Plant material

Clusters of 6 to 12 adventitious shoots of different sizes formed on A. angustifolia stem tissue were the plant material for the experiment. To transfer the shoots to culture media for the propagule multiplication stage, under aseptic conditions in a horizontal laminar flow filtered air chamber and using sterilized tweezers and scalpel, the clusters were extracted from the initial culture medium and placed in a sterilized glass Petri dish, 10×100 mm, where they were divided into smaller clusters of 3 to 4 shoots with 9 to 15 leaves on a base of stem tissue from which they initially emerged. One cluster of shoots (experimental unit) was established in each 145 cm³ of recipient containing 20 mL of one of several variants of culture medium to induce proliferation of the adventitious shoots. These culture medium variants contained 100 mg L⁻¹ myo-inositol, 30 g L⁻¹ sucrose, 0.4 mg L⁻¹ thiamine-HCl, mineral salts (MS), diverse quantities of IAA (0.1, 0.3, 1 mg L⁻¹) and N⁶-benzylaminopurine (BAP; 0.5, 1, 2, 4 mg L⁻¹). The pH of the culture medium variants was adjusted to 5.8 before adding 5.6 g L⁻¹ of agar. The agar was dissolved by heating and shaking and distributed in the 145 cm³ glass recipients which were capped with a polypropylene stopper and sterilized in an autoclave for 17 min at 120 °C and 1.2 kg cm⁻² pressure.

After planting the clusters, the polypropylene stopper was again placed on the recipients and sealed with polyethylene adherent. The cultures were kept for 16 wk in the incubation area, where they were exposed to white fluorescent illumination (35 μmol m⁻² s⁻¹) in photoperiods of 16 h light and 8 h dark; the temperature varied from 15 to 29 °C.

At the end of the experiment, seven experimental units from each treatment were harvested and the following variables were evaluated as follows: 1) number of leaves, the total number of leaves of the shoots that formed the cluster on the common base of stem tissue; 2) number of new shoots; 3) size of the largest shoot, measured using a precision ruler in 0.1 cm increments; 4) fresh weight, determined on an analytical balance with 0.1 mg accuracy (DENVER INSTRUMENT®, USA); and 5) total dry weight, where the samples of plant material were placed in paper bags for drying in a convection oven (Felisa® Fabricantes Feligneo, S. A. de C.V. México) at 70 °C for 72 h and later weighed on an analytical balance.

Experimental design and evaluation

The experiment was set up in a completely randomized design with a 3 (IAA 0.1, 0.3, 1) × 4 (benzylaminopurine 0.5, 1, 2, 4) factorial array. The experimental unit was one cluster of shoots, and there were 12 replications per treatment. The data were subjected to a variance homogeneity test using the Bartlett test and the Shapiro-Wilk normality test. The variable number of shoots was transformed to log10(x) to comply with assumptions of variance normality and homogeneity. The data of each variable were subjected to analysis of variance and a test of comparison of means (Tukey, 0.05). For the variables of increase in number of leaves and shoots, the statistical models of exponential regression y=b0eb1(1/x) were carried out to obtain models that described the increase in shoots or leaves as a function of the variation in growth regulator concentrations. Later, graphs of response areas were constructed using Minitab 15® software (Minitab. Inc., State College PA, USA).

Results and discussion

The results of this experiment indicate that Agave angustifolia Haw. explants subjected to the tested treatments had more shoots and more leaves, but the magnitude of the increases was conditioned by growth regulator concentrations. At the beginning of the experiment, each experimental unit consisted of clusters of 3 to 4 shoots with a total of 9 to 15 leaves and heights of 3 to 4 cm (Figure 1a). The analyses of variance showed that the concentrations of benzylaminopurine (BAP) had highly significant (P ≤ 0.01) effects on leaf number, leaf length, number of shoots, fresh weight and total dry weight, while the concentrations of IAA had highly significant (P ≤ 0.01) effects on total dry weight and a significant (P ≤ 0.05) effect on number of shoots and total fresh weight. The interaction of BAP and IAA had a highly significant effect on number of shoots and total dry weight and a significant effect on total fresh weight (Table 1). There was an increase in the number of shoots that was in a positive relationship with increasing concentration of BAP and IAA (Figure 1b, 1c and 1d). Several studies that have achieved organogenesis in vitro report that the growth regulators are the critical factor for growth, quantity or size of the propagules (^{Ramírez-Malagón et al., 2008}; ^{Meratan et al., 2009}; ^{Shimizu et al., 2009}).

Figure 1 a) Clusters of 6 to 12 adventitious shoots formed on explants from Agave angustifolia stem tissue and transferred to propagule multiplication; b) in vitro cultures of A. angustifolia shoots incubated for 16 weeks; c) A. angustifolia shoot; d) harvest of A. angustifolia shoots 16 weeks after in vitro establishment. 

At the beginning of the experiment, the explants had 3 to 4 shoots with a total of 9 to 15 leaves. After 16 wk of incubation, the explants established in the different variants of culture medium had formed more shoots and leaves, but this increase had a significant (P ≤ 0.001) positive exponential relationship with the BAP and IAA concentration, with an R²-adj for leaves and shoots of 0.76 and 0.61, respectively. The number of new shoots and leaves is conditioned by the BAP concentration in the culture medium, and the synergy that occurred with the IAA auxin increased the effectiveness of the cytokinins (Figures 2a and 2b). In this way, the explants established in the culture medium with 0.5 mg L⁻¹ BAP + 0.1 mg L⁻¹ IAA had 7.8 shoots with 24 leaves, 4.5 g total fresh weight and 0.21 g total dry weight, while the explants established in the culture medium with 4 mg L⁻¹ BAP + 1 mg L⁻¹ IAA had 32.8 shoots with 45 leaves, 9.1 g total fresh weight and 0.58 g total dry weight (Table 2). These figures, in each case, were significantly different (Tukey, 0.05).

Figure 2 Response areas due to the effect of N6-benzylaminopurine (BAP) in combination with indol-3-acetic acid (IAA) in the culture medium on a) an increase in the number of leaves (INL) and b) an increase in the number of shoots (INS). 

In a scheme of in vitro propagation of agaves, during the propagule multiplication stage, it is necessary to know what types and concentrations of growth regulators to use in the culture medium to achieve optimal response, considering the maximum number of quality shoots. Quality shoots are those that are not deformed and do not have hyper-hydrated organs.

For the micropropagation of different Agave species, the dose of any cytokinin has been evaluated to induce the formation of shoots in protocols for in vitro propagation of Agave cupreata, A. difformis, A. karwinskii, A. obscura and A. potatorum. Formation of multiple shoots on basal explants has been achieved in an MS medium supplemented with 30 g L⁻¹ sucrose, 8 g L⁻¹ agar and several types of cytokinins: BA, 2iP, Cin, TDZ and MT, which also varied in concentration (0.5, 1, 1.5, 2, 2.5, and 3 mg L⁻¹). It was found that explants of A. cupreata and A. karwinskii established in a CM with 1.5 and 1 mg L⁻¹ BA formed 10.5 and 6.1 shoots each, respectively, while explants of A. difformis and A. obscura responded best in a CM with 0.2 mg L⁻¹ TDZ, forming 8.5 and 11 shoots per explant, respectively. A. potatorum responded best in a CM with 3 mg L⁻¹ kinetin, forming 6.9 shoots per explant (^{Domínguez-Rosales et al., 2008b}). However, in other works, the best response was obtained when a combination of cytokinin and auxin was used as follows: 0.002 mg L⁻¹ 2,4-D and 9.9 mg L⁻¹ BA for A. americana (^{Reyes-Zambrano et al., 2016}); 0.011 mg L⁻¹ IBA and 1 mg L⁻¹ BA, for A. duranguensis, 0.49 mg L⁻¹ IBA /1 mg L⁻¹ BA for A. oscura, 0.09 mg L⁻¹ IBA / 2.99 mg L⁻¹ BA for A. pygmaea, 0.11 mg L⁻¹ IBA / 1 mg L⁻¹ BA for A. salmiana subspecies crassispina, 0.49 mg L⁻¹ IBA / 0.50 mg L⁻¹ BA for A. victoria-reginae (^{Ramírez-Malagón et al., 2008}), and 0.99 mg L⁻¹ naphthaleneacetic acid plus 0.19 mg L⁻¹ zeatin for A. veracruz Mill (^{Tejavathi et al., 2007}). The aforementioned results show that cytokinin has the most effect in inducing shoot proliferation, but this effect is increased when an auxin was used in the media.

Additionally, in in vitro culture of other plant species, variations in type and concentration of cytokinins and auxins have been used in the culture medium to examine the synergetic effect of the two growth regulators. ^{Siddique et al. (2015)} achieved in vitro propagation of a medicinal species, Cassia angustifolia Vahl, and evaluated two cytokinins (BA, Kin) in the concentration of 1.12 mg L⁻¹ and the auxin IAA (0.01, 0.08 or 0.17 mg L⁻¹). They observed that BA was more effective than kinetin in promoting new shoot formation: the explants established in a culture medium with 1.12 mg L⁻¹ BA and 0.08 mg L⁻¹ IAA developed 8.7 shoots 5.6 cm in length. Although cytokinins have the effect of promoting new shoot formation, it has also been observed that this effect increases when an auxin is included to complement the cytokinins. When ^{Uribe-Moraga and Cifuentes (2004)} propagated Legrandia concinna Phil, they found that the explants established in a culture medium with MS mineral salts and 0.1/0.5 mg l⁻¹ IBA/BAP, and 12.7 shoots per explant were formed with total fresh and dry weights of 9.1 and 0.55 g, respectively, while the explants established in a culture medium with a similar concentration of BAP but when combined with 1.0 mg L⁻¹ IAA formed only 7.8 shoots. ^{Koné et al. (2013)} used explants from Vigna subterranea (L.) Verdc. cotyledons established in culture media with different concentrations of BAP (1, 3, and 5 mg L⁻¹) and NAA (0.01, 0.05, 0.1, and 0.5 mg L⁻¹); on average, 6 to 10 shoots were formed on explants established in a culture medium with 3 mg L⁻¹ BAP alone or combined with 0.5 mg L⁻¹ NAA.

All of the shoots established in the diverse culture media variants grew to lengths of from 7.1 to 10.6 cm, but there were no significant differences (Tukey, 0.05) (Table 2). We thus conclude that the dosage of BAP is a determining factor in inducing formation and proliferation of adventitious shoots, and the type and concentration of auxin can increase the effect (^{Daffalla et al., 2011}). When data are grouped as a function of levels of BAP concentration (Table 3), a positive relationship between the number of new shoots and the concentration of BAP in the culture medium can be observed. Explants established in the culture medium with 0.5 mg L⁻¹ BAP and explants in the culture medium with 4 mg L⁻¹ BAP formed 8.3 and 21.0 shoots, on average, with 27 and 44 leaves, which had a total of 4.7 and 6.8 g total fresh weight and 0.26 and 0.39 g total dry weight, respectively. These magnitudes significantly differed from each other (Tukey, 0.05). This effect could be attributed to BAP, which induces cell division of somatic cells and synthesis of specific proteins (^{Arab et al., 2014}) that lead to later formation of meristematic structures that develop in the shoots. However, the quantity of BAP is related to how much meristematic structure can be induced (^{Siddique and Anis, 2009}).

It should be noted that the shoots obtained in our study in the diverse culture medium variants appeared normal and there were no hyper-hydrated shoots as it is not only the quantity of shoots but also their quality that is desirable.

^{Sánchez-Urbina et al. (2008)} propagated A. grijalvensis in vitro establishing explants in diverse CM in which BA concentrations varied from 0 to 13.51 mg L⁻¹. The highest number of shoots formed on explants established in culture medium with 8.60 mg L⁻¹ BA. ^{Martínez-Palacios et al. (2003)} evaluated indirect organogenesis in A. victoriae reginae by inducing regeneration of multiple shoots from stem segments cultured in MS medium with from 0.49 to 0.99 mg L⁻¹ de BA. ^{Aureoles-Rodríguez et al. (2008)} cultured explants from axillary buds and stem sections of A. inaequidens Kochen in a CM with BA; they achieved 1 to 2 shoots on the explants. ^{Miguel et al. (2014)} established clusters of 2 to 3 shoots with A. americana, while in a CM with different concentrations of BAP in the range of 2 to 8 mg L⁻¹, they induced formation of 8.4 to 11.7 shoots per explant. They found that the number of new shoots was positively related to the concentration of BAP in the culture medium.

^{Nikam et al. (2003)} for 28 d cultured A. sisalana Perr. ex. Engelm explants from embryogenic calluses obtained from a portion of a bulbil shoot. The explants established in a CM with 0.1 mg L⁻¹ kinetin formed somatic embryos, of which 76% germinated.

By ordering the growth data of the cultures as a function of the IAA concentrations in the culture medium (Table 3), a slight but significant trend can be observed in that a larger number of shoots were formed when the concentration of IAA increased in the culture medium. An average of 12.5 shoots formed on explants that were established in a culture medium with 0.5 to 4 mg L⁻¹ BAP supplemented with 0.3 mg L⁻¹ IAA, while 15.5 shoots formed on those established in culture medium with 0.5 to 4 mg L⁻¹ BAP supplemented with 1 mg L⁻¹ of IAA. These numbers were significantly different (Tukey, 0.05). Explants established in culture medium with 0.5 to 4 mg L⁻¹ BAP and 1 mg L⁻¹ IAA had a total fresh weight of 6.5 g and a total dry weight of 0.37 g, which was 1.36 times the fresh weight and 1.85 times the dry weight of the propagules established in a culture medium with 0.5 to 4 mg L⁻¹ BAP and 0.1 mg L⁻¹ IAA, respectively.

Therefore, it can be concluded that in a CM that contains BAP, the auxins in low concentrations contribute to increasing the promoting effect of this cytokinin on adventitious shoot proliferation in A. angustifolia. This agrees with ^{Daffalla et al. (2011)}, who propagated Boscia senegalensis (Pers.) by establishing mature zygotic embryos in a CM with MS inorganic salts and varying concentrations of BA (1 to 5 mg L⁻¹), TDZ (1 to 5 mg L⁻¹), NAA (1 to 5 mg L⁻¹) and 2,4-D (1 to 5 mg L⁻¹). They found that BA induced shoot formation in two ways: adventitious organogenesis and the development of axillary shoots. Explants established in a CM with the auxin 2,4-D and with the cytokinin TDZ tend to produce calluses, while in a CM with the auxin NAA they developed somatic embryos. The auxin NAA and the cytokinin BA were complementary to one another in promoting morphogenesis.

In the present work, BAP promoted shoot proliferation in A. angustifolia, while IAA exhibited synergism with this cytokinin in promoting the shoot formation. The greatest proliferation of A. angustifolia adventitious shoots was obtained on explants established in the culture medium with 4 mg L⁻¹ BAP + 1 mg L⁻¹ IAA, with an average of 32.8 shoots per explant.