Botanical Studies (2008) 49: 189-197.

*

Corresponding author: E-mail: xiech@mail.hzau.edu.cn;

Tel: +86-27-87282727; Fax: +86-27-87396057.

INTRODUCTION

Amorphophallus species, belonging to the family

Araceae, are perennial herbaceous plants, adapted to the

shady and mountainous areas and mainly distributed over

Southeast Asia and Africa (Gandawijaja, 1983). There

are 22 species grown in China, of which Amorphophallus

albus Liu & Wei is a native species and mainly cultivated

in southwest China (Long, 1998). Amorphophallus albus

is a traditional medicinal plant with a large subterranean

corm and the ability to lower blood cholesterol and sugar

levels, help with weight loss, and promote intestinal ac-

tivity and immune function (Zhang et al., 2005). The

corms of A. albus are important source of glucomannan,

which serves as a gelling agent, thicker, film former, and

emulsifier in industry (Nishinari et al., 1992). Recent

years have seen a highly accelerated demand for A. albus

corms, which led to over-harvesting and depletion of its

natural resources (Long et al., 2003). Moreover, natural

genetic variation is lacking in A. albus due to long-term

vegetative propagation (by corm setts) and lack of seeds.

To date, no successful breeding of improved cultivars of

A. albus has been documented, even though this crop has

been cultivated for hundreds of years. Callus cultures and

subsequent regeneration may result in the generation of

useful somaclonal variants not available by conventional

methods.

Plant regeneration from petiole callus of Amorphophallus

albus and analysis of somaclonal variation of regenerated

plants by RAPD and ISSR markers

Jianbin HU

1, 2

, Xiaoxi Gao

1

, Jun LIU

1

, Conghua XIE

1,

*, and Jianwu LI

1,2

1

Key Laboratory of Horticultural Plant Biology, Ministry of Education/National Center for Vegetable Improvement (Cen-

tral China), Huazhong Agricultural University, Wuhan 430070, China

2

College of Forestry and Horticulture, Henan Agricultural University, Zhengzhou 450002, China

(Received October 29, 2007; Accepted February 21, 2008)

ABSTRACT.

A simple procedure has been outlined for plant regeneration of Amorphophallus albus Liu &

Wei, a native medicinal plant in China, from petiole-derived callus. Calli were induced at a high frequency of

76.4¡Ó3.2% from petiole explants excised from two-month-old plants on Murashige and Skoog (MS) medium

supplemented with 5.37 £gM

£\-naphthaleneacetic acid (NAA) and 4.44 £gM 6-benzyladenine (BA). Of the

different types of callus induced, type III callus was selected for morphogenesis induction. Culture of the

callus on MS medium containing proper NAA and BA or KT combinations resulted in formation of corm-like

structure (CLS) that produced shoots and roots during further culture. The optimal morphogenetic response

was observed on the media with a cytokinin/auxin ratio of about 4:1, which resulted in more than 70% CLS

formation and 6~8 CLSs per callus. Complete plantlets with well-developed root systems were obtained from

these CLSs by subculturing them on the original media from which they had been derived without a separate

rooting culture. Transfer of the plantlets with roots to soil resulted in a more than 90% survival rate. Analysis

of 20 regenerated plants by two molecular markers, randomly amplified polymorphic DNA (RAPD) and inter-

simple sequence repeat (ISSR), revealed somaclonal variation in the regenerated plants. The percentage of

polymorphic bands in RAPD and ISSR analysis were respectively 20.8% and 39.0% for the 20 plants. Cluster

analysis indicated that the genetic similarity values calculated on the basis of RAPD and ISSR data among the

21 plants (20 regenerated and one donor plant) were, respectively, 0.973 and 0.917, which allowed classifica -

tion of the plants into distinct groups. A high-frequency somaclonal variation induced in A. albus tissue culture

may help in the selection of useful variants that may be induced to improve this important corp.

Keywords: Amorphophallus albus Liu & Wei; Organogenesis; Molecular marker; Somaclonal variation.

Abbreviations: BA, 6-Benzyladenine; CLS, Corm-like structure; ISSR, Inter simple sequence repeat; KT, Ki-

netin (6-furfuryl-amino purine); MS, Murashige and Skoog medium (1962); NAA, £\-Naphthaleneacetic acid;

RAPD, Randomly amplified polymorphic DNA; UPGMA, Unweighted pair group method with arithmetic

average.

mOleCUlAR BIOlOgy

190

Botanical Studies, Vol. 49, 2008

Plant tissue culture has been successfully used in many

valuable medicinal plants for efficient propagation (Rout

et al., 2000) or for selection of useful mutants (Sato et al.,

1988; Bertin et al., 1997). In Amorphophallus, plant regen-

eration has been achieved in some economically important

species using corm segments (Asokan et al., 1984; Irawati

et al., 1986), leaf (Kohlenbach and Becht, 1988), inflo-

rescence, and bulbil (Zhuang and Zhou, 1987). Wu and

Xie (2001) investigated the effect of growth regulators on

corm explant callusing in A. albus. Successful induction

of adventitious bud from corm-derived callus of A. albus

and establishment of regenerated plants in soil were

reported by Liu et al. (2001) and Yan et al. (2005). To our

knowledge, however, there has been no report on genomic

stability/instability of A. albus regenerated plants.

Recently, several DNA markers have been successfully

employed to assess the genomic stability/instability in

regenerated plants. Among the markers, the randomly

amplified polymorphic DNA (RAPD) and the inter-simple

sequence repeat (ISSR) have been favored because of

their sensitivity, simplicity, and cost-effectiveness (Yang

et al., 1996). In this paper, we aim to develop a simple and

efficient tissue culture method for A. albus and analyze

genomic stability or instability of the regenerated plants

using RAPD and ISSR markers. The information obtained

on genomic variation will be valuable for the specific

purpose of utilizing tissue culture as a means for clonal

propagation, or for the selection of useful mutants in A.

albus.

mATeRIAlS AND meTHODS

Seed corms of Amorphophallus albus Liu & Wei,

obtained from Hubei Konjac Research Institute, germinat-

ed in a greenhouse and were used as source material. The

petioles excised from the 2-month-old plants were surface

sterilized as described by Hu et al. (2005). The basal me-

dium used in all experiments was MS medium (Murashige

and Skoog, 1962) supplemented with 3% (w/v) sucrose

and 0.7% (w/v) Bacto-agar (Difco Laboratories, USA).

The specific concentrations of growth regulators added to

MS medium were indicated for different treatments. All

components of the medium were mixed and adjusted to 5.8

prior to autoclaving. Forty milliliters of medium were dis-

pensed into polyethylene containers (7 cm ¡Ñ 7 cm ¡Ñ 7 cm)

and autoclaved at 121¢XC (1.04 Kg cm

-2

) for 15 min.

Callus induction and plant regeneration

The sterilized petioles were chopped into segments (4-6

mm in length) and incubated on MS medium supplemented

with NAA (2.69, 5.37, 10.74 £gM) alone or in combination

with BA (2.22, 4.44 £gM) for callus induction and subcul-

tured at an interval of 4 weeks. The compact and nodu-

lar calli were selected from those in different types and

cultured on MS medium with combinations of NAA (0,

0.54, 1.07, 2.15 £gM) and BA (2.22, 4.44, 6.66, 8.88, 11.10

£gM) or KT (2.32, 4.65, 6.97, 9.29, 11.62 £gM) for corm-

like structure (CLS) induction and plant regeneration.

All cultures were incubated at 25¡Ó1¢XC in a culture room

in darkness or under a 12-h photoperiod (45 £gmol m

2

s

-1

)

provided by cool-white fluorescent tubes for callus induc-

tion or plant regeneration, respectively. Each treatment

was repeated thrice with about 30 petiole segments or cal-

lus pieces per replicate. Data were statistically analyzed

by analysis of variance (ANOVA) followed by Duncan¡¦s

Multiple Range Test (P = 0.05).

Transplantation

CLS culture would lead to development of both shoot

and root. Regenerated plants with well-developed roots

along with CLSs were isolated from mother tissues and

rinsed under running tap water to remove the gel and then

transferred to a soil mixture containing garden soil and

peatmoss in the ratio of 1:1 (v/v). The plants were watered

every 2-3 days, as needed. A month later, the percentage of

surviving plants was investigated.

DNA extraction and PCR amplification

The 20 regenerated plants randomly selected from the

>300 soil-established plants along with the single donor

mother plant were subjected to RAPD and ISSR analysis.

Total genomic DNA was extracted from fresh leaves of

each individual plant using a CTAB method described by

Porebski et al. (1997). Quantity and quality of DNA was

inspected by both gel electrophoresis and spectrometric

assays.

A total of 62 RAPD and 26 ISSR primers were initially

tested using the donor plant DNA (in two replicates) as

template to screen for suitable primers. Both types of am-

plification were performed in a volume of 15 £gl containing

30 ng total DNA, 1¡ÑPCR buffer, 2.0 mM MgCl

2

, 1 .0

U Ta q DNA polymerase (Sangon Biotech, Shanghai,

China), 0.25 mM dNTPs, and 0.45 £gM or 0.9 £gM primer

(respectively for RAPD or ISSR). Amplifications were

performed in a UNO II Biometra thermocycler. The pro-

gram for RAPD consisted of an initial denaturation of the

DNA at 94¢XC for 4 min, following by 40 cycles of 50 s

at 94¢XC, 50 s at 36¢XC, 1.5 min at 72¢XC, and a final 8-min

extension at 72¢XC. The program for ISSR was: DNA

denaturation at 94¢XC for 4 min; 40 cycles of 30 s at 94¢XC,

45 s at the annealing temperature, and 2 min at 72¢XC, and

an extension at 72¢XC for 8 min. The annealing temperature

was adjusted according to the Tm of the primer being used

in the reaction.

The amplified products of RAPD and ISSR were elec-

trophoresed in a 1% and 2% agarose (Sigma, USA) gel

with 1¡ÑTAE buffer, stained with ethidium bromide, and

photographed under ultraviolet light. PCR reactions were

repeated at least twice to establish reproducibility of the

results.

Data analysis

Only consistently reproducible bands in the size of 150

bp to 2.0 kb were considered for RAPD and ISSR analysis.

Each amplified product was scored 1 for presence and 0

HU et al. ¡X Plant regeneration in

Amorphophallus albus

and somaclonal variation

191

for absence of a band across the 20 regenerated plants and

one donor mother plant of A. albus. Pairwise similarity

matrices were generated using Jaccard¡¦s similarity

coefficient. Data analysis was performed using NTSYSpc,

version 2.10. Dendrograms were created with the

unweighted pair group method using arithmetic averages

(UPGMA) (Sneath and Sokal, 1973).

ReSUlTS

Callus induction

Incubated on callus induction media, most petiole

segments swelled and disorganized, followed by

development of callus masses in the fourth week. Single

NAA containing media stimulated the explant callusing

at a low frequency (less than 15%), regardless of NAA

concentration. In contrast, addition of both NAA and BA to

MS medium resulted in a high-frequency callus formation

(Table 1). Of the media tested, MS medium plus 5.37 £gM

NAA and 4.44 £gM BA was most effective, indicating by

76.4¡Ó3.2% callus induction. Also, the combination of

2.69 £gM NAA and 2.22 £gM

BA gave a positive response,

with an induction rate of 68.8¡Ó6.7%. Further increases

in NAA concentration led to a declining frequency of

callus induction and more explant browning. The induced

callus, after being subcultured for 2~3 weeks, developed

into three different types: (I) translucent and friable, (II)

pale and loose, (III) yellowish in color and nodular and

compact in structure (Figure 1A). In our experiment, only

the type III calli were used for organogenesis induction

because this type of callus multiplied easily and was prone

to producing globular structures.

Plant regeneration

Type III calli (approx. 500 mg) were transferred to MS

medium with various combinations of NAA and BA or

KT for plant regeneration. After 4 weeks of incubation,

most calli revealed a special morphogenesis that could be

divided into two stages: the formation of CLSs from the

calli (Figure 1B) in the first stage, and an almost synchro-

nous development of both shoots and roots from the CLSs

(Figure 1C) in the second one. This led to the formation

of complete plants with root systems (Figure 1D), and

it distinctly differed from the common in vitro pathway,

which only produced shoots.

The frequency of CLS occurrence varied with BA/NAA

or KT/NAA combinations added to MS medium. The use

of proper molar ratios of cytokinin to auxin (. 4:1) was

a key factor promoting CLS formation (Table 2). Among

the phytohormone combinations tested, 1.07 £gM NAA in

combination with 4.44 £gM BA (BA/NAA ratio = 4.15:1)

proved most effective, giving rise to 78.6¡Ó7.4% CLS

formation and a mean 7.8¡Ó1.2 CLSs per callus. The other

phytohormone combinations with a cytokinin/auxin ratio

of about 4:1¡Xi.e. 2.22 £gM BA + 0.54 £gM NAA, 8.88

£gM BA + 2.15 £gM NAA, 2.32 £gM KT + 0.54 £gM NAA,

4.65 £gM KT + 1.07 £gM NAA¡Xalso resulted in more

than 70% CLS formation and 6~8 CLSs per callus. No

statistically significant difference was found among these

Table 1. Effect of different combinations of NAA and BA on

callus induction in A. albus after 4 weeks of culture.

NAA

(£gM)

B A

(£gM)

Callus induction

%

a

Callus feature

2.69 0

13.2¡Ó5.1f Brown, granular

2.69 2.22 68.8¡Ó6.7ab Yellowish, loose

2.69 4.44 39.7¡Ó3.6d Greenish, nodular, compact

5.37 0

12.4¡Ó6.3f Brown, granular

5.37 2.22 40.5¡Ó3.5d Yellowish, loose

5.37 4.44 76.4¡Ó3.2a Yellowish, nodular, compact

10.74 0

8.7¡Ó4.5f Explant browning

10.74 2.22 27.3¡Ó3.7e Brown, granular

10.74 4.44 53.1¡Ó5.7c Brown, friable

a

Means followed by the same letters within a column are not

s ignificantly different us ing Duncan¡¦s Multiple Range Test

(p<0.05). Values represent the means of three replicates¡ÓSE.

Figure 1. CLS formation and plant regeneration from petiole

callus of A. albus. (A) Nodular and compact calli (type III)

induced from petiole segment on MS medium supplemented

with 4.44 £gM BA and 5.37 £gM NAA. Bar = 5 mm; (B)

Numerous CLSs (arrows) formed on the surface of the type III

callus incubated on MS medium supplemented with 8.88 £gM

BA and 2.15 £gM NAA. Bar = 5 mm; (C) Apical buds of CLSs

geminated and roots developed at base of the buds with the com-

bination of 8.88 £gM BA and 2.15 £gM NAA. Bar = 10 mm; (D)

Complete plantlets regenerated from the CLSs on MS medium

plus 8.88 £gM BA and 2.15 £gM NAA. Bar = 10 mm.

192

Botanical Studies, Vol. 49, 2008

combinations. Higher cytokinin/auxin ratios resulted in

repression of CLS formation and stimulated adventitious

bud differentiation. Lower ratios of BA to NAA¡Xe.g. 2.22

£gM BA + 1.07 £gM NAA and 2.32 £gM KT + 1.07 £gM NAA

(cytokinin/auxin ratio . 2:1)¡Xrepressed CLS formation

and promoted callus growth.

Maintained on fresh medium with the same components

as the original media for 2 weeks, most CLSs devel-

oped into complete plantlets (Figure 1D). The percent-

age of CLSs converting to plants exceeded 80% in our

experiment. These plants along with the basal CLSs were

excised from mother tissues and transferred directly to soil

after removal of the gel attached to the roots. One month

later, more than 90% of the plants survived under ex vitro

conditions and grew vigorously. Thus, within a time frame

of about 4 months, a population of >300 CLS-derived

plants, originating from a single mother donor plant, was

established in the field. No discernible morphological

difference was found between the mother plant and the

regenerated plants.

RAPD and ISSR fingerprinting

In order to assess the genetic stability/instability of

the regenerated plants, a comparison of the RAPD and

ISSR fingerprinting of 20 randomly-selected regenerated

plants and their single donor mother plant was carried

out. In an initial screening, 21 decamer oligonucleotides

were selected from 62 decamer RAPD primers due to

the generation of clear and reproducible amplification

products (Table 3). Seven of the 21 primers (33.3%)

produced monomorphic band patterns while each of the

remaining 14 primers generated polymorphic bands in at

least one of the 20 individuals relative to the donor plant.

Of a total of 154 bands scored, which ranged from 150

bp to 2 kb and averaged 7.3 bands per primer, 32 were

polymorphic with 20.8% polymorphism. The highest

percentage of polymorphic bands (50%) was possessed

by primer S437. The changed banding patterns included

both loss of original bands (Figure 2A) and appearance of

novel bands (Figure 2B), which occurred at more or less

equal frequencies. Cluster analysis was done on the basis

of similarity coefficients among the 21 individuals (20

regenerated and one donor plant) generated from RAPD

data. The similarity coefficients ranged from 0.85 to 1.00

among the plants with a mean of 0.973. All 21 plants could

be clustered into one major group at an 85% similarity

level. However, four groups could be obtained at a 93%

similarity level (Figure 3).

Table 2. Effect of different phytohorm one combinations on

CLS formation from calli of A. albus after 4 weeks of culture.

Phytohormone

combinations

(£gM)

Cytokinin/Auxin

molar ratios

Shooting

response

%

a

Number of

shoots per

callus

a

BA + NAA

2.22 + 0

-

12.2¡Ó4.3i 2.6¡Ó0.4g

2.22 + 0.54

4.11

70.2¡Ó5.4a 6.4¡Ó1.1bcd

4.44 + 0.54

8.22 26.5¡Ó7.4gh 3.2¡Ó0.6fg

2.22 + 1.07

2.07 41.8¡Ó7.7def 4.6¡Ó0.8def

4.44 + 1.07

4.15

78.6¡Ó7.4a 7.8¡Ó1.2a

6.66 + 1.07

6.22

32.2¡Ó9.3fg 5.0¡Ó1.1cde

6.66 + 2.15

3.10 54.5¡Ó2.9bcd 6.2¡Ó0.9bcd

8.88 + 2.15

4.13

72.1¡Ó10.8a 7.2¡Ó0.6ab

11.10 + 2.15

5.16

18.7¡Ó6.1hi 2.7¡Ó0.7g

KT + NAA

2.32 + 0

-

10.7¡Ó4.6i 3.1¡Ó0.7fg

2.32 + 0.54

4.29

71.2¡Ó9.3a 7.5¡Ó1.0a

4.65 + 0.54

8.59 35.3¡Ó5.7efg 4.2¡Ó1.7efg

2.32 + 1.07

2.16 48.1¡Ó7.4cde 5.3¡Ó1.3bcde

4.65 + 1.07

4.35

76.2¡Ó7.2a 6.5¡Ó1.3bc

6.97 + 1.07

6.51

57.1¡Ó7.8bc 5.2¡Ó0.9cde

6.97 + 2.15

3.24

65.4¡Ó7.7ab 6.5¡Ó1.1bc

9.29 + 2.15

4.32

56.6¡Ó1.0bc 6.2¡Ó.10bcd

11.62 + 2.15

5.40

16.2¡Ó8.0hi 3.0¡Ó0.8fg

a

Means followed by the same letters within a column are not

significantly different using Duncan¡¦s Multiple Range Test

(p<0.05). Values represent the means of three replicates¡ÓSE.

Figure 2. RAPD profiles of 20 regenerated plants (lane 1~20)

and their donor mother plant (D) produced using the decamer

pri mers S 230 (A) and S 437 (B). M, DNA ladder, DL 2000

marker (Takara, Dalian, China). Polymorphisms included both

absence of original bands (A, arrow) and gain of novel bands (B,

arrow).

HU et al. ¡X Plant regeneration in

Amorphophallus albus

and somaclonal variation

193

In ISSR analysis, 26 an chored microsatellite primers

including di-, tri- and tetranucleotide repeat motifs were

tested individually to amplify the genomic DNA of the A.

albus donor plant, out of which 11 primers generated well

resolved reproducible band patterns (Table 4). The eleven

ISSR primers yielded 136 scorable bands, of which 53

were polymorphic. In comparison to RAPD, ISSR primers

produced a higher percentage of polymorphism (39.0%)

in the same 20 individuals. An average of 12.4 bands per

primer was recorded, and band size varied from 150 bp to

2 kb. The number of bands from each primer varied from

9 bands in UBC840 and UBC873 to 16 bands in UBC866.

All the primers were found to be polymorphic and produce

different percentages of polymorphism, ranging from the

14.3% produced by UBC818 to the 72.7% by UBC846. As

in the RAPD results, the variable bands in the regenerated

plants relative to the donor plant also included both the

loss of original bands (Figure 4A) and the appearance of

Table 3. Description of twenty one RAPD primers used for fingerprint analysis of 20 plants regenerated from petiole callus of A.

albus.

Primers

Sequence

(5¡¦-3¡¦)

Size of the amplified

bands (bp)

Scored bands

Polymorphic bands

Number Frequency (%)

S35

TTCCGAACCC

300-1800

9

4

44.4

S67

GTCCCGACGA

500-2000

7

0

0

S70

TGTCTGGGTG

500-1000

5

2

40.0

S71

TGTCTGGGTG

250-1800

9

4

44.4

S75

GACGGATCAG

250-1500

7

1

14.3

S80

CTACGGAGGA

250-1500

6

0

0

S161

ACCTGGACAC

500-2000

8

2

25.0

S163

CAGAAGCCCA

250-1000

7

0

0

S226

ACGCCCAGGT

250-1500

10

2

20.0

S230

GGACCTGCTG

300-1000

7

1

14.3

S236

ACACCCCACA

300-2000

7

3

42.9

S353

CCACACTACC

250-2000

10

3

30.0

S380

GTGTCGCGAG

150-1000

9

0

0

S387

AGGCGGGAAC

500-2000

9

3

33.3

S425

ACTGAACGCC

250-1500

8

0

0

S427

CAGCCCAGAG

250-1800

7

0

0

S429

TGCCGGCTTG

300-1800

5

1

20.0

S431

TCGCCGCAAA

500-2000

6

2

33.3

S437

CATTGGGGAG

250-2000

6

3

50.0

S504

CCCGTAGCAC

250-1800

6

1

16.7

S507

ACTGGCCTGA

250-1800

6

0

0

Total

154

32

20.8

Figure 3. Dendrogram illustrating coefficient similarities among

20 regenerated plants (1~20) and their donor mother plant (D) of

A. albus based on RAPD data.

194

Botanical Studies, Vol. 49, 2008

novel bands (Figure 4B). In contrast to the RAPD results,

the total number of lost bands in the 20 regenerated plants

was much larger than that of novel bands (202 versus

106). The coefficient of similarity in the dendrogram

generated by the ISSR data among the 20 regenerated

plants and their donor plant ranged from 0.80 to 0.97 with

a mean of 0.917. The associations among the 21 plants

were similar to those of revealed by the RAPD analysis

(a cluster analysis). One major group including all the 21

plants could be obtained with the similarity level set at

80%. However, only at an 85% similarity level could these

plants be clustered into four groups (Figure 5).

DISCUSSION

Amorphophallus albus is a multipurpose plant in China

that has attracted increasing attention due to its economic

and medicinal importance. Its propagation has proven

difficult because the corms have to grow for at least three

years before harvest, and the cormels (usually used as the

succeeding year ¡¦s seed corms) produced by the mother

corms are quite few (Long, 1998). That makes it urgent to

develop a rapid method of multiplication for commercial

cultivation of this important crop. Here, we presented

a new method for propagating A. albus via corm-like

structure (CLS) induction, which differed from previously

reported methods that produced only adventitious buds

(Liu et al., 2001; Wu and Xie, 2001; Yan et al., 2005). In A.

albus tissue culture, a callus phase is required for obtaining

regeneration, and, to date, no information is available

about direct organogenesis from explants. Our results sug-

gest that combinations of NAA and BA at approximately

equal concentrations favour callus induction, indicating

the need for a proper balance of auxin and cytokinin in

petiole explant callusing. Similar findings were described

by Yan et al. (2005) in A. albus and Liu et al. (2001) in

Amorphophallus konjac Koch . The type III callus formed

in the present experiment was a desirable tissue type char-

acterized by a compact texture and nodular structure and

the capacity to produce CLSs (Hu et al., 2004). In some

other species, a similar type of callus was also considered

an alternative tissue type that could multiply and easily

differentiate shoots or other organs (Teng, 1997; Te-Chato

and Lim, 2000).

Table 4. Description of eleven ISSR primers used for fingerprint analysis of 20 plants regenerated from petiole callus of A. albus.

Primer

Sequences Size of the amplified bands (bp) Scored bands

Polymorphism bands

Number Frequency (%)

UBC818

(CA)

8

G

250-1500

14

2

14.3

UBC825

(AC)

8

T

250-1800

15

6

40.0

UBC826

(AC)

8

C

300-1800

13

4

33.3

UBC840

(GA)

8

YT

150-1500

9

5

55.6

UBC846

(CA)

8

RT

500-2000

11

8

72.7

UBC850

(GT)

8

YC

250-2000

11

6

54.5

UBC857

(AC)

8

YG

250-1500

12

5

41.7

UBC858

(TG)

8

RT

250-2000

12

4

33.3

UBC864

(ATG)

6

200-2000

14

2

14.3

UBC866

(CTC)

6

200-1800

16

8

50.0

UBC873

(GACA)

4

300-2000

9

3

33.3

Total

136

53

39.0

R = A/G; Y = G/T.

Figure 4. ISSR profiles of 20 regenerated plants (lane 1~20) and

their donor mother plant (D) produced using the ISSR primers

UBC825 (A) and UBC840 (B). M, DNA ladder, DL2000 marker

(Takara, Dalian, China). Polymorphisms included both absence

of original bands (A, arrow) and gain of novel bands (B, arrow).

HU et al. ¡X Plant regeneration in

Amorphophallus albus

and somaclonal variation

195

As previously observed (Hu et al., 2004), culture of

type III callus resulted in formation of numerous CLSs that

produced complete plantlets during further culture. Similar

morphogenetic events have been reported by Irawati et al.

(1986) in tissue culture of Amorphophallus paeoniifolius

(Dennst.) Nicols., where the globular structures were

developed from the long-time conserved calli and could

produce shoots and roots if transferred to fresh medium.

During A. konjac callus culture, we also found that CLSs

frequently occurred and exhibited the same morphogenetic

pathway as those in present study (Liu et al., 2001; Hu et

al., 2005). From these findings, we might conclude that

CLS widely exists in the tissue culture of Amorphophal-

lus species. Although CLS formation was frequently

accompanied by adventitious bud differentiation in A.

albus, we could adjust the cytokinin/auxin ratio of the

medium to promote CLS formation and reduce adventi-

tious bud differentiation. Maintenance of a molar ratio

of cytokinin to auxin (. 4:1) was a key factor promoting

a high-frequency CLS formation. The proper proportion

of auxin to cytokinin might support a balance of endog-

enous growth substances in callus that favoured corm

organogenesis. Liu et al. (2001) reported that in A. konjac

seed-derived callus culture, cormlets frequently occurred

in the presence of moderate concentrations of BA and low

concentrations of NAA, where the BA/NAA ratio was

approximately 5:1. This finding approximated what was

observed in the present study.

Based on the findings above, a two-step propagation

protocol including callus induction (step 1) and CLS

induction and plant regeneration (step 2) can be devel-

oped. The obvious advantages of the protocol lie in its:

(1) technical simplicity (includes only two steps and

excludes rooting culture); (2) shortened time span of plant

regeneration (about 3 months from explant incubation

to plant regeneration); and (3) development of complete

Figure 5. Dendrogram illustrating coefficient similarities among

20 regenerated plants (1~20) and their donor mother plant (D) of

A. albus based on ISSR data.

regenerated plants. The strong morphogenetic abilities

revealed by the CLSs, i.e., production of both shoots and

roots, might be due to the pre-existing bud and root pri-

mordial in CLS developed in the morphogenesis induction

stage (Hu et al., 2005; unpublished work). It must be

noted that, apart from CLS production in our experi-

ment, adventitious buds also differentiated from some

calli. These buds, however, were not used for A. albus

propagation because of the low frequency of conversion to

normal shoots and the need for rooting culture.

It is well known that in vitro culture conditions act as a

stress factor and elicit genetic instability in cultured cells,

tissues and organs, a phenomenon known as somaclonal

variation (Larkin and Scowcroft, 1981; Karp, 1995). This

instability may be a risk associated with the application of

in vitro techniques to commercial micropropagation and

germplasm conservation. Conversely, somaclonal variation

may provide another source of novel and useful variability,

which can be used in crop improvement, particular for

species with a narrow genetic background (Karp, 1995).

The introduction of valuable variation through somaclonal

variation may help in programmes designed to improve

the characteristics of A. albus. Though the 20 randomly

tagged regenerated plants were phenotypically normal

and essentially identical with their mother donor plant,

they showed apparent genetic variations when subjected

to RAPD and ISSR analysis. Similar findings on genomic

variation in phenotypically normal regenerants have

been well documented in some other plants (Diwan and

Cregan, 1997; Rahman and Rajora, 2001; Kawiak and

Lojkowska, 2004). There were two probable reasons for

the incongruence of phenotypic versus genomic stability in

regenerated plants: (1) the tissue culture-induced genomic

changes mainly take place at non-coding regions, which

hardly affects gene expression, and hence, phenotype;

and (2) even if the changes occur in coding regions, the

probability of simultaneous mutation of two alleles in

a diploid plant is extremely low and has little effect on

phenotypic characteristics.

In our study, the percentage of polymorphic bands

among the 20 regenerated plants detected by ISSR was

much higher than that detected by RAPD, indicating

that the former was more discriminative than the lat-

ter of genomic variations in A. albus. Meanwhile, our

study also shows that A. albus genome seems to be rich

in dinucleotide repeat motifs like (CA)n and (GA)n and

that these repeat regions are prone to change under i n

vitro culture conditions. In a separate experiment, we

found that the same set of ISSR primers could produce a

high-frequency polymorphism in detection of somaclonal

variation in A. konjac (Hu et al., 2007). According to Wang

et al. (1994), dinucleotide microsatellites are prevalent in

plants while mono-, tri-, and tetranucleotide repeats are

less common. As observed in band investigation, the ratios

of band loss to band gain were distinctly different between

RAPD and ISSR, with that of the latter being much higher.

This may be because RAPD and ISSR primers target

different regions of the A. albus genome, and different

196

Botanical Studies, Vol. 49, 2008

regions have different sensitivities to tissue culture

conditions, facts demonstrated in several plant species (Xie

et al., 1995; Devarumath et al., 2002).

In conclusion, we report herein a simple and efficient

tissue culture protocol for obtaining numerous complete

regenerated plantlets of A. albus in a short period of

time. Genomic variations revealed by RAPD and ISSR

marker are apparent in regenerated plants, which may be

an important source of useful variants that can be used

to improve this important crop at the cellar level. The

regenerated plants are currently being screened in the field

for corm size and glucomannan content.

Acknowledgements. The authors thank Yan Huabing for

technical assistance. This research was supported by the

National Key Technologies R&D Programme of China

(2003BA901A05).

lITeRATURe CITeD

Asokan, M.P., S.K. O¡¦Hair, and R.E. Litz. 1984. In vitro plant

regeneration from corm callus of Amorphophallus rivieri

Durieu. Sci. Hortic. 24: 251-256.

Bertin, P., J. Bouharmont, and J.M. Kinet. 1997. Somaclonal

variation and improvement of chilling tolerance in rice:

changes in chilling-induced chlorophyll fruorescence. Crop

Sci. 37: 1727-1735.

Devarumath, R.M., D. Nandy, V. Rani, S. Marimuthu, N. Mura-

leedharan, and S.N. Raina. 2002. RAPD, ISSR and RFLP

fingerprints 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 (Assam-India type). Plant Cell

Rep. 21: 166-173.

Diwan, N. and P.B. Cregan. 1997. Automated sizing of fluo-

rescent labeled simple sequence repeat (SSR) markers to

assay genetic variation in soybean. Theor. Appl. Genet. 95:

723-733.

Gandawijaja, D., S. Idris, P. Nasution, L.P. Nyman, and J. Ar-

ditti. 1983. Amorphophallus titanum Becc: an historical re-

view and some recent observations. Ann. Bot. 51: 269-278.

Hu, J.B., J. Liu, H.B. Yan, and C.H. Xie. 2004. The callus types

and redifferentiation capacities of Amorphophallus Blume.

J. Huazhong Agri. Univ. 23: 654-658.

Hu, J.B., J. Liu, H.B. Yan, and C.H. Xie. 2005. Histological

observations of morphogenesis in petiole derived callus of

Amorphophallus rivieri Durieu in vitro. Plant Cell Rep. 24:

642-648.

Hu, J.B., X.X. Gao, C.H. Xie, and J. Liu. 2007. RAPD and ISSR

analysis of somaclonal variation among regenerated plants

of Amorphophallus rivieri. Sci. Agri. Sin. in press.

Irawati, J. Arditti, and L.P. Nyman. 1986. In vitro propagation of

the elephant yam, Amorphophallus campanulatus Var. hor-

tensis Backer (Araceae). Ann. Bot. 57: 11-17

Karp, A. 1995. Somaclonal variation as a tool for crop

improvement. Euphytica 85: 295-302.

Kawiak, A. and E. Lojkowska. 2004. Application of RAPD in

the determination of genetic fidelity in micropropagation

Drosera pla ntlet s. In Vit ro Ce ll. Dev. Biol.-P lant 40:

592-595.

Kohlenbach, H.W. and C. Becht. 1998. In vitro propagation of

Amorphophallus titanum Becc. and Amorphophallus rivieri

Durieu. Acta Hortic. 226: 65-72.

Larkin, P. and W.R. Scowcroft. 1981. Somaclonal variation,

a novel source of variability from cell cultures for plant

improvement. Theor. Appl. Genet. 60: 197-214.

Liu, J., C.H. Xie, Z.S. Yu, and Y. Liu. 2001. Research on the

propagation of Amorphophallus in vitro. J. Huazhong Agri.

Univ. 20: 283-285.

Long, C.L. 1998. Ethnobotany of Amorphophallus of China.

Acta Bot. Yunnan. 20: 89-92.

Long, C.L., H. Li, Z.Q. Ouyang, X.Y. Ya ng, Q. L i, and B.

Trangmar. 2003. Strategies for agrobiodiversity conserva-

tion and promotion: a case from Yunnan, China. Biodivers.

Conserv. 12: 1145-1156.

Murashige, T. and F. Skoog. 1962. A revised medium for rapid

growth and bioassays with tobacco tissue culture. Physiol.

Plant. 15: 473-497.

Nishinari, K., P.A. Williams, and G.O. Phillip. 1992. Review of

the physico-chemical characteristics and properties of kon-

jac glucomannan. Food Hydrocolloid. 6: 199-222.

Porebski. S, G.L. Bailey, and B.R. Baum. 1997. Modifications of

a CTAB DNA extraction protocol for plant containing high

polysaccharide and polyphenol components. Plant Mol.

Biol. Rep. 15: 8-15.

Rahman, M.H. and O.P. Rajora. 2001. Microsatellite DNA so-

maclonal variation in micropropagated trembling aspen

(Populus tremuloides). Plant Cell Rep. 20: 531-536.

Rout, G.R., S. Samantaray, and P. Das. 2000. In vitro manipula-

tion and propagaton of medicinal plants. Biotech. Adv. 18:

91-120.

Sato, F., Y. Shigematsu, and Y. Yamada. 1988. Selection of an

atrazine-resistant tobacco cell line having a mutant psbA

gene. Mol. Gen. Genet. 214: 358-360.

Sneath, P.H.A. and R.R. Sokal. 1973. Numerical Taxonomy.

Freeman, San Francisco.

Wang, Z., J.L. Weber, G. Zhong, and S.D. Tanksley. 1994. Sur-

vey of plant short-tandem DNA repeats. Theor. Appl. Genet.

88: 1-6.

Wu, Y.X. and Q.H. Xie. 2001. Study on tissue culture of explants

from Amorphophallus albus under different conditions .

Plant Physiol. Commun. 37: 513-514.

Xie, Q.J., J.H. Oard, and M.C. Rush. 1995. Genetic analysis of a

purple-red hull rice mutation derived from tissue culture. J.

Hered. 86: 154-156.

Yan, H.B., J.B. Hu, and J. Liu. 2005. Plant regeneration system

and its sensitivity to antibiotic in Amorphophallus albus in

vitro. Hubei Agri. Sci. 4: 71-74.

Yang, W., A.C. Oliveira, I. Godwin, K. Schertz, and J.L. Ben-

HU et al. ¡X Plant regeneration in

Amorphophallus albus

and somaclonal variation

197

netzen. 1996. Comparison of DNA marker technologies in

characterizing plant genome diversity: variability in Chi-

nese sorghums. Crop Sci. 36: 1669-1676.

Zhuang, J.C. and J.K. Zhou. 1987. Study on callus induction

and plant regeneration of Amorphophallus plants. Acta Bot.

Yunnan. 9: 339-347.

Zhang, Y.Q., B.J. Xie, and X. Gan. 2005. Advance in the appli-

cations of konjac glucomannan and its derivatives. Carbo-

hyd. Polym. 60: 27-31.

Te-Chato, S. and M. Lim. 2000. Improvement of mangosteen

m icropropagation through meristematic nodular callus

formation from in vitro-derived leaf explants. Sci. Hortic.

86: 291-298.

Teng, W.L. 1997. An alternative propagation method of Ananas

through nodule culture. Plant Cell Rep. 16: 454-457.