2021 Volume 86 Issue 3 Pages 201-206
“Híbrido de Timor” (HT) ‘CIFC 4106’ is a natural allotriploid formed from Coffea arabica×C. canephora, which has been used to elucidate morphogenic in vitro responses and to regenerate new individuals, owing to its resistance to coffee pathogens and its semifertile condition. However, seedlings have not been efficiently regenerated from somatic embryogenesis hitherto. This study aimed to adapt a new indirect somatic embryogenesis procedure for HT ‘CIFC 4106,’ and to evaluate the genetic stability of plantlets regenerated. Leaf explants were inoculated in semisolid (M1) and liquid (M2) media for friable calli induction. During six subcultures (three months), the semisolid system yielded more friable calli, with visually greater cell mass. Subsequently, the friable calli of the two systems were randomly transferred to semisolid (M3) or liquid (M4) media for somatic embryo regeneration. The highest somatic embryo regeneration rate was observed in M1–M4. All recovered plantlets showed stable allotriploidy. We reinforce the relevance of adjusting in vitro conditions for indirect somatic embryogenesis establishment to provide plantlets of the different Coffea germplasms. In addition, the proposed indirect somatic embryogenesis protocol in a liquid system enables the propagation of HT ‘CIFC 4106’ plantlets, overcoming the seminal propagation barriers.
Genetic, epigenetic, and physiological aspects of the explant donors as well as in vitro conditions can influence the indirect somatic embryogenesis (ISE) establishment, rendering it a complex morphogenic pathway. Thus, procedures have been described and continuously modified to establish ISE and to accelerate the induction and proliferation stages of embryogenic cells, as well as the regeneration of SE and seedlings for different taxa, including the genus Coffea (van Boxtel and Berthouly 1996, Samson et al. 2006, Loyola-Vargas and Ochoa-Alejo 2016). ISE in Coffea has been described mainly for C. arabica L. and C. canephora Pierre ex Froehner, due to their economic relevance (Staritsky 1970, van Boxtel and Berthouly 1996, Santana et al. 2004). Other Coffea species have been propagated in vitro owing to their relevance as genetic resources for breeding programs, such as “Híbrido de Timor” (HT, C. arabica×C. canephora).
HT ‘CIFC 4106’ is a natural allotriploid with 2C=2.10 pg and 2n=3x=33 chromosomes, formed from the crossing between C. arabica and C. canephora (Clarindo et al. 2013). HT ‘CIFC 4106’ plants are used in Coffea breeding programs as genetic sources of resistance to economically important diseases and pests (Romero et al. 2014). The seminal propagation of this hybrid is restricted. Therefore, the development of an in vitro vegetative propagation protocol is necessary to increase the regeneration of SE and plantlets (Sattler et al. 2016). During ISE establishment, HT ‘CIFC 4106’ shows the rapid and uniform morphogenic response in the induction of friable calli at 15 days, with a mean of 72.4% responsive explants in the semisolid system (Sanglard et al. 2019). However, the regeneration rate of mature cotyledonary somatic embryos (MCSE) is relatively low, with a mean of 16.2% at 180 days, similar to C. canephora (14.3%) and lower than C. arabica (60.1%) and C. eugenioides (30.4%; Sanglard et al. 2019). Thus, it is necessary to develop new protocols and strategies to improve the regeneration of SE in HT ‘CIFC 4106.’
In vitro responses are influenced by the chemical and physical conditions of the tissue culture medium, such as the type, concentration, and combination of growth regulators, inorganic salts, vitamins, carbon and energy sources, organic compounds, pH, gelling agents, and photoperiod (Staritsky 1970, van Boxtel and Berthouly 1996, Papanastasiou et al. 2008). Liquid systems allow for a faster morphogenic response compared to semisolid ones owing to the increased capacity of the cultured cells to assimilate the tissue culture media (Jones and Petolino 1988, van Boxtel and Berthouly 1996, Papanastasiou et al. 2008). In C. arabica and C. canephora, cell aggregates suspended in liquid systems show increased uptake of mineral elements and growth regulators, and consequently greater metabolic activities as well as cell proliferation and SE regeneration responses (van Boxtel and Berthouly 1996, Papanastasiou et al. 2008). Several advantages have led liquid systems to gradually replace semisolid ones: reduced culture medium preparation costs; avoidance of gelling agent impurities; and increased efficiency in the acclimatization of the seedlings to the ex-vitro environment (Savio et al. 2012). However, a recurrent problem in ISE, especially in a liquid system, is the occurrence of somaclonal variation in plantlets regenerated in vitro, promoted by genetic changes such as euploidy, aneuploidy, and DNA sequence alterations (Muzamli et al. 2017).
Nevertheless, in Coffea, the liquid system is only used for cell aggregate proliferation and SE regeneration (van Boxtel and Berthouly 1996). This system is generally consortium or dependent on the semisolid system during the initial stage of friable calli induction. To date, no studies regarding the impact of the liquid system on the establishment of ISE in the HT ‘CIFC 4106’ have been performed. Thus, the present study proposed to establish and compare ISE in HT ‘CIFC 4106’ in liquid and semisolid systems and to evaluate the ploidy level stability of plantlets regenerated in vitro using flow cytometry and cytogenetic observation.
Leaves of HT ‘CIFC 4106’ were collected from plantlets maintained under in vitro conditions at the Laboratory of Cytogenetics and Plant Tissue Culture, Universidade Federal do Espírito Santo-ES, Brazil. The collected leaves were used as explants for ISE establishment. Six leaf explants of 2 cm2 were inoculated in semisolid (M1) and liquid (M2) media (Table 1) for friable calli induction (van Boxtel and Berthouly 1996, Sanglard et al. 2019). The culture media were sterilized in an autoclave at 121°C and 1.5 atm for 20 min and exchanged monthly. Petri dishes containing 15 mL of medium M1 were kept in the dark at 25±2°C, and Erlenmeyer flasks of 120 mL containing 15 mL of medium M2 were shaken at 100 rpm in the dark at the same temperature. Friable calli formation was evaluated fortnightly and the cell mass was measured after 90 days, totalizing six subcultures (S1 to S6). The experimental design was completely randomized, with twelve replicates each for the semisolid and liquid media. Variables used for statistical comparison of the semisolid and liquid systems were the mean number of responsive explants evaluated over successive subcultures (S1 to S6), and the cell mass of friable calli measured at S6. The analysis of variance (ANOVA) and quadratic regression at 5% were performed with the software R (R Core Team 2016, https://www.R-project.org/).
| Compounds | Culture media | ||||
|---|---|---|---|---|---|
| M1 | M2 | M3 | M4 | M5 | |
| MS (Sigma®) | 2.15 g L−1 | 2.15 g L−1 | 4.3 g L−1 | 4.3 g L−1 | 4.3 g L−1 |
| Gamborg’s B5 vitamins | 10 mL L−1 | 10 mL L−1 | 10 mL L−1 | 10 mL L−1 | 10 mL L−1 |
| Sucrose (Sigma®) | 30 g L−1 | 30 g L−1 | 30 g L−1 | 30 g L−1 | 30 g L−1 |
| L-cysteine (Sigma®) | 0.08 g L−1 | 0.08 g L−1 | 0.04 g L−1 | 0.04 g L−1 | — |
| Malt extract (Sigma®) | 0.4 g L−1 | 0.4 g L−1 | 0.8 g L−1 | 0.8 g L−1 | — |
| Casein (Sigma®) | 0.1 g L−1 | 0.1 g L−1 | 0.2 g L−1 | 0.2 g L−1 | — |
| 2,4-D (Sigma®) | 9.06 µM | 9.06 µM | — | — | — |
| BAP (Sigma®) | 4.44 µM | 4.44 µM | 4.44 µM | 4.44 µM | — |
| GA3 (Sigma®) | — | — | — | — | 2.89 µM |
| Phytagel (Sigma®) | 2.8 g L−1 | — | 2.8 g L−1 | — | 2.8 g L−1 |
| Activated charcoal (Isofar®) | — | — | 4.0 g L−1 | 4.0 g L−1 | — |
M1: Semisolid callus induction medium; M2: liquid callus induction medium; M3: semisolid SE regeneration medium; M4: liquid SE regeneration medium; M5: seedling regeneration medium.
For SE regeneration, 0.2 g of friable calli from each system (M1 and M2) was randomly transferred to the semisolid (M3) or liquid (M4) media (Table 1; van Boxtel and Berthouly 1996, Sanglard et al. 2019). The culture media were exchanged fortnightly. Friable calli from the semisolid induction medium (M1) transferred to M3 and M4 were respectively designated SM-SM (semisolid M1-semisolid M3) and SM-LM (semisolid M1-liquid M4); friable calli obtained from liquid induction media (M2) transferred to M3 and M4 were named LM-SM (liquid M2-semisolid M3) and LM-LM (liquid M2-liquid M4), respectively. Petri dishes containing 15 mL of M3 were maintained in the dark at 25±2°C. Erlenmeyer flasks of 120 mL containing 15 mL of M4 were shaken at 100 rpm in the dark at 25±2°C. MCSE was counted biweekly until 17 subcultures were completed (255 days), then transferred to seedling regeneration medium (M5, Table 1). Test tubes with M5 were maintained at 24±2°C under a 16 : 8 h (light/dark) cycle with light irradiation of 36 µmol m−2 s−1 provided by two fluorescent lamps (20 W, Osram). The experimental design was completely randomized. The mean number of MCSE was the analyzed variable to compare the culture systems (SM-SM, SM-LM, LM-SM, and LM-LM) and subcultures (S1 to S17). This comparison was accomplished through descriptive analysis and quadratic regression at 5% using the software R.
Ploidy and chromosome number stabilityThe ploidy level of the regenerated plantlets was previously assessed by nuclear DNA content measurement via flow cytometry, using the leaf of Solanum lycopersicum L. as internal standard (2C=2.0 pg, Praça-Fontes et al. 2011). For this, nuclei suspensions were prepared from leaves of each plantlet and the explant donor plant, according to the procedures described by Clarindo et al. (2013) and Sanglard et al. (2019). The nuclei suspensions were analyzed in a Partec PAS cytometer (Partec GmbH, Münster, Germany). The chromosome number was determined from root meristems of the recovered plantlets. To achieve this, the roots were excised, individually washed in dH2O for 15 min, treated with 3 µM amiprophos-methyl for 4 h at 30°C, fixed in 3 : 1 methanol : acetic acid, and stored at −20°C. After at least 12 h, the roots were washed in dH2O, then macerated for 2 h at 36°C in an enzymatic pool [4% cellulase (Sigma), 0.4% hemicellulase (Sigma), 1% Macerozyme Onozuka R10 (Yakult), 100% pectinase (Sigma)] diluted in dH2O in the proportion 1 : 8 (enzyme pool : dH2O), washed in dH2O, fixed in 3 : 1 methanol : acetic acid, and stored at −20°C (Clarindo et al. 2013). Slides were prepared by cellular dissociation and air-drying techniques, followed by staining with 5% Giemsa. The metaphases were captured with a digital 12-bit CCD video camera (Olympus) coupled to a photomicroscope (Olympus BX-60).
The number of responsive explants, as determined by leaves exhibiting friable calli, was influenced by the semisolid (M1) and liquid (M2) media during the successive subcultures. First responsive explants were observed at 15 days in the M1 and M2 systems, with mean numbers of 5.75 and 1.42, respectively. The mean number of responsive explants was higher in M1 than in M2 at all evaluated subcultures, presenting friable calli in all explants starting at S2. Differently, the highest mean number of responsive explants in liquid media was 4.5, seen at S5 (Fig. 1). The cell mass of the friable calli gradually increased until S6 for both semisolid and liquid systems but differed statistically between them. The mean value for cell mass in the semisolid system was 1.21 g at 90 days (S6), compared to 0.25 g for the liquid medium. Moreover, the friable calli formed in the liquid system were detached from the leaf explant due to orbital agitation. All friable calli of both systems had a yellowish appearance (Fig. 1).
The mean number of MCSE differed over time among the combinations SM-SM, LM-SM, SM-LM, and LM-LM. Globular, heart- and torpedo-shaped SE in HT ‘CIFC 4106’ calli were initially observed at 20 days in LM-LM. At 45 days (S3), SE was found at globular, heart, torpedo, and cotyledonary stages, with a mean number of 1.29 MCSE. The regeneration rate of SE in LM-LM gradually increased during the successive subcultures, with a mean number of 38.86 MCSE by 255 days (S17, Fig. 2). In turn, the first SE in SM-LM was observed at 150 days (S10), with a mean number of 0.36 MCSE. The regeneration rate of SE in SM-LM increased gradually up to 255 days (S17), with a mean number of 4.64 MCSE. Differently, no SE was recovered in the SM-SM and LM-SM systems. Based on these results, the most suitable combination for SE regeneration of HT ‘CIFC 4106’ was LM-LM, which presented the highest mean number of MCSE in a shorter period. Non-responsive calli in SM-SM and LM-SM were characterized as compact, exhibiting pale yellow color. Contrarily, the responsive calli of the SM-LM and LM-LM were friable and exhibited dark coloration (Fig. 2).
The allotriploidy of the recovered plantlets was determined by DNA ploidy level, with all plantlets showing 2C=2.10 pg, in agreement with the HT ‘CIFC 4106’ explant donor. In addition, the chromosome number of 2n=33 chromosomes (Fig. 3) found for these plantlets confirmed the triploid condition of all regenerants.
In this work, the liquid system proved to be a viable, reproducible, and efficient alternative for the establishment of the complete ISE pathway in HT ‘CIFC 4106.’ This result reinforces the importance of in vitro conditions for ISE in Coffea. In addition, the present study demonstrated a protocol for the relatively large-scale propagation of HT ‘CIFC 4106,’ regenerating a mean number of 38.85 MCSE per friable callus. This mean value is higher than previous data for HT ‘CIFC 4106’: 1.51 MCSE per leaf explant via direct somatic embryogenesis (Sattler et al. 2016) and 6.06 MCSE per friable callus via ISE (Sanglard et al. 2019), both conducted in the semisolid system.
The in vitro morphogenic response varied significantly between the liquid and semisolid systems. Moreover, the calli, presenting a pale yellow and friable appearance, were formed in the relatively short time of two months in both culture systems. This result was already expected, as the time of callogenesis establishment in Coffea has been reported to vary from two months for C. canephora, C. arabica, and HT ‘CIFC 4106’ (van Boxtel and Berthouly 1996, Samson et al. 2006, Sanglard et al. 2019) to six months for C. canephora and C. eugenioides (Sanglard et al. 2019). In addition, the cell mass of the friable calli increased over time, particularly in the semisolid system, evincing cell proliferation. This result corroborates previous studies evaluating friable calli multiplication in C. arabica which observed the highest induction of friable calli in the semisolid system (Albarran et al. 2005). In contrast, Teixeira et al. (2004) found no significant difference in the multiplication of friable calli of C. arabica in liquid and semisolid systems.
The SM-SM, SM-LM, LM-LM, and LM-SM combinations influenced the SE regeneration across successive subcultures. Both LM-LM and SM-LM exhibited embryogenic responses from friable calli of HT ‘CIFC 4106.’ However, the highest short-term SE regeneration was observed in LM-LM, where friable calli originated and remained in the liquid system. The LM-LM system likely promoted an increased assimilation capacity due to the greater accessibility of the cultured cells to the compounds of the tissue culture medium (Jones and Petolino 1988, van Boxtel and Berthouly 1996, Papanastasiou et al. 2008). For instance, cytokinins are more effective in the liquid system because they are not conjugated to the gelling agent, which results in more pronounced SE regeneration (Papanastasiou et al. 2008). In addition, the osmotic potential may also have influenced the in vitro response in HT ‘CIFC 4106,’ since the liquid system had lower osmotic potential compared to the semisolid system and regenerated the highest mean number of MCSE. According to Jeannin et al. (1995), osmotic pressure below 400 mOsm kg−1 H2O is associated with organogenesis, while osmotic pressure above this threshold is associated with somatic embryogenesis. However, no difference was observed like the morphogenic pathway between culture systems in the present study, since all in vitro plantlets were regenerated via ISE.
SM-SM and LM-SM combinations were not effective for ISE establishment in HT ‘CIFC 4106,’ as indicated by the lack of embryogenic response. This can be a consequence of the reduced contact area of the friable calli with the culture medium. In addition, the semisolid condition promoted by gelling agents probably reduces the assimilation capacity (Papanastasiou et al. 2008). Nutrient availability in the semisolid system is limited by the osmotic potential of the culture media (Papanastasiou et al. 2008). The low rate and/or non-response of SE regeneration in the semisolid system was also reported by Sanglard et al. (2019) when comparing in vitro responses of Coffea among the diploids C. canephora and C. eugenioides, the allotriploid HT ‘CIFC 4106’ and the true allotetraploid C. arabica.
The 2C nuclear DNA level of plantlets regenerated in vitro via ISE in LM-LM and SM-LM was identical to that of the explant donor plant. The cytogenetic analysis confirmed that these plantlets remained allotriploid, showing the same chromosome number of 2n=3x=33 ascertained in the explant donor plants. Therefore, the strategies for regeneration of HT ‘CIFC 4106’ SE in a liquid system were reproducible, efficient and safe, since the 2C nuclear DNA content and ploidy level remained unchanged and preserved. This result is in agreement with Sattler et al. (2016), who evaluated ploidy level stability in HT ‘CIFC 4106’ plantlets regenerated in vitro via direct somatic embryogenesis and showed that allotriploidy was conserved in all regenerated plantlets. However, somaclonal variation has been reported in Coffea somatic embryogenesis (Campos et al. 2017). Therefore, other genetic aspects should be further investigated, as mutations in nuclear and organellar genes and mobile element activation.
We would like to thank Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq, Brasília-DF, Brazil; grant: 443801/2014-2), Fundação de Amparo à Pesquisa do Espírito Santo (FAPES, Vitória-ES, Brazil; grants: 65942604/2014 and 82/2017), and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES, Brasília-DF, Brazil) for financial support.
