Introduction

Lilium, commonly known as lilies, possesses significant horticultural and commercial value due to its aesthetic appeal and utility as an ornamental plant (Fatihah and Nudin 2018;Kang et al. 2024). The genus Lilium also holds ecological significance, particularly in its native habitats, where species contribute to regional biodiversity and act as key genetic resources for future breeding programs. Lilium miquelianum, known for its ecological significance and vibrant orange flowers, is indigenous to regions of East Asia, including Korea (Lee et al. 2016;Park et al. 2014). This study utilized bulb scales collected from populations of L. miquelianum endemic to Jeju Island, Korea. Jeju Island, characterized by its volcanic soil and distinct climatic conditions (Lee et al. 2024), is a biodiversity hotspot for native Lilium species. The endemic populations of L. miquelianum in Jeju Island not only contribute to the region’s ecological biodiversity but also represent a unique genetic reservoir for breeding programs. The volcanic soil and climate of Jeju Island may have driven specific adaptations, making L. miquelianum a valuable model for studying resilience in changing environmental conditions.

Traditional propagation methods, such as bulb division (or partitioning) (Knippels 2000), are often limited by slow multiplication rates and vulnerability to environmental stressors. This is particularly challenging for L. miquelianum populations, which are at risk of habitat loss and degradation due to climate change and human activities. As a result, in vitro tissue culture techniques have emerged as a viable alternative, enabling rapid multiplication and year-round production of high-quality plants (Chinestra et al. 2015;Kundu et al. 2024;Niimi 1986). Among these techniques, the cultivation of bulb scales in controlled media offers a promising approach for mass propagation due to its efficiency in producing a high number of propagules within a relatively small space (Dhyani et al. 2014;Liu 2012;Varshney et al. 2000). The choice of culture medium is a critical determinant of success in in vitro propagation. The Murashige and Skoog (MS) medium (Murashige and Skoog 1962), widely employed in plant tissue culture, provides an optimal balance of macronutrients, micronutrients, and growth regulators to promote cell division, organogenesis, and growth. Modifications to this medium, such as altering nutrient concentrations or adding specific supplements, have been shown to significantly influence developmental outcomes, including callus formation, shoot induction, and root elongation, in Lilium species when bulb scales are cultured (Ali et al. 2024;Huong et al. 2017;Song et al. 2021;Takayama and Misawa 1979).

Furthermore, this research aims to support the conservation of L. miquelianum and its potential use as a genetic resource in ornamental breeding programs by testing the effects of different media compositions on bulb scale development. This study evaluates two key growth parameters: the surface area of bulb scales and the number of newly formed scales. These parameters serve as indicators of growth performance and regenerative potential. Through periodic observations and quantitative measurements, the research investigates temporal growth dynamics across media types to identify the formulation most conducive to rapid and uniform development. The findings provide insights into optimizing in vitro propagation protocols for Lilium, contributing to enhanced efficiency in both commercial production and conservation efforts.

Materials and Methods

In vitro Plant Materials

Bulb scales of Lilium miquelianum were collected from naturally occurring populations in various regions of Jeju Island, Korea. To ensure sterility and prevent contamination, scales were first washed under running tap water for 10 min to remove surface debris. The cleaned scales were then submerged in 1% sodium hypochlorite solution with 1 - 2 drops of Tween-20 as a surfactant and agitated gently for 10 min. Then scales were rinsed thoroughly 3 - 4 times with distilled water. The bulb scales within similar size range (15 - 20 mm in length) were collected and carefully prepared by removing the roots and the outermost scales to ensure uniformity. Each scale was then cultured in a 100 mm × 40 mm culture dish (SPL, Korea), with five scales placed radially per vessel. A total of 120 - 125 scales were cultured in four different types of media in in vitro facility for the duration of the experiment.

Four different media formulations were tested in this study, all with a base composition of 60 g·L^-1 sucrose and 4 g·L^-1 Gelrite, adjusted to pH 5.8 before autoclaving. The FMS medium (Full MS), commonly used for Lilium propagation, was used as the control. The experimental media were FMS (Full MS): Murashige and Skoog Medium Including Vitamins (MS) at 4.4 g·L^-1, HMS (Half MS): Murashige and Skoog Medium Including Vitamins at 2.2 g·L^-1, AV (Anderson Vitamin): Anderson’s Rhododendron Medium at 1.8 g·L^-1 with Murashige and Skoog Vitamin Mixture at 0.1 g·L^-1, and KV (Knudson Vitamin): Knudson C Orchid Medium at 1.9 g·L^-1 with Murashige and Skoog Vitamin Mixture at 0.1 g·L^-1. The FMS medium served as the control to compare the effects of reduced macronutrients and alternative formulations. The prepared media were distributed into 90 mm Petri dishes, each containing 25 mL of medium. Anderson and Knudson media were included alongside Full MS and Half MS media to evaluate the response of L. miquelianum bulb scales to a broad range of nutrient formulations. Although Anderson and Knudson media were originally designed for rhododendrons and orchids, respectively, their distinct compositions provided an opportunity to explore whether alternative nutrient profiles might yield unique developmental outcomes in Lilium species. This comparative approach was aimed at broadening the understanding of media-specific influences on growth performance.

Statistical Analysis

Throughout the experimental period (0 – 12 weeks), any bulb scales exhibiting severe contamination or necrosis were excluded from data collection to ensure the reliability of the results. Only healthy, contamination-free samples that showed normal development were included in the final dataset for analysis. Quantitative measurements of bulb area and scale count were obtained at 3-week intervals (Weeks 0, 3, 6, 9, and 12). Data collected at Week 12 were analyzed using one-way analysis of variance (ANOVA) to evaluate the effect of media type on growth parameters. To identify specific pairwise differences between media, Tukey’s Honest Significant Difference (HSD) test was conducted with a 95% confidence interval. Statistical analyses were performed using SAS software (SAS Institute Inc., ver. 9.2). All results are presented as the mean ± standard error (SE).

Normalization and Relative Growth Calculation

To account for initial size variation among bulb scales, all growth parameters were normalized to Week 0 values for each individual scale. Relative growth rates were calculated as follows:

This normalization ensured a standardized comparison of growth trends across media types.

Visualization and Correlation Analysis and Data integrity

Data visualization was conducted utilizing matplotlib and seaborn libraries in Python to generate line graphs, heatmaps, and scatter plots. The correlation between bulb area and scale count was evaluated using Pearson correlation coefficients (r) and coefficients of determination (R²). Regression lines were fitted to scatter plots to elucidate trends both across all data and within individual media types. All outliers, such as visibly deformed bulbs, were excluded based on predefined morphological criteria to ensure data reliability. Ethical guidelines were adhered to in data handling and analysis to prevent overrepresentation or underrepresentation of results.

Results and Discussion

Growth Dynamics Across Media Over Time

The representative bulb development of Lilium miquelianum under different media conditions is shown in Fig. 1. Images were taken at Week 0, 3, 6, 9, and 12 for each sample. This visual evidence complements the quantitative measurements presented in Fig. 2 to 4, providing a direct comparison of the growth trends across media over time. FMS demonstrated superior growth performance, yielding the largest bulb sizes and highest scale counts by Week 12 (Fig. 2 and 3).

The relative growth rates of bulb area and scale count showed significant variation among the four media types over the 12-week period (Fig. 2). A detailed breakdown of week-to-week growth rates across media is provided in Table 1, highlighting the steady growth of FMS across all weeks and the strong early performance of HMS, which plateaued after Week 6. FMS consistently exhibited the highest growth rates, particularly after Week 6, while KV demonstrated the lowest performance throughout the experiment. Heatmap analysis further corroborated these trends (Fig. 3), revealing that FMS achieved the highest weekly growth rates in most time points, with HMS showing comparable performance during early growth stages. These results emphasize the critical role of nutrient composition and concentration in determining the growth potential of L. miquelianum under in vitro conditions. FMS supports robust and sustained growth due to its balanced nutrient composition, providing essential macro- and micronutrients for long-term metabolic activity. This aligns with previous studies demonstrating the efficacy of full-strength MS medium for bulbous plants (Furmanowa and Olędzka 2014;Kumar et al. 2016;Ramsay et al. 2003).

Conversely, KV exhibited limited growth, with bulbs remaining small and scales sparse throughout the experiment. However, its moderate correlation between bulb size and scale count (R² = 0.51) suggests a potential for more balanced growth compared to other low-performing media, such as AV and HMS, under specific conditions. These outcomes may reflect the partial suitability of KV for supporting proportional development (Goh 1983). The limited nitrogen and potassium content in Knudson C medium, originally designed for orchids, contrasts with the high nutrient demands of bulbous plants like Lilium. For instance, MS medium provides significantly higher levels of these macronutrients, making it better suited for robust growth in species with high metabolic activity (Knudson 1946; Marschner 2012;Murashige and Skoog 1962).

Similarly, HMS demonstrated strong initial growth, likely due to its reduced nutrient concentration, which may have facilitated early cell division and tissue establishment. However, its performance plateaued after Week 6, suggesting that its nutrient composition supports rapid initial growth but is insufficient for sustained metabolic activity. This characteristic makes HMS a potential candidate for short-term culture systems aimed at rapid establishment or initial multiplication of L. miquelianum, where prolonged growth is not the primary goal (Berhe et al. 2023;Chukwujekwu and Van Staden 2003). AV, on the other hand, displayed moderate but stable growth throughout the study, with less variability across replicates compared to other media. This stability suggests AV may be a reliable choice for propagation systems where uniformity and consistency are critical, such as under resource-limited conditions or in experimental setups requiring controlled variability. Its performance, though lower than FMS, could also make it suitable for stress-testing scenarios or preliminary trials aimed at optimizing growth conditions for L. miquelianum.

These findings collectively underscore the significance of aligning nutrient composition with the specific developmental requirements of L. miquelianum. While KV did not achieve the high growth rates of FMS, its moderate correlation (R² = 0.51) highlights its potential for specific propagation systems requiring balanced growth under less resource-intensive conditions. While FMS is optimal for achieving robust growth over extended periods, HMS may serve as a viable alternative for short-term culture systems aimed at rapid establishment. The consistent performance of AV further elucidates the potential for alternative media formulations tailored for specific propagation objectives.

Variability in Growth Across Media and Samples

The growth intensity patterns of bulb area and scale count across different media and time periods were illustrated through heatmaps (Fig. 3), revealing the cumulative impact of media on growth. Among the tested media, FMS consistently demonstrated the most significant growth trends over time, while KV underperformed across all metrics. HMS exhibited moderate growth, excelling during the early weeks but plateauing as time progressed. AV, while not achieving the highest growth rates, showed stable and uniform growth across replicates, highlighting its reliability under controlled conditions.

Interestingly, the relative growth rates of individual samples (Fig. 4) highlighted significant variations even within the same medium, with some HMS samples surpassing their FMS counterparts during certain intervals. This variability suggests that individual genetic or physiological factors play a critical role in determining growth outcomes, even under identical environmental conditions. While nutrient availability drives overall trends, the observed differences likely stem from genetic diversity inherent to L. miquelianum. Such genetic diversity has been reported in other crops, including lilies, where individual plants exhibit varying responses to identical growing conditions (A et al. 2012;Tabassum et al. 2022;Wang et al. 2016). By Week 12, FMS had clearly outperformed all other media, yielding the largest average bulb size and highest scale count (Fig. 5 and Table 2, 3). Statistical analyses further validated the superior performance of FMS over other media. ANOVA results confirmed significant differences among media types for bulb area development at Week 12 (p = 2.43e-05). To identify specific pairwise differences between media, Tukey’s Honest Significant Difference (HSD) test was conducted. The results showed significant differences between FMS and AV (p = 0.0002) and between FMS and HMS (p = 0.0001), while no significant differences were observed between FMS and KV (p = 0.1502) or between AV and HMS (p = 0.1079; Table 4).

Correlation analysis further illustrated the relationship between bulb size and scale count (r = 0.62, R² = 0.39; Fig. 6 and Table 3). Detailed correlation values for each medium are provided in Table 3, with FMS exhibiting the strongest relationship (R² = 0.61), indicating that its nutrient composition supports balanced growth across these two parameters. In contrast, HMS and AV showed weaker correlations (R² = 0.23 and R² = 0.30, respectively), suggesting that these media influence bulb area and scale count independently rather than proportionally. KV exhibited a moderate correlation (R² = 0.51), suggesting a more balanced relationship compared to HMS and AV, though still less robust than FMS.

These results underscore the critical importance of selecting appropriate media to ensure consistent and efficient propagation outcomes. While FMS offers optimal conditions for balanced growth, KV provides a moderately balanced relationship between bulb size and scale count, making it a potential alternative for propagation scenarios requiring a compromise between growth efficiency and stability. The variability observed in HMS and AV highlights their potential for specific applications, such as short-term propagation or studies requiring controlled variability. This variability also emphasizes the necessity of replication and robust statistical analysis in plant propagation studies, particularly when working with genetically diverse species like L. miquelianum.

Conclusion

This research highlights the importance of media composition in cultivating Lilium miquelianum. Out of the four media evaluated, full-strength MS medium (FMS) consistently yielded the best results in terms of bulb area and scale count, surpassing other media throughout the study period. The interconnection between these parameters was further emphasized by heatmap and correlation analyses, with FMS exhibiting the strongest link between bulb size and scale quantity. KV medium, while originally designed for orchids, demonstrated a moderately strong correlation between these parameters, suggesting its potential as an alternative for propagation scenarios requiring a balance between growth efficiency and stability. Conversely, HMS and AV showed weaker correlations, highlighting their variability and limited suitability for long-term propagation. These findings indicate that FMS is ideal for extended propagation, likely due to its well-balanced nutrient profile suited for bulbous plants. This investigation offers crucial insights for optimizing L. miquelianum propagation conditions, with potential applications in commercial and conservation efforts. Subsequent studies should investigate additional variables such as lighting conditions, hormone treatments, and genetic factors to further improve propagation outcomes and broaden the applicability of these findings.