DE TAI 1 HAT NHAN TAO KHOAI TAY

Seediscussions,stats,andauthorprofilesforthispublicationat: />
Stabilityofpotato(SolanumtuberosumL.)
plantsregeneratedviasomaticembryos,
axillarybudproliferatedshoots,microtubers
andtruepotatoseeds:Acomparative
phenotypic,cy...
ArticleinPlanta·December2007
DOI:10.1007/s00425-007-0583-2·Source:PubMed

CITATIONS

READS

46

490

4authors:
SanjeevSharma

GlennBryan

JamesHuttonInstitute

JamesHuttonInstitute

36PUBLICATIONS1,099CITATIONS

200PUBLICATIONS5,263CITATIONS

SEEPROFILE

SEEPROFILE

MarkOwenWinfield

SteveMillam

UniversityofBristol

52PUBLICATIONS1,042CITATIONS

43PUBLICATIONS2,142CITATIONS

SEEPROFILE

SEEPROFILE

Allin-textreferencesunderlinedinbluearelinkedtopublicationsonResearchGate,
lettingyouaccessandreadthemimmediately.

Availablefrom:SanjeevSharma
Retrievedon:27October2016


Planta (2007) 226:1449–1458
DOI 10.1007/s00425-007-0583-2

O R I G I N A L A R T I CL E

Stability of potato (Solanum tuberosum L.) plants regenerated

via somatic embryos, axillary bud proliferated shoots,
microtubers and true potato seeds: a comparative phenotypic,
cytogenetic and molecular assessment
Sanjeev Kumar Sharma · Glenn J. Bryan ·
Mark O. WinWeld · Steve Millam

Received: 22 January 2007 / Accepted: 26 June 2007 / Published online: 1 August 2007
© Springer-Verlag 2007

Abstract The stability, both genetic and phenotypic, of
potato (Solanum tuberosum L.) cultivar Desiree plants
derived from alternative propagation methodologies has
been compared. Plants obtained through three clonal propagation routes—axillary-bud-proliferation, microtuberisation and a novel somatic embryogenesis system, and
through true potato seeds (TPS) produced by selWng were
evaluated at three levels: gross phenotype and minituber
yield, changes in ploidy (measured by Xow cytometry) and
by molecular marker analysis [measured using AFLP
(ampliWed fragment length polymorphism)]. The clonally
propagated plants exhibited no phenotypic variation while
the TPS-derived plants showed obvious phenotypic
segregation. SigniWcant diVerences were observed with
respect to minituber yield while average plant height, at the
time of harvesting, was not signiWcantly diVerent among
plants propagated through four diVerent routes. None of the
plant types varied with respect to gross genome constitution
as assessed by Xow cytometry. However, a very low level
of AFLP marker proWle variation was seen amongst
the somatic embryo (3 out of 451 bands) and microtuber
(2 out of 451 bands) derived plants. Intriguingly, only
AFLP markers generated using methylation sensitive

restriction enzymes were found to show polymorphism. No

polymorphism was observed in plants regenerated through
axillary-bud-proliferation. The low level of molecular
variation observed could be signiWcant on a genome-wide
scale, and is discussed in the context of possible methylation changes occurring during the process of somatic
embryogenesis.
Keywords AFLP · Clonal propagation · Methylation ·
Polymorphism · Potato (Solanum tuberosum L.) · Somatic
embryogenesis · Stability · Synthetic seeds
Abbreviations
2,4-D 2,4-Dichlorophenoxyacetic acid
ABP
Axillary bud shoot plants
AFLP AmpliWed fragment length polymorphism
DAPI 4Ј-6-Diamidino-2-phenylindole
EMB Embling(s)
ISSR
Inter simple sequence repeat
MS
Murashige and Skoog (1962)
MTP
Microtuber plants
NAA Naphthalene acetic acid
TPS
True potato seedling(s)
RAPD Random ampliWed polymorphic DNA
RFLP Restriction fragment length polymorphism

Introduction

S. K. Sharma · G. J. Bryan (&) · M. O. WinWeld · S. Millam
Genetics Programme, Scottish Crop Research Institute,
Invergowrie, Dundee DD2 5DA, Scotland, UK
e-mail:
Present Address:
S. Millam
Institute of Molecular Plant Sciences, Daniel Rutherford Building,
The University of Edinburgh, The Kings Buildings,
MayWeld Road, Edinburgh EH9 3JH, Scotland, UK

Potato (Solanum tuberosum L.), the most important noncereal food crop in the world, is a polyploid outbreeder
which maintains a high degree of heterozygosity. Consequently, tetraploid commercial cultivars are propagated
vegetatively in order to maintain phenotypic characteristics
and economically important traits. Generally, propagation
is via tubers from the previous crop, but alternative clonal

123


1450

propagation methods have been developed to accommodate
emerging economic production requirements. Tissue culture techniques, mainly micropropagation, have substantially augmented the supply of planting material, and
reduced the time required for the release of new cultivars
from more than 10 to 3–5 years (Struik and Wiersema
1999). Clonal in vitro propagation methods not only maintain the uniformity among oVspring, but also preserve
health status as the planting material has reduced exposure
to soil-borne and other diseases.
Micropropagation techniques in potato are mainly
employed to allow rapid mass propagation of nuclear stock

material, usually in the form of microtubers and axillary
bud shoot plants. Minitubers produced in initial cycles of
multiplication from the in vitro material increase the number of propagules in the Wrst year, and bulks up the prebasic seed required for large-scale production of basic seed
tuber material, which initiates the seed production programme. The multiplication rate in microtuber and axillary
bud based micropropagation systems depends upon the
number of pre-formed meristems present in explants during
each multiplication cycle; inherently, these numbers are
low rendering such systems comparatively resource ineYcient. Somatic embryogenesis, however, oVers the possibility of a novel system for the large-scale multiplication of
plant material and could be more economical because of the
greater multiplication rates, exploiting the potential of
every cell to regenerate into a complete plant. However, in
potato, somatic embryogenesis requires further research
before it can become a feasible option (Naik et al. 2000).
Apart from the developmental aspects of somatic
embryogenesis, successful Weld testing of established plants
is crucial for the success of the system. In addition to evaluating the survival potential and growth performance of
plants derived from somatic embryogenesis, it is vital to
monitor their uniformity. The variation observed from vegetative propagation, either genetic or epigenetic in origin, is
commonly referred to as somaclonal variation (Larkin and
Scowcroft 1981). It is of fundamental importance that
micropropagated plants, irrespective of their development
through either organogenesis or embryogenesis, remain
true-to-parental type. Over the last two decades, the phenomenon of somaclonal variation has been extensively
investigated using cytogenetic, biochemical and molecular
methods (e.g. Karp 1995; Vazquez 2001). Though the
underlying mechanisms that generate somaclonal variation
remain unclear, major contributory factors are thought to be
disorganised meristematic growth, the genetic make-up
(ploidy and genotype) of the stock material, the type and
concentration of plant growth regulators in the culture

medium and explant source (Karp 1991). At the molecular
level, it has been postulated that the tissue culture-induced
stress that cells undergo in the presence of plant growth

123

Planta (2007) 226:1449–1458

regulators could induce alteration(s) in sensitive regions of
the plant genome (Sala et al. 1999), which can be attributed
to DNA methylation, ampliWcation or the activation of
transposable elements (Brar and Jain 1998). In some
instances, somaclonal variation has been exploited to obtain
variant genotypes with new and desirable characteristics,
such as for salt tolerance in potato (Tal 1996; Ochatt et al.
1998). However, somaclonal variation is highly undesirable
where clonal propagation is the goal (Cassells and Curry
2001).
Until recently, measures of whether plants were true to
parental type were rather coarse, being based on phenotypic observations and/or cytogenetic characterisation.
Recent developments in molecular genetic techniques
have made it possible to ‘proWle’ the genome looking,
directly or indirectly, for changes in DNA sequence.
Thus, in principle, single base changes and even methylation events can be observed and interpreted as instances of
tissue-culture-induced somaclonal variation. However, at
its extreme limit, such an approach would involve
sequencing the entire genome, an infeasible task. Highly
multiplex molecular marker techniques [e.g. ampliWed
fragment length polymorphism (AFLP)] are potential
tools for the assessment of uniformity, notwithstanding

the inherent limitations of any genome proWling methodology.
Somatic embryogenesis in potato oVers a potentially
novel method for producing nuclear seed material as compared to the existing two other forms of micropropagation
viz. by production of microtubers and axillary bud shoot
plants. The year round production and utilisation of
microtubers is hampered by their dormancy and constraints
related to physiological age. Axillary bud shoot plants,
though without the constraints that are encountered with
microtubers, are fragile and production, in common with
microtubers, is labour intensive and requires large amounts
of laboratory space. In contrast, somatic embryos are free
from dormancy and constraints related to physiological age,
and require signiWcantly less manual labour and laboratory
space. Thus, encapsulated somatic embryos (synthetic
seeds) hold the potential to overcome the disadvantages of
conventional micropropagation systems, and combine the
advantages of clonal and seed propagation systems. However, in order to conWrm whether or not somatic embryogenesis is a reliable clonal method of propagation in potato,
a Wdelity and yield assessment of plants regenerated
through emblings is required.
Somatic embryogenesis has been previously reported in
potato (de Garcia and Martinez 1995; Seabrook and
Douglass 2001; JayaSree et al. 2001; Vargas et al. 2005;
Sharma and Millam 2004), but post-embryogenesis events,
including the assessment of uniformity, have not been studied in detail. In this investigation, potato cv. Desiree plants


Planta (2007) 226:1449–1458

1451

obtained from four propagation methods—somatic
embryogenesis, axillary bud proliferation, microtubers and
true potato seeds (TPS) obtained by selWng—were tested
for stability and uniformity using phenotypic, cytogenetic
and molecular approaches.

Materials and methods
Plant material
In vitro cultures of S. tuberosum L. cultivar Desiree were
obtained from the Scottish Agricultural Science Agency,
Edinburgh, Scotland. All of the cultures, unless otherwise
stated, were maintained on basal MS (Murashige and
Skoog 1962) medium and incubated under controlled environmental conditions of 19 § 1°C, 16/8 h light/dark cycles
and 90 mol m¡2 s¡1 photon Xux density (400–700 nm).
Development of potato somatic embryos was as described
previously (Sharma and Millam 2004). For experimental
plant material, 3-week-old emblings (embling—a synonym to seedling for a plant obtained via a somatic embryo)
were obtained from encapsulated somatic embryos
(Sharma 2006), axillary bud shoot plants were obtained
from 3-week-old in vitro rooted single-node-cuttings multiplied by axillary bud proliferation, microtuber plants
were produced by sprouting microtubers and seedling
plants were obtained by germinating true potato seeds
obtained by selWng cv. Desiree plants. TPS were highly
uniform in their appearance in contrast to their obvious
heterozygous genetic make-up. For propagules following
vegetative modes of propagation, care was taken to keep
their size range (within their respective categories) as
uniform as possible. For emblings, only somatic embryos
measuring between 1 and 2 mm were used for
encapsulation, axillary bud shoot plants were generated

using single-node-cuttings between 1.5 and 2.0 cm in
length and microtuber plants were obtained using microtubers having 7–9 mm diameter. Once all the plants were
established for transplantation, no screening for any
noticeable variation was performed in order to achieve a
non-biased progression of plant growth and development.
Figure 1 shows the four comparative categories of propagules used in the current study.
Plants obtained through the four propagation routes were
established under glasshouse conditions in a compost mix
of 24:2:1 (by weight) peat, sand and perlite supplemented
with fertiliser and Celcote water retaining gel. The propagules (75 plants) for each category were grown in randomised blocks. Plants were inspected for any noticeable
variation after 30, 60 and 90 days of transplantation. Plant
height measurements were taken at the time of haulm (foliage) cutting (110 days after transplantation). Progeny

Fig. 1 A depiction of four comparative potato planting propagule
types. a Synthetic seeds (encapsulated somatic embryos). b Axillary
bud shoot plants. c Microtubers. d True potato seeds. Bar = 5 mm (a,
c), 10 mm (b, d)

tubers (minitubers) were left in the pots for two more weeks
after haulm cutting, following which the tubers were harvested and used for biometric analysis.
Cytogenetic analysis
Representative leaf samples (obtained from apical buds
only) for all four-plant propagation categories (15 plants
each) were collected from their respective 45-day-old
plants. Small pieces of leaf material (»40 mg) were

123


1452

Planta (2007) 226:1449–1458

macerated, using a sharp razor blade, in 2 ml of an
ice-cold DAPI (4Ј-6-diamidino-2-phenylindole) based
nuclear extraction buVer (Arumuganathan and Earle
1991). Following this, the buVer containing cell constituents and large tissue remnants was passed through 40 m
nylon Wlters and, after 15 min of incubation, the solution
containing stained nuclei was analysed in a Xowcytometer (CyFlow ML, Partec, Germany). Together with each
leaf sample an internal standard (»10 mg of Ilex crenata
‘Fastigiata’ leaf tissue, nuclear DNA content = 2.16 pg/
2°C) was also included and its G0/G1 peak (2°C) was
adjusted to around channel 215 set on a linear scale of
Xuorescence intensity (FL2-DAPI). The axillary bud
shoot plants were used as tetraploid (2n = 4x) controls for
ploidy comparisons. Histograms were constructed using
Flomax version 2.4 d (Partec) software. For each plant,
DNA-ratios were obtained by dividing the mean of the
dominant (G0/G1) peak of the potato sample by the mean
of the G0/G1 peak of the internal standard and ANOVA
was performed.
DNA extraction
Genomic DNA from representative leaf samples (from
every 1st node) of 45-day-old plants from all four-plant
groups was extracted, according to manufacturer’s instructions, using DNeasy Plant Mini Kits (Qiagen, Hilden,
Germany). DNA concentration of each plant sample was
quantiWed using a Nanodrop® Microphotometer (Nanodrop
Technologies, Wilmington, Delaware, USA) by measuring
its absorbance at 260 nm.

Table 1 Details of AFLP primer combinations used and the
corresponding numbers of bands
observed. EMB, emblings; ABP,
axillary bud shoot plants; MTP,
microtuber plants; TPS, true
potato seedlings

Primer
combination

AFLP analysis
AmpliWed fragment length polymorphism (AFLP) assays
were performed using a modiWcation of the protocol of Vos
et al. (1995), as described in Bryan et al. (2002). The 6-bp
cutting restriction enzymes PstI and EcoRI were used in
combination with the 4-bp cutting restriction enzyme MseI.
Table 1 contains primer sequences used. For separation of
labelled fragments, a 3.5 l aliquot of each ampliWcation
product was electrophoresed in 5% denaturing acrylamide
gels which, after drying, were exposed to X-ray Wlm
(Biomax MS, Kodak). Exposed X-ray Wlms were developed through an automatic X-ray Wlm processor (XO Graph
Imaging Systems, Compact X4) and used for AFLP
analysis.
Experimental design and statistical analysis
Data storage and calculations were done using the Genstat
7 statistical package (Payne et al. 1993). The data were
analysed by analysis of variance (ANOVA) and the individual group means were ranked by comparing their mean
diVerences against LSD (least signiWcant diVerence). AFLP
was performed using representative plants from emblings
(EMB, 15 plants), axillary bud shoot plants (ABP, 15

plants), microtuber plants (MTP, 5 plants) and true potato
seedlings (TPS, 5 or 15 plants). In order to facilitate the
intra-group comparisons of plants and keep track of individual plants, the order of plant samples in their respective
groups was kept constant across all gels irrespective of the
relative position of diVerent plant groups, with respect to

Primer sequence (5Ј to 3Ј)

Total number
of bandsa

Number of polymorphic bands
EMB

ABP

MTP

TPS

67

1

0

1

12

66

0

0

0

16

94

2

0

1

19

82

0

0

0

22

76

0

0

0

23

PstI/MseI combination
P12

GACTGCGTACATGCAG AC

M38

GATGAGTCCTGAGTAA ACT

P13

GACTGCGTACATGCAG AG

M36

GATGAGTCCTGAGTAA ACC

P14

GACTGCGTACATGCAG AT

M32

GATGAGTCCTGAGTAA AAC

P15

GACTGCGTACATGCAG CA

M41

GATGAGTCCTGAGTAA AGG

EcoRI/MseI combination

a

Size range ¡150 to 700 bp

123

E32

GACTGCGTACCAATTC AAC

M51

GATGAGTCCTGAGTAA CCA

E35

GACTGCGTACCAATTC ACA

M48

GATGAGTCCTGAGTAA CAC

Total

66

0

0

0

21

451

3

0

2

113

Planta (2007) 226:1449–1458

1453

each other. Variation between samples was observed as the
presence of polymorphic bands.

Results
Phenotypic analysis and assessment of minituber yield
No morphological diVerences were observed among clonally propagated plants (plants derived from axillary bud
shoot plants, microtuber plants and emblings) and all tubers
showed normal morphology (Fig. 2a–c). Initially, embling
derived plants showed uneven vegetative growth compared
to plants derived from axillary bud shoot plants and microtuber plants. However, over a period of 30 days, these
diVerences in growth pattern among the emblings disappeared. The seedling-derived plants were variable in
growth habit and, additionally, showed delayed Xowering
(Fig. 2d) and maturity compared to the other plant types
(Fig. 2a–c). No signiWcant diVerences (P = 0.492) in average plant height, taken at the time of harvesting (110 days
after transplantation), were observed between any of the

Fig. 2 Overview showing fully grown plant populations, inXorescence and minitubers produced from potato cv. Desiree plants derived
from somatic embryogenesis (a), axillary bud proliferation (b),

four-plant groups (Table 2). Nevertheless, as compared to
seedling-derived plants, the clonally propagated plants
were more uniform and displayed less variation both within
and among individual plant groups. As expected, seedlingderived plants were variable in height and ranged from
much shorter (88 cm) to taller (153 cm) plants with an average height of 127.3 (SE 5.1) cm at the time of haulm
destruction (110 days).
The minitubers were harvested 2 weeks after the cutting

of the haulm and data on tuber number and weight were
collected. The mean tuber number per plant was signiWcantly diVerent (P < 0.001) among all plant types except
microtuber plants versus seedling-derived plants. The axillary bud shoot plants yielded the highest average number of
tubers per plant followed by embling-derived plants while
the average per plant tuber number of microtuber plants and
seedling plants was signiWcantly lower (Table 2). While
average tuber weight was signiWcantly higher (P < 0.003)
in microtuber plants and embling-derived plants as compared to axillary bud shoot plants and seedling plants, the
tuber yield per plant was signiWcantly higher (P < 0.001) in
axillary bud shoot plants and embling-derived plants as

microtubers (c), and Desiree-selfed seedlings (d). Bar = 5 cm (for
minitubers only)

Table 2 Biometric analysis of potato plants propagated through four diVerent propagation routes
Plants obtained
from

Mean plant
height (cm)

Mean tuber
number

Mean tuber
weight (gm)

Tuber yield
per plant (gm)

Tuber skin
colour

Emblings

131.5a (1.4)

19.5b (1.1)

31.6a (1.6)

615.0a (31.6)

Reddish

Axillary bud shoots

132.2a (1.1)

24.1a (1.3)

26.8b (1.5)

645.5a (24.8)

Reddish

Microtubers

a

135.4 (0.9)

14.8 (1.6)

True potato seeds

127.3a (5.1)

13.5c (1.1)
<0.001

<0.003

P value

0.492

c

a

35.4 (3.4)

b

524.5 (23.3)

Reddish

25.5b (1.9)

343.5c (28.1)

Variable

<0.001

*

Plant height was taken at the time of haulm cutting (110 days after transplantation) while the tubers were harvested and analysed 2-week after
haulm cutting. n = 15, values in bracket represent SE. Means having same superscripts are not signiWcantly diVerent at the deWned P value for the
respective column/variate category
* Qualitative assessment by visual inspection

123


1454

compared to microtuber and seedling plants (Table 2).
While the tuber shape was maintained in progeny tubers
(minitubers) of all four-plant types, the tuber skin colour in
seedling-derived plants ranged from lighter to darker
shades of a typical cv. Desiree reddish colour (Fig. 2).
Cytogenetic analysis
The ploidy level of plants used in the study was characterised by analysing small pieces of leaves of the representative plants from all four propagation routes using Xow
cytometry. The respective DNA histograms were obtained
and peaks were analysed (Fig. 3a–d). No signiWcant diVerences (P = 0.887) in DNA-ratio, obtained as described in
“Materials and methods”, were observed within and among

ABP, MTP, EMB and TPS plant categories and the respective average DNA-ratios obtained were 1.809 (SE 0.006),
1.811 (SE 0.005), 1.807 (SE 0.006) and 1.805 (SE 0.006).
The DNA-ratio (»1.81), channel position (»390) of the
G0/G1 peak and the total number (2) of peaks of the EMB,
MTP and TPS plant samples corresponded to that of ABP
plants (taken as 2n = 4x tetraploid control) suggesting that
the somatic embryogenesis induced-emblings and other

Fig. 3 Representative Xow-cytometric histograms of DAPI stained
nuclei isolated from leaf tissue of S. tuberosum cv. Desiree plants
originated from four comparative propagation routes. a Axillary bud
shoot plant used as control for the nuclear DNA content of a tetraploid.
b–d Microtuber, embling and true potato seedling plants, respectively,
Fig. 4 AFLP proWles involving
the use of methylation
insensitive enzymes EcoRI and
MseI with primer-pair
combination E32/M51 (a) and
E35/M48 (b). ProWles of EMB,
ABP and MTP were
monomorphic, while proWle of
TPS was polymorphic at various
loci. L Ladder; EMB, ABP,
MTP and TPS stands for the
plants regenerated through
emblings, axillary bud shoot
plants, microtuber plants and
true potato seedlings,
respectively

123

Planta (2007) 226:1449–1458

plant types had maintained their gross cytological stability
and (tetra) ploidy level.
Molecular analysis
The AFLP analysis involved six primer combinations; two
based on the use of EcoRI (a methylation insensitive
restriction enzyme), and four based on the use of PstI
(which is methylation sensitive). The number of clearly
separated fragments for the primer pairs E32/M51, E35/
M48, P12/M38, P13/M36, P14/M32 and P15/M41 was 76,
66, 67, 66, 94 and 82, respectively. No AFLP polymorphism was detected among the clonally propagated plants
using the methylation insensitive (i.e. EcoRI–MseI) primer
combinations (Fig. 4a, b). In contrast, when the methylation
sensitive enzyme PstI was used in the protocol, a small number of polymorphic fragments were identiWed (Fig. 5a, b).
Out of a total of 67 bands, the primer-pair P12/M38
detected polymorphism at one locus in both embling
(Fig. 5c; arrow) and microtuber plants (Fig. 5c; arrowhead).
Out of a total of 94 bands, primer-pair P14/M32 revealed
polymorphism at two and one loci in embling (Fig. 5d, e;
arrows) and microtuber plants (Fig. 5e; arrowhead),

with the same nuclear DNA content and ploidy as control. Counts on
y-axis represent number (£100) of nuclei. G0/G1 and G2 corresponds
to the phases of the cell cycle. IS represents G0/G1 peak of an internal
standard

Planta (2007) 226:1449–1458

1455

Fig. 5 AFLP proWles involving the use of methylation sensitive
enzyme PstI and methylation insensitive enzyme MseI with primerpair combination P12/M38 (a) and P14/M32 (b). Primer pair P12/M38
detected polymorphism at one locus each in EMB and MTP plant
categories (c), while P14/M32 revealed polymorphism for EMB and

MTP at two (d, e) and one (e) loci, respectively. Polymorphism for
EMB and MTP is depicted by white arrows and arrowhead, respectively. L Ladder; EMB, ABP, MTP and TPS stands for the plants
regenerated through emblings, axillary bud shoot plants, microtuber
plants and true potato seedlings, respectively

respectively. The primer-pairs P13/M36 and P15/M41 did
not reveal any polymorphism. Axillary bud shoot plants
were monomorphic for all six-primer combinations
(Figs. 4, 5). In the case of the P12/M38 gel (Fig. 5c), the
single polymorphism is represented by the absence of a
band in somaclonal variants (EMB-1, 3–8 and 15; MTP-2
and 5) with respect to the parental genotype. This fragment
was also absent from the seedling TPS-1. In the P14/M32
gel (Fig. 5e), a small fragment (»174 bp) was found to be
absent among EMB-1, 3–8 and 15, the same samples that
show polymorphism on the P12/M38 gel. Interestingly, in
the aVected embling-derived plants (EMB) a new larger
fragment (»243 bp), not present in any of the other plants,
was observed (Fig. 5d). As expected, seedling-raised plants
displayed polymorphisms regardless of the types of restriction enzyme used in the reaction: percentage polymorphism
was 30, 32, 18, 24, 20 and 27% (overall mean of 25%)

using primer-pairs E32/M51, E35/M48, P12/M38, P13/
M36, P14/M32 and P15/M41, respectively (Figs. 4, 5).
Table 1 summarises the number of AFLP bands obtained
among four-plant categories with respect to the diVerent
sets of primer combinations used.

JayaSree et al. 2001; Vargas et al. 2005) or only presented
in superWcial detail (Fiegert et al. 2000; Schafer-Menuhr
et al. 2003). Seabrook and Douglass (2001) reported the
presence of some ‘oV-type’ tubers from somatic embryogenesis-derived plants, without giving further detail. In the
current study, no variability at the phenotypic level was
observed among clonally propagated plants. There was no
signiWcant diVerence in mean plant height among the plant
groups studied. The axillary bud shoot plants produced the
maximum average number of tubers per plant followed by
embling-derived plants. Minituber production in microtuber plants and seedling plants was signiWcantly lower than
the other two-plant categories. For ‘mean tuber weight’ and
‘tuber yield per plant’ categories, the emblings were
grouped with microtuber plants and axillary bud shoot
plants, respectively. Thus, the growth and yield performance of embling plants, in addition to other potential beneWts, was comparable to the other two clonal forms of
micropropagation.

Discussion
Gross phenotypic analysis and assessment of minituber
yield
In studies of somatic embryogenesis in potato reported to
date, details of plant transfer and uniformity assessment are
either entirely lacking (de Garcia and Martinez 1995;

Cytogenetic analysis of variability

Gross genome stability in tissue culture derived plants has
been studied at the cytogenetic level in other species (Shoyama et al. 1995; Zoriniants et al. 2003). Regarding stability
of embling derived plants, cytogenetic studies have
revealed contrasting observations. Odake et al. (1993)
reported chromosome doubling (from diploid to tetraploid)
in 66.7% and 100% emblings of Asparagus oYcinalis L.
obtained from Gellan Gum-solidiWed medium and liquid
medium, respectively. In contrast, Mamiya et al. (2001)
reported there to be no ploidy changes during somatic

123


1456

embryogenesis in A. oYcinalis. Synthetic auxins, such as
2,4-D (2,4 dichlorophenoxyacetic acid) and NAA (naphthalene acetic acid), used in culture media have been reported
to be associated with somaclonal variation (Karp 1989;
Phillips et al. 1994). Indeed, a reduction in ploidy was
reported in both carrot (Ronchi et al. 1992) and poplar
(Rugh et al. 1993) somatic embryogenesis systems in
which 2,4-D had been used. In contrast, in the current
study, gross DNA content of cells, as determined by Xow
cytometry, remained unaVected. Thus, the original ploidy
level was probably maintained in the emblings and the
plants derived from them. This would suggest that, if 2,4-D
causes changes in ploidy, the doses used in our culture
medium were below harmful levels.
Molecular genetic assessment of uniformity
For the induction of embryogenesis in somatic cells, their

existing gene expression patterns would need to be altered
or replaced with an embryogenic programme. One possible
method for the down-regulation of existing gene expression
is by DNA methylation, which in turn is inXuenced by
endogenous auxins (Lo Schiavo et al. 1989). For any stimulus to become capable of evoking somatic embryogenesis it
must be able to alter, mostly increase, the endogenous auxin
levels and consequently DNA methylation (Lo Schiavo et al.
1989). Under this hypothesis, DNA methylation would
appear to be an indispensable process for initiating somatic
embryogenesis but one that, at times, has the undesirable
consequence of leading to somaclonal variation. Various
molecular approaches such as AFLP, RAPD (rapid ampliWed polymorphic DNA), RFLP (restriction fragment length
polymorphism) have been attempted to identify and measure the level of somaclonal variation in tissue culture
derived plants (Devarumath et al. 2002; Martins et al. 2004;
Sanchez-Teyer et al. 2003; Hale and Miller 2005). Irrespective of the methodology used, however, only a very small
percentage (much less than 1%) of the genome can be
assayed, however proliWc the technique in terms of number
of loci sampled. Of the various techniques mentioned,
AFLP is the most highly multiplex with typically 50–100
loci assayed per primer pair. These loci are thought to be
scattered more or less randomly throughout the genome and
thus AFLPs oVer, perhaps, the best chance for detecting tissue culture-induced changes. Moreover, AFLP is also one
of the more robust molecular techniques for cultivar identiWcation and variability analysis (Hale and Miller 2005),
hence its use in this study.
We detect a very low level of AFLP polymorphism
amongst embling regenerated plants with 3 of 451 AFLP
fragments showing presence/absence polymorphisms. All
of the observed variability pertains to fragments generated
using PstI/MseI, and this raises obvious questions about

123

Planta (2007) 226:1449–1458

whether methylation polymorphisms may be responsible
for the observed variation. For example, with respect to the
polymorphic P14/M32 fragment, the most parsimonious
explanation would be that methylation has eVectively
‘knocked-out’ the PstI restriction site deWning one end of
the »174 bp fragment (Fig. 5e) and that the next PstI site,
approximately 70 bp away, has remained unmethylated.
Therefore, the 174 bp fragment has disappeared and been
replaced by a 243 bp fragment (Fig. 5d). The very small
70 bp fragment was not observable on the gel. The two
AFLP fragments that are variable among emblings are both
present in Desiree and polymorphic amongst Desiree seedlings, suggesting that these markers are present in the ‘simplex’ conWguration in the parent clone. It is possible that
such ‘single-dose’ AFLP markers are more susceptible to
the eVects of methylation or other phenomena. Whatever
the explanation, the fact that several plants show the same
mutation begs another question: how do we explain the
presence of the same mutation in several independent samples? It may be that the genomic region containing the PstI
site in question is in a ‘hotspot’ for methylation or for some
other mutational process. Alternatively, the somatic
embryos that gave rise to the regenerated somaclonal variants all had their origin in a single cell that had undergone
the said variation prior to giving rise to a segment of daughter cells that in turn became embryogenic. However, MTP2 and 5, and TPS-1 also show the absence of the same P12/
M38 fragment (Fig. 5c). An increase in the methylation status during tissue culture has also been reported previously
for tomato callus, as compared to leaves (Smulders et al.
1995) and tissue culture regenerants of pea (Cecchini et al.
1992). This has been the Wrst study to evaluate the applicability of AFLP markers in establishing the clonal Wdelity of
potato plants raised by somatic embryogenesis, and the use

of methylation-sensitive enzymes has allowed the identiWcation of some tissue culture-induced polymorphism in
potato emblings that has not been previously reported.
Somaclonal variation as aVected by diVerent modes
of propagation
Axillary-branching is regarded as the plant micropropagation system with the lowest risk of generating genetic instability, because the presence of preformed and organised
meristem makes them less susceptible to the genetic variation that is more likely to occur during cell division or diVerentiation under in vitro conditions (Shenoy and Vasil 1992).
Nevertheless, the production of oV-types has been previously reported in conventional tissue culture, as well as
plants derived from somatic embryogenesis. The maintenance of genetic integrity among axillary bud-propagated
cassava plantlets was reported using RAPD and ISSR (Inter
simple sequence repeat) markers (Martins et al. 2004), while


Planta (2007) 226:1449–1458

Devarumath et al. (2002), using RAPD, ISSR and RFLP
Wngerprints, reported subtle genetic variation at the DNA
level in micropropagated tea plants derived from organised
meristems. In this study, no polymorphism was detected
among the AFLP proWles of axillary bud-propagated potato
plants. Axillary bud propagation proceeds through a regular
ontogenic progression, without involving any stressful dediVerentiation phase which may be likely to induce mutations. In contrast, the other two tissue culture processes,
somatic embryogenesis and microtuberisation, are stressinduced pathways, involving the formation of natural structures (embryos and tubers) in an atypical and artiWcial manner
compared to their corresponding in vivo pathways. The production of somatic embryos was observed at callused
explant ends (indirect somatic embryogenesis) as well as at
the non-callused longitudinal surface of the explant (direct
somatic embryogenesis; Sharma and Millam 2004). Published results (Henry et al. 1998) suggest that the indirect
mode of somatic embryogenesis is more likely to inXuence
the maintenance of genetic stability among regenerants.
Since the emblings transferred to glasshouse for variability
assessment were not itemised with regard to their origin

from direct or indirect modes, the sampled embling population may possibly have been of mixed embryo origin. This
may explain the presence of two types of plant groups
within emblings, one showing polymorphism and the other
not—the former probably originating from indirect mode of
somatic embryogenesis. In support of this argument,
Sanchez-Teyer et al. (2003) using coVee plants, observed
discrepancies between direct and indirect types of somatic
embryogenesis with regard to DNA re-arrangements, and
detected speciWc bands for each type of somatic embryogenesis. Since the callus phase during potato somatic embryogenesis was markedly less pronounced, even the
polymorphism between emblings and other modes of vegetative propagation was negligible. Moreover, as supported
by corresponding cytogenetic and phenotypic data, these
molecular aberrations did not aVect the agronomic performance, and the ploidy level was maintained.
The overall level of polymorphism in somatic emblings
was three polymorphic bands out of 451 total bands and is
comparable to the two polymorphic bands out of 451 total
bands detected in microtuber plants, which is an accepted
and widely used form of micropropagation and germplasm
exchange. This strongly indicates that the level of variation
observed in somatic embryogenesis-derived plants should
not be an obstacle for the further uptake of this novel form
of propagation. However, the mechanism for this low level
of genetic variability remains an interesting and, as yet,
unresolved question. Future studies will attempt to assess
the potential role of methylation and genetic or other epigenetic diVerences among the regenerated plants, rather than
due to changes in DNA sequence.

1457
Acknowledgments SKS is grateful to the Government of India and
the Commonwealth Scholarship Commission, United Kingdom for his
doctoral Commonwealth Scholarship Award. The authors thank GeoV

Swan (SCRI) for providing the potato cv. Desiree true potato (selfed)
seeds. The authors are grateful to Drs Brian Forster and Gavin Ramsay
(SCRI) for critical reading of the manuscript. SCRI is supported by
grant-in-aid from the Scottish Executive Environment & Rural AVairs
Department (SEERAD).

References
Arumuganathan K, Earle ED (1991) Estimation of nuclear DNA content of plants by Xow cytometry. Plant Mol Biol Rep 9:229–233
Brar DS, Jain SM (1998) Somaclonal variation: mechanisms and applications in crop improvement. In: Jain SM, Brar DS, Ahloowalia
BS (eds) Somaclonal variation and induced mutations in crop
improvement. Kluwer, Boston, pp 17–37
Bryan GJ, McLean K, Bradshaw JE, Phillips M, Castelli L, De Jong
WS, Waugh R (2002) Mapping QTL for resistance to the cyst
nematode Globodera pallida derived from the wild potato species
Solanum vernei. Theor Appl Genet 105:68–77
Cassells AC, Curry RF (2001) Oxidative stress and physiological, epigenetic and genetic variability in plant tissue culture: implications
for micropropagators and genetic engineers. Plant Cell Tiss Org
64:145–157
Cecchini E, Natali L, Cavallini A, Durante M (1992) DNA variations
in regenerated plants of pea (Pisum sativum L.). Theor Appl Genet 84:874–879
de Garcia E, Martinez S (1995) Somatic embryogenesis in Solanum tuberosum L. cv. Desiree from stem nodal sections. J Plant Physiol
145:526–530
Devarumath RM, Nandy S, Rani V, Marimuthu S, Muraleedharan N,
Raina SN (2002) RAPD, ISSR and RFLP Wngerprints as useful
markers to evaluate genetic integrity of micropropagated plants of
three diploid and triploid elite tea clones representing Camellia
sinensis (China type) and C. assamica ssp. assamica (AssamIndia type). Plant Cell Rep 21:166–173
Fiegert AK, Mix-Wagner G, Vorlop KD (2000) Regeneration of Solanum tuberosum L. cv. Tomensa: induction of somatic embryogenesis in liquid culture for the production of “artiWcial seed”.
Landbauforsch Volk 50:199–202
Hale AL, Miller JC (2005) Suitability of AFLP and microsatellite

marker analysis for discriminating intraclonal variants of the potato cultivar Russet Norkotah. J Am Soc Hortic Sci 130:624–630
Henry RJ, Nato A, de Buyser J (1998) Genetic Wdelity of plants regenerated from somatic embryos of cereals. In: Jain SM, Brar DS,
Ahloowalia BS (eds) Somaclonal variation and induced mutations in crop improvement. Kluwer, Boston, pp 17–37
JayaSree T, Pavan U, Ramesh M, Rao AV, Reddy KJM, Sadanandam
A (2001) Somatic embryogenesis from leaf cultures of potato.
Plant Cell Tiss Org 64:13–17
Karp A (1989) Can genetic instability be controlled in plant tissue cultures? Int Assoc Plant Tiss Cult Newsl 58:2–11
Karp A (1991) On the current understanding of somaclonal variation.
In: MiXin BJ (ed) Oxford surveys of plant molecular and cell biology. Oxford University Press, Oxford, pp 1–58
Karp A (1995) Somaclonal variation as a tool for crop improvement.
Euphytica 85:295–302
Larkin PJ, Scowcroft WR (1981) Somaclonal variation—a novel
source of variability from cell cultures for plant improvement.
Theor Appl Genet 60:197–214
Lo Schiavo F, Pitto L, Giuliano G, Torti G, Nutironchi V, Marazziti D,
Vergara R, Orselli S, Terzi M (1989) DNA methylation of
embryogenic carrot cell cultures and its variations as caused by

123


1458
mutation, diVerentiation, hormones and hypomethylating drugs.
Theor Appl Genet 77:325–331
Mamiya K, Sakamoto Y, Ohnishi N, Hirosawa T (2001) Synthetic
seeds of Asparagus oYcinalis L. In: Bhojwani SS, Soh W-Y (eds)
Current trends in the embryology of angiosperms. Kluwer,
Dordrecht, pp 337–352
Martins M, Sarmento D, Oliveira MM (2004) Genetic stability of micropropagated almond plantlets, as assessed by RAPD and ISSR
markers. Plant Cell Rep 23:492–496

Murashige T, Skoog F (1962) A revised medium for rapid growth and
bio assays with tobacco tissue cultures. Physiol Plant 15:473–497
Naik PS, Chakrabarti SK, Sarkar D, Birhman RK (2000) Potato biotechnology: Indian perspective. In: Khurana SMP, Shekhawat
GS, Singh BP, Pandey SK (eds) Proceedings of Global Conference on Potato, New Delhi, December 6–11, 1999: potato, global
research and development. Indian Potato Research Association,
Central Potato Research Institute, Shimla, pp 194–211
Ochatt SJ, Marconi PL, Radice S, Arnozis PA, Caso OH (1998) In vitro recurrent selection of potato: production and characterization
of salt tolerant cell lines and plants. Plant Cell Tiss Org 55:1–8
Odake Y, Udagawa A, Saga H, Mii M (1993) Somatic embryogenesis
of tetraploid plants from internodal segments of a diploid cultivar
of Asparagus oYcinalis L. grown in liquid culture. Plant Sci
94:173–177
Payne RW, Lane PW, Digby PGN, Harding SA, Leech PK, Morgan
GW, Todd AD, Thompson R, TunnicliVe Wilson G, Welham SJ,
White RP (1993) Genstat 5 Release 3: reference manual. Oxford
University Press, Oxford
Phillips RL, Kaeppler SM, Olhoft P (1994) Genetic instability of plant
tissue cultures—breakdown of normal controls. Proc Natl Acad
Sci USA 91:5222–5226
Ronchi VN, Giorgetti L, Tonelli M, Martini G (1992) Ploidy reduction
and genome segregation in cultured carrot cell lines. 1. Prophase
chromosome reduction. Plant Cell Tiss Org 30:107–114
Rugh CL, Parrott WA, Merkle SA (1993) Ploidy variation in embryogenic yellow poplar. In: Proceedings of 22nd Southern Forest
Tree Improvement Conference, pp 493
Sala F, Arencibia A, Castiglione S, Christou P, Zheng Y, Han Y (1999)
Molecular and Weld analysis of somaclonal variation in transgenic
plants. In: Altman A, Ziv M, Izhar S (eds) Plant biotechnology
and in vitro biology in the 21st century. Kluwer, Dordrecht, pp
259–262
Sanchez-Teyer LF, Quiroz-Figueroa F, Loyola-Vargas V, Infante D

(2003) Culture induced variation in plants of CoVea arabica cv.
Caturra rojo, regenerated by direct and indirect somatic embryogenesis. Mol Biotechnol 23:107–115

123

Planta (2007) 226:1449–1458
Schafer-Menuhr A, Mix-Wagner G, Vorlop KD (2003) Regeneration
of plants from cell suspension cultures and encapsulated cell suspension cultures of Solanum tuberosum L. cv. Clarissa. Landbauforsch Volk 53:53–59
Seabrook JEA, Douglass LK (2001) Somatic embryogenesis on various potato tissues from a range of genotypes and ploidy levels.
Plant Cell Rep 20:175–182
Sharma SK (2006) The development of an eYcient somatic embryogenesis system for the production of synthetic seed in potato. University of Dundee, Dundee
Sharma SK, Millam S (2004) Somatic embryogenesis in Solanum tuberosum L.: a histological examination of key developmental
stages. Plant Cell Rep 23:115–119
Shenoy VB, Vasil IK (1992) Biochemical and molecular analysis of
plants derived from embryogenic tissue cultures of napier grass
(Pennisetum purpureum K. Schum). Theor Appl Genet 83:947–
955
Shoyama Y, Matsushita H, Zhu XX, Kishira H (1995) Somatic
embryogenesis in ginseng (Panax species). In: Bajaj YPS (ed)
Biotechnology in agriculture and forestry. Springer, Berlin, pp
344–356
Smulders MJM, Rus-Kortekaas W, Vosman B (1995) Tissue culture
induced DNA methylation polymorphisms in repetitive DNA of
tomato calli and regenerated plants. Theor Appl Genet 91:1257–
1264
Struik PC, Wiersema SG (1999) Seed potato technology. Wageningen
Pers, Wageningen
Tal M (1996) Somaclonal variation for salt tolerance in tomato and potato. In: Bajaj YPS (ed) Biotechnology in agriculture and forestry.
Springer, Berlin, pp 132–145
Vargas TE, De Garcia E, Oropeza M (2005) Somatic embryogenesis in

Solanum tuberosum from cell suspension cultures: histological
analysis and extracellular protein patterns. J Plant Physiol
162:449–456
Vazquez AM (2001) Insight into somaclonal variation. Plant Biosyst
135:57–62
Vos P, Hogers R, Bleeker M, Reijans M, Vandelee T, Hornes M, Frijters A, Pot J, Peleman J, Kuiper M, Zabeau M (1995) AFLP—a
new technique for DNA Wngerprinting. Nucleic Acids Res
23:4407–4414
Zoriniants SE, Nosov AV, Monforte-Gonzalez M, Mendes-Zeel M,
Loyola-Vargas VM (2003) Variation of nuclear DNA content
during somatic embryogenesis and plant regeneration of CoVea
arabica L. using cytophotometry. Plant Sci 164:141–146