Responses to Dehydration and Conservation of the Non-Orthodox Seeds of Warburgia Salutaris

Responses to dehydration and conservation of the non-orthodox seeds of Warburgia salutaris

JI Kioko*, P Berjak and NW Pammenter

School of Life and Environmental Sciences, University of Natal, Durban 4041, South Africa * Corresponding author, e-mail: [email protected]

Received 23 May 2003, accepted in revised form 18 July 2003

Seeds from naturally-ripened fruits of Warburgia salu- iccation and cryopreservation presently reported for taris, for which viability retention on slow air-drying has mature seeds. However, the desiccated mature seeds not been achieved, survived relatively rapid dehydration did not survive for more than a few weeks when stored in silica gel to water contents at which they could be at 16°C or 25°C, and lost viability far more rapidly at 6 ± cryopreserved by immersion in liquid nitrogen. 2°C, revealing their chilling-sensitivity. Seedlings were Subcellular events during natural seed maturation were readily established from seeds retrieved from cryostor- attenuated when premature fruits were harvested and age and germinated in bottom-heated sand-beds at stored, perhaps explaining improved responses to des- 25°C.

Introduction

Warburgia salutaris (Bertol. f.) Chiov. (syn. W. ugandensis; rooted cuttings (from a re-introduction project in South W. breyeri), the only African genus of the Canellaceae, is Africa), and their establishment in the home gardens of local unarguably among the most endangered plant species in farmers (Cunningham 2000). South Africa for two main reasons: the lack of seedling Faced with the impending extinction of W. salutaris in recruitment, and the non-sustainable level of bark stripping South Africa, which meant the irretrievable loss of its sought- which has all but exterminated mature trees in the wild. Lack after resources, a sustainable alternative was envisaged by of seedling recruitment stems from the fact that developing the production of plants via cuttings from as broad a geo- fruits are heavily predated by an (unidentified) insect larva, graphic spread of the limited number of mother plants with- so that mature seeds can be obtained only following special in the country, as was possible (Cunningham 2000). While precautions, as described below. Also, in the unusual event this project was successful, it could not satisfy ex situ con- of mature seeds being formed without human intervention, servation ideals for immediately-available planting stock rep- these lose viability after being extracted from the fruit as resenting a broad genetic base, as is facilitated by seed their originally high water content gradually equilibrates with banking. This situation is exacerbated by the ever-present ambient relative humidity (Kioko et al. 1993, Mbuya et al. problem that mature seeds would still not be available, 1994, personal observations). because of the unsolved predation problem and that, even if The bark of W. salutaris contains a variety of sesquiter- seeds could be nurtured to maturity by human intervention, penoid dialdehydes (Warthen et al. 1983, Taniguchi and apparently they could not be stored. Kubo 1993), with collective anti-feedant and molluscicidal However, in a seed-screening project jointly sponsored by activity (Hutchings 1996, Clark and Appleton 1997) and the IPGRI (International Plant Genetics Resources Institute, broad anti-fungal efficacy (Taniguchi and Kubo 1993). Bark Rome) and DANIDA (The Danish Agency for Development preparations are therefore used for a wide variety of health Assistance), it emerged that, if dried relatively rapidly, the problems (Watt and Breyer-Brandwijk 1962, Hutchings and water content of the seeds of W. salutaris of Kenyan prove- Van Staden 1994, Hutchings 1996), which is the basis of the nance could be reduced to surprisingly low levels, without demand for the product in traditional medicinal practice. This immediate loss of viability (Pritchard and Daws 1997). This has led to the over-exploitation of W. salutaris, and its was followed by work showing that 30% of desiccated seeds endangered status, despite the species being protected by (from the same provenance) survived after cryopreservation law in South Africa (Scott-Shaw 1999). The status of W. in liquid nitrogen (Kioko et al. 1999, 2000). salutaris in Zimbabwe was recorded as ‘locally extinct’ as a While cryopreservation slows metabolism in the explants result of its over-exploitation, although some amelioration of to unmeasurable levels (Kartha 1985) and is reported to the situation has been brought about by the importation of maintain genetic fidelity (e.g. Fretz and Lörz 1994, South African Journal of Botany 2003, 69: 532–539 533

Yoshimatsu et al. 1996, Schäfer-Menuhr et al. 1997) some conventionally-used surface sterilants, NaOCl and HgCl2. studies have shown that genetic aberrations may be associ- While axes survived immersion in alcohol as a surface ster- ated with cryogenic storage (e.g. Ohsako et al. 1997, Aronen ilant, this treatment had no effect on the associated micro- et al. 1999, Harding et al. 2000, Perazzo et al. 2001). In the organisms. Hence, in the present study, surface-sterilisation case of W. salutaris, our preliminary results have indicated was necessarily omitted and extracted seeds were cleaned that neither dehydration nor cryopreservation appears to of fruit pulp only by gentle rubbing with sterile paper towel. affect genomic integrity (Kioko et al. 2003). Storage: Seeds were stored either fully hydrated (wet stor- The present investigation, therefore, set out to: age) or dehydrated. For wet storage, cleaned seeds were • ascertain whether successful seed dehydration could be placed in a monolayer on plastic mesh that had been previ- repeatedly achieved, and if not, the nature of the limiting ously sterilised by soaking in 1% NaOCl and then rinsed thor- parameters; oughly with sterile water. The mesh was supported 200mm • ascertain whether or not the dehydrated seeds could be over wet paper towel in closed white plastic 5-litre buckets, stored, and if so, for what duration and under what condi- all components having been sterilised before use. The buck- tions; ets were maintained at 6°C, 16°C or 25 ± 2°C. • improve survival (including seedling establishment) after For dry storage, seeds that had been dehydrated to a cryopreservation of the seeds water content of about 0.1g g–1 (see below) were enclosed in batches of 30, comprising individual weekly test samples, Materials and Methods in heat-sealed air-tight polythene bags, thus obviating the exposure of the stored seeds to ambient humidity. Bags The material, presently referred to as W. salutaris (T) and W. were then placed in dry, sealed 1-litre plastic buckets and salutaris (SA), was obtained from two geographically widely- stored at 6°C, 16°C or 25 ± 2°C. separated provenances, Lushoto in north-west Tanzania, Dehydration: Whole seeds were buried in activated silica and the Silver Glen Nature Reserve, Durban, South Africa, gel at 25°C and sampled at 6h intervals for up to 6d. respectively. When of full-size, an equal number of fruits of Alternatively, excised embryonic axes were placed on dry fil- W. salutaris (T) were harvested directly from each of ≥30 ter paper in a laminar air flow and sampled at 15min inter- mother trees comprising natural stands.1 These were deliv- vals for 2h. ered by courier to Durban within 48h of harvest. To obtain Water content: These values were determined gravimetri- unpredated material of W. salutaris (SA), after fruit set the cally, separately for embryonic axes and cotyledons from entire canopy of an individual tree was enveloped in shade- individual whole seeds, and for excised axes that had been cloth, which was gathered together and tied round the trunk. rapidly dehydrated (n = 10 in each case), after oven-drying This procedure both reduced the insect predation and pre- at 80°C for 48h, and expressed as g water per g of dry mat- cluded damage to the developing fruits by birds and mon- ter (g g–1). keys. Fruits which had abscised naturally were collected Cryopreservation: Dehydrated whole seeds were from the funnel-like base of the shade-cloth contrivance. enclosed in 1.0ml cryotubes (NUNC®, Daigger, USA) which Upon receipt in Durban, fruits of Tanzanian origin were were rapidly immersed in liquid nitrogen, in which they were divided into three categories on the basis of maturity status kept for at least 1h. Four cryovials containing five seeds as indicated by colour and degree of softness, viz. each were used per sample. Following retrieval from cryos- green/hard; yellow-green/soft; brown/soft. Seeds were torage, the tubes were immediately plunged into a water immediately extracted from all categories, and, in the case bath at 40°C for 1–2min. Seeds were rehydrated at 25°C, of the green/hard fruits, a sub-sample was extracted after concomitant with sowing. storage for 4–6 weeks at 16°C during which the fruits had Assessment of survival: Whole seeds were sown in moist softened. Embryonic axes from seeds of brown/soft fruits sand in bottom-heated beds maintained at 25°C, and were used for electron microscopical studies only, as we watered twice daily. Survival in all cases was scored in terms sought to explain our previous observations (Kioko et al. of shoot emergence. Survival of axes dehydrated after exci- 2003) that a significant diminution of seed quality had sion was assessed by TTZ (2,3,5-triphenyl tetrazolium chlo- occurred by this stage. Fruits of W. salutaris (SA), which ride) staining (ISTA 1999), as the unsterilised material (see were mostly at the yellow/green stage when abscised, were above) could not be cultured in vitro, rapidly being over-run conveyed to the laboratory where the seeds were immedi- by proliferating fungi. ately extracted. Unless otherwise indicated, the results Hardening-off of seedlings and subsequent handling: presently reported are for seeds extracted from mature, yel- Germinated seedlings were transplanted into potting soil, low-green fruits from both provenances. and maintained in the greenhouse for eight weeks under natural lighting at 25°C/18°C day/night temperatures, with Seed processing twice daily mist-spraying. They were then transferred to a shade-house, with 60% light transmittance, watered twice Cleaning: It is usual to surface-sterilise extracted seeds prior weekly, and fertilised with Multifeed® (Plaaskem, South to any other manipulations: however, irrespective of prove- Africa) once every ten days. After six months, when the nance, the seeds — or excised embryonic axes — of W. plants had attained heights of 300mm to 600mm, they were salutaris presently used lost viability after immersion in the transplanted to the field.

1 Courtesy of HP Msanga, National Tree Seed Programme, Morogoro, Tanzania 534 Kioko, Berjak and Pammenter

Processing for transmission electron microscopy (TEM): inadequate. Despite the fact that the W. salutaris seeds Embryonic axes were excised from hydrated seeds extract- were newly extracted from the compact pulp of undamaged, ed from brown/soft fruits, fresh yellow-green/soft and closed fruits, and then stored under otherwise sterile condi- hard/green fruits, as well as from those removed from the tions, their hydrated storage without adequate anti-fungal latter after storage at 16°C for 4 weeks. Apical tips of radi- pre-treatment, was not feasible. This is in keeping with find- cles were fixed in 2.5% phosphate-buffered glutaraldehyde ings that fungal propagules are generally associated periph- containing 0.5% caffeine, post-fixed with 1% osmium tetrox- erally and often internally with non-orthodox seeds of tropi- ide, dehydrated through an acetone series and embedded in cal/sub-tropical provenance (Mycock and Berjak 1990, low-viscosity epoxy resin. After polymerisation, sections Sutherland et al. 2002). Fungi are, however, not invariably were cut using a Reichert Ultracut E microtome, and post- associated with seeds of any one species: in the case of stained on copper grids sequentially with uranyl acetate and those of W. salutaris newly extracted from a batch of fruits of lead citrate. Sections were viewed and photographed using Kenyan provenance, Pritchard and Daws (1997) did not a Jeol 1010 TEM. report problems with fungal proliferation during storage of undried seeds at 16°C over a 7 week period. Results and Discussion Responses to dehydration Hydrated storage Excised embryonic axes: When axes of W. salutaris (T) Water contents of seeds of W. salutaris (T) and (SA) extract- seeds were excised and placed in a laminar airflow, they ed from yellow/green fruits were 2.45 ± 0.23g g–1 and 1.18 ± dried rapidly but lost viability within 60min, during which time 0.28g g–1, respectively. While the disparity might be a func- the water content had reached a mean of 0.15g g–1 (Table tion of the different provenances, significant divergences in 1). This considerable loss of viability at a water content as water content within non-orthodox seeds and seed compo- high as 0.5g g–1 is in marked contrast to viability retention nents of individual species both inter- and intra-seasonally when axes of a variety of species are dehydrated as rapidly are neither unexpected nor surprising (Berjak and as in this study, or are flash dried (e.g. Hevea brasiliensis Pammenter 2001, 2003a, 2003b). Irrespective of the water (Normah et al. 1986), Landolphia kirkii (Pammenter et al. contents of the newly-extracted seeds, none survived in 1991), Quercus rubra (Pritchard 1991), Camellia sinensis hydrated storage as in all cases they became engulfed by (Berjak et al. 1993), Ekebergia capensis (Pammenter et al. proliferating fungi. In the present investigation, therefore, the 1998), Artocarpus heterophyllus (Wesley-Smith et al. thorough manual cleaning of the seed surface emerged as 2001)). The effect of rapid drying of excised axes of W. salutaris was also diametrically different from axis survival when whole seeds were dehydrated in silica gel (see below). It was therefore concluded that the axes of W. salutaris are Table 1: The response of excised embryonic axes of W. salutaris (T) to dehydration. Axes were dehydrated in a laminar air flow, and lethally affected by excision damage, probably exacerbated the viability determined by tetrazolium staining. n = 20 by dehydration. In tandem with the inability of these axes to withstand surface-sterilisation, and the revelation of the fungal status of the seeds, no further attempts were made to Dehydration time (min) Water content g g–1 ± s.d. % viability manipulate excised axes for cryopreservation. 0 2.45 ± 0.23 100 Dehydration of intact seeds: Irrespective of provenance, 15 0.54 ± 0.15 60 whole seeds retained viability to remarkably low water con- 30 0.20 ± 0.04 55 60 0.15 ± 0.03 0 tents as a consequence of burial in activated silica gel 90 0.15 ± 0.02 0 (Tables 2 and 3). For both W. salutaris (T) and (SA) seeds,

Table 2: Axis water content and seed germination after dehydration alone, and dehydration followed by cryopreservation, of W. salutaris (T) seeds extracted from yellow-green fruits. Seeds were dehydrated by burial in activated silica gel. n = 20

% Germination Drying time (h) Axis water content g g-1 ± s.d. Dehydration only Dehydration + Cryopreservation 0 2.45 ± 0.23 100 0 6 1.08 ± 0.14 100 0 10 0.88 ± 0.09 80 0 18 0.44 ± 0.08 90 0 24 0.59 ± 0.16 90 0 48 0.26 ± 0.07 100 0 72 0.12 ± 0.04 90 30 96 0.09 ± 0.04 100 55 108 0.09 ± 0.05 100 40 120 0.08 ± 0.03 95 65 132 0.08 ± 0.01 85 35 144 0.08 ± 0.02 85 45 South African Journal of Botany 2003, 69: 532–539 535

Table 3: Axis water content and seed germination after dehydration alone, and dehydration followed by cryopreservation, of W. salutaris (SA) seeds extracted from yellow-green fruits. Seeds were dehydrated by burial in activated silica gel. n = 20

% Germination Drying time (h) Axis water content g g-1± s.d. Dehydration only Dehydration + Cryopreservation 0 1.18 ± 0.28 90 0 6 0.53 ± 0.11 100 0 10 0.43 ± 0.14 80 0 18 0.32 ± 0.11 90 0 24 0.31 ± 0.08 90 0 48 0.23 ± 0.08 100 0 72 0.15 ± 0.04 90 30 96 0.10 ± 0.04 100 35 108 0.10 ± 0.04 80 40 120 0.09 ± 0.01 80 15 144 0.08 ± 0.01 80 15

80 80

60 60

40 40 GERMINATION (%) GERMINATION (%) 20 20

1 2 3 4 5 6 7 8 9 10 11 12 4942352821147 STORAGE PERIOD (days) WEEKS AFTER SOWING

Figure 1: Germination time course of W. salutaris (T) seeds sown Figure 2: The change in final percentage germination over time, in without drying (), or dehydrated in silica gel for 6h (), 48h (), seeds of W. salutaris (T) dried to 0.1g g–1 and then stored at 6 ± 2°C 72h (), or 144h (x) (), 16°C () or 25°C ()

100% survived dehydration to 0.1g g–1, attained in 96h, 1). Similar stimulatory effects of shorter-term dehydration although in both cases, viability declined with very little addi- and their reversal after drying in the longer-term, have been tional water loss when this drying period was much exceed- reported for the recalcitrant seeds of Ekebergia capensis ed. These results are in keeping with those reported by (Pammenter et al. 1998) and have been observed for a wide Pritchard and Daws (1997), thus demonstrating the repro- range of non-orthodox seeds and excised axes (our unpub- ducibility of the observation that intact seeds of W. salutaris lished observations). While the basis for the stimulation is will withstand relatively rapid dehydration, in contrast to their not yet known, the phenomenon is presently being investi- losing viability concomitant with slow air-drying to equilibri- gated in our laboratory. um with ambient RH (Kioko et al. 1993, Mbuya et al. 1994, our unpublished observations). Seed storage in the dehydrated condition Dehydration of intact seeds for shorter intervals, stimulat- ed germination rate, with this effect being optimal after 48h The above results, reporting viability retention to low water burial in silica gel (Figure 1). This effect was observed both contents as a result of exposure of W. salutaris seeds to sil- in the shorter lag period before shoot emergence, as well as ica-gel-mediated dehydration, were obtained when germina- by the timing of the attainment of full germination of the seed tion assays followed immediately. In contrast, when the samples. The latter was attained within 6 weeks for seeds dehydrated seeds were stored, they maintained high viabili- subjected to 48h of silica-gel-mediated dehydration — at ty for 28d at 16°C and 25°C. The viability of the seed batch which stage only 10% of the non-dehydrated seeds that had at 6 ± 2°C declined to 10% after 28d storage: full ger- been sown immediately following extraction from the fruits, minability was retained for only 7d (Figure 2), indicating the had germinated. For seeds subjected to 144h of dehydra- chilling sensitivity of W. salutaris seeds, at least in the des- tion, however, the lag period again increased before the first iccated condition. Because of the fungal problem, it was not seeds germinated and germination totality was lower (Figure possible to ascertain whether or not the hydrated seeds 536 Kioko, Berjak and Pammenter

Figure 3: The ultrastructural situation in embryonic axes of W. salutaris seeds immediately after dehydration to a water content of ~0.1g g–1 (a) and then stored for 35 days at 6°C (b); stored for 42 days at 16°C (c) and 25°C (d). While the cells in non-stored seeds dehydrated to a water content 0.1g g–1 contained many discrete lipid bodies (a), these became fused into an amorphous mass, most severely after storage at 6°C (b). Even though axis cells of seeds stored at 16°C and 25°C generally had less lipid coalescence (c and d), extensive cellular degen- eration was evident. (e) Cells from axes of seeds from green/hard fruits. The cells had relatively large vacuoles devoid of obvious accumula- tions of insoluble material, a scattering of lipid bodies and clearly-defined organelles including Golgi bodies, mitochondria, plastids and pro- files of rough endoplasmic reticulum — a situation indicative of an actively metabolic state. (f) Axis cells of seeds from fruits harvested at the yellow-green stage of maturity were dominated by masses of lipid bodies, among which the cytomatrix containing dedifferentiated organelles and compressed vacuoles, can be seen to be compacted. The cells had a higher lipid body:organelle ratio than those from seeds ripened in storage (g), probably facilitating improved tolerance to cryopreservation. (h) The ultrastructure of cells from axes of seeds extracted from brown/soft fruits had few recognisable organelles, and showed evidence of deterioration including some coalesced lipid bodies (asterisk), probably the consequence of fungal activity, cytomatrical clearing and inwards plasmalemma withdrawal. G, Golgi body; L, lipid; m, mito- chondrion; v, vacuole; w, cell wall. Scale Bar = 2µm South African Journal of Botany 2003, 69: 532–539 537 were similarly chilling sensitive. the South African seeds was at a water content of 0.1 ± While the chilling sensitivity of W. salutaris seeds in the 0.04g g–1. dehydrated condition suggests their categorisation as an The difference in post-thaw survival of seeds of W. salu- intermediate type, their curtailed lifespan once dehydrated taris (T) and (SA) cannot presently be explained with any implies that desiccation damage sensu stricto (Pammenter certainty, although the necessarily small sample size (n = et al. 1998, Walters et al. 2001) occurs, albeit considerably 20) might partly account for the disparity. However, current more slowly than would take place in highly recalcitrant research in our laboratory indicates that, based on PCR- seeds. Such damage was evident in the ultrastructure of RAPD data, the Tanzanian and South African populations seeds stored at all three temperatures (Figure 3a–d); stor- may represent disparate species. If this is the case, it could age brought about coalescence of lipid bodies, which has account — at least partly — for the disparity in response to been described as a consequence of desiccation damage, immersion and maintenance in liquid nitrogen. With the manifested on rehydration, in lipid-rich desiccation-sensitive often-considerable intra- and inter-seasonal variability in seeds (Leprince et al. 1998). The findings of Daws et al. seed properties within single species (Berjak and (2003) that W. salutaris seeds (of Kenyan origin) at a water Pammenter 2001, 2003a, 2003b), considerably greater dif- content of ~0.2g g–1 survived storage at –20°C for three ferences may be expected among different species. It has months (at which time the water content was 0.18g g–1), but been our experience that responses to the harsh manipula- had lost viability after 12 months, might also be interpreted tions imposed by cryopreservation protocols, demand fine- as the slow, but inexorable effects of desiccation damage tuning of procedures on a species-basis, which may indeed sensu stricto. However, as suggested by those authors, at be a requirement for W. salutaris (T) vs (SA). the initial water content, ~0.2g g–1, it is also possible that ice Besides the provenance, another factor suggested to nucleation and slow crystal growth might have occurred, as influence the amenability of the seeds of W. salutaris to cryo- a small proportion of this water might have been freezable preservation significantly, is the developmental status at the (Vertucci 1990). time of harvest/cryopreservation. Post-cryopreservation sur- Furthermore, the prominent component of lipid bodies in vival of W. salutaris (T) seeds was significantly improved, at axis cells (Figure 3f), may result in an underestimation of the 65%, relative to the 30% obtained previously with seeds of water activity as indicated by the measured water content Kenyan provenance (Kioko et al. 1999, 2000). This is pri- (Sacandé and Hoekstra 1999). Therefore, at a measured marily ascribed to the stage of development of the fruits, and water content in the range of 0.1g g–1 the seeds stored in this hence the seeds, selected in each case, and secondarily, to study may have had a water activity equivalent to that of the different germination protocols used. ‘intermediate water contents’ (Vertucci and Farrant 1995), at The fruits used by Kioko et al. (1999, 2000) were harvest- which injurious, out-of-phase oxidative processes occur. ed in the green/hard condition and stored at 16°C for up to Such processes may be temperature-dependent, as shown 4 weeks before the seeds could be extracted without injury. by Ntuli et al. (1997) and may come into play sooner at 25°C The seeds in that case were not fully mature when the fruits than at 16°C. While seeds stored at both temperatures had were harvested. There is no guarantee that seed maturation deteriorated after 42 days, the degree of degradation was during storage could be equated in all ways with that occur- more extreme in those maintained at 25°C (Figures 3c and ring in seeds completing development in fruits which ripened d). At 6°C the seeds are suggested to have suffered exten- on the parent tree, as was the case for the seeds used in the sive chilling damage during storage, as do non-orthodox present study. seeds of certain other species of tropical origin (Tompsett Ultrastructurally, there were striking differences between 1984, Hong and Ellis 1998). newly-extracted seeds from green/hard, and those from yel- Thus, in terms of their tolerance of relatively rapid dehy- low-green/soft fruits (Figures 3e and 3f, respectively): there dration to water contents ≤0.1g g–1, seeds of W. salutaris was a considerably greater lipid-body:organelle volume ratio cannot be described as being typically recalcitrant; however, in axis cells from the yellow-green/soft fruits. This was considering the limited period (~40d and 49d) within which accompanied by cytomatrical compaction, vacuole shrink- viability declined to 50% and zero, respectively, during storage and organelle dedifferentiation (Figure 3f), and affords age at either 16°C or 25°C, they are certainly non-orthodox. direct evidence of a marked accumulation of lipid reserves and indirectly, of a decline in metabolic activity occurring dur- Cryostorage ing the fruit ripening phase. In comparison, axes from seeds extracted after 4 weeks storage of prematurely-harvested Although the seeds of W. salutaris (T) and (SA) remained fruits had amassed considerably less lipid, with a concomi- 100% viable after desiccation to water contents of 0.26 ± tantly lower degree of cytomatrical compaction and 0.07g g–1 and 0.23 ± 0.08g g–1 , respectively, none survived organelle de-differentiation (Figure 3g). The lower survival storage in liquid nitrogen (Tables 2 and 3), arguing for the after cryostorage of W. salutaris seeds reported by Kioko et presence of freezable water and formation of lethal intracel- al. (1999) is suggested to have been at least partly a conse- lular ice crystals at the relatively slow cooling rate achieved quence of this subcellular situation. Axis intracellular varia- when material is enclosed in cryotubes (Wesley-Smith et al. tions between naturally-matured seeds and those from 1992, Berjak et al. 1999a, 1999b). In contrast, after cryos- unripened fruits stored until soft, reflect differences in physi- torage, 65% of the seeds of Tanzanian origin desiccated to ological status, which might well underlie the differing 0.08 ± 0.03g g–1 retained viability as indicated by seedling responses of the seeds, not only to cryopreservation, but establishment, and comparably, the best survival (40%) of also to vigour and viability retention after dehydration 538 Kioko, Berjak and Pammenter

(Pammenter and Berjak 1999). The ultrastructure of axis Berjak P, Pammenter NW (2001) Seed recalcitrance — current per- cells from seeds extracted from brown/soft fruits showed spectives. South African Journal of Botany 67: 79–89 that various deteriorative events had occurred, including evi- Berjak P, Pammenter NW (2003a) Understanding and handling des- dence of fusion between lipid bodies, areas of cytomatrical iccation-sensitive seeds. In: Smith RD, Dickie JB, Linington SH, degradation, a paucity of identifiable organelles and inwards Pritchard HW, Probert RJ (eds) Seed Conservation: Turning Science into Practice. Royal Botanic Gardens, Kew. (in press) withdrawal of the plasmalemma in places (Figure 3h). These Berjak P, Pammenter NW (2003b) Recalcitrant seeds. In: Benech- degradative changes are suggested to be primarily the Arnold RL, Sánchez RA (eds) Seed Physiology: Applications to result of depredation of the embryo cells by enhanced fun- Agriculture. New York, Haworth Press Inc. (in press). gal activity encouraged by the over-ripe condition of the fruit Berjak P, Vertucci CW, Pammenter NW (1993) Effects of develop- pulp. While inherent deterioration of the seeds having mental status and dehydration rate on characteristics of water accompanied fruit browning cannot be precluded, it is sug- and desiccation-sensitivity in recalcitrant seeds of Camellia gested that the present microscopical observations explain sinensis. Seed Science Research 3: 155–166 the underlying cause of the poor quality of such seeds pre- Berjak P, Walker M, Watt MP, Mycock DJ (1999a) Experimental viously reported (Kioko et al. 2003). parameters underlying failure or success in plant germplasm cry- The results suggest that the stage of seed development is opreservation: a case study on zygotic axes of Quercus robur L. Cryo-Letters 20: 251–262 probably a critical variable that will determine the degree of Berjak P, Kioko JI, Walker M, Mycock DJ, Wesley-Smith J, success of cryopreservation procedures for Warburgia Pammenter, NW (1999b) Cryopreservation — an elusive goal? sp./spp. particularly, and probably for non-orthodox plant In: Marzalina M, Khoo KC, Jayanthi N, Tsan FY, Krishnapillay B germplasm generally. A further, although indirect factor to be (eds) Recalcitrant Seeds. FRIM, Kuala Lumpur, pp 96–109. ISBN considered, however, is inherent in the germination proce- 983–2181–01–1 dure. Clark TE, Appleton CC (1997) The molluscicidal activity of Apodytes dimidiata E. Meyer ex Arn. (Icacinaceae), Gardenia thunbergia Seed germination and seedling establishment L.f. (Rubiceae) and Warburgia salutaris (Bertol. F.) Chiov. (Cannelaceae[sic]), three South African plants. Journal of In the earlier studies on W. salutaris by Kioko et al. (1999, Ethnopharmacology 56: 15–30 Cunningham AB (2000) People and plants online. Available at: 2000) seeds were set to germinate in vermiculite in seedling http://www.rbgkew.org.uk/people[lants/wp/wp1/intro.htm trays at 25°C, but presently bottom-heated sand beds also Daws MI, Omondi W, Pritchard HW, Berjak P (2003) Some ecolog- at 25°C were used. While use of vermiculite was adequate ical and conservation aspects of Warburgia salutaris seed biolo- for vigorous seeds, fungi emanating from individuals that gy. In: Smith RD, Dickie JB, Linington SH, Pritchard, HW, Probert were slower to germinate tended to proliferate on this water- RJ (eds) Seed Conservation: Turning Science into Practice. retaining medium, prompting the shift to the sand beds. This Royal Botanic Gardens, Kew. (in press) proved to be worthwhile, the well-draining sandy medium Fretz A, Lörz H (1994) Cryopreservation of in vitro cultures of barley and wider-apart spacing of seeds curtailing the spread of (Hordeum vulgare L. and H. murinum L.) and transgenic cells of fungus from locations of non-viable individuals. The result- wheat (Triticum aestivum L.). Journal of Plant Physiology 146: ant seedlings were maintained a shade-house for six 489–496 Harding K, Marzalina M, Krishnapillay B, Nashataul-Zaimah NA, months, after which they were transplanted in a random Normah MN, Benson EE (2000) Molecular stability assessments block in the field at Eston, KwaZulu-Natal, South Africa of trees regenerated from cryopreserved mahogany (Swietenia (29°52’S, 30°30’E) and are presently being monitored. macrophylla) seed germplasm using non-radioactive techniques Previously-reported preliminary results have indicated that to examine the chromatin structure and DNA methylation status neither manipulation of W. salutaris seeds for cryopreserva- of the ribosomal RNA genes. Journal of Tropical Forest Science tion, or the freezing and thawing processes themselves, 12: 149–163 appeared to have had deleterious consequences for the Hong TD, Ellis RH (1998) Contrasting seed storage behaviour maintenance of genetic fidelity (Kioko et al. 2003). among different species of Meliaceae. Seed Science and Therefore, as long as fruits and seeds can be husbanded to Technology 26: 77–95 maturity in each of the isolated, remaining populations in Hutchings A (1996) Zulu Medicinal Plants. Natal University Press, Pietermaritzburg. ISBN 0–86980–893–1 protected areas in South/southern Africa, cryopreservation Hutchings A, Van Staden J (1994) Plants used for stress-related ail- offers the possibility of building up planting stocks and facil- ments in traditional Zulu, Xhosa and Sotho medicine. Part 1: itating breeding programmes of Warburgia salutaris that will Plants used for headaches. Journal of Ethnopharmacology 43: ensure conservation of the species and increase the diver- 89–124 sity of the genepool. Ultimately, the sought-after products International Seed Testing Association, ISTA (1999) International used in traditional medicine should again be available, as a rules for seed testing, 1999. Seed Science and Technology 27 result of appropriate silvicultural practices and sustainable (Supplement): 201–244 bark harvesting. Kartha KK (1985) Meristem culture and germplasm preservation. In: Kartha KK (ed) Cryopreservation of Plant Cells and Organs. CRC References Press, Boca Raton, pp 115–134. ISBN 0–8493–6102–8 Kioko J, Albrecht J, Uncovsky S (1993) Seed collection and handling. In: Albrecht J (ed) Tree Seed Handbook of Kenya. GTZ Aronen TS, Krajnakova J, Haggman HM, Ryynanen LA (1999) Forestry Seed Centre, Muguga, pp 30–54. ISBN 9966–9897–2–2 Genetic fidelity of cryopreserved embryogenic cultures of open- Kioko J, Berjak P, Pritchard H, Daws M (1999) Studies of post- pollinated Abies cephalonica. Plant Science Limerick 142: shedding behaviour and cryopreservation of seeds of Warburgia 163–172 salutaris, a highly-endangered medicinal plant indigenous to trop- South African Journal of Botany 2003, 69: 532–539 539

ical Africa. In: Marzalina M, Khoo KC, Jayanthi N, Tsan FY, Pritchard HW, Daws M (1997) Warburgia salutaris. The Project on Krishnapillay B (eds) Recalcitrant Seeds. FRIM, Kuala Lumpur, Handling and Storage of Recalcitrant and Intermediate Tropical pp 365–371. ISBN 983–2181–01–1 Forest Seeds Newsletter — October 1997: 9 Kioko J, Berjak P, Pritchard H, Daws M (2000) Seeds of the African Schäfer-Menuhr A, Schumacher HM, Mix-Wagner G, Altman A pepper bark (Warburgia salutaris) can be cryopreserved after (1997) Cryopreservation of potato cultivars — design of a method rapid dehydration in silica gel. In: Engelmann F, Takagi H (eds) for routine application in genebanks. Acta Horticulturae 447: Cryopreservation of Tropical Plant Germplasm. Current Research 477–482 Progress and Application. International Plant Genetic Research Scott-Shaw R (1999) Rare and Threatened Plants of KwaZulu-Natal Institute (IPGRI), Rome, pp 371–377. ISBN 92–9043–428–7 and Neighbouring Regions: A Plant Red Data Book. KwaZulu- Kioko J, Berjak P, Pammenter N, Lamb J, Gumede Z (2003) Is Natal Nature Conservation Service, Pietermaritzburg. ISBN genetic fidelity maintained following the cryopreservation of the 0–620–24688–X seeds of the endangered African Pepper Bark (Warburgia salu- Sacandé M, Hoekstra F (1999) Improving the storage longevity of taris)? In: Nicolas G, Bradford KJ, Côme D, Pritchard HW (eds) intermediate neem (Azadirachta indica) seeds. In: Marzalina M, The Biology of Seeds: Recent Research Advances. CABI, Khoo KC, Jayanthi N, Tsan FY, Krishnapillay B (eds) Recalcitrant Wallingford, pp 327–335 (in press) Seeds. FRIM, Kuala Lumpur, pp 64–73. ISBN 983–2181–01–1 Leprince O, Van Aelst A, Pritchard HW, Murphy DJ (1998) Oleosins Sutherland JR, Diekmann M, Berjak P (2002) Forest Tree Seed prevent oil-body coalescence during seed imbibition as suggest- Health for Germplasm Conservation. International Plant Genetic ed by a low-temperature scanning electron microscope study of Research Institute (IPGRI), Rome, pp 28–41. ISBN desiccation-tolerant and -sensitive oil seeds. Planta 204: 109–119 92–9043–515–1 Mbuya LP, Msanga HP, Ruffo CK, Birnie A, Tegnäs B (1994) Useful Taniguchi M, Kubo I (1993) Ethnobotanical drug discovery based on Trees and Shrubs for Tanzania. Regional Soil Conservation medicine-men’s trials in the African savanna: screening of East Unit/Swedish International Development Cooperation Agency African plants for antimicrobial activity II. Journal of Natural (RSCU/SIDA), Nairobi, pp 510–511. ISBN 9966–896–22–8 Products 56:1539–1546 Mycock DJ, Berjak P (1990) Fungal contaminants associated with Tompsett PB (1984) Desiccation studies in relation to the storage of several homoiohydrous (recalcitrant) seed species. Araucaria seed. Annals of Applied Biology 105: 581–586 Phytophylactica 22: 413–418 Vertucci CW (1990) Calorimetric studies of the state of water in Normah MN, Chin HF, Hor YL (1986) Desiccation and cryostorage seed tissues. Biophysical Journal 58: 1463–1471 of embryonic axes of Hevea brasiliensis Muell.-Arg. Pertanika 9: Vertucci CW, Farrant JM (1995) Acquisition and loss of desiccation 299–303 tolerance. In: Kigel J, Galili G (eds) Seed Development and Ntuli T, Berjak P, Pammenter NW, Smith MT (1997) Effects of tem- Germination. Marcel Dekker Inc, New York, pp 237–271. ISBN perature on the desiccation responses of seeds of Zizania palus- 0–8247–9229–7 tris. Seed Science Research 7: 145–160 Walters C, Pammenter NW, Berjak P, Crane J (2001) Desiccation Ohsako S, Nagano R, Sugimoto Y, Goto K (1997) Comparison of damage, accelerated ageing and respiration in desiccation toler- the nuclear DNA stability against freezing-thawing and high tem- ant and sensitive seeds. Seed Science Research 11: 135–148 perature treatments between spermatozoa and somatic cells. Warthen JD Jr., Waters RM, Flippen-Anderson JL, Gilardi R (1983) Journal of Veterinary Medical Science 59: 1085–1088 Purification of synthetic warburganal intermediates by open col- Pammenter NW, Berjak P (1999) A review of recalcitrant seed phys- umn and high performance liquid chromatography. iology in relation to desiccation-tolerance mechanisms. Seed Chromatographia 17: 623–626 Science Research 9: 3–37 Watt JM, Breyer-Brandwijk MG (1962) The Medicinal and Pammenter NW, Greggains V, Kioko JI, Wesley-Smith J, Berjak P, Poisonous Plants of Southern and Eastern Africa. Livingstone, Finch-Savage WE (1998) Effects of differential drying rates on London. ISBN 0317297619 viability retention of recalcitrant seeds of Ekebergia capensis. Wesley-Smith J, Pammenter NW, Berjak P, Walters C (2001) The Seed Science Research 8: 463–471 effects of two drying rates on the desiccation tolerance of embry- Pammenter NW, Vertucci CW, Berjak P (1991) Homeohydrous onic axes of recalcitrant jackfruit (Artocarpus heterophyllus (recalcitrant) seeds: dehydration, the state of water and viability Lamk.) seeds. Annals of Botany 88: 653–664 characteristics in Landolphia kirkii. Plant Physiology 96: Wesley-Smith J, Vertucci CW, Berjak P, Pammenter NW, Crane J 1093–1098 (1992) Cryopreservation of recalcitrant axes of Camellia sinensis Perazzo G, Panta A, Rodriguez F, Gomez R, Toledo J, Huaman Z, in relation to dehydration, freezing rate and the thermal properties Ghislain M, Golmirzaie AM, Roca W (2001) Clonal true-to-type of tissue water. Journal of Plant Physiology 140: 596–604 verification of potato accessions retrieved from in vitro conserva- Yoshimatsu K, Yamaguchi H, Shimomura K (1996) Traits of Panax tion and cryopreservation. CIP Program Report, 1999–2000. ginseng hairy roots after cold storage and cryopreservation. Plant International Potato Centre, Lima, Peru, pp 175–183 Cell Reports 15: 555–560 Pritchard HW (1991) Water potential and embryonic axis viability in recalcitrant seeds of Quercus rubra. Annals of Botany 67: 43–49

Edited by J van Staden