PlantCeU Reports

Plant Cell Reports (1996) 1 6 : 1 - 5

© Springer-Verlag 1996

Callus induction, regeneration of haploid plants and chromosome doubling in ovule cultures of pot gerbera (Gerberajamesonii) K. Miyoshi and N. Asakura* Sakata Seed Corp., Kakegawa Breeding Station, Kakegawa, Shizuoka 436-01, Japan * Present address: Graduate School of Science and Technology, Kobe University, Kobe 657, Japan Received 1 November 1995/Revised version received 1 May 1996 - Communicated by G. Phillips

Abstract. Gynogenetic plants of pot gerbera (Gettn~a jamesonii) were successfully produced from cultures of unpollinated ovules in vitro. Genotypic variations in the

number of ovules that formed callus were found among the lines tested. One particularly responsive genotype was found among 17 genotypes tested where the frequency of callus-forming ovules was 17.5%. Four genotypes formed no callus at all. The frequency of shoot formation from the callus varied from 0-19.6% in nine genotypes. Ploidy was determined by flow cytometry, and 37 (80.4%) regenerants were haploid, seven (15.2%) were diploid, and two (4.3%) were mixoploid with both haploid and diploid cells. The doubling of chromosomes was achieved by treatment with 0.05% colchicine for 2-6 d in vitro, and 24.2-34.1% of treated haploid plants were found to have been diploidized. Abbreviations: BA, 6-benzylaminopurine; NAA, 1naphthaleneacetic acid; IAA, indole-3-acetic acid; DAPI, 4' ,6-diamidino-2-phenylindole dihydrochloride; MS, Murashige and Skoog (1962) basal medium.

Introduction

There are two distinctive groups of gerbera plants (Gerbera jamesonii H. Bolus), namely, pot gerbera and cut-flower

gerbera. They are the results of breeding programs aimed at plants with different uses. Pot gerberas have short stems, very small foliage leaves, and they are commonly marketed as potted plants in pots of 12-15 cm in diameter. They are mainly propagated from seeds and some F Z cultivars have been produced. By contrast, cultivars of entflower gerberas produce many flowers with long stems and they are mainly propagated by tissue culture (Rogers and Tjia 1990; Behnke 1991). For better marketability of pot gerberas, further improvements in uniformity in terms of earliness, habit, color, and flower form are desired. However, two or three Correspondence to: K. Miyoshi

cycles of inbreeding result in degenerate progeny, and malformations and sterility occur frequently in this species because of its allogamous mode of reproduction. Therefore, the establishment of inbred lines by conventional selfing methods has been difficult (Cappadocia and Vieth 1990). It has been reported that haploids and doubled haploid plants can be obtained by culture in vitro of unpoUinated ovules of Gerbera, in which androgenesis has been found to be inefficient (Cappadocia and Vieth 1990). Production of gynogenetic plants in this species was first achieved by Cagnet-Sitbon (1981). Attempts were made to improve the method for gynogenetic production of plants in cut gerbera (Meynet and Sibi 1984; Ahmim and Vieth 1986; Cappadocia et al. 1988). In pot gerbera, Tosca et al. (1990) provided the first demonstration of the gynogenetic regeneration of shoots from calli that originated from unpollinated ovules. In gerbera, 76.0-100.0% of plants that regenerated from unpollinated ovule cultures were haploid (Meynet and Sibi 1984; Cappadocia et al. 1988; Tosca et al. 1990; Honkanen et al. 1991). Therefore, doubling of chromosomes in haploids is one of the limiting steps in the application of this technique to practical breeding. In this report, we describe the gynogenetic formation of plants of pot gerbera from cultures of unpollinated ovules and the doubling of chromosomes by treatment with colchicine in vitro. Materials and Methods Plant material: Seventeen genotypes were isolated from several highly

heterozygous populations of pot gerbera that had been maintained vegetatively. The donor plants were grown in a greenhouse at 17-24°C. Vigorous plants were maintained by supplying water-soluble fertilizer that contained 150 mg L a nitrogen every 10 d and by removing senescent leaves. Overhead irrigation was not applied to avoid the contamination of explants by microorganisms during culture in vitro. The capitula were harvested 1 d before anthesis of the first male florels during April-June and September-November. Tile ligules were cut off and ovules from the outermost 1-3 whorls of florets were

dissected under a light microscope (Fig. 1A, IB). Capitula and ovules were not sterilized, and florets were immersed in sterilized distilled water to prevent drying during dissection of the ovules. Eleven to 20 naked ovules were placed on 5 mL of medium in a 6-cm petri dish, and 1-60 dishes were prepared for each treatment.

by flow cytometry in leaves that had newly formed during the 2-month incubation on the drug-free medium.

Results

and D i s c u s s i o n

Induction of callus and regeneration of shoo~ For induction of callus, we used the medium of Ahmin and Vieth (1986), which consisted of the macronutrients and vitamins of Mnrashige and Skoog (1962) medium (MS), half-strength Fe-EDTA of MS, I-Ieller (1953) mieronutrients from which FeC13 was excluded, 100 mg L x myo-inositol, 10 g Lt sucrose, and 8 g La a g a r (Wako Pure Chemical Industries Ltd., Osaka, Japan). The pH of the medium was adjusted to 5.6 before autoclaving at 111*C (0.5 kg/em 2) for 30 rain. Growth regulators, such as IAA (0.1 mg La), NAA (0.1 mg L q) and BA (0.2 mg L-I), were added to the basal medium alone or in combination (Tables 1 and 2). The ealli that developed from the ovules were transferred to a petri dish that contained 5 mL of the regeneration medium, namely the basal medium in which vitamins had been replaced by 80 mg L "1 adenine sulphate, plus 0.1 mg L -1 IAA and 2 mg L : BA. If necessary, the shoots were transferred to medium that contained 3 g L -1 Hyponex® (N:P:K=I:I:3; The Hyponex Company, Inc., Copley, OH, USA), 0.05 mg L -~ IAA, 20 g L -~ sucrose and 10 g L-~ agar to induce formation of roots (Fig. 1E). The frequency of callus formation was calculated from the number of ovules that formed callus as a percentage of the total number of ovules after 2 months of culture. Rooted plantlets were directly transferred to sterilized soil in pots and grown to maturity in a greenhouse, which was air-conditioned to maintain a temperature of 15-23"C.

Within 4 weeks of culture initiation, one callus (or two in rare cases) developed from each of the responding ovules. Only those caUi that were compact and had developed in the middle of the ovule after the ovary wall had been ruptured (Fig. 1C) were subcultured and counted.

0-21 M

Determination of ploidy by flow cytometry: Leaf segments of about 1 cm 2 were excised from plants that had been grown in vitro from cultured ovules and chopped with a razor blade in 1.5 mL of a solution that contained 6.07 g L q NaC1, 0.86 g L-I Tris base, 0.079 g L q CaCI2, 0.086 g L a MgSO4"(7H20), 0.36 g L q bovine serum albumin, 14.3 ml L-: Tergitol NP-40 (Sigma Chemical Company, St Louis, MO, USA), 9.63 g L q MgCI 2"(6HzO), 8.93 mg L q 4',6-diamidino-2-phenylindole dihydrochloride (DAPI; Sigma Chemical Company, St Louis, MO, USA) and 98.2 ml L "1dimethyl sulfoxide. After chopping of tissue, the solution, containing cell constituents and large remnants of tissue, was passed through a nylon filter (40-1xm mesh). Preparations were kept i n darkness on ice. For flow cytometry, we used a CA II eytometer (Partec, MUnster, Germany) equipped with a filter combination of UG1 and GG 435. Trout erythroeyte nuclei (BioSure®; Riese Enterprises, San Jose, CA, USA) were used as a biological standard.

0-1M-B

Chromosome count in root t/ps: Chromosomes were counted in root tips taken from three acclimatized plants for which ploidy had already been determined by flow cytometry. The collected root tips were pretreated with 2 mM 8-hydroxyquinoline at 20"C for 3-4 h. Pretreated root tips were fixed in a mixture of ethanol and acetic acid (3: I, v/v) for at least 24 h. After the root tips had been rinsed three times with distilled water, they were macerated with an enzyme solution that contained 4% Cellulase Onozuka RS (Yakult Pharmaceutical Ind. Co., Ltd., Tokyo, Japan), 1% Pectolyase Y-23 (Seishin Corp., Tokyo, Japan), 7.5 mM KC1 and 7.5 mM EDTA at pH 4.0 (Nishibayashi 1990) at 37"C for 3040 min. After rinsing, the root tips were tapped for a few seconds with the tip o f a needle on a glass slide with a few drops of fixative and then air-dried. The dispersed root-tip cells were stained with a drop of 1% aceto-oreein for 10 min and observed under a light microscope. Colchicine treatment:. Colchicine was dissolved in water to make a stock solution which was then filter-sterilized. It was added to the rootinduction medium at a final concentration of 500 mg L l. A total of 185 haploid plants with 4-6 leaves and well-developed root systems were planted on this medium and incubated in the light (33 ,umol m 2 s -1) at 25"C. Treatments were conducted for 2, 3, and 6 d. The treated plants were transferred to colchicine-free medium and ploidy was assessed

T a b l e l . F r e q u e n c y o f callus induction in ovule cultures o f 17 genotypes o f pot gerbera. F r o m t w o - w a y A N O V A , significance levels were as follows: genotypes, p=0.0001; media, p=0.08. IAA (0.1 mg L:) + BA (0.2 mg L -~)

Genotype

NAA (0.1 mg L -~) + BA (0.2 mg L-:)

No. o f No. of Callus No. of No. of Callus cultured ovules induction cultured ovules induction ovules forming per 100 ovules forming per 100 callus ovules callus ovules

0-8 M-A

40

7

17.5

33

4

12.1

952 0-22 F- B

1119 100

67 4

6.0 4.0

312 87

17 5

5.4 4.7

178

6

3.4

198

6

3.0

206

6

2.9

218

3

1.4

7-29 F

90

2

2.2

76

0

0.0

410F

951

18

1.9

349

13

3.7

470M

663

7

1.1

251

1

0.4

331M

625

6

1.0

129

4

3.2

7-29 M

99

1

1.0

136

0

0.0

7-48M

168

1

0.6

131

0

0.0

0-1M-E

428

2

0.5

411

9

2.2

0-1F-C

77

0

0.0

79

3

3.8

0-30RM-D

57

0

0.0

59

0

0.0

0-22 M-A

33

0

0.0

20

0

0.0

0-1M-A

53

0

0.0

61

0

0.0

0-1M-C

155

0

0.0

137

0

0.0

Genotypic variations in callus formation were observed among the various lines of pot gerbera tested. Thirteen out of 17 genotypes responded with at least one callus among cultured ovules. Seven genotypes had frequencies of callus induction of 3.0% or higher, on media supplemented with IAA or NAA both at 0.1 mg L -t plus 0.2 mg L -1 BA (Table 1). The highest frequency (17.5%) was observed with genotype '0-8M-A' on medium that contained 0.1 mg L1 IAA and 0.2 mg L -1 BA. In cut gerberas, genotypic differences in callus formation from unpolliuated ovules have also been reported, and the highest frequencies under optimal culture conditions were 43.3% for '1(9-8' (Cappadocia et al. 1988), 12.5% for 'Super Gerbera' (Ahmim and Vieth 1986), and 8% for 'Fresultane' (Meynet and Sibi 1984). In the present study, four genotypes ('0-30R M-D', 'O-22M-A', '0-1M-A', '0-1MC') formed no callus at all. For haploid Needing of gerbera, it will be necessary to transfer the putative gene(s)

Figure 1A-H. Gynogenesis of pot gerbera (Gerberajamesonii H. Bolus). (A) Intact floret used in the experiment (bar=-2.0 mm). (B) Dissected ovule, indicated by an arrow (bar=2.0 ram). (C) Callus on a ruptured ovary wall after 4 weeks in culture (bar=-0.5 ram). (D) Plant regenerated from callus (bar=0.5 cm). (E) Doubled haploid plant just before acclimatization (bar=-2 cm). (F) Chromosomes from root tips of a doubled haploid plant from an ovary-derived callus; 2n=2x=50 (bar=5 ixm). (G) Chromosomes from the root tip of a haploid plant from aa ovary-derived callus; 2n=x=25 (bar=10 txm) (H) Anthesis of a doubled haploid plant (bar----4cm).

24q0

that control the

gynogenetic response in unpoUinated ovule culture from positive genotypes, such as ' 0 - 8 M - A ' , to recalcitrant geuotypes. 2. The effects of BA and IAA on callus induction in ovule culture of pot gerbera. From two-way ANOVA, significance levels were as follows: genotypes, p=0.0128; media, p=0.1121 (also included nonresponsive data from IAA alone treatment).

A

2000

Table

IAA (0.1 mgL-:) + BA (0.2 m~ k 1) genotype 0-8 M-A 0-22 F- B 0-21 M 0-22 M-A 0-1M-B 7-29 F 7-29 M 7-48 M 0-1M-E 0-1F -C 0-30R M-D 0-1M-A

No. of No. of culture ovules d forming ovules callus 247 212 117 31 157 63 38 11 36 25 38 47

28 6 5 1 1 0 0 0 0 0 0 0

ISOO[

E

Z

I~00

BA (0.2 mg k 1)

Callus No. of No. of induction culture ovules per 100 d forming ovules ' o v u l e s callus 11.3 2.8 4.3 3.2 0.6

228 198 114 29 149 55 35 20 20 19 39 40

0.0

0.0 0.0 0.0 0.0 0.0 0.0

800

Callus induction per 100 ovules

19 3 2 0 3 0 0 0 0 0 0 0

8,3 1,5 1.8 0,0 2.0 0.0 0,0 0,0 0.0 0.0 0.0 0,0

B A was required for the formation o f callus, and no callus was f o r m e d on m e d i u m that contained only 0.1 m g L 1 I A A as phytohormone (data not shown). However, I A A had a synergistic effect with BA: four out of five genotypes had higher frequencies of callus formation on medium that contained B A and I A A than on medium that contained solely B A (Table 2). Maynet and SiN (1984) included GA3 in their callus-induction medium. In our study, however, G A 3 had no dear-cut promotive effect on the induction of callus from ovules when it was added to the m e d i u m (data not shown).

0

:ao

80

40

DNA

I c 2c

t6o zoo Conzent

B

700 600

~d

E

500

Z

40O

¢3

300t 2

0

0

~

0

o

40

eo

,~o

DNA

2c

zoo

~so

No. of shoots

No. of shoots per 100 calli

952

92

18

19,6

410F

45

8

17.8

0-8M-A

58

7

12.1

470M

10

1

10.0

12

1

8.3

0-22F-B

12

0

0.0

O-1M-B

10

0

0.0

0-1M-E

11

0

0.0

0-21M

19

0

0.0

F i v e out o f nine genotypes regenerated shoots from callus (Fig. 1D). Nevertheless, the regeneration frequency

~00

4c

640

C

480

400

Z

331M

240

Content

5GO

No. of calli transferred

2.80

800

Table 3. Genotype dependent response in the regeneration of shoots from gynogenetic calli in pot gerbera

Genotype

a4o

~ao 340

1SO 80

0 O

3;~

64

SS

DNA

l~S

I6Q

},~;~

a24

2.5G

Content

1 c 2c 4c Figure 2. Analysis by flow cytometryof haploid, diploid and mixoploid leaves from plants regenerated from callus derived from unpollinated ovules of pot gerbera. (A) Haploid: (B) Diploid; (C) Mixoploid with haploid and diploidcells.

of '952' (19.6%) (Table 3) was lower than those of responsive genotypes of cut gerbera. In the latter case, 37.8-45.5% of caUi formed shoots (Ahmim and Vieth 1986). Further studies will be needed to determine whether specific factors (such as phytohormones in the regeneration medium) required for morphogenesis of callus from pot gerbem compared to those for cut gerbem are different, or reflect only the narrow range of genotypes examined. Assessment of ploidy: Chromosome counts were made using root tips of randomly selected plants whose ploidy had been confirmed by flow cytometry. Flow cytometry has been reported to be efficient for determinations of ploidy (De Laat et al. 1987), with a strong corrdafion being obtained between nuclear DNA content chromosome number (Fahleson et al. 1989). This result was confirmed in pot gerbem by examining three regenerants (Fig.lF, 1G, Fig. 2). The ploidy of 46 regenerants was assessed by flow cytometry. Seven (15:2%) regenerants were diploid; 37 (80.4%) were haploid; and two (4.3%) were mixoploid, being both haploid and diploid. These mixoploids were detected for the first time in gynogenetic gerbera plants by flow eytometry. Detailed studies are needed to clarify the origin and mode of occurrence of mixoploids, including analysis of ptoidy from initiation of callus to regeneration of shoots by flow cytometry. The predominance of haploid plants among the regenerants found in the present study resembles the results reported for cut gerbera, in which the frequencies ofhaploids were 76.0-100% (Meynet and Sibi 1984; Cappadocia et al. 1988; Tosca et al. 1990; Honkanen et al. 1991)

50 I

D2 days

~_ 40 -

~

~

~

--

3O

__[]3days • 6 days

~. 2o a, 10 0

,

x

x+2x

,

2x

2×+4x

4x

x+4x/x+ 2x+4x

Ploidy Figure 3. q'he effects of the duration of 0.05% colchicine treatment on diploidization of haploid plants obtained from unpollinated ovule cultures of pot gerbera. Haploid plantlets were treated for 2, 3 or 6 d with 0.05% colchicine, and 24.2-34.1% of treated plants were shown to be diploids by flow cytometry. Mixoploid plantlets with haploid and diploid cells (30.3-41.5%), as well as those with diploid and tetraploid cells (17.121.2%), were also obtained after the treatment with colehicine (Fig. 3). Four out of 37 plantlets treated with

colchicine for 6 d showed necrosis and died eventually. Foliage and capitula of doubled haploid plants were smaller than those of the source plants (Fig. 1H). To increase the frequency of induction of doubled haploids, it will be necessary to evaluate the effects of briefer treatment and/or use of colchicine-supplemented medium during the induction of callus or regeneration of shoots (Alemanno and Guiderdoni 1994). Antimicrotubule herbicides, such as oryzalin (Wan et al. 1991), have been successfully used for diploidization in rice. Tosca et al. (1995) reported, from preliminary investigation of percentages of haploid, diploid, and tetraploid nuclei in leaves of treated plants, that oryzalin may be superior to colchicine because of its lower phytotoxicity and the absence of long term effects. Therefore, the application of oryzalin and other antimicrotubule herbicides should be tested for the ability of such compounds to cause diploidization of pot gerbera haploids. While genotypic differences were apparent, the method reported here seems suitable for production of homozygous lines for breeding of pot gerbera. Comparisons of the productivity of F 1 seeds of gynogenetic plants with those of conventionally derived lines are currently being made.

Acknowledgments. The authors thank Prof. Masahiro Mii of Chiba University for critical reading of the manuscript, Mrs. Y. Izaki and Mrs. K. Anzai-Akiyamafor their technical assistance, and Dr. Tadashi Sato of Tohoku Universityfor help with the statisticalanalysis. References

Alemanno L, GuiderdoniE (1994) Plant Cell Reports 13:432-436 AhmimM, ViethJ (1986)Can J Bot 64:2355-2357 Behnke M (1991) In: Ball V (ed) The Ball RedBook, 15th ed, Greenhouse growing. Geo J Ball Publishing, West Chicago, IL, pp 555-559 Cagnet-SitbonM (1981)Agronomic1:807-812 Cappadocia M, ChretienL, LaublinG (1988) Can J Bot 66:1107-1110 Cappadocia M, Vieth J (1990) In: Bajaj YPS (ed), Biotechnologyin Agriculture and Forestry, vol 12, Haploids in crop improvement I. Springer-Verlag, Berlin Heidelberg De Laat AMM, Gobde W, Vogelzang JDC (1987) Plant Breeding 99: 303-307 Fahleson J, Dixelius J, Sundberg E, Glimelius K (1988) Plant Cell Reports7:74-77 HellerR (1953)Ann Sci Nat Biol Veg 14:1-123 HonkanenJ, Aapola A, DeWitJC, Esendam HF, Seppanen P, Tormala T, StraversLJM (1991) Acta Hortic300:341-346 MeynetJ, Sibi M (1984) Z Ptlanzenzuchtg93:78-85 MurashigeT, SkoogF (1962)PhysiolPlant 15:473-497 Nishibayashi S (1990)Plant TissueCultLeft(Tokyo)7:127-129 Rogers MN, Tjia BO (1990) Gerbera productionfor cut flowers and pot plants: Growers handbookseries, vol 4. Timber Press, Portland, OR ToscaA, LombardiM, Marinoni L, Conti L, Frangi P (1990) Acta Hort 280:337-340 ToscaA, PandolfiR, CitterioS, Fasoli A, SogorbatiS (1995) Plant Cell Reports 14: 455-458. Wan Y, Duncan DR, Raybum AL, Petolino JF, Widholm JM (1991) TheorAppl Genet81:205-211