Study and experimental material location

The experiments were conducted at the Laboratory of In Vitro Culture of Forest Species, Department of Forestry Sciences (DCF), Federal University of Lavras - UFLA, Lavras, MG. The genetic material used to obtain the explants came from microstumps of Eucalyptus dunnii species donated by the Institute of Research and Forest Studies (IPEF).

For the elongation phase, shoots produced in the in vitro multiplication phase were prepared by isolating four 0.5 cm standard shoots and inoculated under aseptic conditions in glass vials (250 mL capacity) grown for 30 days, containing 50 mL of MS medium (^{Murashige and Skoog, 1962}), added with 6 g.L^-1 agar, 0.05 mg.L^-1 BAP (6-benzylaminopurine - Sigma Co.), and 0.5 mg.L^-1 of indole-3-butyric acid (IBA) (Sigma^®) (^{Trueman et al., 2018}).

In vitro rooting, microcuttings with size about 2 cm, were prepared from elongated buds in the in vitro elongation phase. Microcuttings were inoculated under aseptic conditions, cultured for 30 days in glass vials (250 mL) containing 50 mL of MS medium, with the addition of 6 g.L^-1 agar, 1 mg.L^-1 IBA and 0.5 mg.L^-1 of naphthalene acetic acid (NAA).

The culture medium was prepared using deionized water, and its pH was adjusted to 5.8 ± 0.05 with NaOH (0.1 M) and HCl (0.1 M) prior to autoclaving and agar addition. Autoclaving of the culture medium was performed at a temperature of 127°C and a pressure of approximately 1.5 kgf.cm^-2 for 20 min. The treatments were kept in a growth room at 25 ± 1°C for a photoperiod of 16 h light and irradiance of 40 µmol m^-2.s^-1 (quantified by radiometer, LI-COR^®, LI-250A Light Meter).

DESIGN AND EXPERIMENTAL EVALUATIONS

To evaluate packaging and sucrose variations in in vitro elongation, the experiment was organized in a 3×4 factorial arrangements, with three types of sealing: rigid polypropylene lids without membrane (0/M), polypropylene lids with a hole (1 cm in diameter) covered with a membrane of 1.0 cm² (1/M), polypropylene lids with three holes (1.0 cm in diameter each) covered with a membrane of 1.0 cm² (3/M), and under four different sucrose concentrations (0, 10, 20 and 30 g.L^-1).

As for the experiment in the in vitro rooting phase, 3×4 factorial arrangements were used, with three sealing forms: rigid polypropylene lids without membrane (0/M), polypropylene lids with a hole (1 cm in diameter) covered with a membrane of 1.0 cm² (1/M), polypropylene lids with four holes (1.0 cm in diameter each) covered with a membrane of 1.0 cm² (3/M), and three sucrose concentrations (0, 10, 20 and 30 g.L^-1). The natural ventilation systems were obtained by porous membranes manufactured in the lids of the culture vessels, as those in ^{Saldanha et al. (2012}).

For all experiments, a completely randomized design with thirty replicates, composed of a split-plot with one explant, was used. After the inoculation in the in vitro elongation phase, the following parameters were evaluated: oxidation and vigor (Figure 1), shoot length (> 0.5 cm), mean shoot number per explant (> 0.5 cm), photosynthetic pigment, leaf area, and anatomy.

FIGURE 1 Oxidation and vigor assessments according to a scale of notes. (A) 1 = Null: no oxidation; (B) 2 = Average: reduced oxidation at the base of the explants (medium with grayish tonality; (C) 3 = High: complete oxidation of shoots); (D) 1 = Optimum: induction of shoots with active growth, without apparent nutritional deficiency; (E) 2 = Good: induction of shoots, but with leaves of reduced size; (F) 3 = Low: no induction of shoots and, senescence and death). Bar = 1 cm. 

Leaf area was measured using WinFOLIATM software by the EPSON PERFECTION V700 PHOTO scanner. Five explants from each treatment were used. In each seedling, one area was evaluated and a leaf was removed from the second pair, counted from the apex of the aerial part to the root.

The length of adventitious root was assessed at 30 days (roots observed at the lower end of the vial), as well as the root number from the base of the stem, diameter and length of the largest root (cm).

Photosynthetic pigment analysis

Once the in vitro elongation culture (30 days) under different forms of sealing and sucrose concentrations was over, leaf discs (25 mg of fresh leaf matter) were withdrawn and inoculated into 5 mL of DMSO solution (Sigma aldrich) for 48 hours in the dark (^{Lichtenthaler, 1987}). Sample absorbance was determined by triplicate in a 10 mm quartz cuvette of optical path in a Genesys 10UV spectrophotometer (ThermoScientific, USA). The wavelengths (665, 649 and 480 nm) and the equations for calculating chlorophyll concentrations a, b, and total carotenoids were based on the method described by ^{Wellburn (1994}).

Leaf anatomy

To perform the histological sections, leaves collected from each treatment were maintained in a 70% solution of formaldehyde acetic acid for 48 hours, and then the material was transferred to 70% ethanol (^{Johansen, 1940}). The plant material was dehydrated in an alcoholic-ethylic series in increasing concentrations (80, 90 and 100%) for 30 minutes in each solution (Johansen, 1940), and it was finally stored in 100% alcohol and histresin solution (Leica^®) in a 1:1 ratio in a hot oven (overnight). The embedding was with pure hydroxyethyl methacrylate resin and the cross sections obtained with the manual rotary microtome and knife with a thickness of 7 µm. They were then stained with toluidine blue, mounted on histological slides with stained glass finisher (^{Paiva et al., 2006}) and photomicrographs with a coupled digital camera (AxionCam ERc5s) in a 20x and 40x objective micrometer scale.

Data analysis

Data were processed in R Software, version 3.0.3 (R CORE TEAM, 2018), with the help of the ExpDes package, version 1.1.2 (^{Ferreira et al., 2013}). Treatment means were used to perform the statistical analyses and adjustments of regression equations. Non-parametric variables assessed with a 5% significance test were transformed into arsenic. For the significant variables, Tukey’s test was done at 5% of significance.

Mixotrophism effect on in vitro elongation

Variations according to studied characteristics, different types of sealing and sucrose concentrations used on in vitro elongation were observed at 30 days of culture with the clone of Eucalyptus dunnii (Figure 3). Shoot length, number of shoots per explant and leaf area observed on in vitro elongation showed significant interaction (p <0.05) (packaging and sucrose). Oxidation, explants vigor, and pigment content were factors that acted independently. The regression curves showed polynomial behavior of the second degree.

FIGURE 3 Jackknife test for relative importance of bioclimatic variables. 

Shoot length showed the same behavior as the number of shoots per explant and leaf area regarding sucrose concentrations, where treatment with 10 g.L^-1 achieved the best results (Figures 2A, 2B, and 2I). The number of shoots per explant 1/M (on average 6.15) (Figure 2B) was higher with 3/M (average 3.68 cm) shoots (Figure 2A), and leaf area 3/M (on average 7.01 cm²) (Figure 2I). ^{Costa et al. (2017)}, working with Ochroma pyramidale, verified similar results, obtaining higher averages for length, number of shoots per explant and leaf area using the concentration of 10 g.L^-1 of sucrose. ^{Silva et al. (2017}) report their positive effect on different species, which benefits cultivation and reduces costs, and can be applied to Eucalyptus dunnii.

However, in vitro propagation with the use of porous membranes allows the exchange of gases between the external and internal atmosphere of the flasks through natural ventilation, provided that the concentration of CO₂ is adequate, resulting in increased growth (^{Kozai, 2010}; ^{Martins et al., 2015}). In Capsicum annuum, ^{Batista et al. (2017}) verified higher shoot length and number of leaves per explant in membrane system when compared to conventional system. For Vaccinium ashei Reade, the membrane-based system provided higher shoot length, number of shoots, leaf area and chlorophyll content (^{Hung et al., 2016}).

Oxidation evaluations according to grade scale evidenced lowest averages without sucrose supplementation (1.21) (Figure 2D) and S/M (Figure 2C), but no difference was observed otherwise (p > 0,05). In vitro conditions are stressful to plant growth, and high concentrations of exogenous sugar are a major cause of oxidation during in vitro culture (^{Tisarum et al., 2018}).

FIGURE 2 Characteristics observed on in vitro elongation of Eucalyptus dunnii regarding different sealing (0/M, 1/M and 3/M) and sucrose concentrations (0, 10, 20 and 30 g.L-¹). (A) Shoot length; (B) Number of shoots per explant; (C and D) Shoot oxidation; (E and F) Shoots vigor; (G and H) Pigment content; (I) Leaf area. *Averages followed by the same letter do not differ from each other, by Tukey’s test at 5% probability. 

In contrast, the concentration of 30 g L^-1 sucrose (mean 2.85) (Figure 2F) and 3/M (Figure 2E), although not statistically different (> 0.05), stands out. For Alocasia amazonica, according to ^{Jo et al. (2009}), the concentration of 30 g.L^-1 of sucrose achieved the best results in the morphophysiological aspects and also in explants development.

However, desired results with membrane use may be associated with increased photosynthesis caused by high CO₂ availability. The in vitro culture of Plectranthus amboinicus, an improvement in explant vigor was observed with the use of three membranes in the containers (^{Silva et al., 2017}).

Best results regarding pigment content were obtained with the use of membranes (Figure 2G) and 20 g.L^-1 sucrose (Figure 2H), evidencing the importance of greater gas exchange (CO₂), and an external carbohydrate source in plant growth under these conditions. Sucrose levels had an effect on shoot regeneration and bioactive compounds content in Ajuga multiflora (^{Jeong and Sevanesan, 2018}). According to ^{Yuan et al. (2015}), the biosynthesis of chlorophyll and carotenoids is also regulated by gas exchange and sucrose. Prolonged exposure to high concentrations of exogenously applied sucrose in most plants may result in an inhibition of photosynthesis, which is associated with inhibition of chlorophyll biosynthesis (^{McCarthy et al., 2016}).

Results report several effects of membrane use on plants grown in vitro, such as improved growth and increased content of photosynthetic pigments, mainly by maintaining adequate concentrations of CO₂ to stimulate photosynthesis (^{Saldanha et al., 2012}). ^{Bandeira et al. (2007}) studies with the species Thymus vulgaris and Saldanha et al. (2012) with Pfaffia glomerata, this system improved the in vitro growth and increased the photosynthetic pigment content of the seedlings compared to the conventional system. ^{Rodrigues et al. (2012}) verified a lower number of shoots, photosynthetic pigment content and increased senescence in conventional system, in the Azadirachta indica species when comparing membrane use.

The use of the membrane system and sucrose concentration (10 g.L^-1) was directly implicated on in vitro elongation of Eucalyptus dunnii, being determinant for shoot length, number of shoots per explant, pigment content and leaf area, where a reduction of this component in the medium is evidenced, since the most used in the micropropagation of Eucalyptus is 30 g.L^-1 (^{Brondani et al., 2012}). The lack of carbohydrate source in the culture medium showed the worst results in length, number, vigor, leaf area, and pigment content. In vitro culture plants partially lose autotrophism and, consequently, need an exogenous source of carbohydrates, being the most used sucrose in plant tissue culture (^{Parveen and Shahzad, 2014}).

Anatomical structure of the foliar cells can also be influenced by the packaging system of the containers used on in vitro culture. The literature reports that the natural ventilation system, for instance, makes leaf cells more organized and consequently they form more rustic leaves. As stated by ^{Silva et al. (2014}), natural ventilation provided an increase in the thickness of Cattleya walkeriana leaf mesophyll and stomata, making them more functional. ^{Mohamed and Alsadon (2010}) evaluated the anatomy of potato seedlings cultured in vitro, and found that the use of natural ventilation resulted in thicker leaves with more developed xylem than the conventional system. Results from this study show that membrane-sealing systems provided more developed vascular bundles (Figure 4A to 4C) and reduced intercellular spaces of mesophyll cells (Figure 4D to 4F).

FIGURE 4 Pinus pinea L. bioclimatic habitat suitability maps in Turkey (Current, HadGEM2-ES RCP 4.5-8.5 2050, HadGEM2-ES RCP 4.5-8.5 2070). (0-0.5 unsuitable habitat, 0.5-0.75 suitable habitat, 0.75-1 higly suitable habitat). 

Mixotrophic effect on in vitro rooting

Length, number of shoots, and root diameter acted independently regarding sealing, in face of the sucrose concentrations. These factors were significant (p<0.05) for adventitious rooting. Regression curves showed polynomial behavior of the second degree.

At 30 days on in vitro rooting phase, sealing and sucrose concentration showed the same tendency, with the best results obtained for the treatment with 1/M and 30 g.L^-1 (Figure 5), due to length (Fig. 6A and 6B), shoot number per explant averaging 1.70 and 1.98 (Figure 6C and 6D), and mean root diameter of 0.17 and 0.28 cm (Figure 6E and 6F). In Vaccinium ashei, the ventilation system provided greater length and number of roots (^{Hung et al., 2016}). The use of systems that increase the CO₂ supply to in vitro plants are conditions that can increase plant growth, improve physiological characteristics and facilitate seedlings acclimatization to ex vitro conditions, promoting the development of the photosynthetic apparatus (^{Shin et al., 2014}).

FIGURE 5 Explants of the clone of Eucalyptus dunnii at 30 days in the in vitro rooting phase in the treatment with a membrane (1/M) and 30 g.L-1 of sucrose. (A) Rooted explant; (B) Explant without root. Bar = 1 cm. 

FIGURE 6 Characteristics observed in the in vitro rooting of Eucalyptus dunnii as a function of different sealing (0/M, 1/M and 3/M) and sucrose concentrations (0, 10, 20 and 30 g.L^-¹). (A and B) Root length; (C and D) Number of roots per explant; (E and F) Root diameter; (G) Rooting percentage. *Averages followed by the same letter do not differ from each other, by Tukey’s test at 5% probability. 

Adventitious rooting is a challenge in clonal propagation, with variations from 4 to 46.5% (^{Brondani et al., 2011}; Obserschelp et al., 2015). However, for Eucalyptus dunnii, it found a better response to rooting with 66.66% using 1/M and 30g L^-1 sucrose (Figure 5G). The increase in CO₂ concentration promotes rooting and reduces growth anomalies, as it may improve the photosynthetic rate (^{Cha-Um et al., 2011}). This condition is related to the reduction of water loss due to the deposition of epicuticular wax and to the production of functional stomata (^{Martins et al., 2015}; ^{Hoang et al., 2017}). According to ^{Moreira et al. (2013}), the effects of the use of membranes to increase the gas exchange infer in the reduction of relative humidity, increase of aeration, production of more rustic plants and, as a consequence, greater survival and rooting. The use of sucrose was also considered paramount. Using bioreactors in Populus hybrids, sucrose provided higher percentages of survival and rooting when compared without the carbon source (^{Arencibia et al., 2017}).

In contrast to the aforementioned results, for in vitro cultivation and ex vitro acclimatization of Hevea brasiliensis, better results for root length and number, survival and rooting were obtained without sucrose and flasks that allowed natural ventilation (^{Tisarum et al., 2018}). Sometimes gradual adaptation to the ex vitro condition may be required for plants grown in in vitro mixotrophic systems to undergo autotrophic growth in greenhouse (Perez et al., 2015).