Field establishment

The experiment was carried out in 2011 in General Cepeda, Coahuila and in the Cytogenetics Laboratory from the Universidad Autonoma Agraria Antonio Narro in Saltillo Coahuila. The plant materials were three autotetraploid (11, 16 and 20) selected based on yield and quality of fruit and the diploid (19) rendidora. The populations were sown depositing two or three seeds per well in polystyrene trays of 200 cavities, using as substrate peat and perlite at a ratio of 1:1. The trays were carried to a greenhouse where seedlings developed and when these reached a height of 10 to 12 cm and two pairs of true leaves, the seedlings were transplanted in the field on grooves with padded black plastic, 5 m long and 1.8 m wide with a spacing of 60 cm between plants and two rows per groove with spacing of 30 cm between rows. Ten days after transplantation a thinned was performed leaving only one seedling per point. The treatments were established under a randomized block design with four replications, each treatment consisted of 10 plants. Nutrition is applied through irrigation, three times a week; irrigation was daily applying two liters per day per plant until 45 days after transplantation, then increased to four liters per day per plant.

Morphological characterization

At 60 days after transplantation, the following morphological variables were evaluated in diploid and tetraploids: leaf width (AH) and length (LH), stem diameter in cm (DT), flower diameter in cm (DF), plant height in cm (AP), these variables were measured in three plants taken randomly from each treatment in each of the four repetitions, making a total of 12 plants per each population. The measurements were made using a digital precision vernier AutoTEC®, plant height was measured with a tape measure in cm. These variables were evaluated in order to characterize tetraploid populations and to identify morphological changes induced by polyploidization in tomatillo plants.

Histological study

Samples were taken at 60 days after planting three plants per treatment of each replication and taken to the laboratory, cut a piece of tissue (2 cm) from the main root at 3 cm below soil surface, piece of tissue (2 cm) from stem at 10 cm above soil surface, in addition to the base of the first leaf a piece of tissue (approximately 1 cm2) taking as reference the midrib and petiole fragment (2 cm) from the first branch of each plant; the tissues were processed by the histological paraffin technique as follows: to preserve the tissues with minimal alterations were placed in glass jar 15 ml with fixative FAA (Formaldehyde (36-40%) 5cc, ethyl alcohol (70%) 90 cc and 5 cc glacial acetic acid). Then the samples were dehydrated in ethyl alcohol concentrations 50, 60, 70, 85 and 96%, plus eosin, continuing with absolute ethyl alcohol I, absolute ethyl alcohol II, absolute ethyl alcohol plus xylene in volume ratios of 3:1, absolute ethyl alcohol plus xylene in proportion 1:1, absolute ethyl alcohol plus xylene in proportions of 1:3 and finally the tissues passed to pure xylene and remained in each solution for 2 h and embedded in paraffin with melting point at 58 °C then the alcohol was replaced with xylene (^{Hernandez, 1990}; ^{Jauregui, 2003}; ^{Cañizares et al., 2005}).

Transversal cuts of tissue at 20 microns with a rotary microtome were made, which were mounted on a slide. For staining a dual color staining of safranin-fast green were used, passing through different alcohol and xylene solutions, the preparations were pulled out from the last xylol and drained. Sealed with a drop of canada balsam and a slide of the tissue size.

Thus the preparations were put to dry in an oven at 30 °C for a week. The tissues were analyzed with the aid of a microscope with an integrated camera Pixera Wiender Pro; the best samples were selected and photographed with 10X and 40X objectives (^{Wilkinson, 1979}). The measurements were made from digital images, with measurement software Axion Vision Rel. 4.8. The variables were xylem vessels area from the midrib of the leaf (AXH), petiole xylem vessels (AXP), xylem vessels from the main root (AXR), and stem xylem vessels (AXT) palisade parenchyma (LPEH) and lacunar parenchyma in leafs (APLH).

Statistical analysis

An analysis of variance was performed on a randomized block design with four replications, the significance for morphological variables: leaf width (AH) and length (LH), stem diameter (DT) flowers diameter (DF), plant height (AP), and histologic: xylem vessel area from the midrib of the leaf (AXH) petiole xylem vessels (AXP), xylem vessels from the main root (AXR), and stem xylem vessels (AXT) palisade parenchyma (LPEH) and lacunar parenchyma in leaf (APLH), was considered on the ploidy level, where the variation factor were three autotetraploid populations and one diploid population of tomatillo, for mean comparison Tukey test (p≤ 0.05). SAS statistical software version 9.0 was used.

Morphological characterization

Diploid population had on average a width (AH) of 6.81 cm while the three autotetraploid populations had on average 8.51 cm, autotetraploid population 11 showed the highest value with 9.21 cm and was statistically similar (p> 0.05) to the other two tetraploid populations, but different (p< 0.05) from rendidora diploid (Table 1) observing that the leaves of autotetraploid plants showed a darker green color than diploid plants, also the life cycle of autotetraploid plants was longer (about 15 days) than diploid, these results agree with those reported by ^{Chen et al. (2011)} in tetraploid plants of Anthurium andraeanum "Arizona" induced with colchicine since tetraploid plants showed firm petioles, thick leaves and a dark green color, also had a longer cycle compared to diploid plants. Regarding to leaf length (LH) in this study, no significant differences (p> 0.05) were found between diploid and autotetraploid populations.

Table 1 Mean values of morphological characteristics of diploid and autotetraploid tomatillo. 

AH= ancho de hoja, LH=largo de hoja; DF= diámetro de flor; DT= diámetro de tallo; AP=altura de planta; *Letras diferentes dentro de las columnas indican diferencias significativas a nivel (P≤ 0.05); DMS= Diferencia mínima significativa.

In flower diameter (DF) significant differences between autotetraploid and diploid populations were found, the tetraploid populations showed between 2.47 and 2.68 cm, being tetraploid 16 which had the highest value and were statistically similar (p> 0.05) between them. The population that showed the smaller flower diameter was diploid 19 (2.07 cm), agreeing with ^{Liu et al. (2007)} who found morphological differences between diploid and tetraploid on Platanus acerifolia developed with the use of colchicine in Gerbera jamesonii Bolus cv. Sciella, the tetraploids showed slow growth, but had greater vigor, broad and thick leaves, larger flowers, longer stems and improved vase life (^{Gantait et al., 2011}).

In stem diameter (DT), there were no significant differences between autotetraploid and diploid population, the tetraploid population 11 (2.53 cm), showed the highest value and was statistically similar (p> 0.05) to all others, including diploid population 19.

As for plant height (AP) diploid 19 (82.2 cm) had the highest value and was statistically different (p <0.05) to autotetraploid populations showing the lowest values (58.60 cm to 67.13cm) and were statistically equal (p> 0.05). Tomatillo autotetraploids had higher values for leaf width and flower diameter compared to diploid, coinciding with ^{Cequea (2000)} who mentions that autopolyploids have a wider ecological tolerance and larger cell sizes, meanwhile ^{Imery and Cequea (2001)} found that new tetraploid genotypes of A. vera produced greater leaf biomass accumulation, especially due to the increase in width and thickness of the leaves, also, ^{Rego et al. (2011)} found that with the formation of autotetraploid of Passiflora edulis Sims. (2n= 2x= 18) using colchicine and oryzalin obtained more vigorous plants (2n= 2x= 36).

In this research was noted that tetraploids growth was slower because at 60 days when diploid variables were taken outperformed tetraploids in plant height; however, at the end of the cycle tetraploid exceeded diploid, so it is recommend to take these variables at different stages of development and thus be able to match the observed, ^{Molero and Matos (2008)} who found that the gene duplication in Aloe vera increased plant height, length, width , thickness and leaf volume regarding to diploid plants, so it is estimated that polyploidization could be an alternative to increase production of cell biomass in this species. Leaf area is an important variable because it is directly related to the photosynthetic activity and this with the accumulation of reserves (^{Poehlman and Allen, 2005}). A variable that normally is related to fruit size is flower size and in this case according to observed values it can be said that an increase was induced in autotetraploids regarding this characteristic.

Histological study

Histological variable cell area of lacunar parenchyma in leaf (APLH) the autotetraploid populations had on average 6 912.67µm2 being surpassed by diploid population showing a value of 7 398.97µm2, in xylem vessels area from the midrib leaf (AXH) the diploid showed 655.08 µm2 while autotetraploid presented 602.81µm2, length of palisade parenchyma cells (LPEH) autotetraploids exhibited 100.61µm with a value that exceeded the diploid, since this had 97 361µm (Figure 1), xylem vessels area in petioles (AXP) autotetraploid populations showed on average 604.54 µm2 while diploid plants showed a value of 714.56 µm2; however, no significant differences were found between diploid and autotetraploid plants since they were statistically equal (p≥ 0.05). The fact that in the present study no significant differences were found in these variables, may indicate that despite chromosomes and genes duplication, leaf and petioles characteristics were little affected by polyploidization in tomatillo, this may also be due to the fact that autotetraploid have a slow growth and the cycle is longer than diploids and for the present study diploid and autotetraploid samples were taken on the same date.

Figure 1 Xylem vessels of the midrib of the leaf; A) Diploid; autotetraploid. Palisade parenchyma of the leaf; Diploide; and D) autotetraploid in tomatillo 10 X. 

Variance analysis applied to diploid and autotetraploid plants presented significant differences (p< 0.01) between diploid and autotetraploid regarding to xylem vessels area from stem (AXT) where the three autotetraploid populations (1 842.66 µm2) exceeded the diploid (748.4 µm2) in 146.17%, for xylem vessels area in root (AXR) autotetraploid plants showed on average 1 309.68 µm2 while diploid had 738.4 µm2 and were excelled by tetraploids in 77.36%, therefore it was found that doubling the genome of tomatillo, increased xylem vessels area in stem and root (Figure 2 and 3), these results could be positive for this species according to ^{Zahng and Cao (2009)} who assume that it has been proven that the structure of the vascular system of the stem has a strong influence on leaf structure and functioning, assimilation of carbon dioxide, efficiency in the use of nutrients and growth rate.

Figure 2 Area of xylem vessels in roots and stems of the diploid population and population autotetraploid tomatillo. 

Figure 3 Stem xylem vessels, A) diploid; B) autotetraploid. Root xylem vessels; C) Diploide; and D) autotetraploid in tomatillo 10 X.