Materials and Methods s19
In vitro pollen germination in avocado (Persea americana Mill.): optimization of the method and effect of temperature
M.L. Alcaraz, M. Montserrat, J.I. Hormaza*
Instituto de Hortofruticultura Subtropical y Mediterránea “La Mayora”, (IHSM-UMA-CSIC), E-29750 - Algarrobo-Costa, Málaga, Spain
ABSTRACT
An improved in vitro pollen germination method was developed for avocado (Persea americana Mill.). The effect of different concentrations of sucrose, polyethylene glycol (PEG), Mg and Ca on pollen germination was evaluated in order to determine the optimal pollen germination medium, i.e. that maximizing the percentage of pollen germination and minimizing the percentage of bursted pollen grains. Once the germination medium was optimized we used it to study the effect of temperature on in vitro pollen germination and tube growth in different cultivars from the three botanical varieties of avocado, that differ in their adaptation to environmental conditions. Significant differences in percentage of pollen germination and in pollen tube growth were observed among cultivars. These results could have implications not only for optimizing pollen management in avocado but also to select the best pollinizers for a particular cultivar.
Keywords: Lauraceae, pollen viability, polyethylene glycol, sucrose
Abbreviations: PEG, polyethylene glycol
* Corresponding author. Tel.: +34 952548990; fax: +34 952552677
Email address:
1. Introduction
Avocado (Persea americana Mill., Lauraceae) is an evergreen subtropical fruit tree native to Central America and Mexico, where it was domesticated in ancient times (Galindo-Tovar et al., 2008; Chen et al., 2009), that is currently cultivated in tropical and subtropical regions worldwide. World production of avocados in 2009 was estimated at c.a. 3.5 million tons, with Mexico as the main producer with more than 30% of the world production (FAOSTAT, 2011). Avocado flowers exhibit a unique behaviour described as protogynous dichogamy with synchronous daily complementarity (Davenport, 1986). Each avocado flower opens twice, each time for several hours: first, as a functionally female flower; then the flower closes and reopens the following day as a male flower (Davenport, 1986). Based on their flowering behaviour, avocado cultivars are classified in two groups (A or B) (Nirody, 1922). Although the cycle can vary depending on temperature and humidity, in general, flowers of the type A cultivars open in the morning in the female stage, close at midday, and reopen in the afternoon of the following day in the male stage. In type B cultivars, the flowers open in the afternoon in the female stage, close in the evening, and reopen the following morning in the male stage (Stout, 1923). This system facilitates pollen transfer between A and B cultivars.
The currently available protocols to evaluate avocado pollen viability need to be optimized in order to perform basic studies of avocado sexual reproduction such as pollen function, effect of environmental factors on pollen performance, pollen storage and pollination. Pollen viability can be evaluated by several experimental procedures. One approach is to evaluate viability before germination mainly using the fluorochromatic reaction based in fluorescein diacetate (FDA) (Heslop-Harrison and Heslop-Harrison, 1970). However, another method that also takes into account pollen performance is the evaluation of pollen germination in vitro (Shivanna and Johri, 1985; Shivanna et al., 1991). Different media for in vitro pollen germination have been reported in several species (Taylor and Hepler, 1997), mainly using the basic medium developed by Brewbacker and Kwack (1963). Several authors (Sahar and Spiegel-Roy, 1984; Loupassaki et al., 1997) reported a method for in vitro pollen germination in avocado, which yielded germination percentages ranging from 14 to 44%. However, this method produced a high percentage of bursted pollen grains.
In this work we first aimed at developing an optimized method for in vitro pollen germination in avocado that would maximize pollen grain germination and minimize pollen grain bursting. Second, we used the optimized medium to study differences on pollen germination and tube growth at different temperatures among cultivars pertaining to the three botanical varieties, or “races”, of avocado that have been traditionally described, i.e. ‘Mexican’, ‘Guatemalan’ and ‘West Indian’. The Mexican race is the most tolerant, and the West Indian race is the most sensitive, to cold temperatures, whereas the Guatemalan race is intermediate between the two. Knowledge of temperature effects on pollen germination and pollen tube growth in avocado cultivars will increase the understanding of the effect of temperature on fertilization and fruit set.
2. Materials and methods
2.1. Optimization of the pollen germination medium
To optimize the pollen germination medium, pollen from ‘Hass’, the commercially most important avocado cultivar worldwide, was used. Flowers in the male stage were collected from the field immediately after anther dehiscence and maintained at 100% RH during 2 h (Loupassaki et al., 1997).
We used as basic medium that reported by Sahar and Spiegel (1984). The experimental design aimed at evaluating the effect of sucrose, polyethylene glycol (PEG), MgSO47H2O and Ca(NO3)4H2O on the percentage of pollen grain germination and bursting. Experiments were conducted in Petri dishes containing 100 mg L-1 KNO3 and 100 mg L-1 H3BO3 (Sahar and Spiegel, 1984). Initially, the effect of sucrose at six concentrations (0, 5, 10, 15, 20 and 30%) and PEG 8000 at three concentrations (17, 20, and 23%) on pollen grain germination and bursting was evaluated. In a second step, a fine adjustment of PEG 8000 concentrations (10, 15, 17, 20, 23, 26 and 30%) was made using the best sucrose concentration found previously. Pollen germination was then evaluated at the optimal concentration of sucrose and PEG 8000 determined (see Results), but varying the concentration of MgSO47H2O (200, 240 and 300 mg L-1) and Ca(NO3)4H2O (240, 480, 720 and 960 mg L-1).
Pollen from 20 flowers was placed on 35 mm Petri dishes containing 2 mL of liquid germination medium. Pollen germination was evaluated after an incubation period of 20 h at room temperature. Pollen germination was quantified by direct observation using a Leica DML microscope in three Petri dishes that contained at least 200 pollen grains. Pollen was considered germinated when the tube length was at least twice the diameter of the pollen grain. Ungerminated and bursted pollen grains were also counted.
Data were first analyzed with a 2-way MANOVA with percentage of germinated pollen grains and percentage of bursted pollen grains as dependent variables, and sucrose and PEG concentrations as the explanatory variables. If the MANOVA were significant, each of the two dependent variables was subsequently analysed with a 2- way ANOVA, separately. Percentages of germinated and bursted avocado pollen grains were arcsine–transformed prior to analysis. Post-hoc analyses were done using the Tukey HSD test. Pollen germination at different concentrations of either MgSO47H2O or Ca(NO3)4H2O were analyzed with two separate one-way ANOVA. Post-hoc analyses were done using the Tukey HSD test. Statistical analyses were performed using SPSS 17.00 statistical software (SPSS Inc., Chicago, USA).
2.2. Effect of temperature and cultivar on pollen germination and pollen tube growth in vitro.
Six genotypes of different avocado botanical varieties maintained in the experimental station La Mayora (Malaga, Spain) were used for this study: ‘Hass’, ‘Fuerte’, ‘Anaheim’, ‘Topa Topa’, ‘Maoz’ and ‘Gvar 13’ (Table 1). ‘Fuerte’ (Ashworth and Clegg, 2003) and ‘Hass’ (Schnell et al., 2003) are Guatemalan x Mexican hybrids with different level of Mexican heredity; ‘Topa Topa’ (Ashworth and Clegg, 2003, Schnell et al., 2003) and ‘Gvar 13’ (Kadman and Ben-Ya’acov, 1980a) are Mexican; ‘Anaheim’ is Guatemalan (Schnell et al., 2003), and ‘Maoz’ is West Indian (Kadman and Ben-Ya’acov, 1980b). Under our environmental conditions these different genotypes show different flowering times; thus, the blooming season of ‘Topa Topa’ and ‘Fuerte’ is the earliest (flowering in ‘Topa Topa’ starts at the end of February, and that of ‘Fuerte’ at mid March), whereas the rest of genotypes flower in April (Alcaraz and Hormaza, 2009).
To evaluate the effect of temperature and cultivar on in vitro pollen germination and on pollen tube growth we used the optimized medium that was determined in section 2.1. The optimized medium consisted of 23% PEG 8000, 10% sucrose, 100 mg L-1 KNO3, 100 mg L-1 H3BO3, 300 mg L-1 MgSO47H2O and 480 mg L-1 Ca(NO3)4H2O (see Results). Pollen from flowers of the different genotypes was collected at the time of anther dehiscence, between 12:00 and 13:00 h in type B cultivars and between 16:00 and 17:30 in type A cultivars. For in vitro pollen germination, the anthers of 20 flowers were maintained at high relative humidity for 2 h. Then, pollen grains were placed on 35 mm Petri dishes containing 2 mL of the optimized medium. Last, they were incubated under different temperature regimes. Pollen germination was assessed 24 h later on a minimum of 200 randomly chosen pollen grains per Petri dish. A minimum of 20 Petri dishes were examined per cultivar and temperature; each Petri dish was considered a replicate. Pollen germination was quantified as described above. Percentages of pollen germination were evaluated at three temperatures that cover the range of temperatures found in the field during the blooming season in the avocado growing area in Southern Spain: 20ºC, 25ºC and 30ºC.
The effect of temperature on pollen tube growth was also evaluated. Pollen tube length was measured using an ocular micrometer attached to an optical microscope after an incubation of 24 h. A minimum of 20 pollen tubes were measured per Petri dish and the averages were calculated.
Percentage of pollen germination and average pollen tube length were analyzed with two separate two-way ANOVA, with temperature and genotype as explanatory variables. Tukey HSD test was used for means separation in cases of significant differences. Statistical analyses were performed using SPSS 17.00 statistical software (SPSS Inc., Chicago, USA).
3. Results
3.1. Optimization of the in vitro pollen germination medium
To optimize the pollen germination medium, different concentrations of PEG and sucrose were added to the basic medium. Pollen germination was negligible when sucrose concentration was below 5% and above 30% (data not shown).
The multivariate test was significant for both the main factors (Sucrose: Wilks = 0.34, F4,78 = 13.81, P < 0.001; PEG: Wilks = 0.32, F4,78 = 15.25, P < 0.001) and the interaction (Sucrose * PEG: Wilks = 0.23, F8,78 = 10.45, P < 0.001). Subsequent univariate analyses revealed similar statistical significance for each of the two dependent variables (% Pollen grains germinated: Sucrose – F2,40 = 25.30, P < 0.001; PEG: – F2,40 = 5.38, P 0.01; Sucrose * PEG: F4,40 = 13.78, P < 0.001. % Pollen grains bursted: Sucrose – F2,40 = 19.30, P < 0.001; PEG: – F2,40 = 28.35, P 0.001; Sucrose * PEG: F4,40 = 9.75, P < 0.001). Significance of the interactions indicated that variation of the percentage of germinated and bursted pollen grains depending on sucrose concentration was not the same at each of the PEG concentrations (Fig 1). Post-hoc analyses revealed that the optimal medium, i.e. that with the maximum percentage of pollen grain germination and the minimum percentage of pollen grains bursted was obtained when sucrose concentration was 10 % and PEG concentration was 23 % (Fig 1).
The multivariate test for Mg and Ca concentrations in the germination medium was significant (MgSO47H2O: Wilks = 0.64, F4,200 = 12.47, P < 0.001; Ca(NO3)4H2O: Wilks = 0.62, F6,200 = 8.98, P < 0.001; interaction MgSO47H2O * Ca(NO3)4H2O: Wilks = 0.48, F12,200 = 7.44, P < 0.001). Subsequent univariate analyses revealed similar statistical significance for each of the two dependent variables (% germinated pollen grains: MgSO47H2O – F2,101 = 1.08, P = 0.345; Ca(NO3)4H2O: – F3,101 = 6.37, P < 0.001; MgSO47H2O * Ca(NO3)4H2O: F6,101 = 2.598, P = 0.023; % bursted pollen grains: MgSO47H2O – F2,101 = 25.00, P < 0.001; Ca(NO3)4H2O: – F3,101 = 11.06, P < 0.001; MgSO47H2O * Ca(NO3)4H2O: F6,101 = 13.12, P < 0.001). Results revealed that, at the optimal medium, the best concentrations of calcium and magnesium for pollen germination were obtained with 300 mg L-1 of MgSO47H2O and 480 mg L-1 of Ca(NO3)4H2O (Fig. 2).
From these results we concluded that the optimal medium for avocado pollen germination consisted on 23% PEG, 10% sucrose, and the following mixture of mineral salts: 100 mg L-1 KNO3, 100 mg L-1 H3BO3, 300 mg L-1 MgSO47H2O and 480 mg L-1 Ca(NO3)4H2O.
3.2. Effect of temperature and cultivar on pollen germination and pollen tube growth in vitro
After optimizing the germination medium, the effect of three constant temperatures (20, 25 and 30ºC) on pollen germination was evaluated on six avocado genotypes of different botanical varieties.
The analysis of variance revealed significant differences among genotypes (F5,163 = 23.21, P < 0.001), and temperatures (F2,163 = 3.15, P = 0.046), but not in their interaction (F10,163 = 0.387, P = 0.95). The maximum germination percentage was obtained at 30ºC for all the genotypes except ‘Fuerte’ (Fig. 3). At 20ºC the germination percentages ranged from 8.78% in ‘Fuerte’ to 31.2% in ‘Hass’. At 25ºC germination varied from 8.32% in ‘Fuerte’ to 31.58 % in ‘Hass’. At 30ºC the percentage ranged from 7.09 % in ‘Fuerte’ to 34.93% in ‘Gvar 13’. At 20 and 25ºC it was possible to establish two different groups based on pollen performance. Thus, the first group included the genotypes ‘Topa Topa’, ‘Maoz’ and ‘Fuerte’ with the lowest germination and the other group was formed by ‘Gvar 13’, ‘Hass’ and ‘Anaheim’. However, at 30ºC, no clear groups can be defined although the percentage of pollen germination in ‘Fuerte’ was significantly lower than that of the rest of the genotypes (Fig. 3). In fact, all the cultivars except ‘Fuerte’ show the highest germination percentage at 30ºC although those differences were only significant for ‘Maoz’, a West Indian cultivar (Fig. 3).
Pollen tube length was also estimated for the different genotypes at three constant temperatures. The genotypes showed also differences in pollen tube length (Fig. 4). Significant differences were found among genotypes (F5,3937 = 20.19, P < 0.0001), temperatures (F2,3937 = 6.61, P < 0.001) and their interaction (F10,3937 = 4.48, P < 0.001). At 20ºC the length after 24 h ranged from 2.80 mm in ‘Maoz’ to 6.45 mm in ‘Gvar 13’ and significant differences were observed among genotypes (P 0.001). Under this condition, three groups were established. The first group was formed by ‘Maoz’, the second by ‘Hass’, and the rest of the genotypes analyzed were included in the third group. At 25ºC the pollen tube length after 24 h ranged from 2.55 mm in ‘Maoz’ [that shows a significant (P 0.01) difference to the rest of the genotypes] to 5.67 mm in ‘Gvar 13’. At 30ºC the shortest pollen tube length after 24 h was 3.78 mm (‘Gvar 13’) and the highest 6.35 mm for ‘Topa Topa’. At 30ºC significant differences among genotypes on pollen tube length after 24 h were detected (P 0.001) and, based on them, two groups were established: one group with ‘Topa Topa’ only, and a second group with all the rest of genotypes. At this temperature, pollen tube length in ‘Maoz’ was significantly higher (P 0.001) than that observed at lower temperatures.