SOlllll Afncan Journal of Botany 2001 67 74 - 77 Copyrl(}lll 0 NISC Ply Lid Pnnled m Soul/• Afnca - All nghls reserved SOUTH AFRICAN JOURNAL OF BOTANY ISSN 0254-6299

Short Communication

Micropropagation of the River Lily, Crinum variabile (Amaryllidaceae)

CW Fennell'*, NR Crouch2 and J van Staden'

' Research Centre for Plant Growth and Development, School of Botany and Zoology, University of Natal Pietermaritzburg, P/Bag X01 Scottsville 3209, South Africa l Ethnobotany Unit, National Botanical Institute, PO Box 52099, Berea Road, Durban 4007, South Afdca *Corresponding author, e-mail: [email protected]

Received 25 July 2000, accepted in revised form 5 October 2000

Crinum variabi/e (Jacq.) Herb. is restricted to the Plant growth regulators were not required for the induc­ Namaqualand, Bokkeveld Mountains and western Karoo tion of shoots but benzyladenine (BA), in the absence of regions of South Africa. Although its potential as a hor­ naphthalene acetic acid (NAA), promoted their out­ ticultural subject has been recognised, factors relating growth. The inclusion of 5gl·' activated charcoal to the biology of the plant limit large scale propagation. improved development by increasing the bulblet size These include short-lived seed viability and vegetative and frequency with which shoots formed bulblets. offsets that are produced too slowly or infrequently. As These bulblets were used to initiate further cultures by an alternative to conventional methods, C. variabile was splitting them in half. Such a system provided for the successfully propagated in vitro using twin-scale continuous multiplication of C. variabile bulblets which explants. Shoots developed in the axes of twin-scales readily transferred to the soil. when placed on a Murashige and Skoog (MS) medium.

Few South African crinums are as restricted in distribution as elled; the fresh lettuce-green leaves generally only appear Crinum variabile (Jacq.) Herb., known from only a handful of following the onset of the winter rains. The leaves, which are localities around Garies, Nleuwoudtville and Sutherland. strap-shaped and channelled, are at first stiffly erect but arch Such is its localisation that following its description by and broaden as they age, reaching 35cm in length and 4cm Jacquin in 1804, as Amaryllis variabilis Jacq., it was not re­ in breadth (Verdoorn 1964). The margins are minutely located in the wi ld until 1961, when Mr Dave Hardy of the toothed (Manning and Goldblatt 1997). In habitat amongst Botanical Research Insti tute found a population in th e Groen riverbed rocks, the plants benefit from the deep waters that River near Garies. In flower (Figure 1A), the River Lily as it gather during the winter months, some specimens being is known, may reach 1m in height and produce six to 12 temporarily inundated. Verdoorn (1973) recorded that the flowered umbels. Plants bloom between February and April ovoid bulbs reached 26cm in length and 9cm in diameter, in early Autumn, during extremely hot and dry conditions narrowing gradually towards the apex and so without a dis­ (Verdoorn 1964). This is unlike other drought-adapted tinct neck. However, subsequent field observations crinums which flower only after rain (Lehmiller 1987). (Lehmilier 1987) have reported on the existence of bulbs Accordingly, C. variabile should prove to be a useful horti­ with diameters of up to 20cm. Offshoots often sprout from cultural subject in the drier regions of South Africa. The the axiis, and from around the bulb (Verdoorn 1964 ). heads of narcissus-scented (Manning and Goldblatt 1997) Crinums have a long history of cultivation (Bryan 1989). flowers typically mature from white to a deep rose-pink; this Yet there is still much scope for the selection and develop­ colour progression gave rise to its specific epithet. The large ment of cultivars among African species (Stirton 1980) for tru mpet-shaped flowers open wider as they mature, reach­ which the centre of diversity lies south of the Sahara ing up to 80mm across, revealing the white anthers. Crinum (Fangan and Nordal 1993). Their potential development as variabile is one of only three southern African crinums pot plants has been recognised by Jansen Van Vuuren eta/. whose funnel-shaped flowers have segments longer than (1993) and Lehmiller (1996) as they rarely or only sporadi­ the perianth tube, the others being C. lfneare L.f. and C. cally produce offsets. Conventionally, offsets or seeds are macowanii Bak. (Verdoorn 1973). The closest relative of C. used to propagate the plants. There are inherent difficulties variabile is considered to be the rare C. lineare U .. which is with both techniques because natural rates of vegetative restricted to the Eastern Cape Province (Verdoorn 1973). propagation are slow and the recalcitrant seeds, which are During flowering, the leaves are normally brown and shriv- fleshy, germinate readily and without water (Verdoorn 1973) South African Journal of Botany 2001 . 67 74-77 75

ID.

Figure 1: Stages in the micropropagation of Crinum varia bile; (A) C. variabile in flower, a source of twin-scale ex plants; (B) Shoot production in the axis of a twin-scale; (c) Multiple shoot production induced by 1mgl·' SA; (D) Bulblet formation on MS medium supplied with activated charcoal 76 Fennell. Crouch and van Sladen

Table 1: Decontamination and regeneration potential of different explants of Crinum variabile

Explant Decontamination(%) Explants producing shoots (%) Benlate pretreatment + NaOCI NaOCI NaOCI (35 mins) (20 mins) (30 mins) Bulb scale + basal plate 67 0 Bulb scale 75 67 0 Twin-scale 32 78 79 66 Lear 100 0

such that their viability is short-lived. In vitro techniques are manes benzyladenine (BA) and naphaleneacetic acid (NAA) thus worth considering for Crinum variabile: one of just two were used in the range 0-2mgJ·•. Bulblet development was species represented in the Cape flora. monitored on the MS medium with or without 5gl' activated A voucher (Crouch 777, NH), collected on the Groen River charcoal. The propagules were subsequently split, longitudi­ to the south of Garies in Namaqualand, has been lodged for nally, in half for use as secondary explants and placed on verification purposes. media containing BA, in the concentration range 0- 1OmgJ ·'. Bulbs were transferred to pots containing a mixture of fine Mature bulblets were dipped in a systemic fungicide river sand (2 parts) and peat (1 part) and grown in the before planting in sterilised soil, containing one part each of greenhouse where they were watered, intermittently, during sand, peat and fine bark chips, and placed in the misthouse the growing season. As the plants had not reached flowering to acclimatise. size, only bulb and leaf material was used as a source of Despite the fact that twin-scales were more difficult to ster­ explants, such that insufficient replicates were available for ilise than the leaves (Table 1), they were the only explants to statistical analysis. The bulbs were washed and then cut regenerate shoots (Table 1 and Figure 1B). High contami­ lengthwise into halves. The upper, tunicate, portion was dis­ nation rates are often observed in cultures initiated from soil­ carded. Both bulbs and the basal parts of emerging leaves borne organs such as bulbs (Hoi and Van Der Linde 1992). were pretreated by rinsing in 1% SporekiW"' (17-30min) and This is because their open structure allows microorganisms 70% ethanol (30 sec-1 min). This was followed by steriliza­ to move in between the scales. In addition, VA mycorrhizae tion in 3.5% NaOCI for either 20 or 30min. To rid plants of are known to infest the outer leaves (Iqbal and Bareen endogenous fungal contaminants, potted plants were also 1986). Their endogenous habit means that surface sterilants watered with a BenlaterM solution and then soaked in 0.2% are ineffective unless used in combination with alcohols or Benlate (15min) prior to sterilisation in NaOCI. systemic fungicides, like Benlate. Even then . there may be The following explants were dissected from sterilised no consistency in the degree of decontamination achieved, material under aseptic conditions: (1) single bulb scales with as was the case for C. variabile. This may relate to differ­ part of the basal plate, (2) single bulb scales with no basal ences in the quality of the parent bulb. plate tissue, (3) twin-scales, consisting of two adjacent Twin-scales are frequently used in the micropropagation scales joined by the basal plate, and (4) basal leaf seg­ of amaryllidaceous species (Van Aartrijk and Van Der Linde ments. Each was placed on an MS basal medium 1986) since meristems already exist where the leaves join (Murashige and Skoog 1962) with 2% sucrose and 0.2% the basal plate (Thompson 1989). These give rise to shoots. Gelriter"' and grown in the light (70.7J.1molm-2s·•; 16 hours Shoot induction occurred spontaneously in C. variabile, that light and 8 hours dark) at a temperature of 25°C. is, without the addition of hormones (Figure 2). For C. To optimise shoot induction from twin-scales, the hor- macowanii, the average number of plantlets that regenerat­ ed in the absence of plant growth regulators compared favourably with the best hormone treatments (Siabbert eta/. - -7-~~ 1993). By cutting into the basal plate, apical dominance is . destroyed. This effectively stimulates the outgrowth of pre­ 100 i .-- ~------~ .-- existing axillary meristems. However, shoot initiation can be .- enhanced by adding cytokinins to the medium (Figure 1C) . - Ji The involvement of cytokinins in apical dominance has been demonstrated in other bulbous species (Hussey 1976). But = I• I i~ 60~j . -· c unlike the lridaceae, liliaceous and amaryllidaceous species '(j so require higher concentrations of BA to promote branching ;;;) "C 40 ..J 0 (Hussey 1976). The combination of auxin and cytokinin, c. 30 ~ however, reduced shoot initiation (Figure 2). This was Ill accompanied by abnormal organogenesis. In other studies, c 20 { -[ 10 BA and NAA, in interaction, were found to adversely affect ~ 0 . 0:~~~~--,------r--' ?/ the numbers of shoots produced (McAlister et a/. 1998), 0 : 0 0 . 1 0 : 2 0.5 : 0 0.5 . 1 0 5 2 sometimes inhibiting plantlet regeneration (Siabbert et a/. 1 NAA: BA concentration (mg 1" ) 1993). The addition of charcoal to the medium resulted in a greater number (92%) of explants forming bulblets com­ pared to the control (41 %). The bulb lets were, on average, Figure 2: Interactive effect of NAA and BA on in vitro shoot induc­ tion in Crinum variabile 3.9mm in diameter (Figure 1D) , whereas those grown on South African Journal of Botany 2001 . 67. 74-77 77

Table 2: Effect of BA on shoot production from halved bulblets of chloroplast-DNA and distribution within the genus Crinurn Crinum variabile (Amaryllidaceae). Journal of Biogeography 20: 55-61 Hoi GM. Van Der Linde PCG (1992) Reduction or contamination in BA Concentration (mgl ') bulb-explant cultures of NarCissus by a hot-water treatment of 0 0.1 1 10 parent bulbs. Plant C ell. Tissue and Organ Culture 31 : 75-79 Hussey G (1976) In vitro release of axillary shoots from apical dom­ Explants producing shoots (%) 66 0 75 50 inance in monoco tyledonous plantlets. Annals of Botany 40: 1323-1325 charcoal-free media, were smaller at 2.3mm. Charcoal is Hussey G (1980) Propagation of some members or the Lil1aceae, routinely used in the propaga tion of some bulbs as it lridaceae and Amaryllidaceae by tissue culture. In: Brickell CD, reduces browning and the decay of twin-scale bases Cutler OF. Gregory M (eds) Petaloid monocotyledons: horticultur­ (Steinitz and Yahel 1982) and strongly influences bulblet al and botanical research. Academic Press. London. pp 33-42. regeneration and size (e.g. Narcissus tazetta L.; ISBN 0-12-133950-5 Amaryllidaceae) (Steinitz and Yahel 1982, Peck and Iqbal SH. Bareen F (1986) Mycorrhizae of the Liliflorae: I. Vesicular­ Cumming 1986). Its ability to regulate these processes is arbuscular mycorrhizal infections in outer dried sheathing leaves thought to be linked to the adsorption of substances likely to from bulbs of some ornamental lilies. Biologia - Pak. 32: 323-328 antagonise cytokinin activity or even the regulation of Jansen Van Vuuren PJ, Coetzee J, Coertse A (1993) South African flowering plants with a potential as future floriculture crops. Acta cytokinin activity itself (Takayama and Misawa 1980). The Horticullurae 337: 65-71 development of bulblets as propagules has several benefits, Jacquin NJ (1804) Plan/arum Rariorum Hortl Caesarei including: the elimination of a rooting stage and the need for Schoenbrunnensis. Volume 4. CF Wappler, Vienna. p 429 hardening-off (Ziv 1997). They can also be used as a source Lehmiller DJ (1 987) A South African Crinum safari. Herbertia 43: 50- of secondary explants by splitting them longitudinally in half. 59 Bulblets of C. variabile which were divided in this manner Lehmiller DJ (1996) C ultivation of African Crinum in pots and tubs. produced new shoots even without the addition of BA (Table Herbertia 51 : 33-37 2). At concentrations of 1 mgl ', however, shoot regeneration Manning J, Goldblatt P (1997) Nieuwoudtville, Bokkeveld Plateau was enhanced. Greater amounts of BA reduced shooting. and Hanlam. South African Wildflower Guide 9. Botanical Society Cytokinins like BA. thus have a stimulatory effect on shoot of South Africa. Cape Town, p 204. ISBN 1-874999-17-1 McAlister BG. Jager AK, Van Staden J (1998) Micropropagallon of regeneration from both twin-scales and bulblets that devel­ Babiana spp. South Afncan Journal of Bo tany 64: 88-90 op in vitro. Similar effects have been noted in Agapanthus. Murashige T, Skoog F ( 1962) A revised medium for rapid growth and Alstroemeria, Freesia (Hussey 1980), Gladiolus (Dantu and bioassays with tobacco tissue cultures. Physiologia Plantarum Bhojwani 1995) and Nerine cultivars (Custers and 15: 473-479 Bergervoet 1992). Peck DE, Cumming BG (1986) Beneficial effects of activated char­ Bulblets of C. variabile transferred to the soil showed no coal on bulblet produc tion in tissue cultures of Muscari armeni­ dormancy, with 100% survival and leaf elongation during the acum. Plant Cell, Tissue and Organ Culture 6: 9-14 period of acclimatization in the misthouse. Aspects of the pot Slabber! MM, De Bruyn MH. Ferreira 01, Pretorius J (1993) cultivation requirements of this taxon have been dealt with Regeneration of bulb le ts from twin-scales of Crinurn macowanii in by Lehmiller (1996). The benefits of the micropropagation vitro. Plant Cell, Tissue and Organ Culture 33: 133-141 Steinitz B. Yahel H (1 982) In vitro propagation of Narcissus tazetta. scheme, outlined above. include the potential for large scale HortScience 17' 333-334 production of a new horticultural subject. For example, a sin­ S tirton CH (1980) Asp ects of research on South AFrican petaloid gle bulb (measuring 4cm in diameter) may yield approxi­ monocotyledons of h orticultural importance. In: Brickell CD, mately 23 plantlets in 17 weeks, or double this number in 6 Cutler DF, Gregory M (eds) Petaloid monocotyledons: horticullur­ months. al and botanical research. Academic Press, London, pp 191- 197. ISBN 0-12-133950-5. Acknowledgements -The NRF and University of Natal Research Takayama S, Misawa M (1980) Differentiation in Lilium bulbscales Fund are thanked for financial support. The Mazda Wildlife Fund grown in vitro. Effects o f activated charcoal, physiological age or generously supports the Ethnobotany Programme of the NBI Dr bulbs and sucrose con centration on differentiation and scale leaf John Manning kindly allowed the use of his photograph of C. vari­ formation in vitro. Physiologia Plantarum 48: 121-125 abile in habitat. Thompson P (1989) C reative propagation: a grower=s guide. Christopher Helm, London. ISBN 0-7470-3213·0 References Van Aartrijk J, Van Der Linde PCG (1 986) In vitro propagation of flower bulb crops. In: Zimmerman RH. Hammerschlag FA, Bryan JE (1 989) Bulbs. Volume II. A-H. Timber Press. Portland. Lawson RH (eds) Tissue culture as a plant production system for ISBN 07470-0231-2 horticultural crops. M artinus Nijhoff Publishers, Dordrecht, pp Custers JBM. Bergervoet JHW (1992) Differences between Nerine 317-331 . ISBN 90-24 7 -3378-2 hybrids in micropropagation potential. Scienlia Horticulturae 52: Verdoorn IC (1964) Crinum variabi/e. Flowering Plants 36: 1. 1433 247-256 Verdoorn IC (1973) The genus Crinum in southern Africa. Bothalia Dantu PK, Bhojwani SS (1 995) In vitro corm formation and field 11 : 27-52 evalua tions of corm-derived plants or Gladiolus. Scientla Ziv M (1 997) The contribution of biotechnology to breeding, propa­ Horticulturae 61 : 11 5-129 gation and disease resistance in geophytes. Acta Horticullurae Fangan BM. Nordall (1993) A comparative analysis of morphology, 430: 247-258

Edited by TH Thomas