Planning to submit to In Vitro Cell. Dev. Biol.

The development of an efficient plant regeneration system from callus derived from root segments of garlic (Allium sativum L.)

Si-Jun Zheng, Betty Henken, Frans A. Krens & Chris Kik

Plant Research International, Wageningen University and Research Center, P. O. Box 16, 6700 AA Wageningen, The Netherlands

Corresponding author:C. Kik, Tel: 31 317 477001; Fax: 31 317 418094;

E-mail:

Summary

Root segments from in vitro plantlets of four widely cultivated garlic (Allium sativum L.) cultivars in Europe (Messidrome, Morado de Cuenca, Morasol and Printanor) were used for callus induction and plant regeneration. Eight treatments for callus induction with different auxin and cytokinin levels were compared for their effect on callus induction and later plant regeneration. There were no statistically significant differences among cultivars for the number of root segments that induced callus. The average callus induction frequency from root segments was 34.7 %. After callus induction for two months, callus lines were transferred to regeneration medium, i.e. MS30 supplemented with 1 mg/l (4.6 M) kinetin. Regenerated plants were obtained via somatic embryogenesis and organogenesis. Calli derived from different experiments were quite uniform with respect to their regeneration potential. Also it was found that our regeneration system was cultivar independent. Highly statistically significant differences were found in the frequency of shoot regeneration among different callus induction treatments. When cytokinin was present during callus induction, later shoot regeneration was high, ranging from 30.10 % to 47.60 %. Already a small amount of cytokinin (0.1 mg/l 2ip: 0.5 M) in the callus induction medium had a profound effect on later regeneration. All in all, an efficient callus induction and plant regeneration system could be developed from segments taken along the entire length of the roots. This system can be used for garlic transformation.

Key words: callus induction,garlic, in vitro plantlets, plant regeneration, root segments

Introduction

Garlic (Allium satium L.) is a very important and widely cultivated crop that has been domesticated already for a long time and which is known for its culinary and medicinal use. Up till now, garlic breeding has been limited to clonal selection of wild varieties or spontaneous mutants because most of the germplasm of garlic is sexually sterile. To improve the current garlic cultivars two ways are open: a. the development of new garlic cultivars via sexual hybridisation ( Kamenetsky and Rabinowitch, 2001) and b. the development of an efficient transformation system (Kondo et al., 2000). Our research focuses on the development of a transformation system for garlic as all current garlic cultivars propagate asexually and because we have been successful in developing an efficient transformation system for a close relatives of garlic, i.e. onion and shallot (A. cepa; Zheng et al., 2001). Genetic transformation has proven to be a powerful tool for modification of plants with desired traits in many crops. However, the use of this tool depends on the availability of a highly efficient callus production and regeneration system. Regeneration via somatic embryogenesis in garlic was first reported by Abo El-Nil (1977) from calli of bulb leaf discs and stem tips. Subsequently, it was reported from basal plate (Al-Zahim et al., 1999), leaf (Wang et al., 1994; Zheng et al., 1998), receptacle (Xue et al., 1991) and flower buds (Suh and Park, 1988). In the aforementioned cases the efficiency of plantlet regeneration was very low. Other reports on in vitro regeneration of garlic plants by organogenesis or embryogenesis were mainly based on shoot-tip or stem disc explants (Kehr and Schaeffer, 1976; Nagasawa and Finer, 1988; Choi et al., 1993; Ayabe et al., 1995; Ayabe and Sumi, 1998; Myers and Simon, 1999; Kondo et al., 2000). Suh and Park (1995) used abnormal roots derived from anthers, pedicels and bulbils of garlic as explants to regenerate plantlets. Haque et al. (1997, 1999) obtained high frequency shoot regeneration from the root tip of garlic without an intervening callus phase. So, root-tip explants have been commonly used for the development of a garlic regeneration system (Haque et al., 1997, 1998; Barandiaran et al., 1999a, 1999b; Roledo et al., 2000). The problem with this type of explant is that there is only one root tip on a root. Myers and Simon (1998) used root segments from shoot tip-derived plantlets for continuous callus production and regeneration. However, this system was time-consuming and took about eight months. To date, there is as yet no consistent, efficient and reliable regeneration system that could be used effectively for the genetic transformation of garlic. A successful garlic transformation system is highly dependent on an efficient and reliable regeneration system that is applicable to a wide range of cultivars. Here, we report on the development of an efficient callus induction and plant regeneration system not only for root-tips but also for other root segments, which can be used for garlic transformation.

Material and methods

Plant material

Virus-free mature cloves from four most widely cultivated garlic cultivars in Europe, namelyMessidrome, Morado de Cuenca, Morasol and Printanor, were used to initiate in vitro plantlets. These cloves were obtained from INRA Avignon - Montfavet, France.

In vitro shoot induction

The dry outer layers of the garlic cloves were peeled off and the cloves were cut into small cubes that contained the basal part of the stem as described by Ayabe and Sumi (1998). The stems were surface-sterilized with 96 % (v/v) ethanol for 5 min., 10 % (v/v) sodium hypochlorite solution with 1 drop of Tween 20 per 100 ml for another 5 min., and then washed 10 times with sterile distilled water. After removing the residual storage and foliage leaves, the base of the stem disc, approximately 1 mm thick, was excised. Each disc was cut into 2 or 4 pieces depending on the size. These pieces were then placed on a shoot initiation medium consisting of Murashige and Skoog (1962) basal salts and vitamins supplements with 3 % (w/v) sucrose, 0.4 % (w/v) phytagel (Duchefa, the Netherlands), 0.1 mg/l (0.5 M) 1-naphthaleneacetic acid (NAA, Duchefa, the Netherlands) and 0.5 mg/l (2.5 M) 6-(- -dimethylallylamino) purine (2ip, Sigma, USA). After shoot initiation, approximately 4-6 weeks later, small shoots were transferred to a root induction medium consisting of Murashige and Skoog (1962) basal salts and vitamins supplements with 3 % (w/v) sucrose, 0.4 % (w/v) phytagel, without plant growth regulators. The media were adjusted to pH 5.8 prior to autoclaving (103 kPa, 121C; 20 min). NAA was added to the media before autoclaving, while filter sterilised 2ip was added to the media after autoclaving. In vitro plantlets of garlic were also maintained on this last medium. All treatments including shoot initiation, root induction and in vitro plantlet maintenance were carried out at an ambient temperature of 25C with a 16h photoperiod (ca. 60  E s-1 m-2; lamps used: Philips, TLD 50W/840HF, Electronic NG).

Callus induction

Garlic roots, including root tips from stem disc-derived plantlets, were cut into 1 cm segments and transferred to callus induction medium consisting of Murashige and Skoog (1962) basal salts and vitamins supplements with 3 % (w/v) sucrose, 0.4 % (w/v) phytagel with different combinations of auxin and cytokinin (Table 1). Two types of auxins were used, namely 2,4-dichlorophenoxyacetic acid (2,4-D, Duchefa, the Netherlands) and 4-amino-3,5,6-trichloropicolinic acid (picloram, Duchefa, the Netherlands). One concentration of 2,4-D (1 mg/l (4.5 M)) was used because this concentration proved to be optimal in onion and shallot tissue culture (Zheng et al. 1998). Three concentrations of picloram were used, namely 1, 2 and 5 mg/l (4.1 M, 8.3 M and 20.7 M). A low concentration of 2ip (0.1 mg/l: 0.5 M) was used in combination with 2,4-D or picloram. Media were adjusted to pH 5.8 prior to autoclaving (103 kPa, 121C; 20 min.). 2,4-D was added to the media before autoclaving, while filter sterilised picloram and 2ip were added to the media after autoclaving. Twenty root segments per dish were placed on callus initiation medium. Dishes were sealed with parafilm and explants were subcultured to fresh medium every four weeks. Callus induction was carried out in the dark. The frequency of callus induction (percentage of root segments with callus formation) was recorded after eight weeks.

Plant regeneration

All calli from the different treatments were transferred to regeneration medium consisting of Murashige and Skoog (1962) basal salts and vitamins supplements with 3 % (w/v) sucrose, 0.4 % (w/v) phytagel and 1 mg/l (4.6 M) 6-furfurylaminopurine (kinetin; Duchefa, the Netherlands). The medium was adjusted to pH 5.8 prior to autoclaving (103 kPa, 121C; 20 min). The Petri dishes were placed under an ambient temperature of 25C with a 16h photoperiod (ca. 60  E s-1 m-2; lamps used: Philips, TLD 50W/840HF, Electronic NG). Explants were subcultured every four weeks to fresh medium, and shoot regeneration (percentage of explants with at least one shoot) was determined after two months.

Statistical analysis

The experimental set-up used to analyse callus induction and plant regeneration from root segments was a two-factorial design. The main effects were cultivars (4 cultivars used) and callus induction media (8 treatments used). Each treatment consisted of at least five Petri dishes and per Petri dish 20 root segments were present. A generalised linear model (McCullagh & Nelder, 1990) based on a binomial distribution and a logit as link function was used for the analysis.

Results and discussion

Callus induction

A total of 3329 root segments excised from in vitro plantlets of the four garlic cultivars analysed were subjected to eight callus induction treatments (Table 1). A small amount of callus formation from root segments could be observed after two weeks of culture when explants started to swell. Callus was visible from different treatments after four weeks of culture. The explants with tiny calli were then transferred to the same fresh callus induction medium. After eight weeks of culture, callus induced on root segments of all four cultivars tested was present in all eight treatments. There were a total of 1156 callus lines, which were produced from 3329 root segments. The average of callus induction frequency was 34.7 %. There were two ways in which callus could form. One way of callus formation was from the cutting edges of an explant and the other way was callus formation along the entire length of the explant (Fig.1 A and B). All calli formed were compact and exhibited a nodular-like structure. There were no statistically significant differences among cultivars for the percentage of root segments that induced callus (Table 2). This means that root segments from all four cultivars could be used successfully to induce callus. Fellner and Havranek (1994) failed to induce callus using root explants from both A. sativum and its progenitor A. longicuspis. They observed that root explants turned green in the different media they tested and found that the root explants used exhibited only a slight enhancement of their size without producing callus. There was also no statistically significant differences between cultivars x treatments for the percentage of root segments that induced callus (Table 2). However, there were highly statistically significant differences among callus induction treatments for the percentage of root segments that induced callus (Table 2). The frequency of callus induction (24.17 %, 44.43 % and 56.48 %) was statistically significantly promoted when picloram was applied in increasing concentration from T1 (1 mg/l: 4.1 M), T2 (2 mg/l: 8.3 M) to T3 (5 mg/l: 20.7 M), respectively (Table 3). When auxin was used alone in callus induction (T1, T2, T3 and T7), there was a higher frequency of callus induction compared to the situation when auxin was combined with a cytokinin in callus induction (T4, T5, T6 and T8). It was also observed that root segments of cv. Messidrome in the treatment T8 usually became pink and necrotic when a little amount of cytokinin 2ip (0.1 mg/l: 0.5 M) was applied to the callus induction medium with 1 mg/l 2,4-D (4.5 M). The other three cultivars behaved normal for callus induction in treatment T8. Calli were mainly induced from one or both cutting edges of a root segment whenever a cytokinin was applied to the callus induction (T4, T5, T6 and T8; Fig. 1 A). When auxin was used alone in callus induction (T1, T2, T3 and T7), calli were mainly formed along the entire length of the explant and exhibited nodular-like structures (Fig.1 B).

Plant regeneration

After callus induction for two months, a total of 1156 callus lines from the four cultivars used were produced in the eight callus induction treatments. These callus lines were transferred to a regeneration medium, i.e. MS30 supplemented with 1 mg/l (4.6 M) kinetin. The first shoot was produced after two weeks (Fig.1 C ). Most shoots, however, were induced after subculture. Callus lines were transferred to the same fresh regeneration medium after four weeks. The frequency of shoot regeneration was scored after calli were placed on the regeneration medium for two months. A total of 207 callus lines showed regeneration ability. Regenerated plants were obtained via somatic embryogenesis and organogenesis. There were no statistically significant differences between the different experiments concerning the frequency of shoot regeneration. This means that calli derived from the different experiments were quite uniform with respect to their regeneration potential. There were also no statistically significant differences among cultivars for the frequency of shoot regeneration (Table 2). This indicated that the regeneration system was cultivar independent. There were also no statistically significant differences between cultivars x treatments interactions with respect to the frequency of shoot regeneration (Table 2). However, highly statistically significant differences were found in the frequency of shoot regeneration among different callus induction treatments (Table 2). From Table 3, it is clear that a small amount of cytokinin (0.1 mg/l 2ip: 0.5 M) in the callus induction medium had a profound effect on later regeneration potential. The frequency of shoot regeneration ranged from 6.60 % to 8.94 % in T1, T2, T3 and T7 treatments, while in treatments T4, T5, T6 and T8 in which cytokinin was included in callus induction medium, the frequency of shoot regeneration was much higher: it ranged from 30.10 % to 47.60 %. In most cases, multiple plantlets were formed from one piece of callus after two months on regeneration medium (Fig. 1 D). Cytokinin 2ip present in the callus induction medium appears to have a specific effect during the regeneration phase. This is in agreement with the finding reported by Haque et al. (1997), who observed that only auxins in combination with cytokinins induced shoot regeneration. Our results indicated that the auxin picloram can be used in a broad range from 1 - 5 mg/l for callus induction. Our data also fit with previous studies on other Allium species which indicated that picloram is the most suitable auxin to use (Phillips and Luteyn, 1983; Phillips and Hubstenberger, 1987; Eady et al., 1998).

Nowadays, root-tip or shoot-tip explants are widely used for callus induction and plant regeneration in garlic tissue culture (Haque et al., 1997, 1998; Barandiaran et al., 1999a, 1999b; Roledo et al., 2000). The disadvantage using such explants from in vitro culture is the limited number of explants. Myers and Simon (1998) reported on a continuous callus production and regeneration system for garlic using root segments including root tips. However, the whole procedure for callus induction and later plant regeneration obtained in a different way took 7-9 months. The regeneration efficiency reported by Myers and Simon (1998) showed a large variation among different garlic clones and treatments. We used root segments as explants and developed a callus induction and regeneration procedure which took 4 months (2 months for callus induction and another 2 months for regeneration). The critical point in our system is that a little amount of cytokinin (0.1 mg/l 2ip: 0.5 M) should be included in the callus induction medium. Plant regeneration was high among four cultivars, ranging from 30.10 % to 47.60 %. The advantage of using root segments for callus production is that a lot of root segments can be harvested every 4-6 weeks from in vitro plantlets.

The germplasm used in this study represented different types of European cultivars. Although these cultivars behaved quite different in in vitro culture, e.g. the plantlets of cv. Messidrome could easily become vitrified or dormant during subculture, the regeneration system reported here is efficient, reliable and cultivar independent. Based on previous research on onion and shallot using young callus for transformation (Zheng et al. 2001), we are confident that the root segment callus, induced in this study, is a suitable starting point for garlic genetic transformation by Agrobacterium tumefaciens.

Acknowledgements

We would like to thank Paul Keizer (Plant RI, Wageningen) for his help on the statistical analysis and our G&H project partner Véronique Chovelon (INRA Avignon - Montfavet, France) for providing the virus-free garlic bulbs. The list of partners involved in the G&H project can be found at our home page: This research is partially financed by an EU FP 5 grant in the area of key action 1 (QLK1-CT-1999-498).

References

Abo El-Nil, M. M. Organogenesis and embryogenesis in callus cultures of garlic (Allium sativum L.). Plant Science Letters. 9: 259-264; 1977.

Al Zahim, M. A.; Ford Lloyd, B. V. ; Newbury, H. J. Detection of somaclonal variation in garlic (Allium sativum L.) using RAPD and cytological analysis. Plant Cell Reports. 18: 473-477; 1999.

Ayabe, M.; Taniguchi, K.; Sumi, S. I. Regeneration of whole plants from

protoplasts isolated from tissue-cultured shoot primordial of garlic (Allium

sativum L.). Plant Cell Reports. 15: 17-21; 1995.

Ayabe, M.; Sumi, S. Establishment of a novel tissue culture method, stem-disc culture, and its practical application to micropropagation of garlic (Allium sativum L.). Plant Cell Reports. 17: 773-779. 1998.

Barandiaran, X.; Martin, N.; Rodriguez Conde, M. F.; Di Pietro, A.; Martin, J. An efficient method for callus culture and shoot regeneration of garlic (Allium sativum L.). HortScience. 34: 348-349; 1999a.

Barandiaran. X.; Martin, N.; Rodriguez Conde, M. F. Di Pietro, A.; Martin, J. Genetic variability in callus formation and regeneration of garlic (Allium sativum L.). Plant Cell Reports. 18: 434-437, 1999b.

Choi, S. Y.; Paek, K. Y.; Jo, J. T. Plantlet production through callus culture in

Allium sativum L. Journal of the Korean Society for Horticultural Science. 34:

16-28, 1993.

Eady, C. C.; Butler, R. C.; So, Y. Somatic embryogenesis and plant regeneration from immature embryo cultures of onion (Allium cepa L). Plant Cell Reports. 18: 111-116, 1998.