Adventitious root formation in Anacardium occidentale L. in response to phytohormones and removal of roots

Scientia Horticulturae 111 (2007) 164–172 www.elsevier.com/locate/scihorti

Adventitious root formation in Anacardium occidentale L. in response to phytohormones and removal of roots Juliano Saranga, Ross Cameron * School of Biological Sciences, University of Reading, Whiteknights, Reading RG6 6AS, UK Received 28 February 2006; received in revised form 22 September 2006; accepted 19 October 2006

Abstract Despite advances in tissue culture techniques, propagation by leafy, softwood cuttings is the preferred, practical system for vegetative reproduction of many tree and shrub species. Species are frequently defined as ‘difficult’- or ‘easy-to-root’ when propagated by conventional cuttings. Speed of rooting is often linked with ease of propagation, and slow-to-root species may be ‘difficult’ precisely because tissues deteriorate prior to the formation of adventitious roots. Even when roots form, limited development of these may impair the establishment of a cutting. In this study we used softwood cuttings of cashew (Anacardium occidentale), a species considered as ‘difficult-to-root’. We aimed to test the hypothesis that speed, and extent of early rooting, is critical in determining success with this species; and that the potential to form adventitious roots will decrease with time in the propagation environment. Using two genotypes, initial rooting rates were examined in the presence or absence of exogenous auxin. In cuttings that formed adventitious roots, either entire roots or root tips were removed, to determine if further root formation/ development was feasible. To investigate if subsequent root responses were linked to phytohormone action, a number of cuttings were also treated with either exogenous auxin (indole-3-butyric acid—IBA) or cytokinin (zeatin). Despite the reputation of Anacardium as being ‘difficult-to-root’, we found high rooting rates in two genotypes (AC 10 and CCP 1001). Removing adventitious roots from cuttings and returning them to the propagation environment, resulted in subsequent re-rooting. Indeed, individual cuttings could develop new adventitious roots on four to five separate occasions over a 9 week period. Data showed that rooting potential increased, not decreased with time in the propagation environment and that cutting viability was unaffected. Root expression was faster (8–15 days) after the removal of previous roots compared to when the cuttings were first stuck (21 days). Exposing cuttings to IBA at the time of preparation, improved initial rooting in AC 10, but not in CCP 1001. Application of IBA once roots had formed had little effect on subsequent development, but zeatin reduced root length and promoted root number and dry matter accumulation. These results challenge our hypothesis, and indicate that rooting potential remains high in Anacardium. The precise mechanisms that regulate the number of adventitious roots expressed, remain to be determined. Nevertheless, results indicate that rooting potential can be high in ‘difficult-to-root’ species, and suggest that providing supportive environments is the key to expressing this potential. # 2006 Elsevier B.V. All rights reserved. Keywords: Cashew; Cuttings; Adventitious roots; Injury; Auxin; Cytokinin

1. Introduction Clonal propagation of many tree and shrub species is effective through ‘leafy softwood’ cuttings, however, a number of important genera remain difficult to propagate and establish using this technique. The limited potential to form adventitious roots may in part result from inappropriate time of propagation (via stock plant effects, Cameron et al., 2001, 2003) or the physiological state of tissues (Hackett, 1988; Ermel et al., 2000;

* Corresponding author. Tel.: +44 118 378 8379; fax: +44 118 378 8160. E-mail address: [email protected] (R. Cameron). 0304-4238/$ – see front matter # 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.scienta.2006.10.010

Reineke et al., 2002). Some species may be slow to form adventitious roots and the cutting may fail prior to the formation of functional roots (Rose and Pellett, 1994; Stankova and Panetsos, 1997; Voyiatzi et al., 2002). Indeed, species where roots emerge rapidly are often categorised as ‘easy-toroot’ (e.g. 12 days, Euphorbia pulcherrima, Wilkerson et al., 2005) in contrast to those more ‘difficult-to-root’ ones that require longer periods before emergence (e.g. 35 days, Ilex paraguariensis, Tarrago et al., 2005). There is a perception that leafy ‘softwood’ cuttings need to form adventitious roots quickly, otherwise tissues will become dysfunctional through either prolonged exposure to sub-optimal environments (Howard and Harrison-Murray, 1995; Aiello and Graves,

J. Saranga, R. Cameron / Scientia Horticulturae 111 (2007) 164–172

1998) or pathogen activity associated with such environments (e.g. Littlejohn and Gertse, 2001). In slow-to-root species, reasons for cutting failure are frequently associated with tissue dehydration (Grange and Loach, 1983), loss or an inability to photosynthesis new carbohydrates (Reuveni and Raviv, 1981; Del Rio et al., 1991) and possibly a limited response to exogenous auxins (Aminah, 2003). Providing environmental conditions that both minimise water stress and provide light for photosynthesis appears to be critical, and the use of sub-optimal propagation environments can often explain failure to root (Howard and Harrison-Murray, 1995). For those species where success from conventional cuttings remains elusive or inconsistent, tissue culture, grafting or air-layering may be employed. These themselves, however, do not necessarily guarantee success and even when proven useful can be prohibitive due to; expense, the amount of labour required or low reproduction rates. Improving propagation success through the use of conventional cuttings therefore remains a key objective for nurserymen, farmers and foresters worldwide. The precise determination of ‘rooting success’ is controversial (Wilson and Struve, 2003), with reports suggesting it can be based on percentage of the cutting population that form roots, the numbers of adventitious roots per cutting, the number of total roots per cutting (where the numbers may include secondary and tertiary roots) or even the speed in which cuttings root. In addition, rooting success does not necessarily correlate with propagation success as the number of cuttings that eventually establish may vary considerably from those that initially formed roots (Owen et al., 2001; Griffin and Schroeder, 2004). Indeed, it is possible for a cutting to form only a single adventitious root (and constitute success), promote numerous lateral root branches from this first root, yet, finally fail due to the original root being damaged (e.g. at transplanting, Billingsley, 2003). Therefore, it is evident that the rooting of cuttings is a dynamic event and that relationships need to be established that take account of both adventitious root formation and subsequent cutting development. This research explores the rooting potential in a ‘difficultto-root’ species (Cashew—Anacardium occidentale L.). Due to difficulties in rooting of cuttings (Rao, 1985; Duarte et al., 1992), vegetative propagation of superior clones has relied on techniques such as air-layering, grafting (Damodaran, 1985) and tissue culture (Mneney and Mantell, 2002). Previous research demonstrates, however, that rooting of stem cuttings is possible, although success often correlates with; more elaborate preparation techniques such as etiolation or shoot ringing (Rao et al., 1988), the provision of contact polythene or mist to minimise desiccation (Rao et al., 1990; Sen et al., 1991) and in at least one occasion, the provision of a well-aerated rooting medium (Coester and Ohler, 1976). The fact that tissues form roots readily in vitro, and can do so under certain circumstances in vivo, suggests that failure in Anacardium may relate to a slow root formation process and the loss of cutting viability prior to root emergence. This loss of viability being accelerated under sub-optimal propagation environments. Therefore, in addition to improving propagation of this

165

species from a practical viewpoint, the work aimed to elucidate the relationship between propagation duration/environment and rooting potential (both in terms of number of adventitious roots but also subsequent root development). Indeed, we tested the hypothesis that speed of adventitious rooting is critical to the success of propagation and that the potential to form adventitious roots would decrease with time in the propagation environment. This was evaluated by the repeated removal of any adventitious roots that formed over a 9 week period. Subsequently, we wished to examine the extent to which any new root development was regulated by phytohormone action, via removing root tips (e.g. a possible source of endogenous auxins and cytokinins) or exogenously adding these compounds. 2. Materials and methods 2.1. Plant material and cutting preparation Stockplants of cashew (A. occidentale L.) were raised from seed sown in 1999 and grown on in glasshouses at minimum air temperature of 20 8C and natural photoperiods at the University of Reading, UK. Seed was collected from two selected genotypes, based on agronomic characteristics; CCP 1001 with ‘dwarfing’ habit and non-vigorous growth and precocious flowering/fruiting characteristics and AC 10 with a more vigorous growth habit. Stock material was pruned (removing 50% of growth) and re-potted on an annual basis to promote new shoot growth suitable for the selection of softwood leafy cuttings. Cuttings were harvested when stock material was approximately 4–5 years old. Although stock material was considered chronologically young, flower formation was common and tissues selected for propagation were phenotypically mature. Apical cuttings, approximately 100 mm long by 5–8 mm wide (calliper) and comprising three to four nodes were used throughout. Leaves were retained on the cuttings, with the exception of the basal node where they were removed. Cuttings were selected from stock plants and within each genotype randomised to avoid any bias associated with individual mother-plants. Depending on treatment, cuttings would be dipped in 6.2 mM indole-3-butyric acid (IBA), 0.03 mM zeatin (Z) in aqueous acetone solution, or a combination of both for 5 s. Chemicals were dissolved in 10 ml acetone and made up to the appropriate concentration by adding distilled water. Control cuttings were dipped in a 5% (v/v) acetone aqueous solution. Choice of IBA and zeatin concentrations reflected positive responses on root generation in preliminary experiments. Cuttings were then inserted two per 90 mm diameter pots containing perlite (0.5–2.0 mm, grade) and placed into a polythene enclosed mist system (to maintain both high humidity and leaf cooling, Harrison-Murray and Howard, 1992) for periods up to 63 days. Mist application was controlled via an electronic leaf (Mist Irrigation System Controls, Ringwood, UK) and relative humidity maintained at >97% (temperature and humidity were monitored by data-loggers, TGX-3580, Gemini Data Loggers, Chichester, UK). The mist

166

J. Saranga, R. Cameron / Scientia Horticulturae 111 (2007) 164–172

system was maintained within a glasshouse where air temperatures were regulated either through heat application (at 15 8C) or venting (at 25 8C). Irradiance within the propagation environment was monitored once daily between 12:00 and 14:00 h (SKE 500/512/140G, Skye Instruments Ltd., Llandrindod Wells, UK). The use of perlite enabled cuttings to be removed from the pots, inspected and carefully re-planted with minimal disturbance to the root system. During intermediate inspections cuttings were spraying with water at approximately 3 min intervals. 2.2. Data collection and statistical analyses Data were recorded for percentage of cuttings rooting, percentage survival, number of roots per rooted cutting (where appropriate divided into adventitious [primary], secondary and tertiary roots), growth (increments in stem length, diameter and number of leaves) and meristematic activity in the shoot tip (i.e. whether a resting bud was present or not). Cuttings were divided into blocks based on position in the propagation environment and treatments randomised within the blocks. Individual cuttings were used as replicated plots. Where data sets were binomial (e.g. rooting) treatment effects were analysed by logistic regression analysis and data presented as mean values with predicted standard errors (S.E.) using analysis of deviance. Otherwise, analysis of variance was employed to determine significance of treatment effects and calculate least significant differences (L.S.D.) at P = 0.05. 2.3. Experiment 1—effects of initial IBA application and subsequent repeated root removal on cutting development during summer 2003 Cuttings of CCP 1001 and AC10 were collected on 17 June 2003, and bases treated with distilled water (Con) or 6.2 mM IBA (IBA), before placement in enclosed mist. After 21 days (8 July) root formation was evident and cuttings were further subdivided into groups where all the roots were either removed (RootRem) or left intact, before being carefully re-planted. Root removal was repeated again at weekly intervals for up to 63 days after the original collection of the cuttings. To avoid any bias relating to root disturbance rather than root removal, cuttings with roots left intact were also de-potted and re-planted on these occasions. No additional IBA was applied before the re-planting of cuttings. After 63 days (19 August) the number of episodes that root generation or re-generation had been observed on an individual cutting (frequency) was recorded, as were final root numbers and the total number of roots each cutting had initiated (i.e. including the number of roots that had been cumulatively excised over the previous 9 weeks). Cuttings were placed on the propagation bench in a randomised complete block design with 10 cuttings per replication in 3 positional blocks. There were a total of 120 cuttings per genotype and separate statistical analyses were carried out for each genotype.

2.4. Experiment 2—effects of root manipulation and phytohormone treatments on the development of rootedcuttings during late summer/autumn 2004 A second experiment was carried out on 16 August 2004, using cuttings from AC 10. Cuttings were prepared as before but without any phytohormone treatment at the time of propagation. On 16 September, however, once initial rooting had taken place, cuttings were divided randomly into three batches. Treatments were then imposed on these batches of rooted-cuttings, namely adventitious roots were left in place (Con), removed entirely (RootRem) or the root apices alone removed (excision approximately 5–10 mm from the tip) (ApRem). Each batch was further sub-divided into four more groups and root systems (or cutting bases in the case of the RootRem treatment) dipped with distilled water, 6.2 mM IBA (IBA), 0.03 mM zeatin (Z), or 6.2 mM IBA plus 0.03 mM zeatin (IBA + Z). Cuttings were placed back in the propagation environment and assessed on 5 October. The 12 treatment combinations were each represented by 6 cuttings in each of 4 blocks, and treatments randomised within a block. After assessment cuttings were retained but divided into two subsequent sub-experiments. These cuttings were used to determine (i) whether repeated root removal, again influenced adventitious root formation (RootRem cuttings only, but without additional phytohormone treatment); and (ii) the longer term influence of the original phytohormone treatments on establishment and development of the Con and ApRem cuttings. All cuttings were placed back in their blocks within the propagation environment and harvested on 18 October. Data from each sub-experiment were analysed separately. 3. Results 3.1. Experiment 1—effects of initial IBA application and subsequent repeated root removal on cutting development during summer 2003 Temperatures and irradiance levels recorded in the propagation environment fluctuated on a day to day basis, but there were marginal trends upwards for both maximum and minimum temperature over time. In contrast, recorded irradiance levels tended to decrease slightly (Fig. 1). The proportion of cuttings that rooted was relatively high in both genotypes (Table 1), although treatment with IBA appeared to improve rooting percentage in AC 10 (e.g. IBA = 100%; Con = 90%, L.S.D. = 4.62df 101), but have an opposite, but non-significant effect in CCP 1001. Removing adventitious roots resulted in the formation of more adventitious roots on the stem and cuttings could ‘re-root’ on average approximately four to five times over 63 days (Table 1). Indeed, some individual cuttings rooted early on (by day 21), and despite repeated removal of roots, re-formed roots on six separate occasions afterwards. The initial application of IBA had no significant effect on how many times adventitious roots could form. In contrast, in those cuttings where initial roots were left intact, there was little evidence of new adventitious

J. Saranga, R. Cameron / Scientia Horticulturae 111 (2007) 164–172

Fig. 1. Experiment 1. Maximum and minimum daily temperature (8C) and irradiance (mmol m 2 s 1) recorded in the propagation environment during summer 2003. Dates relate to when cuttings were inspected or treated.

roots forming after the initial 21 day period (mean frequency = 1). When analyses were restricted to the RootRem treatments alone, it was evident that the number of new adventitious roots generated generally increased after each episode of root removal (Fig. 2), i.e. the capacity to form new adventitious roots did not decrease with time. As with rooting percentage, the number of new roots formed appeared to be marginally favoured in AC 10 by IBA application (not significant), but inhibited by it in CCP 1001 (significantly different after 42 and 56 days). It was observed that root emergence after excision took place between 8 and 15 days, whereas the original root expression had taken approximately 21 days (Fig. 2). Although RootRem treatments enhanced the total number of adventitious roots formed, greatest numbers of roots of all types (i.e. including secondary and tertiary) was maximised by the non-removal treatments; anywhere between 7 and 21 greater than equivalent RootRem treatments (Table 1). Interestingly, despite IBA improving rooting percentage in AC 10, the total numbers of roots was significantly less in the IBA treatment (358) compared to Con (563) (L.S.D. = 159df 101) by day 63. At final harvest, root lengths were significantly shorter in the RootRem treatment (data not shown). In those cuttings of

167

Fig. 2. Experiment 1. Effect of initial IBA application and subsequent repeated root removal (RootRem) on the mean number of new adventitious roots formed in A. occidentale, CCP 1001 and AC 10 cuttings during summer 2003. Cuttings were inspected after 7, 21, 28, 35, 42, 49, 56 and 63 days, and any roots present removed on each occasion.

CCP 1001 where roots were retained, initial IBA application significantly reduced final root length (Con = 126 mm; IBA = 102 mm, L.S.D. = 20.0df 84). RootRem treatments reduced apical shoot activity in cuttings by the time of the final harvest and both shoot extension and increase in leaf number tended to be reduced in these treatments (Table 2). Radial growth (stem diameter increment) was relatively poor in the RootRem treatment in AC 10, with values being significantly less than the non-treated cuttings (Con). In CCP 1001, treatments corresponding to auxin application resulted in reduced stem dry weight, but this was not the case for AC 10 (Table 2). 3.2. Experiment 2—effects of root manipulation and phytohormone treatments on the development of rootedcuttings during late summer/autumn 2004 Temperatures and irradiance levels tended to decline slowly from early September until the end of the experiment, although temperatures below 16 8C were rarely recorded (Fig. 3). Cuttings of AC 10 rooted readily, with a 96% rooting rate

Table 1 Experiment 1: effects of initial IBA application and subsequent repeated root removal (RootRem) on rooting parameters of A. occidentale, CCP 1001 and AC 10 cuttings, as recorded over a 63 day period in summer 2003 Treatment

% rooted (S.E.)

Mean rooting episodes (frequency)

Total no. of adventitious roots (sum over 63 days)

Roots per cutting at final harvest

CCP 1001 Con IBA RootRem IBA + RootRem L.S.D.df 90

90 (5.3) 73 (7.5) 90 (5.3) 80 (6.9) NA

1.0 1.0 5.0 4.4 0.75

7.7 8.8 36.7 22.3 9.4

436 450 31 21 141

AC 10 Con IBA RootRem IBA + RootRem L.S.D.df 101

90 (1.6) 100 (1.7) 80 (1.5) 97 (1.7) NA

1.0 1.0 5.4 5.1 0.54

10.0 7.7 41.2 48.0 12.2

563 358 31 45 159

168

J. Saranga, R. Cameron / Scientia Horticulturae 111 (2007) 164–172

Table 2 Experiment1: effects of initial IBA application and subsequent repeated root removal (RootRem) on shoot parameters of A. occidentale, CCP 1001 and AC 10 cuttings, as recorded over a 63 day period in summer 2003 Treatment

% active shoot apices (S.E.)

Shoot growth (mm)

Diameter increment (mm)

CCP 1001 Con IBA RootRem IBA + RootRem L.S.D.df 90

52 (13.0) 63 (14.8) 20 (8.0) 19 (8.5) NA

23 28 17 17 12.5

0.8 0.7 1.1 0.7 0.47

2.0 1.8 0.3 1.0 2.36

8.4 5.8 9.1 6.8 1.96

AC 10 Con IBA RootRem IBA + RootRem L.S.D.df 101

83 (16) 63 (14) 34 (11) 43 (12) NA

36 39 16 25 14.5

1.4 1.1 0.7 1.2 0.54

3.2 1.2 0.4 1.2 2.98

8.7 7.3 8.0 8.2 1.81

recorded on the cuttings propagated on 16 August 2004. Only rooted cuttings were used for subsequent root and phytohormone treatments on 16 September. Results recorded 21 days later (5 October) showed that in those cuttings where roots had been excised (RootRem), re-rooting percentages were high (79%, with no significant difference associated with phytohormones). There were no statistical differences in shoot activity between treatments at any stage of the experiment (data not shown). The RootRem treatment reduced the total number of roots and root length recorded on cuttings on 5 October (Table 3), but showed the greatest increase in new adventitious roots compared to the previous record on 16 September. The increase with RootRem and RootRem + IBA + Z being significantly greater than their equivalent Con treatments (Table 3). ApRem had no consistent effect on root number or number of new roots, but there was a reduction in root length (significant in all except Z treated cuttings). There was little direct effect of phytohormone treatments on root numbers, although Z application reduced root length in the Con cuttings (Table 3). By comparison with Experiment 1 shoot growth rates in cuttings were slower and there were no significant effects of root or phytohormone treatments on shoot extension or leaf number (data not shown).

Change in leaf number

Stem dry weight (g)

3.2.1. Experiment 2i—effects of repeated root removal during autumn 2004 After further excision of adventitious roots in the RootRem cuttings on 5 October, cuttings again re-rooted by 18 October (i.e. 100% in all treatments). Adventitious root numbers increased compared to the previous assessment (Fig. 4), but there were no significant effects due to Z, IBA or a combination of the two. Similarly, phytohormone treatments had no significant influence over other growth parameters in the RootRem cuttings (data not shown). 3.2.2. Experiment 2ii—longer term effects of phytohormone treatments on cuttings that retained all or part of their root system During the period from 5 to 18 October cuttings in the Con and ApRem continued to develop in the propagation environment, but there were no significant treatment effects associated with shoot activity, cutting extension or leaf increment (data not shown). Root length, however, showed Table 3 Experiment 2: effects of root manipulation (Con, control; ApRem, root apex removed; RootRem, entire root removed) and phytohormone application (none, IBA, Z or IBA + Z) on mean number and length of roots of A. occidentale AC 10 cuttings as measured on 5 October 2004, and the increase in adventitious root number since the previous recording on 16 September 2004 Phytohormone applied None No. of roots Con ApRem RootRem

6.2 6.1 4.5

Root length (mm) Con 120 ApRem 72 RootRem 35

Fig. 3. Experiment 2. Maximum and minimum daily temperature (8C) and irradiance (mmol m 2 s 1) recorded in the propagation environment during late summer/autumn 2004. Dates relate to when cuttings were inspected or treated.

IBA 8.2 7.2 4.8 105 67 39

No. of new adventitious roots Con 2.1 3.0 ApRem 2.8 1.2 RootRem 4.5 4.8

L.S.D.df 273 Z 6.9 7.1 3.5 89 72 18 3.1 1.8 3.5

IBA + Z 7.9 6.6 4.1 101 59 25 2.1 3.0 4.1

2.12

19.3

1.96

J. Saranga, R. Cameron / Scientia Horticulturae 111 (2007) 164–172

169

Table 4 Experiment 2ii: effect of zeatin (Z) application on mean root number, root and shoot dry weight in A. occidentale AC 10 cuttings that had retained all (Con) or part (ApRem) of their root systems

No. of total roots Root dry wt. (g) Shoot dry wt. (g)

Con

Con + Z

ApRem

ApRem + Z

L.S.D.df 181

283 0.74 4.71

398 0.80 5.56

223 0.51 4.06

266 0.65 5.06

71.4 0.153 0.798

Data recorded on 18 October 2004. Data are pooled across IBA treatment, which showed no significant effects.

Fig. 4. Experiment 2i. Mean number of new adventitious roots recorded in A. occidentale AC 10 cuttings after roots were repeatedly removed (RootRem) during autumn 2004. Root numbers were recorded on 5 and 18 October, after cuttings had been exposed to phytohormone application on 16 September (none, IBA, Z or IBA + Z) and root removal on the 16 September and again, after recording on 5 October 2004.

effects due to root treatment (P < 0.001) and zeatin (P = 0.01), with treatments that involved either ApRem or Z being associated with reduced root extension (Fig. 5). In ApRem cuttings the addition of IBA + Z was particularly detrimental. In contrast, zeatin treatments were often associated with increased final root numbers and shoot dry weight (e.g. Con + Z significantly greater than Con), but there was no significant effect with IBA (data pooled across IBA treatments, Table 4). ApRem treatment tended to reduce root weight, but had no significant effect on shoot weight (Table 4). In the ApRem treatment the addition of zeatin, appeared to counteract the loss of root-tips in that final shoot weights were significantly greater (and root weights almost so) than those not treated with zeatin (Table 4). 4. Discussion The use of enclosed mist combined with a free draining rooting medium (perlite), resulted in high rooting percentages

Fig. 5. Experiment 2ii. Effect of phytohormone application (none, IBA, Z or IBA + Z) on mean root length in A. occidentale AC 10 cuttings that had retained all (Con) or part (ApRem) of their root systems. Data recorded on 18 October 2004.

in both experiments. This supports previous studies in Anacardium where rooting percentages can be improved through the use of supportive environments (Sen et al., 1991) and appropriate drainage (Coester and Ohler, 1976). It also corresponds with studies in other difficult-to-root species where rooting can be enhanced when the use of supportive propagation environments allows cuttings to survive long enough to enable adventitious root formation to take place before tissue viability is lost (Howard and Harrison-Murray, 1995). One assumption from previous studies (and commercial production) has been that cutting viability in finite and rooting in softwood cuttings must occur relatively quickly or the cutting will fail due to dehydration (Mudge et al., 1995; Puri and Thompson, 2003) or insufficient carbohydrates (HarrisonMurray and Thompson, 1988). To some extent, the results presented here challenge that assumption in that removing adventitious roots as they formed over a 63 day period did not result in cutting failure. Despite the frequent removal of roots, however, we cannot exclude the possibility that these roots were contributing to the cutting’s viability when they were present. Even relatively young roots may contribute to shoot development via water uptake or hormonal action, such as the export of cytokinins (Taylor and Van Staden, 1997). Nevertheless, the results suggest that cuttings can tolerate a degree of root injury without failing. Indeed, the removal of roots not only encouraged the emergence of new roots on four to five separate occasions, but that the number of adventitious roots formed increased in each subsequent occasion. In commercial situations, there may some advantages in promoting greater numbers of adventitious primary roots to form directly from the stem. For example, cuttings that generate only one root are often discarded due to the promotion of a weak and one-sided root system. Removing that single root, however, and placing the cutting back in the propagation environment may encourage a more robust root system, and hence a stronger saleable plant. Temperatures within the propagation environment varied with weather conditions, and there was no apparent seasonal trends recorded in Experiment 1, although daily maximum temperatures tended to decrease marginally with time in Experiment 2. There was no consistent trend, however, between temperature and speed or extent of rooting. Indeed, numbers of roots generated increased with time, for example in Experiment 2, between 5 and 18 October 2003 despite reductions in temperature and light levels between the periods preceding these harvests (compare Figs. 3 and 4).

170

J. Saranga, R. Cameron / Scientia Horticulturae 111 (2007) 164–172

In both experiments time for root emergence varied strongly, however, between first root expression and any subsequent root emergence episode. Initial adventitious roots were expressed after approximately 21 days, yet after root removal, new adventitious roots emerged after only 8–15 days, i.e. much faster root expression. The precise reason for faster expression with the second, third, etc., rooting events are not clear, but it may be that these roots had already completed a number of preliminary developmental stages. Adventitious rooting has been divided into three to five distinguishable physiological stages, depending on species (Hartmann et al., 2001). The main difference between the first and the following rooting events is that the first rooting occurred in shoots which had just been shortly excised from the mother-plant. In contrast, the later rooting episodes occurred in cuttings which had already been part re-organised. In ‘difficult-to-root’ species, adventitious rooting has been associated with a period of non-directed (undifferentiated) cell division prior to the induction of competent root forming cells (Lovell and White, 1986). It may be that this period of undifferentiated cell division delayed root expression in the first instance, but at the later episodes, cells were already competent to form roots and hence expression was quicker. Similarly, it is also possible that more root primordia are formed over the first 21 days than are actually expressed. Possibly, the growth of the earliest emerging roots inhibits the further development of the remaining primordia, and these remain latent. Indeed, the 8–15 day root expression period is similar to that found when cuttings are taken of species with naturally pre-formed primordia present, e.g. Populus (Cameron, unpublished) and Salix (Fjell, 1987). If competent cells or even root primordia are formed, yet some are inhibited from further development, the question arises what is the signalling mechanism that switches off development? Presumably there is a stage where preference is given to the development of lateral roots derived from the already emergent roots, rather than promoting greater numbers of adventitious roots directly from the stem. Indeed, this is evident in these studies, where few new adventitious roots were noted, once secondary root branching commenced in those cuttings where roots were not removed. This may relate to competition for carbohydrates, with the root apices of extending adventitious roots providing a greater sink demand and hence, inhibiting further root development from the stem. Alternatively, some other signalling mechanism may be generated by the extending roots themselves, that inhibits any further development of de novo roots or the extension of the remaining primordia (i.e. a form of correlative inhibition). Whether the mechanisms controlling this are analogous to those thought to regulate lateral root development is unclear (Zhang and Hasenstein, 1999). Although the removal of adventitious roots did not induce cutting failure, there was some evidence that carbohydrate (and possibly other) resources were being diverted from other sinks within the cutting. By the end of the first experiment, shoot activity, extension growth and the number of newly expanded leaves were reduced in cuttings where roots were removed

compared to those where they were left intact. Similar trends were apparent in Experiment 2, but values were non significant. Whether this was a consequence of reduced storage carbohydrates (Li and Leung, 2000) or the re-direction of new photosynthates (Rapaka et al., 2005) is not clear. Indeed, it is feasible that continual removal of roots was influencing water balance or phytohormone activity and thereby indirectly affecting CO2 fixation. Photosynthesis and carbon gain in cuttings is often closely associated with the development of roots, with stomata remaining closed and photosynthetic rates low, until root meristem activity increases (Feldman et al., 1989; Thomas, 2000). Continual removal of roots in this case, therefore may not only have depleted existing carbohydrate reserves, but also have inhibited further assimilation. In research by Feldman et al. (1989) and Gay and Loach (1977) stomata conductance remained low until visible root emergence, but in contrast Svenson et al. (1995) working on Euphorbia pulcherrima demonstrated that stomata opened concurrently with the formation of root primordia and two days before root emergence. Their findings, although acknowledging that assimilation increases with progressive root development, suggest that carbon gain can occur with the presence of primordia alone. Stomatal behaviour was not measured in Anacardium, but if responses were similar to Euphorbia then assimilation (albeit at lower levels) may have still been feasible after root excision, due to the presence of the remaining root primordia. Perhaps the presence of undeveloped root primordia contributed to maintaining cutting viability in our studies. Alterations in cutting growth may also have been affected directly by phytohormones without affecting carbon assimilation. Root-derived cytokinins, particularly, have been associated with promoting shoot development and inhibiting leaf senescence (Sitton et al., 1967), and continual removal of roots may have disrupted cytokinin activity in our cuttings. However, more recent evidence suggests cytokinins that regulate such activity, need not necessarily be synthesised in the roots (Carmi and Van Staden, 1983; Emery and Atkins, 2002). Auxins are clearly involved in the root formation process, and although the application of exogenously applied auxins can aid adventitious rooting, it is not a universal requirement across species. In addition to understanding the role IBA may play in root formation of Anacardium, we were also interested in how it affected further root development and cutting establishment. Similarly, cytokinins are sometimes quoted as counteracting the actions of auxins (Van Staden and Harty, 1988), so these were included in our studies. Exposing cuttings to IBA at the time of preparation (Experiment 1) appeared to be favourable for one genotype (AC 10), but not for the other (CCP 1001) in terms of percentage rooting and numbers of adventitious roots generated after each removal episode. Variations in response to auxin, even within the one species are not unusual (Stankova and Panetsos, 1997). The addition of IBA or zeatin once adventitious roots had formed and then been removed (RootRem, Experiment 2) had no significant effect on re-rooting or subsequent cutting development. Phytohormone addition, at least at the concentrations applied, appeared to have little influence on the re-

J. Saranga, R. Cameron / Scientia Horticulturae 111 (2007) 164–172

expression of adventitious roots. In contrast, in those cuttings where the adventitious roots were retained and then exposed to phytohormone treatments, zeatin applications reduced root length, but tended to promote root number, i.e. more, but shorter roots. Zeatin also tended to counteract some of the negative factors associated with removing the root tips, and may have partially substituted for the loss of endogenous cytokinins in these cuttings (Taylor and Van Staden, 1997) To some extent these results with zeatin are surprising as studies exploiting transgenic lines suggest that cytokinins tend to reduce root mass (Faiss et al., 1997; Roeckel et al., 1998), and not only inhibit adventitious root formation, but root growth in general. Nevertheless, this may be a simplification of processes involved in root induction, formation and development. For example, studies by De Klerk et al. (2001) and Soh et al. (1998), imply that cytokinins, particularly at low concentrations, may be essential at early stages of the root induction process. Overall effects of IBA on root development in rooted cuttings were less evident than those of zeatin. There was no overall effect in Experiment 2; and in Experiment 1 total root numbers in AC 10 were significantly reduced, despite IBA optimising percentage rooting. Auxins are closely correlated with increased lateral root formation (Boerjan et al., 1995; Celenza et al., 1995; Reed et al., 1998) but actual responses can vary depending on form of auxin applied, concentration and the sensitivity of tissues to absorb or utilise the exogenous auxin (Chhun et al., 2004). Lloret and Casero (2002) claim that lateral root development is regulated by interactions between auxin and cytokinin, but the exact nature of these, remain undetermined. There was little advantage to be gained in removing the root apices (ApRem), and there was no evidence of increased lateral root formation in this treatment compared to control cuttings. Indeed, the main responses tended to be reduced root length, fewer numbers of total roots and a reduced root dry weight. Removal of the primary root apex is often a stimulus for lateral root growth (Zhang and Hasenstein, 1999), but both Con and ApRem treatments with Anacardium formed lateral roots readily, and possibly the only consequence of removing apices was to delay the development of the root system. Data from this research suggest care should be used when defining ‘rooting or propagation success’, and confirms the earlier report by Wilson and Struve (2003). Those treatments that corresponded with higher percentage rooting, did not always correlate with greater root production per cutting. For example, treatment with IBA to cuttings of AC 10 increased rooting percentage, but decreased final root number (Table 1). Further research is warranted to investigate the relationship between early root development and that of establishment and final plant quality. The primary objectives of this research were to clarify the importance of speed of rooting within softwood leafy cuttings, and to investigate how susceptible cuttings were to loss of viability, should either rooting be slow or if any new roots were damaged shortly after formation. Despite using a reputedly ‘difficult-to-root’ species the results contradicted our hypothesis and demonstrated that rooting potential remained strong

171

over a prolonged period, when supportive propagation environments were employed. Indeed, rather than rooting potential decreasing with time, the numbers of adventitious roots formed increased, when stimulated by repeated excision of previously formed roots. In a commercial context, this characteristic may help offset any injury sustained when cuttings and young plants are removed from their rooting trays and potted-on. Clearly, in Anacardium at least, rooting potential remains high when the appropriate balance of light, humidity, drainage and leaf wetting is provided in the propagation environment. The challenge therefore is to develop simple, effective systems that replicate such conditions in commercial situations and hence, enable cuttings to survive long enough to express their rooting potential. This will help ensure that ‘difficult-to-root’ species can be dealt with cost-effectively and successfully, without the requirement for more intensive or expensive propagation techniques. References Aiello, A.S., Graves, W.R., 1998. Success varies when using subirrigation instead of mist to root softwood cuttings of woody taxa. J. Environ. Hort. 16, 42–47. Aminah, H., 2003. Vegetative propagation of Endospermum malaccense by leafy stem cuttings: effects of indole butyric acid (IBA) concentrations and propagation systems (mist and non-mist). J. Trop. For. Sci. 15, 249–258. Billingsley, G., 2003. Increased rooting percentages and decreased transplant shock using Ellepots. Com. Proc. Int. Plant Prop. Soc. 53, 579–581. Boerjan, W., Cervera, M.T., Delarue, M., Beeckman, T., Dewitte, W., Bellini, C., Caboche, M., Onckelen, H., van Montagu, M., van Inze, D., 1995. Superroot, a recessive mutation in Arabidopsis, confers auxin overproduction. Plant Cell. 1405–1419. Cameron, R.W.F., Harrison-Murray, R.S., Ford, Y.-Y., Judd, H.L., 2001. Ornamental shrubs: effects of stock plant management on the rooting and establishment of cuttings. J. Hort. Sci. Biotechnol. 76, 489–496. Cameron, R.W.F., Harrison-Murray, R.S., Fordham, M., Judd, H.L., Ford, Y.-Y., Marks, T.R., Edmondson, R., 2003. Rooting cuttings of Syringa vulgaris cv. Charles Joly and Corylus avellana cv. Aurea: the influence of stock plant pruning and shoot growth. Trees 17, 451–462. Carmi, A., Van Staden, J., 1983. Role of roots in regulating the growth rate and cytokinin content in leaves. Plant Physiol. 73, 76–78. Celenza, J.L., Grisafi, P.L., Fink, G.R., 1995. A pathway for lateral root formation in Arabidopsis thaliana. Genes Dev. 9, 2131–2142. Chhun, T., Taketa, S., Tsurumi, S., Ichii, M., 2004. Different behaviour of indole-3-acetic acid and indole-3-butyric acid in stimulating lateral root development in rice (Oryza sativa L.). Plant Growth Regul. 43, 135–143. Coester, W.A., Ohler, J.G., 1976. Cashew propagation by cuttings. Trop. Agric. 53, 353–358. Damodaran, V.K., 1985. Vegetative propagation of cashew—review of work done in Kerala. Acta Hort. 108, 51–56. De Klerk, G.J., Hanecakova, J., Jasik, J., 2001. The role of cytokinins in rooting of stem slices cut from apple microcuttings. Plant Biosyst. 135, 79–84. Del Rio, C., Rallo, L., Caballero, J.M., 1991. Effects of carbohydrate content on the seasonal rooting of vegetative and reproductive cuttings of olive. J. Hort. Sci. 66, 301–309. Duarte, O., Nieto, J.M., Suarez, A., 1992. Treatments for improving seed germination and rooting of cashew (Anacardium occidentale L.) cuttings. Proc. Int. Soc. Trop. Hort. 35, 9–14. Emery, R.J.N., Atkins, C.A., 2002. Roots and cytokinins. In: Waisel, Y., Eshel, A., Kafkafi, U. (Eds.), Plant Roots—The Hidden Half. Marcel Dekker, New York, pp. 417–434. Ermel, F.F., Vizoso, S., Charpentier, J.P., Jay-Allemand, C., Catesson, A.M., Couee, I., 2000. Mechanisms of primordium formation during adventitious root development from walnut cotyledon explants. Planta 211, 563–574.

172

J. Saranga, R. Cameron / Scientia Horticulturae 111 (2007) 164–172

Faiss, M., Zalubilova, J., Strnad, M., Schmulling, T., 1997. Conditional transgenic expression of the ipt gene indicates a function for cytokinins in paracrine signaling in whole tobacco plants. Plant J. 12, 401–415. Feldman, W.R., Palzkill, D.A., Dobrenz, A.K., Hogan, L., 1989. Carbon dioxide exchange during mist propagation of jojoba cuttings. Agron. J. 81, 233–236. Fjell, I., 1987. Structure and development of adventitious roots in Salix caprea and S. caprea  S. viminalis. Nord. J. Bot. 7, 311–323. Grange, R.I., Loach, K., 1983. The water economy of unrooted leafy cuttings. J. Hort. Sci. 58, 9–17. Gay, A.P., Loach, K., 1977. Leaf conductance changes on leafy cuttings of Cornus and Rhododendron during propagation. J. Hort. Sci. 52, 509–516. Griffin, J.J., Schroeder, K.R., 2004. Propagation of Ulmus parvifolia ’Emerald Prairie’ by stem cuttings. J. Environ. Hort. 22, 55–57. Hackett, W.P., 1988. Donor plant maturation and adventitious root formation. In: Davis, T.D., Haissig, B.E., Sankhla, N. (Eds.), Adventitious Root Formation in Cuttings. Dioscorides Press, Portland, pp. 11–28. Harrison-Murray, R.S., Howard, B.H., 1992. Innovations in manipulation of environment for propagation research. Acta Hort. 314, 249–256. Harrison-Murray, R.S., Thompson, R., 1988. In pursuit of a minimum stress environment for rooting leafy cuttings: comparison of mist and fog. Acta Hort. 227, 211–216. Hartmann, H.T., Kester, D., Davies, F., Geneve, R., 2001. Plant Propagation: Principles and Practices, seventh ed. Pearson Higher Education, USA. Howard, B.H., Harrison-Murray, R.S., 1995. Responses of dark-preconditioned and normal light-grown cuttings of Syringa vulgaris ‘Madame Lemoine’ to light and wetness gradients in the propagation environment. J. Hort. Sci. 70, 989–1001. Li, M., Leung, D.W.M., 2000. Starch accumulation is associated with adventitious root formation in hypocotyl cuttings of Pinus radiata. J. Plant Growth Regul. 19, 423–428. Littlejohn, G.M., Gertse, S., 2001. Genetic variation in rootability of Leucospermum hybrids. Acta Hort. 545, 187–190. Lloret, P.G., Casero, P.J., 2002. Lateral root initiation. In: Waisel, Y., Eshel, A., Kafkafi, U. (Eds.), Plant Roots—The Hidden Half. Marcel Dekker, New York, pp. 127–155. Lovell, P.H., White, J., 1986. Anatomical changes during adventitious root formation. In: Jackson, M.B. (Ed.), New Root Formation in Plants and Cuttings. Martinus Nijhoff Publishers, Dordrecht, pp. 111–140. Mneney, E.E., Mantell, S.H., 2002. Clonal propagation of cashew (Anacardium occidentale L.) by tissue culture. J. Hort. Sci. Biotechnol. 77, 649–657. Mudge, K.W., Mwaja, V.N., Itulya, F.M., Ochieng, J., 1995. Comparison of 4 moisture management-systems for cutting propagating of Bougainvillea, Hibiscus, and Kei apple. J. Am. Soc. Hort. Sci. 120, 366–373. Owen, J., Johnson, W., Maynard, B., 2001. Cutting propagation screening trials at University of Rhode Island. Com. Proc. Int. Plant Prop. Soc. 51, 514–516. Puri, S., Thompson, F.B., 2003. Effect of soil and plant water relations on rooting of Populus  euramericana stem cuttings. New For. 25, 109– 124. Rao, S.N., 1985. Vegetative propagation of cashew—review of work done at Baptla. Acta Hort. 108, 64–66. Rao, M.B.N., Satyanarayana, G., Raj, A.S., Kumari, N.G., Padmanabham, V., 1988. Interaction of source-plant age and shoot ringing on rooting of Cashew (Anacardium occidentale L.) cuttings. J. Hort. Sci. 63, 517–519. Rao, M.B.N., Satyanarayana, G., Raj, A.S., Gnanakumari, N., Padmanabham, V., 1990. Effect of some propagation structures on rooting of stem

cuttings of cashew (Anacardium occidentale L.). Ind. Cashew J. 20, 17–20. Rapaka, V.K., Bessler, B., Schreiner, M., Druege, U., 2005. Interplay between initial carbohydrate availability, current photosynthesis and adventitious root formation in Pelargonium cuttings. Plant Sci. 168, 1547–1560. Reed, R.C., Brady, S.R., Muday, G.K., 1998. Inhibition of auxin movement from the shoot into the root inhibits lateral root development in Arabidopsis. Plant Physiol. 118, 1369–1378. Reineke, R.A., Hackett, W.P., Smith, A.G., 2002. Lignification associated with decreased adventitious rooting competence of English ivy petioles. J. Environ. Hort. 20, 236–239. Reuveni, O., Raviv, M., 1981. Importance of leaf retention to rooting of avocado cuttings. J. Am. Soc. Hort. Sci. 106, 127–130. Roeckel, P., Oancia, T., Drevet, J.R., 1998. Phenotypic alterations and component analysis of seed yield in transgenic Brassica napus plants expressing the tzs gene. Physiol. Plantarum 102, 243–249. Rose, N., Pellett, H., 1994. Rhododendron ‘Northern Hi-Lights’. J. Environ. Hort. 12, 179–180. Sen, S.K., Debnath, S., Bandyopadhyay, A., Poi, A., 1991. Propagation of Cashewnut Anacardium occidentale L. from cuttings. Environ. Ecol. 9, 268–271. Sitton, D., Itai, C., Kende, H., 1967. Decreased cytokinin production in the roots as a factor in shoot senescence. Planta 73, 296–300. Soh, W., Choi, P., Cho, D., 1998. Effects of cytokinin on adventitious root formation in callus cultures of Vigna unguiculata (L.) Walp. In Vitro Cell. Dev. Pl. 34, 189–195. Stankova, T., Panetsos, K., 1997. Vegetative propagation of Cupressus sempervirens L. of Cretan origin by softwood stem cuttings. Silvae Genet. 46, 137–144. Svenson, S.E., Davies, F.T., Duray, S.A., 1995. Gas exchange, water relations, and dry weight partitioning during root initiation and development of poinsettia cuttings. J. Am. Soc. Hort. Sci. 120, 454–459. Tarrago, J., Sansberro, P., Filip, R., Lopez, P., Gonzalez, A., Luna, C., Mroginski, L., 2005. Effect of leaf retention and flavonoids on rooting of Ilex paraguariensis cuttings. Sci. Hortic. 103, 479–488. Taylor, J.L.S., Van Staden, J., 1997. Variation in the level and type of cytokinin with the stage of root development in Impatiens wallerana Hook. F. stem cuttings. Plant Growth Regul. 22, 175–180. Thomas, P., 2000. Microcutting leaf area, on the stock shoot influence root vigour, shoot weight and position growth and incidence of shoot tip necrosis in grape plantlets in vitro. Plant Cell. Tiss. Org. Cul. 61, 189–198. Van Staden, J., Harty, A.R., 1988. Cytokinins and adventitious root formation. In: Davis, T.D., Haissig, B.E., Sankhla, N. (Eds.), Adventitious Root Formation in Cuttings. Dioscorides Press, Portland, pp. 185–201. Voyiatzi, C., Petridou, M., Pritsa, T., Sotiriou, M., Voyiatzis, D., 2002. Rooting capacity of cuttings as a criterion for the evaluation of the progeny of five olive cultivars. Acta Hort. 586, 927–930. Wilkerson, E.G., Gates, R.S., Zolnier, S., Kester, S.T., Geneve, R.L., 2005. Transpiration capacity in poinsettia cuttings at different rooting stages and the development of a cutting coefficient for scheduling mist. J. Am. Soc. Hort. Sci. 130, 295–301. Wilson, P.J., Struve, D.K., 2003. Rooting variables for stem cuttings. J. Hort. Sci. Biotechnol. 78, 29–31. Zhang, N.G., Hasenstein, K.H., 1999. Initiation and elongation of lateral roots in Lactuca sativa. Int. J. Plant Sci. 160, 511–519.