Euphytica DOI 10.1007/s10681-014-1273-3 Viability, storage and ultrastructure analysis of Aechmea bicolor (Bromeliaceae) pollen grains, an endemic species to the Atlantic forest Everton Hilo de Souza • Fernanda Vidigal Duarte Souza • Mônica Lanzoni Rossi • Nathalia Brancalleão • Carlos Alberto da Silva Ledo • Adriana Pinheiro Martinelli Received: 31 July 2014 / Accepted: 29 September 2014 Ó Springer Science+Business Media Dordrecht 2014 Abstract Bromeliaceae is a large family, and many species are valued ornamentally for their bright colorful flowers. Pollen grain conservation is important for plant breeding and genetic resource conservation increasing the possibilities of crosses between allogamous species, and further hybrid production. The present study aimed to evaluate pollen conservation methodologies for Aechmea bicolor L.B.Sm. using different viability and germination tests and to characterize conservation effects in pollen morphology and ultrastructure. Pollen grains were collected from flowers at anthesis and both (flowers and pollen) were morphologically characterized. Preliminary studies were done to define the best germination medium and dehydration condition. Pollen samples were then subjected to storage under three conditions: freezer (-5 °C), ultra-freezer (-80 °C), or liquid nitrogen (-196 °C), with or without dehydration, at different intervals. In vitro germination and pollen tube length were assessed at 1, 24 h, 8, 30, 180 and 365 days. Pollen grain morphology and ultrastructure E. H. de Souza (&)  M. L. Rossi  N. Brancalleão  A. P. Martinelli University of São Paulo, Av. Centenário 303, Piracicaba, SP 13416-903, Brazil e-mail: hilosouza@gmail.com Present Address: E. H. de Souza  F. V. D. Souza  C. A. da Silva Ledo Embrapa Cassava & Fruits, Brazilian Agricultural Research Corporation, Cruz das Almas, BA 44380-000, Brazil were assessed at 24 h, 30 and 365 days. The experimental design was completely randomized in a 2 9 3 ? 1 factorial design (2 dehydration conditions, 3 storage conditions, and 1 control). The plot was subdivided by storage time with plots defined by the factorial design, and subplots by storage time and their interaction with plot treatments. The best results were obtained with dehydration and storage in liquid nitrogen (-196 °C) with regard to in vitro germination, pollen tube length, in vivo fertilization and other variables studied, including morphological and ultrastructural integrity. Fruits produced developed normally and produced viable seeds, with germination rates above 92 %. Keywords Bromeliads  In vitro pollen germination  In vivo fertilization  Liquid nitrogen  Microscopy Abbreviations ND Non-dehydrated D Pre-dehydrated FR -5 °C Freezer UF -80 °C Ultra-freezer LN -196 °C Liquid nitrogen Introduction Bromeliads are distinguished in floriculture especially by their exotic appearance, exuberance, beauty, 123 Euphytica postharvest longevity and high market acceptance worldwide. Flower innovation may result from the identification of new species or genotypes with ornamental potential in existing genetic resources or through the generation of new hybrids in breeding programs that utilize currently known species (Souza et al. 2009a). The limited available data on bromeliad breeding potential and compatibility emphasizes the need for detailed studies. Pollen grain storage requires research efforts due to a lack of flowering synchrony among species. In addition, these studies may be useful for genetic resource conservation, haploid production and biochemical, physiological, allergenic and recalcitrant pollen studies (Connor and Towill 1993; Bajaj 1995; Grout and Roberts 1995). Conventional methods of pollen storage involving low temperature and humidity are described for several plant species; papaya (Ganeshan 1986a), pecan (Yates and Sparks 1989), tomato (Sacks and St Clair 1996), Delphinium (Honda et al. 2002), cherimoya (Lora et al. 2006), Dendrobium (Vendrame et al. 2008), annonas (Bettiol Neto et al. 2009) and Picea omorika (Batos and Nikolić 2013). Pollen cryopreservation studies were also applied to different species using various cryoprotection methods; sugarcane (Tai and Miller 2002), Dendrobium (Vendrame et al. 2008), olive (Alba et al. 2011) and Lilium 9 siberia (Xu et al. 2014). Only a few studies have evaluated bromeliad pollen conservation, despite the importance of this family for ornamental and conservation purposes. Parton et al. (1998, 2002) showed successful results using ultra-low temperatures for cryopreservation of A. fasciata, A. chantinii, Vriesea ‘Leen’, Vriesea ‘Christiane’, Tillandsia cyanea and Pitcairnia herdee pollen grains, with viability assessed by pollen grain in vitro germination. In most studies on pollen conservation from different species, viability tests are limited to in vitro germination or histochemical evaluations, rather than seed production after hybridization, which ultimately confirms the efficiency of the process. The possibility of maintenance of pollen viability through the conservation process depends on factors such as flower physiological stage, storage temperature and relative humidity and pollen moisture content (Akihama et al. 1979; Soares et al. 2008). Monitoring viability before, during and after storage is critical for an efficient protocol. Thus, establishing the maximum 123 period for which pollen can be stored without losing germinability and fertility is an important tool for breeding and germplasm exchange (Damasceno et al. 2008; Ganeshan et al. 2008). When applying ultra-low temperatures (-80 °C), the pollen moisture content is crucial for survival and viability because ice crystal formation during storage can disrupt tissue structure (Yates and Sparks 1989; Taylor and Hepler 1997; Ganeshan et al. 2008). Therefore, desiccation processes are recommended, but this step should be done carefully not to turn the pollen unviable due to excessive removal of tissue water (Benson 2008). Preliminary studies should be performed to achieve optimal dehydration according to the species. Parton et al. (1998, 2002) used a 4-hour incubation in silica gel to reduce anther moisture and obtained good results. Towill (1985) reported that pollen moisture content must be reduced to 15–20 % for efficient pollen germination following cryopreservation. Monitoring stored pollen viability can be accomplished by various methods, including histochemistry, in vitro germination, in vivo pollination followed by fruit-set percentage (Galletta 1983). Stored pollen grains should show 50–80 % viability and welldeveloped pollen tubes (Scorza and Sherman 1995). Normally, germination percentage and pollen tube length decreases with time. The goal of conservation is to avoid this aging process and maintain the maximum pollen viability for later use. However, there are several considerations when choosing the viability tests. Some stains may react with cell structures or constituents, leading to confusing results (Stanley and Linskens 1974). Moreover, in vitro germination is not a conclusive method, since in vitro germination does not ensure fertilization. In vivo fertilization is the most efficient method to determine crossing efficiency. A. bicolor L. B. Sm. is a bromeliad species native to Brazil and is distributed in the Atlantic Forest Central Corridor of Biodiversity (Martinelli et al. 2008). Aechmea species are among the most commercially valuable bromeliads (Zhang et al. 2012), with many species in this genus listed as endangered (Martinelli et al. 2008). The present study aimed to study different pollen grain conservation conditions for A. bicolor (Bromeliaceae) with results assessed by viability tests and pollen morphology and ultrastructural evaluations. Initially, flower morphometrical measurements were made and pollen grain germination and viability Euphytica assayed using different culture media and histochemical stains. Pollen grain partial desiccation was also assayed by three methods to determine the best desiccation treatment to be used as a cryopreservation pre-treatment. Pollen conservation methods were tested with or without previous desiccation, followed by evaluation of conservation efficiency by percent of pollen germination, and ultrastructural analyses were performed to characterize the damage caused by the conservation of pollen grains under different treatments. Materials and methods Aechmea bicolor L. B. Sm. plants were grown in a greenhouse at room temperature with approximate relative humidity of 70 %. Flowering was induced by spraying 30 ml EthrelÒ solution (240 ppm) to the rosette of each plant, according to Souza et al. (2009b). Morphometric analysis was done to characterize A. bicolor flowers, using ten flowers obtained from three plants. Anthers were collected from flowers at anthesis, which occurs around 7:30 am. The number of pollen grains per flower was estimated according to the methodology described by Kearns and Inouye (1993). For pollen grain size measurements, grains were acetolyzed through ACLAC 40 (Raynal and Raynal 1971), placed in histological slides and micrographs were obtained in a digital camera under a transmitted light microscope. Both polar and equatorial diameters were measured from 25 pollen grains, randomly chosen, using the ImageJ 1.47r software (Rasband and Image 1997–2012). In vitro pollen grain germination Pollen grains were inoculated in Petri dishes with 25 mL of one of the following culture media: BM: 1.62 mM H3BO3, 584 mM sucrose, 5 g L-1 agar, pH 6.5 (Parton et al. 2002); BK: 1.62 mM H3BO3, 1.27 mM Ca(NO3)24H2O, 0.81 mM MgSO47H2O, 0.99 mM KNO3, 292 mM sucrose, 5 g L-1 agar, pH 6.5 (Brewbaker and Kwack 1963); MBK: 1.62 mM H3BO3, 1.27 mM Ca(NO3)24H2O, 0.81 mM MgSO47H2O, 0.99 mM KNO3, 584 mM sucrose, 5 g L-1 agar, pH 6.5 (BK with twice the amount of sucrose); SM: 1.62 mM H3BO3, 1.27 mM Ca(NO3)24H2O, 0.81 mM MgSO47H2O, 0.99 mM KNO3, 438 mM sucrose, 8 g L-1 agar, pH 6.5 (Soares et al. 2008); MSM: 1.62 mM H3BO3, 1.27 mM Ca(NO3)24H2O, 0.81 mM MgSO47H2O, 0.99 mM KNO3, 350 mM sucrose, 8 g L-1 agar, pH 6.5. Pollen grains were spread on culture medium and incubated at 27 ± 1 °C, each plate divided into four quadrants for further pollen germination and pollen tube growth analysis. After 24 h of inoculation, the cultures were observed under the stereomicroscope (Leica EZ4D) and images were taken using a digital camera. The experimental design for pollen germination was completely randomized with five culture media (BM, BK, MBK, SM, MSM) in 12 replicates, each replicate consisting of one Petri dish quadrant. All grains were counted in each replicate and germination percentage was calculated. Pollen grains were considered germinated when the pollen tube was longer than the pollen grain diameter. Pollen tube length was determined by the average length of five pollen tubes in each of the 12 replicates, randomly chosen. The germination percentage data was transformed by arc sen (Hx/100) prior to statistical analysis. For means comparisons, data was submitted to analysis of variance (ANOVA) followed by Tukey’s test (P B 0.05), using the SAS program (SAS Institute 2004). Histochemical analysis of pollen grain viability For pollen viability analysis, three stains were used: Alexander solution (222 mM lactic acid, Alexander 1980), acetocarmine (10 g L-1, Kearns and Inouye 1993) and Sudan IV (26 mM, Baker and Baker 1979). All six anthers from each flower were placed on a glass slide and pollen grains were removed from the anthers, with anther remains discarded. One drop of a given stain solution was added per slide, and the samples were covered with a coverslip for observations under the light microscope. One hundred pollen grains were counted per slide, with three replications, each consisting of one slide. The percent of viable pollen grains was calculated as an average of the three replicates. The viability data was transformed by arc sen (Hx/ 100) prior to statistical analysis. For means comparisons, data was submitted to analysis of variance (ANOVA) followed by Tukey’s test (P B 0.05), using the SAS program (SAS Institute 2004). 123 Euphytica Pollen grain dehydration Whole anthers were individually placed and wrapped in previously weighed aluminum foil squares, weighed, and anther wet weight calculated. Samples (whole anthers wrapped in aluminum foil) were subjected to dehydration under four desiccating conditions: incubator at 27 ± 1 °C, laminar flow cabinet, silica gel in a desiccator, and incubator at 37 ± 1 °C, during different periods of time (1, 3, 6, 12, 24, 48, 72 h and 8 days). After each period, the samples were weighed to determine the anther weight after each period of desiccation. The pollen grains were then placed to germinate in the best culture medium determined in the preliminary tests mentioned above. Anther moisture was calculated according to Pixton (1966): M ¼ ½ðw  dÞ=ðw  tÞ  100 where M = moisture content of the anther (%), w = anther wet weight, d = anther weight after desiccation and t = tare weight. Pollen grain dehydration tests was composed of 32 treatments, in completely randomized (4 9 8) factorial design, with four desiccating conditions and eight periods of desiccation, with 12 replicates each, with each replicate composed of the pollen grains in one quadrant of a Petri dish. For pollen germination and pollen tube growth evaluations, one representative image was taken under the stereomicroscope, at 3509 magnification, all pollen grains in each image were counted to determine the average percent germination, and five pollen tubes were measured in each quadrant, totaling 60 pollen tubes per treatment, and the average was calculated. Pollen grains were considered germinated when pollen tube length was larger than the pollen grain diameter. The germination percentage data was transformed by arc sen (Hx/100) prior to statistical analysis. For means comparisons, data was submitted to analysis of variance (ANOVA) followed by Tukey’s test (P B 0.05), using the SAS program (SAS Institute 2004). Pollen grain conservation Anthers were removed from flowers, wrapped in aluminum foil, and inserted in 3 mL cryogenic tubes. After selecting the best dehydration condition and 123 culture medium (above), pollen grain conservation was done under three temperatures: -5 °C (freezer); -80 °C (ultra-freezer), and -196 °C (liquid nitrogen), with or without prior dehydration, for periods of 1 h, 24 h, 8, 30, 180 and 365 days. After each period of time, the pollen grains were prepared for in vitro germination. Three plates were used for each treatment, each divided into four quadrants, totaling 12 replicates. Germination percentage and average pollen tube length was determined for each replicate, as previously described. The experimental design was completely randomized, in a 2 9 3 ? 1 factorial (two dehydration treatments, three conservation conditions and one control condition). The control consisted of recently collected pollen (room temperature, 27 ± 1 °C) that was immediately processed for subsequent analysis. Pollen viability was assessed through in vitro germination, in MSM medium, and was carried out before and after the conservation treatments. Fresh pollen was used as an absolute control. The data was subjected to analysis of variance considering the completely randomized design in plots subdivided in time, in which the plots were defined by the factorial scheme and the subplots by time and their interaction with plot treatments. The percentage data was transformed by arc sen (Hx/100) to meet the analysis of variance assumptions. Means of different conservation conditions were compared by Tukey’s test, and means comparison for conservation with or without dehydration was done by the F test. Dunnett’s test was used to compare the control and the treatments. The SAS program was used to perform all statistical analysis, and differences were considered significant if P B 0.05. Pollen grain ultrastructure Pollen morphology, integrity and ultrastructure were evaluated after 24 h, 30 and 365 days of pollen conservation. Pollen samples were placed in three metal stubs for scanning electron microscopy (SEM), and ultrathin sections in copper grids for transmission electron microscopy (TEM) analysis, each subdivided in four quadrants. Twelve replicates were used to calculate the percentage of intact pollen grains. For pollen grain size and exine thickness, 25 grains were randomly measured. Euphytica For SEM, pollen grains were fixed in a modified Karnovsky solution (Karnovsky 1965) (199 mM glutaraldehyde, 2 g L-1 paraformaldehyde, 1 mM CaCl2, 5 mM sodium cacodylate buffer, pH 7.2) for 48 h and dehydrated in an ethanol series, followed by drying in HMDS (hexamethyldisilazane). Dried pollen grains were mounted on metal stubs and sputter coated with gold. Observations were made under the scanning electron microscope (LEO 435VP, Carl Zeiss, Germany) and digital micrographs were taken. For TEM, pollen grains were fixed in the same modified Karnovsky solution (Karnovsky 1965) for 24 h, washed three times in sodium cacodylate buffer (100 mM), and post-fixed in osmium tetroxide (39 mM) in sodium cacodylate buffer (100 mM), washed in saline solution (154 mM sodium chloride) and ‘‘en block’’ stained in uranyl acetate (59 mM). Samples were dehydrated in an acetone series, embedded in Spurr resin, and polymerized for 48 h, at 70 °C. Semi-thin sections were placed on glass slides, stained with toluidine blue (1.85 mM) and covered with EntelanÒ and a coverslip, for light microscopy observations (Zeiss Axioskop 2 inverted microscope) and images were captured with a digital camera. Ultrathin sections were obtained on an ultramicrotome (Sorvall Porter Blum MT2, Norwalk, USA), placed on copper grids previously coated with formvar and post-stained in uranyl acetate (59 mM) and lead citrate, according to Reynolds (1963). Observations were made under a Zeiss EM900 transmission electron microscope (Carl Zeiss, Germany) and micrographs taken with a digital camera. In vivo pollen grain viability test Flowers from different plants of A. bicolor were used as the female parent for controlled pollination with pollen grains from each conservation treatment, at least 12 flowers per treatment. Anthers were removed prior to anthesis to prevent self-pollination, and flowers were bagged to avoid contamination from pollen from other plants. After emasculation, flowers were hand pollinated using pollen grains from each conservation treatment, and each flower was individually tagged and bagged. Number of fruits and seeds formed were assessed after the fruits ripened. The seeds were germinated in vitro, in MS medium (Murashige and Skoog 1962), according to Souza et al. (2009c). Results Aechmea bicolor develops a small indeterminate inflorescence with the flowers organized in a spike (Fig. 1a). Flowers are tubular; sepals are orange, and white petals become visible at anthesis (Fig. 1b–c). The androecium has six dorsifixed rimose anthers, with filaments adnate to petals. The gynoecium is composed a single pistil with a tricarpelate inferior ovary, with numerous ovules in an axillar placentation, with interlobular septal nectary present. Ovules are anatropous, with attenuated micropyle, ovule appendage in the chalazal region is absent. Stigma is conduplicate-spiral, and papillae are slightly prominent. The fruit is an ellipsoid-shaped berry. The morphometric analysis of the inflorescence, floral organs and pollen grains are presented (Table 1). A. bicolor pollen grains are medium sized, biporate, with bilateral symmetry, prolate shape, psilate (Fig. 1d–e), with a tectate exine. The estimated production of pollen grains per flower was 32,916 grains, with an average of 5,486 pollen grains per anther. Pollen germination varied according to the culture media used. No germination was observed in BM medium (Figs. 1f, 2a). The highest pollen germination percentage was observed in SM and MSM media (Figs. 1i–j, 2a), and significant longer pollen tubes occurred in MSM medium (Figs. 1j, 2b). Therefore, MSM was used for subsequent analysis of pollen grains after conservation assays. Histochemical tests showed that all three stains used were efficient in determining pollen viability for this species, with easy detection of viable and nonviable pollen grains (Fig. 1k–m). Pollen grain viability was above 94 % (Fig. 2c), which was similar to the results obtained for the germination percentage when SM or MSM media were used. There was no significant difference in pollen grain viability percentage among the three histochemical tests used. In the preliminary tests for pollen grain desiccation, the use of silica gel in a desiccator for 3 h gave the best overall results, with good results for pollen germination (95 %) and pollen tube growth (0.6 mm after 24 h in culture medium) and a (Table 2). In these conditions, pollen grain humidity was around 40 %, which is considered a safe condition for pollen conservation (data not shown). Longer periods of desiccation reduced pollen germination and humidity gradually, 123 Euphytica Fig. 1 Aechmea bicolor flower, pollen grains, and pollen grain viability tests. a A. bicolor inflorescence. b General view of the flower. c Detailed view of the flower showing the floral organs. d–e Pollen grains in polar view (pv) and equatorial view (ev) under the scanning electron microscope (d) and acetolyzed pollen grains in equatorial view, showing the pores (arrowhead). f–j In vitro pollen germination showing different germination percentages in different culture media: f BM, g BK, h MBK, i SM, j MSM. k–m Histochemical tests showing viable and non-viable (arrows) pollen grains. Alexander (k), Acetocarmine (l), Sudan IV (m). an anther, fi filament, ov ovary, oe ovule, pe petal, se sepal, st stigma, sy style. Bars: b–c 2 mm; d, e, k–m 20 lm; f–j 0.5 mm with rates of pollen germination lower than 15 % after 3 days. Hence, a period of 3 h in a desiccator was used as a pre-desiccation treatment in subsequent tests. There was a significant interaction among the different methods of dehydration, conservation and the periods of time of exposure of anthers and pollen grains for the parameters percent of pollen germination, pollen tube length (Table 3) and morphological characters (Table 4). The control treatment (room temperature, 27 ± 1 °C) showed pollen viability up to 24 h, with a drastic reduction at day 8 and total loss of viability at day 30, both for in vitro germination and 123 Euphytica Table 1 Morphometric characterization of Aechmea bicolor flowers, floral parts and pollen grains Flower* Petal length (mm) 9.42 ± 0.29 Petal width (mm) 3.32 ± 0.16 Sepal length (mm) 6.36 ± 0.35 Sepal width (mm) Number of flowers per inflorescence 3.28 ± 0.07 46.90 ± 4.04 Androecium Pollen grain (n = 25) Polar diameter in equatorial view (lm) 21.95 ± 1.56 Equatorial diameter in equatorial view (lm) 31.63 ± 1.34 Tectum (lm) 0.64 ± 0.06 Columella (lm) Sexine (lm) 0.23 ± 0.03 0.86 ± 0.07 Nexine (lm) 0.54 ± 0.08 Exine (lm) 1.41 ± 0.09 Anther* Length (mm) 4.01 ± 0.07 Width (mm) 0.97 ± 0.02 Number 6 Filament* Length (mm) 6.68 ± 0.70 Diameter (mm) 0.49 ± 0.02 Gynoecium* Ovary Length (mm) 4.29 ± 0.16 Diameter (mm) 4.44 ± 0.33 Ovule Length (lm) Diameter (lm) 356.61 ± 13.42 170.39 ± 10.34 Number per locule 71.00 ± 5.29 Micropyle width (lm) 43.63 ± 3.60 Stigma Length (mm) 1.07 ± 0.12 Diameter (mm) 0.70 ± 0.06 Style Length (mm) 5.10 ± 0.06 Diameter (mm) 0.72 ± 0.02 * The results correspond to an average of measurements of ten flowers obtained from three plants ± standard deviation pollen tube length. This result was similar to pollen grain conservation in the freezer (-5 °C) independently of the dehydration, while pre-dehydrated pollen grains showed slightly higher viability values (Tables 3, 4, 5). Ultra-freezer conservation (-80 °C) resulted in a higher loss of pollen germination after the first hour of conservation, these values remained stable for up to 1 year of conservation, differently from the control, which showed progressive loss of pollen germination (Table 3). Higher in vitro germination and pollen tube length were observed in the conservation treatment composed of dehydration followed by storage in liquid nitrogen (-196 °C) (Fig. 3d, Table 3). Morphology and ultrastructure of the pollen grains in this treatment (Fig. 3l, t, z, ab, ac) were comparable to the control treatment (Fig. 3e, u; Tables 4, 5). After 24 h of conservation treatments the pollen grains already showed some morphological damage, observed under the scanning electron microscope, with differences among treatments. At this stage the conservation treatment under liquid Nitrogen, with or without pre-dehydration were similar to the control. However, as storage time increased, there a higher number of damaged pollen grains for all treatments, except pollen grains conserved in liquid nitrogen after dehydration (Table 4). Control treatment at day 30 showed 100 % damaged pollen grains and size reduction, confirming the need for the definition of appropriate A. bicolor pollen conservation methods (Tables 3, 4; Fig. 3f, m). When pollen conservation was done in the freezer (-5 °C) a detachment of the intine from the exine was observed and pollen grain size was reduced (Fig. 3g, p, r, w). Freezing pollen grains directly in liquid nitrogen (-196 °C), without prior dehydration, caused pollen grain rupture in approximately 20 % of the grains (Fig. 2i, o, s, y). On the other hand, predehydration of pollen grains prior to liquid nitrogen immersion yielded 96.25 % of undamaged pollen grains, highlighting the need for the pre-dehydration procedure (Table 4; Fig. 3l, t, a, ab and ac). Cytoplasmic degradation was observed under the transmission electron microscope, in conservation treatments without previous dehydration, at both 24 h, and day 30th (Fig. 3v, w, x), with the exception of the freezer treatment (-5 °C) that showed degraded cytoplasmic content even after dehydration. The exine morphology and thickness of the different layers were not significantly different independently of the conservation and dehydration treatment (Table 5). 123 Euphytica Fig. 2 Aechmea bicolor pollen grain viability and pollen tube length. a Pollen grain germination percentage in different culture media. b Pollen tube length (mm) after 24 h in different culture media. c Percent viability of pollen grains in histochemical tests. Means followed by different letters in each bar graph are significantly different according to the Tukey’s test (P B 0.05) The controlled pollination experiments done with pollen grains stored for 365 days in different treatments resulted in different percentages of fruit set and number of seeds produced per fruit (Table 6). A higher percentage of fruit set, as well as number of seeds, was obtained when pollen grains were predehydrated and conserved under liquid Nitrogen (196 °C). These results were similar to the control (fresh pollen grains). The number of seeds produced either in the control, or when pollen grains were predehydrated and conserved in liquid nitrogen were lower than the average of number of ovules developed per flower (210 ovules/flower). Absence of fruit set was observed in the control treatment (storage without conservation treatment) and when pollen grains were conserved in the freezer (-5 °C) without previous dehydration. All other treatments resulted in fruit set with 92 % of seed germination in culture medium 8 days after inoculation. Ultra-freezer (-80 °C) treatments with and without dehydration formed 67 and 83 % of fruits, respectively, but the percentages of seeds produced were only 5.44 and 30.15 %, respectively, when compared to the control and dehydration/liquid nitrogen conservation -96 °C) treatments. other than leaf color and size, and flower morphology, which may be more prone to changes when compared to pollen characteristics, which is considered to be highly conserved (Benzing 2000). Pollen grains are also key factors for breeding and species conservation. The availability of pollen grains for interspecific and intergeneric crosses is highly dependent on the possibility of long-term conservation, and it is important to develop efficient methods to evaluate pollen viability and conservation methods. In this study, we assessed the quality of pollen grain conservation methods through in vitro germination, histochemical tests to determine pollen grain viability and in vivo fertilization using the conserved pollen grains of A. bicolor. Although these tests can be used to compare different conditions, in vitro germination does not fully reproduce in vivo pollen tube growth and its fertilization ability. The best in vitro pollen germination and pollen tube length results were obtained in MSM medium, which has the complete nutrient composition and 350 mM sucrose. We found the lowest germination values in BM medium, which is in contrast to the results obtained by Parton et al. (2002), who worked with seven bromeliad species, different from the species studied in the present work. The histochemical viability tests presented similar results to those of in vitro germination experiments, with values above 94.33 % viability, demonstrating that both approaches can give reliable results for pollen grain evaluations. According to Galletta (1983), in order to maintain pollen grain integrity there must exist an osmotic balance between the cytoplasmic content and the culture medium. This balance is achieved by the Discussion Detailed morphometric, morphological and ultrastructural characterization of species are important for species delimitation. The use of pollen grain morphology can be an alternative for taxonomic studies, 123 Euphytica Table 2 Aechmea bicolor pollen grain characteristics after exposure to different drying treatments and periods of time Period of time Incubator at 27 ± 1 °C Laminar flow cabinet–room temperature Desiccator with silica gel at room temperature Oven at 37 ± 1 °C 1h 84.74aA 83.55aA 70.92aB 47.26aC 3h 62.78bA 39.96bB 38.93bB 39.65bB 6h 49.78cA 30.64cB 24.73cC 35.23bB 12 h 40.95dA 30.92cB 24.30cC 34.19bB 1 day 38.73dA 30.86cB 23.09cC 39.90bA 2 days 38.73dA 30.69cB 22.15cC 38.31bA 3 days 38.64dA 30.43cB 21.51cC 37.95bA 8 days 38.00dA 29.88cB 20.79cC 29.79bB CV (%) 12.44 Anther humidity (%) Pollen grain germination (%) 1h 98.08aA 97.16aA 98.08aA 91.83aB 3h 97.58aA 91.66bB 95.16aA 69.00bC 6h 84.91bA 72.00cB 67.66bC 56.58dD 12 h 1 day 85.33bA 73.50cA 69.83cB 69.41cA 68.41bB 71.41bA 61.25cC 40.00eB 2 days 42.91 dB 63.41dA 37.50cC 23.66fD 3 days 11.50eA 12.08eA 13.33dA 3.08gB 8 days 3.25fA 0.66fB 4.91eA 0.00hB CV (%) 8.10 Pollen tube length (mm) Means followed by different uppercase letter in the row and lowercase letter in the column within the same factor are significantly different according to Tukey’s test (P B 0.05) 1h 0.63 ± 0.11aA 0.58 ± 0.10aB 0.59 ± 0.10aB 0.52 ± 0.10aC 3h 0.58 ± 0.09bA 0.51 ± 0.08bB 0.60 ± 0.08aA 0.41 ± 0.10bC 6h 0.53 ± 0.12cA 0.41 ± 0.10cC 0.44 ± 0.11bB 0.38 ± 0.10bC 12 h 0.52 ± 0.10cA 0.42 ± 0.11cB 0.41 ± 0.11bB 0.38 ± 0.11bB 1 day 0.52 ± 0.07cA 0.41 ± 0.11cB 0.40 ± 0.10bB 0.28 ± 0.09cC 2 days 0.27 ± 0.08 dB 0.40 ± 0.09cA 0.27 ± 0.08cB 0.19 ± 0.08dC 3 days 0.20 ± 0.10eA 0.19 ± 0.11dA 0.20 ± 0.10dA 0.16 ± 0.08dA 8 days 0.14 ± 0.10fA 0.14 ± 0.14eA 0.16 ± 0.13eA 0.00eB CV (%) 26.70 concentrations of sucrose and other substances in the culture medium, including boric acid and calcium nitrate. Thus, an excess or deficiency of any of these components may promote pollen grain disruption, which affects the test results. In addition to being essential for the osmotic balance, sucrose plays a significant role as an energy source for the pollen tube growth (Stanley and Linskens 1974). Some authors believe that in addition to necessary carbohydrates, culture medium should contain elements such as boric acid, calcium nitrate, potassium nitrate and magnesium sulfate (Brewbaker and Kwack 1963; Kearns and Inouye 1993; Taylor and Hepler 1997). Other environmental factors can interfere with pollen viability. When anther opening coincides with high humidity, high osmotic pressure of the cellular content of the pollen combined with a low resistance of the exine, reduces pollen viability (Kearns and Inouye 1993). The viability results of the present study obtained with the different stains was similar to the in vitro germination results. Reports from other studies, however, mention the low reliability of the histochemical tests, which overestimate the viability results (Parfitt and Ganeshan 1989; Soares et al. 2008). Parfitt and Ganeshan (1989) made pollen unviable by exposure to high temperatures, and, yet, the histochemical tests 123 123 Table 3 In vitro germination of Aechmea bicolor pollen grains and pollen tube growth after different conservation and dehydration treatments and periods of time in each treatment Conservation Treatments 1h ND 24 h 8 days 30 days 180 days 365 days D ND D ND D ND D ND D ND D 4.83cA** Pollen germination (%) FR (-5 °C) 75.17bB** 85.58bA** 43.00bB** 67.42bA** 3.75cB** 8.75cA** 0.00cB** 4.25cA** 0.00cB** 4.42cA** 0.00cB** UF (-80 °C) 28.58cB** 77.67cA** 28.58cB** 77.67bA** 25.42bB** 59.25bA** 22.50bB** 56.42bA** 21.75bB** 55.83bA** 22.25bB** 54.91bA** LN (-196 °C) 86.17aB** 94.00aAns 80.42aB** 89.25aAns 65.42aB** 87.17aAns 61.50aB** 86.50aAns 61.58aB** 86.92aAns 60.08aB** 87.58aAns CV (%) 12.19 Pollen tube length (mm) FR (-5 °C) 0.38cA** 0.34bA** 0.34cA** 0.35bA** 0.17cA** 0.19cA** 0.00cB** 0.15cA** 0.00cB** 0.15cA** 0.00cB** 0.15cA** UF (-80 °C) 0.52aA* 0.36bB** 0.56aA* 0.32bB** 0.49aA* 0.32bB** 0.51aA* 0.35bB** 0.52aA* 0.36bB** 0.51aA* 0.35bB** LN (-196 °C) 0.45bB** 0.59aAns 0.44bB** 0.58aAns 0.39bB** 0.60aAns 0.42bB** 0.58aAns 0.42bB** 0.60aAns 0.43bB** 0.61aAns CV (%) 30.92 ND non-dehydrated; D pre-dehydrated; FR -5 °C freezer; UF -80 °C ultra-freezer; LN -196 °C liquid. ns non-significant; ns non-significant Means followed by the same uppercase letter in the row and lowercase in the column within the same factor are not significantly different according to the Tukey’s test (P B 0.05). Dunnett’s test (P B 0.05) * Significantly different at P B 0.05; ** significantly different at P B 0.01 Euphytica Euphytica Table 4 Aechmea bicolor pollen grains evaluated through scanning electron microscopy after different conservation treatments and periods of time in each treatment Conservation Treatments 24 h ND 30 days 365 days D ND D ND D 40.08cA** Undamaged pollen grains (%) FR (-5 °C) 78.25bA** 65.75cB** 20.58cB** 39.50cA** 21.08cB** UF (-80 °C) 68.08cB** 89.42bA** 66.83bB** 89.17bA** 65.00bB** 89.08bA** LN (-196 °C) 91.25aB** 96.66aAns 84.33aB** 96.25aAns 83.50aB** 96.08aAns CV (%) 8.32 Polar diameter (lm) FR (-5 °C) 19.86 ± 1.71aA 19.28 ± 1.55bB 19.66 ± 1.47bB UF (-80 °C) 20.58 ± 1.62aA 20.83 ± 1.61aA 21.02 ± 1.32aA LN (-196 °C) 21.88 ± 1.71aA 21.50 ± 1.94aA 21.26 ± 1.44aA CV (%) 6.06 Equatorial diameter (lm) FR (-5 °C) 28.53 ± 2.64bA 27.67 ± 1.84bA 28.13 ± 1.84bA UF (-80 °C) 29.61 ± 2.11bA 29.73 ± 1.99aA 29.63 ± 1.88bA LN (-196 °C) 31.65 ± 1.78aA 31.92 ± 1.77aA 32.08 ± 1.56aA 3.69 Non-dehydrated 20.74 ± 1.77aA 20.24 ± 2.03bB 20.55 ± 1.89bB Pre-dehydrated 20.80 ± 1.97aA 20.83 ± 1.81aA 20.31 ± 1.62aA CV (%) 6.06 ND non-dehydrated; D pre-dehydrated; FR -5 °C freezer; UF -80 °C ultra-freezer; LN -196 °C liquid. ns non-significant Means followed by the same uppercase letter in the row and lowercase letter in the column within the same factor are not significantly different according to the Tukey’s test (P B 0.05). Dunnett’s test (P B 0.05) * Significantly different at P B 0.05; ** significantly different at P B 0.01 detected high levels of viability in the pollen grains. Histochemical overestimation of pollen viability was also found by Parton et al. (2002), who assessed the pollen grains of bromeliads before anthesis and after conservation at 21 °C. Einhardt et al. (2006) stated that although staining is a straightforward and inexpensive procedure, it may overestimate viability values. Pollination with pollen grains that had been dehydrated and cryopreserved for 365 days produced satisfactory amounts of fruits and seeds that were comparable to pollination with fresh pollen, demonstrating that the storage method maintains pollen integrity and viability. Ganeshan (1986b) found similar results for onion, with 81.05 % fruit formation using pollen that had been conserved in liquid nitrogen for 365 days. Vendrame et al. (2008) carried out Dendrobium orchid crosses using pollen grains conserved in liquid nitrogen with and without a vitrification treatment, and obtained successful hybridizations with both methods. Stanley and Linskens (1974) found a high correlation between in vitro germination and controlled fertilization in the field. However, they claimed that viability is not accurately estimated even in in vivo pollination because fertilization and number of seeds in maize do not strictly depend on pollen viability and fertility, but also in the nutritional status of the mother plant, style-stigma receptivity and environmental conditions in which pollination was performed. Interestingly, in our work pollen was collected from inflorescences at the same physiological state. According to Franzon and Raseira (2006), breeding programs must use pollen in the proper maturity stage so that pollen maintains viability and is able to germinate when hybridization is performed. With regard to storage results, the assessment of different dehydration methods was fundamental for establishing which 123 Euphytica conditions should be used in the subsequent experiments. Among the different dehydration methods assessed, silica was the most promising. It showed the highest reproducibility out of all of the methods assessed, which is similar to that found for several other species of bromeliads (Parton et al. 2002). According to Engelmann et al. (2008), dehydration methods in laminar flow hood yield variable rates and are dependent on several factors; therefore, they are not recommended. We chose a treatment with silica for 3 h not only because of the good results in in vitro germination and pollen tube length after dehydration, but also because these results were associated with a lower moisture content when compared to the others. The low moisture content is significant considering that the probability of keeping pollen intact after freezing can be increased by avoiding ice crystal formation and the subsequent cellular structure disruption. The lower the water content within the cell the lower the chance of structural damages. Table 5 Morphometric data of exine layer thickness through SEM of different storage treatments and assessment times Treatments Fig. 3 Aechmea bicolor pollen grains after different conserva- c tion treatments. a–d In vitro pollen germination in MSM medium: control at 6 months (a), ultra-freezer (-80 °C) (b), LN (-196 °C) (c), dehydration ? LN (-196 °C) (d); e–p pollen grains after different conservation treatments observed by scanning electron microscopy: control at 1 h (e), control at day 30 (f), freezer (-5 °C) at day 30 (g), ultra-freezer (-80 °C) at day 30 (h), LN (-196 °C) at day 30 (i), dehydration ? freezer (-5 °C) at day 30 (j), dehydration ? ultrafreezer (-80 °C) at day 30 (k) dehydration ? LN (-196 °C) at day 30 (l), m–p details of damages in pollen grains: pollen grain shrinkage in the control at day 30 (m), exine rupture in ultrafreezer (-80 °C), at day 30 (n), exine rupture and exposure of the cytoplasm in grains without prior dehydration and conserved in liquid nitrogen (-196 °C), at day 30 (o), detachment of the cytoplasm from the exine in freezer (-5 °C), at day 30 (p); q–ac pollen grains observed under light (q–t) and transmission electron microscopy (u–ac) after different conservation treatments demonstrating the cytoplasmic and exine integrity: control at day 30 (q), freezer (-5 °C) at day 30 (r), LN (-196 °C) at day 30 (s), dehydration ? LN (-196 °C) at day 30 (t), control at 1 h (u), control at day 30 (v), freezer (-5 °C) at day 30 (w), ultra-freezer (-80 °C) at day 30 (x), LN (-196 °C) at day 30 demonstrating exine rupture (er) (y), dehydration ? LN (-196 °C) at day 30 (z, ab, ac). po pore. Bars: a– d 0.5 mm; e–t 20 lm; u–z, ab 2 lm; ac 10 lm Tectum (lm) Columella (lm) Sexine (lm) Nexine (lm) Exine (lm) Control 0.64 ± 0.06 0.23 ± 0.03 0.87 ± 0.07 0.54 ± 0.09 1.41 ± 0.09 ND 1 FR (-5 °C) 0.66 ± 0.05 0.23 ± 0.05 0.89 ± 0.07 0.53 ± 0.08 1.42 ± 0.10 ND 1 UF (-80 °C) 0.64 ± 0.07 0.23 ± 0.05 0.87 ± 0.09 0.58 ± 0.09 1.44 ± 0.16 ND 1 LN (-196 °C) 0.64 ± 0.06 0.24 ± 0.06 0.88 ± 0.10 0.53 ± 0.07 1.41 ± 0.13 D 1 FR (-5 °C) 0.63 ± 0.10 0.25 ± 0.04 0.89 ± 0.09 0.55 ± 0.08 1.44 ± 0.13 D 1 UF (-80 °C) 0.65 ± 0.06 0.25 ± 0.05 0.90 ± 0.08 0.57 ± 0.06 1.47 ± 0.09 D 1 LN (-196 °C) 0.63 ± 0.09 0.25 ± 0.04 0.88 ± 0.11 0.52 ± 0.08 1.40 ± 0.14 Control 0.62 ± 0.08 0.24 ± 0.05 0.86 ± 0.10 0.55 ± 0.09 1.41 ± 0.10 ND 1 FR (-5 °C) 0.64 ± 0.06 0.24 ± 0.05 0.87 ± 0.09 0.53 ± 0.08 1.41 ± 0.11 ND 1 UF (-80 °C) 0.61 ± 0.08 0.25 ± 0.06 0.85 ± 0.12 0.57 ± 0.13 1.42 ± 0.24 ND 1 LN (-196 °C) 0.61 ± 0.07 0.26 ± 0.06 0.86 ± 0.11 0.53 ± 0.07 1.40 ± 0.14 D 1 FR (-5 °C) 0.63 ± 0.10 0.25 ± 0.04 0.89 ± 0.09 0.55 ± 0.09 1.44 ± 0.15 D 1 UF (-80 °C) 0.65 ± 0.07 0.24 ± 0.05 0.90 ± 0.08 0.59 ± 0.07 1.49 ± 0.11 D 1 LN (-196 °C) 0.61 ± 0.10 0.24 ± 0.04 0.86 ± 0.11 0.59 ± 0.16 1.45 ± 0.21 Control 0.61 ± 0.07 0.24 ± 0.05 0.85 ± 0.09 0.53 ± 0.20 1.42 ± 0.10 ND 1 FR (-5 °C) 0.62 ± 0.07 0.24 ± 0.05 0.86 ± 0.09 0.52 ± 0.11 1.38 ± 0.09 ND 1 UF (-80 °C) 0.61 ± 0.09 0.25 ± 0.06 0.86 ± 0.09 0.55 ± 0.08 1.41 ± 0.11 ND 1 LN (-196 °C) 0.61 ± 0.07 0.24 ± 0.07 0.85 ± 0.07 0.55 ± 0.09 1.40 ± 0.10 D 1 FR (-5 °C) 0.62 ± 0.09 0.24 ± 0.06 0.86 ± 0.09 0.56 ± 0.10 1.42 ± 0.15 D 1 UF (-80 °C) 0.61 ± 0.07 0.24 ± 0.06 0.85 ± 0.06 0.58 ± 0.09 1.43 ± 0.11 D 1 LN (-196 °C) 0.63 ± 0.07 0.25 ± 0.04 0.88 ± 0.08 0.55 ± 0.11 1.43 ± 0.10 12.58 20.12 11.11 16.88 10.34 24 h 30 days 12 months ND non-dehydrated; D predehydrated; FR -5 °C freezer; UF -80 °C ultra-freezer; LN -196 °C liquid nitrogen 123 CV (%) Euphytica A comparison of treatments with or without dehydration, allowed for the observation that this procedure was essential for promoting higher in vitro germination and pollen tube length, as well as the maintenance of the pollen grain ultrastructural integrity. 123 Euphytica Table 6 Fruits and seeds produced after controlled cross-pollination of Aechmea bicolor flowers using pollen grains conserved for 365 days under different conservation treatments Conservation treatments Number of pollinated flowers Percent of fruit set Seeds per fruit (x ± S) Control (fresh pollen grains) 25 100 Control (storage w/o treatment) 18 0 0 133.33 ± 22.55 ND ? FR (-5 °C) 12 0 0 ND ? UF (-80 °C) 12 67 7.25 ± 2.55 ND ? LN (-196 °C) 12 58 9.14 ± 1.95 D ? FR (-5 °C) 12 25 12.00 ± 5.29 D ? UF (-80 °C) 12 83 40.20 ± 8.18 D ? LN (-196 °C) 18 100 129.33 ± 23.23 ND non-dehydrated; D pre-dehydrated; FR -5 °C freezer; UF -80 °C ultra-freezer; LN -196 °C liquid nitrogen The conservation of pollen grains directly at -80 °C produces the formation of ice crystals. This is possibly the result of the physical properties of water at ultra-low temperatures and the slower freezing at -80 °C, compared to -196 °C, which allows ice crystal nucleation. Ice crystal formation affects the osmotic, colligative and structural integrity of cells, which leads to physical ruptures and mechanical damage (Benson 2008). In our study, microscopical analysis confirmed the pollen grain damage in conservation treatments without prior dehydration (Fig. 3q, r, v, w and x). Besides cytoplasmic degradation, exine rupture was also observed, which explains the loss of pollen viability following these treatments (Fig. 3s, y). Regarding the best pollen conservation method for A. bicolor, we found that freezing in liquid nitrogen with previous dehydration in silica for 3 h showed the best results for all parameters analyzed. The final assessment of in vivo pollination after 365 days of pollen conservation unequivocally corroborated these results. Finally, Dunnett’s t test, which compares experimental means with the control, showed that only the treatments in liquid nitrogen were not significantly different. These results are similar to those obtained by Parton et al. (2002) for other species of Bromeliaceae, including A. fasciata, A. chantinii, Vriesea ‘Leen’, Vriesea ‘Christiane’, Tillandsia cyanea and Pitcairnia herdee, when assessing pollen conservation in liquid nitrogen. The cessation of cellular metabolism can be confirmed after 24 h of freezing, when the values of pollen germination or tube length were no longer 123 changing. This characteristic of the ultra-freezing method (-80 °C) with prior maintenance of cellular integrity is essential for long-term conservation (Withers 1991). Morphological and ultrastructural changes were observed in pecan pollen grains that were dehydrated and conserved (Yates and Sparks 1989). These authors observed pollen grain shrinkage with intine separation, also found in the present study. Other deformations, including pollen rupture and cytoplasmic content release in the pore region, were detected when pollen was subjected to -196 °C without previous dehydration. Similar observations, such as cytoplasmic content release and exine layer disruption, were noted in A. bicolor (Fig. 3o, s, y). Ultrastructural differences in cytoplasmic content were observed for the different treatments. In ultrafreezer treatments at temperatures above -80 °C, the cytoplasmic content was irregularly distributed. This change was not observed in the treatment with dehydration and subsequent conservation in liquid nitrogen (-196 °C), which remained similar to the control treatment for 1 h. Yates and Sparks (1989) used light microscopy and observed differences in the cytoplasmic content of pecan pollen grains conserved at -10 °C, however not in pollen that was dehydrated and stored in liquid nitrogen. With regard to the exine layers, the deformations were due to their rupture; there were no significant differences in the thickness of the layers among the different treatments. The sturdy and rigid layer of sporopollenin (oxidized polymers and esters of carotenoids) that forms the exine explains this, since the intine is formed by cellulose and pectin, which are Euphytica less resistant and more homogeneous, thus not prone to a reduction in thickness (Hesse et al. 2009). No pollen conservation studies are reported in the literature, with detailed analysis done by electron microscopy, which allows for the observation of the damage here reported. Thus, these techniques are an innovation to demonstrate the damage effects of the inadequate pollen conservation methods. These results confirm the possibility of conserving A. bicolor pollen grains and provide breeders with an alternative for carrying out crosses with asynchronous flowering material. Moreover, long-term pollen storage may become feasible for carrying out various physiological studies, among others. Similar research efforts should be encouraged for bromeliad species, for their increasing importance in the ornamental agribusiness, with tremendous potential for the creation of novelties through breeding. Acknowledgments The authors acknowledge the support of Fundação de Amparo a Pesquisa do Estado de São Paulo— FAPESP (2009/18255-0) and Conselho Nacional de Desenvolvimento Cientı́fico e Tecnológico—CNPq (305.785/2008-7 and 476.131/2008-1), for financial support, and Núcleo de Apoio à Pesquisa em Microscopia Eletrônica Aplicada à Agricultura, Escola Superior de Agricultura ‘‘Luiz de Queiroz’’, Universidade de São Paulo, for the use of the microscopic facilities. References Akihama T, Omura M, Kosaki I (1979) Long-term of fruit tree pollen and its application in breeding. Trop Agric Res 13:238–241 Alba V, Bisignano V, Alba E, Stradis A, Polignano GB (2011) Effects of cryopreservation on germinability of olive (Olea europaea L.) pollen. Genet Resour Crop Evol 58:977–982. doi:10.1007/s10722-011-9736-z Alexander MP (1980) A versatile stain for pollen, fungi, yeast and bacteria. Stain Technol 55:13–18 Bajaj YPS (1995) Cryopreservation of plant cell, tissue and organ culture for the conservation of germplasm and biodiversity. In: Bajaj YPS (ed) Biotechnology in agriculture and forestry 32. Cryopreservation of plant germplasm I, vol 32. Springer-Verlag, New York, pp 3–18 Baker HG, Baker I (1979) Starch in angiosperm pollen grains and its evolutionary significance. Am J Bot 66:591–600 Batos BZ, Nikolić BM (2013) Variability of in vitro germination of Picea omorika pollen. Dendrobiology 69:13–19 Benson EE (2008) Cryopreservation theory. In: Reed BM (ed) Plant Cryopreservation, A practical guide. Springer, New York, pp 15–32 Benzing DH (2000) Bromeliaceae: profile an adaptive radiation. Cambridge University Press, Cambridge Bettiol Neto JM, Nero MD, Kavati R, Pinto-Maglio CAF (2009) Viability and conservation of pollen from tree commercial annonas. Bragantia 68:825–837. doi:10.1590/S000687052009000400002 Brewbaker JL, Kwack BH (1963) The essential role of calcium ion in pollen germination and pollen tube growth. Am J Bot 50:859–865 Connor KF, Towill LE (1993) Pollen-handling protocol and hydration/dehydration characteristics of pollen for application to long-term storage. Euphytica 68:77–84. doi:10. 1007/BF00024157 Damasceno Junior PC, Pereira TNS, Pereira MG, Silva FF (2008) Conservation of papaya (Carica papaya L.) pollen grain. Rev Ceres 55:433–438 Einhardt PM, Correa ER, Raseira MC (2006) Comparação entre métodos para testar a viabilidade de pólen de pessegueiro. Rev Bras Frutic 28:5–7. doi:10.1590/S0100-29452006000 100003 Engelmann F, Arnao MTG, Wu Y, Escobar R (2008) Development of encapsulation dehydration. In: Reed BM (ed) Plant Cryopreservation. A practical guide. Springer, New York, pp 59–76 Franzon RC, Raseira MCB (2006) Germinação in vitro e armazenamento do pólen de Eugenia involucrata DC (Myrtaceae). Rev Bras Frutic 28:18–20. doi:10.1590/S0100-2945200 6000100007 Galletta GJ (1983) Pollen and seed management. In: Moore JN, Janinick J (eds) Methods in fruit breeding. Purdue University Press, West Lafayette, pp 23–47 Ganeshan S (1986a) Cryogenic preservation of papaya pollen. Sci Hortic 28:65–70. doi:10.1016/0304-4238(86)90125-1 Ganeshan S (1986b) Viability and fertilizing capacity of onion pollen (Allium cepa L.) stored in liquid nitrogen (-196 & #xB0;C). Trop Agric 63:46–48 Ganeshan S, Rajasekharan PE, Shashikumar S, Decruze W (2008) Cryopreservation of Pollen. In: Reed BM (ed) Plant cryopreservation: a practical guide. Springer, New York, pp 281–332 Grout BWW, Roberts AV (1995) Storage of free pollen, pollen embryos and the zygotic embryos of seed by cryopreservation and freeze drying. In: Grout BWW (ed) Genetic preservation of plant cells in vitro. Springer-Verlag, Berlin, pp 63–74 Hesse M, Halbritter H, Weber M, Buchner R, Frosch-Radivo A, Ulrich S (2009) Pollen terminology an illustrated handbook. Springer Wien, New York Honda K, Watanabe H, Tsutsui K (2002) Cryopreservation of Delphinium pollen at -30 & #xB0;C. Euphytica 126:315– 320. doi:10.1023/A:1019931304742 Karnovsky MJ (1965) A formaldehyde-glutaraldehyde fixative of high osmolality for use in electron microscopy. J Cell Biol 27:137A Kearns CA, Inouye DW (1993) Techniques for pollination biologist. University of Colorado, Niwot Lora MAJ, Oteyza P, Fuentetaja P, Hormaza JI (2006) Low temperature storage and in vitro germination of cherimoya (Annona cherimola Mill) pollen. Sci Hortic 1:91–94. doi:10.1016/j.scienta.2005.12.003 Martinelli G, Vieira CM, Gonzalez M, Leitman P, Piratininga A, Costa AF, Forzza RF (2008) Bromeliaceae da Mata 123 Euphytica Atlântica brasileira: lista de espécies, distribuição e conservação. Rodriguesia 59:209–258 Murashige T, Skoog FA (1962) A revised medium for a rapid growth and bioassays with tobacco tissues cultures. Plant Physiol 15:473–479 Parfitt D, Ganeshan S (1989) Comparison of procedures for estimating viability of Prunus pollen. HortScience 24: 354–356 Parton E, Deroose R, De Proft MP (1998) Cryostorage of Aechmea fasciata pollen. Cryo-Letters 19:355–360 Parton E, Vervaeke I, Delen R, Vandenbussche B, Deroose R, De Proft MP (2002) Viability and storage of bromeliad pollen. Euphytica 125:155–161. doi:10.1023/A:1015884019619 Pixton SW (1966) Moisture content-its significance and measurement in stored products. J Stored Prod Res 3:35–47. doi:10.1016/0022-474X(67)90085-9 Rasband WS, Image J (1997–2012) US National Institutes of Health, Bethesda, Maryland, USA. http://imagej.nih.gov/ij/ Raynal A, Raynal J (1971) Une technique de préparation des grains de pollen fragiles. Adansonia 1:77–79 Reynolds ES (1963) The use of lead citrate at high pH as an electron-opaque stain in electron microscopy. J Cell Biol 17:208–212 Sacks EJ, St Clair DA (1996) Cryogenic storage of tomato pollen: effect on fecundity. HortScience 31:447–448 SAS Institute (2004) Inc. Sas/Stat useŕs guide: statistics, Version 9.1.3 ed. SAS Institute, Cary, NC, USA Scorza R, Sherman WB (1995) Peaches. In: Janik J, Morre JN (eds) Fruit breeding. John & Sons, New York, pp 325–440 Soares TL, Silva SO, Costa MAPC, Santos-Serejo JA, Souza AS, Lino LSM, Souza EH, Jesus ON (2008) In vitro germination and viability of pollen grains of banana diploids. Crop Breed Appl Biotechnol 8:111–118 Souza EH, Rosa SS, Souza FVD, Melo VC (2009a) Indução floral em Neoregelia carolinae (Beer) L. B. Sm e Aechmea fasciata (Lindley) Baker (Bromeliaceae). Magistra 21: 305–310 Souza FVD, Cabral JRS, Souza EH, Ferreira FR, Santos OSN, Silva MJ (2009b) Evaluation of F1 hybrids between 123 Ananas comosus var. ananassoides and Ananas comosus var. erectifolius. Acta Hortic 822:79–84 Souza FVD, Souza AS, Santos-Serejo JA, Souza EH, Junghas TG, Silva MJ (2009c) Micropropagação do abacaxizeiro e outras bromeliáceas. In: Junghas TG, Souza AS (eds) Aspectos práticos da micropropagação de plantas, 1st edn. Embrapa Mandioca e Fruticultura Tropical, Cruz das Almas, pp 177–205 Stanley RG, Linskens HF (1974) Pollen: biology biochemistry and management. Springer-Verlag, Berlin Tai PYP, Miller JD (2002) In vivo viability assay of sugarcane pollen stored at ultra low temperature following preservation treatments. Sugar Cane Int., pp 5–9 Taylor LP, Hepler PK (1997) Pollen germination and tube growth. Annu Rev Plant Physiol Plant Mol Biol 48: 461–491 Towill LE (1985) Low temperature and freeze-vacuum-drying preservation of pollen. In: Kartha KK (ed) Cryopreservation of plant cells and organs. CRC Press, Boca Raton, pp 171–198 Vendrame WA, Carvalho VS, Dias JMM, Maguire I (2008) Pollination of Dendrobium hybrids using cryopreserved pollen. HortScience 43:264–267 Withers LA (1991) Biotechnology and plant genetic resources conservation. In: Paroda RS, Arora RK (eds) Plant genetic resources-conservation and management, concepts and approaches. New Delhi, India Xu J, Liu Q, Jia M, Liu Y, Li B, Shi Y (2014) Generation of reactive oxygen species during cryopreservation may improve Lilium 9 siberia pollen viability. In Vitro Cell Dev Biol Plant 50:369–375 Yates IE, Sparks D (1989) Hydration and temperature influence in vitro germination of pecan pollen. J Am Soc Hortic Sci 114:599–605 Zhang F, Ge Y, Wang W, Yu X, Shen X, Liu J, Liu X, Tian D, Shen F, Yu Y (2012) Molecular characterization of cultivated bromeliad accessions with Inter-Simple Sequence Repeat (ISSR) markers. Int J Mol Sci 13:6040–6052. doi:10.3390/ijms13056040