Plant Cell Reports

Plant Celt Reports (1993) 12:139-143

9 Springer-Verlag 1993

Production of gynogenic haploids of Hordeum vulgate L. A.M.

Castillo and L. Cistu6

Estacitn Experimental de Aula Dei, Ap. 202. Zaragoza, Spain Received May 18, 1992/Revised version received November 13, 1992 - Communicated by I.K. Vasil

AbstraeL Haploid plants were regenerated from cultured unfertilized ovaries of Hordeum vulgate L. (barley). Optimal response was obtained by the addition of 0.6 pM 4-chloro-2methylphenoxyacetic acid (MCPA), 2.8 pM indole-3-acetie acid (IAA) and 4.4 pM 6-benzyladenine (BA) in the N 6 medium. Further increase in the rate of callus formation and the number of green plants produced was possible with the addition of 90 g/l sucrose and 100 g/1 coconut water. The stage of development of the ovaries at the time of culture was critical; the largest number of plants being produced by ovaries from flowers at the trinucleate stage of pollen.

Key Words. Hordeum vulgare L., gynogenesis, haploids, ovary culture. Abbreviations. 6-benzyladenine (BA); 4-chloro-2methylphenoxyaceticacid (M CPA);2,4-diehlorophenoxyacetieacid (2,4-D); gibberellic acid (GA3); indole-3-acetie acid (IAA). Introduction Anther culture (Clapham 1973) and interspeciflc crossing with Itordeum bulbosum (Kasha and Kao 1970) are the two methods of choice for obtaining haploid plants of barley. However, both methods present some problems. Many of the plants recovered from anther culture are albinos (Foroughi-Wehr et al. 1982), although the latest improvements in nutrient media seem to have partly solved this problem (Hunter 1987, Olsen 1987, Kuhlmann and Foroughi-Wehr 1989). Crossing with H. bulbosum is limited as it depends on the degree of compatibility among genotypes (Bj0rnstand 1986). In addition, the yield of haploid plants by both methods is strongly genotype dependent (Foroughi-Wehr et al. 1982, Pickering 1984, Poweli 1988).

stages of development. Material and Methods Hordeum vulgate L ev. Berenice and the dihaploid line C001374 were used as source material. The dihaploid line, obtained through interspecifie crossing with H. bulbosum, was kindly supplied by Prof. N. O. Bosemark (Hilleshtg AB, Sweden). The plants were grown in a greenhouse under 16 h photoperiod at 20+4"C. Spikes were collected, unless stated otherwise, when they were still in the sheath, with half the length of the awn coming out of the sheath. They were surface disinfected with a 2.8% solution of sodium hylxxzhlorite for 5 min, and were rinsed several times with sterile distilled water. Ten flowers from one ear were plated on each Petri dish, using 200 ovaries for each treatment. Ovaries and anthers were plated together (Fig. la). N 6 medium (Chu et al. 1975), solidified with Difco Baeto Agar (6.5 g/l) and containing sucrose (80 g/l), was used as an induction medium. The effect of auxins (MCPA, IAA, 2,4-D), eytokinins (BA, kinetin), organic compounds, sucrose and the stage of ovary development were studied in five experiments: Experiment 1: MCPA alone or in combination with other growth regulators. The N6 medium was supplemented with MCPA (0, 0.6 or 2.5 pM) in combination with 5.7 pM IAA, 4.4 pM BA, and 4.6 ~M kinetin, or IAA plus BA, or IAA plus kinetin. The ovaries were maintained on the same medium until calli appeared. ]~xperhnent 2: Comparison between MCPA and 2~4-D. The N6 medium was supplemented with 2.5 pM MCPA and 2.3 pM 2,4-D, or 0.6 pM MCPA and 0.6 pM 2,4-D, in combination with 2.8 pM IAA and 4.4 pM BA. The ovaries were kept on these media for 21 days and then transferred onto fresh N 6 with 2.8 pM IAA and 4.4 pM BA. Four hundred ovaries were plated per genotype and treatment.

Following the first report of haploid plants from the culture of unfertilized ovaries of barley (San Noeum 1976, 1979), this method has proven successful in crops like rice (Zhou and Yang 1981), sunflower (Gelebart and San 1987), and sugarbeet (Hosemans and Bossoutrot 1983). Despite the low frequency of haploid production in barley, about one plantlet for one hundred ovaries plated, the system has the advantage that all regenerated plants are haploid as well as green.

Experiment 3: Carbohydrate source. Sucrose (90, 50 and 10 g/l), and melibiose (80 g/l) plus sucrose (10 g/l), were assayed. MCPA (0.6/~M), IAA (2.8 pM) and BA (4.4 pM) were used as growth regulators. After 21 days the ovaries were transferred to fresh medium without MCPA.

In this report we describe improved frequency of haploid production from cultured ovaries of barley by the use of various plant growth regulators and the culture of ovaries at specific

Experiment 4: Developmental stage of the ovary. The stage of pollen grain development was used as an indirect test to determine the age of the ovary. One anther was removed from

Correspondence to: L. Cistu6

140 the central part of the ear and was stained with aeetoearmine (1%). Three different developmental stages were defined: uni-, bi- and trinueleate (90% of the pollen grains had one, two or three nuclei, respectively). A fourth stage was used as the mature stage. This stage was obtained by removing the anthers from the flowers and maintaining the ears in modified Hoagiand solution (Jensen 1975) for two days. The culture media used were the same as in Experiment 3 but with 90 g/l sucrose. Experiment 5: Organic compounds. Casein hydrolysate (500 mg/l) plus malt extract (100 toga), coconut water (10%), potato extract (10% wA,), barley flower extract (10 ovaries/8 ml medium), gibberellic acid (2 mg/l) plus ascorbie acid (50 rag/l), and arginine (500 mga) plus proline (500 rag/l), were assayed. The flower extract was obtained as follows: mature flowers from different barley genotypes were homogenized with culture medium using Ommi-mixer 17106, six pulses at full speed. The homogenate was centrifuged in Sorvall RC5C for 5 min at 5000 rpm with SS34 rotor, q-he supernatant was sterilized by filtration and added to the medium. The calli obtained were subeultured on N6 medium with 2.8 /xM IAA, 4.4 t~M BA and 30 g]l sucrose, except in Experiment 3 where calli induced in media with 10 g/l sucrose, and 80 g/l melibiose plus 10 g/l sucrose, were subcultnred in medium with 10 g/l sucrose. The organic compounds added to the regeneration media were the same as in the induction medium in Experiment 5. N6 medium with 5.37/~M NAA was used as rooting medium if necessary. Chromosome counts were performed using tips of young roots that were treated with a saturated solution ofa-bromonaphthalene for 5 h at 22"C, then fixed in ethanol-giacial acetic acid (3:1) solution for 24 h. Hydrolysis was done in 1 N hydrochloric acid for 12 rain at 60"C, Root tips were stained in laetopropionie orcein (Dyer 1963). Analysis of variance was used to analyze the rate of calli induced and regenerated. Data were transformed with arc sine t/x; ten ovaries coming from the same ear were considered as an experimental unit. Results Experiments 1 and 2: Most growth regulator combinations induced formation of gynogenic calli within 30 to 90 days, but most frequently after about 45 days. Plantlets regenerated only in a few combinations (Fig. lb). The calli obtained after two months did not form plants. Some of the haploid plants originated directly from the ovary on induction medium (Fig. lc), without being plated on a regeneration medium. The largest number of calli were obtained at the highest concentration of MCPA (Table 1). Although media containing IAA did not yield any plants, the rate of callus formation was higher than the control. In the absence of IAA, BA produced plants at the highest concentrations of MCPA tested. Kinetin induced high frequencies of callus induction with the dihaploid line, but not with the cultivar Berenice. However, kinctin inhibited regeneration of plantlets. The combination of IAA and BA gave a high rate of gynogenic calli and the highest number of regenerated haploid green plantlets. MCPA produced a higher frequency of callus induction and a higher number of plants than 2,4-D (Table 2).

Experiment 3: High concentrations of sugars seemed to be more effective than low ones. The highest rate of callus induction was obtained using sucrose (90 g/l) and melibiose (80 g/l) plus sucrose (10 g/l), although plants from the two genotypes were produced only when sucrose alone was used (Table 3). As the concentration

Fig la. Ovaries and anthers two weeks after plating on culture medium. Fig lb. Regeneration of two haploid plantlets from a gynogenic callus (scale bar = 2 mm). Fig lc. Plantlet obtained directly from an ovary on the inducation medium (scale bar = 2 mm).

141 Table 1. Mean rate (%) of induction of callus (a), and number of regenerated green haploid plantlets (b) in the first experiment. Concentration of growth regulators in/~M.

Genotypes

Growth regulators MCPA

Treatment 0

DH C001374

IAA 5.7

BA 4.4

Kinetin 4.6

IAA 5.7 BA 4.4

IAA 5.7 Kinetin 4.6

a

b

a

b

a

b

a

b

a

b

a

b

0.0

0.0

0

0.5

0

0.9

0

0.5

0

0.5

1

3.4

0

0.6

0.0

0

0.6

0

1.9

2*

1.6

0

1.6

3

5.8

0

2.5

3.9

0

4.3

0

2.6

0

3.7

2.2

0

2.3

0

0.0

1.0

0

0.7

0

0.5

0

0.0

0

4.0

2

1.0

0

0.6

0.6

0

2.1

0

0.9

0

0.6

0

2.0

1

0.0

0

2.5

3.2

1

1.0

0

5.5

2

0.0

0

0.5

0

0.0

0

Berenice

0

*Albino haploid plants.

Table 2. Mean rate (%) of induction of callus (a), and number of regenerated green haploid plantlets (b) in the second experiment. Concentration in/~M.

Genotypes

Growth regulators MCPA 2.5

MCPA 0.6

2,4-D 2.3

IAA 2.8

BA 4.4

2,4-D 0.6 IAA 2.8

BA 4.4

a

b

a

b

a

b

a

b

DH C001374

3.2

2

4.0

2

2.5

4

3.0

0

Berenice

2.4

0

2.5

4

0.0

0

1.1

0

Table 3. Influence of carbon source on gynogenesis. The mean rate (%) of calli and plants are expressed per ovary plated.

Treatment

Genotype

Ovaries plated

Calli (%)

Plants (%)

DH

210

5.24

0.95

BE

200

5.00

0.50

DH

200

3.00

---

BE

200

3.50

2.00

DH

200

2.00

---

BE

200

1.50

---

Melibiose 80 g/l +

DH

200

7.50

0.50

Sucrose 10 g/1

BE

200

2.00

---

Sucrose 90 g/1

Sucrose 50 g/1

Sucrose 10 g/1

142 Table 4. Influence of stage of ovary development on gynogenesis. Mean rates (%) of calli, regenerated calli and plants are expressed per ovary plated.

Treatment

Ovaries plated

Calli (%)

Regenerated calli (%)

Plants (%)

Uninucleate

400

1.00

Binucleate

160

3.75

1.25

0.62

Trinucleate

180

6.11

2.78

1.66

Mature

200

3.00

1.00

0.50

. . . . . .

Table 5. Influence of organic compounds on gynogenesis. Mean rates (%) of call, regenerated calli and plants are expressed per ovary plated. Treatment

Ovaries plated

Calli (%)

Regenerated calli (%)

Plants (%)

Control

120

5.00

1.67

2.50

Casein hydrolysate + malt extract

140

7.14

0.71

---

Potato extract

120

2.50

0.83

---

Coconut milk

130

7.69

3.84

10.0

Ovary extract

130

0.77

GA 3 + ascorbic acid

110

2.73

0.91

---

Arginine + proline

140

2.86

1.43

2.14

of sucrose was reduced, the induction as well as regeneration rates decreased, Plants could not be obtained with 10 g/l sucrose. Experiment 4: Highest callus induction and plant regeneration, 6.11% and 1.66% respectively, were obtained using ovaries corresponding to trinucleate pollen grain stage (Table 4). Younger developmental stages produced a lower response. The use of mature ovaries produced lower induction rates than ovaries corresponding to trinucleate pollen grain. The analysis of variance showed meaningful differences between treatments for the rate of callus induction (data not shown).

. . . . . .

Discussion MCPA seems to be as effective for haploid production by 8ynogenesis in barley as it was for gynogenesis of rice (Zhou and Yang 1981). The results indicate that MCPA and IAA produced the best results when a cytoldnin is present during the induction process, in order to obtain subsequent plantlet regeneration. Of the two eytokinins used, ldnetin does not seem to be effective on barley ovaries, as only BA produced green haploid plantlets. High concentrations of growth regulators seem to be harmful. Thus, 2.5 t~M MCPA, added to 5.7 #M IAA and 4.4 /~M BA, either decreased callus induction (with the cultivar Berenice) or induced nonregenerable callus (with the dihaploid line).

Experiment 5: With coconut milk in the medium, the rate of callus formation was 7.69%, and half of the calli could be regenerated. More than one plant was obtained from some calli, so 10 plants were produced per 100 ovaries plated (Table 5). Addition of casein hydrolysate plus malt extract yielded a higher rate of induction than the control (without organic compounds), but no plants could be regenerated. Arginine plus proline had no effect on callus induction, but appeared useful for plant regeneration.

To obtain gynogenic haploid plants of barley, San Noeum (1976) used 2,4-D at a concentration of 9.0 pM. Our results suggest that, for the concentrations tested and the genotypes used, the rate of callus induction was higher for MCPA than for 2,4-D, whether used alone or in combination with another growth regulator. The number of regenerated plants was also higher for MCPA tharl for 2,4-D.

All plants, either obtained by direct embryogenesis or from callus during the first 45 days in culture, were haploid and green. However, chromosome counts carried out on calli that were kept on a regeneration medium for a period longer than two months showed cells with different numbers.

In line with previous results with barley (Singh 1986), calli which stayed for more than two or three months in regeneration medium were not able to regenerate. This can be related to induced chromosome variability. Altogether, a balance between auxins and cytokinins seems to be necessary. Furthermore, their total concentration should be as low as possible, since an excess of growth regulators may produce chromosomal variations in

143 tissue culture (Bayliss 1980, D'Amato 1985). Haploid cell lines have a greater tendency to increase their ploidy level than diploid ones (Sacristan 1971). High levels of sucrose are considered important for embryogenesis in barley anther culture (Dunwel11985, Sorvari and Scheider 1987). According to our results, barley ovaries also need high concentrations of sucrose for best results. However, it would be interesting to know if other sugars such as maltose can improve the production of green plantlets with concentrations of 50 or 60 g/l, as has been shown in anther culture (Hunter 1987). Coconut water was clearly beneficial for increasing the number of embryogenic calli formed and also for plants obtained. Glutamine enhanced embryogenesis in barley anther culture (Olsen 1987). Glutamine in coconut milk could be the cause of embryogenesis in ovary culture, too. In agreement with data obtained by other authors in rice (Zhou et al. 1983) and sugarbeet (D'Halluin and Keimer 1986), there was a strong genotypic effect for barley ovary culture, with the rate of callus induction of the dihaploid line being higher than that of Berenice. It is difficult to determine the developmental stage of the embryo sac. However, a correlation between the development of embryo sac and pollen exists in barley (Huang et al. 1982). Accordingly, pollen developmental stage was used as an indirect test to estimate ovary developmental stage. Ovary developmental stage seems to be one of the most important factors for gynogenesis. Ovaries plated when pollen was in trinueleate stage gave rise to calli at a rate of 6.11% and 45% of the calli obtained could be regenerated. The rates of callus induction and plantlet regeneration at the trinucleate stage were almost 2- or 3-fold

higher, respectively, than the binucleate stage. Huang et al. (1982) reported that binucleate and trlnudeate stages produced similar rates of callus induction (1.85%) when a high concentration of sucrose was added to the medium. Attached anthers perhaps supply some useful compounds to the ovary because the rate of callus induction as well as plant regeneration was lower when the anthers were removed. By improving the medium with growth regulators, organic compounds and a high sucrose concentration, and selecting the spikes at the trinucleate pollen grain stage, it was possible to improve the production of gynogenic haploid plants of barley. Acknowledgements. ACwas the recipient of a fellowship from the Ministerio de Edueaci6n y Ciencia, Spain. This research was supported by Project No. P CA-11/88 from Consejo Asesor de Investigaci6n de la Diputaci6n General de Arag6n, and by Project No. AGR 89-0494 from Comisi6n Interministerial de Ciencia y Tecnologla.

References

Bayliss MW (1980) Int Rev Cytol Supp 11a:113-114 BjOrnstand A (1986) Hereditas 104:171-191 Chu CC, Wang CC, Sun CS, Hsu C, Yin KC, Chu Cu (1975) Sci Sinica 18:659-668 Clapham D (1973) Z Pflanze~ftcht 69:142-155 D'Amato F (1985) Crit Rev Plant Sci 3:73-112

Production of gynogenic haploids of Hordeum vulgare L.

Haploid plants were regenerated from cultured unfertilized ovaries of Hordeum vulgare L. (barley). Optimal response was obtained by the addition of 0...
665KB Sizes 0 Downloads 0 Views