Scientia Horticulturae 63 ( 1995) 47-55
SCIENTIA
HORTICULTURR
Development of axillary buds of rose in vitro
C.A.M. Marcelis-van Acker a,b**,H.J. Scholten a
aDepartment of Horticulture, Wageningen Agricultural University,Haagsteeg 3, 6708 PM Wageningen.
Netherlands
bDepartment of Plant Cytology and Morphology, Wageningen Agricultural Vniversiry, Arboretumlaan 4,
6703 BD Wageningen, Netherlands
Accepted 3 February 1995
Abstract
An in vitro model system has been developed to grow axillary buds into shoots, which are mor-
phologically comparable to those grown in vivo. In the development of the shoot three stages can be
distinguished: sprouting of the bud, unfolding of leaves already present in the bud, and new formation
of leaves and a flower bud. A low concentration of benzyladenine (BA), sugar (preferably glucose),
and a cultivar-dependent concentration of a suitable agar was necessary to obtain a complete shoot.
The size of the in vitro shoot positively correlated with the size of the explant. The presence of the
petiole inhibited the outgrowth of lateral buds, which were already present in the inoculated bud.
Using this in vitro system the growth potential of axillary buds, apart from influences of other plant
parts, can be studied.
&words: Axillary bud; Micropropagation; Model system; Morphology; Rosa hybrida
1. Introduction
The flower yield of roses depends on the willingness of axillary buds to sprout and their
subsequent growth into flowering shoots (Zieslin et al., 1973). The development of axillary
buds on intact plants depends on three partly interdependent factors, i.e. the intrinsic growth
potential of the buds, their position on the plant (Zieslin et al., 1976) and influences of
other plant parts (Zieslin and Halevy, 1976). This latter point, especially, makes it difficult
to assess the growth potential of the buds themselves in intact plants. In vitro culture proved
useful to study bud and shoot growth, isolated from the interactions to which the buds are
* Corresponding author.
Abbreviations: BA = Benzyladenine; GA = Gibberellic acid; IBA = Indole butyric acid; IPA = Indole propionic
acid; 2iP = 2-isopentenyladenosine; MS = Murashige and Skoog; NAA = Naphthaleneacetic acid.
0304-4238/95/$09.50 0 1995 Elsevier Science B.V. All rights reserved
SsDIO304-4238(95)00786-5
48 C.A.M. Marcel&van A&r, H.J. Scholten /Scientia Horticulturae 63 (1995) 47-55
subjected in the intact plant. Moreover, it offers a technique to study the physiology and
requirements for growth and development of isolated organs (Dutcher and Powell, 1972;
Halim et al., 1988; Nadel et al., 1991).
In each leaf axil of a rose shoot an axillary bud is present. A quiescent axillary bud of a
flowering shoot contains the lower part of the future shoot, i.e. six to seven scale-like leaves
and four to five compound leaves. When the correlative inhibition of the axillary bud is
released by removing the stem above the bud, the bud will sprout and approximately seven
new leaves and a flower will be formed (Marcelis-van Acker, 1994).
In vitro culture of rose has been described previously, but has focused on micropropa-
gation (Jacobs et al., 1969; Hasegawa, 1979; Bressan et al., 1982), somatic embryogenesis
and adventitious shoot formation (Tweddle et al., 1984). The studies on in vitro culture of
rose have been reviewed by Short and Roberts ( 1991). The methods of micropropagation
of rose are always based on the system of multiple shoot formation, starting with axillary
buds or shoot tips, which are regularly subcultured. In most cases a standard Murashige and
Skoog (MS) medium (Murashige and Skoog, 1962) is used with a rather high benzylad-
enine (BA) concentration. The shoot proliferation may decline after a number of subcultures
at a high BA concentration, as reported by Norton and Norton ( 1986). Furthermore, in
vitro derived plants often show more branching after hardening than in vivo plants (Dubois
et al., 1988; Vijaya and Satyanarayana, 1991), which might be a carry-over effect of the
relatively high BA concentration applied in the multiple shoot system. The multiple shoot
system is not a suitable system to study the growth potential of axillary buds in vitro.
The aim of the present study was to develop an in vitro model system for physiological
studies on the development of axillary rose buds. In this system axillary shoots of rose are
grown as single elongated shoots at low cytokinin concentration. The effects of medium
components (agar, plant growth regulators and sugars), petiole and explant size on bud and
shoot development were evaluated. Furthermore, the growth characteristics of the in vitro
shoot were compared with those of the in vivo shoot.
2. Materials and methods
From two cultivars of Rosa hybrida, `Sweet Promise' and `Motrea', stem segments about
1 cm long, bearing a quiescent axillary bud including a petiole stump of 1 cm were cut from
the middle part of flower stems in the harvestable stage (sepals reflexing) . Flower stems
were used immediately after harvest. The explants were surface-sterilised in 70% alcohol
for a few seconds, followed by 20 min in 1% NaOCl. Explants were then washed three
times with sterile water. Before inoculation the buds were excised, leaving as few stem and
petiole tissue as possible. The effect of explant size was investigated by varying the stem
length (0,0.5 and 2.0 cm) or by varying the petiole length (0,0.5, 1.O, 1.5 and 2.0 cm).
The basic culture medium consisted of MS salts with 37.5 mg l- ' NaFeEDTA. Unless
stated otherwise, 45 g 1-i glucose, 0.1 mg 1-l BA and 5 g 1-l agar (MC 29; Lab M, Bury,
UK) were used. Preliminary results showed that, in contrast to `Sweet Promise', `Motrea'
required the addition of MS vitamins and glycine. The effect of sugar type was evaluated
by comparing sucrose with glucose at a concentration of 45 g 1-I. The effect of BA
concentration was investigated by applying 0,0.044,0.22,0.44,0.88 and 2.2 PM BA. The
C.A.M. Marcelis-van Acker, H.J. Scholten /Scientia Horticulturae 63 (1995) 47-55 49
effect of cytokinin type was studied by comparing equimolar concentrations (0.44 PM) of
the cytokinins BA, zeatin, zeatin riboside, 2-isopentenyladenosine (2iP), indole propionic
acid (IPA) and kinetin. The effect of agar brand was investigated by testing six different
commercially available agars at a concentration of 5 g l-`, i.e. Daichin (Brunschwig
Chemie, Amsterdam, Netherlands), Difco Bacto (Difco, Detroit, USA), MC 29 (Lab M,
Bury, UK), BD (Becton Dickinson, Cockeysville, USA) grade A, BD granulated, and BD
purified. Furthermore, for agar MC 29 a concentration range (5, 6, 7 and 8 g 1-l) was
applied. The pH was adjusted to 5.8. Culture tubes were filled with 15 ml medium, closed
with a cotton plug, and autoclaved for 20 min. Tubes with an explant were sealed with
Vitafilm (Good Year) and incubated in a culture room at 23°C with a 16 h photoperiod (5-
7 W m-*, Philips TL 54).
Routinely, the effects of the treatments on growth and development of the axillary buds
into shoots were evaluated after 4 weeks by determining length, weight and number of
compound leaves of the main shoot, and number of lateral shoots. The experiment on
cytokinin type was evaluated after 5 weeks. For calculation of the means only sprouted
buds were taken into account.
The effects of medium components and explant size were investigated in one or two
replicate experiments. Per treatment 12, 18, or 24 replicate explants were used. When one
experiment was performed the mean and the standard error of the mean (SE) were calculated
per treatment. In the case of two experiments analysis of variance was applied and the
significance of differences was determined by Student's t-test (P = 0.05).
3. Results
3.1. Morphology
In general, at least 90% of the axillary buds inoculated in vitro sprouted (length over 0.5
cm; stage I), On most media, the four to five leaves and internodes which were already in
the bud at the time of inoculation, unfolded (stage II). When cultured under optimal
conditions, approximately seven additional leaves and internodes and a flower were formed
(stage III). Comparative studies revealed that this developmental process under optimal
conditions in vitro is very similar to that in vivo, resulting in a miniature version of the in
vivo plant. The'number of leaves preceding the flower and the leaf form (number of leaflets
per leaf) of in vitro grown shoots were very similar to those of in vivo grown shoots.
3.2. Medium components
3.2.1. Plant growth regulators
For both cultivars the presence of a cytokinin was necessary for prolonged shoot growth
of axillary buds; without cytokinin, axillary buds did sprout, but hardly developed further
than stage II, since only a few leaves were formed and little shoot elongation occurred
(Table 1) . A concentration of 0.44 PM BA resulted in elongated shoots with a flower bud,
whereas higher concentrations induced a cluster of shoots. At high concentrations weight
and length of the main shoot decreased (Table 1). Comparing equimolar concentrations
50 C.A.M. Marcelis-van Acker, H.J. Scholten/Scientia Horticulturae 63 (1995) 47-55
Table 1
Effect of BA concentration on axilhuy shoot development from single node explants of rose `Sweet Promise'.
Values are the mean of 48 plants
BA
cont.
(@vu
0
0.044
0.22
0.44
0.88
2.2
No. of Length
leaves (cm)
6.0 1.0
6.9 1.2
10.1 2.3
11.3 3.4
11.6 3.0
11.5 1.7
Weight of
main shoot
(g)
0.24
0.23
0.25
0.30
0.28
0.24
No. of
laterals
0.1
0.1
0.5
1.2
2.3
2.7
Total
weight
(8)
0.24
0.23
0.25
0.34
0.37
0.46
LSD (P=O.O5) 0.9 0.6 0.06 0.9 0.08
(0.44 PM) of the cytokinins BA, zeatin, zeatin riboside, 2iP and IPA showed that all
cytokinins, except kinetin, stimulated the growth of axillary buds into shoots of `Sweet
Promise' (Fig. 1). BA, zeatin and zeatin riboside resulted in the longest shoots with the
highest weight (Fig. 1). Increasing the concentration of kinetin to 9.3 PM had no effect on
weight, length and number of leaves of `Motrea' (data not shown).
3.2.2. Agar
Both cultivars appeared to be sensitive to the agar brand. In several experiments the agars
Daichin (agar 1) , MC 29 (agar 3), and BD purified (agar 6) gave the best results. In Fig.
2 a representative experiment is shown. Since MC 29 showed least symptoms of necrosis
of the apex, this agar was routinely used.
With respect to the optimal agar concentration large differences between the cultivars
were observed. `Motrea' preferred high concentrations of agar. At 7 g 1-l completely
developed shoots were formed (Fig. 3). `Sweet Promise', however, showed the best results
0.4 I-
0.3 I-
3
5 0.2
s
0.1
0.0
No K 2 2iP ZR IPA BA
Fig. 1.Effect of cytokinin type (0.44 PM) on number of leaves (hatched bars), length (dotted bars) and weight
(cross-hatched bars) of axillary shoots from single node explants of rose `Sweet Promise'. Values are the mean
of 18explants. Bars indicate SE. No, no cytokinin added; K, kinetin; Z, zeatin; ZR, zeatin riboside.
C.A.M. Marcelis-van A&r, H.J. Scholten /Scientia Horticulturae 63 (1995) 47-55 51
5 0.3
E
m
g 0.2
0.1
0.0
1 2 3
Agar type4
5 6
Fig. 2. Effect of the agar brand on weight of axillary shoots from single node explants of rose `Motrea' (cross-
hatched bars) and `Sweet Promise' (dotted bats). Values are the mean of 12 plants. Bars indicate SE. Agar 1,
Daichin; agar 2, Difco Bacto; agar 3, MC 29; agar 4, BD grade A; agar 5, BD granulated; agar 6, BD purified.
with extremely low concentrations (4 g 1-r) or when grown on small rockwool plugs with
liquid medium (data not shown).
3.2.3. Sugar
For both cultivars glucose gave better growth than sucrose (Table 2). For `Motrea',
development of the buds was more enhanced with glucose in the medium. The time of
addition of 45 g l- ' sucrose, before or after autoclaving the medium, did not make any
difference. Chemical analysis with HPLC showed that after autoclaving only 3% of the
sucrose was decomposed while glucose could be recovered completely.
3.3. Explantfactors
Shoot growth increased with increasing stem tissue. No effect on developmental stage
was found since the number of leaves was similar (Table 3). Absence of the petiole induced
1.5
1.0
E
E
.y
3
0.5
:..i:\l10.0 L
5 6 7 a
Agar concentration (g 1-t)
Fig. 3. Effect of agar (MC 29) concentration on weight ( A ), length (0) and number of leaves (0) of axillary
shoots from single node explants of rose `Motrea'. Values are the mean of 12 plants. Bars indicate SE.
52 C.A.M. Marcelis-van A&r, H.J. Scholten /Scientia Horticulturae 63 (1995) 47-55
Table 2
Effect of sugar type at 45 g I-' on axillary shoot development from single node explants of rose `Sweet Promise'
and `Motrea'. Values are the mean of 48 plants
Cultivar sugar No. of
leaves
Length
(cm)
No. of
laterals
Total
weight
(g)
`Sweet Promise'
LSD (P=O.O5)
sucrose 10.3 2.5 1.6 0.25
Glucose 10.8 3.4 1.6 0.33
1.3 0.4 0.7 0.03
`Motrea' Sucrose 8.9 0.5 0 0.10
Glucose 14.0 1.9 0 0.27
LSD (P=O.O5) 1.9 0.6 0.08
Table 3
Effect of explant size on axillary shoot development from single node explants of rose `Sweet Promise'. Values
are the mean of 48 plants
Explant No. of
leaves
Length
(cm)
No. of
laterals
Total
weight
(g)
Axillary bud 8.8 1.7 0.9 0.22
Stem slice of 0.5 cm 8.3 2.0 0.6 0.26
Stem slice of 2.0 cm 8.4 2.5 0.7 0.44
LSD (P=O.O5) 0.7 0.4 0.6 0.02
a
e
4
s 1.0
3
z"
sE 0.5
.P
s
t
L
9
8 c
0
0.0 - 0
0.0 0.5 1.0 1.5 2.0
Petiole length(cm)
Fig. 4. Effect of petiole length on weight ( A ) , number of laterals (n), length (0) and number of leaves (0) of
axillary shoots from single node explants of rose `Sweet Promise'. Values are the mean of 24 plants. Bars indicate
SE.
C.A.M. Marcelis-van Acker, H.J. Scholten /Scientia Horticulturae 63 (1995) 47-55 53
growth of lateral shoots, whereas the main shoot remained short (Fig. 4). The pattern of
the effect of BA concentration on explants was not influenced by the presence of the petiole,
but the optimum concentration was lower in its absence (0.22 $vI instead of standard 0.44
PM). An increase in petiole length resulted in a slight increase in both shoot length and
shoot weight, while the numbers of leaves and laterals were not substantially affected (Fig.
4).
4. Discussion
Axillary buds are not dormant but correlatively inhibited by upper plant parts (Zieslin
and Halevy, 1976). As soon as the inhibition is released in vivo, by pruning the stem part
above the bud, the bud will sprout and develop into a shoot. The medium requirements for
bud break and unfolding of preformed stem parts in vitro appeared not very specific. As the
sprouting bud is a strong sink for assimilates (Mor and Halevy, 1979)) carbohydrates were
supplied to the bud by addition of sugar to the medium. The effect of different sugars on
growth of rose in vitro has not been reported before. Commonly sucrose is supplied, only
Ghashghaie et al. ( 1992) used glucose. The present study shows that glucose induces more
vigorous growth and especially for `Motrea', use of glucose should be strongly advised.
In vivo, the apical meristem completes its developmental programme, after release from
correlative inhibition, by forming several additional leaves and a terminal flower bud. In
vitro, the composition of the medium determines which developmental stage is reached.
Addition of cytokinin to the culture medium appeared necessary for the bud to complete its
developmental programme isolated from the intact plant, Accordingly, axillary buds of pea
did not form new leaves and grew only slightly on a medium without cytokinin (Gould et
al., 1987). Excised floral buds are also reported to require cytokinin for the growth and
development of floral organs (Rastogi and Sawhney, 1989). In vivo, cytokinin will pri-
marily be supplied by the roots, that are absent in vitro. Of the several cytokinins tested,
BA appeared most suitable, although it should be noted that the optimum concentration
might be dependent on the cytokinin used. The BA concentration applied in the multiple
shoot system is supra optimal for the single shoot system, since also the buds, which are
already present in the axils of the bud scales (Marcelis-van Acker, 1994)) were released
from inhibition, resulting in a cluster of shoots. Outgrowth of the buds was accompanied
by a smaller main shoot (Table 1)) indicating that the main shoot experienced competition
of the axillary shoots.
The agar brand may affect the growth and development of in vitro plants (Debergh,
1983). Rose also appeared to be very sensitive to agar brand. Several explanations have
been given for the agar effect on in vitro cultures, e.g. availability of water or nutrients
(Debergh, 1983; Bornman and Vogelmann, 1984; Scherer et al., 1988; Ghashghaie et al.,
1991). In case of rose the nutrient availability may be a likely explanation. For, a reduced
nutrient supply at higher agar concentration agrees very well with the observation that
`Sweet Promise', in contrast to `Motrea', is not able to grow at lower nutrient concentration
than the standard MS medium (H.J. Scholten, unpublished data, 1994). Moreover, addition
of liquid medium to the agar medium after 3 weeks of culture (Maene and Debergh, 1985)
had a positive effect on growth of `Sweet Promise'.
54 CAM. Marc&s-van Acker, H.J. Scholten/Scientia Horticulturae 63 (1995) 47-55
The effect of explant size on growth of the axillary bud might be nutritional and/or
hormonal. The data on the effect of stem tissue attached to the excised axillary bud (Table
3) suggest a nutritional effect of the stem tissue. The effect of the petiole might also be
hormonal, since absence of the petiole induced release of the buds in the axils of the bud
scales of the inoculated bud.
The single shoot system is suitable for physiological studies on axillary bud development.
By use of this system the effects of influencing factors imposed during initiation and early
development of the axillary bud on its subsequent growth potential can be studied and
quantified. This knowlegde will lead to a better understanding of axillary bud development
and of the growth of a rose plant. Furthermore, the single shoot system offers possibilities
for micropropagation by single-node culture. In this way, carry-over effects after transfer
to in vivo conditions, resulting from the high cytokinin concentration in the culture medium,
may be reduced.
Acknowledgements
We would like to thank Prof. Dr. R.L.M. Pierik and Prof. Dr. J. Tromp for valuable
comments on the manuscript.
References
Bomman, C.H. and Vogelmaun, T.C., 1984.Effect of rigidity of gel medium on benzyladenine-induced adven-
titious bud formation and vitrification in vitro in Picea abies. Pbysiol. Plant., 61: 505-512.
Bressan, P.H., Kim, Y.-J., Hyndman, SE., Hasegawa, P.M. and Bressan, R.A., 1982. Factors affecting in vitro
propagation of rose. J. Am. Sot. Hortic. Sci., 107:979-990.
Debergh, P.C., 1983.Effects of agar brand and concentration on the tissue culture.medium. Pbysiol. Plant., 59:
270-276.
Dubois, L.A.M., Roggemans, J., Soyeurt, G. and De Vries, D.P., 1988.Comparison ofthe growthand development
of dwarf rose cultivars propagated in vitro and in vivo by softwood cuttings. Sci. Hortic., 35: 293-299.
Dutcher, R.D. and Powell, L.E., 1972. Culture of apple shoots from buds in vitro. J. Am. Sot. Hottic. Sci., 97:
511-514.
Ghashghaie, J., Brenckmann, F. and Saugier, B., 1991.Effects of agar concentration on water status and growth
of rose plants cultured in vitro. Physiol. Plant., 82: 73-78.
Ghashghaie, J., Brenckmann, F. and Saugier, B., 1992.Water relations and growth of rose plants cultured in vitro
under various relative humidities. Plant Cell Tiss. Org. Cult., 30: 51-57.
Gould, KS., Cutter, E.G., Young, J.P.W. and Charlton, W.A., 1987.Positional differences in size, morphology,
and in vitro performance of pea axillary buds. Can. J. Bot., 65: 4mll.
Halim, H., Edwards, G.R., Coombe, B.G. and Aspinall, D., 1988.The dormancy of buds of Citrus sinensis (L.)
Gsbeck inserted into rootstock stems: Factors intrinsic to the inserted bud. Ann. Bot., 61: 525-529.
Hasegawa, P.M., 1979. In vitro propagation of rose. HortScience, 14:610-612.
Jacobs, G., Allan, P. and Bomman, C.H., 1969.Tissue cultnre studies on rose: use of shoot tip explants. I. Auxin:
cytokinin effects. Agroplautae, 1: 179-188.
Maene, L. and Debergh, P., 1985.Liquid medium additions to established tissue cultures to improve elongation
and rooting in vivo. Plant Cell Tiss. Org. Cult., 5: 23-33 .
Marcelis-van Acker, C.A.M., 1994. Ontogeny of axillary buds and shoots in roses: Leaf initiation and pith
development. Sci. Hortic., 57: 111-122.
C.A.M. Marcel&van Acker, H.J. Scholten /Scientia Horticulturae 63 (1995) 47-55 55
Mor, Y. and Halevy, A.H., 1979. Translocation of "C-assimilates in roses. I. The effect of the age of the shoot
and the location of the source leaf. Physiol. Plant., 45: 177-182.
Mumshige, T. and Skoog, F., 1962. A revised medium for rapid growth and hioassays with tobacco tissue cultures.
Physiol. Plant., 15: 473-497.
Nadel, B.L., Altman, A., Pleban, S. and Hthtermann, A., 1991. In vitro development of mature Fugus syluatica
L. buds. I. The effect of medium and plant growth regulators on bud growth and protein profiles. J. Plant
Physiol., 138: 596601.
Norton, M.E. and Norton, CR., 1986. Change in shoot proliferation with repeated in vitro subculture of shoots of
woody species of Rosaceae. Plant Cell Tiss. Org. Cult., 5: 187-197.
Rastogi, R. and Sawhney, V.K., 1989. In vitro development of angiosperm floral buds and organs. Plant Cell Tiss.
Org. Cult., 16: 145-174.
Scherer, P.A., Muller, E., Lippert, H. and Wolff, G., 1988. Multielement analysis of agar and gelrite impurities
investigated by inductively coupled plasma emission spectrometry as well as physical properties of tissue
culture media prepared with agar or the gellan gum gelrite. Acta Hortic., 226: 655-658.
Short, K.C. and Roberts, A.V., 1991. Rosa spp. (Roses): In vitro culture, micropropagation, and the production
of secondary products. In: Y.P.S. Bajaj (Editor), Biotechnology in Agriculture and Forestry, Vol. 5. Medicinal
and Aromatic Plants III. Springer, Berlin/Heidelberg, pp. 376-397.
Tweddle, D., Roberts, A.V. and Short, K.C., 1984. In vitro culture of roses. In: F.J. Novak, L. Have1 and J. Dolezel
(Editors), Plant Tissue and Cell Culture Application to Crop Improvement. Czechoslovak Academy of
Sciences, Prague, pp. 529-530.
Vijaya, N. and Satyanarayana, G., 1991. Effect of culture media and growth regulators on in vitro propagation of
rose. In: J. P&ash and R.L.M. Pierik (Editors), Horticulture-New Technologies and Applications. Kluwer,
Dordrecht, pp. 209-214.
Zieslin, N. and Halevy, A.H., 1976. Components of axillary bud inhibition in rose plants. I. Effect of different
plant parts (correlative inhibition). Bot. Gaz., 137: 291-296.
Zieslin, N., Halevy, A.H. and Biran, I., 1973. Sources of variability in greenhouse rose flower production. J. Am.
Sot. Hortic. Sci., 98: 321-324.
Zieslin, N., Haaze, H. and Halevy, A.H., 1976. Components of axillary bud inhibition in rose plants. II. The effect
of bud position on degree of inhibition. Bot. Gaz., 137: 297-300.