Six leaf lettuce cultivars were used in this study: Chima-sanchi and Chirimen-chisya (purchased from Tohoku Seed), Red fire and Green wave (purchased from Takii Seed), and Fringe green and Shiki-beni (purchased from Sakata Seed). Chima-sanchi, Red fire, and Shiki-beni are white seed cultivars; Chirimen-chisya, Green wave, and Fringe green are brown seed cultivars . The seeds were stored in a constant humidity chamber (SD-302‐01, Toyo Living) at 25 °C and a relative humidity of 0%‐1% until sowing. Seeds were surface-sterilized via immersion in 70% ethanol for 1 minute. The seeds were treated with 20% commercial bleach (Kao) containing 6% sodium hypochlorite, resulting in a final NaOCl concentration of 1.2%, for 15 minutes. The seeds were then rinsed 3 times with sterile distilled water. The sterilized seeds were placed on a germination medium containing half-strength Murashige and Skoog medium (2.3 g/L, Wako Pure Chemical Industries) [13] supplemented with 10 g/L sucrose and 2.5 g/L Phytagel (Sigma-Aldrich) in Petri dishes with a diameter of 9 cm. The pH of the medium was adjusted to 5.8 with 1N KOH and 1N HCl. The medium was then autoclaved at 120 °C and 0.1 MPa for 20 minutes. Seeds were germinated in an environmentally controlled growth chamber (LPH-411S, NK systems) fitted with fluorescent light (FLR40SW/M/36, NEC) at a photosynthetic photon flux density of 300 μmol photons/m²/s under continuous white light conditions at 20 °C. All experiments were conducted at Takasaki University of Health and Welfare, Takasaki City, Gunma Prefecture, Japan (36.33°N, 139.00°E) between September 2021 and September 2022, in a humid subtropical climate (Köppen climate classification: Cfa).

Shoot regeneration efficiency was examined using a medium supplemented with different basal salt mixtures, sugars, and concentrations of BAP and NAA (Nacalai Tesque) following a previously described method [14] (Table 1). NAA was dissolved in a 10 mM NaOH solution and BAP was dissolved in a 10 mM HCl solution before either was added to the culture media. All media contained 0.5 g/L polyvinylpyrrolidone (Nacalai Tesque) and 2.5 g/L Phytagel (Sigma-Aldrich) at pH 5.8. The medium was sterilized via autoclaving at 121 °C for 20 minutes. Cotyledons from 7-day-old seedlings were used as explants and placed on the medium in Petri dishes with a diameter of 9 cm. Each treatment included 16 explants that were cultured per dish, with 8 dishes per treatment, for a total of 128 explants. The explants were maintained for 4 weeks under continuous white light conditions (photosynthetic photon flux density=300 μmol photons/m²/s) at 25 °C and transferred to fresh medium every 2 weeks.

The color parameters of the seeds from each cultivar were measured with an SD 7000 spectrophotometer (Nippon Denshoku Industries) using the CIELAB L*, a*, and b* color scale. The L* axis represents the degree of brightness ranging from black (L*=0) to white (L*=100). The a* and b* axes represent redness (positive number) to greenness (negative number) and yellowness (positive number) to blueness (negative number), respectively [15].

The total flavonoid content was analyzed according to a previously described method, with some modifications [16]. A 50 µg aliquot of seeds was homogenized in 0.5 mL of 80% methanol with 5.0 mm stainless beads (Biomedical Science) at 1100 rotations per minute for 45 seconds using a Shake Master (Biomedical Science). The homogenized solutions were incubated for 15 minutes at 70 °C, then centrifuged at 10,000 × g for 10 minutes at 4 °C. The resulting supernatants were incubated at 60 °C, and the dried pellets were dissolved in 20 µL of 80% methanol. The extracts were spotted on a 5 × 5 cm TLC Silica gel 60 F₂₅₄ plate (Merck). For staining, the blots were sprayed with a methanolic solution containing 1% diphenylboric acid 2-aminoethylester (DPBA, Tokyo Chemical Industry), then sprayed with a methanolic solution containing 5% PEG 4000 (Nacalai Tesque). The fluorescence was visualized using an iBright CL1000 imaging system (Thermo Fisher Scientific).

All statistical analyses were performed using EZR software [17], a free graphical interface for R that is widely used for standard biostatistical analyses. Significance was determined using a Student t test for two-group comparisons or a one-way ANOVA followed by a Tukey test for multiple group comparisons. The statistical significance was set at P<.01 for Student t tests and P<.05 for one-way ANOVA. All values are expressed as means and SE.

BAP and NAA are commonly used as plant growth regulators for regenerating lettuce shoots [8,10,14,18]. We examined the effects of different concentrations of BAP and NAA on the efficiency of shoot regeneration in Chima-sanchi and Chirimen-chisya cultivars (Figure 1). We used different concentrations of BAP with 0.1 mg/L NAA (M1, M3, and M8). The shoot regeneration efficiency was highest in Chima-sanchi (Figure 1C) and Chirimen-chisya (Figure 1D), with 0.05 mg/L BAP (M3) and 0.5 mg/L BAP (M1), respectively. We then tested different concentrations of NAA with 0.5 mg/L BAP (M1, M5, and M6). The efficiency was highest in Chima-sanchi using 0.1 mg/L NAA (M1) (Figure 1C). The Chirimen-chisya shoot regeneration efficiency did not differ between the 0.1 mg/L (M1) and 1 mg/L (M6) NAA treatments (Figure 1D). The shoots of both cultivars weakly regenerated when treated with 5 mg/L BAP (M7-9; Figure 1). The BAP concentration strongly and cultivar-dependently influenced the shoot regeneration efficiency, whereas the basal salt mixture and sugar composition did not (Multimedia Appendices 1 and 2).

The seed coat color is a key phenotypic trait in many crops such as lettuce [19-21]. The Chima-sanchi seeds were lighter, redder, and yellower than those of Chirimen-chisya (Figure 2). The ratio of shoot regeneration efficiency under 0.05 mg/L BAP to that under 0.5 mg/L BAP (M3/M1) positively correlated with seed brightness (L*) and yellowness (b*) in all 6 cultivars (r=0.834 and 0.722, respectively; Figure 3B and D; Multimedia Appendix 3). The brown-seeded cultivars contained more flavonoids than the white-seeded cultivars (Multimedia Appendix 4). The regeneration efficiency of the white-seeded types was higher than that of the brown-seeded types, indicating that flavonoid accumulation negatively modulates cytokinin responsiveness.

Flavonoids are secondary metabolites that play multiple roles in auxin transport, oxidative stress tolerance, and cell division [22]. In Arabidopsis, transparent testa (tt) mutants are defective in enzymes or regulatory factors involved in the flavonoid biosynthetic pathway. These mutants have a pale seed coat and show altered responses to phytohormones [23]. The Arabidopsis tt4 mutant is deficient in chalcone synthase, a key enzyme in the flavonoid biosynthetic pathway. The Arabidopsis tt4 mutant has a pale yellow seed coat and markedly reduced shoot regeneration efficiency compared with the wild type [24]. This finding suggests that certain downstream flavonoid metabolites produced after chalcone synthase catalysis are required for efficiently regenerating shoots in Arabidopsis. The brown seed cultivars contained higher levels of flavonoids than the white seed cultivars in this study (Multimedia Appendix 4); however, the shoot regeneration efficiencies of the brown seed cultivars were considerably lower (Figure 3). These patterns imply that although flavonoid metabolism may influence regenerative competence, the mechanisms underlying this metabolism differ between lettuce and Arabidopsis. However, the strong correlation we observed between seed coat color and the optimal BAP concentration provides a practical morphological marker for predicting the cytokinin requirements of leaf lettuce.

Seed coat color is a simple morphological marker for predicting the most appropriate cytokinin concentration required for efficiently regenerating shoots regardless of lettuce cultivar. This approach can be quickly and easily used for optimizing regeneration conditions, thereby increasing the efficiency of transgenic and genome-edited lettuce production. However, this study was limited to 6 leaf lettuce cultivars, which were tested under controlled laboratory conditions. Further validation with additional genotypes and under different environmental conditions is necessary. This marker-based method is more cost- and labor-efficient for laboratories; however, scaling up tissue culture systems for industrial use remains challenging because maintaining aseptic conditions, controlling the environment, and performing manual subculture steps substantially increases labor and energy costs. Further technical improvements and validation studies are required before genotype-specific protocols can be applied to large-scale propagation or transformation systems. However, integrating visible traits, such as seed coat color, into regeneration and transformation workflows could help balance the overall costs and benefits of cultivar-specific plant biotechnologies and support the broader use of lettuce as a bioproduction platform.

Seed coat color is strongly correlated with the cytokinin required for efficiently regenerating shoots in leaf lettuce. The maximum shoot regeneration occurred with 0.05 and 0.5 mg/L BAP for white and brown seed cultivars, respectively. Pigmentation-related metabolites, possibly flavonoids, may modulate cytokinin responsiveness during the regeneration process. This simple visual marker can accelerate the optimization of regeneration systems, reduce experimental costs, and facilitate cultivar-specific transformation. Our findings directly support the development of lettuce cultivars with specific traits, such as enhanced yield, stress tolerance, and nutritional value, by enabling the rapid establishment of efficient regeneration and transformation protocols. Additionally, understanding the interplay between flavonoid metabolism and cytokinin signaling may provide new options for engineering regenerative competence in other horticultural crops. Our findings not only clarify the physiological basis of genotype-dependent regeneration but also provide a practical framework that connects the fundamental knowledge of plant regeneration mechanisms to their application in lettuce breeding and commercial production.