmann, S.M. 1988. Enhancement of the stomatal response to blue light by red light, reduced intercellular concentrations of CO2, and low vapor pressure differences. Plant Physiol. 87:226– 231.

Baroli, I., G.D. Price, M.R. Badger, and S. von Caemmerer. 2008. The contribution of photosynthesis to the red light response of stomatal conductance. Plant Physiol. 146:737–747.

Beaman, A.R., R.J. Gladon, and J.A. Schrader. 2009. Sweet basil requires an irradiance of 500 µmol∙m−2∙s−1 for greatest edible biomass production. HortScience 44:64−67.

Briggs, W.R. and J.M. Chistie. 2002. Phototropins 1 and 2: versatile plant blue-light receptors. TRENDS Plant Sci. 7:204–210.

Briggs, W.R. and E. Huala. 1999. Blue-light photoreceptors in higher plants. Ann. Rev. Cell Dev. Biol. 15:33–62.

Brown, C. and S. Miller. 2008. The impact of local markets: a review of research on farmer markets and community supported agriculture (CSA). Am. J. Agr. Econ. 90:1298–1302.

Brown, S. 1991. Culinary herb use in southern California restaurants. CA Agric. 45:4−6. 102

Bugbee, B. 2016. Toward an optimal spectral quality for plant growth and development: the importance of radiation capture. Acta Hort. 1134:1–12.

Chang, X., P.G. Alderson, and C.J. Wright. 2005. Effect of temperature integration on the growth and volatile oil content of basil (Ocimum basilicum L.). J. Hort. Sci. Biotechnol. 80:583−598.

Chang, X., P.G. Alderson, and C.J. Wright. 2008. Solar irradiance level alters the growth of basil (Ocimum basilicum L.) and its content of volatile oils. Environ. Expt. Bot. 63:216–223.

Chapin III, F.S. 1991. Integrated responses of plants to stress. BioScience. 41:29–36.

Christie, J.M. 2007. Phototropin blue-light receptors. Ann. Rev. Plant Bio. 58:21–45.

Christie, J.M., L. Blackwood, J. Petersen, and S. Sullivan. 2014. Plant flavoprotein photoreceptors. Plant Cell Physiol. 46:401–413.

Cook, N.C. and S. Samman.1996. Flavonoids- chemistry, metabolism, cardioprotective effects, and dietary sources. Nutr. BioChem. 7:66–76.

Cope, K.R. and B. Bugbee. 2013. Spectral effects of three types of white light-emitting diodes on plant growth and development: absolute versus relative amounts of blue light. HortScience. 48:504–509.

Craufurd, P.Q. and T.R. Wheeler. 2009. Climate change and the flowering time of annual crops. J. Expt. Bot. 60:2529–2539.

De Pascale, S., A. Maggio, F. Orsini, and G. Barbieri. 2006. Nutrients influence on ready to eat sweet basil quality. Acta Hort. 718:523−530.

Dixon, R.A. and N.L. Paiva. 1995. Stress-induced phenylpropanoid metabolism. Amer. Soc. Plant Phsiol. 7:1085–1097. 103

Dougher, T.A. and B.G. Bugbee. 1998. Is blue light good or bad for plants? Life Support. Biosphere Sci. 5:129–136.

Dougher, T.A.O. and B. Bugbee. 2001. Differences in the response of wheat, soybean and lettuce to reduced blue radiation. Photochemistry Photobiology. 73:199–207.

Emerson, R. and E. Rabinowitch. 1960. Red drop and role of auxiliary pigments in photosynthesis. Plant Physiol. 35:447–485.

Folta, K.M. and K.S. Childers. 2008. Light as a growth regulator: controlling plant biology with narrow-bandwidth solid-state lighting systems. HortScience 43:1957–1964.

Folta, K.M. and S.A. Maruhnich. 2007. Green light: a signal to slow down or stop. J. Expt. Bot. 58:3099–3111.

Forster, M.A. and S.P. Bonser. 2009. Heteroblastic development and shade-avoidance in response to blue and red light signals in Acacia implexa. Photochemistry Photobiology 85:1375– 1383.

Franklin, K.A. and G.C. Whitelam. 2005. Phytochromes and shade-avoidance response in plants. Ann. Bot. 96:169–175.

Frąszczak, B., A. Golcz, R. Zawirska-Wojtasiak, and B. Janowska. 2014. Growth rate of sweet basil and lemon balm plants grown under fluorescent lamps and LED modules. Acta Sci. Pol. Hortorum Cultus 13:3–13.

Frechilla, S., L.D. Talbott, R.A. Bogomolni, and E. Zeiger. 2000. Reversal of blue light- stimulated stomatal opening by green light. Plant Cell Physiol. 41:171–176.

Gerovac, J.R., J.K. Craver, J.K. Boldt, and R.G. Lopez. 2016. Light intensity and quality from sole-source light-emitting diodes impact growth, morphology, and nutrient content of Brassica microgreens. HortScience. 51:497–503.

Hernández, R. and C. Kubota. 2012. Tomato seedling growth and morphological responses to supplemental led lighting red:blue ratios under varied daily solar light integrals. Acta Hort. 956:187–194.

Hernández, R. and C. Kubota. 2016. Physiological responses of cucumber seedlings under different blue and red photon flux ratios using LEDs. Environ. Expt. Bot. 121:66–74.

Hollman, P.C.H. 2001. Evidence for health benefits of plant phenols: local or systemic effects? J. Sci. Food Agric. 81:842–852.

Huang, W.Y., Y.Z. Cai, and Y. Zhang. 2009. Natural phenolic compounds from medicinal herbs and dietary plant: potential use for cancer prevention. Nutr. Cancer. 62:1–20.

Humble, G.D. and T.C. Hsiao. 1970. Light-dependent influx and efflux of potassium of guard cells during stomatal opening and closing. Plant Physio. 46:483–487.

Inada, K. 1976. Action spectra for photosynthesis in higher plants. Plant Cell Physiol. 17:355– 365.

Jensen, M.H. 1999. Hydroponics worldwide. Acta Hort. 481:719–730.

Kasajima, S.Y., N. Inoue, and R. Mahmud. 2009. Response spectrum for green light-induced acceleration of head in wheat cv. Norin 61. Plant Production Sci. 12:54–57.

Kasajima, S.Y., N. Inoue, R. Mahmud, and M. Kato. 2008. Developmental responses of wheat cv. Norin 61 to fluence rate of green light. Plant Production Sci. 11:76–81.

Khoddami, A., M.A. Wilkes, and T.H. Roberts. 2013. Techniques for analysis of plant phenolic compounds. Mol. 18:2328–2375.

Kim, H.H., G.D. Goins, R.M. Wheeler, and J.C. Sager. 2004. Green-light supplementation for enhanced lettuce growth under red- and blue-light-emitting diodes. HortScience 39:1617–1622.

Li, Q. and C. Kubota. 2009. Effects of supplemental light quality on growth and phytochemicals of baby leaf lettuce. Environ. Expt. Bot. 67:59–64.

Litvin, A.G. 2019. Light intensity and quality affect growth, morphology, gas exchange, and secondary metabolite accumulation in culinary herbs. Ph.D. Diss., Iowa State Univ., Ames.

Massa, G.D., H. Kim, R.M. Wheeler, and C.A. Mitchell. 2008. Plant productivity in response to LED lighting. HortScience 43:1951–1956.

McCree, K.J. 1972. Test of current definitions of photosynthetically active radiation against leaf photosynthesis data. Agr. Meteorol. 10:443–453.

Mickens, M.A. and Z. Assefa. 2014. Tunable luminescence and white light emission of novel multiphase sodium calcium silicate nanophosphors doped with Ce3+, Tb3+, and Mn2+ ions. J. Luminescence 145:498–506.

Moe, R., S.O. Grimstad, and H.R. Gislerod. 2006. The use of artificial light in year round production of greenhouse crops in Norway. Acta Hort. 711:35–42.

Morrow, R. 2008. LED lighting in horticulture. HortScience. 43:1947–1950.

Ohashi-Kaneko, K., R. Matsuda, E. Goto, K. Fujiwara, and K. Kurata. 2006. Growth of rice plants under red light with or without supplemental blue light. Soil Sci. Plant Nutr. 52:444–452.

Pandey, K.B. and S.I. Rizvi. 2009. Plant polyphenols as dietary antioxidants in human health and disease. Oxidative Medicine Cellular Longevity 2:270–278.

Parks, B.M., K.M. Folta, and E.P. Spalding. 2001. Photocontrol of stem growth. Curr. Opin. Plant Biol. 4:436–440.

Ritchie, R.J. 2006. Consistent sets of spectrophotometric chlorophyll equations for acetone, methanol, and ethanol solvents. Photo. Res. 89:24–41.

Sager, J.C., W.O. Smith, J.L. Edwards, and K.L. Cyr. 1988. Photosynthetic efficiency and phytochrome photoequilibria determination using spectral data. Amer. Soc. Agr. Eng. 31:1882– 1889. 106

Samuolienė, G., A. Brazaitytė, A. Urbonavičiūtė, G. Šabajevienė, and P. Duchovskis. 2010. The effect of red and blue light component on the growth and development of frigo strawberries. Zemdirbyste-Ag. 97:99–104.

Schmitt, J. and R.D. Wulff. 1993. Light spectral quality, phytochrome and plant competition. Trends in Ecol. Evo. 8:47–51.

Shacklock, P.S., N.D. Read, and A.J. Trewavas. 1992. Cytosolic free calcium mediates red light- induced photomorphogensis. Nature. 358:753–755.

Sharrock, R.A. 2008. The phytochrome red/far-red photoreceptor superfamily. Genome Biol. 9:230

Snowden, M.C., K.R. Cope, and B. Bugbee. 2016. Sensitivity of seven diverse species to blue and green light: interactions with photon flux. PLOS ONE 11:e0163121.

Stutte, G.W. 2009. Light-emitting diodes for manipulating the phytochrome apparatus. HortScience 44:231–234.

Taiz, L. and E. Zeiger. 2010. Plant physiology, 5th ed. Sinauer Associates. Sunderland, MA.

Talbot, L.D., G. Nikolova, A. Ortiz, I. Shmayevich, and E. Zeiger. 2002. Green light reversal of blue-light-stimulated stomatal opening is found in a diversity of plant species. Amer. J. Bot. 89:366–368.

Tixier, P. and H. de Bon. 2006. Urban horticulture. Cities Farming for the Future, Urban Agriculture for Green and Productive Cities. RUAF Foundation, IDRC and IIRR, Silang, The Philippines.

U.S. Dept. of Agr. 2011. USDA database for the flavonoid content of selected foods. Release 3. Nutrient Data Laboratory. Accessed Feb. 19 2016.

U.S. Dept. of Agr. 2014. Census of horticultural specialties. Census of Agriculture. Accessed April 6 2019. https://www.nass.usda.gov/Publications/AgCensus/2012/Online_Resources/Census_of_Horticult ure_Specialties/hortic_1_001_001.pdf/ van Wyk, B.E. 2014. Culinary herbs and spices of the world. Briza Publications. Pretoria, South Africa.

Wang, Y. and K.M. Folta. 2013. Contributions of green light to plant growth and development. Amer. J. Bot. 100:70–78.

Wheeler, R.M., C.L., Mackowiak, and J.C. Sager. 1991. Soybean stem growth under high- pressure sodium with supplemental blue lighting. Agron. J. 83:903–906.

Winkel-Shirley, B. 2002. Bioysnthesis of flavonoids and effects of stress. Current Opinion Plant Biol. 5:218–223.

Wittwer, S.H. and N. Castilla. 1995. Protected cultivation of horticultural crops worldwide. HortTechnology. 5:6–23.

Zhang, T., S.A. Maruhnich, and K.M. Folta. 2011. Green light induces shade avoidance symptoms. Plant Physiol. 157:1528–1536.

Zobayed, S. and P.K. Saxena. 2004. Production of St. John’s wort plants under controlled environment for maximizing biomass and secondary metabolites. In Vitro Cell. Dev. Biol. Plant 40:108–114.

Table 1. Light treatment composition of blue (B), green (G), and red (R) for B light fraction of 0 to 100% and white (W) light for ratio (%) of total, photosynthetic photon flux density (PPFD), and light source phytochrome stationary state (PSS) and relative quantum efficiency (RQE). Parameter Units B light fraction

0% 20% 40% 60% 80% 100% W

B (400-500 nm) % 0% 20% 40% 60% 80% 100% 21%

µmol·mol–2·s–1 0.0 60.0 120.0 180.0 240.0 300.0 61.9

G (500-600 nm) % 0% 0% 0% 0% 0% 0% 39%

µmol·mol–2·s–1 0.0 0.0 0.0 0.0 0.0 0.0 117.1

R (600-700 nm) % 100% 80% 60% 40% 20% 0% 38%

108

µmol·mol–2·s–1 300.0 240.0 180.0 120.0 60.0 0.0 113.9

FR (700-800nm) % 0% 0% 0% 0% 0% 0% 2%

µmol·mol–2·s–1 0.0 0.0 0.0 0.0 0.0 0.0 7.1

PSS* Pfr/Ptotal 0.88 0.88 0.87 0.84 0.80 0.48 0.86

RQE* PPFD (µmol·mol–2·s–1) 278.5 266.5 256.7 244.2 234.5 222.8 262.3

*Phytochrome photostationary state and relative quantum efficiency (Sager et al., 1988) 109

Fig. 1. Spectral distribution of LED lighting for blue light (B):red light (R) photon flux ratios of 0% B (0), 20% B (20), 40% B (40), 60% B (60), 80% B (80), 100% B (100), and white (W). Spectra was recorded with a spectroradiometer at plant height across growing media at 9 points per light source to confirm uniformity, and repeated at each repetition of the experiment. 110

Fig. 2. Height, internode length, and node number of basil (Ocimum basilicum) grown in deep-flow technique hydroponic systems under treatments of increasing blue (B) light or white (W). Each point represents the mean of five replications and bars represent the SE of the mean. Regression lines represent plant responses to B light fraction of 0% to 80% B light. *, **, or ***indicate significant differences in treatment means compared to control (white) using Holm-Sidak at P ≤ 0.05, 0.01, or 0.001, respectively. 111

Fig. 3. Total fresh mass (TFM), total dry mass (TDM), leaf fresh mass (LFM), and stem fresh mass (SFM) of basil (Ocimum basilicum) grown in deep-flow technique hydroponic systems under treatments of increasing blue (B) light or white (W). Each point represents the mean of five replications and bars represent the SE of the mean. Regression lines represent plant responses to B light fraction of 0% to 100% B light. *, **, or ***indicate significant differences in treatment means compared to control (white) using Holm-Sidak at P ≤ 0.05, 0.01, or 0.001, respectively. 112

Fig. 4. Leaf dry mass (LDM), stem dry mass (SDM), root dry mass (RDM), leaf mass ratio (LMR), shoot mass ratio (SMR), and root mass ratio (RMR) for basil (Ocimum basilicum) grown in deep-flow technique hydroponic systems under treatments of increasing blue (B) light or white (W). Each point represents the mean of five replications and bars represent the SE of the mean. Regression lines represent plant responses to B light fraction of 0% to 100% B light. *, **, or ***indicate significant differences in treatment means compared to control (white) using Holm-Sidak at P ≤ 0.05, 0.01, or 0.001, respectively. 113

Fig. 5. Individual leaf area, total leaf area, leaf area ratio (LAR), and specific leaf area (SLA) for basil (Ocimum basilicum) grown in deep-flow technique hydroponic systems under treatments of increasing blue (B) light or white (W). Each point represents the mean of five replications and bars represent the SE of the mean. Regression lines represent plant responses to B light fraction of 0% to 100% B light. *, **, or ***indicate significant differences in treatment means compared to control (white) using Holm-Sidak at P ≤ 0.05, 0.01, or 0.001, respectively. 114

Fig. 6. Photosynthesis (Pn), stomatal conductance (gs), transpiration (E), and intracellular CO2 concentration (Ci) compiled from morning (0600–0800 HR), midday (1200–1400 HR), and evening (2000–2200 HR) measurements within a day for basil (Ocimum basilicum) grown in deep-flow technique hydroponic systems under treatments of increasing blue (B) light or white (W). Each point represents the mean of five replications and bars represent the SE of the mean. Regression lines represent plant responses to B light fraction of 0% to 100% B light. *, **, or ***indicate significant differences in treatment means compared to control (white) using Holm-Sidak at P ≤ 0.05, 0.01, or 0.001, respectively. 115

Fig. 7. Chlorophyll fluorescence (Fv/Fm), chlorophyll a, chlorophyll b , and total chlorophyll (bottom) content for basil (Ocimum basilicum) grown in deep-flow technique hydroponic systems under treatments of increasing blue (B) light or white (W). Each point represents the mean of five replications and bars represent the SE of the mean. Regression lines represent plant responses to B light fraction of 0% to 100% B light. *, **, or ***indicate significant differences in treatment means compared to control (white) using Holm-Sidak at P ≤ 0.05, 0.01, or 0.001, respectively. 116

Fig. 8. Macronutrient leaf concentrations and content, calculated from leaf dry mass, for basil (Ocimum basilicum) grown in deep-flow technique hydroponic systems under treatments of increasing blue (B) light or white (W). Each point represents the mean of five replications and bars represent the SE of the mean. Regression lines represent plant responses to B light fraction of 0% to 100% B light. *, **, or ***indicate significant differences in treatment means compared to control (white) using Holm-Sidak at P ≤ 0.05, 0.01, or 0.001, respectively. 117

Fig. 9. Micronutrient leaf concentrations and whole canopy micronutrient content, calculated from leaf dry mass, for basil (Ocimum basilicum) grown in deep-flow technique hydroponic systems under treatments of increasing blue (B) light or white (W). Each point represents the mean of five replications and bars represent the SE of the mean. Regression lines represent plant responses to B light fraction of 0% to 100% B light. *, **, or ***indicate significant differences in treatment means compared to control (white) using Holm-Sidak at P ≤ 0.05, 0.01, or 0.001, respectively. 118

Fig. 10. Relative leaf concentrations and content, calculated from leaf fresh mass, of estragole, orientin, and vicenin for basil (Ocimum basilicum) grown in deep-flow technique hydroponic systems under treatments of increasing blue (B) light or white (W). Each point represents the mean of five replications and bars represent the SE of the mean. Regression lines represent plant responses to B light fraction of 0% to 100% B light. *, **, or ***indicate significant differences in treatment means compared to control (white) using Holm-Sidak at P ≤ 0.05, 0.01, or 0.001, respectively. 119

Fig. 11. Relative leaf concentrations and content, calculated from leaf fresh mass, of myricetin, kaempferol, and quercetin for basil (Ocimum basilicum) grown in deep-flow technique hydroponic systems under treatments of increasing blue (B) light or white (W). Each point represents the mean of five replications and bars represent the SE of the mean. Regression lines represent plant responses to B light fraction of 0% to 100% B light. *, **, or ***indicate significant differences in treatment means compared to control (white) using Holm-Sidak at P ≤ 0.05, 0.01, or 0.001, respectively. 120

CHAPTER 5: GENERAL CONCLUSIONS

I sought to measure how light quantity and quality affect culinary herb growth. This section of my dissertation distills my results into practical conclusions intended to provide a pathway for increasing the efficiency of electric lighting for optimizing culinary herb growth.

Chapter 2: Modeling of culinary herbs to daily light integral

In Chapter 2, we characterized hydroponic culinary herb growth in response to a range of daily light integrals (DLIs) for basil (Ocimum basilicum), dill (Anethum graveolens), parsley (Petroselinum crispum), mint (Mentha sp.), oregano (Origanum vulgare), sage

(Salvia officinalis), cilantro (Coriandrum sativum), and thyme (Thymus vulgaris). This is the first work to comprehensively evaluate herb species specific responses to DLI and, as a result, we classify five herb species as being high- or very high-light species (basil, dill, oregano, cilantro, and thyme), with another three species (parsley, mint, and sage) classified as medium-light species. My findings indicate increasing DLI with supplemental lighting from high-pressure sodium (HPS) will proportionally increase mass up to 20 mol·m–2·d–1 for most herbs with little to no decrease in the incremental mass per additional mole of DLI.

More importantly, these results indicate excessive lighting is possible for at least mint, parsley, and sage, with supra-optimal DLI conditions reducing yield.

With this work, additional questions arise about the appropriate methods for lighting.

To emphasize seasonal variability across experimental runs, supplemental lighting was turned off during the middle of the day, when ambient light is brightest, in accordance with commercial greenhouse practices; there is usually no need to provide supplemental light during the brightest times of the day. While this helped provide data for more a broader range 121 of DLI, it also reduced instantaneous light intensity during the brightest portion of the day.

Therefore, given photosynthesis may become saturated at higher light intensities, increasing non-photochemical quenching, it leads to questions about the effect on optimal DLI configurations. For example, what would be the effect on these growth curves if supplemental lighting were applied continuously from 0600 to 2200 HR? Additionally, while

HPS lamps have been the traditional method for greenhouse supplemental lighting, use of narrow-spectra light-emitting diodes (LEDs) has increased for both research and applications. Because of this, how would growth responses to DLI vary under different spectral compositions?

Chapter 3: Supplemental light source spectra

In Chapter 3, supplemental light from broad-spectrum HPS lighting was evaluated against dichromatic narrow spectra LEDs, with either low or high blue light proportions, for basil, dill, and parsley. Comparing growth under two proportions of red and blue spectra of

LED lamps to that of HPS, my findings indicate light source has little to no effect on dry mass accumulation, though plants grown high-blue LEDs have less water content than HPS- grown plants. This work further establishes modern LED supplemental lighting in greenhouses as a plant-efficient light source. Goals for efficient crop production aim to improve morphological and biomass traits, but emphasizing methods for increasing aroma and flavor profiles, which enhance nutritional value, can also increase crop value. My findings show LED lighting increases gas exchange, improves metabolic function, and enhances essential oil and flavonoid content, all of which correspond to desirable aroma and flavor attributes in culinary herbs. 122

During the course of the study, several questions arose about the environmental effect on the performance of supplemental lighting. In this study, all light sources were provides

~100 µmol·m–2·s–1. Bugbee (2016) reported secondary effects of light quality on growth, such as production of photoprotective pigments, are increased under higher light intensity.

Given this interactive effect of light intensity and light quality, how would plant responses differ under varying ambient and supplemental light DLI?

Chapter 4: Blue light fraction effect on basil plant functions

Chapter 4 delves into the spectral significance of blue light for promoting or restricting specific elements of plant growth and metabolism. Basil was selected as the model crop based on economic importance to growers, short life cycle, and physiological responses.

Although some work on blue light fraction on plant morphology or physiology has been conducted, this was the first work comprehensively looking at the effect of increasingly higher blue light proportions on everything from developmental growth to gas exchange and secondary metabolite accumulation of photoprotective compounds (flavonoids).

Additionally, because there are several reports on the benefits of broad-spectrum lighting from white LEDs for plant growth (Chen et al., 2016; Mickens et al., 2018), my work quantified the effects of white LEDs and compared them to the dichromatic spectra of red with increasing blue light fractions. Despite the morphological benefits of increased size and mass under white LEDs and high relative quantum efficiency (RQE) of the white LEDs compared to other treatments (Table 1), plants with moderate blue light proportions had improved photosynthetic efficiency and gas exchange, leading to higher mineral nutrient concentrations and flavonoid concentrations than those under white light. 123

With the increase in controlled-environment crop production for research and commercial purposes, the implications of this research are multi-faceted. First, while extremes of blue light fractions (0% to 100%) have their advantages and disadvantages, this work highlights the importance of balancing light spectra to maximize physiological responses and minimize disadvantages, with respect to yield. Second, the physiological response to blue light to increasing CO2 uptake, photosystem II (PSII) efficiency, and crop space efficiency per unit area presented in these findings hold potential for future work on photoreceptor efficiency, mitigating available ambient CO2 concentrations for increased metabolic efficiencies, and increasing plant densities. Lastly, this study is the first to quantify the white light’s deleterious effect on basil growth and metabolism. Because of the energy inefficiencies for producing white light LEDs from phosphor coated blue LEDs (Mickens and Assefa, 2014), plant growth and metabolism under white LEDs is similar or less than under standard dichromatic red and blue light highlights gaps in light optimization with white

LEDs.

During the course of this study, several questions arose pertaining to the physiological cause of responses, and possible additional parameters to investigate for future studies. While the photostationary state (PSS; Sager et al., 1988) is considered an acceptable method for estimating phytochrome response to lighting spectra, it is still unknown what the actual phytochrome responses were, and if there was any difference in the overall quantity of these photoreceptors per leaf area across lighting treatments. As such, if phytochromes red and far-red were quantified in this study, while their proportions could serve as updated support of PSS calculations, their quantification may offer potential for assessing red and far- red light sensitivity of plants adapted to various light spectra. Additionally, while flavonoid 124 concentrations were analyzed by HPLC, these data present only the final accumulation of these photoprotective compounds. Without information on gene expression regulation along the synthesizing pathways, it is not known if the final accumulation of flavonoids is due to the promotion of their synthesis, or the lack of catabolism. Because of this, additional future work on the effects of blue light on flavonoid biosynthesis could include characterization of key enzymatic activity along their pathways. If done, how does enzyme activity and gene expression along this pathway change with spectra, and are there key steps that can be up- or down-regulated to emulate the effect regardless of the spectral composition of the light source? Similarly, Assmann (1988) reported intracellular CO2 increases with increasing blue light. As a regulation of CO2 assimilation under enrichment, how does blue light affect plant growth under increasing CO2 concentrations? Lastly, further research into green light is necessary to increase the photosynthetic efficiency of plants under white LEDs. While some spectra, such as red and far-red light, can act synergistically by amplifying their combined effect on photosynthesis (Emerson and Rabinowitch, 1960), the antagonistic effect of green light on blue light-mediated responses is detrimental, but can be wavelength specific, with lower wavelengths less antagonistic than higher. Thus, by evaluating white LEDs with different peak spectra, can alternative spectral compositions be identified for photobiology that minimize the contrasting effects of individual spectra, while improving overall growth and metabolism?

Future suggestions

The findings of this research contribute to the scientific understanding of plant-light relations by modeling plant responses to increasing DLI and changing spectra. While these data contribute to optimization of environmental conditions, it is limited to the tools and 125 technology currently available. New developments in technology will continue to increase the ability to observe, measure, discriminate, and ultimately improve the methods by which crops in CEA are grown. My work establishes a necessary foundation for supplemental and sole-source lighting of culinary herbs, but also highlights the complexity of different physiological responses of herbs to light quantity and quality. These data can now be implemented into future CEA environmental studies and commercialization for continuing optimization of environmental factors to target and control plant growth.

References

Assmann, S.M. 1988. Enhancement of the stomatal response to blue light by red light, reduced intercellular concentrations of CO2, and low vapor pressure differences. Plant Physiol. 87:226–231.

Bugbee, B. 2016. Toward an optimal spectral quality for plant growth and development: the importance of radiation capture. Acta Hort. 1134:1–12.

Chen, X., X. Xue, W. Guo, L. Wang, and X. Qiao. 2016. Growth and nutritional properties of lettuce affected by mixed irradiation of white and supplemental light provided by light- emitting diode. Scientia Hort. 200:111–118.

Emerson, R. and E. Rabinowitch. 1960. Red drop and role of auxiliary pigments in photosynthesis. Plant Physiol. 35:447–485.

Sager, J.C., W.O. Smith, J.L. Edwards, and K.L. Cyr. 1988. Photosynthetic efficiency and phytochrome photoequilibria determination using spectral data. Amer. Soc. Agr. Eng. 31:1882–1889.