In Vitro Propagation of Easter Island Curcuma longa from Rhizome Explants Using Temporary Immersion System

: Curcuma longa ( C. longa ) is widely known for its medicinal properties. However, the potential overexploitation of this plant raises doubts about its long-term survival on Rapa Nui. Micropropagation using a temporary immersion system (TIS) could be the basis for developing a cost-effective and highly productive method of large-scale cultivation of this plant. Our objective was to develop and reﬁne the in vitro multiplication system for mass propagation of C. longa, and thus help restore the fragile ecosystem of Rapa Nui. Three parameters were evaluated: number of explants per ﬂask, ﬂask capacity, and LEDs spectrum. For each parameter evaluated, four aspects were analyzed: fresh weight per plant, number of shoots, percentage of non-sprouting explants, and the proliferation rate. The use of 30 explants per two-liter ﬂask results in more plants with high fresh biomass than other conﬁgurations. In addition, LEDs with a red:blue ratio of 2:1 provided the best lighting conditions for in vitro propagation and positively affected C. longa proliferation and rooting. Therefore, our results show that 30 explants per two-liter ﬂask and an LED source with a red:blue ratio of 2:1 allow a higher number of C. longa plants to be obtained using TIS.


Introduction
Easter Island or Rapa Nui is one of the most isolated inhabited places in the world. Ancestral medicine is very important to the people of Rapa Nui, who rely on several plants to make preparations widely used for medical purposes [1]. However, due to excessive harvesting, several plant species have disappeared and others, among them, Curcuma longa L., (Pua in the Rapa Nui language), are overexploited. The natives consider it a magical plant and use it to treat diseases such as cancer, diabetes, muscle pain, and others [2]. The widespread use of this species has put it in danger of extinction, and it is now under state protection [3]. A study of the biological properties of the Rapa Nui ecotype of C. longa revealed high relative concentrations of polyphenols and curcumin, highlighting its antioxidant and anti-inflammatory properties [3]. The results demonstrate the importance of conservation, reproduction, protection, and proper management of this species. Particular attention should be given to efforts aimed at optimizing its propagation on Rapa Nui through in vitro multiplication.
The rhizome of C. longa is one of the plant materials from which adult plants can be produced by in vitro micropropagation, preserving the characteristics of the mother plant [4]. The micropropagation method that we used in this study is known as the Temporary Immersion System (TIS) and involves intermittent contact between liquid culture medium and explants [5,6]. In addition, this system improves the proliferation of shoots in a liquid medium with a temporary immersion of the explant in a liquid medium for a specific interval, thus providing the medium at one interval and oxygenation during the other [7]. However, a few studies about the development of a method of immersion using fed-batch liquid culture as bioreactors to micropropagation of C. longa were analyzed the response to minerals in the culture medium [8]. In TIS, the nutrient medium is pneumatically propelled to reach and immerse the explants while air is pumped to completely refresh the explants headspace with each immersion, renewing oxygen and avoiding CO 2 and ethylene accumulation, as occurs in semi-solid and solid media, where it can negatively affect plant morphogenesis [9]. Short immersion times allow good control of water content to reduce the possibility of plant vitrification [10]. In addition, this method allows for adequate nutrient mixing, relatively low frequency of subculturing, easy medium replacement, and minimal damage to explants. For these reasons, several authors have described TIS as the most appropriate approach to tissue culture [11,12].
TIS is a complex system influenced by several factors such as light, temperature, and medium composition [12]. Explant density is also important for the multiplication and growth of plants within the flask. The higher the density of explants per flask, the more nutrients and space needed for plant movement and proper development. An excessive number of explants in a limited space can reduce the proliferation rate and the quality of the plants produced [13,14]. Another parameter that affects plant growth is the light source. A common light source used for in vitro propagation is fluorescent lamps. However, they have the problem of high power consumption and produce a wide range of wavelengths (350-750 nm) that are not necessary for plant development [15]. Light-emitting diodes (LEDs) have recently become a viable alternative for commercial micropropagation. LEDs have several advantages: low thermal radiation, monochromatic spectrum, high durability, and low power consumption. LED lighting systems for in vitro culture provide light in the spectral range involved in photosynthesis and photomorphogenic responses in plants [16]. In the case of C. longa, LED light has been reported to affect the production of metabolites such as curcumin and its derivatives, thus increasing the antioxidant capacity of the plant [17]. However, the effect of this type of light on the proliferation and rooting of C. longa in vitro has not been reported so far.
TIS micropropagation of various plants species is an alternative method that offers advantages over mini-cutting, such as: higher multiplication rate, higher rooting rate; plant health, effective control of pests and diseases, environmental control, and reduced need for physical space [18]. Several TIS setups have been successfully used for micropropagation of various plant species of agricultural, medicinal and conservational value in recent years. In this study, we used the Twin-Flask system (BIT ® ) described by Escalona [19]. This system consists of a container for growing plants and a reservoir for the liquid medium. When the solenoid valve is opened, and the air compressor is turned on, the airflow pushes the medium into the flask with the plants and the plants are immersed. This process is reversed when another solenoid valve is opened, and the medium is returned to the original reservoir by air pressure. The system is connected to an automated irrigation mechanism, which controls the frequency and duration of each immersion [19,20]. The operating parameters of the TIS are adjusted for each species to be propagated in vitro in order to achieve optimal results.
In this study, we successfully propagated C. longa in vitro by TIS and evaluated the effects of three factors: number of explants per flask, flask capacity (volume), and red:blue ratio of LEDs. The ultimate goal is to establish and refine an in vitro propagation system with reasonable production costs that can be used for the mass propagation of C. longa to restore the compromised Rapa Nui ecosystem.

Explant Collection, Disinfection and Establishment of Rhizome Cultures
Rhizomes of C. longa L were harvested from the Mataveri Otai nursery on Easter Island (geographical coordinates: 27 • 09 50, 17 S-109 • 26 24, 63 O). The excised long creeping rhizomes were thoroughly washed with water and Tween-20 for 10 min. The rhizomes were then cut into 2-3 cm pieces and placed under water flow for 30 min. Subsequently, the rhizomes were immersed in the Phyton solution (3 mL/L) under gentle agitation for 15 min. Next, the rhizomes were cut into 1 cm pieces and active buds were separated and washed with 1.5% sodium hypochlorite solution, 0.5 g/L citric acid, and 0.5 g/L ascorbic acid for 15 min under constant agitation. The pieces were transferred to a laminar flow hood and washed four times with a sterile solution of 400 mg/L citric acid and 400 mg/L ascorbic acid. Finally, the pieces were placed in a Murashige and Skoog [21] (MS) culture medium supplemented with 0.1 mg/L thiamine, 200 mg/L glycine, 170 mg/L NaH 2 PO 4 , 2.5 mL/L plant preservative mixture (PPM), agar 6.5 g/L, sucrose 30 g/L, at pH 5.8. The flasks were incubated in vitro in the dark at 25 • C for 10 days and then placed under 1R:1B LEDs for one month.

In Vitro Growth of Explants
The shoot rhizomes were transferred to MS growth medium, supplemented with 0.1 mg/L thiamine, 200 mg/L glycine, 170 mg/L NaH 2 PO 4 , 1 mg/L 1-Naphthaleneacetic acid (NAA), 2 mg/L N 6 -Benzylaminopurine (BAP), 6.5 g/L agar, 30 g/L sucrose, at pH 5.8. The plant material was subcultured three times (30 days each time) to obtain more explants for TIS. The initial explants for TIS were 1-2 cm long, rootless and showed no symptoms of vitrification (hyperhydricity), or oxidation ( Figure 1).
Rhizomes of C. longa L were harvested from the Mataveri Otai nursery on E land (geographical coordinates: 27°09′50, 17′S-109°26′24,63O). The excised long c rhizomes were thoroughly washed with water and Tween-20 for 10 min. The rh were then cut into 2-3 cm pieces and placed under water flow for 30 min. Subse the rhizomes were immersed in the Phyton solution (3 mL/L) under gentle agita 15 min. Next, the rhizomes were cut into 1 cm pieces and active buds were separa washed with 1.5% sodium hypochlorite solution, 0.5 g/L citric acid, and 0.5 g/L acid for 15 min under constant agitation. The pieces were transferred to a lamin hood and washed four times with a sterile solution of 400 mg/L citric acid and 4 ascorbic acid. Finally, the pieces were placed in a Murashige and Skoog [21] (MS) medium supplemented with 0.1 mg/L thiamine, 200 mg/L glycine, 170 mg/L NaH mL/L plant preservative mixture (PPM), agar 6.5 g/L, sucrose 30 gr/L, at pH 5.8. Th were incubated in vitro in the dark at 25 °C for 10 days and then placed under 1R:1 for one month.

In Vitro Growth of Explants
The shoot rhizomes were transferred to MS growth medium, supplemented mg/L thiamine, 200 mg/L glycine, 170 mg/L NaH2PO4, 1 mg/L 1-Naphthaleneac (NAA), 2 mg/L N 6 -Benzylaminopurine (BAP), 6.5 g/L agar, 30 g/L sucrose, at pH plant material was subcultured three times (30 days each time) to obtain more e for TIS. The initial explants for TIS were 1-2 cm long, rootless and showed no sy of vitrification (hyperhydricity), or oxidation ( Figure 1).

In Vitro Proliferation of C. Longa by TIS
TIS was used for the proliferation and rooting of C. longa shoots. Ten flasks set of conditions were prepared and three independent assays were performed were maintained under the following conditions during proliferation and rooting otherwise noted): 15 mL of culture medium (proliferation or rooting) per explant, sion of 4 min each time in a 4 h cycle, the temperature of 23  1 °C under for 16 h/d white LED illumination. For proliferation, MS culture medium was used supple with 200 mg/L glycine, 0.1 mg/L thiamine, 0.5 mg/L nicotinic acid, 0.5 mg/L pyr 100 mg/L myoinositol, 170 mg/L NaH2PO4, BAP 3 mg/L, 30 g/L sucrose, at pH 5.8 root stage, the same medium was used and BAP was replaced with 1 mg/L NA the proliferation stage, the flask of proliferation medium was replaced with the rooting medium under a laminar flow hood, while the explants were not distu handling. After both stages, all shoots were placed in a greenhouse to acclimate.

Evaluation of the Number of Explants per Flask
We evaluated 20, 30, and 40 explants of C. longa per culture flask with a capa L. The cultures were grown for 90 days (60 days for proliferation and 30 days for r

In Vitro Proliferation of C. Longa by TIS
TIS was used for the proliferation and rooting of C. longa shoots. Ten flasks for each set of conditions were prepared and three independent assays were performed. Flasks were maintained under the following conditions during proliferation and rooting (unless otherwise noted): 15 mL of culture medium (proliferation or rooting) per explant, immersion of 4 min each time in a 4 h cycle, the temperature of 23 ± 1 • C under for 16 h/day, and white LED illumination. For proliferation, MS culture medium was used supplemented with 200 mg/L glycine, 0.1 mg/L thiamine, 0.5 mg/L nicotinic acid, 0.5 mg/L pyridoxine, 100 mg/L myoinositol, 170 mg/L NaH 2 PO 4 , BAP 3 mg/L, 30 g/L sucrose, at pH 5.8. For the root stage, the same medium was used and BAP was replaced with 1 mg/L NAA. After the proliferation stage, the flask of proliferation medium was replaced with the flask of rooting medium under a laminar flow hood, while the explants were not disturbed by handling. After both stages, all shoots were placed in a greenhouse to acclimate.

Evaluation of the Number of Explants per Flask
We evaluated 20, 30, and 40 explants of C. longa per culture flask with a capacity of 1 L. The cultures were grown for 90 days (60 days for proliferation and 30 days for rooting), at a temperature of 23 • C, under 16 h of light per day, and white LED illumination. Subsequently, the resulting plants were acclimated in a greenhouse. Four parameters were measured prior to acclimation: fresh plant weight (total weight of shoots/number of shoots), number of shoots, percentage of non-sprouting explants (explants that did not germinate), and proliferation rate (number of shoots/number of initial explants).

Evaluation of Flask Capacity
Two different volumes of culture flasks were evaluated: 1 L and 2 L. We chose 30 explants per flask as this quantity produced a greater number of shoots. The cultures were grown for 60 days (45 days for proliferation and 15 days for rooting), at a temperature of 23 • C under 16 h of light per day and white LED illumination. Subsequently, the resulting plants were acclimated in a greenhouse. Four parameters were measured prior to acclimation: fresh plant weight (total weight of shoots/number of shoots), number of shoots, percentage of non-sprouting explants (explants that did not germinate), and proliferation rate (number of shoots/number of initial explants).

Evaluation of Light Spectra
Considering the available data from previous research, we used 30 explants in 2 L flasks for this study. Three different sources were used for light treatments: white LED, red:blue LED 1 (RB1, model LX-GLT8), and red:blue LED 2 (RB2, PARALED system PAR-ALED_GRB_20W, Ciencia Pura SpA. Santiago, Chile). Each light source had different proportions of Blue (420-460 nm) and Red (660 nm). These wavelengths are consistent with the known maximum absorptions for chlorophyll an and b, indicating optimal photosynthetic efficiency [22]. The culture was illuminated from the top of the culture vessel and cultured in a 16 h photoperiod (long day). The experimental conditions and parameters were the same as in the previous experiment. In addition, Yield Photon Flux Density (YPFD, µmol m −2 s −1 ), Red:Blue ratio, and Daily Light Integral (DLI) (mol m −2 ) were measured using a compact spectrophotometer (Lighting Passport Standard Pro model, Allied Scientific Pro TM , Gatineau, QC, Canada). To avoid light contamination, all assays were separated from each other by white screens made by a local manufacturer.

Statistical Analysis
Comparisons were made using one-way ANOVA with Tukey multiple comparison tests or Mann-Whitney test analysis. A value of p < 0.05 was considered as statistically significant difference. All statistical analyses were calculated using GraphPad Prism 9 computer software (GraphPad Software, San Diego, CA, USA).

Evaluation of the Number of Explants per Flask
The number of C. longa explants per flask was evaluated. After 90 days of incubation, plants were removed and fresh weight, the number of shoots, percentage of non-sprouting explants, and proliferation rate were determined. The results are shown in Figure 2, where the assay with 30 explants yielded an average fresh weight of 13.8 g/plant, which was higher than the other treatments ( Figure 2a). The same number of explants also yielded the highest number of shoots, with green leaves and roots, averaging 135.8 plants per flask (Figure 2b). In the assay using 20 and 30 explants, the percentage of non-sprouting explants was comparable. However, in the assay with 40 explants, this parameter increased twofold (Figure 2c). Figure 2d shows the proliferation rate, which decreased as the number of explants used per flask increased. Figure 3 is a photographic record of the plants produced. These are representative of the assay. At the end of proliferation and rooting, all shoots had green leaves and roots (Figure 3a), with an average size of 10 cm (Figure 3b). The non-sprouting explants had no green leaves and oxidation and vitrification were observed (Figure 3c). Subsequently, all the obtained shoots were successfully acclimated in a temperature-and humidity-controlled greenhouse (Figure 3d).   (Figure 3b). The non-sprouting explants had no green leaves and oxidation and vitrification were observed (Figure 3c). Subsequently, all the obtained shoots were successfully acclimated in a temperature-and humidity-controlled greenhouse (Figure 3d). Average ± SD was plotted for all replicates analyzed. Tukey's multiple comparison tests were applied at the 95% confidence level to compare the differences. Bars with different letters show significant differences at p < 0.05 for each parameter studied. Figure 3 is a photographic record of the plants produced. These are representative of the assay. At the end of proliferation and rooting, all shoots had green leaves and roots (Figure 3a), with an average size of 10 cm (Figure 3b). The non-sprouting explants had no green leaves and oxidation and vitrification were observed (Figure 3c). Subsequently, all the obtained shoots were successfully acclimated in a temperature-and humidity-controlled greenhouse (Figure 3d). In all the tests carried out, it was observed that in the middle of the proliferation phase (day 30 of 60) and the rooting phase (day 15 of 30), the plants reached maximum Agronomy 2021, 11, 2121 6 of 11 development, so in the following experiments, the incubation time for both phases were reduced by half.

Evaluation of the Volumetric Capacity of the Flasks
In the second assay, we evaluated the effect of flask volumetric capacity on plant growth. The results showed that the volumetric capacity of the flask affected the proliferation and rooting of C. longa in vitro (Figure 4). In the 2 L flasks, the average fresh weight of plants was 5.7 g/plant, which was greater than the value in the 1 L flasks (4.7 g/plant) (Figure 4a). The system produced an average of 127 plants in the 2 L flask, and 117 plants in the 1 L flask (Figure 4b). No significant differences were observed in the percentage of non-sprouting explants during the experiment (Figure 4c). For the proliferation rate, we obtained values of 4.2 for the 2 L flask and 3.9 for the 1 L flask (Figure 4d). Thus, it was concluded that the use of a 2 L flask would produce a greater number of plants with a higher fresh weight than the use of a 1 L flask. In all the tests carried out, it was observed that in the middle of the proliferation phase (day 30 of 60) and the rooting phase (day 15 of 30), the plants reached maximum development, so in the following experiments, the incubation time for both phases were reduced by half.

Evaluation of the Volumetric Capacity of the Flasks
In the second assay, we evaluated the effect of flask volumetric capacity on plant growth. The results showed that the volumetric capacity of the flask affected the proliferation and rooting of C. longa in vitro (Figure 4). In the 2 L flasks, the average fresh weight of plants was 5.7 g/plant, which was greater than the value in the 1 L flasks (4.7 g/plant) (Figure 4a). The system produced an average of 127 plants in the 2 L flask, and 117 plants in the 1 L flask (Figure 4b). No significant differences were observed in the percentage of non-sprouting explants during the experiment (Figure 4c). For the proliferation rate, we obtained values of 4.2 for the 2 L flask and 3.9 for the 1 L flask (Figure 4d). Thus, it was concluded that the use of a 2 L flask would produce a greater number of plants with a higher fresh weight than the use of a 1 L flask.

Evaluation of Light Spectra
To evaluate the last parameter, we used three different LED light sources: white LED, red:blue LED 1 (RB1) and red:blue LED 2 (RB2). The plants obtained were extracted for analysis and acclimated in a greenhouse. The results, summarized in Figure 5, show that the red:blue LEDs had a positive effect on the in vitro proliferation and rooting of C. longa. When RB1 was used, an average fresh weight of 6.05 g/plant was obtained, which is higher than that of white LED and RB2 (5.14 and 4.53 g/plant, respectively) (Figure 5a). In most

Evaluation of Light Spectra
To evaluate the last parameter, we used three different LED light sources: white LED, red:blue LED 1 (RB1) and red:blue LED 2 (RB2). The plants obtained were extracted for analysis and acclimated in a greenhouse. The results, summarized in Figure 5, show that the red:blue LEDs had a positive effect on the in vitro proliferation and rooting of C. longa. When RB1 was used, an average fresh weight of 6.05 g/plant was obtained, which is higher than that of white LED and RB2 (5.14 and 4.53 g/plant, respectively) ( Figure 5a). In most of the assays performed under any sets of conditions, the percentage of non-sprouting explants at the end of the experiment was minimal or zero. However, in terms of productivity, the RB2 spectrum was the most productive, yielding 154 plants (Figure 5b). Similarly, RB2 gave a proliferation rate of 5.12 plants per explant, while white LED and RB1, produced 4.4 and 4.3 plants per explant, respectively. of the assays performed under any sets of conditions, the percentage of non-sprouting explants at the end of the experiment was minimal or zero. However, in terms of productivity, the RB2 spectrum was the most productive, yielding 154 plants (Figure 5b). Similarly, RB2 gave a proliferation rate of 5.12 plants per explant, while white LED and RB1, For each LED light source analyzed, the Yield Photon Flux Density (YPFD) for red and blue photons, the ratio of red:blue photons, and the Daily Light Integral (DLI) were determined. The results are summarized in Table 1. The white LED had the lowest photosynthetically active radiation (PAR) output performance compared to other light sources; RB1 had ten times more red photons and two times more blue photons than the white LED light source. As for the RB2 ratio, the difference in red photons is similar to the ratio of RB1. However, the number of blue photons is two and five times higher than that of the RB1 and the white LED, respectively. The other parameter analyzed, DLI, shows a similar pattern to that of YPFD. The DLI represents the number of photosynthetically active radiations (PARs) delivered to a specific area over a 24 h period. This variable is particularly useful for describing the light environment of plants; RB2 is the light source with the most photons per area (high DLI). After the proliferation and rooting period, the plants were acclimated in a greenhouse. The plants exhibited green leaves, abundant rootlets and no vitrification was observed ( Figure 6a). After four months of greenhouse cultivation, the plants were about 20 cm in size, had developed the characteristic orange-colored rhizomes (Figure 6b,c), and were ready for shipment back to Rapa Nui. For each LED light source analyzed, the Yield Photon Flux Density (YPFD) for red and blue photons, the ratio of red:blue photons, and the Daily Light Integral (DLI) were determined. The results are summarized in Table 1. The white LED had the lowest photosynthetically active radiation (PAR) output performance compared to other light sources; RB1 had ten times more red photons and two times more blue photons than the white LED light source. As for the RB2 ratio, the difference in red photons is similar to the ratio of RB1. However, the number of blue photons is two and five times higher than that of the RB1 and the white LED, respectively. The other parameter analyzed, DLI, shows a similar pattern to that of YPFD. The DLI represents the number of photosynthetically active radiations (PARs) delivered to a specific area over a 24 h period. This variable is particularly useful for describing the light environment of plants; RB2 is the light source with the most photons per area (high DLI).
After the proliferation and rooting period, the plants were acclimated in a greenhouse. The plants exhibited green leaves, abundant rootlets and no vitrification was observed (Figure 6a). After four months of greenhouse cultivation, the plants were about 20 cm in size, had developed the characteristic orange-colored rhizomes (Figure 6b,c), and were ready for shipment back to Rapa Nui.

Discussion
The success of micropropagation by TIS depends on the volume of the culture vessel and liquid medium container compared to the amount of explant biomass in the early and

Discussion
The success of micropropagation by TIS depends on the volume of the culture vessel and liquid medium container compared to the amount of explant biomass in the early and late stages of cultures, the immersion regime to which the culture is exposed, the type and composition of light, and the effect of each of these parameters on the others [12]. Liquid media cultivation with TIS is considered a technology with great potential in terms of automation and, therefore, reduction of labor costs. However, to optimize the in vitro production of C. longa, it is necessary to establish the appropriate experimental conditions.
The first two variables analyzed are closely related. As the number of explants per flask increases, the area available for growth declines. In addition, the movement of the explants during the immersion cycle in the liquid medium is also reduced. Likewise, the flask volumetric capacity had a similar effect. The number of shoots was proportional to the number of explants initially used per TIS flask. The use of greater volumetric capacity flasks provides a larger distribution area for explants during proliferation and rooting in TIS. The round bottom area of a 1 L flask is half that of a 2 L flask, and the use of flasks with smaller round bottom areas may limit propagation. Using a flask with a small bottom area will limit plant growth and result in underdeveloped plants. Therefore, if the number of explants per flask is high (small distribution area) or the capacity is low (small round bottom area), the percentage of C. longa explants that do not sprout will increase and TIS productivity will drop.
Another factor to consider is the amount of headspace air available to in vitro plants. In studies on the Musa system [7], it was shown that the supply of air in TIS improved the shoot quality compared to growing in solid media. This may be due to the increase in O 2 (oxygen) concentration and the simultaneous decrease in CO 2 (carbon dioxide) and C 2 H 4 (ethylene) concentration by the supply of fresh air during immersions. There are few studies on the effect of the immersion frequency on the micropropagation of C. longa in vitro. Previous reports reveal that decreasing ventilation through low number of immersions produces a high fresh weight of plants [23]. In addition, in another study, the authors compared different methods for micropropagation of Curcuma zedoaria. They concluded that using TIS with a low immersion frequency generates a better shoot proliferation and could be achieved with less labor input than other systems [24]. Although our results were satisfactory and the immersion conditions were effective, this may be a variable that should be considered in future studies.
Next, three different LED light sources were tested. The main wavelengths required for photosynthesis are in the blue and red regions of visible light. The LEDs used in this study have a spectral energy distribution with an emission peak at 420 and 440 nm for the blue LEDs and at 660 nm for the red LEDs. These wavelengths coincide with the known peak absorption wavelengths for chlorophyll a (430 and 663 nm) and b (453 and 642 nm), indicating optimal photosynthetic efficiency [22]. Plants have photoreceptors to control their responses to these physiologically important wavelengths. Phytochromes are primarily red-light photoreceptors that distinguish between red and far-red wavelengths to control physiological responses such as node elongation [25], floral transition [26], and flowering [27]. Blue light, on the other hand, inhibits cell growth. Blue light photoreceptors can regulate and alter gene expression to inhibit stem elongation [28,29] or control chloroplast movements to maximize light absorption [30]. White LEDs have a complex absorption spectrum that includes blue, green, yellow, orange, red, and far-red wavelength photons (Supplementary Figure S1a). However, because the photon composition is more complex, the PAR value per day, i.e., the daily light integral (DLI) that the plant receives in 24 h, is significantly lower than that of other light sources. Similarly, the red and blue YPDF values of the white LED (Table 1) are the lowest of all light sources, prioritizing survival over proliferation and multiplication of C. longa in vitro.
For RB1, the absorption spectrum peaks at blue, red, and far-red wavelengths, a simpler configuration than that of the white LED (Supplementary Figure S1b). In addition, the DLI value of RB1 was four times higher than that of the white LED. The total number of photons is distributed in a limited number of wavelengths that are more useful to plants during the growth process, resulting in plants with higher fresh weight (Figure 5a) compared to other light sources. However, the number of plants and the proliferation rate were lower than those of other light sources (Figure 5b,c), suggesting that red light promotes plants growth and not the multiplication of initial explants. Furthermore, C. longa leaves exposed to RB1 showed progressive etiolation, which is consistent with what has been described in the literature. This is because red LEDs have a more pronounced effect on the plant, acting similar to a growth regulator [31]. Therefore, to improve in vitro propagation of C. longa and obtain a large number of plants, RB1 LED light is not sufficient.
On the other hand, blue photons also showed an effect on the multiplication and growth of C. longa in vitro. The RB2 light source had more blue photons, which promoted explant multiplication rather than growth. There was an increase in plant number and a decrease in fresh weight compared to the other light sources. The DLI value was the highest among all the light sources studied, which could be attributed to the composition of the absorption spectrum. RB2 has a limited spectral width with only three peaks, two in the blue (420 y 460 nm) and one in the red (660 nm) of the visible spectrum ( Figure S1c). Kosell et al. reported that blue light can modulate the concentration of nutritionally important secondary metabolites, such as chlorophyll [32]. These pigments in leaf tissue absorb the maximum amount of light, especially at 430 and 453 nm, and higher concentrations of chlorophyll pigments are produced in response to light with stronger blue wavelengths [32,33]. The RB2 LED light generated more shoots and higher proliferation rates. However, the mechanism by which RB2 and other red and blue photon ratios promotes the multiplication of explants in vitro remains unknown. Therefore, future studies on the effect of LED light on the in vitro culture of C. longa by TIS will have to examine secondary metabolism in detail.

Conclusions
The number of explants per flask, flask volume, and light source affect in vitro propagation of C. longa. The use of 30 explants in 2 L flasks will promote gas exchange (e.g., CO 2 , O 2 , and C 2 H 4 ) and explant movement, resulting in a greater number of plants (higher proliferation rate). The use of RB2 had a positive effect on the in vitro propagation of C. longa. However, other factors such as leaf photosynthetic activity and the synthesis of secondary metabolites produced by C. longa should be taken into account in future studies.