Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Photoautotrophic Culture of Coffea arabusta Somatic Embryos: Development of a Bioreactor for Large‐scale Plantlet Conversion from Cotyledonary Embryos

Photoautotrophic Culture of Coffea arabusta Somatic Embryos: Development of a Bioreactor for... Abstract Somatic embryos were developed from in vitro‐grown leaf discs of Coffea arabusta in modified Murashige and Skoog medium under 30 µmol m–2 s–1 photosynthetic photon flux (PPF). Cotyledonary stage embryos were selected from the 14‐week‐old cultures and were placed under a high (100 µmol m–2 s–1) PPF for 14 d. These pretreated embryos were grown photoautotrophically in three different types of culture systems: Magenta vessel; RITA‐bioreactor (modified to improve air exchange); and a specially designed temporary root zone immersion bioreactor system (TRI‐bioreactor) with forced ventilation. The aims of the study were to achieve large‐scale embryo‐to‐plantlet conversion, and to optimize growth of plantlets under photoautotrophic conditions. The plantlet conversion percentage was highest (84 %) in the TRI‐bioreactor and lowest in the modified RITA‐bioreactor (20 %). Growth and survival of converted plantlets following 45 d of photoautotrophic culture in each of the three culture systems were studied. Fresh and dry masses of leaves and roots of plantlets developed in the TRI‐bioreactor were significantly greater than those of plantlets developed in the modified RITA‐bioreactor or Magenta vessel. The net photosynthetic rate, chlorophyll fluorescence and chlorophyll contents were also highest in plantlets grown in the TRI‐bioreactor. Normal stomata were observed in leaves of plantlets grown in the TRI‐bioreactor, whereas they could be abnormal in plantlets from the modified RITA‐bioreactor. Survival of the plants after transfer from culture followed a similar pattern and was highest in the group grown in the TRI‐bioreactor, followed by plants grown in the modified RITA‐bioreactor and Magenta vessel. In addition, ex vitro growth of plants transferred from the TRI‐bioreactor was faster than that of plants from the other culture systems. Key words: CO2 enrichment, embryo‐to‐plantlet conversion, ex vitro, forced ventilation, in vitro, stomata, survival percentage. Received: 17 December 2001; Returned for revision: 2 February 2002; Accepted: 19 March 2002 INTRODUCTION Somatic embryogenesis is a highly effective technology for mass propagation of elite clones. The quality of a somatic embryo is determined by its maturation and germination ability. One of the challenges preventing the wider application of somatic embryogenesis to clonal propagation is the low rate of embryo germination and conversion to plantlets. One means of increasing germination and the vigour of seedlings from somatic embryos is to provide a synthetic endosperm as a coating to the somatic embryos (Redenbaugh et al., 1988; Redenbaugh and Walker, 1990). This approach has achieved very little success, perhaps because of the leaching of the added nutrients during germination, poor uptake of the added nutrients by the embryo axis or toxicity of the coating. Physiologically, plant conversion or seedling development involves a transition from the heterotrophic embryo to an autotrophic plant. Despite the far‐ranging scope of earlier work, aspects of germination and conversion physiology, and biochemistry in somatic embryos remain unexplored. In a previous study (Afreen et al., 2002), we suggested that once Coffea somatic embryos have developed chlorophyllous cotyledons exhibiting active photosynthesis, they could be successfully cultured photoautotrophically. We anticipate that the ability of somatic embryos to grow photoautotrophically will simplify embryo‐to‐plantlet conversion procedures, allow automation and contribute to reducing production costs specifically by reducing the labour input, and thus improving the plant quality. In commercial micropropagation, labour usually accounts for 70–80 % of the total in vitro and ex vitro costs (Aitken‐Christie et al., 1991). Therefore, in the present study, an attempt was made to culture cotyledonary stage Coffea arabusta somatic embryos photoautotrophically for large‐scale embryo‐to‐plantlet conversion. To this end, a specially designed bioreactor with a temporary root zone immersion system (TRI) was developed and used for embryo‐to‐plantlet conversion. The plantlet conversion percentage, growth and physiology, and ex vitro survival of plants were studied and compared with those of plants grown in modified RITA‐bioreactors and in Magenta vessels. MATERIALS AND METHODS Development of a bioreactor with a temporary root zone immersion system (TRI‐bioreactor) The newly designed bioreactor consisted of two main chambers (Fig. 1): the lower chamber was used as a reservoir for the nutrient solution, and the upper one for culturing embryos. A narrow air distribution chamber was located between these two chambers. Two air‐inlet tubes (internal diameter 5 mm; length 10 mm) opened into the air distribution chamber and were connected directly to an air pump (Non noise S200; Artem Co. Ltd, Osaka, Japan) via a filter disc (pore diameter 0·45 µm, diameter 45 mm; Nippon Millipore Co. Ltd, Yonezawa, Japan) to prevent microbes entering the culture vessel. The top of the air distribution chamber had several narrow tubes that were fitted vertically in between the rows of the cell tray and opened in the culture chamber headspace. CO2‐enriched air entered the culture chamber from the air distribution chamber by means of these vertical tubes. The culture chamber contained a six × nine cell autoclavable cell tray (Minoru Sangyo Co. Ltd, Okayama, Japan). Outflow was through four Millipore membranes (pore diameter 0·45 μm; Nippon Millipore Co. Ltd) covering the outlet holes (10 mm diameter) on the side walls of the bioreactor. An air inlet tube (tube ‘a’ in Fig. 1) connected an air pump to the headspace of the nutrient reservoir chamber; an electric timer operated the pump. A second tube (tube ‘b’) ran from close to the base of the reservoir to the culture chamber. To supply nutrient solution to the culture chamber the air pump was switched on, thereby raising the pressure in the headspace of the reservoir and forcing the nutrient solution from the reservoir into the culture chamber. The nutrient solution immersed the root zone temporarily for a total of 15 min every 6 h. After 15 min, the air pump was switched off and the excess nutrient solution flowed back into the reservoir under gravity. Plant material, culture conditions and treatments Establishment of the culture. Nodal cuttings of coffee plantlets (Coffea arabusta) were cultured in Magenta vessels containing hormone‐free MS (Murashige and Skoog, 1962) medium supplemented with 20 g l–1 sucrose. After 4 weeks of culture, regenerated leaves were collected and cut into pieces (10 × 5 mm) and were placed in a Magenta vessel containing modified MS medium (for details, see Afreen et al., 2002). Agar (8 g l–1; Kanto Chemical Co., Tokyo, Japan) was used as a gelling agent and 30 g l–1 sucrose was added in the medium. Cultures were placed in a growth chamber with an air temperature of 23 °C and a 16 h d–1 photoperiod provided by cool‐white fluorescent lamps (National Co., Tokyo, Japan). The ambient CO2 concentration was 400 µmol mol–1 and the photosynthetic photon flux (PPF) was 30 µmol m–2 s–1 measured on the empty culture shelf. Somatic embryos developed within 9–12 weeks of culture. Cotyledonary stage embryos were selected for use as experimental materials. Methods for measuring CO2 concentration and net photosynthetic rate CO2 concentration in the culture headspace was measured using the method described by Afreen et al. (2002). Net photosynthetic rate per plantlet, Pn (mol h–1 per plantlet) was calculated following the method of Fujiwara et al. (1987) using the following equation: Pn = [k × E × V × (Cout – Cin)]/N where k is the conversion factor of CO2 from volume to mole, E is the number of air exchanges per hour of the culture vessel (h–1), V is the air volume of the culture vessel (l), Cin and Cout are the CO2 concentrations (mol mol–1) inside and outside the culture vessel under steady‐state conditions during the photoperiod, and N is the number of plantlets per vessel. Measuring chlorophyll fluorescence, chlorophyll contents and stomatal development A fibre‐optic based chlorophyll fluorimeter (Hansatech, UK) was used to analyse the photochemical activity of the somatic embryos on day 45. In dark‐adapted samples (2 h), the maximal quantum yield of photochemistry through PSII (Φpmax) was calculated from the ratio (Fm – Fo)/Fm (Kitajima and Butler, 1975). The actual quantum yield (Φp) of PSII photochemistry in light‐adapted leaves was calculated from the steady‐state level of chlorophyll fluorescence (Fs) and maximal fluorescence level: Φp = (Fm – Fs)/Fm (Havaux et al., 1991). Samples for chlorophyll measurement were collected on day 45 from plantlets in each of the vessels, and in every case an appropriate mass of leaves was soaked in 80 % ice‐cold acetone for 3 d, centrifuged at 300 r.p.m. for 10 min and measured for light absorption between 400 and 700 nm in a spectrophotometer (Hitachi, Japan). From the absorption curves, the proportion of chlorophyll contents was evaluated according to the formula of Lichtenthaler and Wellburn (1983). For studying stomata (on day 40), the first true leaves of plantlets from different treatments were collected and epidermal peels were taken from the abaxial (lower) surface. Stomata were studied and photographed using a pre‐calibrated digital microscope (Keyence Corporation, Osaka, Japan). Stomatal density (number per mm2) and stomatal length and width (with guard cells) were measured directly under the microscope for each treatment. Stomatal length refers to the distance between the ends of the guard cells, and the width is the distance transversely across them. Plantlet conversion percentage and growth of plantlets Somatic embryos were established as described earlier. After 14 weeks of culture the vessels were placed under a high PPF (100 µmol m–2 s–1) for 14 d (for details, see Afreen et al., 2002). In the first attempts to achieve embryo‐to‐plantlet conversion, 16‐week‐old cotyledonary stage embryos (including 2 weeks pretreatment) were selected for use as experimental material. The embryos were cultured under photoautotrophic conditions (in sugar‐free medium with CO2 enrichment in the culture headspace and high PPF) in three different types of culture systems: (1) Magenta vessel; (2) a modified RITA‐bioreactor with temporary immersion system; and (3) a newly developed bioreactor with a temporary root zone immersion system (TRI‐bioreactor). A mixture of vermiculite and paper pulp (as described by Afreen et al., 2000) was used as supporting medium in the Magenta vessels and in TRI‐bioreactors. For modified RITA‐bioreactors, MS liquid nutrient solution was used and the immersion frequency was 5 min every 6 h, achieved by connecting an air pump through an electric timer. The planting density for all the treatments was 2·4 × 103 plantlets m–2. To provide natural ventilation in the Magenta vessels, two gas‐permeable Millipore filter membranes (pore diameter 0·45 µm) were attached over holes (10 mm diameter) in the lid of the vessels. RITA‐bioreactors were modified by using three gas‐permeable filter membranes with a pore diameter of 0·45 µm to cover the holes (10 mm diameter) in the lid of each of these vessels. The number of air exchanges was 2·6 h–1 in both Magenta vessels and modified RITA‐bioreactors throughout the experiment (measured according to Kozai et al., 1988). In the TRI‐bioreactor, forced ventilation was introduced by using an air pump connected to the headspace of the air distribution chamber; initial flow rates were 50 ml min–1 (1·6 air exchanges h–1) and were gradually increased every 2 or 3 d to maintain the CO2 concentration in the culture headspace at approx. 1000 µmol mol–1. The maximum flow rate was 200 ml min–1 on day 45 (5·8 air exchanges h–1). For all treatments, hormone‐free MS medium was used as a basal medium; sucrose, vitamins and amino acids were eliminated from the formulation to ensure conditions were photoautotrophic. Vessels were placed in a growth chamber with an enriched CO2 concentration (1000–1100 µmol mol–1) and with a PPF of 100 µmol m–2 s–1 during the 16 h photoperiod; ambient relative humidity (RH) was 80–85 % and the air temperature was 23 °C. Experiments were conducted for 45 d and plantlet conversion percentage, fresh and dry masses of the plantlets, and rooting percentage were recorded at harvest. For chlorophyll fluorescence, chlorophyll content and stomatal studies, ten replicates were taken from each treatment. CO2 concentration in the culture headspace was measured throughout the culture period and the net photosynthetic rates were calculated as described earlier. Plantlets were transplanted in a glasshouse (average temperature 29 ± 2 °C; 60–70 % RH) and the survival percentage was recorded after 15 d. Thirty days after transplanting, plants were harvested and fresh and dry masses of surviving plants were recorded. When necessary, statistical significance was determined by one‐ or two‐way ANOVA and least significant difference test (LSD). Experiments were conducted twice. RESULTS AND DISCUSSION Plantlet conversion percentage and growth of plantlets In terms of plantlet conversion percentage under photoautotrophic conditions, the results (Table 1) revealed very distinct differences among plantlets grown in either a temporary root zone immersion bioreactor with forced ventilation, a modified RITA‐bioreactor or Magenta vessels with natural ventilation. In the TRI‐bioreactor, almost 84 % of the embryos successfully produced plantlets, whereas in Magenta vessels and in the modified RITA‐bioreactor the conversion rates were 53 and 20 %, respectively. Considering all parameters of growth and development, it was evident that embryos cultured in the TRI‐bioreactor produced more vigorous shoots and normal roots than those grown in a modified RITA‐bioreactor or in Magenta vessels (Table 1; Fig. 2). Plantlets grown in the TRI‐bioreactor had the highest number of leaves per plantlet (6·6 including cotyledons), 2·5 and 2·1 times more, respectively, than those of plantlets grown in the modified RITA‐bioreactor and Magenta vessels. The leaf area of plantlets grown in the TRI‐bioreactor (2·9 cm2) was 3·6 and 4·8 times that of plantlets grown in the modified RITA‐bioreactor and Magenta vessels, respectively (Table 1). Fresh and dry masses of leaves of plantlets grown in the TRI‐bioreactor were significantly higher than those of plantlets grown in other treatments (2·8 and 3·5 times, respectively, those of the plantlets grown in a modified RITA‐bioreactor, and 4·4 and 6·3 times those of plantlets grown in Magenta vessel; Table 1). Stem fresh and dry masses followed a similar pattern: stem dry mass was 1·7 and 2·5 times that of plantlets grown in a modified RITA‐bioreactor and Magenta vessel, respectively (Table 1). In general, apart from leaf number, the growth attained in a modified RITA‐bioreactor was intermediate between that of plantlets grown in the TRI‐bioreactor and in Magenta vessels; however, most of the growth parameters of plantlets grown in Magenta vessels were only marginally different from those grown in the modified RITA‐bioreactor. The most remarkable difference observed among the treatments was in rooting percentage. In the TRI‐bioreactor, 90 % of plantlets developed roots, 3 times more than plantlets grown in a modified RITA‐bioreactor and 1·6 times more than plantlets in Magenta vessels; some roots of plantlets in the TRI‐bioreactor produced laterals (data not shown). It should be noted that roots that developed in a few plantlets in the modified RITA‐bioreactor remained very small. Plantlets grown in the TRI‐bioreactor had the greatest root fresh and dry masses (6·5 and 12 times those of plantlets grown in the modified RITA‐bioreactor; Table 1). Among the treatments, plantlets cultured in Magenta vessels had a root growth pattern (fresh and dry masses of 4·8 and 0·4 mg, respectively) intermediate between that of plantlets grown in TRI‐ and modified RITA‐bioreactors. CO2 concentrations in the headspace and net photosynthetic rate As plantlets in the TRI‐bioreactor grew, the CO2 concentration in the culture headspace was controlled by increasing the number of air exchanges (Fig. 3A). Thus, despite the increase in biomass, CO2 concentrations were nearly the same throughout the experimental period (approx. 1280 µmol mol–1); in contrast, in Magenta vessels and in the modified RITA‐bioreactor, the number of air exchanges could not be controlled, and were thus 3·3 h–1 throughout the experiment (under natural ventilation). A reduction in CO2 concentration was noted in the culture vessel headspace in the Magenta vessels during the photoperiod: on day 7 the CO2 concentration was 1280 µmol mol–1, while on day 42 it was 1254 µmol mol–1. In the modified RITA‐bioreactor, the CO2 concentration in the headspace fell from 1278 µmol mol–1 on day 7 to 1266 µmol mol–1 on day 42 despite the low air exchange rate; possible reasons for this low consumption of CO2 by plantlets include: (1) total CO2 consumption is low due to the small size of plantlets; (2) chlorophyll contents of the plantlets are lower than those of plantlets in other treatments; and most importantly (3), as the chlorophyllous plant material remained moist almost all the time due to immersion of whole plantlets every 6 h and the high humidity in the culture headspace, these plantlets were probably virtually unable to fix any CO2 from the atmosphere for in vitro metabolism. As expected, among the treatments the highest net photosynthetic rate was observed in plantlets grown in the TRI‐bioreactor (Fig. 3B). The result clearly shows that in this treatment, the forced ventilation system provided the best conditions throughout the experiment for the assimilation of CO2. As a result, the net photosynthetic rate, which is a closer reflection of normal in vitro metabolism, was 8·3‐ and 3·2 times greater than that of plantlets grown in the modified RITA‐bioreactor and in Magenta vessels, respectively, on day 42 (Fig. 3B). Chlorophyll contents, chlorophyll fluorescence and stomatal development In general, the highest chlorophyll content based on the fresh mass of leaves was observed in plantlets grown in the TRI‐bioreactor (Fig. 4A and B). Chloro phyll a and b contents were 606 and 241 µg g–1 fresh mass, respectively, in plantlets grown in the TRI‐bioreactor, 2 and 1·6 times those of leaves of plantlets grown in the modified RITA‐bioreactor. In the case of Magenta vessels, chlorophyll a and b contents of leaves were intermediate between those of plantlets grown in TRI‐ and modified RITA‐bioreactors. The potential activity of PSII (Φpmax), as estimated in the dark, was nearly the same in leaves of plantlets grown in the TRI‐bioreactor (Φpmax = 0·89) and in Magenta vessels (Φpmax = 0·83) (Fig. 4C); in contrast, Φpmax was low in leaves of plantlets grown in a modified RITA‐bioreactor (0·76). A similar pattern was observed for actual photochemical efficiency of PSII (Φp) (Fig. 4D), which is known to be a good estimate of the quantum yield of photosynthetic electron transport (Genty et al., 1989, 1992). An increase in the quantum yield for electron transport was noted in leaves of plantlets grown in both the TRI‐bioreactor (Φp reaching 0·35) and in Magenta vessels (Φp = 0·32). As noted earlier, due to their low PSII activity, leaves of plantlets grown in the modified RITA‐bioreactor exhibited comparatively lower electron transport activity (Φp = 0·25) than that of plantlets in the other two treatments (Fig. 4D). Microscopy highlighted the differences among treatments with respect to stomatal density (Fig. 2E–G), which was highest in leaves of plantlets grown in the TRI‐bioreactor (8·3 mm–2 leaf area) followed by those of plantlets from the modified RITA‐bioreactor (7·5 mm–2 leaf area) (Fig. 4E). Compared with the other treatments, stomatal density was lowest in leaves of plantlets grown in Magenta vessels (5·9 mm–2 leaf area) (Fig. 4E). Average stomatal length was nearly the same in leaves of all three treatments (Fig. 4F). The most noticeable feature was that some stomata that developed in the leaves of plantlets grown in the modified RITA‐bioreactor were open wide (Fig. 2G), while others were distorted or still morphologically immature. It is possible that these stomata may not function properly, although no specific attempt was made to investigate this in the present study. Ex vitro survival and growth After transplanting under glasshouse conditions, a similar trend was noted in terms of survival percentage and growth (Fig. 2M–O). Ex vitro survival, which was recorded on day 15 of transplanting, was highest (89 %) in plantlets grown in the TRI‐bioreactor. Plantlets grown in Magenta vessels had a survival percentage of 67 %, although their growth was much slower than that of plants grown in the TRI‐bioreactor. When plantlets from the modified RITA‐bioreactor were transferred ex vitro only 33 % survived. In terms of ex vitro growth, it was noticeable that plants from the TRI‐bioreactor exhibited much faster growth (Fig. 2M) and, as a consequence, after 30 d of transplanting almost all the growth parameters were significantly greater than those of plants grown in modified RITA‐bioreactors and Magenta vessels (Fig. 5). In the TRI‐bioreactor treatment, leaf number and leaf area were 2·7 and 2·8 times greater, respectively, than those of plantlets from the modified RITA‐bioreactor, and 2·0 and 2·7 times greater, respectively, than those of plants from the Magenta vessel on day 30 after transplanting (Fig. 5A and B). Similarly, leaf and root fresh masses were also enhanced and were 69 and 19 mg per plant, respectively, in the TRI‐bioreactor treatment, compared with 19 and 7·3 mg per plant in the Magenta vessel treatment and 25 and 2·4 mg per plant in plantlets from the modified RITA‐bioreactor (Fig. 5C and E). During the present study, it became increasingly apparent that the vigorous growth (Fig. 5F) and higher survival percentage observed in plants from the TRI‐bioreactor could be the result of many environmental and physiological conditions during the in vitro culture period: for example, the relative humidity under forced ventilation was lower (85–90 %) than that in the modified RITA‐bioreactor (95–99 %) or in Magenta vessels (≤95 %). Smith et al. (1992) suggest that reducing the relative humidity in the culture headspace could improve resistance to wilting of micropropagated grapevine. In a previous experiment (Zobayed et al., 2000), we found that lowering the relative humidity in the culture headspace by introducing forced ventilation can increase the deposition of epicuticular wax on the leaf surface, which can, in turn, prevent water loss after transplanting and thus increase the chance of survival and subsequent growth. The results provide clear evidence that Magenta vessels and modified RITA‐bioreactors gave the lowest growth both in and ex vitro in terms of plantlet conversion from cotyledonary stage embryos under photoautotrophic conditions. RITA‐bioreactors are normally used for the development of plantlets from embryogenic cell suspension cultures using a sugar‐containing medium. In addition, this vessel is also claimed to be suitable for embryo‐to‐plantlet development without handling the plant material (Berthouly and Etienne, 1999). However, our results show that for embryo‐to‐plantlet conversion under photoautotrophic conditions, the use of a modified RITA‐bioreactor is less effective at promoting growth (shoot and root) compared with the newly developed TRI‐bioreactor. This is probably because in the modified RITA‐bioreactor the plant material is either never completely dry or it takes a long time to dry out after each immersion with nutrient solution because the humidity inside the vessel is normally high (95–99 %). Thus, there is a thin layer of water surrounding the plant material which acts as a boundary layer, impeding the exchange of gases between the plant and the surrounding environment, and possibly preventing CO2 fixation by green plant material which is vital for the photoautotrophic growth of embryos. In the case of conventional photomixotrophic systems, the medium contains sugar and therefore the lack of air exchange may not have such serious consequences as it does for plants that depend on CO2 in the atmosphere for their growth (photoautotrophy). Again, it should be noted that the RITA‐bioreactor system has not been developed for culturing plant material under photoautotrophic conditions, although in the present study the vessel was modified by attaching three gas permeable filter membranes; the same was done for Magenta vessels. Furthermore, the results have also established that in the TRI‐bioreactor plantlets not only exhibited the best growth, but they were physiologically normal, survived well and grew faster ex vitro. In general, the following steps are necessary for embryo‐to‐plantlet conversion in conventional somatic embryogenesis systems: (1) embryo selection and transfer on the germination medium; (2) germinated and rooted plantlet selection and transfer to soil; (3) acclimatization (Gupta et al., 1993). Each of these steps is usually time consuming, especially for mass propagation, and labour intensive because the embryos are selected individually, in most cases by hand under a stereomicroscope. The use of machine vision (Cazzulino et al., 1990) and image analysis (Harrell and Cantliffe, 1991) is still limited, and can be used only to classify and sort the embryos (step 1, above). These selected embryos are transferred onto a semi‐solid medium for germination and plantlets with an epicotyl are then selected by hand and transferred to soil for acclimatization and growth. However, we were able to simplify these stages in our experiment by using the TRI‐bioreactor. Although in our system cotyledonary stage embryos were selected by hand, it is possible to select these embryos using automation, and once the embryos have been transferred to the TRI‐bioreactor, germination, shoot and root development, and acclimatization take place in the same bioreactor without any need to handle the plant material or change the culture medium. Therefore, the new system should substantially reduce production costs, and the density limitations encountered in Magenta vessels and in the modified RITA‐bioreactor can also be overcome by increasing the number of cells in the culture cell tray. ACKNOWLEDGEMENT We are grateful for financial support from the Japanese Society for the Promotion of Science (JSPS) Research‐for‐the‐Future Program. View largeDownload slide Fig. 1. Schematic diagram of the temporary root zone immersion (TRI) bioreactor with forced ventilation system. View largeDownload slide Fig. 1. Schematic diagram of the temporary root zone immersion (TRI) bioreactor with forced ventilation system. View largeDownload slide Fig. 2. A, Coffee somatic embryos regenerated from leaf discs after 14 weeks of culture under low light (30 µmol m–2 s–1) followed by 2 weeks under high light (100 µmol m–2 s–1) (×0·5). B–D, 45‐d‐old plantlets developed from cotyledonary stage embryos under photoautotrophic conditions in a temporary root zone immersion (TRI) bioreactor (B, ×0·2), a Magenta vessel (C, ×0·7) and a modified RITA‐bioreactor (D, ×0·2). E–G, Stomata from the abaxial (lower) surface of the first true leaves of plantlets developed photoautotrophically in a TRI‐bioreactor (E), a Magenta vessel (F) and a modified RITA‐bioreactor (G). H and I, Individual plantlets immediately before transplanting ex vitro grown in a TRI‐bioreactor (H) and a Magenta vessel (I). J–L, Root development of plantlets grown in a TRI‐bioreactor (J), a Magenta vessel (K) and a modified RITA‐bioreactor (L). M–O, On day 30 after transplanting plantlets previously grown in a TRI‐bioreactor (M), a Magenta vessel (N) and a modified RITA‐bioreactor (O). View largeDownload slide Fig. 2. A, Coffee somatic embryos regenerated from leaf discs after 14 weeks of culture under low light (30 µmol m–2 s–1) followed by 2 weeks under high light (100 µmol m–2 s–1) (×0·5). B–D, 45‐d‐old plantlets developed from cotyledonary stage embryos under photoautotrophic conditions in a temporary root zone immersion (TRI) bioreactor (B, ×0·2), a Magenta vessel (C, ×0·7) and a modified RITA‐bioreactor (D, ×0·2). E–G, Stomata from the abaxial (lower) surface of the first true leaves of plantlets developed photoautotrophically in a TRI‐bioreactor (E), a Magenta vessel (F) and a modified RITA‐bioreactor (G). H and I, Individual plantlets immediately before transplanting ex vitro grown in a TRI‐bioreactor (H) and a Magenta vessel (I). J–L, Root development of plantlets grown in a TRI‐bioreactor (J), a Magenta vessel (K) and a modified RITA‐bioreactor (L). M–O, On day 30 after transplanting plantlets previously grown in a TRI‐bioreactor (M), a Magenta vessel (N) and a modified RITA‐bioreactor (O). View largeDownload slide Fig. 3. A, Carbon dioxide concentrations in the culture headspace of a TRI‐bioreactor, Magenta vessel and modified RITA‐bioreactor; B, net photosynthetic rates of coffee plantlets grown in a TRI‐bioreactor, Magenta vessel and in a modified RITA‐bioreactor. View largeDownload slide Fig. 3. A, Carbon dioxide concentrations in the culture headspace of a TRI‐bioreactor, Magenta vessel and modified RITA‐bioreactor; B, net photosynthetic rates of coffee plantlets grown in a TRI‐bioreactor, Magenta vessel and in a modified RITA‐bioreactor. View largeDownload slide Fig. 4. Chlorophyll a (A) and b (B) contents, chlorophyll fluorescence (C and D), stomatal density (E) and stomatal length (F) of leaves of 45‐d‐old coffee plantlets grown from cotyledonary stage embryos under photoautotrophic conditions. Significant difference between treatments at P ≤0·05 indicated by a, b, c was determined by Student–Newman–Keuls test. View largeDownload slide Fig. 4. Chlorophyll a (A) and b (B) contents, chlorophyll fluorescence (C and D), stomatal density (E) and stomatal length (F) of leaves of 45‐d‐old coffee plantlets grown from cotyledonary stage embryos under photoautotrophic conditions. Significant difference between treatments at P ≤0·05 indicated by a, b, c was determined by Student–Newman–Keuls test. View largeDownload slide Fig. 5. Growth parameters of 30‐d‐old coffee plantlets after transplanting ex vitro. A, Leaf number; B, leaf area; C, leaf fresh mass; D, stem fresh mass; E, root fresh mass; F, total increase of fresh mass after transplanting ex vitro. Significant difference between treatments at P ≤0·05 indicated by a, b, c was determined by Student–Newman–Keuls test. View largeDownload slide Fig. 5. Growth parameters of 30‐d‐old coffee plantlets after transplanting ex vitro. A, Leaf number; B, leaf area; C, leaf fresh mass; D, stem fresh mass; E, root fresh mass; F, total increase of fresh mass after transplanting ex vitro. Significant difference between treatments at P ≤0·05 indicated by a, b, c was determined by Student–Newman–Keuls test. Table 1. Plantlet conversion percentage and growth of plantlets from cotyledonary embryos of coffee grown photoautotrophically in a temporary root zone immersion bioreactor (TRI‐bioreactor), a modified RITA‐bioreactor and in Magenta vessels for 45 d Treatment  Leaf number  Leaf area (cm2)  Leaf f. wt (mg)  Leaf d. wt (mg)  Stem f. wt (mg)  Stem d. wt (mg)  Root f. wt (mg)  Root d. wt (mg)  Percentage rooting (%)  Plantlet conversion (%)  TRI‐bioreactor  6·6 ± 1·4a  2·9 ± 1·2a  57 ± 19a  8·2 ± 2a  27 ± 7a  3·7 ± 1a  11 ± 7a  1·2 ± 0·7a  90 ± 9a  84 ± 1a  RITA‐ bioreactor  2·6 ± 1·2b  0·8 ± 0·3b  20 ± 7b  2·3 ± 1b  18 ± 9b  2·1 ± 1a,b  1·7 ± 0·1c  0·1 ± 0c  29 ± 2c  20 ± 1c  Magenta vessel  3·14 ± 0·9b  0·6 ± 0·4b  12·9 ± 8c  1·3 ± 0·8b  13·6 ± 8b  1·5 ± 1·0b  4·8 ± 2·1b  0·4 ± 0·1b  57 ± 5b  53 ± 6b  Treatment  Leaf number  Leaf area (cm2)  Leaf f. wt (mg)  Leaf d. wt (mg)  Stem f. wt (mg)  Stem d. wt (mg)  Root f. wt (mg)  Root d. wt (mg)  Percentage rooting (%)  Plantlet conversion (%)  TRI‐bioreactor  6·6 ± 1·4a  2·9 ± 1·2a  57 ± 19a  8·2 ± 2a  27 ± 7a  3·7 ± 1a  11 ± 7a  1·2 ± 0·7a  90 ± 9a  84 ± 1a  RITA‐ bioreactor  2·6 ± 1·2b  0·8 ± 0·3b  20 ± 7b  2·3 ± 1b  18 ± 9b  2·1 ± 1a,b  1·7 ± 0·1c  0·1 ± 0c  29 ± 2c  20 ± 1c  Magenta vessel  3·14 ± 0·9b  0·6 ± 0·4b  12·9 ± 8c  1·3 ± 0·8b  13·6 ± 8b  1·5 ± 1·0b  4·8 ± 2·1b  0·4 ± 0·1b  57 ± 5b  53 ± 6b  Each value represents a mean ± s.d. of 20 replicates. Means within a column followed by different superscripts are significantly different at P < 0·05 by the least significant difference test. All parameters were significant at P < 0·01 (ANOVA). View Large References Afreen F , Zobayed SMA, Kozai T. 2002. Photoautotrophic culture of Coffea arabusta somatic embryos: photosynthetic ability and growth of different stage embryos. Annals of Botany  90: 11–19. Google Scholar Afreen F , Zobayed SMA, Kubota C, Kozai T, Hasegawa O. 2000. A combination of vermiculite and paper pulp supporting material for the photoautotrophic micropropagation of sweet potato. Plant Science  157: 225–231. Google Scholar Aitken‐Christie J , Kozai T, Takayama S. 1991. Automation in plant tissue culture–general introduction and overview. In: Aitken‐Christie J, Kozai T, Smith L, eds. Automation and environmental control in plant tissue culture. Dordrecht, The Netherlands: Kluwer Academic Publishers, 1–18. Google Scholar Berthouly M , Etienne H. 1999. Somatic embryogenesis of coffee. In: Jain SM, Gupta PK, Newton RJ, eds. Somatic embryogenesis in woody plants, Volume 5. Dordrecht, The Netherlands: Kluwer Academic Publishers, 259–287. Google Scholar Cazzulino D , Pederson H, Chin CK. 1990. Bioreactors and image analysis for scale‐up and plant propagation. In: Vasil IK, ed. Scale‐up and automation in plant propagation. New York: Academic Press, 147–175. Google Scholar Fujiwara K , Kozai T, Watanabe I. 1987. Fundamental studies of environments in plant tissue culture vessels. (3) Measurement of carbon dioxide gas concentration in closed vessels containing tissue cultured plantlets and estimates of net photosynthetic rates of plantlets. Journal of Agriculture Meteorology  43: 21–30. Google Scholar Genty B , Briantais JM, Baker NR. 1989. The relationship between the quantum yield of photosynthetic electron transport and quenching of chlorophyll fluorescence. Biochimica et Biophysica Acta  990: 87–92. Google Scholar Genty B , Goulas B, Dimon B, Peltier G, Moya I. 1992. Modulation of efficiency of primary conversion in leaves, mechanisms involved at PSII. In: Murata N, ed. Research in photosynthesis IV. Amsterdam: Kluwer Academic Publishers, 603–610. Google Scholar Gupta PK , Timmis R, Carlson WC. 1993. Somatic embryogenesis: A possible tool for large scale propagation of forestry species. In: Soh WY, Liu JR, Komamine A, eds. Advances in development biology and biotechnology of higher plants. Proceedings of the First Asia‐Pacific Conference on Plant Cell and Tissue Culture, Tacjon, Korea, The Korean Society of Plant Tissue Culture, 18–37. Google Scholar Harrell RC , Cantliffe DJ. 1990. Automated evaluation of somate embryogenesis in sweet potato by Machine Vision. In: Vasil IK, ed. Scale‐up and automation in plant propagation. New York: Academic Press, 179–195. Google Scholar Havaux M , Strasser RJ, Greppin H. 1991. A theoretical and experimental analysis of the qp and qN coefficients of chlorophyll fluorescence quenching and their relation to photochemical and nonphotochemical events. Photosynthetic Research  27: 41–55. Google Scholar Kitajima M , Butler WL. 1975. Quenching of chlorophyll fluorescence and primary photochemistry by dibromothymoquinone. Biochimica et Biophysica Acta  376: 105–111 Google Scholar Kozai T , Koyama Y, Watanabe I. 1988. Multiplication of potato plantlets in vitro with sugar free medium under high photosynthetic photon flux. Acta Horticulturae  230: 121–127. Google Scholar Lichtenthaler HK , Wellburn AR. 1983. Determinations of total carotenoids and chlorophylls a & b of leaf extracts in different solvents. Biochemical Society Transactions  11: 591 – 592. Google Scholar Murashige T , Skoog F. 1962. A revised medium for rapid growth and bioassays with tobacco tissue culture. Physiologia Plantarum  15: 473–497. Google Scholar Redenbaugh K , Walker K. 1990. Role of artificial seeds in alfalfa breeding. In: Bhojwani SS, ed. Plant tissue culture. Application and limitations. New York: Elsevier Publishing, 102–135. Google Scholar Redenbaugh K , Fuji JA, Slade D. 1988. Encapsulated plant embryos. In: Biotechnology in Agriculture. Great Britain Agriculture Food Research Council, 226–229. Google Scholar Smith EF , Gribaudo I, Roberts AV, Mottley J. 1992. Paclobutrazol and reduced humidity improve resistance to wilting of micropropagated grapevine. HortScience  27: 111–113. Google Scholar Zobayed SMA , Afreen F, Kubota C, Kozai T. 2000. Water control ability of Ipomoea batatas grown photoautotrophically under forced ventilation and photomixotrophically under natural ventilation. Annals of Botany  85: 603–610. Google Scholar Author notes 1Department of Bioproduction Science, Chiba University, Matsudo, Chiba 271‐8510, Japan http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Annals of Botany Oxford University Press

Photoautotrophic Culture of Coffea arabusta Somatic Embryos: Development of a Bioreactor for Large‐scale Plantlet Conversion from Cotyledonary Embryos

AFREEN, F.; ZOBAYED, S. M. A.; KOZAI, T.
Annals of Botany , Volume 90 (1) – Jul 1, 2002
Download PDF
9 pages

Loading next page...
 
/lp/oxford-university-press/photoautotrophic-culture-of-coffea-arabusta-somatic-embryos-NrxznKAlFl

References (24)

Publisher
Oxford University Press
ISSN
0305-7364
eISSN
1095-8290
DOI
10.1093/aob/mcf151
Publisher site
See Article on Publisher Site

Abstract

Abstract Somatic embryos were developed from in vitro‐grown leaf discs of Coffea arabusta in modified Murashige and Skoog medium under 30 µmol m–2 s–1 photosynthetic photon flux (PPF). Cotyledonary stage embryos were selected from the 14‐week‐old cultures and were placed under a high (100 µmol m–2 s–1) PPF for 14 d. These pretreated embryos were grown photoautotrophically in three different types of culture systems: Magenta vessel; RITA‐bioreactor (modified to improve air exchange); and a specially designed temporary root zone immersion bioreactor system (TRI‐bioreactor) with forced ventilation. The aims of the study were to achieve large‐scale embryo‐to‐plantlet conversion, and to optimize growth of plantlets under photoautotrophic conditions. The plantlet conversion percentage was highest (84 %) in the TRI‐bioreactor and lowest in the modified RITA‐bioreactor (20 %). Growth and survival of converted plantlets following 45 d of photoautotrophic culture in each of the three culture systems were studied. Fresh and dry masses of leaves and roots of plantlets developed in the TRI‐bioreactor were significantly greater than those of plantlets developed in the modified RITA‐bioreactor or Magenta vessel. The net photosynthetic rate, chlorophyll fluorescence and chlorophyll contents were also highest in plantlets grown in the TRI‐bioreactor. Normal stomata were observed in leaves of plantlets grown in the TRI‐bioreactor, whereas they could be abnormal in plantlets from the modified RITA‐bioreactor. Survival of the plants after transfer from culture followed a similar pattern and was highest in the group grown in the TRI‐bioreactor, followed by plants grown in the modified RITA‐bioreactor and Magenta vessel. In addition, ex vitro growth of plants transferred from the TRI‐bioreactor was faster than that of plants from the other culture systems. Key words: CO2 enrichment, embryo‐to‐plantlet conversion, ex vitro, forced ventilation, in vitro, stomata, survival percentage. Received: 17 December 2001; Returned for revision: 2 February 2002; Accepted: 19 March 2002 INTRODUCTION Somatic embryogenesis is a highly effective technology for mass propagation of elite clones. The quality of a somatic embryo is determined by its maturation and germination ability. One of the challenges preventing the wider application of somatic embryogenesis to clonal propagation is the low rate of embryo germination and conversion to plantlets. One means of increasing germination and the vigour of seedlings from somatic embryos is to provide a synthetic endosperm as a coating to the somatic embryos (Redenbaugh et al., 1988; Redenbaugh and Walker, 1990). This approach has achieved very little success, perhaps because of the leaching of the added nutrients during germination, poor uptake of the added nutrients by the embryo axis or toxicity of the coating. Physiologically, plant conversion or seedling development involves a transition from the heterotrophic embryo to an autotrophic plant. Despite the far‐ranging scope of earlier work, aspects of germination and conversion physiology, and biochemistry in somatic embryos remain unexplored. In a previous study (Afreen et al., 2002), we suggested that once Coffea somatic embryos have developed chlorophyllous cotyledons exhibiting active photosynthesis, they could be successfully cultured photoautotrophically. We anticipate that the ability of somatic embryos to grow photoautotrophically will simplify embryo‐to‐plantlet conversion procedures, allow automation and contribute to reducing production costs specifically by reducing the labour input, and thus improving the plant quality. In commercial micropropagation, labour usually accounts for 70–80 % of the total in vitro and ex vitro costs (Aitken‐Christie et al., 1991). Therefore, in the present study, an attempt was made to culture cotyledonary stage Coffea arabusta somatic embryos photoautotrophically for large‐scale embryo‐to‐plantlet conversion. To this end, a specially designed bioreactor with a temporary root zone immersion system (TRI) was developed and used for embryo‐to‐plantlet conversion. The plantlet conversion percentage, growth and physiology, and ex vitro survival of plants were studied and compared with those of plants grown in modified RITA‐bioreactors and in Magenta vessels. MATERIALS AND METHODS Development of a bioreactor with a temporary root zone immersion system (TRI‐bioreactor) The newly designed bioreactor consisted of two main chambers (Fig. 1): the lower chamber was used as a reservoir for the nutrient solution, and the upper one for culturing embryos. A narrow air distribution chamber was located between these two chambers. Two air‐inlet tubes (internal diameter 5 mm; length 10 mm) opened into the air distribution chamber and were connected directly to an air pump (Non noise S200; Artem Co. Ltd, Osaka, Japan) via a filter disc (pore diameter 0·45 µm, diameter 45 mm; Nippon Millipore Co. Ltd, Yonezawa, Japan) to prevent microbes entering the culture vessel. The top of the air distribution chamber had several narrow tubes that were fitted vertically in between the rows of the cell tray and opened in the culture chamber headspace. CO2‐enriched air entered the culture chamber from the air distribution chamber by means of these vertical tubes. The culture chamber contained a six × nine cell autoclavable cell tray (Minoru Sangyo Co. Ltd, Okayama, Japan). Outflow was through four Millipore membranes (pore diameter 0·45 μm; Nippon Millipore Co. Ltd) covering the outlet holes (10 mm diameter) on the side walls of the bioreactor. An air inlet tube (tube ‘a’ in Fig. 1) connected an air pump to the headspace of the nutrient reservoir chamber; an electric timer operated the pump. A second tube (tube ‘b’) ran from close to the base of the reservoir to the culture chamber. To supply nutrient solution to the culture chamber the air pump was switched on, thereby raising the pressure in the headspace of the reservoir and forcing the nutrient solution from the reservoir into the culture chamber. The nutrient solution immersed the root zone temporarily for a total of 15 min every 6 h. After 15 min, the air pump was switched off and the excess nutrient solution flowed back into the reservoir under gravity. Plant material, culture conditions and treatments Establishment of the culture. Nodal cuttings of coffee plantlets (Coffea arabusta) were cultured in Magenta vessels containing hormone‐free MS (Murashige and Skoog, 1962) medium supplemented with 20 g l–1 sucrose. After 4 weeks of culture, regenerated leaves were collected and cut into pieces (10 × 5 mm) and were placed in a Magenta vessel containing modified MS medium (for details, see Afreen et al., 2002). Agar (8 g l–1; Kanto Chemical Co., Tokyo, Japan) was used as a gelling agent and 30 g l–1 sucrose was added in the medium. Cultures were placed in a growth chamber with an air temperature of 23 °C and a 16 h d–1 photoperiod provided by cool‐white fluorescent lamps (National Co., Tokyo, Japan). The ambient CO2 concentration was 400 µmol mol–1 and the photosynthetic photon flux (PPF) was 30 µmol m–2 s–1 measured on the empty culture shelf. Somatic embryos developed within 9–12 weeks of culture. Cotyledonary stage embryos were selected for use as experimental materials. Methods for measuring CO2 concentration and net photosynthetic rate CO2 concentration in the culture headspace was measured using the method described by Afreen et al. (2002). Net photosynthetic rate per plantlet, Pn (mol h–1 per plantlet) was calculated following the method of Fujiwara et al. (1987) using the following equation: Pn = [k × E × V × (Cout – Cin)]/N where k is the conversion factor of CO2 from volume to mole, E is the number of air exchanges per hour of the culture vessel (h–1), V is the air volume of the culture vessel (l), Cin and Cout are the CO2 concentrations (mol mol–1) inside and outside the culture vessel under steady‐state conditions during the photoperiod, and N is the number of plantlets per vessel. Measuring chlorophyll fluorescence, chlorophyll contents and stomatal development A fibre‐optic based chlorophyll fluorimeter (Hansatech, UK) was used to analyse the photochemical activity of the somatic embryos on day 45. In dark‐adapted samples (2 h), the maximal quantum yield of photochemistry through PSII (Φpmax) was calculated from the ratio (Fm – Fo)/Fm (Kitajima and Butler, 1975). The actual quantum yield (Φp) of PSII photochemistry in light‐adapted leaves was calculated from the steady‐state level of chlorophyll fluorescence (Fs) and maximal fluorescence level: Φp = (Fm – Fs)/Fm (Havaux et al., 1991). Samples for chlorophyll measurement were collected on day 45 from plantlets in each of the vessels, and in every case an appropriate mass of leaves was soaked in 80 % ice‐cold acetone for 3 d, centrifuged at 300 r.p.m. for 10 min and measured for light absorption between 400 and 700 nm in a spectrophotometer (Hitachi, Japan). From the absorption curves, the proportion of chlorophyll contents was evaluated according to the formula of Lichtenthaler and Wellburn (1983). For studying stomata (on day 40), the first true leaves of plantlets from different treatments were collected and epidermal peels were taken from the abaxial (lower) surface. Stomata were studied and photographed using a pre‐calibrated digital microscope (Keyence Corporation, Osaka, Japan). Stomatal density (number per mm2) and stomatal length and width (with guard cells) were measured directly under the microscope for each treatment. Stomatal length refers to the distance between the ends of the guard cells, and the width is the distance transversely across them. Plantlet conversion percentage and growth of plantlets Somatic embryos were established as described earlier. After 14 weeks of culture the vessels were placed under a high PPF (100 µmol m–2 s–1) for 14 d (for details, see Afreen et al., 2002). In the first attempts to achieve embryo‐to‐plantlet conversion, 16‐week‐old cotyledonary stage embryos (including 2 weeks pretreatment) were selected for use as experimental material. The embryos were cultured under photoautotrophic conditions (in sugar‐free medium with CO2 enrichment in the culture headspace and high PPF) in three different types of culture systems: (1) Magenta vessel; (2) a modified RITA‐bioreactor with temporary immersion system; and (3) a newly developed bioreactor with a temporary root zone immersion system (TRI‐bioreactor). A mixture of vermiculite and paper pulp (as described by Afreen et al., 2000) was used as supporting medium in the Magenta vessels and in TRI‐bioreactors. For modified RITA‐bioreactors, MS liquid nutrient solution was used and the immersion frequency was 5 min every 6 h, achieved by connecting an air pump through an electric timer. The planting density for all the treatments was 2·4 × 103 plantlets m–2. To provide natural ventilation in the Magenta vessels, two gas‐permeable Millipore filter membranes (pore diameter 0·45 µm) were attached over holes (10 mm diameter) in the lid of the vessels. RITA‐bioreactors were modified by using three gas‐permeable filter membranes with a pore diameter of 0·45 µm to cover the holes (10 mm diameter) in the lid of each of these vessels. The number of air exchanges was 2·6 h–1 in both Magenta vessels and modified RITA‐bioreactors throughout the experiment (measured according to Kozai et al., 1988). In the TRI‐bioreactor, forced ventilation was introduced by using an air pump connected to the headspace of the air distribution chamber; initial flow rates were 50 ml min–1 (1·6 air exchanges h–1) and were gradually increased every 2 or 3 d to maintain the CO2 concentration in the culture headspace at approx. 1000 µmol mol–1. The maximum flow rate was 200 ml min–1 on day 45 (5·8 air exchanges h–1). For all treatments, hormone‐free MS medium was used as a basal medium; sucrose, vitamins and amino acids were eliminated from the formulation to ensure conditions were photoautotrophic. Vessels were placed in a growth chamber with an enriched CO2 concentration (1000–1100 µmol mol–1) and with a PPF of 100 µmol m–2 s–1 during the 16 h photoperiod; ambient relative humidity (RH) was 80–85 % and the air temperature was 23 °C. Experiments were conducted for 45 d and plantlet conversion percentage, fresh and dry masses of the plantlets, and rooting percentage were recorded at harvest. For chlorophyll fluorescence, chlorophyll content and stomatal studies, ten replicates were taken from each treatment. CO2 concentration in the culture headspace was measured throughout the culture period and the net photosynthetic rates were calculated as described earlier. Plantlets were transplanted in a glasshouse (average temperature 29 ± 2 °C; 60–70 % RH) and the survival percentage was recorded after 15 d. Thirty days after transplanting, plants were harvested and fresh and dry masses of surviving plants were recorded. When necessary, statistical significance was determined by one‐ or two‐way ANOVA and least significant difference test (LSD). Experiments were conducted twice. RESULTS AND DISCUSSION Plantlet conversion percentage and growth of plantlets In terms of plantlet conversion percentage under photoautotrophic conditions, the results (Table 1) revealed very distinct differences among plantlets grown in either a temporary root zone immersion bioreactor with forced ventilation, a modified RITA‐bioreactor or Magenta vessels with natural ventilation. In the TRI‐bioreactor, almost 84 % of the embryos successfully produced plantlets, whereas in Magenta vessels and in the modified RITA‐bioreactor the conversion rates were 53 and 20 %, respectively. Considering all parameters of growth and development, it was evident that embryos cultured in the TRI‐bioreactor produced more vigorous shoots and normal roots than those grown in a modified RITA‐bioreactor or in Magenta vessels (Table 1; Fig. 2). Plantlets grown in the TRI‐bioreactor had the highest number of leaves per plantlet (6·6 including cotyledons), 2·5 and 2·1 times more, respectively, than those of plantlets grown in the modified RITA‐bioreactor and Magenta vessels. The leaf area of plantlets grown in the TRI‐bioreactor (2·9 cm2) was 3·6 and 4·8 times that of plantlets grown in the modified RITA‐bioreactor and Magenta vessels, respectively (Table 1). Fresh and dry masses of leaves of plantlets grown in the TRI‐bioreactor were significantly higher than those of plantlets grown in other treatments (2·8 and 3·5 times, respectively, those of the plantlets grown in a modified RITA‐bioreactor, and 4·4 and 6·3 times those of plantlets grown in Magenta vessel; Table 1). Stem fresh and dry masses followed a similar pattern: stem dry mass was 1·7 and 2·5 times that of plantlets grown in a modified RITA‐bioreactor and Magenta vessel, respectively (Table 1). In general, apart from leaf number, the growth attained in a modified RITA‐bioreactor was intermediate between that of plantlets grown in the TRI‐bioreactor and in Magenta vessels; however, most of the growth parameters of plantlets grown in Magenta vessels were only marginally different from those grown in the modified RITA‐bioreactor. The most remarkable difference observed among the treatments was in rooting percentage. In the TRI‐bioreactor, 90 % of plantlets developed roots, 3 times more than plantlets grown in a modified RITA‐bioreactor and 1·6 times more than plantlets in Magenta vessels; some roots of plantlets in the TRI‐bioreactor produced laterals (data not shown). It should be noted that roots that developed in a few plantlets in the modified RITA‐bioreactor remained very small. Plantlets grown in the TRI‐bioreactor had the greatest root fresh and dry masses (6·5 and 12 times those of plantlets grown in the modified RITA‐bioreactor; Table 1). Among the treatments, plantlets cultured in Magenta vessels had a root growth pattern (fresh and dry masses of 4·8 and 0·4 mg, respectively) intermediate between that of plantlets grown in TRI‐ and modified RITA‐bioreactors. CO2 concentrations in the headspace and net photosynthetic rate As plantlets in the TRI‐bioreactor grew, the CO2 concentration in the culture headspace was controlled by increasing the number of air exchanges (Fig. 3A). Thus, despite the increase in biomass, CO2 concentrations were nearly the same throughout the experimental period (approx. 1280 µmol mol–1); in contrast, in Magenta vessels and in the modified RITA‐bioreactor, the number of air exchanges could not be controlled, and were thus 3·3 h–1 throughout the experiment (under natural ventilation). A reduction in CO2 concentration was noted in the culture vessel headspace in the Magenta vessels during the photoperiod: on day 7 the CO2 concentration was 1280 µmol mol–1, while on day 42 it was 1254 µmol mol–1. In the modified RITA‐bioreactor, the CO2 concentration in the headspace fell from 1278 µmol mol–1 on day 7 to 1266 µmol mol–1 on day 42 despite the low air exchange rate; possible reasons for this low consumption of CO2 by plantlets include: (1) total CO2 consumption is low due to the small size of plantlets; (2) chlorophyll contents of the plantlets are lower than those of plantlets in other treatments; and most importantly (3), as the chlorophyllous plant material remained moist almost all the time due to immersion of whole plantlets every 6 h and the high humidity in the culture headspace, these plantlets were probably virtually unable to fix any CO2 from the atmosphere for in vitro metabolism. As expected, among the treatments the highest net photosynthetic rate was observed in plantlets grown in the TRI‐bioreactor (Fig. 3B). The result clearly shows that in this treatment, the forced ventilation system provided the best conditions throughout the experiment for the assimilation of CO2. As a result, the net photosynthetic rate, which is a closer reflection of normal in vitro metabolism, was 8·3‐ and 3·2 times greater than that of plantlets grown in the modified RITA‐bioreactor and in Magenta vessels, respectively, on day 42 (Fig. 3B). Chlorophyll contents, chlorophyll fluorescence and stomatal development In general, the highest chlorophyll content based on the fresh mass of leaves was observed in plantlets grown in the TRI‐bioreactor (Fig. 4A and B). Chloro phyll a and b contents were 606 and 241 µg g–1 fresh mass, respectively, in plantlets grown in the TRI‐bioreactor, 2 and 1·6 times those of leaves of plantlets grown in the modified RITA‐bioreactor. In the case of Magenta vessels, chlorophyll a and b contents of leaves were intermediate between those of plantlets grown in TRI‐ and modified RITA‐bioreactors. The potential activity of PSII (Φpmax), as estimated in the dark, was nearly the same in leaves of plantlets grown in the TRI‐bioreactor (Φpmax = 0·89) and in Magenta vessels (Φpmax = 0·83) (Fig. 4C); in contrast, Φpmax was low in leaves of plantlets grown in a modified RITA‐bioreactor (0·76). A similar pattern was observed for actual photochemical efficiency of PSII (Φp) (Fig. 4D), which is known to be a good estimate of the quantum yield of photosynthetic electron transport (Genty et al., 1989, 1992). An increase in the quantum yield for electron transport was noted in leaves of plantlets grown in both the TRI‐bioreactor (Φp reaching 0·35) and in Magenta vessels (Φp = 0·32). As noted earlier, due to their low PSII activity, leaves of plantlets grown in the modified RITA‐bioreactor exhibited comparatively lower electron transport activity (Φp = 0·25) than that of plantlets in the other two treatments (Fig. 4D). Microscopy highlighted the differences among treatments with respect to stomatal density (Fig. 2E–G), which was highest in leaves of plantlets grown in the TRI‐bioreactor (8·3 mm–2 leaf area) followed by those of plantlets from the modified RITA‐bioreactor (7·5 mm–2 leaf area) (Fig. 4E). Compared with the other treatments, stomatal density was lowest in leaves of plantlets grown in Magenta vessels (5·9 mm–2 leaf area) (Fig. 4E). Average stomatal length was nearly the same in leaves of all three treatments (Fig. 4F). The most noticeable feature was that some stomata that developed in the leaves of plantlets grown in the modified RITA‐bioreactor were open wide (Fig. 2G), while others were distorted or still morphologically immature. It is possible that these stomata may not function properly, although no specific attempt was made to investigate this in the present study. Ex vitro survival and growth After transplanting under glasshouse conditions, a similar trend was noted in terms of survival percentage and growth (Fig. 2M–O). Ex vitro survival, which was recorded on day 15 of transplanting, was highest (89 %) in plantlets grown in the TRI‐bioreactor. Plantlets grown in Magenta vessels had a survival percentage of 67 %, although their growth was much slower than that of plants grown in the TRI‐bioreactor. When plantlets from the modified RITA‐bioreactor were transferred ex vitro only 33 % survived. In terms of ex vitro growth, it was noticeable that plants from the TRI‐bioreactor exhibited much faster growth (Fig. 2M) and, as a consequence, after 30 d of transplanting almost all the growth parameters were significantly greater than those of plants grown in modified RITA‐bioreactors and Magenta vessels (Fig. 5). In the TRI‐bioreactor treatment, leaf number and leaf area were 2·7 and 2·8 times greater, respectively, than those of plantlets from the modified RITA‐bioreactor, and 2·0 and 2·7 times greater, respectively, than those of plants from the Magenta vessel on day 30 after transplanting (Fig. 5A and B). Similarly, leaf and root fresh masses were also enhanced and were 69 and 19 mg per plant, respectively, in the TRI‐bioreactor treatment, compared with 19 and 7·3 mg per plant in the Magenta vessel treatment and 25 and 2·4 mg per plant in plantlets from the modified RITA‐bioreactor (Fig. 5C and E). During the present study, it became increasingly apparent that the vigorous growth (Fig. 5F) and higher survival percentage observed in plants from the TRI‐bioreactor could be the result of many environmental and physiological conditions during the in vitro culture period: for example, the relative humidity under forced ventilation was lower (85–90 %) than that in the modified RITA‐bioreactor (95–99 %) or in Magenta vessels (≤95 %). Smith et al. (1992) suggest that reducing the relative humidity in the culture headspace could improve resistance to wilting of micropropagated grapevine. In a previous experiment (Zobayed et al., 2000), we found that lowering the relative humidity in the culture headspace by introducing forced ventilation can increase the deposition of epicuticular wax on the leaf surface, which can, in turn, prevent water loss after transplanting and thus increase the chance of survival and subsequent growth. The results provide clear evidence that Magenta vessels and modified RITA‐bioreactors gave the lowest growth both in and ex vitro in terms of plantlet conversion from cotyledonary stage embryos under photoautotrophic conditions. RITA‐bioreactors are normally used for the development of plantlets from embryogenic cell suspension cultures using a sugar‐containing medium. In addition, this vessel is also claimed to be suitable for embryo‐to‐plantlet development without handling the plant material (Berthouly and Etienne, 1999). However, our results show that for embryo‐to‐plantlet conversion under photoautotrophic conditions, the use of a modified RITA‐bioreactor is less effective at promoting growth (shoot and root) compared with the newly developed TRI‐bioreactor. This is probably because in the modified RITA‐bioreactor the plant material is either never completely dry or it takes a long time to dry out after each immersion with nutrient solution because the humidity inside the vessel is normally high (95–99 %). Thus, there is a thin layer of water surrounding the plant material which acts as a boundary layer, impeding the exchange of gases between the plant and the surrounding environment, and possibly preventing CO2 fixation by green plant material which is vital for the photoautotrophic growth of embryos. In the case of conventional photomixotrophic systems, the medium contains sugar and therefore the lack of air exchange may not have such serious consequences as it does for plants that depend on CO2 in the atmosphere for their growth (photoautotrophy). Again, it should be noted that the RITA‐bioreactor system has not been developed for culturing plant material under photoautotrophic conditions, although in the present study the vessel was modified by attaching three gas permeable filter membranes; the same was done for Magenta vessels. Furthermore, the results have also established that in the TRI‐bioreactor plantlets not only exhibited the best growth, but they were physiologically normal, survived well and grew faster ex vitro. In general, the following steps are necessary for embryo‐to‐plantlet conversion in conventional somatic embryogenesis systems: (1) embryo selection and transfer on the germination medium; (2) germinated and rooted plantlet selection and transfer to soil; (3) acclimatization (Gupta et al., 1993). Each of these steps is usually time consuming, especially for mass propagation, and labour intensive because the embryos are selected individually, in most cases by hand under a stereomicroscope. The use of machine vision (Cazzulino et al., 1990) and image analysis (Harrell and Cantliffe, 1991) is still limited, and can be used only to classify and sort the embryos (step 1, above). These selected embryos are transferred onto a semi‐solid medium for germination and plantlets with an epicotyl are then selected by hand and transferred to soil for acclimatization and growth. However, we were able to simplify these stages in our experiment by using the TRI‐bioreactor. Although in our system cotyledonary stage embryos were selected by hand, it is possible to select these embryos using automation, and once the embryos have been transferred to the TRI‐bioreactor, germination, shoot and root development, and acclimatization take place in the same bioreactor without any need to handle the plant material or change the culture medium. Therefore, the new system should substantially reduce production costs, and the density limitations encountered in Magenta vessels and in the modified RITA‐bioreactor can also be overcome by increasing the number of cells in the culture cell tray. ACKNOWLEDGEMENT We are grateful for financial support from the Japanese Society for the Promotion of Science (JSPS) Research‐for‐the‐Future Program. View largeDownload slide Fig. 1. Schematic diagram of the temporary root zone immersion (TRI) bioreactor with forced ventilation system. View largeDownload slide Fig. 1. Schematic diagram of the temporary root zone immersion (TRI) bioreactor with forced ventilation system. View largeDownload slide Fig. 2. A, Coffee somatic embryos regenerated from leaf discs after 14 weeks of culture under low light (30 µmol m–2 s–1) followed by 2 weeks under high light (100 µmol m–2 s–1) (×0·5). B–D, 45‐d‐old plantlets developed from cotyledonary stage embryos under photoautotrophic conditions in a temporary root zone immersion (TRI) bioreactor (B, ×0·2), a Magenta vessel (C, ×0·7) and a modified RITA‐bioreactor (D, ×0·2). E–G, Stomata from the abaxial (lower) surface of the first true leaves of plantlets developed photoautotrophically in a TRI‐bioreactor (E), a Magenta vessel (F) and a modified RITA‐bioreactor (G). H and I, Individual plantlets immediately before transplanting ex vitro grown in a TRI‐bioreactor (H) and a Magenta vessel (I). J–L, Root development of plantlets grown in a TRI‐bioreactor (J), a Magenta vessel (K) and a modified RITA‐bioreactor (L). M–O, On day 30 after transplanting plantlets previously grown in a TRI‐bioreactor (M), a Magenta vessel (N) and a modified RITA‐bioreactor (O). View largeDownload slide Fig. 2. A, Coffee somatic embryos regenerated from leaf discs after 14 weeks of culture under low light (30 µmol m–2 s–1) followed by 2 weeks under high light (100 µmol m–2 s–1) (×0·5). B–D, 45‐d‐old plantlets developed from cotyledonary stage embryos under photoautotrophic conditions in a temporary root zone immersion (TRI) bioreactor (B, ×0·2), a Magenta vessel (C, ×0·7) and a modified RITA‐bioreactor (D, ×0·2). E–G, Stomata from the abaxial (lower) surface of the first true leaves of plantlets developed photoautotrophically in a TRI‐bioreactor (E), a Magenta vessel (F) and a modified RITA‐bioreactor (G). H and I, Individual plantlets immediately before transplanting ex vitro grown in a TRI‐bioreactor (H) and a Magenta vessel (I). J–L, Root development of plantlets grown in a TRI‐bioreactor (J), a Magenta vessel (K) and a modified RITA‐bioreactor (L). M–O, On day 30 after transplanting plantlets previously grown in a TRI‐bioreactor (M), a Magenta vessel (N) and a modified RITA‐bioreactor (O). View largeDownload slide Fig. 3. A, Carbon dioxide concentrations in the culture headspace of a TRI‐bioreactor, Magenta vessel and modified RITA‐bioreactor; B, net photosynthetic rates of coffee plantlets grown in a TRI‐bioreactor, Magenta vessel and in a modified RITA‐bioreactor. View largeDownload slide Fig. 3. A, Carbon dioxide concentrations in the culture headspace of a TRI‐bioreactor, Magenta vessel and modified RITA‐bioreactor; B, net photosynthetic rates of coffee plantlets grown in a TRI‐bioreactor, Magenta vessel and in a modified RITA‐bioreactor. View largeDownload slide Fig. 4. Chlorophyll a (A) and b (B) contents, chlorophyll fluorescence (C and D), stomatal density (E) and stomatal length (F) of leaves of 45‐d‐old coffee plantlets grown from cotyledonary stage embryos under photoautotrophic conditions. Significant difference between treatments at P ≤0·05 indicated by a, b, c was determined by Student–Newman–Keuls test. View largeDownload slide Fig. 4. Chlorophyll a (A) and b (B) contents, chlorophyll fluorescence (C and D), stomatal density (E) and stomatal length (F) of leaves of 45‐d‐old coffee plantlets grown from cotyledonary stage embryos under photoautotrophic conditions. Significant difference between treatments at P ≤0·05 indicated by a, b, c was determined by Student–Newman–Keuls test. View largeDownload slide Fig. 5. Growth parameters of 30‐d‐old coffee plantlets after transplanting ex vitro. A, Leaf number; B, leaf area; C, leaf fresh mass; D, stem fresh mass; E, root fresh mass; F, total increase of fresh mass after transplanting ex vitro. Significant difference between treatments at P ≤0·05 indicated by a, b, c was determined by Student–Newman–Keuls test. View largeDownload slide Fig. 5. Growth parameters of 30‐d‐old coffee plantlets after transplanting ex vitro. A, Leaf number; B, leaf area; C, leaf fresh mass; D, stem fresh mass; E, root fresh mass; F, total increase of fresh mass after transplanting ex vitro. Significant difference between treatments at P ≤0·05 indicated by a, b, c was determined by Student–Newman–Keuls test. Table 1. Plantlet conversion percentage and growth of plantlets from cotyledonary embryos of coffee grown photoautotrophically in a temporary root zone immersion bioreactor (TRI‐bioreactor), a modified RITA‐bioreactor and in Magenta vessels for 45 d Treatment  Leaf number  Leaf area (cm2)  Leaf f. wt (mg)  Leaf d. wt (mg)  Stem f. wt (mg)  Stem d. wt (mg)  Root f. wt (mg)  Root d. wt (mg)  Percentage rooting (%)  Plantlet conversion (%)  TRI‐bioreactor  6·6 ± 1·4a  2·9 ± 1·2a  57 ± 19a  8·2 ± 2a  27 ± 7a  3·7 ± 1a  11 ± 7a  1·2 ± 0·7a  90 ± 9a  84 ± 1a  RITA‐ bioreactor  2·6 ± 1·2b  0·8 ± 0·3b  20 ± 7b  2·3 ± 1b  18 ± 9b  2·1 ± 1a,b  1·7 ± 0·1c  0·1 ± 0c  29 ± 2c  20 ± 1c  Magenta vessel  3·14 ± 0·9b  0·6 ± 0·4b  12·9 ± 8c  1·3 ± 0·8b  13·6 ± 8b  1·5 ± 1·0b  4·8 ± 2·1b  0·4 ± 0·1b  57 ± 5b  53 ± 6b  Treatment  Leaf number  Leaf area (cm2)  Leaf f. wt (mg)  Leaf d. wt (mg)  Stem f. wt (mg)  Stem d. wt (mg)  Root f. wt (mg)  Root d. wt (mg)  Percentage rooting (%)  Plantlet conversion (%)  TRI‐bioreactor  6·6 ± 1·4a  2·9 ± 1·2a  57 ± 19a  8·2 ± 2a  27 ± 7a  3·7 ± 1a  11 ± 7a  1·2 ± 0·7a  90 ± 9a  84 ± 1a  RITA‐ bioreactor  2·6 ± 1·2b  0·8 ± 0·3b  20 ± 7b  2·3 ± 1b  18 ± 9b  2·1 ± 1a,b  1·7 ± 0·1c  0·1 ± 0c  29 ± 2c  20 ± 1c  Magenta vessel  3·14 ± 0·9b  0·6 ± 0·4b  12·9 ± 8c  1·3 ± 0·8b  13·6 ± 8b  1·5 ± 1·0b  4·8 ± 2·1b  0·4 ± 0·1b  57 ± 5b  53 ± 6b  Each value represents a mean ± s.d. of 20 replicates. Means within a column followed by different superscripts are significantly different at P < 0·05 by the least significant difference test. All parameters were significant at P < 0·01 (ANOVA). View Large References Afreen F , Zobayed SMA, Kozai T. 2002. Photoautotrophic culture of Coffea arabusta somatic embryos: photosynthetic ability and growth of different stage embryos. Annals of Botany  90: 11–19. Google Scholar Afreen F , Zobayed SMA, Kubota C, Kozai T, Hasegawa O. 2000. A combination of vermiculite and paper pulp supporting material for the photoautotrophic micropropagation of sweet potato. Plant Science  157: 225–231. Google Scholar Aitken‐Christie J , Kozai T, Takayama S. 1991. Automation in plant tissue culture–general introduction and overview. In: Aitken‐Christie J, Kozai T, Smith L, eds. Automation and environmental control in plant tissue culture. Dordrecht, The Netherlands: Kluwer Academic Publishers, 1–18. Google Scholar Berthouly M , Etienne H. 1999. Somatic embryogenesis of coffee. In: Jain SM, Gupta PK, Newton RJ, eds. Somatic embryogenesis in woody plants, Volume 5. Dordrecht, The Netherlands: Kluwer Academic Publishers, 259–287. Google Scholar Cazzulino D , Pederson H, Chin CK. 1990. Bioreactors and image analysis for scale‐up and plant propagation. In: Vasil IK, ed. Scale‐up and automation in plant propagation. New York: Academic Press, 147–175. Google Scholar Fujiwara K , Kozai T, Watanabe I. 1987. Fundamental studies of environments in plant tissue culture vessels. (3) Measurement of carbon dioxide gas concentration in closed vessels containing tissue cultured plantlets and estimates of net photosynthetic rates of plantlets. Journal of Agriculture Meteorology  43: 21–30. Google Scholar Genty B , Briantais JM, Baker NR. 1989. The relationship between the quantum yield of photosynthetic electron transport and quenching of chlorophyll fluorescence. Biochimica et Biophysica Acta  990: 87–92. Google Scholar Genty B , Goulas B, Dimon B, Peltier G, Moya I. 1992. Modulation of efficiency of primary conversion in leaves, mechanisms involved at PSII. In: Murata N, ed. Research in photosynthesis IV. Amsterdam: Kluwer Academic Publishers, 603–610. Google Scholar Gupta PK , Timmis R, Carlson WC. 1993. Somatic embryogenesis: A possible tool for large scale propagation of forestry species. In: Soh WY, Liu JR, Komamine A, eds. Advances in development biology and biotechnology of higher plants. Proceedings of the First Asia‐Pacific Conference on Plant Cell and Tissue Culture, Tacjon, Korea, The Korean Society of Plant Tissue Culture, 18–37. Google Scholar Harrell RC , Cantliffe DJ. 1990. Automated evaluation of somate embryogenesis in sweet potato by Machine Vision. In: Vasil IK, ed. Scale‐up and automation in plant propagation. New York: Academic Press, 179–195. Google Scholar Havaux M , Strasser RJ, Greppin H. 1991. A theoretical and experimental analysis of the qp and qN coefficients of chlorophyll fluorescence quenching and their relation to photochemical and nonphotochemical events. Photosynthetic Research  27: 41–55. Google Scholar Kitajima M , Butler WL. 1975. Quenching of chlorophyll fluorescence and primary photochemistry by dibromothymoquinone. Biochimica et Biophysica Acta  376: 105–111 Google Scholar Kozai T , Koyama Y, Watanabe I. 1988. Multiplication of potato plantlets in vitro with sugar free medium under high photosynthetic photon flux. Acta Horticulturae  230: 121–127. Google Scholar Lichtenthaler HK , Wellburn AR. 1983. Determinations of total carotenoids and chlorophylls a & b of leaf extracts in different solvents. Biochemical Society Transactions  11: 591 – 592. Google Scholar Murashige T , Skoog F. 1962. A revised medium for rapid growth and bioassays with tobacco tissue culture. Physiologia Plantarum  15: 473–497. Google Scholar Redenbaugh K , Walker K. 1990. Role of artificial seeds in alfalfa breeding. In: Bhojwani SS, ed. Plant tissue culture. Application and limitations. New York: Elsevier Publishing, 102–135. Google Scholar Redenbaugh K , Fuji JA, Slade D. 1988. Encapsulated plant embryos. In: Biotechnology in Agriculture. Great Britain Agriculture Food Research Council, 226–229. Google Scholar Smith EF , Gribaudo I, Roberts AV, Mottley J. 1992. Paclobutrazol and reduced humidity improve resistance to wilting of micropropagated grapevine. HortScience  27: 111–113. Google Scholar Zobayed SMA , Afreen F, Kubota C, Kozai T. 2000. Water control ability of Ipomoea batatas grown photoautotrophically under forced ventilation and photomixotrophically under natural ventilation. Annals of Botany  85: 603–610. Google Scholar Author notes 1Department of Bioproduction Science, Chiba University, Matsudo, Chiba 271‐8510, Japan

Journal

Annals of Botany – Oxford University Press

Published: Jul 1, 2002

There are no references for this article.