Environmental influences on the frequency and viability of meiotic and apomeiotic cells of a diploid mutant of alfalfa.
Apomixis as a whole has not been detected in the genus Medicago, but components of apomixis are present. The formation of unreduced eggs through apomeiosis in a diploid plant of M. sativa subsp. falcata (L.) Arcang., named PG-F9 (Tavoletti, 1994) is an extremely interesting feature of apomixis. Similarly, the occurrence of parthenogenesis in tetraploid alfalfa (Bingham, 1971) is another feature which has been widely exploited to reduce cultivated tetraploid alfalfa to the diploid level via haploidy (Bingham and McCoy, 1979). Our goal is to eventually combine all the components of functional apomixis in alfalfa.
In sexuals and facultative apomicts, the course of meiosis is under polygenic control (Kaul and Murthy, 1985) and also depends on environmental conditions (Hermsen, 1984; Asker and Jerling, 1992). Photoperiod had a significant effect on the relative frequency of meiotic and apomeiotic embryo sacs in ovules of Dichanthium species (reviewed by Bashaw, 1980). In diploid Solanum clones, temperature levels also influenced the formation of unreduced pollen (Veilleux and Lauer, 1981; McHale, 1983). In addition, the frequency of 2n gametes has been found to change not only in the same plants at different ages (Ramanna, 1974), but also between ramets of a clone (Jacobsen, 1976), and even among flowers of the same plant (Veilleux et al., 1982). Stein (1970) listed several species in which the frequency of 2n gametes varied with the cultivation conditions. These findings emphasize the importance of knowing the potential sources of variation in 2n gamete formation, as well as the characteristic behavior of specific germplasm in succeeding generations and under diverse environments.
The present paper reports the influence of the environment on the frequency and viability of normally reduced (n), restitutional, and apomeiotic (2n) cells in the mutant PG-F9 grown in the field versus the growth chamber and provides cytological data on the arrangements exhibited during post-meiotic stages of megasporogenesis of PG-F9 and the relative fertility of PG-F9 in terms of seed sets and seeds per pod in controlled matings.
MATERIALS AND METHODS
Plant Materials and Growth Conditions
The mutant PG-F9 was selected by screening for 2n egg producers in a diploid population of M. sativa subsp. falcata (L.) Arcang. (2n=2x=16) based on the seed set in 2x-4x crosses (Veronesi et al., 1988) and later characterized for the occurrence of restitutional and apomeiotic mechanisms of 2n egg formation (Tavoletti, 1994). Three ramets of PG-F9 were reared at a field site in Perugia, Italy, (from November 1992-July 1993) and in a growth chamber (from November 1993-April 1994) together with three 2n pollen producers (plants 2-P; 5-P and 9-P) obtained after two cycles of phenotypic recurrent selection for 2n pollen formation (Tavoletti et al., 1991b), three normal diploids of M. sativa subsp. coerulea (Less.) Schm. (plants 2-DI; 6-DI and 7-DI), and three normal tetraploids of M. sativa subsp. sativa L. (plants 10-TE; 16-TE and CSE-1). Field and growth chamber environmental conditions (minimum and maximum temperatures, photoperiod, and light intensity) are summarized in Fig. 1. The maximum day and minimum night temperature means in the two situations were 15.6 [degrees] C and 8.1 [degrees] C in the field and 22.4 [degrees] C and 13.2 [degrees] C in the growth chamber.
[Figure 1 ILLUSTRATION OMITTED]
Cytological Analyses of Megasporogenesis
Flower buds from the three ramets of PG-F9 were collected twice, at the beginning and at about the middle of the flowering period in both environments [10 and 25 June 1993 in the field, 5 and 20 Feb. 1994 in the growth chamber (Fig. 1)]. Different sized flower buds, spanning the stages from archesporial cells to mature embryo sacs, were fixed in formalin-acetic acid-alcohol (FAA). Cytological investigations were conducted by means of stain-clearing (Stelly et al., 1984) and sectioning techniques (Johansen, 1940) on 46 flower buds collected in the field and 44 flower buds collected in the growth chamber for a total of 312 and 316 ovules analyzed, respectively.
Environmental effects on meiotic processes of mutant PG-F9 were determined by analyzing the frequencies of normal meiotic and apomeiotic arrangements observed in the field and growth chamber by means of a 2 X 2 contingency test (Snedecor and Cochran, 1967). Stability of expression of apomeiosis was established by comparing the frequency of apomeiosis in field grown plants of PG-F9 in 1992 and 1993.
The viability of 2n eggs and the overall fertility of mutant PG-F9 in the field and growth chamber were determined by generating self, intraploid (diploid pollen donor), and interploid (tetraploid pollen donor) cross pollinations in both environments. The use of different pollen sources permitted a detailed analysis of the reproductive behavior of PG-179. The pod set and seed set values (number of pods and seeds produced per flower pollinated, respectively) and the seeds per pod were determined in both environments with 40 to 60 flowers of PG-F9 for each cross combination and selfing.
Pod set and seed set values from selfing and controlled crosses of PG-179 in the field and growth chamber were subjected to analysis of variance (Statistical Analysis System, SAS Institute Inc., Cary, NC) to estimate genotype X environment interactions, to estimate differences among types of pollen donor and among pollen donors within a type, and to evaluate the influence of the two different environments on egg cell viability. The model included as source of variation environments, ploidy level, environment X ploidy level, pollen donors-ploidy level, and environment X pollen donors-ploidy level. All factors were considered fixed. Statistical differences among means were tested by LSD at the 5% level.
RESULTS AND DISCUSSION
The analysis of PG-F9 ovules revealed normal meiosis resulting in a tetrad of megaspores, basically either linear or T-shaped (Pfeiffer and Bingham, 1983) and showed that embryo sac development is monosporic of the Polygonum type (Reeves, 1930). The production of 2n megaspores in PG-F9 resulted from various abnormalities of meiosis. The lack of the second meiotic division, which produced megaspore dyads with two sets of chromosomes, reported by Tavoletti (1994) was confirmed. Dyads and triads, including binucleated functional megaspores, were also identified and resulted from the absence of cytokinesis after telophase II in both chalazal and micropylar cells (Fig. 2a,b,c) or just in the chalazal cell (Fig. 2d,e).
[Figure 2 ILLUSTRATION OMITTED]
The developmental timing and the positioning and appearance (nucleus size, cytoplasmic vacuolization) of apomeiotic cells are cytological features that allowed the production of these cells to be ascribed to a diplosporic pathway, as proposed by Bimal et al. (1995). On the basis of integument growth, all PG-F9 ovules should have displayed tetrads with developing functional megaspores; however, some ovules exhibited enlarged and vacuolized MMCs (Fig. 2f,g) which apparently did not undergo meiosis. It is unlikely that such cells were uninucleated sexual embryo sacs, since degenerated megaspores were never observed and the nucleus and nucleolus of those cells were markedly larger than those of normally reduced megaspores. The finding that the position within the ovule and the appearance and cell volume of the sexual embryo sac always differed (Fig. 2g,h) supports the occurrence of diplospory.
Aposporic initials developing from the nucellar tissue and in competition with sexual cells were never detected. In the aposporic pathway, the unreduced embryo sacs usually develop from somatic cells lying in the center of the nucellus (Nogler, 1984; Asker and Jerling, 1992). Some PG-F9 ovules displayed a single anomalous cell with enlarged nucleus and nucleolus located deep in the chalaza (Fig. 2c and 3a). However, these cells were never vacuolized and usually were compressed by the adjacent generative cells, which demonstrated an unfavorable growth condition. The finding of cells with similar features and growth pattern derived from mitotic division with normal cytokinesis in the chalazal tissue (Fig. 3b) is evidence that these anomalous cells could not have been aposporic initials. Moreover, the lack of aposporous embryo sacs in competition with MMCs, megaspores, or sexual embryo sacs also suggests the absence of apospory.
[Figure 3 ILLUSTRATION OMITTED]
On the whole, the cytological analysis of PG-F9 ovules revealed that the most frequent anomalies of megasporogenesis were second division restitution (average 49.05%) and diplospory (average 21.31%) which resulted in 2n functional megaspores. The presence of two distinct patterns of megasporogenesis was concordant with the molecular differences found between PG-F9 and wild-type plants in terms of poly(A) + RNA accumulation in ovules at the MMC stage of development (Birnal et al., 1995). The question of the occurrence of meiotic or mitotic diplospory in PG-F9 remains. In meiotic diplospory, the MMC differentiates from the nucellus and begins meiosis, which is, however, inhibited at a particular stage by unknown mechanisms and the nucleus is restored to a form that enables mitosis to occur (Nogler, 1984). In mitotic diplospory, the MMC appears to be inhibited and does not start meiosis; if entry to meiosis occurs it is inhibited at a very early stage (Koltunow, 1993). These mechanisms would be evident only at the molecular level and not by cytological examination. Apomeiotic 2n eggs produced through diplospory are interesting because they should retain and transmit the whole (mitotic diplospory) or a large part (meiotic diplospory) of the maternal genotype, while 2n eggs formed by SDR mechanisms generally conserve and transfer a small part of maternal gene combinations.
The relative frequencies of different arrangements exhibited during post-meiotic stages of megasporogenesis in PG-F9 are reported in Table 1. The same postmeiotic arrangements occurred in both environments, with the exception of T-shape tetrads which were present only in the field grown materials. The overall frequency of normal reduced chalazal megaspores in the field (25.32%) was about three times higher than in the growth chamber (9.80%), as was the frequency of normal tetrads (23.08% in the field vs. 7.91% in the growth chamber) (Table 1, columns a, b, and f). The frequency of restitutional cells (dyads and triads with 2n chalazal megaspores) was virtually the same in the two environments (49.68% in the field vs. 48.42% in the growth chamber) (Table 1, columns c, d, and e). Omission of normal meiosis resulting in diplosporic cells was greater in the growth chamber than in the field (16.35% vs. 26.27%). The occurrence of normal tetrads and the degree of apomeiosis were significantly different ([chi square] = 38.464, P [is less than or equal to] 0.001, d.f. = 1) in the field and growth chamber. The frequencies of normal meiotic and diplosporic cells in PG-F9 raised in the field in 1993 were similar to those observed by Tavoletti (1994) in the field in 1992.
[TABULAR DATA 1 NOT REPRODUCIBLE IN ASCII]
The unstable expression of different post-meiotic arrangements suggests that alfalfa meiotic mutants typically produce variable products. Thus, stable versus variable expression of restitutional meiosis and diplosporic apomeiosis may reflect a female gametophyte development influenced by various modifying genes whose expression is sensitive to the environment. However, investigations carried out on plants of the same alfalfa clone grown in the field site under natural environmental conditions during two different years provide evidence that a stable expression of meiotic abnormalities is attainable. The differences in terms of temperature, photoperiod, and light intensity occurred in the field and established in the growth chamber could underlie the increased diplosporic events associated with the decreased occurrence of normal meiosis. If so, the implications are of extreme importance for the use of 2n egg mutants in cultivated alfalfa breeding, since instability of meiotic products in a range of environments would delay progress toward routine use of this valuable trait in wild diploid germplasm introgression and new cultivar development.
Controlled Selfing and Crosses
PG-F9 produced functional 2n eggs in high frequencies in both environments. The self-fertility rate of PG-F9 was very low. The pod set and seed set values, and the seeds per pod frequency from self- and cross-pollination of PG-F9 in the field and growth chamber are reported in Table 2 and Fig. 4.
[TABULAR DATA 2 NOT REPRODUCIBLE IN ASCII]
[Figure 4 ILLUSTRATION OMITTED]
Statistical analysis of pod and seed sets registered in the field and in the growth chamber revealed no interaction between the environment and type of pollen donor. The average values of PG-F9 fertility for selfing and cross combinations were generally lower in the growth chamber than in the field with both normal and 2n pollen producer diploid pollen donors and normal tetraploid pollen donors. The differences between set means of 2x-2-x and 2.x-4x crosses were statistically significant for both variables but were more pronounced for seed set (1.79 vs. 0.93) than for pod set (0.76 vs. 0.62).
The differences in pod set values between the two environments were small but significant (on average, 0.68 in the field and 0.62 in the growth chamber). Under field conditions, the pod set from selfing of PG-F9 (average 0.12) was greater than in the growth chamber (average 0.07). Of the crosses between PG-F9 and the diploid pollen donors, plant 5-P, which had the largest average rate of pod set (0.81), did not differ significantly from the normal diploid plants 6-DI and 2-DI (Table 2). The differences among tetraploid pollen donors were also significant (Table 2). The differences in seed set values were highly significant between field and growth chamber environments (on average, 1.47 in the field and 1.25 in the growth chamber). Selfed seed set of PG-F9 under field conditions was higher (average 0.16) than in the growth chamber (average 0.07). The average seed set values of PG-F9 when crossed with diploid and tetraploid pollen donors was also higher under field conditions, with the exception of plant 5-P (Table 2). Highly significant differences were recorded among types of pollen donor and for single pollen donors within a type. Within diploid pollen donors, the highest rate of seed set (2.07) was documented in plant 5-P, which differed significantly from the other pollen donors (Table 2). The highest seed set rate within the tetraploid pollen donors (0.99) was recorded in plant 16-TE (Table 2).
Although the overall formation of n megaspores was much lower than 2n megaspores (Table 1), the greatest values of seeds per pod were recorded in intraploid crosses, on average, 2.18 to 2.46 in 2-x-2x crosses vs. 1.40 to 1.57 in Zx-4x crosses (Fig. 4).
The differences between fertility data in 2x-2x crosses registered in the two environments agreed with the higher production of n megaspores and with the lower number of aborted ovules observed in the field (Tables 1 and 2). The discrepancy between cytological analyses (rate of functional 2n megaspores) and fertility (seed set and seeds per pod values) in 2x-4x crosses, particularly evident in the growth chamber (Tables 1 and 2; Fig. 4), may indicate that some 2n megaspores were unable to develop into functional embryo sacs (Calderini and Mariani, 1995), and it may also be accentuated by the presence of residual n eggs that could have caused triploid embryo abortion. Moreover, although the differences among overall set means were significant, they were restricted for the pod set but pronounced for the seed set. This fact could confirm the presence of an effective triploid block (Veronesi et al., 1986) which eliminated most of the triploid embryos in intraploid (PG-179-20) and interploid (PG-F9-TE) crosses.
In conclusion, as many as one of five megaspore mother cells in PG-F9 failed normal meiosis and gave rise directly to apomeiotic embryo sacs containing unreduced nuclei. Consistent expression of diplospory in a specific environment provides a unique opportunity for verifying the occurrence of parthenogenesis in PG-F9. Studies conducted in diplosporous species have reported that apomeiosis and parthenogenesis are associated processes (i.e., most of unreduced eggs have parthenogenetic capability) and that the degree of apomixis depends on competition between reduced and unreduced megaspores (Kojima and Nagato, 1992a,b). In consequence, mutant PG-F9 was also characterized by RAPD (Barcaccia et al., 1994) and RFLP markers (Tavoletti et al., 1996) and the parthenogenetic capability of its diplosporic eggs will be assessed by molecular progeny tests. As apomictic seed formation has never been observed in Medicago and because of its potential in alfalfa breeding, the attempt to search for apomictic reproduction in PG-F9 seems to be a worthwhile aim.
Thanks are due to Prof. E.T. Bingham, Agronomy Department, University of Wisconsin, Madison, WI, and three anonymous reviewers for critically reviewing the manuscript.
Asker, S., and L. Jerling. 1992. Apomixis in plants. CRC Press, London.
Barcaccia, G., S. Tavoletti, M. Pezzotti, M. Falcinelli, and F. Veronesi. 1994. Fingerprinting of alfalfa meiotic mutants using RAPD markers. Euphytica 80:19-25.
Barcaccia, G., N. Tosti, E. Falistocco, and F. Veronesi. 1995. Cytological, morphological and molecular analyses of controlled progenies from meiotic mutants of alfalfa producing unreduced gametes. Theor. Appl. Genet. 91:1008-1015.
Bashaw, E.C. 1980. Apomixis and its application in crop improvement. p. 45-63. In W.R. Fehr and H. H. Hadley (ed.) Hybridization of crop plants. ASA, Madison, WI.
Bimal, R., J.H.M. Schel, S. Tavoletti, and M.T.M. Willemse. 1995. Localization of poly (A) + -containing RNA during female gametophyte development in Medicago sativa and the diploid mutant M. sativa subsp. falcata using digoxigenin labelled oligo-dT probes. Sex. Plant Reprod. 8:18-26.
Bingham, E.T. 1971. Isolation of haploids of tetraploid alfalfa. Crop Sci. 11:433-435.
Bingham, E.T., and T.J. McCoy. 1979. Cultivated alfalfa at the diploid level: Origin, reproductive stability, and yield of seed and forage. Crop Sci. 19:97-100.
Bingham, E.T. 1980. Maximizing heterozygosity in autopolyploids. p. 471-490. In W.H. Lewis (ed.) Polyploidy, biological relevance. Plenum Press, New York.
Calderini, O., and A. Mariani. 1995. Megagametophyte organization in diploid alfalfa meiotic mutants producing 4n pollen and 2n eggs. Theor. Appl. Genet. 90:135-141.
Harlan, J.R., and J.M.J. deWet. 1975. On 6 Winge and a prayer: The origins of polyploids. Bot. Rev. 41:361-390.
Hermsen, J.G. 1984. Mechanisms and genetic implications of 2n gamete formation. Iowa State J. Res. 58(4):421-434.
Jacobsen, E. 1976. Cytological studies on diplandroid production in a dihaploid potato clone and its correlation with seed set in 4x-2x crosses. Z. Pflanzenzuchtg. 77:10-15.
Johansen, D.A. 1940. Plant microtechnique. McGraw-Hill, Co., London.
Jongedijk, E. 1991. Desynapsis and FDR 2n megaspore formation in diploid potato; potentials and limitations for breeding and for the induction of diplosporic apomixis. Ph.D. Thesis, University of Wageningen, Wageningen, the Netherlands.
Kaul, M.L.H., and T.G.K. Murthy. 1985. Mutant genes affecting higher plant meiosis. Theor. Appl. Genet. 70:449-466.
Kojima, A., and Y. Nagato. 1992a. Diplosporous embryo sac formation and the degree of diplospory in Allium tuberosum. Sex. Plant Reprod. 5:72-78.
Kojima, A., and Y. Nagato. 1992b. Pseudogamous embryogenesis and the degree of parthenogenesis in Allium tuberosum. Sex. Plant Reprod. 5:79-85.
Koltunow, A.M. 1993. Apomixis: Embryo sacs and embryos formed without meiosis or fertilization in ovules. Plant Cell 5:1425-1437.
McCoy, T.J., and D.E. Rowe. 1986. Single cross alfalfa (Medicago sativa L.) hybrids produced via 2n gametes and somatic chromosome doubling: Experimental and theoretical comparisons. Theor. Appl. Genet. 72:80-83.
McHale, N.A. 1983. Environmental induction of high frequency 2n pollen formation in diploid Solanum. Can. J. Genet. Cytol. 25: 609-615.
Nogler, G.A. 1984. Gametophytic apomixis. p. 475-518. In B.M. Johri (ed.) Embriology of angiosperms. Springer-Verlag, Berlin.
Pfeiffer, T.W., and E.T. Bingham. 1983. Abnormal meiosis in alfalfa, Medicago sativa: Cytology of 2n eggs and 4n pollen formation. Can. J. Genet. Cytol. 25:107-112.
Ramanna, M.S. 1974. The origin of unreduced microspores due to aberrant cytokinesis in the meiocyte of potato and its genetic significance. Euphytica 23:20-30.
Reeves, R.G. 1930. Development of the ovule and embryo sac of alfalfa. Am. J. Bot. 17:239-246.
Snedecor, G.W., and W.G. Cochran. 1967. Sampling from the binomial distribution. In Statistical methods, 6th ed., The Iowa State University Press, Ames.
Stanford, E.H., W.M. Clement, and E.T. Bingham. 1972. Cytology and evolution of the Medicago sativa-coerulea-falcata complex. p. 87-102. In C.H. Hanson (ed.) Alfalfa science and technology. ASA, Madison, WI.
Stein, M.. 1970. Polyploidie und Umwelt. In Die bedeutung der polyploidie fur die evolution und die pflanzenzuchtung. Tagungsberichte Deutschen Akad. Wiss 101:51-68.
Stelly, D.M., S.J. Peloquin, R.G. Palmer, and C.F. Crane. 1984. Mayer's hemalum-methyl salycilate: A stain clearing technique for observations within whole ovules. Stain Technol. 59:155-161.
Tavoletti, S., A. Mariani, and F. Veronesi. 1991a. Cytological analysis of macro- and microsporogenesis of a diploid alfalfa clone producing male and female 2n gametes. Crop Sci. 31:1258-1263.
Tavoletti, S., A. Mariani, and F. Veronesi. 1991b. Phenotypic recurrent selection for 2n pollen and 2n egg production in diploid alfalfa. Euphytica 57:97-102.
Tavoletti, S. 1994. Cytological mechanisms of 2n egg formation in a diploid genotype of Medicago saliva subsp. falcata. Euphytica 75:1-8.
Tavoletti, S., F. Veronesi, and T.C. Osborn. 1996. RFLP linkage map of an alfalfa diploid meiotic mutant based on a F1 population. J. Hered. 87:167-170.
Veilleux, R.E., and F.I. Lauer. 1981. Variation for 2n pollen production in clones of Solanum phureja Juz. and Buk. Theor. Appl. Gen. 59:95-100.
Veilleux, R.E., N.A. McHale, and F.I. Lauer. 1982. 2n gametes in diploid Solanum: Frequency and types of spindles anormalities. Can. J. Genet. Cytol. 24:301-314.
Veilleux, R.E. 1985. Diploid and polyploid gametes in crop plants: Mechanisms of formation and utilization in plant breeding. p. 253-288. In J. Janick (ed.) Plant Breed. Rev., Vol. 3, Avi Publish. Co., Wesport, CT.
Veronesi, F., A. Mariani, and E.T. Bingham. 1986. Unreduced gametes in diploid Medicago and their importance in alfalfa breeding. Theor. Appl. Genet. 72:37-41.
Veronesi, F., A. Mariani, and S. Tavoletti. 1988. Screening for 2n gamete producers in diploid species of the genus Medicago. Genet. Agr. 42:187-200.
Vorsa, N., and E.T. Bingham. 1979. Cytology of 2n pollen formation in diploid alfalfa, Medicago saliva. Can. J. Genet. Cytol. 21:525-530.
Abbreviations: MMC, megaspore mother cell; SDR, second division restitution; FDR, first division restitution; 2nP, 2n pollen mutant; RAPD, random amplified polymorphic DNA; RFLP, restriction fragment length polymorphism.
Gianni Barcaccia,(*) Stefano Tavoletti, Mario Falcinelli, and Fabio Veronesi
G. Barcaccia, M. Falcinelli and F. Veronesi, Istituto di Miglioramento Genetico Vegetale, Facolta di Agraria, Universita degli Studi di Perugia, Borgo XX Giugno, 06100 Perugia, Italy; S. Tavoletti, Dipartimento di Biotecnologie Agrarie ed Ambientali, FacoltA di Agraria, Universita degli Studi di Ancona, Via Brecce Bianche, 60100 Ancona, Italy. Research supported by the Ministry of University, Research, Science and Technology (funds 40%). Received 22 Feb. 1996. (*) Corresponding author (imgvsas@ ipguniv.unipg.it).
|Printer friendly Cite/link Email Feedback|
|Author:||Barcaccia, Gianni; Tavoletti, Stefano; Falcinelli, Mario; Veronesi, Fabio|
|Date:||Jan 1, 1997|
|Previous Article:||Disease reaction changes from tandem selection for multiple disease resistance in two maize synthetics.|
|Next Article:||RFLP detection of 2n pollen formation by first and second division restitution in perennial ryegrass.|