Plant Morphology and a Phytochrome B Model Reveal That the Effects of Far‐Red Light on Shade‐Avoidance‐Like Responses Persist Under High Light Intensity

ABSTRACT

Several studies have concluded that high photon flux density (PFD) attenuates the effects of the red (R; 600–699 nm) to far‐red (FR; 700–750 nm) light ratio on morphology. However, the suppressive effects can depend on individual wavebands that modulate photoreceptor activity. We postulated that morphological responses of shade‐avoiding plants to the FR fraction (FR‐PFD divided by R + FR PFD) act independent of total PFD (TPFD; 400–750 nm) when TPFD increases are only from R and FR light. We grew kale and lettuce under three FR fractions and four TPFDs while maintaining a constant blue (B; 400–499 nm) PFD. An increase in the R + FR PFD reduced leaf elongation and specific leaf area (SLA). However, higher light did not suppress the FR‐fraction effects on leaf elongation and SLA. We estimated PHYB activity with a three‐state PHYB model to mechanistically explain the suppressive effects of high light on leaf elongation and SLA but not on FR‐mediated leaf elongation and SLA increase. PHYB model predictions were in accordance with the morphological responses of kale and lettuce. This study is the first to apply the three‐state PHYB model to explain photon‐spectrum‐induced morphology of light‐grown whole plants, demonstrating its potential use to crops and for applications.

Summary statement

Increases in light intensity did not attenuate far‐red light mediated shade‐avoidance‐like responses of lettuce and kale when the blue light intensity was constant. Growth responses were in accordance with predictions from a three‐state phytochrome B model developed with Arabidopsis.

1. Introduction

Far‐red (FR; 700–750 nm) light and light intensity are environmental cues that plants utilize to respond to light environments (Casal 2012). Chlorophylls in plants absorb a greater proportion of red (R; 600–699 nm) photons than FR photons in incident light to excite photosynthesis (Caffarri et al. 2009; Hogewoning et al. 2012). Therefore, under canopy shade, the photon flux density (PFD) ratio of R and FR light (R:FR) is relatively low, and the FR fraction (FR PFD divided by the R + FR PFD) is relatively high (Kusuma and Bugbee 2021a). Plant species have evolved to acclimate their morphology to cope with shade when the R:FR is low, that is, when the FR fraction is high. The morphological acclimation includes petiole (or leaf) elongation, stem elongation, and an increase in specific leaf area (SLA; leaf area divided by leaf dry mass), which are collectively called shade‐avoidance responses (Casal 2012; Gommers et al. 2013). Such responses to FR light are not limited to the natural ranges of FR fraction created by canopy shade. For example, plants exhibit elongation growth in response to increased FR fraction, even under artificially low (less than sunlight) FR fractions (≤ 0.33) (Kong and Nemali 2021; Kusuma and Bugbee 2021a; Jeong et al. 2024a). To distinguish these responses to relatively low FR fractions (≤ 0.33) from those occurring under typical shade conditions outdoors (> 0.33), we refer to the morphological changes as shade‐avoidance‐like responses (SALR), rather than shade‐avoidance responses. Along with the FR fraction, light intensity also regulates such morphological responses (Ballaré and Pierik 2017). For example, an increase in total light intensity, such as photosynthetic PFD (PPFD; photon flux integral between 400 and 700 nm) and total PFD (TPFD; photon flux integral between 400 and 750 nm), inhibits extension growth, and vice versa. These inhibitory effects can occur through increased blue (B; 400–499 nm) PFD, which increases cryptochrome (CRY) activity (Cashmore et al. 1999; Djakovic‐Petrovic et al. 2007; Sasidharan et al. 2008; Keuskamp et al. 2011). Additionally, the suppressive influence of high light intensity on extension growth can be elicited by increased R or R + FR PFD, which activates phytochrome B (PHYB) (Ballaré et al. 1991; Chen et al. 2003; Allen et al. 2006; Khanna et al. 2007; Rausenberger et al. 2010; Trupkin et al. 2014; Johnson et al. 2020). Therefore, FR light and light intensity regulate plants' acclimation to light environments.

The influences of the FR fraction and light intensity in regulating plant morphology can depend on each other. In some studies, high light intensity diminished shade‐avoidance responses or SALR induced by the FR fraction. For example, an increase in the PPFD attenuated stem and/or leaf elongation induced by FR photons in lettuce (Lactuca sativa) and sunflower (Helianthus annuus) (Kurepin et al. 2007; Meng and Runkle 2019). Similarly, a high TPFD diminished stem elongation of lettuce elicited by FR photons (Kusuma and Bugbee 2023; Jeong et al. 2024b). Also, FR‐mediated internode elongation in white mustard (Sinapis alba) was negated without neutral shade (Ballaré et al. 1991). These studies, which reported the suppressive effects of high light intensity on the FR‐mediated elongation growth, kept relative spectral composition constant, which indicates that the B‐PFD and R (or R + FR) PFD were higher in the high‐light‐intensity treatments than in their lower‐intensity treatments. Thus, the suppressive effects of high light could, at least in part, be caused by the increased B‐PFD and/or R (or R + FR) PFD, which inhibit elongation growth by activating CRYs and/or PHYB (Ballaré et al. 1991; Chen et al. 2003; Allen et al. 2006; Khanna et al. 2007; Sasidharan et al. 2008; Keuskamp et al. 2011; Johnson et al. 2020).

PHYB, a photoreceptor that predominantly mediates plant responses to R and FR light, exists in three different forms of dimers, which consist of inactive (P_R) and/or active (P_FR) PHYB monomers: P_RP_R (D₀), P_RP_FR (D₁), and P_FRP_FR (D₂) (Klose et al. 2015). R photons convert D₀ to D₁ and D₁ are imported into the nucleus (Chen et al. 2003; Li et al. 2022). D₁ in the nucleus can be converted to D₂ upon R photon absorption. While not fully substantiated, D₂ appears to be an active PHYB dimer, which degrades and sequestrates transcriptional factors that contribute to morphological responses such as phytochrome interacting factors (PIFs) (Al‐Sady et al. 2006; Leivar et al. 2008; Park et al. 2012; Klose et al. 2015). Thus, the proportion of D₂ to the sum of D₀, D₁, and D₂ (D₂/D_tot), in part, represents the activity of PHYB dimers (Klose et al. 2015; Sellaro et al. 2019). Contrary to the action of R photons, FR photons can revert the conversion induced by R photons. For this reason, the relative proportion of R and FR photons (the FR fraction or R:FR) predominantly regulates the photoequilibria of PHYB dimers and their downstream signaling.

Increased B‐PFD not only inhibits elongation growth through CRY activation but also diminishes FR‐fraction effects on elongation growth by suppressing PIFs via activated CRYs (Bouly et al. 2007; Foreman et al. 2011; Fraser et al. 2016; Pedmale et al. 2016; Wang et al. 2018). However, while increased R (or R + FR) PFD can inhibit elongation growth by activating PHYB (Ballaré et al. 1991; Chen et al. 2003; Allen et al. 2006; Khanna et al. 2007; Rausenberger et al. 2010; Trupkin et al. 2014; Johnson et al. 2020), it remains unclear whether this also suppresses FR‐fraction effects on morphology. Therefore, this study aimed to determine if the suppressive effects of high light intensity on FR‐mediated morphological responses occur even when the B‐PFD is constant.

Phytochrome photoequilibria (PPE; Kusuma and Bugbee 2021b) and phytochrome photostationary state (PSS; Sager et al. 1988; Jishi 2024) are two‐state PHYB models that are often used in horticulture research to correlate plant growth responses with light quality. The PPE (or PSS) estimates PHYB activity based on photoconversion and does not consider other PHYB kinetics such as thermal reversion or light intensity (Mancinelli 1988). However, because of the interaction between photoconversion and thermal reversion, PHYB activity also depends on light intensity (Klose et al. 2020). For example, increasing the PFD with a given FR fraction activates PHYB (Chen et al. 2003; Trupkin et al. 2014). Contrary to PPE, the three‐state PHYB model incorporates thermal reversion and thus accounts for the influence of PFD. For this reason, the three‐state PHYB model is more appropriate than using PPE, especially under different light intensities. The three‐state PHYB model also considers the dimerization of PHYB monomers, unlike the PPE. Therefore, we explain the morphological responses to the FR fraction at varied light intensity with a constant B‐PFD by estimating PHYB activity (D₂/D_tot) with a three‐state PHYB model, which estimates in vivo D₂/D_tot (Klose et al. 2015; Smith and Fleck 2019; Sellaro et al. 2019). We postulated that morphological responses to the FR fraction will act independently of light intensity when the B‐PFD is constant, so in this study, the light intensity changes are only of R and FR photons. In addition, we hypothesized that high light intensity will not attenuate the FR‐mediated regulation of PHYB activity if the light intensity changes are only of R and FR photons.

2. Materials and Methods

2.1. Preparation of Plant Materials

Kale (Brassica oleracea var. sabellica) ‘Red Russian’, green butterhead lettuce ‘Rex’, and red oakleaf lettuce ‘Rouxai’ were selected for experimentation based on their sensitivity to photon spectra (i.e., SALR) and their commercial significance to indoor farming. Kale and lettuce are responsive to FR light and thus have been used to test a wide range of light responses in controlled‐environment agriculture (Meng et al. 2019; Kong and Nemali 2021; Kang et al. 2023; Kusuma and Bugbee 2023; Jeong et al. 2024b). We sowed seeds of these three cultivars (Johnny's Selected Seeds, Winslow, ME, United States) into 200‐cell (individual plug dimensions 2.5 cm × 2.5 cm × 4.0 cm) Rockwool plugs (Grodan AO Plug 25/40; Grodan, Milton, ON, Canada), one seed per cell, on June 30, 2023, August 16, 2023, and September 5, 2023, for the first, second, and third replication (day 0), respectively. We presoaked the plugs with deionized water adjusted with diluted sulfuric acid (J.Y. Baker Inc. Phillipsburg, NJ, United States) to provide a pH = 4.5 and electrical conductivity (EC) = 0.0 mS∙cm⁻¹ based on values measured with a pH and electrical conductivity meter (HI9814; Hanna Instruments, Woonsocket, RI, United States). We germinated (day 0–2) and propagated seedlings (day 3–8) under light with a PPFD of 180 µmol∙m⁻² ∙ s⁻¹ and a photoperiod of 24 h ∙ d⁻¹ from warm‐white (peak = 639 nm, correlated color temperature = 2700 K) light‐emitting diodes (LEDs) (Phytofy RL; OSRAM Opto Semiconductors, Beverley, MA, United States) in walk‐in rooms of the Controlled Environment Lighting Laboratory (CELL; Michigan State University, East Lansing, MI, United States). We covered the plug trays with transparent plastic domes to increase humidity and promote germination from day 0 to day 2. After germination, we sub‐fertigated the seedlings with deionized water supplemented with a water‐soluble fertilizer (12N‐4P‐16K RO Hydro FeED; JR Peters Inc. Allentown, PA, United States) and magnesium sulfate (Epsom salt, Pennington Seed Inc. Madison, GA, United States) at a pH = 5.6 and EC = 1.6 mS∙cm⁻¹. The nutrient solution contained the following (in mg∙L⁻¹): 125 N, 42 P, 167 K, 73 Ca, 49 Mg, 39 S, 1.7 Fe, 0.52 Mn, 0.56 Zn, 0.13 B, 0.47 Cu, and 0.13 Mo. The air temperature was set 23°C and controlled by an automatic air conditioning unit (HBH030A3C20CRS; Heat Controller LLC. Jackson, MI, United States). The air temperature was measured near the middle of the room every 10 s by four thermocouples (0.13‐mm type E; Omega Engineering Inc. Stamford, CT, United States) connected to a datalogger (CR1000; Campbell Scientific Inc. Logan, UT, United States) and the hourly average was logged. The actual air temperature was 22.9 ± 0.6°C, 23.2 ± 0.7°C, and 23.2 ± 0.5°C (mean ± standard deviation of hourly averages) for the first, second, and third replications, respectively. The CO₂ concentration measured with two nondispersive infrared CO₂ sensors (GMD20, Vaisala Inc. Vantaa, Finland) was 429 ± 25 ppm, 415 ± 31 ppm, and 444 ± 19 ppm for the first, second, and third replications, respectively.

2.2. FR Fraction and TPFD Treatments

When the first true leaf of the plants was expanding and their roots reached or emerged below the base of the plugs (day 9), we selected 216 seedlings with similar growth for each cultivar and transplanted them into floating rafts (60.9 × 121.9 × 2.5 cm with 72 holes; each with a diameter of 1 cm; Beaver Plastics Ltd. Acheson, AB, Canada). There was a 10.0 cm horizontal and 14.1 cm diagonal distance between each hole and thus a density of 97 plants∙m⁻². 18 seedlings per cultivar and treatment grew under each of twelve sole‐source lighting treatments that consisted of three levels of FR fraction (0.02, 0.16, or 0.33) and four TPFDs (85, 170, 255, or 340 µmol∙m⁻² ∙ s⁻¹) for 24 h ∙ d⁻¹ (Figures 1 and 2 and Table 1). TPFD, rather than PPFD, was similar across FR fraction treatments at each light intensity because FR light contributes equivalently to photosynthesis as traditional photosynthetically active radiation when the FR‐PFD remains below 40% of the PPFD (Zhen and Bugbee 2020). We chose the TPFDs to encompass those generally used in controlled‐environment research for indoor leafy‐green vegetable production, which are typically 100–300 µmol∙m⁻² ∙ s⁻¹ (Li and Kubota 2009; Son and Oh 2013; Meng et al. 2019; Pennisi et al. 2019; Spalholz et al. 2020; Kong and Nemali 2021; Jeong et al. 2024a). In addition, these TPFDs delivered daily light integrals ranging from 6.3 to 29.4 mol∙m⁻² ∙ d⁻¹, which are representative of those that occur outdoors in temperate climates (Faust and Logan 2018). We chose the FR fraction as the FR treatment metric because it better correlates with the morphological responses of plants than the R:FR, especially under a wide range of FR fractions or R:FRs (Kusuma and Bugbee 2021a). The highest FR fraction treatment (0.33) was similar to the FR fraction of unfiltered sunlight during the day (Fig. S1). Therefore, the FR fractions of 0.02 and 0.16 used in this study were artificially low. The B‐PFD was 50 µmol∙m⁻² ∙ s⁻¹ in all treatments and the TPFD changes were achieved by only changing the R and FR PFD. We delivered the photon spectra and flux density treatments by using the B, R, and FR narrowband LEDs in the fixtures used for propagation. We set the photon spectra and flux density treatments at initial seedling height based on measurements by a portable spectroradiometer (LI‐180, LI‐COR Biosciences Inc. Lincoln, NE, United States) at 15 locations for each floating raft. For each FR fraction and TPFD treatment, we calculated the PPFD, TPFD, FR fraction, estimated internal phytochrome photoequilibria (iPPE; Kusuma and Bugbee 2021b), and the proportion of the active phytochrome dimer (D₂) to the total phytochrome dimers (D_tot), or D₂/D_tot (Klose et al. 2015; Smith and Fleck 2019).

Figure 1.

Average spectral distributions of twelve lighting treatments with three different levels of far‐red fraction and four different sums of red (600–699 nm) and far‐red (700–750 nm) photon flux densities (R + FR PFD; A–D) delivered by blue (peak at 447 nm), red (peak at 660 nm), and far‐red (peak at 733 nm) light‐emitting diodes.

Figure 2.

Visual representation of twelve lighting treatments with three different levels of far‐red fraction and four different sums of red (R; 600–699 nm) and far‐red (FR; 700–750 nm) photon flux densities (PFD; in µmol∙m⁻² ∙ s⁻¹) delivered by blue (peak at 447 nm), R (peak at 660 nm), and FR (peak at 733 nm) light‐emitting diodes. The value following B, R, and FR indicates measured average blue (B; 400–499 nm), R, and FR PFD in µmol∙m⁻² ∙ s⁻¹, respectively. [Color figure can be viewed at wileyonlinelibrary.com]

Table 1.

Measured and calculated spectral characteristics of twelve lighting treatments (mean ± standard deviation from three experimental replications). The treatment value following R + FR indicates the target sum of red and far‐red photon flux density (in µmol∙m⁻² ∙ s⁻¹) and the value following FRF indicates the target far‐red light fraction of each treatment.

Treatment	Photon flux density (µmol∙m−2 ∙ s−1)					FRFf	iPPEg	D2/Dtot h	D2/Dtot therm (−) i	D2/Dtot blue (−) j
	Bluea	Red (R)b	Far red (FR)c	R+FRd	Totale
R + FR35FRF0.02	50.5 ± 0.2	35.3 ± 2.8	1.0 ± 0.8	36.3 ± 3.2	87.6 ± 2.7	0.03 ± 0.02	0.79 ± 0.03	0.35 ± 0.04	0.62 ± 0.05	0.40 ± 0.06
R + FR35FRF0.16	50.6 ± 0.9	29.0 ± 2.0	5.4 ± 0.2	34.4 ± 2.2	85.6 ± 2.4	0.16 ± 0.00	0.61 ± 0.01	0.16 ± 0.01	0.37 ± 0.01	0.17 ± 0.01
R + FR35FRF0.33	50.7 ± 0.5	22.8 ± 1.2	11.5 ± 1.0	34.3 ± 1.5	85.6 ± 0.9	0.34 ± 0.02	0.43 ± 0.02	0.07 ± 0.01	0.18 ± 0.02	0.07 ± 0.01
R + FR120FRF0.02	50.7 ± 0.8	119.7 ± 2.5	2.0 ± 0.9	121.7 ± 2.0	173.5 ± 3.0	0.02 ± 0.01	0.83 ± 0.01	0.58 ± 0.02	0.70 ± 0.02	0.61 ± 0.03
R + FR120FRF0.16	51.1 ± 0.6	101.4 ± 2.0	20.5 ± 0.3	121.9 ± 2.3	174.0 ± 1.9	0.17 ± 0.00	0.61 ± 0.00	0.28 ± 0.01	0.38 ± 0.01	0.28 ± 0.01
R + FR120FRF0.33	50.1 ± 1.3	81.0 ± 1.1	40.9 ± 1.2	121.9 ± 2.2	172.9 ± 2.2	0.34 ± 0.00	0.42 ± 0.01	0.12 ± 0.01	0.18 ± 0.01	0.12 ± 0.01
R + FR205FRF0.02	49.4 ± 0.6	204.9 ± 1.3	1.7 ± 0.4	206.6 ± 1.5	257.7 ± 1.1	0.01 ± 0.00	0.85 ± 0.00	0.66 ± 0.01	0.73 ± 0.01	0.67 ± 0.01
R + FR205FRF0.16	49.2 ± 2.5	171.3 ± 1.6	34.9 ± 0.3	206.2 ± 1.5	256.7 ± 1.7	0.17 ± 0.00	0.61 ± 0.00	0.31 ± 0.00	0.37 ± 0.00	0.31 ± 0.00
R + FR205FRF0.33	50.2 ± 1.4	137.2 ± 1.9	68.7 ± 0.6	205.9 ± 2.3	257.5 ± 1.0	0.33 ± 0.00	0.43 ± 0.01	0.14 ± 0.01	0.18 ± 0.01	0.14 ± 0.01
R + FR290FRF0.02	51.4 ± 1.3	284.8 ± 2.9	2.9 ± 0.3	287.7 ± 2.6	341.1 ± 1.2	0.01 ± 0.00	0.85 ± 0.00	0.67 ± 0.00	0.72 ± 0.00	0.68 ± 0.00
R + FR290FRF0.16	51.9 ± 0.5	243.6 ± 3.0	48.6 ± 0.7	292.2 ± 3.7	346.0 ± 3.4	0.17 ± 0.00	0.62 ± 0.01	0.33 ± 0.01	0.38 ± 0.01	0.33 ± 0.01
R + FR290FRF0.33	51.8 ± 0.8	193.8 ± 1.8	96.9 ± 0.7	290.7 ± 1.3	344.2 ± 1.9	0.33 ± 0.00	0.42 ± 0.01	0.15 ± 0.00	0.18 ± 0.01	0.15 ± 0.00

^a 400 to 499 nm.

^b 600 to 699 nm.

^c 700 to 750 nm.

^d 600 to 750 nm.

^e 400 to 750 nm.

^f FR fraction: Fraction of far‐red photon flux density (FR‐PFD; 700–750 nm) to the sum of FR‐PFD and red photon flux density (R‐PFD; 600–699 nm).

^g Estimated two‐state model‐based phytochrome monomer photoequilibria adjusted based on Kusuma and Bugbee (2021a).

^h Estimated phytochrome dimer equilibria [P_FRP_FR/(P_RP_R + P_RP_FR + P_FRP_FR)] based on the three‐state model proposed by Klose et al. (2015) and modified by Smith and Fleck (2019), which was calculated using adjusted cross‐section data from Kusuma and Bugbee (2021a).

ⁱ Estimated phytochrome dimer equilibria when the thermal reversion is excluded.

^j Estimated phytochrome dimer equilibria when the background blue light is excluded.

2.3. Additional Environmental Conditions During Treatments

We fertigated plants with the same nutrient solution used for seedling propagation. Each nutrient solution was oxygenated with a 60‐W air pump (Active Aqua AAPA70L; Hydrofarm, Petaluma, CA, United States) connected to a round flat air stone (20.3 × 2.5 cm; Active Aqua AS8RD; Hydrofarm). The actual air temperature was 23.2 ± 0.8°C, 23.3 ± 0.6°C, and 23.5 ± 0.5°C for the first, second, and third replications, respectively. The CO₂ concentration was 417 ± 36 ppm, 427 ± 37 ppm, and 441 ± 38 ppm for the first, second, and third replications, respectively. The air conditioning units operated continuously to circulate the air and improve environmental homogeneity. For each replication in time, we re‐randomized the location of each lighting treatment and the positioning of each cultivar under a lighting treatment to reduce any locational effects.

2.4. Biometric Data Collection

We collected biometric data from nine randomly selected plants of each cultivar and treatment, avoiding ones at the edge and excluding a few outliers with atypical growth, on day 16 (kale ‘Red Russian’), day 17 (lettuce ‘Rex’), and day 18 (lettuce ‘Rouxai’). The data collection date was based on the growth rate of each cultivar to avoid plant‐plant interactions (i.e., before canopy closure). We measured the length (mm), width (mm), and area (cm²) of the largest mature leaf (leaf blade and petiole) using a ruler and a leaf area meter (LI‐3100; LI‐COR, Lincoln, NE, United States), respectively. When measuring the leaf area, we dissected each leaf into several pieces to reduce the error from leaf curvature and folding. We calculated the ratio of leaf length to leaf width, which is often called the leaf‐shape index (Dinkins 2002; Son and Oh 2013; Chen et al. 2022; Jianxia et al. 2024). We used the leaf‐shape index to evaluate SALR because the ratio is less confounded by growth (e.g., different biomass at each R + FR PFD treatment) than absolute leaf length. We measured the fresh mass of shoots (excised at the surface of each plug) and the largest leaf (the same used for length and area measurements) using an analytical balance (GR‐200, A&D Store Inc. Wood Dale, IL, United States). We measured the dry mass of shoots and the largest leaf using the same analytical balance after drying them in a drying oven (Blue M, Blue Island, IL, United States) at 70°C for 5 days. Leaf area and dry mass of the largest leaf were used to calculate specific leaf area (SLA). In lettuce, petiole dry mass was included in the SLA calculation because there is no distinct boundary between the leaf blade and the petiole. In contrast, petiole dry mass was excluded from the kale SLA calculation because it has distinct petioles. The number of leaves (> 20 mm) at harvest was also counted. A photograph of each plant was taken from a top‐down view with a digital camera (NX300M, Samsung, Suwon, Republic of Korea) in a portable photo studio equipped with cool‐white LEDs.

2.5. The Three‐State PHYB Model

We used the three‐state PHYB model proposed by Klose et al. (2015) and modified by Smith and Fleck (2019) to estimate the impact of the photon spectra and PFD on the steady‐state (after prolonged exposure to the light environment) proportion of PHYB dimers: D₀, D₁, or D₂. The three‐state PHYB model considers the dimerization, degradation, photoconversion, and thermal reversion of PHYB, and thus accounts for the effect of PFDs. We assumed only D₂ is active and thus the proportion of D₂ to the sum of D₀, D₁, and D₂ was used to estimate the activity of PHYB (Sellaro et al. 2019; Klose et al. 2020). We used the steady‐state three‐state PHYB model because our lighting treatments were continuous and the air temperature was constant. We calculated the steady‐state proportion of active PHYB dimers [D₂/(D₀ + D₁ + D₂)] using the following equation (Smith and Fleck 2019):

$$\frac{D_2}{(D_0+D_1+D_2)}=\frac{D_2}{D_{\mathit{tot}}}=\frac{2{k_1}^2}{2{k_1}^2+(2k_1+k_2+k_{r1})[2\left(k_2+k_{r2}\right)+k_{\mathit{dfr}}]}$$

where $k_1$ is the photoconversion rate of $D_0$ to $D_1$ and $D_1$ to $D_2$, $k_2$ is the photoconversion rate of $D_2$ to $D_1$ and $D_1$ to $D_0$, $k_{r1}$ is the thermal reversion rate of $D_1$, $k_{r2}$ is the thermal reversion rate of $D_2$, and $k_{\mathit{dfr}}$ is the rate of P_FR‐specific degradation. We calculated $k_1$ and $k_2$ using the two equations:

$$k_1=\sum _{{\lambda }_{\min }}^{{\lambda }_{\max }}{6\times {10}^{-5}\sigma }_{\lambda }^{\mathit{Red}}{N}_{\lambda }$$

$$k_2=\sum _{{\lambda }_{\min }}^{{\lambda }_{\max }}6\times {10}^{-5}{\sigma }_{\lambda }^{\mathit{Far\; red}}{N}_{\lambda }$$

using photoconversion cross‐section data (${\sigma }^{\mathit{Red}}$ and ${\sigma }^{\mathit{Far\; red}}$ in m²∙mol⁻¹), and PFD measured at each wavelength at each experimental replication ($N_{\lambda }$ in µmol∙m⁻² ∙ s⁻¹; Figure. 1 and Table 1). We used photoconversion cross‐section data adjusted by Kusuma and Bugbee (2021a), which accounts for the spectral distortion that occurs in light‐grown leaves due to photon scattering and photon absorption by phytopigments such as chlorophylls. We obtained $k_{r1}$ and $k_{r2}$ from Viczián et al. (2020) using a plot digitizer (WebPlotDigitizer v.4.6, Ankit Rohatgi, Pacifica, CA, USA). $k_{r1}$ and $k_{r2}$ exponentially increase with temperature. The average temperature during the FR fraction and R + FR PFD treatments in our study was 23.3°C. However, the $k_{r1}$ and $k_{r2}$ for 23.3°C were not specifically reported by Viczián et al. (2020). Thus, we interpolated the $k_{r1}$ and $k_{r2}$ reported for 22°C and 27°C to estimate $k_{r1}$ and $k_{r2}$ at 23.3°C, which were 6.234 min⁻¹ and 0.0048 min⁻¹, respectively. We used $k_{\mathit{dfr}}$ of 0.0024 min⁻¹ as measured by Rausenberger et al. (2010) under continuous lighting (the same as in our study). $k_{\mathit{dfr}}$ is sometimes ignored in steady‐state three‐state PHYB model calculations for simplification (Sellaro et al. 2019). In contrast, a very high $k_{\mathit{dfr}}$ (~50 min⁻¹) is sometimes used partly because $k_{\mathit{dfr}}$ is highly influenced by light duration (Rausenberger et al. 2010; Klose et al. 2015).

Thermal reversion is a major factor that interacts with light intensity to regulate PHYB activity (Klose et al. 2020). Thus, thermal reversion may alter the influence of the FR fraction at different light intensities. We simulated the steady‐state three‐state model to estimate the effects of thermal reversion in regulating the influences of the FR fraction at different R + FR PFDs. We calculated the D₂/D_tot with and without thermal reversion using the photon spectra and PFD used to grow plants (Figure 1 and Table 1). We compared the D₂/D_tot calculated with and without thermal reversion to evaluate the influence of thermal reversion.

B photons regulate PHYB photoequilibria along with R and FR photons (Sager et al. 1988; Kusuma and Bugbee 2021a). Thus, the 50 µmol∙m⁻² ∙ s⁻¹ of B photons delivered in our study could affect the influence of the FR fraction on PHYB activity at different R + FR PFDs. Therefore, we used the model to estimate the effects of the B light in regulating the effects of the FR fraction at different R + FR PFDs. We calculated D₂/D_tot with and without the photon spectra and PFD of B light used in our treatments and compared the calculations to evaluate the influence of B light.

2.6. Influence of TPFD With Proportional B‐PFD Change

We conducted an additional experiment to determine whether high TPFD with a proportional increase in B‐PFD suppresses FR‐fraction effects on the morphology in the same three crops. The range of FR fraction also increased from 0.01 to 0.33 to 0.01–0.50. Except for the lighting treatments, experimental methods were as previously described. Seedlings grew under each of six sole‐source lighting treatments that consisted of three levels of FR fraction (0.01, 0.33, or 0.50) and two TPFDs (85 or 340 µmol∙m⁻² ∙ s⁻¹), all of which had a B light fraction of 35% (Figure 3 and Table S1). We delivered the photon spectra and flux density treatments by using B, R, and FR narrowband LEDs (GreenPower LED production module; Koninklijke Philips N.V., Amsterdam, The Netherlands).

Figure 3.

Average spectral distributions of six lighting treatments with three different levels of far‐red fraction and two different total photon flux densities (TPFD; 400–750 nm; A, B) delivered by blue (peak at 445 nm), red (peak at 660 nm), and far‐red (peak at 736 nm) light‐emitting diodes. Blue photon flux density (B‐PFD; 400–499 nm) was proportionally changed with TPFD (35%). [Color figure can be viewed at wileyonlinelibrary.com]

2.7. Statistical Analysis

For the main experiment (TPFD changes only from R + FR PFD), we assigned the treatments in a complete randomized design with three experimental replications (n = 3; each with nine biological replications per cultivar). We used the same design for the second experiment (TPFD changes with proportional B‐PFD change) but with two experimental replications (n = 2; each with nine biological replications per cultivar). Statistical analysis was conducted using the package emmeans (Lenth 2022) and agricolae (de Mendiburu 2021) in R statistical analysis software (R Core Team 2021). To check the assumptions for the analysis of variance (ANOVA) test of the main experiment, we tested normality of residuals and homogeneity of variance using Shapiro‐Wilk's test and Levene's test, respectively. All measured and calculated biometric data except for the SLA of kale and the leaf number of lettuce ‘Rex’ fulfilled the assumptions. For the SLA of kale and the leaf number of lettuce ‘Rex’, we identified outliers based on Cook's distance threshold of four divided by the total number of observations (Fig. S2). Three and two identified outliers were removed for the SLA of kale and the leaf number of lettuce ‘Rex’, respectively. After the outlier removal, these two biometric data sets became unbalanced and satisfied the assumptions. We also tested the normality of residuals for the second experiment. The main and interaction effects of the treatments were evaluated using type Ⅲ two‐way ANOVA, followed by Tukey's honestly significant difference test, except for unbalanced data, which were evaluated by Tukey‐Kramer test instead. We used type Ⅲ ANOVA because the statistical inferences from type Ⅰ ANOVA were consistent with those from type Ⅲ ANOVA and because some of the data were unbalanced. Values of p < 0.05 were considered statistically significant.

3. Results

3.1. SALR to the R + FR PFD and the FR Fraction

Increasing light intensity by increasing the R + FR PFD decreased the leaf‐shape index of all cultivars at all FR fractions (Figures 4 and 5). For example, increasing the R + FR PFD from 35 to 290 µmol∙m⁻² ∙ s⁻¹ at a given FR fraction decreased the leaf‐shape index by 9–11% in kale, by 24–29% in lettuce ‘Rex’, and by 21–24% in lettuce ‘Rouxai’. In contrast, increasing the FR fraction increased the leaf‐shape index of kale and green‐leaf lettuce ‘Rex’ at all R + FR PFDs. For example, increasing the FR fraction from 0.02 to 0.33 at a given R + FR PFD increased the leaf‐shape index by 17–23% in kale and by 30–38% in lettuce ‘Rex’. The degree of increase in the leaf‐shape index elicited by high FR fraction was similar at each R + FR PFD in kale and lettuce ‘Rex’ on an absolute and relative basis. However, the response of red‐leaf lettuce ‘Rouxai’ to the FR fraction was less consistent than the other cultivars.

Figure 4.

Representative photographs taken 16, 17, or 18 days after seed sow of kale ‘Red Russian’, lettuce ‘Rex’, and lettuce ‘Rouxai’, respectively. Plants were grown under twelve lighting treatments. Far‐red fraction refers to the fraction of the far‐red (FR; 700–750 nm) photon flux density (PFD) relative to the sum of the red (R; 600–699 nm) and FR PFD. The red + far‐red photon flux density refers to the photon flux integral between 600 nm and 750 nm, in µmol∙m⁻² ∙ s⁻¹. All plants received a PFD of blue light at 50 µmol∙m⁻² ∙ s⁻¹. The bar on the right‐top corner applies to all images. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 5.

Influence of twelve lighting treatments on the leaf‐shape index (the ratio of leaf length to leaf width) and specific leaf area of the largest leaf of kale ‘Red Russian’ and lettuce ‘Rex’ and ‘Rouxai’. Far‐red fraction (FR fraction) refers to the fraction of the far‐red (FR; 700–750 nm) photon flux density (PFD) relative to the sum of the red (R; 600–699 nm) and FR PFD. R + FR PFD refers to the photon flux integral between 600 nm and 750 nm, in µmol∙m⁻² ∙ s⁻¹. Absolute values (A, B, C, G, H, I) were normalized to the average value at the lowest FR fraction for each R + FR PFD (D, E, F, J, K, L). DM refers to dry mass. Means with different letters are significantly different based on Tukey's HSD test (p < 0.05), except for kale specific leaf area, which was evaluated using Tukey‐Kramer test because of an unbalanced data set. Error bars represent standard error for n = 3 replicates unless marked with *, where standard error is based on n = 2 replicates due to outlier removal. The p values in the tables indicate the significance of the main and interaction effects based on type Ⅲ two‐way analysis of variance test. [Color figure can be viewed at wileyonlinelibrary.com]

Increasing the R + FR PFD generally decreased the SLA of all cultivars at each FR fraction. For example, increasing the R + FR PFD from 35 to 290 µmol∙m⁻² ∙ s⁻¹ at a given FR fraction decreased the leaf‐shape index by 39–44% in kale (except at an FR fraction of 0.02, where the decrease was statistically insignificant), by 31–50% in lettuce ‘Rex’, and by 33–43% in lettuce ‘Rouxai’. Increasing the FR fraction from 0.02 to 0.33 did not significantly affect the SLA of kale at any of the R + FR PFDs. The increase in the FR fraction increased the SLA of lettuce ‘Rex’ and ‘Rouxai’ at the R + FR PFDs of 120, 205, and 290 µmol∙m⁻² ∙ s⁻¹ but not at the R + FR PFD of 35 µmol∙m⁻² ∙ s⁻¹. The increase in the FR fraction increased the SLA of lettuce ‘Rex’ and ‘Rouxai’ by 27–39% and 15–28%, respectively, at the R + FR PFDs > 35 µmol∙m⁻² ∙ s⁻¹. Overall, increasing the R + FR PFD did not attenuate the FR‐fraction effects on SLA on an absolute and relative basis.

3.2. Biomass Accumulation, Leaf Expansion, and Development Responses to the R + FR PFD and the FR Fraction

Increasing the R + FR PFD increased the shoot dry mass of all cultivars at all FR fractions (Figure 6 and S3). Increasing the FR fraction did not significantly affect the shoot dry mass of kale at any R + FR PFD except for the highest one. Similarly, the change in the FR fraction did not significantly affect the shoot dry mass of lettuce ‘Rex’ and ‘Rouxai’. Increasing the R + FR PFD expanded the leaves of all cultivars at all FR fractions. Increasing the FR fraction did not affect the leaf area of kale at any R + FR PFD (Figures 4 and 6). Lettuce leaf size was more responsive than kale and generally increased with the FR fraction. Overall, the FR fraction and the R + FR PFD did not interact to regulate the shoot dry mass and leaf area of any crop. The FR fraction did not influence the leaf number (per plant) of any cultivar at any of the R + FR PFDs tested (Figure 6).

Figure 6.

Influence of twelve lighting treatments on the shoot dry mass (A, B, C), leaf area of the largest leaf (D, E, F), and leaf number per plant (G, H, I) of kale ‘Red Russian’ and lettuce ‘Rex’ and ‘Rouxai’. Far‐red fraction (FR fraction) refers to the fraction of the far‐red (FR; 700–750 nm) photon flux density (PFD) relative to the sum of the red (R; 600–699 nm) and FR PFD. Means with different letters are significantly different based on Tukey's HSD test (p < 0.05), except for the leaf number of lettuce ‘Rex’, which was evaluated using Tukey‐Kramer test because of an unbalanced data set. Error bars represent standard error for n = 3 replicates unless marked with *, where standard error is based on n = 2 replicates due to outlier removal. The p values in the tables indicate the significance of the main and interaction effects based on type Ⅲ two‐way analysis of variance test. [Color figure can be viewed at wileyonlinelibrary.com]

3.3. PHYB Activity (D₂/D_Tot) Responses to the R + FR PFD and the FR Fraction

We estimated PHYB activity with the three‐state PHYB model developed with Arabidopsis (Klose et al. 2015; Smith and Fleck 2019) to mechanistically explain the suppressive effects of a high R + FR PFD on leaf elongation and SLA and the lack of a suppressive effect of a high R + FR PFD on FR‐mediated leaf elongation and SLA increase. We also conducted a sensitivity analysis of the model under simulated FR fractions, R + FR PFDs, and temperatures to illustrate the relative effects of each other (Figure 7). A greater D₂/D_tot indicates increased PHYB activity and vice versa.

Figure 7.

Sensitivity analysis of the three‐state phytochrome B model developed for Arabidopsis (Klose et al. 2015; Smith and Fleck 2019), which estimates the proportion of active phytochrome B dimer (D₂/D_tot) under simulated red (R; 660 nm) and far‐red (FR; 730 nm) light. The FR fraction refers to the fraction of the FR photon flux density (PFD) relative to the sum of the R and FR PFD (R + FR PFD; in µmol∙m⁻² ∙ s⁻¹). In the simulations, 24°C was used in sensitivity to the FR fraction and R + FR PFD (A, D); an R + FR PFD of 290 µmol∙m⁻² ∙ s⁻¹ was used in the FR fraction and temperature (B, E) sensitivity; and an FR fraction of zero was used in the R + FR PFD and temperature (C, F) sensitivity. The R + FR PFD, the FR fraction, or the temperature in each subfigure indicates stable conditions. [Color figure can be viewed at wileyonlinelibrary.com]

The estimated D₂/D_tot calculated with the photon spectral distribution of our lighting treatments (Figure 1 and Table 1) indicate that higher R + FR PFDs increased D₂/D_tot (but at a decreasing rate) at each FR fraction (Figure 8A). We also calculated and plotted changes of D₂/D_tot relative to those at the lowest FR fraction (Figure 8B). The decrease in D₂/D_tot induced by a higher FR fraction was not affected by the R + FR PFD.

Figure 8.

The estimated proportion of active phytochrome B dimer (D₂/D_tot; Klose et al. 2015; Smith and Fleck 2019) at three levels of far‐red fraction (A) and the effect of a high far‐red fraction in decreasing the D₂/D_tot relative to the lowest far‐red fraction (B) at four different red + far‐red photon flux densities (PFDs). Far‐red fraction (FR fraction) refers to the fraction of the far‐red (FR; 700–750 nm) PFD relative to the sum of the red (R; 600–699 nm) and far‐red PFD. Data points and error bars represent the mean and standard error of D₂/D_tot, estimated using actual photon spectra measured at each of the three experimental replications (n = 3). Means with different letters are significantly different based on Tukey's HSD test (p < 0.05). The p values in the tables indicate the significance of the main and interaction effects based on type Ⅲ two‐way analysis of variance test. [Color figure can be viewed at wileyonlinelibrary.com]

Next, we estimated D₂/D_tot with and without PHYB thermal reversion (Figure 9A). The estimated D₂/D_tot was generally lower in the presence of thermal reversion than in its absence at low R + FR PFDs. However, this thermal reversion effect on decreasing D₂/D_tot was negated as the R + FR PFD increased. We also estimated the D₂/D_tot with and without the 50 µmol∙m⁻² ∙ s⁻¹ of B light that was delivered in all treatments (Figure 9B). The background B light did not significantly influence the effect of the FR fraction and R + FR PFD in regulating the D₂/D_tot in the tested range.

Figure 9.

The estimated proportion of active phytochrome B dimer (D₂/D_tot; Klose et al. 2015; Smith and Fleck 2019) at three levels of far‐red fraction with (closed symbol and solid line) and without (open symbol and dashed line) thermal reversion (A) or background blue light (B) at four different red + far‐red photon flux densities (PFDs). Far‐red fraction (FR fraction) refers to the fraction of the far‐red (FR; 700–750 nm) PFD relative to the sum of the red (R; 600–699 nm) and far‐red PFD. Data points and error bars represent the mean and standard error of D₂/D_tot, estimated using actual photon spectra measured at each of the three experimental replications (n = 3). Means with different letters are significantly different based on Tukey's HSD test (p < 0.05). [Color figure can be viewed at wileyonlinelibrary.com]

3.4. Influence of TPFD With Proportional B‐PFD Change on Shade‐Avoidance Responses or SALR

In the first experiment, light intensity was increased by only increasing the R + FR PFD; in the second experiment, the B light fraction was constant and thus, the B‐PFD increased with the TPFD. The higher TPFD (and higher B‐PFD) reduced leaf elongation and SLA at all FR fractions (Figure 10). In contrast, increasing the FR fraction from 0.01 to 0.50 elongated the leaves at both TPFDs in all three crops. The FR‐mediated leaf elongation was less pronounced under high light than under low light. Similarly, while FR‐mediated SLA increases were significant at the lower TPFD in kale and green‐leaf lettuce ‘Rex’, they were not at the higher TPFD. However, the higher TPFD did not diminish the effects of the FR‐fraction on the SLA of red‐leaf lettuce ‘Rouxai’. Therefore, the higher TPFD (with a higher B‐PFD) generally suppressed FR‐mediated shade‐avoidance responses or SALR, particularly leaf elongation.

Figure 10.

Influence of six lighting treatments on the leaf‐shape index (the ratio of leaf length to leaf width; A, B, C) and specific leaf area (D, E, F) of the largest leaf of kale ‘Red Russian’ and lettuce ‘Rex’ and ‘Rouxai’. Far‐red fraction (FR fraction) refers to the fraction of the far‐red (FR; 700–750 nm) photon flux density relative to the sum of the red (R; 600–699 nm) and far‐red photon flux density. The total photon flux density (TPFD) refers to the photon flux integral between 400 and 750 nm, in µmol∙m⁻² ∙ s⁻¹. The fraction of blue (400–499 nm) photon flux density to TPFD was identical at both TPFDs (35%). DM refers to dry mass. Means with different letters are significantly different based on Tukey's HSD test (p < 0.05). Error bars represent standard error for n = 2 replicates. The p values in the tables indicate the significance of the main and interaction effects based on type Ⅲ two‐way analysis of variance test. [Color figure can be viewed at wileyonlinelibrary.com]

4. Discussion

Shade‐avoiding plants acclimate to shade by increasing stem and leaf elongation and SLA to increase light capture (Casal 2012; Gommers et al. 2013). In our study, increasing the TPFD from 85 to 340 µmol∙m⁻² ∙ s⁻¹ with only R and FR photons did not diminish the effects of the FR fraction in inducing SALR (i.e., the leaf‐shape index and SLA increase) in all three crops studied (Figure 5). Consistent with these responses, the three‐state PHYB dimer model predicted that the increase in TPFD with only R and FR photons would not suppress the influence of a high FR fraction in decreasing PHYB activity (D₂/D_tot) (Figure 8B). In all cultivars, there was an inverse linear relationship between the leaf‐shape index and SLA with the estimated D₂/D_tot at the FR fractions and R + FR PFDs tested in our study (Figure 11). The inverse linear model provided an excellent fit for the leaf‐shape index of all three crops based on the normalized root mean square error (NRMSE) rating criteria (e.g., excellent if NRMSE < 10% and good if 10% ≤ NRMSE < 20%) (Jamieson et al. 1991). The model fit for the SLA of all three crops was classified as good. Therefore, the D₂/D_tot estimated through the three‐state PHYB model developed with Arabidopsis can be used to estimate and explain SALR of shade‐avoiding species such as kale and lettuce in a range of low to moderate FR fractions and R + FR PFDs. This study is the first to apply the three‐state PHYB model to explain the morphology of light‐grown whole plants, demonstrating its potential use with other species and lighting applications.

Figure 11.

Influence of the estimated phytochrome B dimer activity (D₂/D_tot; Klose et al. 2015; Smith and Fleck 2019) of twelve lighting treatments on the leaf‐shape index (the ratio of leaf length to leaf width) and specific leaf area of kale ‘Red Russian’ and lettuce ‘Rex’ and ‘Rouxai’. The far‐red fraction refers to the fraction of the far‐red (FR; 700–750 nm) photon flux density (PFD, in µmol∙m⁻² ∙ s⁻¹) to the sum of red (R; 600–699 nm) and FR PFD. DM refers to dry mass. Each data point and error bar represent the mean and standard error for n = 3 replicates, respectively, except the ones for kale specific leaf area at the R + FR PFD of 35 µmol∙m⁻² ∙ s⁻¹, which were for n = 2 replicates due to outlier removal. The r² indicates the coefficient of determination. RMSE indicates root mean square error. Normalized RMSE is the RMSE divided by the mean of observed values. [Color figure can be viewed at wileyonlinelibrary.com]

Our results contrast with some previous research, which demonstrated that high light intensity attenuated shade‐avoidance responses or SALR, particularly elongation growth, induced by the FR fraction (Ballaré et al. 1991; Kurepin et al. 2007; Meng and Runkle 2019; Kusuma and Bugbee 2023; Jeong et al. 2024b). While the suppressive effects of high light intensity on FR‐mediated leaf elongation occurred when the B‐PFD proportionally increased with TPFD (Figures 3 and 10), these did not occur at a constant B‐PFD (Figures 1 and 5). Thus, we attribute the suppressive effects of high light intensity in the previous studies to the B‐PFD, which increased along with the TPFD or PPFD. A high B‐PFD can attenuate shade‐avoidance responses or SALR induced by a high FR fraction through the suppression of E3 ligase CONSTITUTIVE PHOTOMORHPOGENIC 1, HYPOCOTYL IN FAR‐RED 1, and DELLA proteins on PIFs (Achard et al. 2007; Foreman et al. 2011; Crocco et al. 2015; Fraser et al. 2016; Pedmale et al. 2016; Wang et al. 2018). Similar to the results of our study, the effects of the FR fraction (or R:FR) in regulating the stem length of a few ornamental species were independent of PPFD when the B‐PFD was constant (Park and Runkle 2018).

Increased R + FR PFD can inhibit elongation growth by activating PHYB, which suppresses PIFs (Ballaré et al. 1991; Chen et al. 2003; Allen et al. 2006; Khanna et al. 2007; Rausenberger et al. 2010; Trupkin et al. 2014; Johnson et al. 2020). However, the influence of the R + FR PFD has been primarily investigated in seedling hypocotyls or adult plants over the short term (e.g., leaf hyponasty responses in a few hours). We investigated relatively long‐term responses of adult plants to the R + FR PFD and demonstrated that increased R + FR PFD suppresses leaf elongation and SLA increase of mature plants (Figure 5). Consistent with the plant morphological responses in this study, D₂/D_tot estimated with the three‐state PHYB model predicted that increasing the R + FR PFD would increase PHYB activity (Figure 8A). An increase in R + FR PFD can activate PHYB by enhancing the influence of photoconversion on PHYB activity relative to thermal reversion (Figure 9A).

The SLA of both lettuce cultivars showed less of an effect of the FR fraction on SALR at the lowest R + FR PFD than at higher R + FR PFDs (Figure 5). This does not align with the predicted PHYB activity (Figure 8B). It is likely that plants have an inherent maximum SLA (i.e., minimum leaf thickness) to preserve leaf functionality (Aneja et al. 2025). This might have been almost reached at the lowest PFD tested, regardless of the FR fraction. As a result, the plasticity of SLA to the FR fraction could be low under a low PFD. Compared to the SLA of lettuce, the SLA of kale was less responsive to the FR fraction. Morphological acclimations to light environments vary among plant species, and are often attributed to origin or genetic variations (Weijschedé et al. 2006; Ikeda et al. 2021). The low responsiveness of kale SLA in our study implies that kale often responds to FR light through leaf (or petiole) elongation rather than SLA increase.

In the past, experimental measurement of the active PHYB (P_FR or D₂) concentration in vivo was impossible (Mancinelli 1988). However, advancements in transgenics and gene‐expression regulation technologies have enabled the measurement of active PHYB in vivo (Sweere et al. 2001; Chen et al. 2003; Rausenberger et al. 2010; Klose et al. 2015). For example, transgenic lines expressing PHYB‐green fluorescence (PHYB‐GFP or PBG) and PHYB‐yellow fluorescence transgenes in a PHYB‐deficient mutant (Yamaguchi et al. 1999; Gil et al. 2000; Rausenberger et al. 2010) enabled the microscopic fluorescence measurement of PHYB nuclear body in vivo in Arabidopsis and tobacco (Nicotiana tabacum). Also, PHYB‐overexpressing (PHYBox) lines enabled the measurement of active PHYB through a dual‐wavelength spectrophotometric technique in a few plant species (Wagner et al. 1991; Sweere et al. 2001; Rausenberger et al. 2010; Klose and Hiltbrunner 2024). However, such PHYB‐fluorescence transgenic lines have been developed for few plant species (Yamaguchi et al. 1999; Gil et al. 2000). In addition, the dual‐wavelength spectrophotometric technique can be used only for chlorophyll‐deficient plants because chlorophyll interferes with PHYB measurement (Jabben and Deitzer 1978; Klose and Hiltbrunner 2024). Thus, at least for now, direct measurement of PHYB in vivo is impossible for most light‐grown plant species. For this reason, the utilization of the PHYB model, which is constructed based on the PHYB kinetics measured in Arabidopsis, remains the only viable method to estimate in vivo PHYB activity in plant species other than a few model plant species.

PHYB activity estimated with a PHYB model can be different from actual in vivo PHYB activity because of the non‐photochemical kinetics of PHYB (e.g., thermal reversion and degradation) and the spectral distortion within plants (Mancinelli 1988). We addressed this issue by using the three‐state PHYB model that accounts for thermal reversion and PHYB degradation (Klose et al. 2015; Smith and Fleck 2019). We estimated the PHYB thermal reversion and degradation rates for the experimental conditions (e.g., temperature and photoperiod) based on previous studies (Rausenberger et al. 2010; Viczián et al. 2020). In addition, we used the PHYB photochemical cross‐section, which accounts for light distortion that occurs in light‐grown plants (Kusuma and Bugbee 2021a). Nevertheless, because many of the kinetics parameters used in the three‐state PHYB model were measured or calculated in the PHYBox‐line Arabidopsis, the estimates can be different from the actual in vivo PHYB activity in other plant species (Klose et al. 2015; Smith and Fleck 2019; Viczián et al. 2020). For example, overexpression of PHYB can induce hypersensitivity of PHYB activity to light intensity (Wagner et al. 1991). Thus, caution is needed when utilizing PHYB estimation to predict the morphological responses of plants. This is consistent with variation in the magnitude of shade‐avoiding responses among species, sometimes even among cultivars or ecotypes of the same species.

The D₂/D_tot estimated through the PHYB model and applied to our study predicted that background B light would not significantly influence the PHYB activity at any of the tested FR fractions and R + FR PFDs (Figure 9B). P_R and P_FR monomers have the greatest absorption of R and FR light, respectively (Kelly and Lagarias 1985; Lagarias et al. 1987; Sager et al. 1988; Mancinelli 1994). P_R and P_FR monomers also absorb B light; however, the B‐light‐absorption peaks of P_R and P_FR are somewhat similar and thus, B light is less effective in activating PHYB than R photons and in deactivating PHYB than FR photons (Fig. S4). For this reason, background B light theoretically can decrease PHYB activity, particularly if R light is predominant over FR light. In contrast, because light and temperature interact to regulate PHYB activity (Klose et al. 2020), the inclusion of background B light, which increased the overall PFD, could have increased photoconversion relative to thermal reversion. Therefore, PHYB activity might not have been appreciably influenced by background B light because the slightly decreased photoequilibria, and greater photoconversion than thermal reversion caused by B background B light, offset the effects of each other.

The utilization of the three‐state PHYB model to mechanistically explain the SALR in our study is possible because the B‐PFD was similar in all lighting treatments. CRYs, phototropins (PHOTs), and PHYB are the photoreceptors that primarily regulate the morphological responses of plants to photon spectra and PFD. Inactive CRYs absorb ultraviolet (UV; 200–399 nm) and B photons to become active and suppress extension growth (Battle et al. 2020). While activated CRYs absorb short‐wavelength R photons (e.g., 600–650 nm), the significance of their influence on CRY activity is unknown. Similar to CRYs, PHOTs absorb UV and B photons to regulate plant morphology. Therefore, because of the constant B‐PFD, our lighting treatments were unlikely to have disparate effects on morphology through CRYs and PHOTs.

Sunlight has an FR fraction of around 0.33 (Fig. S1), which slightly increases at twilight and exponentially under a plant canopy with specific patterns varying with canopy shape and density (Casal 2012; Sellaro et al. 2019). Although PHYB and plant morphological responses often appear more sensitive at artificially low FR fractions (i.e., < 0.33) compared to natural‐range FR fractions (i.e., ≥ 0.33) (Figures 3 and 10), the responses to the natural and artificial FR fraction are consistent. The FR fraction of unfiltered sunlight is comparable to the highest FR fraction in our primary experiment (Figure 1 and Table 1), which significantly induced leaf elongation and/or SLA increase in all cultivars tested. The similarity between the FR fraction of sunlight and our experimental conditions indicates that our interpretations of morphological responses and predicted PHYB activity can be applied to plants growing under sunlight. The persistent FR‐mediated PHYB deactivation and SALR under a high R + FR PFD in our study suggest that even though the R + FR PFD of sunlight is high, PHYB‐mediated SALR may already be triggered under sunlight. Thus, our findings imply that FR‐mediated morphological responses occurring under canopy‐filtered sunlight, but not under unfiltered sunlight, may be an oversimplification given the relatively high range of FR fractions present in the natural environment. In addition, our findings imply that wavebands with short wavelengths (e.g., B and UV light) might suppress FR‐mediated SALR through other photoreceptors such as CRYs and UV RESISTANCE LOCUS 8, thereby maintaining plant morphology under sunlight.

5. Conclusions

Contrary to the paradigm that high light intensity diminishes the effects of the FR fraction, the influence of the FR fraction on SALR and predicted PHYB activity persist under a high intensity of R and FR light. The suppressive effects of high light intensity proposed by previous studies are likely attributable to the increased intensity of individual wavebands, particularly the B‐PFD, and not to the total light intensity. The findings of our study not only clarify the effects of light intensity on FR‐mediated morphological responses but also suggest a method of inducing SALR with FR light to elicit desired morphology even under high light conditions. In addition, we applied the three‐state PHYB model to explain morphological responses of light‐grown whole plants, demonstrating its potential use to crops and for applications.

Supporting information

supmat.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.