Experiment I - Gingko biloba mini-cuttings

Gingko biloba shoots were collected on October 16, 2017 from ~30 mini-stumps conducted in a mini-garden system (Fig. 1A) at the Forest species propagation laboratory (Embrapa Forests, Colombo-PR, 25°19' S and 49°09' W, 950 m a.s.l.). After collection, the plant material was kept in styrofoam boxes with water to avoid dehydration. The material was then transported to the Macropropagation Laboratory at the research and study group on cuttings (GEPE), located at the Federal University of Paraná (UFPR), in Curitiba - PR (25°44' S and 49°23' W, 920 m a.s.l.). From the shoots of the mini-stumps, stem mini-cuttings were obtained with 4 ± 1 cm length, with a bevel cut in the base and straight cut at the tip, maintaining a leaf in the apical portion (Fig. 1B).

FIGURE 1 A) Mini-stumps, B) mini-cutting, and C) experiment installed in the greenhouse. 

After preparation, mini-cutting bases were immersed in the indole butyric acid (IBA) plant growth regulator in 50% hydroalcoholic solution for 10 sec (^{Bitencourt et al., 2010}), according to the treatments 0 mg L^-1, 1000 mg L^-1, 2000 mg L^-1 and 3000 mg L^-1. The planting was carried out in polypropylene tubes with 53 cm³ capacity, using two substrates separately: vermiculite of fine granulometry and the commercial substrate Tropstrato®. The cuttings were planted at about % the depth of the base and maintained for 60 d in a greenhouse with intermittent mist (temperature of 24°C ± 2°C and 80% relative humidity) (Fig. 1C), located in the Biological Sciences Department of the Federal University of Paraná.

The experiment was arranged in a completely randomized design with a 2 x 4 factorial scheme (substrates x IBA concentrations) for a total of 8 treatments with 4 replicates and 14 mini-cuttings per experimental unit, totaling 448 mini-cuttings.

After 60 d, the following variables were evaluated: percentage of rooted mini-cuttings (mini-cuttings that emitted roots of at least 1 mm in length) (Fig. 2A), number of roots per mini-cutting, average length of three major roots per mini-cutting (cm), percentage of mini-cuttings with calluses (mini-cuttings alive, without roots, with formation of undifferentiated cell mass in the base) (Fig. 2B), percentage of live mini-cuttings (mini-cutting without calluses and without roots) (Fig. 2C), percentage of dead mini-cuttings (mini-cuttings with necrotic tissues); percentage of mini-cutting shoots (live mini-cuttings with or without roots and callus, showing shoots of new leaves) (Fig. 2A), and percentage of mini-cuttings that retained their original leaves (live mini-cuttings, with or without roots and callus, which retained the original leaves at the time of evaluation) (Fig. 2A).

FIGURE 2 A) Rooted mini-cutting, B) mini-cutting with callus and C) live mini-cutting. LM - leaf maintenance; Sh - shoots of new leaves. 

Experiment II - Biochemical composition of Gingko biloba mini-cuttings

For the biochemical analysis, ten Ginkgo biloba mini-cuttings were used at the time of the rooting and ten mini-cuttings at the time of evaluation at 60 d. For the quantification of the total protein content, the method described by ^{Bradford (1976)} was used and the total soluble sugar contents were determined by the phenol-sulfuric method described by ^{Dubois et al. (1956)}.

Statistical analysis

For both experiments, the treatment variances were tested for homogeneity using the Bartlett's test (P<0.05) and normality test [Shapiro-Wilk (W)]. Subsequently, the variables were subjected to analysis of variance (ANOVA) and those with significant differences in relation to the F test had their means compared with the Tukey test (P<0.01 and P<0.05). Only the variables live mini-cuttings, percentage of mortality, callus formation, and percentage of mini-cuttings that retained their original leaves did not show normality but were homogeneous with the Bartlett test.

Experiment I - Gingko biloba mini-cuttings

There was no significant interaction (P<0.05) between the substrate x IBA concentrations for all the evaluated variables, indicating that these factors can act alone in the adventitious rooting of the species. The concentrations of IBA did not influence the induction and development of Ginkgo biloba roots (Tab. 1). In a previous conventional cutting study of the species, ^{Valmorbida and Lessa (2008)} observed that the IBA influenced adventitious rooting formation, when compared to the treatment without the use of this plant growth regulator, with rooting percentage results up to 80.55%. ^{Bitencourt et al. (2010)} found statistical differences with higher concentrations of IBA (4000 and 8000 mg L^-1) bound in talc on ginkgo cutting rooting (45.00 and 46.25% rooting percentages, respectively).

TABLE 1 Rooting percentage (R), number of roots per mini-cutting (NR), average length of three major roots per mini-cutting (LR), live mini-cuttings (A), percentage of mortality (M), callus formation (C), percentage of mini-cutting shoots (Sh), percentage of mini-cuttings that retained their original leaves (LM), subjected to substrates and concentrations of indole butyric acid (IBA) in Ginkgo biloba mini-cuttings. 

Means followed by the same letter In each variable do not differ according to the Tukey test at 5% probability. CV: coefficient of variation.

Thus, since the effect of IBA on the rooting of the species has not been observed, the results obtained in this study are promising considering the high rooting rates (mean of 62.05%). This result shows a positive effect of the rejuvenation of the plant material in the adventitious rooting of Ginkgo biloba and indicates the potential of the minicutting technique for the species, resulting in a reduction in the cost of production since plant regulators are not used for root induction.

One of the advantages of the mini-cutting technique is precisely the reduced need for exogenous auxin application due to the reinvigoration of parent plants (^{Stuepp et al., 2015}; ^{Stuepp et al., 2017b}). Similarly, ^{Stuepp et al. (2017a)} confirm the efficiency of the mini-cutting technique for the species, resulting in a high rooting index (92.5%). The low percentages of live and dead mini-cuttings in all treatments are a consequence of the high levels of rooting with fast formation, demonstrating the adaptation of the material to greenhouse conditions. The high potential of rhizogenesis of Ginkgo biloba mini-cuttings validates the efficiency of the mini-cutting technique for the species.

Regarding the variable mini-cuttings with callus, the Trop-strato^® substrate promoted a higher percentage of callus that differed statistically from the vermiculite substrate. The treatment without IBA application and with 1000 mg L^-1 showed a higher percentage of callus when compared to the others; however, these treatments did not have an effect on the rooting of cuttings. The treatments that showed the highest percentage of calluses may be promising, indicating that a longer permanence of the mini-cuttings in the greenhouse could result in root induction. This fact is justified since the treatments with higher percentages of calluses were the same with lower percentages of rooting (Tab. 1). This inverse relationship between the percentage of callus and adventitious root formation has already been reported by other authors who observed that the formation of adventitious roots in Ginkgo biloba can occur from the callus tissue formed at the base of cuttings or mini-cuttings. This suggests that root induction could result from a longer permanence of mini-cuttings in the greenhouse (^{Valmorbida & Lessa, 2008}; ^{Bitencourt et al., 2010}; ^{Stuepp et al., 2017a}).

There was a significant difference between the concentrations of IBA in the percentage of shoots, with the highest percentages observed in the treatments without IBA application and in the concentration of 3000 mg L^-1 (Tab. 1). According to ^{Moubayidin et al. (2010)}, root growth occurs when, in the apical meristem, cell division prevails over differentiation. This occurs when there is a greater concentration of auxins promoting cell division, in contrast to the cytokinin concentrations that promote cellular differentiation.

The auxin: cytokinin ratio regulates tissue morphogenesis. While a high auxin:cytokinin ratio stimulates the formation of adventitious roots, a low ratio leads to higher growth of the shoot (^{Taiz et al, 2017}). According to ^{Agullo-Anton et al. (2011)} and ^{Agullo-Anton et al. (2014)}, exogenous auxin supplementation in cuttings affect the endogenous cytokinin concentration, influencing adventitious rooting. ^{Souza and Miranda (2006)} found in an in vitro study that the proliferation of shoots and root formation in Gerbera jamesonni depends on the balance between auxin and cytokinin. Although a greater number of shoots was observed in the treatment without synthetic auxin application, adventitious rooting was not influenced, indicating that the presence of leaves induced an auxin-favorable balance.

It is possible to verify a close relationship between the rooting percentage and adventitious root development by maintaining mini-cutting leaves in all treatments (Tab. 1). The importance of leaf maintenance for these two variables has been reported in the literature, especially related to the presence of certain leaf compounds, such as carbohydrates, auxins and rooting cofactors, which can be translocated by phloem to the base of the cuttings, thus stimulating root formation (^{Bona & Biasi, 2010}; ^{Fragoso et al, 2015}).

Experiment II - Biochemical composition of Gingko biloba mini-cuttings

Adventitious rooting is influenced by the biochemical composition of propagules (^{Yan et al, 2017}). Some studies have demonstrated the correlation between rooting and carbohydrate content (^{Aslmoshtaghi & Reza-Shahsavar, 2010}; ^{Ragonezi et al, 2010}; ^{Denaxa, et al, 2012}).

The carbohydrates translocated by phloem are non-reducing sugars such as sucrose, because they are less reactive than reducing sugars (^{Taiz et al, 2017}). This fact can be seen in Figure 3, where it is possible to verify a decrease in the non-reducing sugar concentration at the time of mini-cutting evaluation (60 d after experimental set up). Additionally, it is possible to observe a high level of non-reducing sugars at the moment of root evaluation.

FIGURE 3 Biochemical composition of different organs of Ginkgo biloba and collection time. I - Installation; E - evaluation. Means followed by the same letter do not differ according to the Tukey test (P<0.05). 

These carbohydrates translocated by phloem can be used as an energetic resource in root induction and development (^{Aslmoshtaghi & Reza-Shahsavar, 2010}; ^{Souza et al, 2015}) and as a carbon source for biosynthesis of amino acids and nucleic acids (^{Fachinello et al, 2005}). They can also act in the regulation of gene expression (^{Wang & Ruan, 2013}). Thus, the reduction of endogenous carbohydrate levels indicates that these sugars were used during root emission and growth (^{Husen & Pal, 2007}), as demonstrated in the present study.

The quantity of protein in the mini-cuttings observed at the moment of evaluation was also lower than that observed at the moment of setting up the experiment (Fig. 3). Proteins are biochemical compounds associated with rooting, that can help in the induction, formation and development of adventitious roots (^{Taiz et al, 2017}), and maybe involved in the signaling and biosynthesis of auxins (^{Franklin et al, 2011}; ^{Hornitschek et al, 2012}; ^{Zhang et al, 2017}).

When evaluated separately from cuttings, the reduction in the levels of non-reducing sugars and proteins during rooting that is associated with the high contents of these compounds in the roots (Fig. 3), indicates an ideal redirection of the source of drain energy. This fact reassures the importance of the nutritional status of the stock plant since it is directly related to the formation of the root system (^{Hartmann et al, 2011}), as confirmed by the high rooting indexes in the present study. Furthermore, these results reinforce the importance of the nutritional status of the stock plant that directly affects the mini-cutting rooting and survival, and that is justified by the high rooting results observed in the present study.