Location and description of the experimental area

The experiment was carried out in a greenhouse of the Institute of Agricultural and Forestry Research (IIAF) for its acronym in Spanish, of the Michoacan University of San Nicolás de Hidalgo located in the Morelia-Zinapécuaro highway, km 9.5, 58880, Tarímbaro, Michoacán. With geographical coordinates: 19.7681622, - 101.1513041. Elevation 1 860 m.

Plant material

The plant material consisted of cuttings between 7 and 10 cm in length with three axillary buds that were obtained from lateral branches, selected during a pruning of healthy 2-year-old plants of Vaccinium corymbosum variety Biloxi. These stem sections were disinfected by immersion in sodium hypochlorite at 35 ppm for 1 min, then rinsed five times with deionized water and immersed in solution with fungicide (LucavFlow^® 10 ml L^-1) for 1 min, finally rinsed 10 times with deionized water.

The basal part of the cuttings (2 cm) was placed for 2 h in a solution of 24-Epibrasinolide (24-EBL) and a control solution [dimethyl sulfoxide (DMSO) + polyoxyethylenesorbitan monolaurate surfactant agent (Tween 20) + distilled water], then they were transplanted into cubic black plastic containers of 1 L capacity containing the substrate mixtures, prior to planting, the substrates were irrigated with deionized water until drained and then the cuttings were planted, the containers were irrigated daily at field capacity for 30 days until the end of the experiment.

Conditions in the greenhouse

The greenhouse where the experiment was installed is of the curved roof type with milky white plastic of gauge 720, which provides 30% shade and allows maintaining a uniform temperature inside the greenhouse of 27 ° C on average during the day and a relative humidity of 47%.

Experimental design

A completely randomized experimental design with factorial arrangement was used, with 4 repetitions in each pot, the experimental unit consisted of a cutting. Twenty treatments were generated from the combination of the two factors, factor substrate: types of substrates Sphagnum^® Peat Moss (PM); mixture of PM + perlite in a ratio 2:1 (V/V) (PMP); commercial coconut fiber (Power Forteco) (CF); mixture of CF + perlite in a ratio 2:1 (V/V) (CFP) and mixture of river sand + tezontle in a ratio 1:1 (V/V) (ST) and the factor concentration of 24-EBL (0, 20, 40, 60 μg L^-1).

Phytohormone preparation

The solutions of 24-EBL at 20, 40 and 60 μg L^-1 were obtained by dilutions with distilled water of a stock solution of 24-Epibrasinolide (24-EBL) at a concentration of 0.01 g L^-1. This solution was prepared by dissolving 10 mg of 24-EBL (Sigma-Aldrich Epibrassinolide, ≥85%) in 1 ml of dimethyl sulfoxide (DMSO), then 25 ml of double distilled water and 0.1% (V/V) of polyoxyethylenesorbitan monolaurate (Tween 20) as surfactant were added, the final volume was adjusted to 1 L with double distilled water.

Physical and chemical characterization of substrates

The substrates were disinfected with sodium hypochlorite solution at 35 ppm and washed three times with deionized water, placed in black plastic containers of cubic design with a capacity of 1 L. The electrical conductivity (EC) of the substrate was determined with the saturation extract method (^{Vargas et al., 2008}). The concentration of N in the substrates was performed by the microKjeldahl method. P, K, Ca and Mg were determined after wet digestion and quantified by atomic absorption spectrophotometry (Varian ICP-AES Plasma 96).

These analyses were carried out in triplicate at the laboratory of the Center for Innovation and Agrifood Development of Michoacán (CIDAM) for its acronym in Spanish, located in the city of Morelia, Michoacán. Bulk density (Db) was determined by the graduated cylinder method (^{Acosta et al., 2014}) and real density (Dr) by fluid pycnometry (^{Díaz et al., 2012}). From these determinations, the total porosity of the substrate was obtained using the equation Pt (%) = 100 (1-Db/Dr). The determination of moisture retention was performed by the hanging column method (^{Crespo et al., 2018}) and from this measurement, air porosity (Pa) was obtained. These tests were done in the Laboratory of Plant Nutrition at the Institute of Agricultural and Forestry Research (IIAF) for its acronym in Spanish.

Variables evaluated

The variables evaluated were percentage of survival of cuttings, number of rooted cuttings (those that had a main root of 2 mm in length), number of roots per cutting, main root length and total length of the root system. To analyze the root system, photographs of the rooted cuttings were taken for each treatment and the images were processed with the Root System Analyser software - analysis of images and sequences of plant roots, of the University of Vienna (^{Leitner et al., 2014}). The analyses were performed 30 days after (DAE) the experiment was established.

Statistical analysis

From the data obtained, the analysis of variance was performed and in the variables with significant effect, the comparison of means was carried out with the Tukey test, in both analyses it was used at a probability of 95% with the statistical program IBM SPSS^® v26 2021.

Characterization of substrates

The substrates used in this work presented the following characteristics: Peat Moss, it is a substrate with almost zero salinity effects (EC) of 0.5 dS m^-1, a high concentration of N of 0.84 g kg^-1, low in P with 5.2 ppm, poor in K with 15 ppm, poor in Ca with 10 ppm and poor in Mg 7 ppm, a high moisture retention of 64.5%, these results are similar to those reported by ^{Gómez et al. (2013)}.

Perlite was not characterized in this study, the reference values indicated in the supplier’s technical sheet (Multiperl^® Hotyicola) were taken, which indicates the following characteristics: expanded volcanic siliceous rock with a pH: 7, it has a total pore volume of 95%, the approximate density is 0.12 kg L^-1.

Regarding the results obtained for the mixture of PMP, it is a substrate with very low salinity effects (EC) of 0.1 dS m^-1, as mentioned by ^{Sánchez et al. (2008)}, a high N concentration of 1 g kg^-1, low in P with 4 ppm, very low in K with 3 ppm, limited in Ca with 3 ppm, a high moisture retention of 60%. The coconut fiber presented a mean concentration of N, high P content, very high in K, slightly high in Ca and Mg, similar results reported by Pardo and Pardo (2008) (Table 1).

The mixture of river sand and tezontle, with almost zero salinity effects (EC), because it has a mineral origin, has a very poor content of organic matter in nutrients, as mentioned by ^{Luna et al. (2021)}, this mixture is extremely low in N and P, very high in K, slightly high in Ca and Mg. Regarding the physical characteristics of this mixture, it presented a high bulk density, percentage of water retention of 43.3% (Table 2).

Substrate effect

The rooting capacity of cuttings from Ericaceae plants is limited because blueberries have lower nutrient requirements than most crops and thrives in acidic soils (pH of 4.5-5.5) with limited availability of essential nutrients such as nitrate-nitrogen (NO₃ -N), phosphorus (P), potassium (K), calcium (Ca) and magnesium (Mg) (^{Braha and Rama, 2018}). When evaluating the effect of the concentration of 24-EBL and the type of substrate separately and the interaction of the factors, significant differences (p= 0.05) were found in the variables evaluated in blueberry (Vaccinium corymbosum) cuttings, these results are shown in (Table 3).

In the percentage of survival of cuttings, the mixture of ST had the highest mean of 48.88%, the PM presented the lowest mean of 16.84%, while the mixtures of PMP and CFP did not show significant differences (p< 0.05), with these results the effect that perlite has on the mixture of substrates was observed, these results are similar to the studies reported by ^{De-Boodt and Verdonck (1972}), where they indicate that the values of aeration porosity are higher in perlite with 37.79%, it is a value that is considered favorable, in this research the mixture of perlite and peat (Peat Moss^®) had an aeration of 31.6%.

On the other hand, the results obtained in the present research support the suitability of the coconut fiber substrate for blueberry propagation, as reported by ^{Machado et al. (2014}), when mixed with perlite. It is not recommended to use this substrate at 100% due to the chemical characteristics of coconut fiber, with high concentrations of K, Na and Cl, it can cause stress on plants and delay rooting.

The effect of the interaction of factors in the rooting of cuttings was observed at day 30, most studies reported that the rooting of blueberry cuttings occurs in a time of 55 to 60 days, with a success percentage of 40 to 52% (^{Castro and Guzmán, 2013}). ^{Pego et al. (2019}) reported that some species of Ericaceae root better in mixture of peat and river sand; however, with the results obtained in the present work, the mixture of river sand and tezontle presented good physicochemical properties for the rooting of blueberry, this mixture maintains adequate ventilation in the rooting environment.

Effect of the brassinosteroid 24-EBL

When the effect of 24-EBL on the percentage of survival of cuttings was analyzed, a significant effect of this factor was observed, depending on the concentration, when the cuttings were treated with 60 μg L^-1, this variable was 54.75%, while with 0 μg L^-1 it was only 5.11%. ST treatments with 20 and 40 μg L^-1 presented 36% and 63.7% respectively.

Regarding the percentage of rooted cuttings and the number of roots per cutting, the concentration of brassinolide was a significant factor (p= 0.042 and p= 0.013 respectively). When a concentration of 60 μg L^-1 24- EBL was used, the percentage of rooting was 67% in the ST mixture, in this same mixture with 40 μg L^-1, 60% of the cuttings rooted, while with the concentration of 20 μg L^-1, 28% of the cuttings rooted and without the brassinolide only 8% rooted (Table 3).

These results are similar to those presented by ^{Ahammed et al. (2017}) in Arabitodpsis, where the effect of brassinolides as bioactive products that potentiate and induce early physiological mechanisms thanks to intermembrane receptors called BRASSINOSTEROID INSENSITIVE1 (BRI1), BRI1-LIKE 1 (BRL1) and BRL3 was observed. BRI1 binds to brasinolide (BL), the most active form of brassinolides, and leads to ligand-mediated activation of the BRL1 receptor, resulting in transphosphorylation events, which also involve coreceptors, such as BRI1 ASSOCIATED KINASE1 (BAK1) (^{Ayub et al., 2020}), these results indicate that cuttings of blueberry variety Biloxi have these receptors and that when they are in the presence of 24-EBL, these are activated triggering cascades of signals that induce the formation of root tissue.

It was observed that the concentration of 24-EBL and the number of roots have a direct relationship, at a higher concentration of 24-EBL, the number and length of roots increase (Table 3). Exogenous application of different brassinolide analogues is known to be used to improve seed germination and induce growth promotion of hypocotyls, cotyledons, leaf blades, main and lateral root elongation (^{Hussain et al., 2020}); nevertheless, until now they had not been used as root-inducing agents in blueberry cuttings.

The combination of the concentration of 40 μg L^-1 with CFP and CF substrates did not show significant differences. At a concentration of 20 μg L^-1, the substrates CF, PM, CFP and PMP did not present significant differences and showed values of 50% to 34% less with respect to the treatment of ST (Figures 1A, 1B).

Figure 1 A) Percentage of rooted cuttings; and B) main root length (LRP) of blueberries, propagated on different substrates and treated with 24-EBL. Substrates= sphagnum^® Peat Moss (PM); mixture of Sphagnum^® Peat Moss and Perlite in a ratio 2:1 (PMP); coconut fiber c (Power Forteco) (FC); mixture of coconut fiber (Power Forteco) and perlite in a ratio 2:1 (FCP) and a mixture of river sand and tezontle in a ratio 1:1 (V/V) (AT). Columns with different letters indicate significant differences (Tukey p< 0.05). Means ± standard error. 

^{Mouchel et al. (2006}) mention that there is a synergism between exogenous brassinolides and other plant phytohormones, such as auxins and ABA (abscisic acid), which are responsible for the induction of primary roots. Specifically, 24-EBL also regulates root development through an interaction with auxin in a dose-dependent manner (^{Sun et al., 2020}).

In this study, it was determined that Biloxi roots well in substrate of river sand and tezontle (80%), is acceptable in mixtures of coconut fiber and perlite (50%), peat (Peat Moss) and perlite (52%). For propagation by blueberry cuttings, it is not recommended to use substrates such as peat (Peat Moss) and coconut fiber at 100%, it is recommended to use them in mixture with perlite or river sand.

Blueberry rooting is commonly performed under in vitro conditions using auxins as a root inducer (^{Hung and Trueman, 2012}). The presence of 24-EBL was effective for root formation in blueberry cuttings as shown in the present study, using an alternative substrate to those that are frequently used for propagation decreases cost by directly rooting cuttings under non-aseptic conditions using commercial substrates and solutions of brassinolides at low concentrations.