Shoot proliferation, leaf anatomy and pigment content of Eugenia dysenterica growing in conventional and natural ventilation systems

Submitted on June 17 , 2019 and accepted on September 9 , 2019. 1 This work is part of the first author’s Master Dissertation. 2 Universidade Federal de Goiás, Escola de agronomia, Goiânia, Goiás, Brazil. andreiaac2@gmail.com; liviacristy@gmail.com 3 Universidade Federal de Goiás, Instituto de Ciências Biológicas, Goiânia, Goiás, Brazil. leticia.icb.ufg@gmail.com; elienaibio@gmail.com; nauanyssales@gmail.com; stsibov@gmail.com *Corresponding author: andreiaac2@gmail.com Shoot proliferation, leaf anatomy and pigment content of Eugenia dysenterica growing in conventional and natural ventilation systems


INTRODUCTION
Eugenia dysenterica (Mart.) DC. (popularly known as cagaiteira) is a fruit tree species that occurs in Cerrado regions (Brazilian savanna) and has the potential for economic use. Exploited extractively, its main use is in the commercialization of its by-products such as ice cream, sweets and juices (da Silva et al., 2017). The species also possesses medicinal properties, being used as an antioxidant (Takao et al., 2015); a laxative, from fruit consumption (Silva et al., 2015); and antidiarrheal, from leaf extract consumption (Galheigo et al., 2016), among others (Lima et al., 2010;Lima et al., 2011;Souza et al. 2012).
The period of fruiting of E. dysenterica coincides with the beginning of the first rains after drought in Brazilian savannah regions. However, the seed viability period is considered short (approximately 50 days), which characterizes the seed recalcitrance. (Farias Neto et al., 1991;Andrade et al., 2003). Thus, the conventional production of seedlings of this species at any time of the year becomes infeasible, and requires alternative methods of vegetative propagation, such as micropropagation.
Micropropagation is a set of techniques that provides large-scale seedling production. If properly applied, micropropagation results in uniform individuals very similar to those from seeds and is up to a thousand times more productive than conventional propagation techniques. Rev. Ceres, Viçosa, v. 66, n.5, p. 363-371, sep/oct, 2019 However, the application of conventional propagation to woody plants remains limited by the high cost of production, which is largely due to relatively low growth rates, significant losses due to contamination, lack of rooting and low percentages of in vitro plant survival after transfer to ex vitro (Kozai & Kubota 2001).
Thus, in addition to developing micropropagation methods, it is also necessary to improve these methods as a way to guarantee the propagation of plant material at any time of the year. Studies on propagation techniques for the native wood species of Brazil (including E. dysenterica) are still scarce . This kind of propagation is a way to overcome difficulties in seed propagation and the cloning of genotypes of native forest species, thus making it possible to use the species for commercial purposes, as well as to help the conservation of forest genetic resources (Hawkes et al., 2012). Considering the unbridled deforestation of the Cerrado domain (Brazilian savanna) (Brandão et al., 2017), it is necessary to develop in vitro propagation protocols for this species in order to obtain in vitro conservation protocols in the future.
In the "heterotrophic" conventional micropropagation system, seedlings are stored in sealed flasks with low light intensity and no internal ventilation. Plants grown under this regime have several morphological disorders caused by the high relative humidity and the accumulation of gases inside the flasks. These morphological disorders include anatomical changes such as deformed stomata. Studies describe that the reduction of gas exchange in the culture flask may cause stomatal abnormalities, which may compromise, the pigment content as well as stomatal conductance. (Monja-Mio et al., 2015;Gago et al., 2016). Other disorders related to the conventional system include: hyperhydricity, reduced leaves, premature death of explants and, consequently, high losses in acclimatization (Kozai & Kubota, 2005;Hazarika, 2006). Different cultivation systems have been used to increase the seedling nutritional quality, and the reduction of losses of micropropagated plants. The use of pigment quantification to monitor photosynthetic performance in plants is widespread (Baker, 2008). In the natural ventilation system, gas exchange is added to the system, thus improving the transpiration rate and, consequently, nutrient absorption. It is also known that rates of photosynthesis can be increased and significantly promoted by increasing CO 2 concentrations and by decreasing relative humidity in the culture vessel (Kozai & Kubota, 2005;Saldanha et al., 2012). Considering that E. dysenterica has economic potential (Camilo et al., 2013), it is necessary to develop protocols for in vitro propagation of the species aiming at large-scale seedling production. It also becomes necessary that these seedlings have morphological and nutritional quality, and thus, the improvement of micropropagation techniques is necessary. This work had the objective of developing a shoot proliferation protocol for E. dysenterica, by comparing two in vitro growth conditions: a conventional and a natural ventilation system, and observing leaf anatomy and chlorophyll production.

Disinfection and germination
E. dysenterica fruits were collected in October 2013 from the Coleção de Espécies Frutíferas do Cerrado of the Escola de Agronomia, da Universidade Federal de Goiás. Five hundred seeds were removed from the ripe fruits, washed and placed on absorbent paper for two days to eliminate excess moisture (Duarte, 2006). Seed coats were removed with the aid of a scalpel (Martinotto et al., 2007) and the seeds were washed with running water for 20 minutes.
For decontamination, the seeds were placed in a sterilized flask and placed in a laminar flow chamber, where they were immersed in 70% alcohol for one minute. The seeds were then placed in a sterile flask with 0.5% sodium hypochlorite solution of active chlorine for 20 minutes, after which they were passed through a triple wash in autoclaved distilled water. Once decontaminated, the seeds were inoculated into test tubes containing 20 ml of WPM medium (Lloyd & McCown, 1980) supplemented with 3% sucrose (p/v) and 2.5% (p/v) of Gelsan®, and maintained in a growth room with a temperature of 25 ± 1 ºC, luminosity of 42 µmol.m -2 .s -1 and a photoperiod of 16 h light / 8 h dark for at least ninety days and/or until the plants reached at least 100 mm in height with at least three axillary buds.

Shoot proliferation
An experiment was performed with flasks with sealed lids (conventional system) and flasks with lids that allowed gas exchange with the external environment (natural ventilation system -NV). Each lid of the latter contained a 5.0 mm hole, that was covered with two layers of microporous tape and a layer of thread-sealing tape (PTFE tape). Lateral seals for both types of lids were made with two layers of PVC film (Saldanha et al., 2012).
Under aseptic conditions, axillary and apical buds (0.5 mm diameter) were sectioned and inoculated in WPM medium supplemented with 3% sucrose (w/v) and exposed to treatments of four concentrations of 6-benzylaminopurine (BAP) [0.00 µM (control); 4.44 µÌ; 11.10 µÌ; and 17.76 µM] with thirty-two replicates. The juvenility of the tissue within each concentration of BAP was considered, and the position of the buds relative to the root and caulinar apex of the mother plant was identified. In this sense, each of the following types of buds had eight replicates: 1 -Axillary buds proximate to the root;
The test tubes containing the explants were kept in a growth room under the same conditions of growth described previously for three months. The culture medium with the respective treatments was replaced every fortyfive days. The height of the largest shoot, the number of shoots and the number of leaves were evaluated. When evaluating the height of the largest shoot in axillary buds, the shoot was considered from the axillary bud to the insertion of the last leaf.

Leaf anatomy analysis
In order to evaluate the influence of the conventional and natural ventilation systems on the development of the plants, a study of leaf anatomy was carried out. Leaves from buds type 2 and 3 were used. The material was fixed in FAA [37-40% formaldehyde, glacial acetic acid, and 70% ethanol (1:1:18, v/v/v)] and stored in 70% ethanol until processing. Part of this material was used for diaphanization (Johansen, 1940) and part for embedding in hydroxyethyl methacrylate (Jung's Historesin -Leica) according to the manufacturer's instructions.
The material embedded in Historesin (Leica Historesin-Leica TM ) was used to obtain cross-sections of 9-µm thickness using a Leica RM2245 rotary microtome. These cross-sections were then subjected to toluidine blue (O'Brien et al. 1964) and were mounted in Vitral Incolor 500 Acrilex ® varnish (Paiva et al., 2006). The diaphanized material was placed in 0.1% aqueous safranin for 24 h, and mounted in hydrated glycerin.
The following characteristics were evaluated: total leaf blade thickness, thickness of palisade parenchyma, thickness of spongy parenchyma, stomatal frequency, stomatal index, diameter of the polar axis of the stomata and diameter of the equatorial axis of the stomata. Four replicates were used per treatment, with each replicate being considered a leaf taken from a plant. To determine thicknesses, 10 sections per leaf were used, while for stomatal analysis 20 quadrants per leaf were used.

Chlorophyll and carotenoid contents
Samples of 150 mg of leaves of E. dysenterica resulting from explants by conventional and natural ventilation systems were taken for the analysis of chlorophyll and carotenoids. The leaves were fragmented and conditioned in 2-mL Eppendorfs microcentrifuges. These were then immersed in 1.5 mL of acetone, sealed with Parafilm®, and maintained in the dark under agitation (100 rpm) for 24 h at a temperature of 10 ºC. The cetonic solution was removed from the tubes and subjected to spectrophotometric reading for the quantification of chlorophyll a, chlorophyll b and carotenoids. The calculation of the concentrations of chlorophyll a, chlorophyll b and carotenoids was performed according to the equations of Lichtenthaler (1987). The experimental design was completely randomized with 2 treatments (two types of lids) and 10 repetitions.

Statistical analysis
The data were submitted to analysis of variance with a significance level of 5%. Data analysis was performed using R software (version 3.4.3) (R Core Team, 2017), while graphics were generated using Graph Pad Prism (version 5.01) (Radushev, 2007).

Shoot proliferation
In the conventional system, the best treatment as measured by the variables of height, number of shoots and number of leaves was WPM supplemented with 4.44 µM of BAP with Type 2 buds (means of 42.00 mm, 4.33 shoots and 4.33 leaves, respectively) ( Figure 1). In the natural ventilation system, the best result for the variable of height occurred in the treatment of 11.1 µM of BAP with Type 2 buds (mean 20.66 mm), while for the number of shoots and number of leaves the best treatment was 17.76 µM of BAP with Type 3 buds (means of 6.60 shoots and 9.20 leaves, respectively).
Shoots with lower mean height were reported in the natural ventilation system, compared with the conventional system. In conventional systems, the accumulation of gases such as ethylene occurs within the culture flask. Ethylene can inhibit or promote stem elongation (Suzuki & Kerbauy, 2006). In this second case, it is promoted by cell division, cell elongation, auxin transport, and auxin action. This occurs, for example, in rice, which showed a greater distance between the internodes when the stem was subjected to ethylene (Abeles et al. 1992;Pierik et al. 2006). Another factor to be considered is the fact that the growth rate of E. dysenterica is considered slow relative to other species (Souza et al., 2002;Silveira et al., 2013). It seems that the plants in flasks of the natural ventilation system probably had a tendency to behave in a manner similar to plants in normal conditions of ex vitro growth; the aerial part of the plants enters a state of latency for at least one year, and only after this period does it begin to grow slowly (Silveira et al., 2013).
Reduced or absent leaves were observed in most of the explants in the conventional system, representing 76.37% of all the buds inoculated in this group. In contrast, Rev. Ceres, Viçosa, v. 66, n.5, p. 363-371, sep/oct, 2019 only 26% of the plants in the natural ventilation system did not produce leaves, with an increase of 448.53% in the number of leaves in relation to the number grown in the conventional system. Cytokinins can cause an increase in ethylene inside the culture flasks in heterotrophic systems (Abeles et al., 1992;Chae, 2003), which can cause decrease on leaf limb, senescence (Pan et al., 2016) and a decrease in chlorophyll production . Thus, with the increase in gas exchange inside the culture flasks, the accumulation of ethylene likely did not occur, which favored the increase of leaves.
In the present study, shoots coming from intermediate positions (bud type 2 and 3) had higher means in the analyzed variables, in both the conventional system and the natural ventilation system (Figure 1). In shoot proliferation, juvenile explants generally have a better regeneration response under in vitro conditions than to mature explants (Ahuja, 1993). This is probably due to the fact that the highest concentrations of free auxins in plants are found in the apical meristems of the aerial part and in young leaves, therefore forming a gradient of concentration along the stem. (Taiz & Zeiger, 2010). Thus, the balance between endogenous and exogenous hormones and their interactions, probably contributed to promote positive results in intermediate buds. In addition, in the natural ventilation system with gas exchange, nutrients and growth regulators were probably better absorbed by the plant because of better transpiration (da Silva et al., 2014). In a study involving guava micropropagation, the bud position influenced the shoot proliferation, leading to intermediate buds with a higher percentage of shoots (Shekafandeh & Khosh-Khui, 2008). Similarly, in another study involving passion-fruit, the authors related that plants with intermediate buds also had higher mean plant height; however, the authors did not observe differences in plant leaf number (Faria et al., 2007). Intermediate buds were superior in a study of Corylus avellana L., in which higher percentages of viability were reported, which consequently increased the proliferation of shoots (Hand et al., 2016).

Leaf anatomy
In both systems, leaves of E. dysenterica had a uniseriate epidermis covered by a thin cuticle and stomata on the abaxial face ( Figures. 2a, 2b). Epidermal cells had slightly undulating anticlinal walls (Figures 2a, 2b), on both faces, and dorsiventral mesophyll with a single layer of palisade parenchyma and four to five layers of spongy parenchyma (Figure 2c). Tector trichomes occurred in both treatments. Secretory cavities (Figure 2d) with large lumens and uniseriate epithelium were observed in the mesophyll throughout the extension of the leaf blade.
Eugenia dysenterica cultivated in vitro possesses an organized tissue structure in both evaluated systems. However, plant leaves grown in containers from the conventional system may be thinner and have little differentiated mesophyll (Mohamed & Alsadon, 2010).
In the conventional system, some larger stomata (Figure 2a), called "giant" stomata were reported, with polar and equatorial diameters of approximately 70 x 55 ìm, respectively. These stomata contained fully opened ostioles and were different from the common stomata, as they projected above the other epidermal cells ( Figure  2e). The remaining stomata of this system had mean polar and equatorial diameters of 36 x 28 ìm. In the natural ventilation system, all stomata were located at the same level as the ordinary epidermal cells (Figure 2f). The occurrence of abnormal stomata in micropropagated plants from heterotrophic systems are reported in some species, such as grapevine (Machado & Biasi 2011), apple (Chakrabarty et al., 2006) and eucalyptus (Picoli et al., 2008), In these species, these abnormalities are due to hyperhydricity, which causes chemical damage to the cell wall surrounding the stomata orifice, creating cell walls with less callose, that lack the characteristic orientation of cellulose myofibrils (Vasconcelos et al., 2012). Thus, in culture conditions and factors associated with leaf hyperhydration seem to be linked to stomatal malfunction (remaining open) (Ziv & Ariel, 1994). Junior and Scherwinski-Pereira (2009) reported an increase in the size of the stomata of Tectona grandis L. cultivated in vitro, which may be related to the lack of control in the mechanism of opening and closing of the stomata. Stomata that projected slightly above other epidermal cells were also observed in the leaves of Cymbidium Hort. grown in vitro (Mayer et al., 2008).
The stomatal index and stomatal frequency were greater in the conventional system (Table 1). However, no differences were observed regarding the polar and equatorial diameters of the stomata, or the ratio between the two (Table 1). Additionally, no differences were reported in the thickness of the leaf blade, the palisade parenchyma or the spongy parenchyma (Table 1). For in vitro culture, the decrease in relative humidity with the higher gas exchange in the flask increased the transpiration rate and, consequently, the absorption of water and nutrients by the plant (da Silva et al., 2014). In addition, the high relative humidity inside the culture flask reduces the development of functional stomata, which can lead to losses during acclimatization (Chandra et al., 2010). Further work involving in vitro cultivated E. dysenterica hyperhydricity should be developed in the future to confirm the association of abnormal stomata with this condition.  Although we report the presence of some abnormal stomata in plants from the conventional system, the high values reported for the ratio between polar diameter and equatorial diameter of the stomata (PD/ED ratio) in the two bottle-sealing conditions indicate functional stomata. According to Rocha (2005), the greater the relationship between PD and ED, the more ellipsoid the shape of the stomata, and thus the greater its functionality.
In general, a thin cuticle was observed in both culture systems. The cuticle protects the plant against excessive water loss and excess luminosity of solar radiation. However, micropropagated plants in heterotrophic systems usually have an undeveloped cuticle that is related to the high concentration of moisture inside the culture vessel (Johansson et al., 1992).

Chlorophyll and carotenoid contents
The natural ventilation system promoted an increase of 85.64% in chlorophyll a levels and 74.90% in chlorophyll b levels ( Figure 3). This can be observed in the leaves of the treatments from the natural ventilation system, which presented intense green coloration (Figure 4).
Stomata play a very important role in regulating the entry and exit of water in plants, as well as the exchange of CO 2 (Jones, 1998). Most likely, the pigment accumulation in plants of the natural ventilation system resulted from a more active photosynthetic machinery as the mixotrophic system allows for CO 2 interchanges stimulating photosynthesis (Morison, 1985;Anyia & Herzog, 2004) Chlorophyll increase was also observed in a strawberry (Fragaria x Ananassa) study using a natural ventilation system, a study in which the authors also reported root growth in plants grown under this system (Pandey et al., 2015). In a study using Pfaffia glomerata (Spreng.), the authors reported that the natural ventilation system pigment synthesis tripled compared to the control, which was reflected in the nutritional quality of the plant (Saldanha et al., 2012). The increase in pigment production was also observed in potato explants submitted to the natural ventilation system, in which a significant increase in explant dry mass was also observed compared to potatoes grown in a conventional system (Mohamed & Alsadon, 2010).

CONCLUSION
It is possible to obtain an in vitro shoot proliferation protocol for Eugenia dysenterica with shoots above 20 mm in height and with a mean number of five leaves per stem. The natural ventilation system is superior at the moment in the formation of new shoots with regard to the number of shoots (greater), number of leaves per shoot (more), chlorophyll production and leaf anatomical characteristics.
The type of bud has an effective influence on micropropagation, with positive responses in plants with buds of intermediate position, regardless of the type of system -conventional or natural ventilation.
The occurrence of a greater number of stomata and the presence of stomata considered "giant" in the plants grown in the conventional system is probably associated with the difficulty of transpiration for these plants due to the high humidity accumulated in the internal environment of the flasks.
Our results indicate that the mixotrophic system can be used in future protocols for rooting E. dysenterica, which will contribute to the maintenance of the genetic resources of this species.

AKNOWLEDGEMENTS, FINANCIAL SUPPORT AND FULL DISCLOSURE
The authors thank Capes for their financial support in granting a master's degree.

CONFLICT OF INTEREST
The authors declare that they have no conflict of interest.