 

LIBRARY
Mlchlgan State
Unlverslty

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

 

 

 

 

 

DATE DUE DATE DUE DATE DUE I

j
——J

.J
tL____1
~ :1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

———_7

 

 

 

———

  

 

 

 

 

 

 

 

 

 

‘ msu Is An Affirmative Action/Equal Oppottunity Institution
CWMM1

 

 

 

 

GENOTYPE AND GROWTH REGULATOR EFFECTS ON SHOOT REGENERATION
FROM PRIMARY AND SERIALLY-SUBCULTURED
HORMONE-AUTONOMOUS CALLUS OF SUGARBEET (BETA VULGARIS L.).

By
William Paul Doley

A DISSERTATION
Submitted to
Michigan State University
in partial mlﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Crop and Soil Sciences
1990

(90 - VoS'X

ABSTRACT
GENOTYPE AND GROWTH REGULATOR EFFECTS ON SHOOT REGENERATION
FROM PRIMARY AND SERIALLY-SUBCULTURED
HORMONE-AUTONOMOUS CALLUS OF SUGARBEET (BETA VULGARIS L.).
BY

William Paul Doley

Many in vitro genetic manipulations require a reliable shoot regeneration system.
Some sugarbeet genotypes are capable of one-step shoot regeneration from honnone-
autonomous callus (i.e. without subculture) when leaf disks are incubated on a Murashige and
Skoog (MS) medium containing 1 mg/L N‘-benzyladenine (BA)(Bl medium). Leaf disks from
plants from 16 populations were incubated on Bl at 31°C in darkness for 10 wk. Among and
within populations differences were signiﬁcant for frequencies Of callus production (CALLUS)
and shoot regeneration (REGEN), time to callus (CI'IME), and lag period (LAG) between
CALLUS and REGEN. Of 3018 leaf disk explants, 52.6% initiated callus in an average time
of 43.0 d, and 30.1% Of the calli regenerated shoots after a mean LAG Of 14.7 d. All 16
populations produced callus, but 5 populations failed to regenerate shoots. The 8 monogenn
populations had REGEN three times that Of the 8 multigenn populations (41.9% vs. 13.6%).
Clustering Of the 16 populations by CALLUS and REGEN resulted in 7 response types, each
of which may require a different medium to Optimize REGEN.

To Optimize REGEN within the response types, genotype x growth regulator
interactions were investigated in 21 sugarbeet genotypes. Bl medium was used as the control
in all experiments. Interactions Of genotype with BA, naphthaleneacetic acid (NAA) and
gibberellic acid (6A,) suggest that a single medium is inadequate when screening gennplasrn
for REGEN. Although CALLUS was somewhat independent Of [BA], the [BA] needed for
Optimal REGEN varied with genotype. Optimal [BA] for REGEN was above 1.0 mg/L for 4
of 8 genotypes tested. Some genotypes had enhanced REGEN when NAA (0.01-0.1 mg/L)
was included in Bl, but for most genotypes, N AA (0.1 mg/L) inhibited and delayed callus

formation and reduced REGEN. Inclusion Of GA3 (1.0 to 3.0 mg/L) in B1 increased shoot
number (SHOOTS) per leaf disk, and simultaneously inhibited root regeneration from callus.
RBGEN declined when calli were subcultured every 3 wk on Bl. To investigate this
decline in REGEN, calli of 3 genotypes were initiated on Bl and subcultured to various media
after 3 wk growth. Competence was assessed by REGEN and SHOOTS on maintenance (M)
media and challenge (C) media. After 15 wk on B1, more than half of BL 45/2-108 calli were
still regenerating shoots, while regeneration by calli of REL-1 and FC 607-020 was
approaching zero. REEEN from calli maintained on Bl was increased after subculture to C
medium B3 (MS «I» 3 mg/L BA). REGEN and SHOOTS were both enhanced by repeatedly
doubling the [BA] at each subculture or by maintenance on B1 + 1 mg/L 2,3,5-triiodobenzoic
acid (TIBA). Increases in REGEN were greater when both [BA] and [TIBA] were higher in
the C medium relative to the M medium. Calli maintained in a non-regenerating state on
honnone-free medium were induced to regenerate by transfer to B3. Manipulation Of shoot

regeneration with BA and TIBA is compatible with a model involving auxin/cytokinin ratio.

Copyright by
WILLIAM PAUL DOLEY

1992

TO the love of life,

and to all the good things which make life worth living.

TO the study of life,

for each bit Of understanding sheds a little light on our place on earth and in the universe.

TO the possibility,
that through the study Of life,
we will achieve a level Of understanding that allows us to live in harmony with all the diverse

life forms on earth, and to sustain forever the live-giving properties Of the earth.

ACKNOWLEDGEMENTS

The author wishes to acknowledge all those who helped to make this dissertation a
reality. Thanks are due to the guidance committee which consisted Of Drs. Joseph W.
Saunders, Kenneth C. Sink. Thomas G. Isleib, George L. Hosﬁeld, and J. Clair Theurer. Their
advice and suggestions were greatly appreciated. Special thanks are due to my dissertation
advisor, Dr. Joseph W. Saunders, who always made time tO discuss all those things which
needed to be discussed, and some other things too. He also provided the facilities and supplies
utilized in this research as well as ﬁnancial support beyond the second year of my program.
Financial support for the ﬁrst two years was provided by a grant from the Plant Breeding and
Genetics program at Michigan State University.

Thanks are due to all those who made media and/or assisted in the initiation Of some
of these experiments. They include: Martha Andrews. Steve Bamard, Connie Bobrovsy, Jeff
Klamer and Molly Stark.

Finally, a very special thank you to four wonderful women whose love and

encouragement provided the incentive to get out of bed on a regular basis: my daughters,
Rachael and Kira, my wife, Helene, and her dog, Ukiah.

vi

TABLE OF CONTENTS

LIST OF TABLES ......................................................................................................................... ix
LIST OF FIGURES ...................................................................................................................... xii
PAGE

INTRODUCTION ............................................................................................................................ 1
LITERATURE REVIEW ................................................................................................................ 3
SUGARBEET BREEDING ............................................................................................... 3
SUGARBEET PLANT REGENERATION SYSTEMS ................................................... 5
Regeneration from Conventional Callus .............................................................. .6

Regeneration from Honnone-Autonomous Callus ............................................... 7

Regeneration from Self-Regenerating Lines 7

Direct Regeneration from Explants ...................................................................... 8

GENETIC VARIATION FOR IN VITRO RESPONSE .................................................. 8
GENOTYPE x MEDIUM INTERACTION FOR IN VITRO RESPONSE ................... 1 l
LONG-TERM REGENERATION FROM CELL CULTURES ..................................... 13

CHAPTER 1 Genetic Variability for Frequency and Chronology Of Hormone-Autonomous
Callus Initiation and Subsequent Shoot Regeneration in Sugarbeet (Beta
vulgaris L.).

SUMMARY ......................................................................................................... 18
INTRODUCTION ............................................................................................... .19
MATERIALS AND METHODS ........................................................................ 20
Plant Material ......................................................................................... 20
Culture Procedures ................................................................................. 21
Experimental Design and Data Analysis .............................................. .23
RESULTS ............................................................................................................ 24
Gennplasrn Screening ............................................................................ 25
Variability Within and Between Populations ........................................ 31
In Vitro Response Types ....................................................................... .31
DISCUSSION ...................................................................................................... 35

CHAPTER 2 Genotype x Growth Regulator Interaction for Callus Initiation and Shoot
Regeneration in Sugarbeet (Beta vulgarr's L.).

SUMMARY ......................................................................................................... 38
INTRODUCTION ............................................................................................... 39
MATERIALS AND METHODS ........................................................................ 40
Plant Material ......................................................................................... 40
Culture Procedures 41

vii

Experimental Designs ............................................................................. 4 2

Data Analyses ......................................................................................... 4 6
RESULTS ............................................................................................................ 47
Effects Of BA ......................................................................................... .47
Effects Of NAA ...................................................................................... 52
Effects Of GA, ........................................................................................ 61
DISCUSSION ................................................... - .................. 63

 

CHAPTER 3 Factors affecting Shoot Regeneration from Serially-Subcultured Hormone-
autonomous Sugarbeet (Beta vulgaris L.) Callus.

SUMMARY ......................................................................................................... 68
INTRODUCTION ............................................................................................... 69

MATERIALS AND METHODS ........................................................................ 72

Plant Material 72

Culture Procedures ................................................................................. 72

Experimental Designs ............................................................................. 7 4

Data Analyses ......................................................................................... 76

RESULTS ............................................................................................................ 78

Effects of Genotype ................................................................................. 80

Subculture Interval ................................................................................. 83

BA Concentration ................................................................................... 84

Effects of TIBA ...................................................................................... 85

DISCUSSION ...................................................................................................... 95

GENERAL DISCUSSION .......................................................................................................... 101
RECOMMENDATIONS ................................................................................................ 103
Genetic Variability ............................................................................................ 103

Genotype x Medium Interaction ....................................................................... 104

Long-Tenn Regeneration .................................................................................. .104

LIST OF REFERENCES ............................................................................................................ .106

viii

Table

10

11

12

Table

LIST OF TABLES

Omen.
Seed Sources and characteristics of populations studied.

Frequency Of callus formation from leaf disks on B1, and number of
genotypes which produced callus from leaf disks.

Frequency of shoot regeneration from callus on BI, and number of
genotypes which regenerated shoots from callus.

Mean time to callus initiation from leaf disks cultured on B1.

Mean lag period between callus initiation and shoot production from leaf
disks cultured on B1.

Conelations among population means for four parameters of in vitro
behavior.

Comparison of in vitro behavior Of monogenn populations vs multigenn
populations.

Analysis of variance of frequency of callus production from leaf disks
cultured on B1.

Analysis of variance Of frequency of shoot regeneration from callus on
B1.

Analysis of variance of time to callus initiation from leaf disks cultured
on Bl.

Analysis of variance Of lag period between callus initiation and shoot
regeneration from leaf disks cultured on B1.

In vitro response types, including member populations and their

characteristics, generated by clustering populations by frequency of
callus production and shoot regeneration.

Chapter 2
Summary of some components Of experimental design for the seven

experiments to evaluate genotype x growth regulator interaction in
sugarbeet.

ix

I???

26

27

29
30

30

32

32

33

33

33

Table

Genotypes sampled in the seven experiments to evaluate genotype x
growth regulator interaction in sugarbeet.

Effect Of concentrations Of BA, NAA and GA, on time to callus (Exp
7).

Effect Of concentrations of BA and GA3 on shoot number (Exp 7).

Effect of 0.1 mg/L NAA in Bl medium on frequency of callus
initiation, frequency Of shoot regeneration and time to callus (Exp 2).

Effect of GA, concentration on frequencies of shoot and root
regeneration from callus Of three genotypes (Exp 5).

Effect of GA3 concentration on frequency Of shoot regeneration from
callus (Exp 6).

Effect of GA, concentration on shoot number variables (Exp 6).

Chapter 3

Summary of treatments and sampling structure of four experiments to
evaluate long term regeneration in sugarbeet.

BA and TIBA contents of the various media used throughout the four
experiments.

Pooled mean values of in vitro response of leaf disks Of three sugarbeet

genotypes from cultures used to initiate Exp 1, 2 and 3 (N=108).
Mean values of shoot regeneration frequency and shoot number per
callus Of three genotypes when challenged at 9 and 15 wk after
maintenance with differing subculture interval in Exp 1.

Mean values Of shoot regeneration frequency and shoot number per

callus Of l2-wk-Old calli Of three genotypes maintained with and without

BA in Exp 2. At 9 wk, each callus was subcultured to both
maintenance medium and challenge medium (B3).

Mean values Of shoot regeneration frequency and shoot number per

callus Of 15-wk-Old calli Of three genotypes maintained with and without

BA in Exp 2. At 12 wk, each callus was subcultured to both
maintenance medium and challenge medium (B3).

Mean values Of shoot regeneration frequency and shoot number per
callus of 12-wk-Old calli Of two genotypes maintained on Bl with and
without TIBA in Exp 3. At 9 wk, each callus was subcultured to both
maintenance medium and challenge medium (B3).

Mean values of shoot regeneration frequency and shoot number per

callus Of 15-wk-old calli Of two genotypes maintained with and without

56

62

87

88

91

Table

10

11

TIBA in Exp 3. At 12 wk. each callus was subcultured to both
maintenance medium and challenge medium (B3).

Mean values Of shoot regeneration frequency and shoot number per
callus Of two genotypes maintained on B1 with and without TIBA in
Exp 3. The response Of calli after 15 wk on maintenance media was
compared with the same calli after 3 wk on challenge medium B3Tl
(l8-wk Old). At 15 wk, all calli were subcultured from their respective
maintenance media to the challenge medium.

Mean values of shoot regeneration frequency and shoot number per
callus of 11.6-wk-old calli Of two genotypes maintained with and
without BA and/or TIBA in Exp 4. At 8.7 wk, each callus was
subcultured to both maintenance medium and challenge medium (B3Tl).
Data presented is for long SC plates only.

Mean values of shoot regeneration frequency and shoot number per
callus of 14.4-wk-Old calli of three genotypes maintained with and
without BA and/or TIBA in Exp 4. At 11.6 wk. each callus was
subcultured to both maintenance medium and challenge medium (B3Tl).
Data presented is for long SC plates only.

xi

96

Figpre

LIST OF FIGURES

ELEM

Graph Of population means by in vitro response type for frequencies Of
callus initiation and shoot regeneration.

mtg;

The effect Of BA concentration on (a) Mean callus initiation, shoot
regeneration and root regeneration, and (b) time to callus, lag time to
shoot regeneration and lag time to root regeneration summed over eight
genotypes; and on (c) callus initiation frequencies, (d) time to callus, (e)
shoot regeneration frequencies, and (1) root regeneration frequencies for
five genotypes (EL 45/2-96, EL 4999 and SP 6822-91 are not
shown)(Exp 1). Vertical bars represent 1: SE when larger than symbol.

Effect of concentrations of BA and GA, on mean frequency of callus
initiation summed over six genotypes (Exp 7).

The effect Of NAA on (a) mean callus initiation and shoot regeneration,
and (b) time to callus and lag time to shoot regeneration summed over
three genotypes; and on (c) shoot regeneration for three genotypes (Exp
3). Vertical bars represent :t SE when larger than symbol.

The effect Of NAA on (a) mean callus initiation and shoot regeneration,
and (b) time to callus, lag time to shoot regeneration summed over four
genotypes; and on (c) frequencies of shoot regeneration for four
genotypes (Exp 4). Vertical bars represent :1: SE when larger than
symbol.

Chapter 3

Flow chart depicting maintenance and challenge subculture regimes
utilized in the four experiments. The value adjacent each arrow
represents the total age of the callus at the time Of subculture. Exp 1
had no second challenge due to asynchrony Of treatments, and callus age
values in Exp 4 were slightly less due to 20 d subculture intervals.

(a) Shoot regeneration frequency and (b) shoot number per callus over

time for three genotypes in Exp 1. Vertical bars represent :1: SE when
larger than symbol.

xii

54

S7

59

It?

77

(a) Shoot regeneration frequency and (b) shoot number per callus over
time for three genotypes in Exp 2. Vertical bars represent :1: SE when
larger than symbol.

(a) Shoot regeneration frequency and (b) shoot number per callus over
time for three BA regimes in Exp 2. Vertical bars represent :1: SE when
larger than symbol.

(3) Shoot regeneration frequency and (b) shoot number per callus over
time for three levels of TIBA in Exp 3. Vertical bars represent :1: SE
when larger than symbol.

(a) Shoot regeneration frequency and (b) shoot number per callus over

time for four BA x TIBA treatments in Exp 4. Vertical bars represent 1
SE when larger than symbol.

xiii

INTRODUCTION

Plant breeding is the genetic manipulation of crop species by humans to fulﬁll human
needs. It’s successful application is trulyan artpasseddownoverthe centuriesfrom thetime
the ﬁrst cereals and legumes were domesticated. With the discoveries of Mendel and the
theories Of quantitative genetics, plant breeding developed a solid scientiﬁc foundation.
Modem plant breeders now have at their disposal a new array Of molecular and
biotechnological tools with which to enhance the efﬁciency of gene transfer. Some Of these
new tools, such as genetic transformation and protoplast techniques, allow for the movement Of
germplasm across natural biological barriers. In many cases, the application Of these new tools
to crop improvement is dependent on the ability to regenerate whole plants from in vitro cell
cultures. Thus, research efforts to develop plant regeneration systems are necessary
components in the progression of plant breeding into the biotechnological age.

Sugarbeet breeders are currently seeking methodologies to utilize the tools of
biotechnology in genetic improvement programs. A reliable sugarbeet shoot regeneration
system is needed for the application Of some in vitro genetic manipulations such as somatic
cell selection and protoplast techniques. A system Of shoot regeneration utilizing honnone-
autonomous callus Of sugarbeet has been developed at Michigan State University (Saunders
and Daub, 1984). The callus forms when leaf disks are incubated on Murashige and Skoog
(MS) medium containing 1 mg/L N‘-benzyladenine (BA) as the sole growth regulator (Bl
medium). For some genotypes, this callus subsequently regenerates shoots without subculture
(i.e. the system is one-step).

The following research on shoot regeneration from hormone-autonomous callus of
sugarbeet was divided into three manuscript style chapters, each of which addressed a distinct

1

2
aspect of the system: (1) genetic variation, (2) genotype x growth regulator interaction. and (3)
long-term regeneration. The ﬁrst two chapters involved the use of the one-step regeneration
system developed by Saunders and Doley (1986). In Chapter 1, a wide range Of germplasrn
was evaluated for in vitro behavior using the standard B1 system. Chapter 2 examined the
effects of BA concentration and supplementation of B1 with naphthaleneacetic acid and/or
gibberellic acid on shoot regeneration in a number Of sugarbeet genotypes. In this chapter the
essential features, i.e. one-step and honnone-autonomy, of the shoot regeneration system were
maintained. The ﬁnal chapter examined the effects of genotype, subculture interval, BA
concentration and 2,3,5-triiodobenzoic acid on shoot regeneration from serially subcultured
callus of three sugarbeet genotypes. The callus in Chapter 3 was hormone-autonomous, but
the regeneration was no longer one-step since subculturing was involved.

In addition to potential applications by sugarbeet breeders and geneticists, this research
also touched upon several concepts that are Of general interest to plant developmental
biologists and physiologists. These included genotype and growth regulator effects on the
initiation of the habituated (hormone-autonomous) state, and the regulation Of morphogenesis

from undifferentiated tissue.

LITERATURE REVIEW

Sugarbeet (Beta vulgaris L.), a root crop grown mainly in temperate regions, supplies
about 40% of the world supply of sucrose (Smith, 1987). Sucrose, or table sugar, is extracted
fromtherootsatlarge processingplantsnearthegrowing areas. In 1987, sugarbeets were
harvested from 8.6 million ha around the world, with an average yield of 31.2 Mg/ha (USDA,
1988). The leading sugarbeet producing countries include the Soviet Union, France, West
Germany, the United States and Poland. In the USA, sugarbeets are mainly grown in North
Dakota, Minnesota, Idaho, California and Michigan. Total USA acreage in 1987 was 3.9
million (1.6 million ha), with an average yield of 22.3 T/A (50.0 Mg/haXUSDA, 1988).

A member Of the Orempodiaceae, sugarbeet is an allogamous species with a typically
biennial life cycle. In the ﬁrst season, the plant grows a foliar rosette and sucrose is stored in
a large ﬂeshy taproot. Following an ovenvirrtering experience and exposure to long-day
conditions, a large bushy ﬂower stalk emerges on which as many as 5000 seed Often ripen
under favorable conditions. The fruit is a nutlet, and in the case of multiple ﬂowers at a node,
an aggregate fruit containing several seed is produced. The terms monogerm and multigerm
refer to fruits containing either one or several seed, respectively. Precocious production of a
ﬂower stalk during the ﬁrst season is refened to as bolting, and is highly undesirable in
sugarbeet production.
§QGARBEET BREEDING

Having been domesticated only in the past 200 yr, the sugarbeet is a relatively modern
crop. Following the discovery by Marggraf in 1747 that beets contained sucrose, Achard
developed a process for sucrose extraction and recovery from beets. Mass selection carried out
in fodder beet from 1786 to 1830 by Achard and the von Koppy family improved the sugar

3

4
corner! ofbeets from 6% to 9% (Smith, 1987). This resulted in the cultivar White Silesian,
droughtbymanytobetheprogenitorofallsugarbeets. ThehypothesisthatWhiteSilesian
arosefromintercrossesbetweenfodderbeetandchardmothB. vulgarisL.)issupportcdby
the efforts of Fischer (1989) to resynthesize the progenitor type.

Today’s modem sugarbeet hybrids can approach 20% sucrose under favorable growing
conditions. The genetic improvements facilitated by sugarbeet breeders within this short time
frame are thus a remarkable achievement. Although these improvements are partially due to
advancmentsincmmralpracticesmlambmdinghasmtdoubtedlyplayed akeyrolein
enhancing sugarbeet productivity.

Before the 1950’s, all commercial sugarbeet production involved Open-pollinated
multigerm populations. Due to the nature of the multigerm seed, thinning (stand reduction)
was a routine aspect of sugarbeet cultivation. Two key genetic discoveries, i.e. cytoplasmic
male sterility (CMS) and monogermness, led tO the evolution of today’s monogerm hybrid
sugarbeet cultivars. After the discovery of CMS and its maintainer (O-type) by Owen (1945),
and the monogerm trait by Savimky (1950), monogenn hybrid sugarbeet cultivars ﬁrst became
available in 1958 (Hecker and Helmerick, 1985). Monogerm sugarbeet hybrids with high rates
of germination allow precision planting, eliminating the need for the thinning Operation. A
typical hybrid results from a three-way cross involving: (a) a monogerm CMS inbred, (b) a
monogerm O-type inbred, and (c) a broad—based multigenn pollinator. The pollinator may be
diploid or tetraploid, giving rise to diploid and triploid hybrids, respectively. For detailed
descriptions Of sugarbeet breeding procedures, the reader is refened to reviews by Hecker and
Helrncrick (1985) and Smith (1987).

Breeding objectives in USA sugarbeet improvement programs include: sugar yield and
quality, disease resistance (e.g. Cercospora leaf spot incited by Cercospora beticola Sacc.,
Rhizoctonia root rot incited by Rhizoctom’a solaru' Kuhn, black root incited by Aphanomyces
cochlr’oides Drechsl., powdery mildew incited by Erysr'phe polygonr' DC., bacterial rot incited
by Erwinia carotovora (Jones) Bergey et al. asp. betavasculorum Thomson et al., curly top

5

virus, the yellows viruses and rhizomania), resistance to sugarbeet cyst nematode (Hererodera
schachtii Schmidt), and bolting resistance. Secondary or minor breeding Objectives include
herbicide tolerance, cold tolerance and resistance to insects (sugarbeet root maggot (Teraaops
nryopaefonnis Rider) and sugarbeet root aphid (Pemphigus populivenae Fitch). Additionally,
use Of a single source of CMS has resulted in worldwide cytoplasmic uniformity in sugarbeet
hybrids rendering the crop potentially vulnerable to genetic disaster. Thus, sugarbeet breeders
are searching fornew sourcesofCMS to alleviate this concern. Using traditional approaches,
plantbreedershavemade signiﬁcantprogressinmostoftheseareas,butinthefuturemanyof
these objectives will be addressed through a combination Of traditional and biotechnological
strategies In vitro genetic manipulations, arch as genetic transformation and somatic cell
selection, require that an efficient and reliable method is available for the regeneration of
whole plans from cell cultures.
SU ARBEET PLANT REGENERATION SY TE

Systems of in vitro plant regeneration can broadly be divided into somaclonal,
gametoclonal or protoclonal, relative to the starting material being somatic cells, garnetic cells
or protoplasts, respectively. Two recent reviews by Atanassov (1986a; 1986b) summarize the
various in vitro manipulations which have been applied to sugarbeet. The following discussion
of sugarbeet regeneration systems will be restricted to somaclonal systems, but the alternative
systems deserve brief mention. Largely unsuccessful efforts to regenerate sugarbeet plants
from anther culture (Rogozinska et al., 1977: Welander, 1974) have led to the develotlnent of
gynogenic systems for the production Of gametoclonal haploids. Following reports of low
frequency plant regeneration fmrn cultured ovules and ovaries (Bossotrout and Hosemans,
1985; Hosemans and Bossotrout, 1983), reliable protocols for the production of gynogenic
sugarbeet haploids have been developed (Doctrinal et al., 1989; Goska, 1985; Van Geyt et al.,
1987). Sugarbeet protoplasts have been successfully isolated and cultured (Bhat et al., 1985;
Bhat et al., 1986: Szabados and Gaggero, 1985). Although no protocol for regeneration from

protoplasts has been published, a report Of variation among sugarbeet protoclones implies that

6
the technology has been developed (Steerr et al., 1986).

Prior to 1984, shoot formation from somatic sugarbeet explarrts cultured in vitro was
unpredictable and obtained only sporadically (De Greef and Jacobs, 1979; Hooker and Nabors,
1977). Systems for reliable regeneration of whole sugarbeet plans are now available and
these can be grouped into four distinct approaches: regeneration fmm conventioml callus (Tetu
et al., 1987): legcneration from hormone-autonomous callus (Saunders and Daub, 1984;
Saunders and Doley, 1986: Saunders and Shin. 1986); production of self-regenerating lines
(Van Geyt and Jacobs, 1985); and direct regeneration from explants (Detrez et al., 1988;
Freytag et al., 1988; Ritchie et al., 1989). There have been no reports of regeneration Of
sugarbeet plants from cell suspension cultures.

Rmration from Conventional Callus

Conventional callus, as deﬁned here, requires an exogenous source Of both auxin and
cytokinin for induction and conﬁnued growth, while hormone-autonomous callus is capable of
sustained growth on medium devoid Of growth regulators. Tire infrequent regeneration
obtainedbyHOOkerand Nabors(1977)occurredaﬂercompact.greencallusinducedonauxin
and cytokinin was transferred to media with much lower levels of auxin. Regeneration did not
occur on the compact callus, but on a soft callus derived from it after transfer.

Tetu et al. (1987) Imve reported regeneration from sugarbeet callus by three distinct
protocols,andineachcaseanauxinwasusedtoinitiatethecallus. Compact,greencallus
initiated on various levels of cytokinin and auxin became friable and formed buds at low
frequency alter transfer to a regeneration medium also containing auxin and cytokinin High
intensity shoot organogenesis was Obtained when friable, green callus initiated convendonally
and transferred to media containing cytokinin and the anti-auxin 2.3.5-triiodobenzoic acid
(TIBA). They also reported a procedure for Obtaining somatic embryogenesis, but it included
several media and subcultures and appeared somewhat impractical.

7
n from H A C Ius

Hormone-autonomous sugarbeet callus is im'tiated on medium containing only
cytokinin, but is capable of sustained growth on hormone-free medium (Saunders and Daub,
I984). Initiation can also occur on hormone-free medium (Doley and Saunders, 1989), but the
time required to initiate callus is considerably longer in the absence Of the cytokinin stimulus.
The white, friable callus is morphologically distinct from the compact, conventional callus
described above. High-frequency regeneration can now be obtained using shoot culture leaf
explams (Saunders and Shin, 1986) or whole plant leaf explants (Saunders and Doley, 1986)
on MS medium containing 1 mg/L N‘-benzyladenine (BA). Shoot regeneration is one-step
(i.e. subculture is not required). When explants are incubawd at 28 to 31°C, callus initiation
typically occurs 4 to 6 wk post-inoculation. and subsequent shoot regeneration from callus
occurs after a lag period Of 1 to 3 wk (Sarmders and Doley, 1986). Using leaf blade and
petiole explants from in vitro shoot cultures, Saunders and Shin (1986) demonstrated that the
regeneration system involving hormone-autonomous callus is applicable to a wide range Of
sugarbeet germplasm.

Freytag et al. (1988) and Ritchie et al. (1989) described procedures for two-step
regeneration from callus morphologically similar to the hormone-autonomous callus described
above. Both procedures utilized shoot culture explants plated on medium containing BA, but
the procedure Of Freytag et a1. (1988) also included 0.1 mg/L indole-3-acetic acid (1AA).

R ration from If-R eratin L1

De Greef and Jacobs (1979) fortuitously obtained what they refer tO as a self-
regenerating line. The cell line was capable of continued growth on hormone-free medium,
butrequiredacytoldninforsuccessﬁrlshootproduction. Attemptstorepeattheprocedure
were notsuccessful. Plants have been regenerated from thiscellline aftermore thaneight
years in culture (Jacobs, personal communication).

A reproducible protocol for the initiation Of self-regenerating lines from a number Of
genotypes has been reported by Van Geyt and Jacobs (1985). The lines were produced when

8
callusatthebaseofprimaryshootswasusedforhabituatiorr. Theseauthorsobtained
habituation by successively halving the hormone concentrations in the medium. The self
regeneratinglineswererrotmrecallus,butwerecomposedofamixnrreofundifferentiated
cells, promcristems, meristerns and leaf-like structures. Total dedifferentiation of dress lines
resulted in loss of the regenerative capacity.

M Rmaﬂon from Emants

Early reports Of direct adventitious shoot regeneration in sugarbeet from in vitro shoot
cultures (Hussey and Hepher, 1978), from shoot culture derived leaf (Rogozimka and Goska,
1978) or petiole explants (Rogozirrska and Goska, 1978; Saunders and Shin, 1986) or using
ﬂower buds ﬁom greenhouse plants (Miedema, 1982) stimulated interest in these procedures
for clonal micropropagation as well as their potertial use in genetic transformation. Since
then, three systems Of high frequency direct regeneration have been developed (Detrez et al.,
1988; Freytag et al., 1988; Ritchie et al., 1989). These reports all utilized petiole explants
from in vitro shoot cultures, but differed considerably in the growth regulator regimes
employed. The medium used by Ritchie et a1. (1989) contained 10 mg/L BA, that of Detrez et
al. (1988) contained 3 mg/L BA and 3 mg/L NAA, while that Of Freytag contained 0.4 mg/L
BAand0.1mg/LIAA. Allthreereportsalsostressedtheimportanceofdonorculture
environment to Obtain Optimum response. For example, the procedure Of Detrez et al. (1988)
involved the use Of TIBA in the shoot culture medium.

Although the production of large numbers of regerrerants has been demonstrated using
these protocols, the likelihood that many of the shoot initials are preformed makes them
unlikely components of in vitro gene transfer systems (Krens and Jamar, 1989). However, the
genetic ﬁdelity of plants regenerated directly from explants (Detrez et al., 1989) suggests that
these procedures may ﬁnd application in sugarbeet breeding programs as a cloning vehicle.
GENETIC VARIATION FOR IN VITRO RESPONSE

Considering the quantitative natrrre of plant growth regulator interactions which result
in organ formation (Skoog and Miller, 1957), the numerous reports of genetic variation for in

9
vitro behavior should come as no surprise. Genetic differences for in vitro response have
apparentlybeenreportedineveryspecieswheletheyhavebeensought. Many studies,ranging
fromaralysesofprogenyofafewcromstofulldiallelmatingdesigns,have attemptedto
elucidatedregeneticcontrolofthevariousinvitro responses. Interestingly,theresu1tssuggest
thatthegeneticcontrolofinvitroresponserangesfromasfewastwomajorgenestoafully
polysenic systeln-

Examples Of crops where major gene activity has been reported include alfalfa (Reisch
and Bingham, 1980), cucumber (Nadolska—Orczyk and Malepszy, 1989), petunia (Izlrar and
Power, 1977) and tomato (Koornneef et al., 1987). In both alfalfa and tomato, two dominant
major genes were postulated to control shoot regeneration ability, while three dominant loci
werepostulatedincucumber. Inpetunia. ’afewgenes’ weretlrouglrttocontrolregerreration
from protoplasts.

On the quantitative side several ﬁxed effects diallels have been performed. Signiﬁcant
general and speciﬁc combining abilities have been reported for regeneration from anther
cultureinrlce (MiahetaL,1985)andwheat(Lazaretal.,1984), andforregerrerationfrom
callus in maize (Beckett and Qing, 1984), pigeonpea (Suresh Kumar et al., 1985) and tomato
(Frankenberger et al., 1981). In most reports, additive genetic variance was more important
than rronadditive genetic variance in controlling in vitro response. Heterosis has been reported
for callus growth in alfalfa (Keyes and Bingham, 1979), maize (Tomes and Smith, 1985),
pigeorrpea (Suresh Kumar et al., 1985) and tobacco (Keyes et al., 1981). Heritability estimates
for shoot regeneration include: 0.09 in cauliﬂower (Buiatti et al., 1974), 0.43 to 0.77 in
cucrrmber (Nadolska-Orczyk and Malepszy, 1989), 0 to 0.62 in maize (Beckert and Qing,
1984), 0 to 0.19 in sweet potato (Templeton-Somers and Collins (1986), 0.98 in tomato
(Frankenberger et al., 1981), and 0.60 to 0.72 in wheat (Lazar et al., 1984). Ranges of
eaimates usually involve different genetic material, btrt in some cases, detection of signiﬁcant
heritability was dependent on the medium that was used (Beckett and Qing, 1984)

The large number and variety Of cytogenetic stocks available in wheat have facilitated

10
detailed analysis of in vitro genetics in that crop. Studies involving substitution analysis,
monosomic analysis, arreuploids and trisomics suggest that loci effecting in vitro behavior can
be found on several chromosomes. Group 2 chromosomes (Felserrburg et al., 1987; Kaleikau
et al., 1989a; Kaleikau et al., 1989b; Szakacs ct al., 1988), chromosome 43 (Felsenburg et al.,
1987; Mathias and Fukrri, 1986) and chromosome 5B (Agache ct al., 1989; Felsenburg et al.,
1987) have all been reported to signiﬁcantly effect regeneration from callus.

In addition to nuclear control, cytoplasmic effects on in vitro behavior have been
reported in barley (Powell and Caligari, 1987), rapeseed (Narasimhulu et al., 1989) md wheat
(Felsenburg et al., 1987: Mathias et al., 1986).

From the many reports of genetic control of in vitro behavior, a few generalizations
may be drawn. In wheat, where breeding behavior and cytogerretic stocks have facilitated the
most thorough examination, it seems clear that major and minor nuclear genes, as well as
cytoplasmic factors, are involved in the in vitro response. One might speculate that major
genes affecting growth regulator metabolism interact with the overall quantitative nature of
growthitself,anditwouldbeexpectedthatasimilardualityofcontrolisinvolvedinthein
vitro response Of most species.

Further evidence for the genetic control of in vitro behavior comes from reports of
progress made through selection. In alfalfa, Bingham et a1. (1975) increased regeneration
respome from 12% to 67% in two cycles of recunerrt selection using vigorous regenerated
plants as parents. This selected population was released as 'Regen-S’. Petolirro et a1. (1988)
were able to increase anther culture response in wheat ﬁorn 3.5% to 23.4% in a single cycle of
selection involving a cross between two antIrer-derived plants. In tomato, Koomneef et a1.
(1986) successfully introgressed the shoot regeneration capacity of Lycopersicon peruvianunr
into L. escalentum.

In sugarbeet, information on the genetics of in vitro response is limited. Many strrdies
have involved explants derived from seedlings of hybrid cultivars germinated in vitro (Hooker
and Nabors, 1977; Krens and Jamar, 1989; Tern et al., 1987), and thus by deﬁnition are

1 1
analyses Of population phenomena. Irrterspeciﬁc variation within the genus Beta has recently
been described for ability to regenerate shoots from callus (Yu,1989)md fordirect shoot
regeneration (Majewska-Sawka and Jassem, 1988; Mikami et al., 1989).

Saunders and Shin (1986) evaluated the in vitro response Ofa broad array of sugarbeet
germplasm,aswellasscme fodderbeetleafbeetandredbeetmaterial. Usingleafbladearrd
petiole explants from in vitro shoot cultures, they observed extereive genetic variation for the
ability to initiate hormone-autonomous callus and subsequently regenerate shoots from that
callus. The response Of the sugarbeet material was superior to the small sample of related
subspecies, and within the sugarbeet germplamn, callus Of the monogerm material sampled had
a greater ability to regenerate shoots.

Developmcrx of the regerrerator clone REL-1 by conventional breeding procedures
(Saunders, unpublished) demonstrated that in vitro behavior in sugarbeet is heritable. Isolation
Of a somaclorral variant for in vitro behavior in sugarbeet provides further evidence of genetic
comrol (Saunders and Doley, 1986).

The availability of a sugarbeet monosomic addition series carrying chromosomes of B.
procumbens (Lange et al., 1988; Van Geyt et al., 1988), as well as a full series of trisomic
lines (Romagosa et al., 1986), should facilitate future efforts to elucidate the genetic control of
in vitro behavior in sugarbeet. Additionally, advances in isozyme (Nagamine et al., 1989a:
Smed et al., 1989; Van Geyt and Smed, 1984) and RFLP (Nagamine et al., 1989b) techniques
as molecular markers in sugarbeet may lead to localization and characterization of the genes
involved in response in culture.

GENOTYPE x MEDIUM INTERACTION FOR IN VITRO RESPONSE

Although reports of genetic effects in crrlture are common, reports of genotype x
medium (G x M) interaction are somewhat rare. Many reports of genetic variation in vitro
involve evaluation on two or more media, and thus contain information pertaining to G x M
interaction (Behki and Lesley, 1976; Izhar and Power, 1977; Oelck and Sehbder, 1983: Ozawa
and Komamine, 1989; Seitz et al., 1987; Zelcer et al., 1984). Although it is difﬁcult to

12
estimate the frequency of this type of interaction. it might safely be assumed that any crop
whicbdisplaysgeneticvariationforinvitrobehavioralsowilldispIayGxMinteraction. Gx
M interaction for regeneration has been reported in alfalfa (Nagarajan et al., 1986; Saunders
and Binglram, 1975), barley (Dunwell, 1981; Hanzel et al., 1985; Powell and Dunwell, 1987),
petunia (Skvirsky et al., 1984), rice (Quimio and Zapata, 1990) and wheat (Mathias and
Simpson, 1986). The practical effect of G x M interaction is to eliminate the possibility that a
singleinvitroculturesystemwillbeapplicabletotherangeofgermplasmwithinthespecies.

Mathias and Simpson (1986) reported G x M interaction in wheat tissue culture.
When they initiated callus from immature embryos with and without coconut milk in the
medium, they found that a single genotype had enhanced callus initiation with the coconut
milk. All other genotypes tested were inhibited by this addition to the medium.

Powell and Durrwell (1987) utilized a quantitative approach for the analysis of G x M
interactioninbarley. ThepotentialimpactofoMinteractionintissueculturewas
exempliﬁed by the character dry weight, where more than 50% of the variation was explained
bytherMcomponentofvariance. Ajointregressionapproachsimilartothatusedby
plarrtbreeders forstability analysis wasusedtocharacterize the ireeractions. Regression Of
genotype means acm media means wasusedtodetermirre whetherG xM interactions were
additive for each response parameter, and therefore of predictive value in a linear model. For
response parameters where the interaction was rronadditive, an analysis Of variance of the
square root Ofthe variance was used to estimate in vitro sensitivity. Thus the in vitro
performance of individual genotypes was assayed by both the mean response and the square
root of the variance component over four 2,4-dichlorophenoxyacetic acid (2,4-D)
concentrations.

There have been few reports of G x M interaction in sugarbeet Some studies have
involved a single genotype or hybrid (DeGreef and Jacobs, 1979; Hooker and Nabors, 1977;
Krebs md Jamar, 1989). In a study involving 15 sugarbeet genotypes, Jarl and Borrrrnarr
(1986) found signiﬁcant G x M interaction for callus growth. Evidence for G x M interaction

13
for shoot regeneration in sugarbeet comes from a recent report by Doley and Saunders (1989).
Threeplantsof’L53’,whichhadfailedtoregenerateshootsfromcallusinpreviousstudies
(see Chapter 1) utilizing 81 medium, regenerated shoots from callus on hormone-free medium.
LQNQT‘ERM REGENERATION FROM CELL CULTURES

The ability ofserially subculturedcallus orcell suspensionstomaintaintheir
regenerative capacity is refened to as long-term regeneration. The cultures may be maintained
inacontinuous stateofregeneration, orthey maybe inducedto regenerate by transferto a
regeneration medium. Alternatively, the ability to regenerate under standard regeneration
conditions may be lost and then restored by some manipulation of the culture environment that
was previously not required to obtain regeneration.

The successful application to plants of new genetic technologies such as DNA transfer
and somatic cell selection is dependent on the availability of reliable plan regeneration
systems. Regeneration of whole plants from protoplasts, suspensions or callus must be
maintained over a period of several weeks to momhs, most likely involving several
subcultures.

Loss of regenerative ability aher serial subculture could be the result of altered genetic
constitution of the culture or could result from the development of some new stable
physiological state in which the culture no longer responds to conditions that previously
induced morphogenesis (Smith and Street. 1974). The critical difference between these two
explanations for loss of competence is that the genetically altered cell line will remain
incompetent while the physiologically altered cell line should be capable of having competency
restored through manipulation of the culture environment.

Changes in the genetic constitution of plant cells in culture have been reported in
many species (Ahloowalia, 1983; Balzan, 1978; Bayliss, 1975; Lee and Phillips, 1988; McCoy
et al., 1982; Meins, 1983; Singh, 1975; Skirvin, 1978; Vapper and Kallak, 1986) and their
frequency of occurrence is apparently high. Variation resulting from these changes is refened
to as somaclonal variation (Larkin and Scowcroft, 1981), and the frequency of this variation

14
has been reported to increase with time in culture (Armstrong and Phillips, 1988; Kasperbauer
et al., 1979; Sutter and Langhans, 1981). Loss of organ-forming capacity in long-term cultures
has been directly related to increased frequency of aneuploidy and polyploidy (Torrey, 1967).
to increased nuclear DNA content (Chandlcr and Dodds, 1933). and to number of cell
divisions (Meyer-Teuter and Reinert, 1973). It should be mentioned that although somaclonal
variation is quite common, there are also examples of long-term cell cultures with relatively
stable chromosomal constitutions (Evans and Gamborg, 1982; Franklin et al., 1989; Mo et al..
1989; Nagl and Pfeifer, 1988).

The regeneration of somaclonal variant plants implies that the genetic constitution has
maintained some minimum level of organization relative to regenerative ability. In other
cases, the genetic changes occurring in vitro are sufﬁcient to preclude any further
morphogenesis. These typically involve gross cytogenetic alterations such as aneuploidy and
polyploidy. Habituation, the ability of a. callus or mspension to continue growth without the
addition of exogenous growth regulators, is an example of the development of a new stable
physiology (for review, see Meins, 1982). Habituation is an epigenetic phenomenon that is
inducible and often reversible. It seems likely tint many other stable physiologies develop
during cell culture that are not as easily detectable as habituation but might interfere with the
regenerative ability of the culture. The genetic totipotency of the cultured cells can be masked
by epigenetic events induced by the culture conditions and these events should be reversible by
modifying thou conditions.

Continuous regeneration in long-term culture has been reported in several species,
including barley (Weigel and Hughes, 1985), birdsfoot trefoil (Orshinsky and Tomes, 1985),
maize (Armstrong and Green. 1985), Microcitnrs spp. (Vardi et al., 1986), oats (l-leyser and
Nabors, 1982), pea (Hussey and Gunn, 1984), rice (Heyser et al., 1983; Ozawa and
Komamine, 1989), wheat (I-Ieyser et al., 1985) and white clover (White, 1984). Culture age in
these species does not seem to affect the ability to regenerate. In some cases, little effort was

necessary to maintain the regenerative state (Orshinsky and Tomes, 1985; Vardi et al.. 1986;

15
White, 1984), but in most reports competence was maintained by preferential subculture of a
pardcuhrcalhstypeorwasdepmdahmsomespecifmaspectofdmculmmsystem.

Competence is frequently maintained by preferential subculturing of regenerative
callus. In oats, embryogenic callus is morphologically distinct from non-embryogenic callus
(Heyser and Nabors. 1982). Embryogenic callus is also easily identiﬁed in wheat, proso millet
and pearl millet (Nabors et al., 1983). A similar situation exists in maize where Type I and
Type II callus, organogenic and embryogenic respectively, are also distinguished by callus
morphology (Armstrong and Green, 1985), and selection for callus type has prolonged
longevity (Lowe et al., 1985). Such morphological distinctions among callus types, while
undoubtedly quite practical. are somewhat fortuitous and seem unlikely to be of general
application.

Frequemly, variatiom in the culture system are found to affect the longevity of the
culture. Choice of explant can have a pronounced effect, as in the case of Stylosanthes
guyanensis (Aubl.) Sw., for which leaf-derived callus had lost tomatovc capacity after two
years in culture, but hypocotyl-derived callus was still highly morphogenic (Meijer, 1984). In
Afglurnpine,regenemﬁonﬁomcalhrshasbeenmahrtainedforuptomreeyearsmndthe
mgareraﬁmcapadtyisummdbycychngﬂncmwbetweenmebudmdmdonandﬂe
bud maturation media (Gladfelter and Phillips, 1987). Subculture frequency is another cultural
consideration which has been reported to affect long-term regeneration in begonia (Cassells
and Morrish, 1987), carrot (Bayliss, 1975), tobacco (Evans and Gamborg, 1982) and rice
(Ozawa and Kornamine, 1989).

The high frequency of genotype-dependent regeneration systems in many species
suggests that genotypic effects may play a role in successful long-term regeneration systems.
Although genotypic effects are common in cell cultures, there have been few reports of
genotypic effects on long-term regeneration. Callus of two pea genotypes continually
regenerated for three years, while three other genotypes produced little or no callus (Hussey
and Gunn, 1984). White (1984) identiﬁed a single white clover genotype, out of 200 tested,

16
that had a high regeneration capacity. This cell line maintained its morphogenic capacity aﬁer
two years in culture. These two examples of genetic differences in long term regeneration
both reﬂect genetic differences in ability to initiate callus and regenerate shoots in primary
cultures, rather than genetic variation for regeneration from long term cultures.

There have been at least three reports of genotypic effects on long-term regeneration.
Of 16 pea genotypes tested, Malmberg (1979) found that six regenerated shoots after two
monthsinculture. Aﬁerfourandsixmonthsinculmremnly fourandtwoofthesegenotypes,
respectively, were still capable of shoot regeneration. Locy (1983) evaluated six Lycopersicon
spp. mdfoundModytwospedesfomedsmotsaﬁermbcmwm,mdomofﬂnseretaimd
its regenerative capacity for more than one year. In maize, 6-month-old type 2 callus of B73 x
A188 regenerated more plants than A188 or A188 x B73 (Kamo and Hodges, 1986).

The most interesting reports of long-term regeneration are those in which the
regenerative ability had been lost and then restored by some manipulation of the culture
environment. In alfalfa, buds were induced on 32-month-old callus by a high cytokinin/low
auxin ratio (Stavarek et al., 1980). The callus had continually been challenged to regenerate
without success on a medium developed for alfalfa regeneration. Lowering the snow
concentration from 3% to 1% promoted further development of the buds. The authors suggest
thatthehighersucrosewasnecessaryasanenergysoureeforbudinduction,butthissame
level inhibited shoot elongation when the sucrose was thought to play an osrnoregulatory role.

Competence has been restored in long-term rice callus by osmotic manipulation of the
medium (Kavi Kishor and Reddy, 1986a; Kavi Kishor and Reddy, 1986b). When the rice
callus was maintained on medium with 2% sucrom and 3% sorbitol or mannitol, regeneration
did not decline after 1400 days (56 subcultures). Callus growing on 2% sucrose alone lost its
' shoot producing ability after 75 to 300 days, but was revived after 50 days on 2% sucrose plus
3% sorbitol or mannitol.

Callus cultures of haploid rape showed a decline in shoot production after the first
subculture (Sacristan, 1981). After nearly three years in culture, regeneration was induced by

17
transfer of the callus to a new medium under high light intensity, followed by tramfer back to
the original medium while maintaining the high light intensity. Since the maintenance medium
included 2,4-D and the induction medium did not. restoration of competence in this case may
ﬁt the 2,4-D withdrawal model of regeneration.

In the several reports of revival of regenerative ability in long-term callus, no clear
pattem emerges. In all cases, some form of stable metabolism was developed and then
overcome by an environmental modiﬁcation, but in each report the effective modiﬁcation was
different.

Little information is available regarding regeneration ﬁorn subcultured sugarbeet
callus. Research on regeneration in sugarbeet has focused on regeneration from primary callus
(Krens and Jamar, 1989; Saunders and Doley, 1986; Saunders and Shin, 1986) and from
secondary callus (Freytag et al., 1989; Hooker and Nabors. 1977; Tetu et al., 1987). Plants
have been regenerated ﬁom the original self-regenerating line of De Greef and Jacobs (1979)
after more than 8 yr in culture. However, because the self-regenerating lines are not true
callus but a leafy callus consisting of various levels of differentiation, it seems unlikely that
they will be useful for such procedures as genetic transformation or somatic cell selection.

There have been no reports of regeneration from conventional sugarbeet callus beyond
the ﬁrst subculture. Callus derived from excised ovules lost regeneration capacity after the
second subculture (Van Geyt et al., 1987). Saunders and Daub (1984) reported that hormone-
autonomous sugarbeet callus regenerated occasional shoots when callus maintained for three
monthly subcultures on hormone-free medium was transfened to medium containing 1 mg/L
BA and 0.3 mg/L IAA. Hormone-autonomous callus maintained on honnone-free medium
was capable of sustained growth for three 3 wk subcultures, but no shoot regeneration was
observed (Saunders and Doley, 1986).

CHAPTER 1
Genetic Variability for Frequency and Chronology of

Hormone-Autonomous Callus Initiation and Subsequent
Shoot Regeneration in Sugarbeet (Bear vulgaris L.).

W

A reliable shoot regeneration system is a prerequisite for the application of many in
vitro genetic manipulations. Some genotypes of sugarbeet (Beta vulgaris L.) are capable of
one-step shoot regeneration from honnone-autonomous callus (i.e. without subculture) when
leafdisksﬁomwlmleplmtsamhwubatedonMumshigeandSkwgmedimncontaiMngl
mg/L N‘-benzyladenine (Bl medium). The frequency of this ability within the sugarbeet
germplasrnpoolwasinvestigatedbytestingleafdisksfrom4t05plantseachfrom 16
sugarbeet populations. 8 multigenn and 8 monogenn. Leaf disks were incubated on Bl
medium at 31°C in darkness for 10 wk. Signiﬁcant differences among populations were found
for frequencies of callus production and shoot regeneration time to callus, and lag period
between callus initiation and shoot regeneration. Signiﬁcant differences among plants within
populations were also found for all four parameters, suggesting that selection for in vitro
behavior should be effective both within and between populations. Of 3018 leaf disk explants,
52.6% initiated callus in an average time of 43.0 d, and 30.1% of the calli regenerated shoots
after a mean lag period of 14.7 d. All 16 populations produced at least some callus, but ﬁve
populations failed to regenerate shoots. Overall, 76% of the plants produced some callus and
59% of these plants regenerated at least one shoot. The monogenn material had a shoot
regeneration frequency three times that of the multigerm material (41.9% vs 13.6%). The
conceptofinvitroresponsetypesisproposedhereasamechanisrntoefﬁcientlyoptimize
shoot regeneration from all members of a gennplasm pool. Clustering of the 16 powlations

18

19
byﬁequarciesofcaﬂushﬁﬂaﬁmmdmmgmemdonrendtedin7mspmuetypes
Differeru media may be required to optimize regeneration within each response type.

INTR ON

Sugarbeet (Beta vulgaris L.) supplies about 40% of the world’s supply of sucrose
(Smith, 1987). In 1987, sugarbeets were harvested ﬁom 8.6 million ha worldwide and yielded
a total of 296 million Mg (USDA, 1988). T‘he'continued production of superior sugarbeet
cultivars through genetic improvemem will partly depend on the ability to efﬁcierlly apply the
techniques of tissue culture and molecular biology in breeding programs.

mdwpastdecade,sugarbeaﬁssueculmmhaspmgressedﬁomamcalcimwstem
with minimal shoot regeneration to the availability of at least three distinct protocols for high
frequency shoot regeneration from callus. These include: (1) the production of self-
regenerating lines (De Greef and Jacobs, 1979; Van Geyt and Jacobs, 1985), (2) regeneration
from conventional auxin-dependent callus ('I‘etu et al., 1987) and (3) regeneration from
hormone-autonomous callus (Saunders and Doley, 1986; Saunders and Shin. 1986). It is not
clear whether somatic embryogenesis as reported by T‘etu et al. (1987) was from conventional
or hormone-autonomous callus. The regeneration system which utilizes hormone-autonomous
callus has two major advantages. First, regeneration can be one-step, i.e. callus initiation and
shoot regeneration can occur without subculture (Saunders and Doley, 1986). Secondly, the
technique seems to be applicable to a wide range of sugarbeet germplasm (Saunders and Shin,
1986).

SaundersandShin(1986)evaluatedtheinvitroresponseofawiderangeofsugarbeet
germplasm using petiole and blade explants derived from shoot cultures. Since then.
methodology has evolved to obtain one-step regeneration using leaf disks from whole plants
(Saunders and Doley, 1986). A regeneration system utilizing whole plants as explant donor's
has potential advantages relative to a breeding program. Application of gene tramfer or
somatic cell selection within elite germplasm may be more efﬁcient if leaf disk explants could
beobtaineddhecﬂyﬁomplantsgrowinginthegreenhouseorindreﬁeld. Furthermore, ifa

20
mmﬂmsystancmudbedevdopedwhichisappﬁabkmdnmgeofgamplmmme
breedingpool, it could alsobeused forvegetatlve propagation. Regenerant plantsthus
obtained could provide information on qualitative traits, such as some forms of disease
resistancebyessentially allowingaplanttobeintwoplacesatonetime. Theprocedure
would be restricted to the evaluation of qualitative traits since somaclonal variation a speciﬁc
locishouldbeinfrequent. Thistypeofeariyscreeningwouidbeparticularlyusefulin
sugarbeet breeding where initial seed supplies are ﬁequently limited, and where evaluation of
floral characteristics requires a minimum 10 wk vemalization treatmeru.

Theone-stepregenerationprocedurehastheunusual characteristicofa4to6wk
delay between explant inoculation and callus initiation, and a l to 4 wk lag period betweat
callus initiation and shoot regeneration (Saunders and Doley, 1986). If genetic variability
exists for this in vitro chronology, it may be possible to select for rapid, as well as higir-
frequency, regeneration.

The objective of this research was to evaluate the frequency and chronology of
initiation of hormone-summons callus, and subsequent shoot regeneration, from leaf disks
taken from greenhouse grown plants of a wide range of sugarbeet germplasm.

MATERIALS AND METHODS
Mt Magrig

Sugarbeet plants used as donorsofleafexplants were grown in soil in 15 cm plastic
pots in a greenhouse with supplemental lighting (provided by high pressure sodium lamps)
only during the ﬁrst 4 montiui. The soil used was a 2;2:1:1 mixture of Baccto professional
planting mix (Michigan Peat Co, Houston, TX, USA), a local greenhouse mix, vermiculite and
perilte, respectively. The local greenhouse mix consisted of a 53:2 mixture of ﬁeld soil of
variable origin, peat and sand, respectively. Plants were fertilized every 2 wk with 200 mL
Peters 20-20-20 water soluble commercial nutrient mix and monthly with Snyder‘s (1974)
nutrient formulation. At approximately 6 months of age, the plants were transfened to 25 cm
plastic pots and moved to a greenhouse without supplemental lighting.

21

AnplantsweregrownfmmseedplamedhrDecemberl985mdrepreseruadiveme
sample of USA monogerm and multigenn sugarbeet germplasm, including several parental
lines which have been used in commercial seed production. Five plants were sampled from
each of eight monogerm and eight multigerm populations. Seed sources and characteristics for
thesepopulationsarelistedinTable 1. Withtheexceptionsowaen’sAnnual(OA)md
84M5-20, each of these lines is a genetically heterogeneous population.
W

Explants were taken in March 1986 and December 1986 from approximately three-
month-old and one-yr-old donor plants, respectively. Two mall partially expanded leaves,
varyinginlengthfromSto 15 cm,perplantwereusedasthesourceofexplants. Detached
leaves were surface sterilized with two 20 min soakings in 15% commercial hypochlorite
bleach solution with 0.01% sodium laurylsulfate, followed by six rinses with sterile distilled
water. Squareexplantsvaryinginsizefrom4t05mm’werecutﬁomleafbladetissuewitha
scalpel. The same plants were sampled for the two experiments, and the two samples were
considered distinct environments, since the ﬁrst sample was in the Spring with plants
maintained under supplemental lighting, while the second was in the Winter without
supplemental lighting. .

The culture medium was MS (Murashige and Skoog, 1962) salts, 3% (w/v) sucrose,
100 mg/L myo-inositol, 0.5 mg/L nicotinic acid, 0.5 mg/L pyridoxine-HCI, 1.0 mg/L
thiamineHCl, 0.9% (w/v) Difco Bacto agar (Difco Laboratories, Detroit, MI, USA) and 1.0
mg/L N‘-benzyladenine (BA)(Sigma Chemical Co, St Louis, MO, USA), herein refened to as
B1. The pH was adjusted to 5.95 with KOH prior to autoclaving for 20 min at 121°C.
Thirty-ﬁve mL of medium was dispensed into each 15 x 100 mm Falcon Optilux dimosable
plastic Petri dish (Becton Dickinson & Co, Lincoln Park, NJ, USA) alter autoclaving. Ten
expiarrts were taken per leaf and a single explant was placed in each dish. The dishes were
sealed with two strips of Paraﬁlm (American National Can. Greenwich, CT, USA). Cultures
weremaintainedat31°Cinthedark. Atotalof1560and1500disheswereinoculatedinthe

22

 

 

Table 1. Seed sources and characteristics of populations studied.

ti 1 Ex 2 ‘ C
EL 36‘ 4 4 mm Breeding line; O—type
EL 401 5 5 MM Pollinator of USH 23
EL 442 5 5 mm Parent of cultivar; O-type
EL 45/21 5 5 mm Breeding line; O-type
EL 48’ 5 5 mm Breeding line
r=roos3 s 5 MM Breeding line
FC 506‘ 5 5 mm Breeding line; O-type
PC 607‘ 5 5 mm Breeding line; O-type
FC 701/5‘ 5 5 MM Breeding line
PC 708‘ 4 5 mm Breeding line
GWK’ 5 4 MM Commercial mangel beet
1.53‘ 5 4 MM High sugar breeding line
0A"2 5 4 MM Annual CMS tester
SP 68222 5 5 MM Pollinator of USH 20
SP 6926‘ 5 4 mm Parent of cultivar; O-type
84M5-20‘ 5 5 MM Trout pigmented genetic stock
Total 78 75

 

‘ MM denotes multigerm, mm denotes monogerm.

2 from W Saunders, East Lansing, MI, USA

3 from DF Cole, Fargo, ND, USA

‘ from GA Smith and RJ Hecker, Fort Collins, CO, USA

’ Garton’s White Knight from I Linde-Laursen, Roskilde, Denmark
‘ from JC Theurer, East Laming, MI, USA

7 Owen’s Annual

‘ from G Coe, Beltsville, MD, USA

 

23

two experiments, respectively.

 

The experimental design was completely randomized. Sampling was hierarchical and
completely nested, i.e. plants within populations, leaves within pints and samples (leaf pieces)
within leaves. For all four variables analyzed, the experimental unit was the leaf and the
responsewascalculatedasthemeanofthesamples withintheleaf. Dataoncallusirritiatiorr
and shoot regeneration were recorded every other day for 10 wk after the experiments were
iru'tiated. Only the moist. friable callus described in previous reports (Saunders and Daley.
1986; Saunders and Shin, 1986) was scored as callus; compact callus, which arose from
vascular tissue of some explants and did not continue togrow, was ignored Time to cirrus
was deﬁned as the number of days from explant irrocrrlation to ﬁrst visual observation of
callus. Lag time was deﬁned as the number of days between callus initiation and ﬁrst visual
observationofshootregeneration. Duetoexplantorpiamdeamdataarepresentedfroma
total of 78 and 75 plants for the two experiments, respectively. Only plants that were sampled
in both experiments were included in the combined analyses.

Statistical Analysis System Release 5.18 (SAS lrrstitute, 1985) was used for most
analyses. Data were generally unbalanced and handled by the General linear Models (GLM)
procedure. Populationsweretreatedasﬁxedeffects, whileplantsarrdleavesweretreatedas
random effects. The two frequency variables displayed variance heterogeneity and were
transformed using the arcsine function (Steel and Torrie, 1980). All means presented are
actual values, but mean separation was accomplished using least squares marginal means
computedbytheLSMEANSoptionofGLM. Leastsquaresmeansarethevaluesexpected
had the design been balanced. The PDIFF option of LSMEANS performed all possible t-tests
in a manner analogous to using an LSD. To avoid deﬂation of error terms, populatiom which
gave no response for a variable were not included in the analysis for that variable.
Approximate F tests involving synthesized mean squares were necessary in some cases
(Satterthwaite, 1946).

24

Poprdadanwaechsteredbyﬁequarciesofcaﬂusproducummdsimmgaremdon
from callusintire twoexperimerrts, i.e. byfourvariables. T‘imetocallus andlagtime were
notincluded. ClusteringwasaccomplishedusingtheaveragelinkagemethodofPROC
CLUSTER ofSAS. Thesmallestnumberofclusters which accounted foratleast90% ofthe
variation among populations was chosen.

W

LeafdiskexplantsplacedonBl mediumbegantoexpandwithin24h. Finalexplant
size,upto35timestheoriginalarea.wasmuallyreachedwithin 1 wk. Callusinitiationwas
ﬁrstobserved after21 d,butthe average timetocallus wasabout6wk. Themoist, white
friablecallusappearedtobetiresameasthatreportedbySaundersandDoley(1986)and
Sarmders and Shin (1986). The callus was hormone-autonomous; calli from all 16 populations
was capable of sustained growth on hormone-free medium. When shoot regeneration occurred.
it was after an average lag time of 2 wk. Blackening of leaf disks did not preclude callus
initiation or shoot regeneration. In some cases, rapid explant death occurred; these explants
failedtoexpandandlostallgreenpigmentationwithinZto3d. Rapidexplantdeathis
though to involve moisture and/or heat stress of donor plans.

Since the cultures were incubated in darkness, etiolation of regenerant shoots was
common. Shootswere frequentlyvitreous,buttheproportionvaried withgenotype. Some
non-vitreous shoots were readily converted into whole plants in all eleven populations which
produced shoots. Since rooting of shoot cultures is routine (Saunders. 1982), the critical step
for conversion to plants is the establishment of healthy, non-vitreous shoot cultures from the
regenerant shoots. Healtiry shoot cultures were sometimes obtained from semi-vitreous shoots,
but rarely so from vitreous regenerants. ‘

Reports of externive bacterial contamination of sugarbeet explants (Hooker and
Nabors, 1977; Miederna et al., 1980) have resulted in a general preference for the use of
aseptically cultured seedlings as explant sources (Ritchie et al., 1989; Tetu et al., 1987). In
our research. contamination of leaf disk explants taken from vegetative one-yr-old greenhouse

25
grownplantswasnotasigniﬁcantproblem. Overallcontaminationrateswerelesstim 1.3%
and no cases of systemic bacterial infection were observed. The use of young expanding
tissuemayhavelrelpedtomirrimizecontamination. Theseresultsdemonstratethatintact
vegetative sugarbeet plants are a viable alternative as a source of high quality. uniform
explains.
We

A1116 populations produced at least some callus in the ﬁrst experiment. and only OA
failed to initiate any callus in the second. The overall frequency of explants producing callus
was 55.5% and 49.5% for the ﬁrst and second experiments, respectively, with more than 75%
oftheplantsproducingsomecallusinbothCl‘ableZ). Mostpopulationsresporrdedsimilarly
in both experiments. Some exceptions included a marked increase in callus initiation by L53
andnotabledecreasesbyEL45/2andEL48,bothofwhichhadhighlankingsintheﬁrst
experiment.

More than half of the genotypes which initiated honnone-autonomous callus
regenerated shoots from the callus. Regeneration frequencies ranged from zero for ﬁve of the
populations to 92.5% for EL 45R (Table 3). Overall, 30% of the explants which produced
callus regenerated at least one shoot. Regeneration response was very consistent across the
two experiments, with the notable exception of SP 6926 where regeneration increased from
24.5% to 53.2%.

Data on shoot number were not recorded, but some populations were noticeably more
proliﬁc than others. Some populations such as EL 48 typically produced only a few shoots
per Petri dish, while others such as EL 45/2 were capable of producing dozens of shoots.
There was an inverse relationship between callus mass and shoot mass. Explants which
produced a large callus mass tended to regenerate few shoots, while those producing marry
siroots had a much smaller callus mass. This partitioning of in vitro biomass was highly
dependentongenotype. T‘heincidenceofvitreousshootsseemedtoincreasewithproliﬁcacy.

Of the four variables, time to callus was the least consistent across experiments. The

26

 

 

Table 2. Frequency of callus formation ﬁom leafdidrs on B1, and number ofgenotypes
which produced callus ﬁom leaf disks.
W WNW r f
1 Ex 2 M N E52 1 gm 2

SP 6926-O 95.9 98.8 97.2 a2 178 5/5 4/4
GWK 81.0 98.7 88.8 ab 178 5/5 414
EL 36 90.0 74.7 82.4 abc 159 4/4 4/4
EL 45/2 92.9 70.1 81.6 ab 196 5/5 SIS
PC 506 81.6 69.0 75.3 abc 198 5/5 5/5
EL 48 86.0 56.6 71.4 abc 199 5/5 5/5
FC 701/5 64.0 70.1 67.0 be 197 5/5 5/5
PC 607 69.7 62.0 65.8 be 199 4/5 4/5
SP 6822 59.8 54.1 56.9 ed 195 5/5 5/5
84M5-20 60.6 45.0 52.8 cde 199 ‘ 5/5 5/5
EL 40 39.4 40.4 39.9 cde 198 215 3/5
PC 708 26.6 16.0 20.7 def 179 2/4 4/5
F1003 26.0 11.0 18.6 ef 199 2/5 215
L53 2.0 37.9 15.6 of g 179 1/5 4/4
OA 7.2 0 4.0 f 177 2/5 0/4
EL 44 6.1 1.1 3.7 f 188 215 1/5
Mean 55.5 49.5 52.6

Total 3018 59/78 60ﬂ5

 

‘ Total number of explants mpled over both experiments.

2 Means in the same column followed by the same letter are not signiﬁcantly different
basedonat—testatPS0.05.

 

27

 

 

Table 3. Frequency of shoot regeneration from callus on Bl, and amber of genotypes
which regenerated shoots ﬁom callus.
R 'on ‘ Num r f
E 1 E Mean Egg 1 Egg 2

EL 45/2 97.8 85.3 92.5 a3 160 5/5 5/5
PC 506 67.5 59.4 63.8 b 149 .5/5 5/5
PC 607 60.9 62.9 61.8 b 131 4/4 4/4
EL 44 50.0 0 42.9‘ 7 2/2 - 011
SP 6926-0 24.5 53.2 37.6 be 173 5/5 4/4
SP 6822 24.1 22.6 23.4 c 111 4/5 3/5
FC 701/5 23.4 14.7 18.9 c 132 2/5 215
EL 48 14.0 12.5 - 13.4 c 142 4/5 3/5
PC 708 9.5 6.3 8.1 37 1/2 114
GWK 6.2 3.9 5.1 c 158 2/5 2/4
H. 36 6.9 0 3.8 c 131 1/4 0/4
84M5-20 0 0 0 105 0/5 0/5
EL 40 0 0 0 79 OR 0/3
F1003 0 0 0 37 0/2 on
L53 0 0 0 28 W1 W4
OA 0 0 0 7 0/2 0/0
Mean 30.9 29.2 30.1

Total 1587 35/59 29/60

 

‘ Frequency of shoot regeneration from callus ﬁom those explants which produced
callus.

2 Total number of explants which produced callus over both experiments.

3 Means in the same column followed by the same letter are not signiﬁcantly different
basedonat-testatPS0.05.

‘ Means not followed by a letter had non-estimable least squares means and could not
be statistically compared to other means.

 

28
overallmemdiffemdbySdmdﬂlemwassomemamnganauinﬂremltorderofﬂle
populationsCI‘able4). MeantimetocallusbyFC506,EL45l2andEL48 wasgreaterthan
10dlongerinthesecondexperiment. T‘imetocallusisviewedasthetime requiredbythe
cells intheleafdisktodevelopauxinautonomy. Thesigniﬁcantly longertime tocallusinthe
aecondexperimentmaybepartiallyduetotheincreasedageoftheplimtsandlor
envirormentally induced physiological variation.

Population means for lag time were highly consisteru across the two experiments (r =
0.90“). LagtimevariedfromjustmorethanaweekforEL45/2toonemonthforSP6822
(Table 5). Although the callus initiation frequency for EL 44 was quite low, EL 44 and EL
45/2 exhibited a rapid regeneration response distinct ﬁom the other populations. The three
populationswhichhadthelongestlagtimesallproducedverylargemassesofcallusandhad
only poor to moderate regeneration frequencies.

Althougir the population respomes for the four variables were highly correlated in the
two experiments (r = 0.79-0.90), the overall means for three variables changed signiﬁcantly.
Inthesecondexperiment,callusinitiatiorrwas6%lower,timetocalluswas$dlongerandlag
time was 1.3 d shorter (Tables 2, 4 and 5, respectively). In addition to differences in plan
age, the two environments differed in temperature, light intensity and day length, any one of
which may have inﬂuenced the altered response.

Iruerrelationships among the four variables were examined by correlation analysis
(Table 6). Populations with high callus initiation frequencies tended to initiate callus more
rapidly and have higher shoot regeneration frequencies. Populations with higher shoot
regeneration frequencies tended to have sirorter lag times. Although genetic parameters were
nmesdmawdmwcmuistaumspmueofﬂnpoptﬂaﬁommﬂnmoexpedmmmsuggestsﬂm
genotype exerts a major inﬂuence on in vitro behavior in sugarbeet.

The in vitro performance of the monogerm populations was generally superior to that
of the multigerm group. The monogerm material had higher ﬁequencies of callus initiation
and shoot regeneration. and when regeneration occurred it was after a sirorter

29

 

 

 

 

Table4. MeantimetocallusinitiationfromleafdisksculnuedonBl.

5g 1 Exp 2 Mm
Pomlgtion N TimegM) N T‘rmegdaﬁ) N Timgdgﬁ)
SP 6926-0 94 35.4 79 38.2 173 36.7 a2
GWK 81 34.0 77 40.0 158 36.9 ab
EL 48 86 34.7 56 45.2 142 38. 8 abc
FC 506 80 33.4 69 45.0 149 38.8 abc
SP 6822 58 40.1 53 39.6 111 39.9 bd
EL 45!). 92 37.5 68 49.0 160 42.4 ab
84M5-20 60 40.9 45 47.4 105 43.7 cd
PC 607 69 42.0 62 46.7 131 44.2 ab
FC 701/5 64 43.5 68 45.6 132 44.6 b
EL 40 39 47.0 40 50.9 79 490’
0A 7 49.0 0 -- 7 49. 0
EL 36 72 53.4 59 53.6 131 53. 5 d
F1003 26 53.8 . 11 52.8 37 53.5 d
L53 2 44.0 26 54.7 28 53.9
EL 44 6 52. 7 1 65.0 7 54.5
PC 708 21 59. 0' 16 56.7 37 58.0
Mean 40.5 45.9 , 43.0
Total 857 730 1587

 

Number of explants which produced callus

Means' in the same column followed by the same letter are not signiﬁcantly different
basedonat-testatPSOﬂS.

Means not followed by a letter had non-estimable least squares means and could not
be statistically compared to other means.

 

' 30

 

 

 

 

Table5.‘ Meanlagperiodbetwwncanushuﬁaﬁonandshootproducﬁonﬁomleafdisks

cultured on B1.

Exp 1 Exp 2 Mm

m N W N “111411.153 ) N MAME).
EL 44 3 7.3 0 -- 3 7.32
EL 45/2 90 8.0 58 6.9 148 7.6 a’
PC 506 54 16.8 41 15.2 95 15.4 b
EL 36 5 15.6 0 -- 5 15.6
PC 607 42 17.1 39 11.7 81 16.2 b
FC 701/5 15 19.6 10 14.2 25 17.4
SP 6926-0 23 17.9 42 17.6 65 17.7 b
PC 708 2 21.0 1 12.0 3 18.0 ab
GWK 5 25.2 3 21.3 8 23.7
EL 48 12 26.4 7 27.4 19 26.8
SP 6822 14 30.2 12 g 30.2 26 30.2
Mean 15.3 14.0 14.7
Total 265 213 478

 

‘ Number of explants which regenerated shoots from callus.

2 Meansnotfollowedbyaletterhadnon—estimableleastsquaresmeansarrdcouldnot
be statistically compared to other means.

3 Means in the same column followed by the same letter are not signiﬁcantly different
based on a t-test at P S 0.05.

 

Table 6. Correlations among population means for four parameters of in vitro behavior.

Regeneration (%) Time to callus (d) Qg time (p)
Callus (%) 0.56“ -0.69""" 0.12

 

Regeneration (%) 046“ -0.48"'
Time to callus (d) -0.39

 

*, “ Signiﬁcant at P S 0.05 and P S 0.01, respectively.

 

31

lag period (Table7). Genetic factors conditioning in vitro response may be localized in the
chromosomal region controlling the monogerm trait. or the gerrnrress locus could exert a
pleiotropic effect on in vitro behavior.
V ill W1 Po '1

Signiﬁcant differences were found botir among and within populations for all four
variables in botir experiments (Tables 8, 9, 10. and 11). In addition to the observed
quantitative variation, qualitative effects on callus initiation and shoot regeneration were also
observed within populations. In EL 40, two plants produced callus at high frequency (95 to
100%) in both experiments, while the other three plants failed to produce callus. The same
pattern occurred in F1003. In FC 701/5, one ofthe ﬁve plants accormted for 92% of the
regenerating calli, and in PC 607, four plants regenerated shoots in botir experiments, wirile the
ﬁftil plant never initiated callus. The same patterns of callus initiation were observed when
explants from these same plants were tested on hormorre-ﬁee medium (Doley and Saunders,
1989). Such bimodal responses indicate that genes with major effects on callus initiation and
shoot regeneration are segregating in some sugarbeet populations. This extensive genetic
variation should allow selection for in vitro behavior both within and between populations.
W V

The concept of in vitro response types .is proposed here as a tool for the classiﬁcation
of the members of a germplasm pool into discrete groups, each of which may require a
different protocol for optimum shoot regeneration. It was obvious from a cursory examination
of the data that similarities existed between some of the populations with regard to their in
vitro proﬁles. When the relationships among the 16 populations analyzed for in vitro behavior
were evaluated using cluster analysis, the populations fell into seven semi—discrete clusters,
hereafter refened to as response types 1-7 (Figure 1; Table 12). Response types 1 and 2
respond well to the B1 regeneration system which was utilized in this research, while the
regenerationresponsesoftypes 3,4 and5 populations mayrrotbeoptimal urrderthese
conditions. Types3and4tendedtorapidlyinitiatelargecalliandproduceonlyafewshoots

32

 

Table 7. Comparison of in vitro behavior of monogerm populations vs multigerm
populations.
M ' errn ' ' M rm
hm
9320.812)
Experiment 1 43. 5 ** 68.3
Experiment 2 44. 2 NS 56. 9
Combined 43.8 NS 62. 6
W
Experiment 1 15. 1 “ 43. 0
Experiment 2 12.1 " 40. 9
Combined 13. 6 NS 41.9
T1111; to ﬁns ((1)
Experiment 1 43.4 NS 41.5
Experiment 2 49.0 ~-‘ 48. 6
Combined 46.3 NS 45. 0
Qg Tim; (g)
Experiment 1 24. 7 "‘ 16.7
Experiment 2 l9. 6 -- 14.6
Combined 22. 4 NS 15.7

 

NS, *, “ Not signiﬁcant or signiﬁcant at P S 0.05 and P S 0.01, respectively.

Indicates comparison involved non-estimable least squares mean.

 

 

 

 

 

Table 8. Analysis of variance of frequency of callus production ﬁom leaf disks cultured
on B 1.
Egp 1 Exp 2
m (if M§(arcsine) df Mgarcsine)
Populations 15 2.263” 15 1.802“
Plants/Population 62 0.341“ 59 0.344“
Leaves/Plant 78 0.059 74 0.039

 

u Signiﬁcant at P s 0.01.

 

33

 

 

 

 

Table 9. Analysis of variance of frequency of shoot regeneration from callus on B1.
Egp 1 Exp 2

m M arcsine df Mﬂm’ )

Populations 10 1.858“ 8 1.353“

Plants/Population 36 0.232" 32 0.243“

leaves/Plant 44 0.031 37 0.041

 

** Signiﬁcant at P s 0.01.

 

Table 10. Analysis of variance of time to callus initiation from leaf disks cultured on B1.

 

 

 

Exp 1 Exp 2
m 9f ME) __df MS
Populations 15 390.7“ 14 2838‘
Won 43 86.0“ 45 128.6“
Leaves/Plant 53 17.8 51 20.8

 

‘, “ Signiﬁcant at P S 0.05 and P S 0.01, respectively.

 

 

 

 

 

 

Table 11. Analysis of variance of lag period between callus initiation turd shoot
regeneration from leaf disks cultured on B1.
Exp 1 Exp 2
m df MS g MS
Populations 10 240.3“ 8 260.4"
Plants/Population 24 75.7“ 20 81.6“
Leaves/Plant 23 18.6 19 14.7

 

‘, *" Signiﬁcant at P S 0.05 and P S 0.01, respectively.

 

Regeneration (%)

 

100

 

 

 

 

 

 

 

 

100
Callus ('5)
0 TYPE 1 A TYPE 2 0 TYPE 3 0 TYPE 4
t TYPE 5 X TYPE 0 V TYPE 7
Figure 1. Graph of population means by in vitro response type for frequencies of callus

initiation and shoot regeneration.

Table 12. In vitro response types, including member populatiom and their characteristics,
generated by clustering populations by frequency of callus production and

 

 

 

shoot regeneration.
Rm tyne Poml_a_ti_9_ns_ Callus (%) Regeneration (%)
1 EL 45/2 Good‘ Excellent
2 PC 506, PC 607 Good Good
3 SP 6926-O Excellent Moderate
4 EL 36, EL 48, FC 7015, Good Poor
GWK, SP 6822
5 EL 44 Poor Moderate
6 EL 40, 84M5-20 Moderate Poor
7 F1003, PC 708, CA, 1.53 Poor Poor

 

‘ Excellent = > 90%, Good = 55-90%, Moderate = 25-55%, Poor = < 25%.

 

35
atteralonglagperiod. Theﬁvepopulationswhichdidnotregenerateshootsineiﬂierofthe
samplings make up the type 6 and type 7 clusters. For'these populations, inalbation ofleaf
disksonBl iscenaiMysub-opﬁmal,hltdwymaybecapableofmre-stepsimmgareraﬁonif
the appropriue medium were employed.
D roar *

Reproducible, high-frequency shoot regeneration from hormone-autonomous callus was
observed in several sugarbeet populations, most noticeably EL 45l2. PC 506 and PC 607
(Table3). Plantsﬁomﬂresesamepoptnanonswemalsofomuimbegoodregenemmmwhen
explains were derived from in vitro shoot cultures (Saunders and Shin, 1986). One-step shoot
regeneration as described here was applicable to a wide range of germplasm; plants
represeruing 11 ofthe 16 populations testedregenerated shoots ﬁom callus. Although plants
of EL 45/2 were the most proliﬁc regenerators, the high incidence of vitreousness detracts
from their potential for in vitro genetic manipulatiors.

Overall favorable response rates coupled with minimal levels of contamination
demmmedﬂntgreaunmegrownvegetaﬁvealgameaplmmcmﬂdbemumofmgr
quality uniform explants. Use of whole plants as explant donors has distinct advantages in
breeding programs. In vitro behavior of breeding clones could be tested using available plant
material without the need to initiate shoot cultures. Breeding efﬁciency could be eniranced by
using regenerant plants as propagules to evaluate qualitative traits, such as some forms of
disease resistance, where bias due to somaclonal variation would be minimaL

Extensive genotypic variation was found for in vitro behavior in the populations
evaluated. The presence of signiﬁcant variation botir within and between populations, and the
corrsistentresponseoftirepopulationsinthetwoexperiments, suggestedthatgeneticgainfrom
selection for in vitro traits in sugarbeet should be possible. Effective selection for in vitro
performance has been practiced in alfalfa (Bingham et al., 1975), maize (Beckert and Qing,
1984), tomato (Koonureef et al., 1986) and wheat (Petolino et al., 1988). In sugarbeet. the
largely quantitative variation may be conditioned by qualitative factors as well. Populations

36
with signiﬁcant within population variability could provide good starting material to evaluate
the inheritance of in vitro behavior in sugarbeet, and might eventually allow isolation of genes
which condition callus initiation turd shoot regeneration.

The superior in vitro response of the monogenn populations over the multigerm
populationsCI‘able7)isconsistentwiththeresultsofSaundersandShin(l986). Allﬁve
populations which failed to regenerate shoots were multigerm, but some multigerm individuals
hadregenerationratescomparabletothebestmonogermgenotypes. Thismaybeareﬂection
ofanarrowergerieticbaseintiremonogerrnmateriaLbutcouldalsobeduetolirrkageor
pleiomriy-

Inviuoresponsetypesampmposedasamecimusmtoefﬁcienﬂyopdmizeslnot
regeneration from all members ofa germplasm pool. Each response type might require a
speciﬁc growth regulator regime to promote optimal levels of shoot regeneration. In theory,
once the speciﬁc medium required for each response type is identiﬁed, it should be possible to
screengennplasmonBl,assigneachtooneoftheresponsegrotmsandtoeffectivelypredict
themediumtobeemployedtooptimizeregerreration. Futureresearchendeavorsarenecessary
todeterminethevalidityofthe proposed concept.

We hypotiresize that shoot regeneration from hormone-autonomous callus of sugarbeet
involves complementatiorr between endogenous auxin physiology and exogenous cytokinin,
and is tirerefore compatible with the classic auxin/cytokinin ratio model (Skoog and Miller,
1957). Thus, genotypes ﬁom response types 1 and 2 are postulated to have an endogenous
auxin physiology which when combined with 1 mg/L BA produces an auxin/cytokinin ratio in
the range favorable for shoot regeneration. Genotypes in response types 3 to 7 may require
higher concentrations of BA and/or additional growth regulators for optimal shoot
regeneration.

The response of EL 44 provides some evidence that callus initiation and shoot
regeneration may be partially independent events. Explants from EL 44 initiated callus at a
low frequency (3.7%), but these rare calli had good regeneration potential (42.9%). In contrast

37
tosomeofthenon-regeneratingpopulations,theproblem withEL44seemstoliewithcallus
initiation rather than shoot regeneration. independent inheritance of callus initiation and plant
regeneration has been reported forwheatantherculture (Deatonetal.,1987;Lazaretal.,
1984).

Some of the unexplained variability encountered in this research may have been due to
gradients of endogenous growth regulators within the sampled leaves, as well as temporal
variability due to circadian rhythms. \Vrthin leaf gradients of indole-3—acetic acid have been
reported in orchardgrass (Wenck et al., 1988), and circadian rhythms have been shown to
inﬂuence endogenous levels of cytokinin in canot (Stiebeling and Neumanrr, 1987). In Lolilrm
multiﬂorum, Joarder et al. (1986) found that leaf segment position inﬂuenced the frequency of
callus initiation. In our research, no attempt was made to standardize the time of day when
leaves were harvested or to examine the response of explants from various positions on the
leaf.

Interest in regeneration systems involving leaf disks has been enhanced by successful
Agrobacterium—mediated transformation of leaf disk explants (Horsch et al., 1985). Sugarbeet
is susceptible to infection by A. fumefaciens (Huizing et al., 1988; Krens et al., 1988), and the
sugarbeet leaf disk regeneration system appears to be a suitable candidate for incorporation
into a transformation protocol. Because of its applicability to a wide range of germplasm, this
system of regenerating siroots from callus is also appealing for somatic cell selection within
elite breeding material.

CHAPTER 2
Genotype x Growth Regulator Interaction for

Callus Initiation and Shoot Regeneration
in Sugarbeet (Beta vulgaris L).

W
Leaf disk explants of many sugarbeet (Beta vulgaris L.) genotypes are capable of one-

step shoot regeneration from callus (i.e. without subculture) when placed on Murashige and
Skoog (MS) medium containing cytokinin. Genotype x medium (G x M) interaction was
investigated in 21 sugarbeet genotypes, representing 12 germplasm sources. as part of efforts
to expand the range of sugarbeet germplasm which can be regenerated from callus. We
investigated the effects of N‘-benzyladenine (BA) concentration and inclusion of
naphthaleneacetic acid (NAA) and/or gibberellic acid (6A,) on callus initiation and shoot
regeneration. An MS medium with 1.0 mg/L BA (Bl) was used as the control in all
experiments. Signiﬁcant G x M interactions were found in all seven experiments, including
interactions with BA, NAA and 6A,. Although frequency of callus initiation was somewhat
independent of BA concentration, genotypes varied in the BA concentration needed for optimal
shoot regeneration. Optimal levels of BA for shoot regeneration were above 1.0 mg/L for four
of eight genotypes tested. Some genotypes had enhanced rates of shoot regeneration from
callus when Bl medium was supplemented with low concentrations of NAA (0.01.0.1 mg/L),
but for most genotypes, NAA (0.1 mg/L) both inhibited and delayed callus formation and
reduced regeneration frequency. Addition of GA, (1.0 to 3.0 mg/L) to Bl medium
signiﬁcantly increased shoot number per leaf disk, up to 3—fold in some genotypes, and
signiﬁcantly inhibited root regeneration from callus. The frequency of signiﬁcant G x M
interactions detected in this research suggests that a single medium is inadequate when

38

39
screening sugarbeet germplasm for the ability to regenerate shoots from callus. Manipulation
of G x M interaction should lead to protocols for optimizing regeneration within in vitro
response types.

Abbreviatiog

BA, N‘-benzyladenine

6A,, gibberellic acid A3

MS. Murashige and Skoog medium
NAA, 1-naphthaleneacetic acid
INTRODUCTION

In a breeding program, efﬁcient application of molecular and cellular technologies may
require that the entire germplasm pool be amenable to in vitro manipulation. However, in
many crop species, only a subset of the breeding pool is generally capable of plant
regeneration from cell cultures (Hanzel et al., 1985; Malmberg, 1979; Ozawa and Komamine,
1989; Saunders and Bingham, 1972; White, 1984). Genetic transformation and somatic cell
selection would be more efﬁcient if accomplished within elite breeding lines, but this strategy
requires that plants can be regenerated from cells of such germplasm.

Genotype x medium (6 x M) interaction describes a nonadditive relationship between
the in vitro response of germplasm to different tissue culture media. If variability in shoot
regeneration frequency is due to G x M interaction, different genotypes may require different
media for optimum regeneration. G x M interaction for regeneration has been reported in
alfalfa (Nagarajan et al., 1986; Saunders and Bingham, 1975). busy (Dunwell, 1981; Hanzel
et al.. 1985; Powell and Dunwell. 1987), petunia (Skvirsky et al., 1984), rice (Quimio and
Zapata. 1990) and wheat (Mathias and Simpson, 1986), and might be an important tool for
broadening the range of genotypes within a species that can be regenerated.

Application of cell culture and recombinant DNA techniques to genetic improvement
of sugarbeet (Beta vulgaris L.) has been limited by low regeneration frequencies. Several

recem reports demonstrated that shoot regeneration from callus can be achieved over a wide

range of sugarbeet genotypes (Saunders and Shin. 1986; Chapter 1). In Chapter 1, plants from

40
ll of 16 sugarbeet populations sampled regenerated shoots when leaf disks were placed on MS
medium with 1 mg/L BA (Bl medium). Significant variation was reported for four parameters
that characterize sugarbeet regeneration from callus: callus initiation frequency, time to callus.
shoot regeneration frequency, and lag time between callus initiation and shoot regeneration.
Although many sugarbeet genotypes were regenerated from callus, shoot regeneration
frequencies varied by genotype from poor to excellent, and a portion of those tested failed to
regenerate.

Jarl and Bomman (1986) reported G x M interaction for callus growth in sugarbeet,
and suggested that this may partially explain the apparent recalcitrance of the crop in vitro.
Further evidence of G x M interaction in sugarbeet was reported by Doley and Saunders
(1989). They observed shoot regeneration from callus of three genotypes of the population
1.53 when leaf disk explants were incubated on hormone-free medium. These same plants
failed to regenerate shoots in two experiments utilizing B1 medium (Chapter 1). Thus,
different sugarbeet genotypes may require different media for optimal shoot regeneration.

Two unusual features characterize the sugarbeet regeneration system: (1) One-step (i.e.
without subculture) callus production and subsequent shoot regeneration. and (2) hormone
autonomy. Throughout this research, various media were explored with the intent of
maintaining these two attributes of the system. The objectives of this research were: (1) to
assess the magnitude of G x M interaction for in vitro performance in several sugarbeet
genotypes on a variety of media, and (2) to elucidate the medium components that might
enhance the in vitro performance of speciﬁc sugarbeet genotypes.

MATERIALS AND METHODS
Plgnt Material

Sugarbeet plants used as donors of leaf explants were grown in soil in 15 cm plastic
pots in a greenhouse without supplemental lighting or in 17 cm plastic pots in a controlled
environment chamber (CBC) at 25°C under a 12 h photoperiod from cool white ﬂuorescent
bulbs (100 to 200 pFJn'zs'l). The soil used was a 2:2:1:1 mixture of Baccto professional

41
planting mix (Michigan Peat Co, Houston, TX, USA), a local greenhouse mix. vermiculite and
perlite, respectively. The local greenhouse mix consisted of a 53:2 mixture of ﬁeld soil of
variable origin, peat and sand, respectively. Plants were fertilized weekly with 200 mL Peters
20-20-20 water soluble commercial nutrient mix and monthly with Snyder’s (1974) nutrient
formulation.

During the course of these investigations, the experimental procedure evolved from the
use of greenhouse grown plants to the use of shoot culture derived ramets (Saunders, 1982)
grown in the CEC. Three types of plant material were evaluated: plants derived from seed and
grown in the greenhouse, plants micropropagated by shoot culture and grown in the
greenhouse. and micropropagated plants grown in the CBC.

The genotypes evaluated were chosen on the basis of diversity for in vitro
performance. With the exceptions of BV-O and REL-1, the genotypes were individuals
isolated from the following breeding populations: EL 36, BL 44, BL 4512. EL 48 and SP 6822
(JW Saunders, East Lansing, MI, USA); FC 607-O, FC 701/5 and PC 708 (GA Smith and RI
Hecker, Fort Collins, CO, USA); ’Garton’s White Knight' (GWKXI Linde-Laursen, Roskilde.
Denmark); and SP 6926-O (G Coe, Beltsville, MD, USA). BV-O (TA Thorpe, Calgary,
Alberta, Canada) is an individual isolated from ’Primahill’, and REL-1 (JW Saunders, East
Lansing, MI, USA) is a clone bred and selected for superior regeneration ability from callus
and suspension cultures.

Culture Procedures

The seven experiments were initiated over a period of 1.5 yr with plants of variable
age and size between experiments, but uniform within experiments. Partially expanded leaves,
varying in length from 3.5 to 11 cm, per plant were used as the source of explants. Detached
leaves were surface sterilized with two 20 min soakings in 15% commercial hypochlorite
bleach solution with 0.01% sodium laurylsulfate, followed by six rinses with sterile distilled
water. Leafdisk explants were cut from leaf blade tissue with a No. 3 cork borer (7 mm i.d.),

except in Exp 1 where square explants varying in size from 4 to 5 mm2 were cut with a

42
scalpel.

The culture medium consisted of MS (Murashige and Skoog, 1962) salts, 3% (w/v)
sucrose, 100 mg/L myo—inositol, 0.5 mg/L nicotinic acid, 0.5 mg/L pyridoxine-HCI, 1.0 mg/L
thiamineilCl, 0.9% (w/v) Difco Bacto agar (Difco Laboratories, Detroit, MI, USA) and
various concentrations of BA, NAA and GA3 (all ﬁom Sigma Chemical Co, St Louis, MO,
USA). The pH was adjusted to 5.95 with KOH prior to autoclaving for 20 min at 121°C. All
growth regulators were added prior to autoclaving, except in Exp 6 where GA3 for 2
treatments was ﬁlter sterilized using a 0.22 p Acrodisc disposable 25 mm syringe ﬁlter
(Gelman Sciences, Ann Arbor, MI, USA). Bl medium (1 mg/L BA) was used as the control.
Thirty-ﬁve mL of medium was dispensed into each 15 x 100 mL Falcon Optilux disposable
plastic Petri dish (Becton Dickinson & Co, Lincoln Park, NJ, USA) alter autoclaving. The
number of explants taken per leaf varied (see Table 1) and a single explant was placed in each
dish. The dishes were sealed with two strips of Paraﬁlm (American National Can, Greenwich,
Cl‘, USA). Cultures were maintained at 28°C under dim light provided by cool-white
ﬂuorescent bulbs (20-40 uEm'zs“) in a controlled environment room.

Emrimental Desimg

The experimental designs were completely randomized genotype x medium factorials,
with hierarchical and completely nested sampling of plants within genotypes, leaves within
plants and samples within leaves. Table 1 summarizes some aspects of the experimental
designs, and the genotypes sampled are listed in Table 2. The details of the seven experiments
were as follows:

Experiment 1. Eight genotypes (EL 36-93, EL 45/2-96, EL 48-99, FC 607-O-92, FC 701/5-
94, FC 708-91, GWK-97 and SP 6822-91) were evaluated over a concentration gradient of BA
(0, 0.1, 0.3, 1.0, 3.0 or 10 mg/L). Plants were grown ﬁom seed and maintained in the
greenhouse, and all were part of germplasm previously evaluated on Bl medium (Chapter 1).
Explants were taken in August 1987. Because of variability in leaf size, two to six partially

expanded leaves were sampled ﬁom each genotype. Samples per leaf varied ﬁom six (one set

43

 

.2830 .85 B Baggage—28 u Um “853 8:8 5296 Ho 02:8
.5255 EoE=§>eu 3:258 u 08 ”8:9...on n =0 ”858:3:— 203 3:29 3:8 Emacs 223 «Snap—ram
4:25.35 :80 E com: 3598» 058% we a: .8 N 032. com

neon—ego :08 5 com: 238 828% no .385: 38.

 

 

 

 

Um emu m N o N _.<Q._<<z.._<a N
0m emu N N m n 38. e
33 :0 m _ N m 3.9 n
Um :0 m _ a e .52. v
8 =0 e _ N e :25 m
88 no N _ a N .55 N
88 :0 9N _ m e Ea _
l8§5> _ ”Bang”.—
.uuoEuwa 5 839.35

.5552 £32m x R598» 8.2.38 9 88859.» :38 2. he canoe Egg—toque he 3:88:80 088 no .5685 A 032.

44

Table 2. Genotypes sampled in the seven experiments to evaluate genotype x growth
regulator interaction in sugarbeet.

 

Exmriment

92m 1 3 4 § 6 7

IN

BV-O X

EL 3693 X

EL 36-98“ X

EL 44-90 X
EL 44-90“ X
EL 45/2-96 X
EL 48-97‘
EL 48-99
FC 607-0-20
FC 607-G92
FC 701/5-94
FC 701/5-94" X

PC 701/5-116 X

FC 708-91 X

GWK-93*

GWK-97 X

REL-l X X

SP 6822-90" X

SP 6822-91 X

SP 6926-0-3 X
SP 6926-0-94 X

><><><><
><><

 

* indicates genotype was a regenerant from callus.

 

45
of treatments) to 66 (eleven sets), and a total of 20 samples per genotype was initiated on each
treatment. The experiment was terminated after 14 wk.
Exmriment 2. Four genotypes (BL 45/2-96, EL 48-99. FC 607-092 and SP (5926-0-94) were
evaluated on Bl with NAA at 0 or 0.1 mg/L. Plants were derived from seed and maintained
in the greenhouse, and all were previously evaluated on Bl medium (Chapter 1). Explants
were taken in July 1987. Two leaves per plant and 12 explants per leaf were sampled,
resulting in six treatment sets per leaf. The experiment was terminated aﬁer 14 wk.
Bxperimglt 3. Time genotypes (BV-0, EL 44-90“ and REL-l) were evaluated on B1 with a
concentration gradient of NAA (0, 0.01, 0.03, 0.1, 0.3 or 1.0 mg/L). Plants were derived from
shoot cultures and maintained in the greenhouse. Explants were taken in June 1988. Four
leaves per plant and 12 explants per leaf were sampled, resulting in two treatment sets per leaf.
The experiment was terminated after 10 wk.
Exgrimgt 4: Four genotypes (EL 36-98", EL 48-97“, FC 701/5-94* and SP 6822-90*) were
evaluated on Bl with a concentration gradiem of NAA (0, 0.01, 0.03, 0.1, 0.3 or 1.0 mg/L).
Plants were derived from shoot cultures and maintained in the greenhouse, and all were
regenerants hour a previous evaluation (Chapter 1). Three leaves per plant and 12 explants
per leaf were sampled, resulting in two treatment sets per leaf. The experiment was terminated
after 10 wk.
W Three genotypes (EL 4512-96, FC 607-O-92 and EC 7015-94) were evaluated
on B1 with 0, 1.0 or 3.0 mg/L 6A,. Plants were derived from seed and maintained in the
greenhouse, and all were previously evaluated on Bl medium (Chapter 1). Explants were
taken in September 1987. Three leaves per plant and six explants per leaf were sampled
resulting in two treatment sets per leaf. The experiment was terminated after 14 wk.
merimﬂ 6. Three genotypes (FC 607-0-20, FC 701/5-116 and REL-l) were evaluated on
B1 with GA, at concentrations of 0, 1.0 or 3.0 mg/L autoclaved or 0.1 or 1.0 mg/L ﬁlter
sterilized. Plants were derived ﬁom shoot cultures and maintained in the CBC. Two plants

per genotype, two leaves per plant and ten explants per leaf were sampled, resulting in two

45
treatment sets per leaf. The experiment was terminated after 10 wk.
meriment 2. Six genotypes (EL “-90, FC 607-0-20, FC 701/5-116, GWK-93*, REL-l and
SP 6926-03) were evaluated with a 23 factorial involving 1.0 or 3.0 mg/L BA, 0 or 0.1 mg/L
NAA and 0 or 1.0 mg/L 6A,. Plants were derived from shoot cultures and maintained in the
CBC. Two plants per genotype, three leaves per plant and eight explants per leaf were
sampled. A single set of treatments was derived from each leaf, and each leaf was considered
a replication. Individual dishes in the experiment were terminated 3 wk after shoot
regeneration was observed. Dishes without callus were terminated after 10 wk, and those
which initiated callus but did not regenerate shoots were terminated 8 wk after callus initiation.
mg Analyses

Data on callus initiation and shoot regeneration were recorded weekly until the
experiments were terminated. The term callus will be used to describe friable callus which is
typically hormone-autonomous, in contrast to compact callus which is typically hormone-
dependent. Time to callus was deﬁned as the number of days from initiation of the
experiment to visual observation of friable callus. Lag time was deﬁned as the number of
days between initiation of friable callus and shoot regeneration. Shoot regeneration from
compact callus did not occur. Number of shoots regenerated per explant was determined at
the termination of most experiments, except in Exp 7 where they were counted 3 wk after
shoot regeneration was ﬁrst observed to provide a more valid comparison of factors effecting
shoot number. Bach shoot was judged for quality, with quality shoots deﬁned as having
relatively normal appearance, i.e. not developmentally deformed or vitreous, with a high
probability of resulting in a healthy plantlet.

In some experiments data on the following variables were also recorded: frequency of
initiation of compact callus, frequency of root regeneration from friable callus, and lag time
between initiation of friable callus and root regeneration.

Statistical Analysis System (SAS Institute, 1985) was used for most analyses. Data
were generally unbalanced and handled by the General Linear Models (GLM) procedure.

47

Genotypes and media were treated as ﬁxed effects, while plants, leaves and samples were
treated as random effects. The experimental unit for all variables was the leaf and respome
frequencies were calculated as the mean of the samples within the leaf. All frequency
variables displayed variance heterogeneity and were transformed using the arcsine function
(Steel and Torrie, 1980). All means presented are actual values, but mean separation was
accomplislwd using least squares marginal means computed by the LSMEANS option of
GLM. Least squares means are the values expected had the design been balanced. Mean
separation was accomplished using the PDIFF option of LSMEANS which performs all
possible t-tests in a marmer analogous to using an LSD. Approximate F tests involving
synthesized mean squares were necessary in some cases (Satterthwaite, 1946).
RESULT

Signiﬁcant G x M interactions were detected for some of the response parameters in
each of the seven experiments. Time to callus and frequencies of callus initiation and shoot
and root regeneration were each inﬂuenced by G x M interactions in three experiments. Lag
time from callus initiation to shoot regeneration was the only variable not affected by G x M
interactions.
El'fgLs of BA

Previous studies (Chapter 1) demonstrated that there was a diversity of in vitro
response types among sugarbeet populations when leaf disk explants were placed on MS
medium containing 1 mg/L BA (B1). In Exp 1, signiﬁcant differences among the eight
genotypes and the six BA concentrations were found for callus frequency, shoot regeneration
frequency, time to callus and lag time. All eight genotypes initiated callus and subsequently
regenerated shoots and roots.

The intermediate four BA concentrations (0.1, 0.3, 1.0 and 3.0 mg/L) resulted in the
induction of similar mean levels of callus frequency averaged over the eight genotypes (Figure
la). The frequency of callus production on hormone-free medium (53.8%) demonstrated that

high levels of callus initiation were possible without exogenous cytokinin. In contrast, the

Figure 1.

The effect of BA concentration on (a) Mean callus initiation, shoot
regeneration and root regeneration, and (b) time to callus, lag time to shoot
regeneration and lag time to root regeneration summed over eight genotypes;
and on (c) callus initiation frequencies, (d) time to callus, (e) shoot
regeneration frequencies, and (0 root regeneration frequencies for ﬁve
genotypes (BL 45/2-96, EL 49-99 and SP 6822-91 are not shown)(Exp 1).
Vertical bars represent 3: SE when larger than symbol.

A

Response Frequency ('5) 'r I

 

 

 

60" r ‘

 

 

 

 

4O
2O
0 ‘F i i i i 'l'
O 0.1 0.3 1.0 3.0 10.0
BA Concentration (mg/L)
‘9' Callus “9" Shoot Regeneration "9- Root Regeneration

J1me (weeks)

 

rm

 

 

 

 

 

 

0 I i i i
O 0.1 0.3 1.0
BA Concentration (mg/L)

4 __ l 3 ..
#F 4% .2

2
i t
3.0 10.

O

 

'9' Callus + Shoot Regeneration '9'- Root Regeneration

49

C

0 PIIIUI Initiation (Va) .

AA..—

   
   

 

 

 

 

 

0 ‘ i 1 1 l
O 0.1 0.3 1.0 3.0 10.0
BA Concentration (mg/L)

 

'0‘ EL 30-03 + F0 007-0-02 ‘9' F0 70115-04

+ F0 700-01 + GWK-OT

 

D

6 Time to Callus (wk)

 

 

 

 

l I
0 0.1 0.3 1.0 3.0 10.0
BA Concentration (mg/L)

 

‘9' EL 36-03 + F0 607-0-02 "3" F0 70115-04

+ F0 708-01 + GWK-O‘I

 

50

E

Shoot Regeneration (In)
100 ‘

 

  
 
 
 
  

so--

601"

40-—

20a

 

 

100

BA Concentration (mg/L)

 

'9' EL “-03 + F0 007-0-02 "9" F0 70115-04

4" F0 700-01 "5 GWK-OT

 

F

Root Regeneration (1a)
100’ “

a
V

80

'L

 

 

 

 

0 DJ 03
BA Concentration (mg/L)

 

'9' EL 38-03 + F0 807-0-02 ‘9' F0 701l5-04

+ F0 708-01 "b GWK-07

 

51
level of BA in the medium had a signiﬁcant effect on the frequency of calli which regenerated
shoots (Figure la). The overall optimum concentration was between 1.0 and 10.0 mg/L, with
the greatest respome at 1.0 and 3.0 mg/L. No shoot regeneration was observed on hormone-
free medium. In contrast to shoot regeneration, root regeneration from callus (Figure la) was
enhanced at lower levels of BA (0.1 to 1.0 mg/L). Direct root regeneration from leaf disk
explains was occasionally observed, particularly on media with S 1.0 mg/L BA (data not
shown).

The chronology of the in vitro events discussed above is depicted in Figure lb. Callus
initiation was most rapid at 1.0 and 3.0 mg/L BA, and was slower as BA concentrations
approached either 0 or 10.0 mg/L. The lag time between callus initiation and shoot
regeneration was slightly longer at higher levels of BA, while the lag time to root initiation
was signiﬁcantly shorter on either 0 or 10.0 mg/L BA, the same two concentrations which
delayed callus initiation.

G x M interaction was signiﬁcant in Exp 1 for callus initiation and shoot regeneration
frequencies and time to callus, but not for lag time. Since some genotypes had similar
response curves, only ﬁve of the eight genotypes are shown in Figure lc-f. The four
genotypes with the lowest rates of callus initiation (i.e. BL 3693, EL 45/2-94, FC 708-91 and
SP 6822-91) on hormone-free medium also had the lowest responses on 10.0 mg/L BA (Figure
1c). The three genotypes (i.e EL 48-99, FC 701/5-94 and GWK-97) with high rates of callus
initiation on hormone-free medium also had high rates of callus initiation on 10.0 mg/L BA.
The ability to initiate callus on hormone-free medium may be related to the ability to tolerate
high levels of BA.

G x M interaction for time to callus mainly involved differences in the delay of callus
initiation on 0 and 10.0 mg/L BA (Figure 1d). The two genotypes with the slowest callus
initiation on honnone-free medium (FC 607-O-92 and FC 708-91) also had low frequencies of
callus initiation on that medium (Figure 1c). FC 708-91 and SP 6822-91 both had delayed
callus initiation on 10.0 mg/L BA, while FC 701/5-94 was the only genotype whose time to

52
callus was not reduced on that medium.

The BA concentration which optimized shoot regeneration frequency varied with
genotype (Figure le). Two genotypes, FC 607-0-92 and FC 708-91 had their highest level of
shoot regeneration on 1.0 mg/L BA, while three others, BL 36-93, SP 6822-21 and GWK-97,
had optimum shoot regeneration on 3.0 mg/L BA. If these genotypes had been evaluated on
B1 medium, EL 36-93 may have been classiﬁed as a non-regenerator. Thus, Bl medium is
not the optimal level for all genotypes. The response of FC 701/5-94 suggests that some
genotypes may respond optimally to levels of BA greater than 3.0 mg/L.

Although root regeneration from callus was greatest on 0.1 mg/L BA, distinct response
patterns were observed (Figure 10. FC 607-0-92 and GWK-97 both had high rates of
rhizogenesis within the 0.1 to 1.0 mg/L BA range. Inhibition of rhizogenesis on BA 2 1.0
mglL was common for most genotypes, but FC 701/5-94 had its highest frequency of
rhizogenesis on 3.0 mg/L BA.

BA effects (1.0 vs 3.0 mg/L) were also evaluated in Exp 7. BA at 3.0 mg/L
signiﬁcantly reduced callus initiation, delayed lag time and reduced shoot mlmber. Although
the main effect of BA was not signiﬁcant for time to callus, REL-l callused 1.5 wk sooner on
3.0 mg/L BA (Table 3). The reduced callus production on 3.0 mg/L BA was mainly due to a
BA x GA, interaction. Medium containing 3.0 mg/L BA and 1.0 mg/L GA, were toxic
regardless of genotype or NAA concentration (Figure 2). In the absence of GA, in the
medium, the two BA concentrations gave almost identical levels of callus initiation. In
contrast, the reduction in shoot number by 3.0 mg/L BA vs 1.0 mg/L BA can be directly
attributed to the BA since GA, had a net stimulatory effect on shoot number (Table 4).
Effggts of NAA

We thought that addition of small quantities of NAA to Bl might enhance shoot
regeneration in some genotypes, without disturbing the hormone-autonomous nature of the
callus. Exp 2 evaluated the response of four genotypes on Bl with and without 0.1 mg/L
NAA. Supplementation of B1 with NAA signiﬁcantly increased frequencies of callus

53

 

 

 

 

 

 

 

 

 

 

 

 

 

533898.. .56 w a .8 86 w a a ﬁance—we .5 “sauce—we 8: «508 e823 8§obE 3.....mz
Boga»: £5 95.205 SEE co 3:8 83:5 9 88¢ 250:...» as we magnum a
o3. a... 8d an mz n: med mz £6 :82
8.Q «tilt ~rrtr 8.9 urri 8.Q at; tum
:k. a and mes a 8.0 end ml 3.0 STQSN. Um
m5. a... oﬂn 86 ml 3.0 . v06 m2 were 00.3% mm
8.» e... “6.? :6 m2 ~v.m mwé so and nit—mm
3.5 a... $6 $6 a New 86 m2 26 8.0.80 U"—
86 a... an 24. m2 5;. new a... n5” cad—30
Eu no...” no...
Julia 3 :1: c a 8 r j. a an .. ﬂat. 3 oﬁsltccoe
(lo 5.: <m
All; 3 9. roe—IL.

 

.ﬁnumvsaaosoaucoécsaﬁzeec ”88383.85

.m 039—.

54

 

Callus Initiation We)

.........................................................................

1007 \\\ .............. Ex , .............
80 it”.

 

 

 

 

 

 

..............

 

 

 

 

 

 

 

 

1.0 rug/L BA

\\\V 3.0 mg/L BA

0 mg/L GA 1.0 mg/L GA

 

 

 

 

 

 

 

 

 

 

 

 

 

. ////
/

 

Figure 2. Effect of concentrations of BA and GA, on mean frequency of callus initiation
summed over six genotypes (Exp 7).

 

 

 

 

 

 

 

 

 

Table 4. Effect of concentrations of BA and GA, on shoot number (Exp 7).
SW
EA GA;
Genome 1.0 mg]; 3.0 gig/L M 1.0 mgé
t~te§t t-tc
REL-l 22.17 “ 14.50 16.62 “ 25.90
SP 6926-03 19.10 ** 5.33 13.58 NS 16.36
GWK 93* 11.69 * 6.85 9.52 NS 9.35
FC 701/5-116 8.12 NS 5.22 4.38 ** 14.00
FC 607-020 7.44 NS 5.22 4.25 """ 12.33
BL 44-90 1.00 ----‘ ---‘ 1.00
Mean 13.94 ** 7.67 9.93 ** 14.61
‘ Explants of this genotype failed to initiate callus on media including this
treatment.
N83,“ Difference between means not signiﬁcant or signiﬁcant at P S 0.05 or P s

0.01, respectively.

 

55
initiation and shoot regeneration, and reduced time to callus (Table 5). Time to callus was
sharply reduced by more than 2 wk. G x M interaction was also signiﬁcant for callus
initiation frequency and shoot number. The overall increase in callus initiation from 0.1 mg/L
NAA was mainly due to the 91.7% response of SP 6926-0-94, vs no response on B1. The
other tluee genotypes had similar levels of callus initiation on both media, but all had non-
signiﬁcant increases in shoot regeneration frequency on 0.1 mg/L NAA (Table 5).

Exp 3 and 4 were designed to determine if there is an optimal level of NAA in B1,
and if this level varies with genotype. Both experiments involved Bl supplemented with
either of six NAA concentrations from 0 to 1.0 mg/L. In Exp 3, there was a linear decline in
callus initiation and shoot regeneration with increase in NAA concentration (Figure 3a). In
contrast to the friable callus, initiation of compact callus was enhanced by higher levels of
NAA. The initiation of both types of callus was strongly affected by genotype. REL-1
initiated friable callus at high frequency on NAA S 0.3 mg/L and only formed compact callus
on NAA 2 0.1 mg/L, with explants on 0.3 mg/L NAA commonly initiating both types of
callus. In contrast, every explant of EL-40* initiated compact callus, but friable callus was
only initiated on NAA s 0.01 mg/L. In vitro chronology was not affected by NAA, except
time to callus was longer on 1.0 mg/l NAA (Figure 3b). Although the G x M interaction for
shoot regeneration was not signiﬁcant, Figure 3c suggests that 0.01 mg/L NAA stimulated
regeneration in BV-O. Shoot regeneration from compact callus did not occur.

Exp 4 was identical to Exp 3 except that a different set of genotypes was evaluated.
Again, friable callus initiation and shoot regeneration declined with increasing NAA (Figure
4a), but the overall levels were higher than in Exp 3. Higher levels of NAA enhanced
production of compact callus (Figure 4a), and slowed the initiation of friable callus by explants
(Figure 4b). As in Exp 3, the effects on regeneration were mainly genotypic, but FC 701/5-
94" had increased regeneration on 0.03 mg/L NAA (Figure 4c).

NAA had few signiﬁcant effects in the factorial experiment (Exp 7). The only
signiﬁcant main effects were a reduction in shoot number and enhanced production of compact

56

 

 

 

 

 

 

53:8nt .86 w m no 3.: w m 8 88:83» 8 “88:86 8: 888 88:80 :5 8880888 8833 882085 3.....mz
.8388 8: .5 88:8: 8:8 02 n
.<<z .85 3 + 8 1.2838 .88 3 + m: u 8 a
8:8 .888: 883 3588 89: 88.: 8:8 88.: 80:8ng .85. Co 88:02"— _
SA 2. £6 Que ... N; :8 a... 98 :82
a.” at... he: at... Se a c 3.3% mm
whm s end 08 m2 an name m2 8: 8.80 U"—
mm.n m2 :2. 3m mz mdm 8: m2 0.3 8.: 4m
and a... QR 8: m2 m2. 8: m2 9% 0?an 4m
83..” l i. 82.: l i
:25 E :25 a 85080
33: mﬂmu 9 m8? 3% 8:88— 8:5

AN 9.8

8:8 9 08: :5 8:89.32 .8850 8838.: 828:8: usage >838: no 85:88 E 8 <42 .58 ﬁche .85 .n 035.

57

A

Rooponoo Frequency ('5)

 

 

 

 

 

 

 

80
60 ‘
40 ..
20 “r
0 t t t i u
0 0.01 0.03 0.1 0.3 1.0

NAA Concentration (mg/L)

 

'9' Frioblo Callus + Shoot Regeneration '9' Compact Callus

 

B

Tlmo (wk)
8

 

 

 

. N 4%

 

 

4
2
0 n i i t i
O 0.01 0.03 0.1 0.3 1.0

NAA Concentration (mg/L)

| '0'- Tirno to Callus 4" Lag Time I

 

: hoot Regeneration (1.)

100 I

 

58

 

 

 

80 : \
eo—~ '\
40::
20 '

0 l % '1- $

0.01 0.03 0.1 0.3 1.0
NAA Concentration (mg/L)
'9' REL-1 + EL 44-90- ‘9' BV-O
Figure 3. The effect of NAA on (a) mean callus initiation and shoot regeneration, and

(b) time to callus and lag time to shoot regeneration summed over three
genotypes; and on (c) shoot regeneration for three genotypes (Exp 3). Vertical
bars represent 1 SE when larger than symbol.

59

A

Reeponee Frequency ('5)

100 f V%

3+ { J

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2O ‘-
0 ‘L r i i i
0 0.01 0.03 0.1 0.3 1.0
NAA Concentration (mg/l.)
-9' Friabie Callue + Shoot Regeneration '9' Compact Cailue
Time (wit)
7
6 .- .E
5 j-
E i
4 ~r-
a -- L I
2 . N l 1-
T I .
1
o f r .L 1‘ i
0 0.01 0.03 0.1 0.3 1.0

NAA Concentration (mg/L)

‘ -0- Time to Caliue + Lag Time I

 

C

1 O _8hoot Regeneration (%)

 

 

 

 

 

 

 

I I I ‘
O 0.01 0.03 0.1 0 < 1.0
NAA Concentration (mg/L)

 

'9' EL 30-08' "7' EL 48-970 + F0 70115-940 ‘9’ SP 0822-000

Figure 4. The effect of NAA on (a) mean callus initiation and shoot regeneration. and
(b) time to callus. lag time to shoot regeneration summed over four genotypes;
and on (c) frequencies of shoot regeneration for four genotypes (Exp 4).
Vertical bars represent :t SE when larger than symbol.

61
callus. G x M interaction due to NAA involved signiﬁcantly delayed callus initiation for FC
6070-20 and FC 701/5-116. but not for the other genotypes tested (Table 3).
Effects f A

Sugarbeet shoots regenerated from friable, hormone-autonomous callus are frequently
vitreous. GA, was explored as a possible remedy to this problem (Miedema, 1984). Although
GA, did not seem to affect vitreousness. other interesting effects on in vitro behavior were
observed. In Exp 5, Bl supplemented with 1.0 or 3.0 mg/L GA, signiﬁcantly enhanced shoot
regeneration while inhibiting root regeneration (Table 6). For both of these variables, G x M
interaction was also signiﬁcant. Two genotypes had high levels of regeneration on all three
media, while FC 701/5-94 failed to regenerate without GA,.

In Exp 6. treatments with ﬁlter-sterilized GA, were included and the genotypes were
different. G x M interaction was again signiﬁcant for shoot regeneration. Shoot regeneration
by callus of FC 701/5-116 was increased from 50% to 100% by all four GA, treatments (Table
7). This genotype is a member of the same population as the genotype whose shoot
regeneration was enhanced by GA, in Exp 5.

Shoot number per leaf disk was signiﬁcantly enhanced by all four media which
contained GA, (Table 8). Regardless of whether it was ﬁlter-sterilized or autoclaved, GA, at
1.0 mg/L had similar results. The number of quality shoots per leaf disk was also enhanced
by GA, (Table 8). Since the ratio of quality to total shoots was not greater with GA,, the
increase in quality shoots reﬂects an increase in total shoot number rather than a reduction in
vitreousness.

In the factorial experiment (Exp 7), GA, signiﬁcantly inhibited and delayed callus
initiation, and inhibited mot regeneration from callus. Overall, shoot regeneration frequency
was excellent (94.7%) and none of the variables affected rate of shoot regeneration. GA,
signiﬁcantly enhanced the number of shoots (Fable 4) and reduced the lag time to shoot
regeneration.

GA, was involved in several interactions in Exp 7, including the toxic interaction with

62

 

 

Table 6. Effect of GA, concentration on frequencies of shoot and root regeneration
from callus of three genotypes (Exp 5).

Shoot Regeneration (% T) Root Regeneration (%)I
We l_31 BlGl l_31G3 Bl Bl 1 BIG
EL 45/2-96 100 a3 100 a 100 a 0 a 0 a 0 a
PC 607-092 87.5 a 100 a 87.5 a 87.5 a 12.5 b 50.0 ab
FC 701/5-94 0 a 75.0 b 87.5 b 75.0 a 50.0 ab 0 b
Mean 62.5 a 91.7 b 91.7 b 54.2 a 20.8 b 16.7 b

 

Frequency of shoot or root regeneration from callus from those explants which
produced callus.

1 Bl = MS +1.0 mg/L BA; 3161 = Bl+1.0 mg/L GA,; BlG3 = 131+ 3.0 mg/L GA,.

For either shoot or root regeneration, means in the same row followed by the same
letter are not signiﬁcantly different based on a t-test at P s 0.05.

 

 

 

 

 

 

Table 7. Effect of GA, concentration on frequency of shoot regeneration from callus of
three genotypes (Exp 6).
Shoot Regeneration‘
Genotype
QA, (mg) REL-l FC 607-0-20 FC 70115-116 Mean
0 100 a2 100 a 50.0 b 81.8 a
0.1 FS3 100 a 100 a 100 a 100 b
1.0 FS 100 a 100 a 100 a 100 b
1.0 A3 100 a 100 a 100 a 100 b
3.0 A 100 a 85.7 a 100 a 95.7 b
Mean 100 a 97.2 ab 90.0 b 95.7

 

Frequency of shoot regeneration from callus from those explants which produced
callus.

Means in the same row or column followed by the same letter are not signiﬁcantly
different based on a t-test at P S 0.05.

3 FS=ﬁlter sterilized; A=autoclaved

 

1*
, \

63

 

 

 

 

 

 

Table 8. Effect of GA, concentration on shoot number variables (Exp 6).
Shoot Ngmber
QA, (mgé) TotalI Quality2 Q-Ratio (%)3
0 14.0 c‘ 1.58 b 11.29
0.1 FS‘ 21.1 be 1.81 b 8.58
1.0 FS 29.5 a 2.78 a 9.42
1.0 A5 2&7 ab 2.83 a 10.60
3.0 A 22.6 ab 1.71 b 7.57
Mean 22.8 2.14
‘ Total number of shoots per leaf disk.

2 Number of quality shoots per leaf disk.

3 Ratio of number of quality shoots to total shoot number.

‘ Means in the same column followed by the same letter are not signiﬁcantly different
based on a t-test at P S 0.05.

5 FS=ﬁlter sterilized; A=autoclaved

 

BA mentioned above (Figure 2). Genotypes interacted with GA, for time to callus, shoot
number and root regeneration from callus. The interaction for root regeneration was primarily
due to GWK-93*, where rhizogenesis was reduced from 100% to 4.2% by 1.0 mg/L GA,.
DISCU§SION

Concentrations of BA as low as 0.1 mg/L were enough to substantially increase callus
production over hormone-free medium, however BA at 1.0 to 3.0 mg/L was needed to
optimize shoot production from the callus. The BA in the medium may serve to smd up the
initiation of an inevitable event, i.e. habituation. The explant requires only a minimal level of
cytokinin to stimulate callus initiation, but the highly organogenic hormone-autonomous callus
that is produced requires a 10-fold greater level of cytokinin in order to shift the organogenic
response in favor of shoot production. In general, the BA concentrations which stimulate

organogenesis are the same as those which stimulate callus production. Shoot production is

64
optimized on higher levels of BA, while root production is optimized on lower levels of BA.
Although somatic embryos were not encountered in this study, we might speculate that somatic
embryogenesis would occur near the BA concentration where root and shoot production are
equally likely (Figure la).

Two distinct callus types. friable or compact, were initiated by sugarbeet leaf explants,
but only the friable callus type regenerated shoots. Although some genotypes (e.g. EL 44-90)
initiated compact callus on B1 medium, supplementation with NAA greatly enhanced its
production, while inhibiting production of friable callus. In our hands, the friable callus has
always proven hormone-autonomous (Doley and Saunders, 1989; Saunders and Daub, 1984;
Saunders and Doley, 1986; Saunders and Shin, 1986; Chapter 1; Chapter 3). Three other
studies have reported the initiation of both friable and compact callus by sugarbeet explants
(Hooker and Nabors, 1977; Ritchie et al., 1989; Tetu et al., 1987), and, as in our research,
shoot regeneration from compact callus did not occur. In two of these (Hooker and Nabors,
1977; Tetu et al., 1987), compact callus was initiated on medium containing auxin and
cytokinin, but the shoot regemration obtained was from a friable callus derived from the
compact callus. Ritchie et al. (1989) obtained a white friable callus when leaf disks were
incubated on MS + 0.5 mg/L BA. This callus developed green areas which developed into
shoots after transfer to MS + 1.0 mg/L BA. The possibility that these regenerating friable calli
were hormone-autonomous was not tested in these three studies.

Sugarbeet genotypes appear to differ in the BA concentration required for optimal
shoot regeneration from friable callus. Cytokinin x genotype interactions affecting direct shoot
regeneration from explant have been reported in petunia (Skvirsky et al., 1984) and
cottonwood (Coleman and Ernst, 1989). Skvirsky et al. (1984) found that two inbred petunia
cultivars differed in the concentration of BA needed to optimize shoot regeneration, but the
response curves were fully reversed when zeatin was used. We have speculated that variation

in endogenous hormone levels and/or sensitivities conditions in vitro response in sugarbeet

65
(Doley and Saunders, 1989; Chapter 1; Chapter 3), but we have not tested genotype—cytokinin
combinations. Cytokinin structure-activity relationships, such as reported in Phaseolus spp.
(Mok et al., 1978) and possibly operative in petunia (Skvirsky et al., 1984), may be involved
in G x M interaction in sugarbeet.

Interactions between genotype and BA concentration may be useful in designing media
to optimize regeneration within the response types proposed in Chapter 1. The three
genotypes whose shoot regeneration was highest on 3.0 mg/L BA are all members of response
type 4. Thus, the poor regeneration frequencies observed for members of response type 4 in
Chapter 1 may have been enhanced had they been tested on higher levels of BA. A single
medium is therefore inadequate for evaluating sugarbeet germplasm for the ability to
regenerate shoots from callus.

Some genotypes had enhanced response when low levels of NAA (0.01 to 0.1 mg/L)
were included in B1 medium (Table 2; Figure 3c), but the overall effect seemed to be neutral
or possibly inhibitory. Although the effects of NAA on callus initiation and time to callus
observed in Exp 2 were not realized in the other experiments involving NAA, two genotypes
in Exp 7 had faster callus initiation with NAA. Thus, rather than routinely supplementing B1
with NAA, we recommend NAA use only for those genotypes or response types where a
stimulatory effect has been demonstrated.

The inconsistent results obtained with GA, suggest that this hormone plays diveme
roles in sugarbeet regeneration from callus. Supplementation of Bl with 1.0 mg/L GA,
signiﬁcantly enhanced shoot regeneration in Exp 5 and 6 mainly by boosting the regeneration
frequency of plants of FC 701/5. Since GA, inhibited shoot formation from tobacco callus
(Murashige, 1961; Murashige, 1964; Murashige, 1974), these results were unexpected. Janet
et al. (1981) found that GA, inhibited shoot meristem initiation from potato tuber disk
explants, but was required for shoot development. A similar phenomenon could explain the
response of sugarbeet callus to GA,. GA, may be stimulating shoot development, rather than

66
shoot initiation. Consistent increases in shoot number by GA, in Exp 6 and 7, as well as
reduction in lag time in Exp 7, support this hypothesis.

Strong inhibition of rhizogenesis suggested that the role of GA, was more complex
than simply organ elongation. If GA, interferes with the initiation of root meristems, their
development may be more semitive than that of shoot meristems. Regulation of
phytomorphology by the interaction of GA, and cytokinin has been reported for leaf shape irt
tobacco (Engelke et al., 1973) and for normal shoot development in beans with restricted root
systems (Carmi and Heuer, 1981). The simultaneous inhibition of roots and stimulation of
shoots by GA, resembles a cytokinin effect and may partially explain the toxic interaction
between BA and GA, observed in Exp 7.

Donor plant effects could have been involved in some of the inconsistent results with
NAA and GA,. Moderate shoot regeneration rates on B1, such as in Exp 2, might reﬂect less
than optimal condition of the donor plants, and might leave room for a stimulatory effect from
NAA or GA,. In Exp 7, the overall shoot regeneration rate was excellent and supplementation
with NAA or GA, could therefore not result in any improvement.

Further studies of G x M interaction in sugarbeet might involve inclusion of abscisic
acid (ABA), triiodobenzoic acid (TIBA) or ethylene antagonists. ABA has been reported to
enhance shoot production by cotyledonary explants of muskmelon (Neidz et al., 1989) and
loblolly pine (Sen et al., 1989), and TIBA appears to enhance regeneration from sugarbeet
callus (Tetu et al., 1987). Use of ethylene antagonists have been reported to increase shoot
production from explants of Chinese cabbage (Chi and Pua, 1989), somatic embryogenesis by
canot suspension cultures (Roustan et al., 1989) and shoot regeneration from maize callus
(Songstad et al., 1988). Sugarbeet leaf disks rapidly produce high levels of ethylene (Aharoni
and Yang, 1983). This ethylene production may vary with genotype and inhibit in vitro
response by the leaf disks.

Powell and Dunwell (1987) utilized a joint regression analysis to analyze G x M

67
interaction in barley. Regression of genotype means across medium means allowed
characterization of speciﬁc interactions. A similar approach might facilitate analysis of G x M
interaction in sugarbeet, and ultimately lead to physiological genetic interpretations of the G x
M phenomenon.

In conclusion, G x M interaction was found to be a major component of in vitro
response in sugarbeet. Information on G x M interaction should allow identiﬁcation of media
which will optimize regeneration of speciﬁc genotypes or of in vitro response types (Chapter
1). For example, it appears that shoot regeneration of response type 4 germplasm can be
enhanced by incubating leaf disks on 3.0 mg/L BA rather than the standard 1.0 mg/L BA.
Although only genotypes with a history of regeneration were evaluated in this research,
manipulation of G x M interaction may lead to regeneration of previously recalcitrant

sugarbeet germplasm.

CHAPTER 3
Effects of Genotype, Subculture Interval and Growth Regulators on

Shoot Regeneration from Serially-Subcultured
Hormone-Autonomous Sugarbeet (Beta vulgaris L) Callus.

W

Many genotypes of sugarbeet (Beta vulgaris L.) initiate hormone-autonomous callus
when leaf disks are incubated on Murashige and Skoog (MS) medium with 1.0 mg/L N‘-
benzyladenine (BA) as the sole growth regulator (B1 medium). When this callus is serially
subcultured on B1 medium, shoot regeneration frequency rapidly declines. We investigated
the effects of genotype, subculture interval, BA concentration and 2,3,5-triiodobenzoic acid
(TIBA) on shoot regeneration from serially-subcultured calli up to 18 wk old. Calli of three
genotypes were initiated from leaf disks on B1 medium and subcultured to various MS based
media after 3 wk growth. Competence was assessed by shoot regeneration frequency and
shoot number per callus on maintenance media as well as after subculture to challenge media.
When calli were subcultured every 3 wk on Bl, genotypes differed signiﬁcantly in rate of
decline in shoot regeneration. After 15 wk on B1, more than half of EL 45(2-108 calli were
still regenerating shoots, while regeneration by calli of REL-1 and FC 607-0-20 was
approaching zero. Although EL 45l2-108 had excellent long term regeneration, the extreme
vitriﬁcation of the regenerant shoots coupled with a very high shoot to callus ratio make this
genotype a poor choice for applications involving serially-subcultured callus. Subculture
interval did not effect subsequent shoot regeneration frequency, but calli subcultured more
frequently were lighter in color and appeared less senescent. Regeneration frequency from
calli maintained on B1 was increased after subculture to MS + 3 mg/L BA. The frequency of
calli regenerating shoots and the number of shoots per callus were both signiﬁcantly enhanced

68

69
by repeatedly doubling the BA concentration at each subculture or by maintenance on Bl + 1
mg/L TIBA. Calli of REL-l were generally more responsive than calli of FC 607-0-20 to
maintenance on TIBA. Increases in regeneration frequency were greater when concentrations
of both BA and TIBA were higher in the challenge medium relative to the maintenance
medium. Calli maintained in a non-regenerating state on honnone-free medium were induced
to regenerate by transfer to challenge medium containing 3 mg/L BA. Manipulation of shoot
regeneration with BA and TIBA appears to be compatible with a regeneration model involving
Abbrgviations
BA, N°-benzyladenine
MS, Murashige and Skoog medium
SC, subculture interval
TIBA, 2,3,5-triiodobenzoic acid
INTRODUCTION

In less than 200 yr, sugarbeet (Beta vulgaris L.) breeders have increased the sucrose
content of sugarbeets from 6% to more than 15%. Future increases in productivity will
partially depend on the ability of plant scientists to apply the new anay of biotechnologies to
sugarbeet. Many of these new procedures are dependent on the ability to regenerate whole
plants from cell cultures, and some, such as somatic cell selection and protoplast
manipulations, require that this regenerative ability be maintained over a period of several
subcultures in vitro.

Serially subcultured cell lines frequently lose their ability to regenerate whole plants
(Lustinec and Horak, 1970; Malmberg, 1979; Smith and Street. 1974; Toney, 1967). Loss of
regenerative ability alter serial subculture could be the result of altered genetic constitution of
the culture or could result from the development of some new stable physiological state in
which the culture no longer responds to conditions that previously induced morphogenesis
(Smith and Street, 1974). The critical difference between these two explanations for loss of

competence is that the genetically altered cell line will remain incompetent while the

70
physiologically altered cell line should be capable of having competency restored through
manipulation of the culture environment.

The most interesting reports of long term regeneration are those in which the
regenerative abith had been lost and then restored by some maniprrlation of the culture
environment. In alfalfa, buds were induced on 32-month-old callus by a high cytokinin/low
auxin ratio (Stavarek et al., 1980). The callus had continually been challenged to regenerate
without success on a medium developed for alfalfa regeneration. Competence was restored in
long term rice callus by osmotic manipulation of the medium (Kavi Kishor and Reddy, 1986a;
Kavi Kishor and Reddy, 1986b). Callus growing on 2% sucrose alone lost its shoot producing
ability after 75 to 300 days, but was revived alter 50 days on 2% sucrose plus 3% sorbitol or
3% mannitol. Callus cultures of haploid rape showed a decline in shoot production after the
ﬁrst subculture (Sacristan, 1981). After nearly three years in culture, regeneration was induced
by transfer of the callus to a new medium under high light intensity, followed by transfer back
to the original medium while maintaining the high light intensity.

In the several reports of restoration of regenerative ability in long term callus, no clear
pattent emerges. In all cases, some form of physiological adaptation was overcome by an
environmental modiﬁcation, but in each report the effective modiﬁcation was different.
Alternatively, when genetic instability is involved in loss .of competence, it may be possible to
minimize these changes through cultural procedures. Somaclonal variation has been shown to
increase with time in culture (Armstrong and Phillips, 1988; McCoy et al., 1982), and
shortening the subculture interval may be one means of reducing the variation (Cassells and
Morrish, 1987; Evans and Gamborg, 1982). Chandler and Vasil (1984) found that shorter
subculture intervals greatly enhanced the frequency of embryogenic callus formation in Napier
grass cultures.

Leaf disk explants from whole plants of some sugarbeet genotypes are capable of one-
step shoot regeneration. i.e. without subculture, from hormone-autonomous callus (Saunders

and Doley. 1986). Callus forms alter a minimum of 3 wk on B1 medium (MS 4» 1.0 mg/L

71
BA), and subsequent shoot regeneration occurs after a lag period of 1 to 4 wk. The response
of sugarbeet germplasm to the B1 regeneration system varies extensively (Saunders and Shin.
1986; Chapter 1). One-step shoot regeneration of some genotypes can be enhanced by
inoculating leaf disks onto medium containing 3 mg/L BA (Chapter 2). This suggests that the
explams, and possibly the callus initiated from them, differ in endogenous auxin physiology
and/or sensitivity to cytokinin.

The only report of shoot regeneration from serially subcultured hormone-autonomous
sugarbeet callus was by Saunders and Daub (1984). They observed occasional shoots when
callus maintained for three monthly subcultures was transfened to MS «I» 1 mg/L BA + 0.3
mg/L indole-S-acctic acid (1AA). Saunders and Doley (1986) reported increases in fresh
weight of calli subcultured to honnone-free medium, but no shoot regeneration was described.
Tetu et al. (1987) obtained regeneration from sugarbeet callus through three distinct protocols,
all of which utilized an auxin for callus initiation and at least one subculture before
regeneration was observed. One procedure, referred to as intense morphogenesis, consisted of
subculturing green friable callus initiated with naphthaleneacetic acid and BA to media
containing a cytokinin (BA or zeatin) and the anti-auxin TIBA. Since no treatments without
TIBA were reported, the effect of the TIBA was not clear. Hooker and Nabors (1977) also
reported a beneﬁcial effect of TIBA on sugarbeet shoot regeneration, but shoot regeneration
frequencies were quite low.

One plausible explanation for the decline in regenerative ability in honnone-
autonomous sugarbeet callus is a buildup of tolerance or decreased sensitivity to BA.
Altematively, the same decline might be expected if the endogenous auxin level in the callus
increases with time in culture, possibly as a direct result of habituation. It should be possible
to remedy either situation by transfer of the calli to media with higher levels of BA. Addition
of TIBA to the medium might also be capable of shifting the effective auxin/cytokinin ratio
into the range required for shoot regeneration.

The objectives of this research were: (1) to characterize the decline in shoot

72

regeneration over time of serially-subcultured, hormone-autonomous callus of three sugarbeet
genotypes, and (2) to evaluate the effects of subculture interval, BA concentration and TIBA
on maintenance of the shoot regeneration ability of these calli.
MATERIAQ AND METHOD§
Plant Material

Sugarbeet plants used as donors of leaf explants were grown in soil in 18-cm pots in a
controlled environment chamber at 25°C under a 12 h photoperiod from cool white fluorescent
bulbs (100-200 uEm‘s"). The soil used was a 2:2:1:1 mixture of Baccto professional planting
mix (Michigan Peat Co, Houston, TX, USA), a local greenhouse mix, vermiculite and perlite,
respectively. The local greenhouse mix consisted of a 53:2 mixture of ﬁeld soil of variable
origin, peat and sand, respectively. Plants were fertilized weekly with 200 mL Peters 20-20-20
water soluble commercial nutrient mix and monthly with Snyder’s (1974) nutrient formulation.

The three genotypes evaluated are all known to have good regeneration capacity
(Saunders and Shin, 1986; Chapter 2). FC 607-0-20 and EL 45/2-108 are individual plants
isolated from the breeding poprrlations FC 607-O (GA Smith, Fort Collins, CO, USA) and EL
45l2 (JW Saunders, East Lansing, MI, USA), respectively. REL-l (JW Saunders. East
Lansing, MI, USA) is a clone bred and selected for superior regeneration ability. Three plants
of each genotype were produced from shoot cultures by the procedure developed by Saunders
(1982), and the same plants were used in the four experiments. Data for EL 45/2-108 is
presented for Exp 1 and 2 only. Use of EL 45’2-108, an unusually proliﬁc regenerator, was
discominued midway through Exp 3 because the time required to subculture callus and count
shoots (up to 200 per dish) of this genotype became prohibitive.

ulture Procedures

Explants for Exp 1, 2, 3 and 4 were taken in March, June, August and October 1988,
respectively, from 4-, 7-, 9- and ll-month—old donor plants, respectively. Two small partially
expanded leaves (three in Exp 4), varying in length from 3.5 to 10 cm. per plant were used as

the source of explants. Detached leaves were surface sterilized with two 20 min soakings in

73
15% commercial hypochlorite bleach solution with 0.01% sodium laurylsulfate, followed by
six rinses with sterile distilled water. Leaf disk explants were cut from leaf blade tissue with a
No. 3 cork borer (7 mm i.d.).

The culture medium consisted of MS (Murashige and Skoog, 1962) salts, 3% (w/v)
sucrose, 100 mg/L myo-inositol, 0.5 mg/L nicotinic acid, 0.5 mg/L pyridoxine-HCI, 1.0 mg/L
thiamineilCl, 0.9% (w/v) Difco Bacto agar (Difco Laboratories, Detroit, MI, USA) and
various concentrations of BA and TIBA (both Sigma Chemical Co, St Louis, MO, USA). The
pll was adjusted to 5.95 with KOH prior to autoclaving for 20 min at 121°C. Thirty-ﬁve mL
of medium was dispensed into each 15 x 100 mm Falcon Optilux disposable plastic Petri dish
(Becton Dickinson & Co, Lincoln Park, NJ, USA) after autoclaving. Eight explants (ten in
Exp 4) were taken per leaf and a single explant was placed on each dish. The dishes were
sealed with two strips of Paraﬁlm (American National Can. Greenwich, Cl‘, USA). Cultures
were maintained at 28°C under dim light provided by cool-white ﬂuorescent bulbs (20-40
pEm"s") in a controlled environment room.

Individual leaf disk explants initiated callus at from one to many sites 3 to 6 wk after
inoculation. For subculturing. all callus on a leaf disk was treated as a single unit. Callus
formed on individual explants was subcultured to fresh media when it was 21 (1 old. Each
initial callus was removed from the explant and subdivided into three calli of 1 to 1.5 cm
diameter which were placed equidistantly around the perimeter of the Petri dish. Calli from
two explants per leaf were subcultured to each of the three treatments, except in Exp 4 where
callus from one explant per leaf was subcultured to each of the eight treatments. Calli were
subcultured to fresh media every 21 d unless otherwise noted. Callus age refers to the total
length of time since callus was ﬁrst observed on the leaf disk, regardless of the number of
subcultures performed. All references to time (e.g. regeneration alter 15 wk) are synonymous
with callus age.

Media on which calli were maintained were refened to as maintenance media, while

media used to assess regenerative ability were refened to as challenge media. Bl medium was

74

the maintenance control medium in all experiments. In many cases, portions of calli were
simultaneously subcultured to both maintenance and challenge media. Regenerative ability on
challenge media was assessed alter 3 wk. Through the course of the four experiments, the
challenge medium evolved from Bl to MS + 3 mg/L BA (B3) and ﬁnally to B3 supplemented
with TIBA.
Egpgriggntal Desimg

The experimental design for callus initiation was completely randomized, with
hierarchical and completely nested sampling, i.e. plants within genotypes, leaves within plants
and samples within leaves. In each of the four experiments, the calli were randomly
subcultured to the various genotype x medium treatments, which were arranged factorially with
the same nested sampling structure used for callus initiation. Table 1 summarizes the design
of the four experiments. while Table 2 provides the BA and TIBA contents of the various
media which were used. See Figure 1 for a ﬂow chart of the callus maintenance and challenge
regimes. The details of the four experiments were:
Exgriment 1. Shoot regeneration of three genotypes was evaluated after maintenance on MS
+ 1.0 mg/L BA (Bl), with calli subcultured at 7, 14 or 21 d. Shoot regeneration ability was
assessed at 9 wk by challenge on B1 and at 15 wk by challenge on MS + 3.0 mg/L BA (B3).
Exgriment 2. Shoot regeneration of three genotypes was evaluated after maintenance on
honnone-free MS (M80), B1 or a regime in which BA concentration in the medium was
doubled at each subculture. The doubling regime began with 2.0 mg/L BA (82) at the ﬁrst
subculture and ended with calli transferred to 32.0 mg/L BA (B32) at 15 wk. Calli maintained
on MS or B1 were challenged to regenerate at 9, 12 and 15 wk by transfer to B3.
Exmriment 3. Shoot regeneration of three genotypes was evaluated after maintenance on Bl
supplemented with 0, 0.1 or 1.0 mg/L TIBA (B1, B1T.1 and BlTl, respectively). Evaluation
of EL 45/2-108 was terminated after 9 wk of maintenance. All calli were challenged at 9 and
12 wk by transfer to B3, and at 15 wk by transfer to B3 + 1.0 mg/L TIBA (B3Tl).

Exmriment 4. Shoot regeneration was evaluated in a 2‘ factorial experiment involving two

75

5.3 a sea N .5 son 8.5.85 33 we NE. gm .
.38 mm .438 2 JR... w .438 v :58 u :38 _ .3 83:68» 88 s conga». <m 2: 8:95: .8 25m»: a 9 E38 xm N

.3338...“ e828 88 No cones: 2: ”3:35 23383 n Um .

 

 

 

 

 

 

 

 

 

gas 3 .o 2E:
.88 3 .o 2a
e 8 .2 8 N N N v
.38 3 .8 .c 2a.: N N .N N
.35 .xN .3 .o 2...: N N N N
e _N .3 .N .8 N N m _
mange... §€E> .3284 0?: gal: 6e 7: Ease am
8:8 .893?»

wee—898-388 82.. 8222ch .85 38:26 9 8:08:85 38 Lo 2325» «sag is $8.58: No .5883 ._ 058.

76

 

 

 

Table 2. BA and TIBA contents of the various media used throughout the four
experiments.
Medium BA (my!) TIBA m
M80 0 0
B1 1.0 0
B2 2.0 0
B3 3.0 0
B4 4.0 0
B8 8.0 0
816 16.0 0
B32 32.0 0
B1T.1 1.0 0.1
BlTl 1.0 1.0
B3Tl 3.0 1.0
8312 3.0 2.0
T1 0 1.0

 

genotypes (FC 607-020 or REL-1), two subculture intervals (10 or 20 d), two BA
concentrations (0 or 1.0 mg/L)and two TIBA concentrations (0 or 1.0 mg/L). Subculture
intervals of 10 and 20 d were chosen so that regeneration ability of the calli could be assessed
every 20 d. Calli were challenged at 61 and 81 d by transfer to B3T1 and at 101 d by transfer
to B3 with 2.0 mg/L TIBA (B3'I‘2).
Data Analysis

Data on callus initiation and shoot regeneration were recorded every other day until the
callus was ﬁrst subcultured. Time to callus was deﬁned as the number of days from explant
inoculation to visual observation of callus. Lag time was deﬁned as the number of days
between callus initiation and visual observation of shoot regeneration. Number of shoots
regenerated on individual calli was recorded at each subculture. Each shoot was judged for
quality. with quality shoots deﬁned as having relatively normal appearance, i.e. not
developmentally deformed or vitreous, with a high probability of resulting in a healthy
plantlet. The ratio of quality shoot number to total shoot number was refened to as the quality
ratio. Quality shoot number and quality ratio were both calculated only from those calli which

regenerated shoots.

Figure 1. Flow chart depicting maintenance and challenge subculture regimes utilized in
the four experiments. The value adjacent each arrow represents the total age
of the callus at the time of subculture. Exp 1 had no second challenge due to
asynchrony of treatments. and callus age values in Exp 4 were slightly less due
to 20 d subculture intervals.

77

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

INITIATION MAINTENANCE
“"“3 1st subculture
from 3 Wk
Ieal disks
3 wk 3 wk
6 Wk CHALLENGE
2nd subculture! 1st challenge
9 wk
3 wk 3 wk
[9 wk
3rd subculture 2nd challenge
12 wk
3 wk 3 wk
12 wk
4th subculture 3rd challenge
15 wk

 

 

 

 

3wk 3wk

 

 

 

78

Statistical Analysis System (SAS Institute, 1985) was used for most analyses. Data
were generally unbalanced and were handled by the General Linear Models (GLM) procedure.
Genotypes and treatments were treated as ﬁxed effects, while plants, leaves and samples were
treated as random effects. The experimental unit for callus initiation and shoot regeneration
frequencies was the leaf and response frequencies were calculated as the mean of the explants
within a leaf. All frequency variables displayed variance heterogeneity and were transformed
using the arcsine function (Steel and Torrie, 1980). For time to callus. lag time and shoot
number, the experimental unit was the Petri dish and explants within a leaf were an additional
source of variation. After the calli were subcultured, the experimental unit was the Petri dish
for shoot regeneration frequency and the individual callus for shoot number. Shoot
regeneration frequency was calculated as the mean of the three calli within the dish, i.e. 0,
33.3. 66.7 or 100%. Approximate F tests requiring synthesized mean squares were used when
necessary (Satterthwaite, 1946).

When data were balanced, mean separation was accomplished using a standard LSD
test. Data on shoot regeneration frequency and shoot number per callus were frequently
balanced, but data for quality shoot number and quality ratio were by definition unbalanced.
For unbalanced data, mean separation was accomplished using least squares marginal means
computed by the LSMEANS option of GLM. Least squares means are the values expected
had the design been balanced. Mean separation was accomplist using the PDIFF option of
LSMEANS which performs all possible t-tests in a manner analogous to using an LSD.
REQLTS

The three genotypes studied were chosen on the basis of their in vitro proﬁles. Table
3 summarizes the mean leaf disk responses in Exp 1, 2 and 3. Since only those leaf disks
whose callus was subcultured are included, callus frequency was 100% in all cases and the
values for time to callus were biased downward. Leaf disk explants of all three genotypes
initiated callus and regenerated shoots at high frequency, but the genotypes differed in time to
callus, lag time, shoot number, quality shoot number and quality ratio. Explants of EL 45/2-

79

 

  

h: wﬁm Rune «Av n.» mam Q8 9: 87va Am
3.9 who and new v.3 Yam «.8 8— ou¢§ Un—
3.Nm 9.." 3d odm a; — hem wdo 8— 74mm

 

 

 

ea .58.

. 25800
89:32 .85

 

 

 

.83qu m
93 N .~ mum 8:25 9 v8: “23.3 82.. «8593 Soﬁa?» 3:20 82“. ado—.8 8:88.. 95> 5 «o 82.; :88 .328.” .m 05.?

80

108 were slow to callus and regenerated a large number of shoots alter a short lag period, but
the shoots were mostly of poor quality resulting in a very low quality ratio. Although lag time
was longer, rapid callus initiation by REL-1 led to shoot regeneration 3 d earlier than EL 45/2-
108. Slow callus initiation coupled with a long lag period by explants of FC 607-020 led to
shoot regeneration 8 d later than REL-1. Calli of REL-l and FC 607-0-20 produced similar
numbers of shoots, but shoots of REL-1 were generally of better quality as indicated by the
quality ratio.

The genotypes also differed in their callus morphology on the initiation medium.
REL-1 callus tended to be bright white to yellowish and friable, but sometimes had green
sectors which were more moist. The callus of FC 607-020 was off-white to tan with dark
specks, friable and rarely green. EL 45/2-108 callus was tan to brown, crumbly and nodular
and very difficult to extract from the large mass of shoots typically present.
Effﬂ of Genotm

All four experiments included genotypes as a variable. Genotypes differed
signiﬁcantly for frequency of shoot regeneration and number of shoots per callus at each
subculture in the four experiments, except the ﬁnal two subcultures in Exp 4. Figures 2a and
2b depict the decline in shoot regeneration frequency and shoot number per callus,
respectively, of the three genotypes when subcultured every 3 wk on Bl in Exp 1. Similar
data from Exp 2 are displayed in Figures 3a and 3b. Maintenance on Bl was the control in
each experiment and the relative performance of the genotypes was always similar. After one
subculture (6 wk), callus of all three genotypes still regenerated shoots at high frequency. By
9 wk, regeneration frequencies of REL-1 and FC 607-0-20 had sharply declined, but all calli
of EL 45/2-108 continued to produce shoots. Shoot regeneration by calli of BL 45/2-108
remained above 50% after 15 wk on Bl, while regeneration by REL-1 and FC 607-0-20 was
near zero. The long term regeneration proﬁles of REL-1 and FC 607-0-20 were quite similar,
but REL-1 typically had higher levels of shoot regeneration and shoot number per callus.

The high shoot number of EL 45/2-108 was the reason that this genotype was dropped

Figure 2. (a) Shoot regeneration frequency and (b) shoot number per callus over time for
three genotypes in Exp 1. Vertical bars represent 1: SE when larger than
symbol. .

81
A

100

Sheet Regeneratlon (%)

 

 

 

 

Callus Age (wk)
"9" REL-1 + F0 607-0-20

'9' EL 4512-108

5 Sheet Number per Callue

 

 

 

 

 

Owl’

I

9 12 15
Callue Age (wk)
"9" REL-1 "9" FC 807-0-20

 

 

‘9’ EL 4512-108

Figure 3. (a) Shoot regeneration frequency and (b) shoot number per callus over time for
three genotypes in Exp 2. Vertical bars represent :1: SE when larger than
symbol. .

82

IX

Shoot Regeneratlon ('5)
100 ‘ a

 

 

 

 

 

 

 

 

. . . .2 “i

Callus Age (wk)

 

'9' REL-1 + F0 507-0-20 ‘9‘ EL 4512-108

E3

Sheet Number per Callue

 

 

 

 

 

 

 

ao~—

20--

10 + i

0" : —9~ ee
3 e e 12 15

Callue Age (wk)

 

'9' REL-1 "'3'" FC 807-0-20 '9’ EL 4512-108

 

83
from this research midway through Exp 3. When EL 4512-108 was subcultured on media
containing BA, the intense regeneration observed on initiation dishes continued for several
subcultures. resulting in up to 200 shoots on many dishes. The amount of time required to
counttheshootsonthethreecallioneachdishandtosubculturethesmallamountofcallus
proved to be prohibitive. Although shoots were removed at each subculture, subcultured callus
of EL 4512-108 may have contained undetected shoot primordia.
lt Interval

Routine callus maintenance was performed at 3 wk intervals, but at the end of the 3
wk period some calli were darker and approaching senescence. We thought that more
frequently subcultured callus might be less senescent, and therefore more likely to be capable
of shoot regeneration. Exp 1 investigated the effects of subculturing callus on B1 at 1, 2 or 3
wk intervals. The three treatments could only be compared at 9 and 15 wk since these were
the only times the calli were all the same age since initiation. Regeneration frequency and
shoot number data from 9 and 15 wk calli challenged to B1 and B3, respectively, for 3 wk are
presented in Table 4. in both cases, genotypes (G) differed signiﬁcantly for both traits, but
subculture interval (SC) and G x SC interaction had no effect. Calli subcultured more
frequently were lighter in color and appeared less senescent. When calli were challenged on
B1 at 9 wk, the challenge medium was the same as the maintenance medium. In this case the
longer SC resulted in more frequent regeneration and more shoots per callus. The calli
subcultured more frequently had been exposed to more BA and may have developed a greater
tolerance to BA. B3 medium was used for the challenge at 15 wk to see if the shorter SC
treatments had a regeneration potential not realized with the B1 challenge. On the B3
challenge, the 1 wk SC gave the highest values for both traits. Although these were not
significant differences, it suggested that shorter SC might have a beneﬁcial effect on
subsequent shoot regeneration EL 4512-108 was most notable in this regard; shoot number
per callus was 5.0, 1.6 and 0.9 from the calli subcultured at 1, 2 and 3 wk intervals,

respectively.

34

Table 4. Mean values of shoot regeneration frequency and shoot number per callus of
three genotypes when challenged at 9 and 15 wk after maintenance with
differing subculture interval in Exp 1.

 

 

B1 Challenge at 9 wk B3 Challenge at 15 wk
Shoot Shoot Shoot Shoot
no S wk R eneration % Number Re eneration % N ber
REL-1 1 0 0 13.9 0.14
2 0 0 5.6 0.08
3 5.6 0.08 0 0
FC 607-0-20 . l 8.3 0.11 2.8 0.03
2 2.8 0.03 0 0
3 25.0 0.33 0 0
EL 4512-108 1 80.0 2.80 60.0 4.97
2 88.9 3.36 55.6 1.64
3 91.7 4.72 47.2 0.92
Means by SC 1 26.5 0.86 23.5 1.52
2 30.6 1.13 20.4 0.57
3 40.7 1.71 16.2 0.31

 

SC was also evaluated in Exp 4. In this case, SC of 10 and 20 d were used so that the
calli could be compared every 20 d. No signiﬁcant effects involving SC were found in the
three challenges in Exp 4, but, as in Exp 1, there were some instances where shorter SC
appeared beneﬁcial. When ﬁrst challenged to B3Tl at 81 d, the 10 (1 SC calli of REL-1
maintained without BA had a shoot regeneration rate more than double that of the 20 d SC
calli. When challenged again to B3T1 at 101 d, the effect was not seen. Maintenance of calli
with SC less than 3 wk, while clearly not of signiﬁcant beneﬁt, does not appear to be
detrimental.

BA goncentration

Chronic exposure to BA by maintenance of calli on B1 could possibly intensify
development of tolerance to BA. The effect of BA concentration was examined in Exp 2.
Genotypes, media and their interaction had signiﬁcant effects on regeneration frequency and

shoot number at each subculture except at 6 wk. Doubling the BA concentration at each

85
subculture maintained the regeneration ﬁequency (Figure 4a) and shoot number (Figure 4b) at
levels signiﬁcantly greater than maintenance on B1 or M80. As expected, maintenance on
M80 resulted in a more rapid decline in both variables. The signiﬁcant genotype x medium
interactions were mainly due to EL 4512-108 not regenerating more on the BA doubling
regime vs. on Bl. Since EL 45/2-108 maintained high-frequency regeneration on Bl (Figure
3a), any boost due to the BA doubling regime was more restricted than for the other two
genotypes.

When calli maintained on M80 or B1 were challenged to B3 at 9, 12 and 15 wk,
signiﬁcant increases in regeneration and shoot number were observed. In all cases the boost
was greater for dishes maintained on M80 vs. those on B1. The response of calli challenged
on B3 at 9 and 12 wk was compared to calli of the same age on maintenance media. These
comparisons for 12- and lS-wk-old calli are presented in Tables 5 and 6, respectively. Calli
maintained on M80 had regeneration rates near zero, but transfer to B3 produced regeneration
rates similar to calli maintained on B1. Shoot number of M80 calli was also greatly enhanced
by B3, but it was still less than half the shoot number of B1 calli. The stimulation in
regeneration from the B3 challenges did not bring these two treatments up to the regeneration
level observed on dishes in the BA doubling regime (Figures 4a and 4b).

The effects of BA concentration were also examined in Exp 4, but a doubling regime
was not included. As in Exp 1, regeneration by calli maintained on MSO declined more
rapidly than calli maintained on B1, but regeneration rates of M80 calli approached those of
B1 calli when challenged to B3T1. Challenge data from both experiments demonstrated that
the lack of regeneration by calli maintained on M80 or B1 was not indicative of loss of
competence.

Effects of TIBA

If continuous exposure to a constant level of BA results in a decreased sensitivity to

BA, we thought it might also be possible to counteract this with the anti-auxin TIBA. Calli

were maintained on B1 with three levels of TIBA in Exp 3. When calli of two genotypes

Figure 4. (a) Shoot regeneration frequency and (b) shoot number per callus over time for
three BA regimes in Exp 2. Vertical bars represent :1: SE when larger than
symbol. .

86

 

 

 

 

 

 

 

 

A
Shoot Regeneration (%)
100
80 “P
50 “F
4o-~
20 '-
0 i i i ‘
3 8 9 12 15
Callus Age (wk)
'9' M80 ‘5'- 81 ‘9" doubling BA
B

Sheet Number per Callus

 

 

 

 

O
.0—
—1— ('

3 8 9 12 15
Callus Age (wk)

 

'9‘ M80 "5" B1 ‘5‘ doubling BA

87

Table 5. Mean values of shoot regeneration frequency and shoot number per callus of
12-wk-old calli of three genotypes maintained with and without BA in Exp 2.
At 9 wk, each callus was subculnued to both maintenance medium and
challenge medium (B3).

 

 

 

 

Maintenance Sl_i_o_ot Regenerationggg) §Mt Ngmggr

m Medim Maintenm Challenge Maintenance Challenge
REL-l M80 5.6 *" 33.3 0.06 . NS 0.58

Bl 25.0 NS 24.2 0.28 NS 0.30
FC 607-0-20 MSO 2.8 NS 16.7 0.03 NS 0.22

Bl 5.6 NS 22.2 0.06 NS 0.36
EL 45/‘2-108 M50 16.7 ** 88.9 0.19 " 7.67

Bl 94.4 NS 91.7 17.03 NS 16.58
Means by MSO 8.3 ** 46.3 0.09 ** 2.82
Medium Bl 41.7 NS 46.7 5.79 NS 5.90
Grand Mean 25.0 * 46.5 2.94 " 4.34
N83,“ Difference between maintenance and challenge means not signiﬁcant or

signiﬁcant at P S 0.05 or P S 0.01. respectively.

 

88

Table 6. Mean values of shoot regeneration frequency and shoot number per callus of
15-wk-old calli of three genotypes maintained with and without BA in Exp 2.
At 12 wk. each callus was subcultured to both maimenance medium and

 

 

 

challenge medium (B3).
Maintenance Shoot Regenerationg‘lb) Shoot Number
m Medium Maintenance Challggg Mainm gigging;
m1 M80 0 ** 38.9 0 * 0.81
Bl 0 NS 13.9 0 NS 0.14
FC 607-0-20 M80 0 NS 2.8 0 NS 0.03
_ Bl 2.8 NS 5.6 0.03 NS 0.06
EL 45/2-108 M80 3.0 ** 61.1 0.03 "' 1.64
B1 80.6 NS 88.9 9.83 ** 8.72
Means by M80 1.0 ** 34.3 0.01 ** 0.82
Medium Bl 27.8 NS 36.1 3.29 NS 2.97
Grand Mean 15.0 ** 35.2 1.67 NS 1.90
NS,*,** Difference between maintenance and challenge means not signiﬁcant or

signiﬁcant at P S 0.05 or P S 0.01. respectively.

 

89
(REL-1 and FC 607-0-20) were maintained on BlTl, regeneration at each subculture was
signiﬁcantly greater than from calli maintained on Bl (Figures 5a and 5b). Calli maintained
on BlT.1 were generally intermediate in response. but did not regenerate signiﬁcantly more
than those on B1. Signiﬁcant genotype x medium interactions were detected for shoot
regeneration at 12 wk, for shoot number at 12 and 15 wk, and for both variables alter the ﬁrst
B3 challenge (12 wk). In all cases, these were due to a greater response to TIBA by REL-1
vs PC 607-0-20.

As in Exp 2, calli challenged at 9 and 12 wk could be compared to calli of the same
age on maintenance media. With the exception of 12-wk-old calli of FC 607-0-20 on B1T.1.
calli transfened to B3 had increased levels of shoot regeneration (Tables 7 and 8). In contrast.
shoot number per callus was not enhanced by the B3 challenges. and calli maintained on BlTl
frequently had reduced shoot numbers after transfer to B3. The possibility that the removal of
TIBA was responsible for the shoot number reduction in the challenged calli led to the
inclusion of 1.0 mg/L TIBA in the medium used for the third challenge at 15 wk. Statistical
comparison of the response of 15-wk-old calli before and after challenge to B3T1 was
confounded by callus age. but maintenance of the 15 wk response levels in the 18-wk-old
challenged calli could be viewed as a reversal or delay of the expected decline in response
(Table 9). Increased regeneration of the calli maintained on B1 and B1T.1, but not of those
maintained on BlTl, suggests that calli respond to increases in TIBA as well as increases in
BA in the challenge medium.

In Exp 4, TIBA at levels of 0 and 1.0 mg/L were evaluated in combination with BA at
the same concentrations. Calli maintained on BlTl again had signiﬁcantly higher regeneration
rates than calli maintained on B1 (Figure 6a), and shoot number was enhanced for three
subcultures (Figure 6b). The regeneration promoting effects of TIBA were only realized in the
presence of BA, with calli on TIBA alone (Tl) performing similar to those on hormone-free
medium (M80).

Neither of the B3T1 challenges signiﬁcantly increased regeneration rates or shoot

Figure 5. (a) Shoot regeneration frequency and (b) shoot number per callus over time for
three levels of TIBA in Exp 3. Vertical bars represent :t SE when larger than
symbol. .

A

Shoot Regeneratlon (In)
100 . _

 

80‘

80'"

40"

 

20‘

 

 

 

 

Callus Age (wk)

 

'9‘ 81 + B1T.1 '9' B1T1

B

Shoot Number per Callus

 

 

 

\,‘

 

 

0
-rr-
qr— I
:i'
y

I.

3 6 9 12 15
Callus Age (wk)

 

"9" Bl "9" B1T.1 '9’ B1T1

91

Table 7. Mean values of shoot regeneration frequency and shoot number per callus of
12-wk-old calli of two genotypes maintained on B1 with and without TIBA in
Exp 3. At 9 wk. each callus was subcultured to both maintenance medium and
challenge medium (B3).

 

 

 

Maintenw Shoot Regenerationgqb) 8m Number

Medium Mainten Challen Maintenance Chall n
REL-1 Bl 11.1 NS 19.4 0.14 NS 0.25
B1T.l 19.4 NS 27.8 0.25 * 0.61
BlTl 72.2 NS 77.8 3.39 “ 2.50
FC 607-0-20 B1 0 NS 25.9 0 NS 0.30
B1T.l 19.4 NS 16.7 0.50 NS 0.25
BlTl 16.7 NS 24.2 0.58 NS 0.51
Means by B1 5.6 NS 22.2 0.07 NS 0.27
Medium B1T.l 19.4 NS 22.2 0.38 NS 0.43
B1T1 44.4 NS 52.2 1.99 ** 1.55
Grand Mean 23.2 “ 32.4 0.81 NS 0.76

N833” Difference between maintenance and challenge means not signiﬁcant or

signiﬁcant at P S 0.05 or P S 0.01, respectively.

 

92

Table 8. Mean values of shoot regeneration frequency and shoot number per callus of
lS-wk-old calli of two genotypes maintained with and without TIBA in Exp 3.
At 12 wk. each callus was subcultured to both maintenance medium and
challenge medium (B3).

 

 

 

 

Maintenance Shoot Regenerationgqo) Shoot Number
anotvne M ' Maintenance en Maintenan n
REL-1 Bl 11.1 NS 18.5 0.19 NS 0.22
B1T.l 11.1 NS 26.7 0.19 NS 0.57
BlTl 47.2 NS 55.5 1.33 NS 1.11
FC 607-0-20 B1 2.8 NS 5.6 0.03 NS 0.06
B1T.l 5.6 NS 13.9 0.08 NS 0.25
BlTl 8.3 NS 16.7 0.11 NS 0.42
Means by B1 6.9 NS 11.1 0.11 NS 0.13
Medium BlT.1 8.3 NS 19.7 0.14 NS 0.39
BlTl 27.8 NS 33.3 0.72 NS 0.71
Grand Mean 14.4 NS 21.4 0.32 NS 0.41

 

NS indicates that maintenance and challenge means are not signiﬁcantly different at P S 0.05.

 

93

Table 9. Mean values of shoot regeneration frequency and shoot number per callus of
two genotypes maintained on B1 with and without TIBA in Exp 3. The
response of calli after 15 wk on maintenance media was compared with the
same calli after 3 wk on challenge medium B3T1 (18-wk old). At 15 wk. all
calli were subcultured from their respective maintenance media to the

 

 

 

 

challenge medium.
Maintenance Shoot Regeneration(%) SMt Numgr

@notype Medium Maintenance Challenge Maintenance Challenge
REL-1 Bl 11.1 NS 22.2 0.19 NS 0.28

BlT.1 11.1 * 33.3 0.19 NS 0.64

BlTl 47.2 * 27.8 1.33 1" 0.64
FC 607-0—20 B1 2.8 NS 0 0.03 NS 0

B1T.l 5.6 NS 8.3 0.08 NS 0.08

B1T1 8.3 NS 5.6 0.11 NS 0.06
Means by B1 6.9 NS 1 1.1 0.1 1 NS 0.06
Medium B1T.l 8.3 * 20.8 0.14 NS 0.36

BlTl 27.8 NS 16.7 0.72 * 0.35
Grand Mean 14.4 NS 16.2 0.32 NS 0.28
N83." Difference between maintenance and challenge means not signiﬁcant or

signiﬁcant at P S 0.05 or P S 0.01, respectively.

 

Figure 6. (a) Shoot regeneration frequency and (b) shoot number per callus over time for
four BA x TIBA treatments in Exp 4. Vertical bars represent t SE when
larger than symbol.

A

0 Shoot Regeneration (%)

 

 

 
   

 

 

 

 

 

 
 
   

80 ~ ‘.
BO --
‘0 d- If
t
.L
20 “P \ .
o i i 7“ F
3 6.9 8.7 11.6 14.4
Callus Age (wk)
'9" M80 "5' T1 '9' 81 + 8111
B
14 Shoot Number per Callus
12 -

10‘

 

6‘

4.4

2q

 

 

 

 

3 5.9 8.7 11.6 14.4
Callus Age (wk)

 

+MSO +T1 '9‘31 +8111

95
numbers over the maintenance levels (Tables 10 and 11). These tables include only data from
the long SC calli since valid maintenance regeneration data could not be obtained from the
short 8C calli. This reduction in sample size may partially explain why signiﬁcant differences
were not detected even though some large increases in regeneration frequency were observed.
The poor response to B3Tl by calli maintained on BlTl provides further evidence that calli
respond to increases in TIBA. The ﬁnal challenge in Exp 4 utilized B3T2, but regeneration
levels were too low to draw any conclusions.
DISCUSSION

The model proposed for shoot regeneration from hormone-autonomous sugarbeet callus
involves complementation of endogenous auxin physiology and exogenously supplied
cytokinin (Chapter 1). We hypothesize that as the calli are subcultured on Bl there is a
gradual decrease in sensitivity to exogenous cytokinin such that the effective auxin/cytokinin
ratio exceeds the range in which shoot regeneration is possible. That is, the cells are still
competent, but the stimulus required for determination is no longer perceived. Bl medium is
clearly an effective stimulator of regeneration in primary callus. but its effectiveness declines
with repeated subculture. The decreased sensitivity, or tolerance, to cytokinin could be due to
an altered auxin physiology in the habituated callus. Kinetin was capable of eliminating the
auxin requirement of tobacco callus by maintaining endogenous auxin levels (Syono and
Funrya, 1972). BA might have a similar effect on hormone-autonomous sugarbeet callus.
This shift in the effective auxin/cytokinin ratio can apparently be avoided by either steadily
increasing the exogenous cytokinin level or by maintenance on medium containing the anti-
auxin TIBA.

Information on endogenous hormone levels in habituated soybean callus (Wyndaele et
al.. 1988) provides some precedent for this hypothesis. Two fully-habituated soybean cell
lines were found to contain higher levels of endogenous 1AA than an auxin-habituated cell
line, which in turn had higher levels of IAA than a non-habituated cell line. These authors

also found that exogenously applied kinetin increased endogenous 1AA in a fully-habituated

96

Table 10. Mean values of shoot regeneration frequency and shoot number per callus of
11.6-wk-old calli of two genotypes maintained with and without BA and/or
TIBA in Exp 4. At 8.7 wk. each callus was subcultured to both maintenance
medium and challenge medium (B3Tl). Data presented is for long SC plates

 

 

 

only.
Maintenance Shoot Regeneration(%) Shoot Numgr
Genotype Medium Maintenance Challengg Maintenance Qallengg

REL-1 M80 7.4 NS 1 1.1 0.07 NS 0.1 1
T1 0 NS 7.4 0 NS 0.07
B1 18.5 NS 40.7 0.70 NS 0.70
BlTl 63.0 NS 63.0 1.00 NS 1.41

FC 607-0-20 M80 0 NS 0 0 NS 0
T1 3.7 NS 7.4 0.04 NS 0.1 1
B1 11.1 NS 14.8 0.15 NS 0.22
BlTl 29.6 NS 33.3 0.44 NS 0.52
Means by M80 3.7 NS 5.6 0.04 NS 0.06
Medium T1 1.9 NS 7.4 0.02 NS 0.09
B1 14.8 NS 27.8 0.43 NS 0.46
B 1T1 46.3 NS 48.2 0.72 NS 0.96
Grand Mean 16.7 NS 22.2 0.30 NS 0.39

 

NS indicates that maintenance and challenge means are not signiﬁcantly different at P S 0.05.

 

97

Table 11. Mean values of shoot regeneration frequency and shoot number per callus of
14.4-wk-old calli of three genotypes maintained with and without BA and/or
TIBA in Exp 4. At 11.6 wk, each callus was subcultured to both maintenance
medium and challenge medium (BBTl). Data presented is for long SC plates

 

 

 

only.

Maintena_n_g 8m Regeneration(%) Shoot Number
ﬁnotype Medium Maintenance Challenge Maintenance Challgngg
REL-1 M80 0 NS 11.1 0 NS 0.11

T1 0 * 25.9 0 *‘l‘ 0.70
B1 3.7 NS 11.1 0.07 NS 0.11
BlTl 22.2 NS 48.2 0.30 NS 0.59
FC 607-0-20 M80 0 NS 0 0 NS 0
T1 0 NS 7.4 0 NS 0.07
B1 3.7 NS 18.5 0.04 NS 0.30
BlTl 22.2 NS 7.4 0.33 NS 0.07
Means by M80 0 NS 5.6 0 NS 0.06
Medium T1 0 NS 16.7 0 NS 0.39
Bl 3.7 NS 14.8 0.06 NS 0.20
BlTl 22.2 NS 27.8 0.31 NS 0.33
Grand Mean 6.5 NS 16.2 0.09 NS 0.25
N83,“ Difference between maintenance and challenge means not signiﬁcant or

signiﬁcant at P S 0.05 or P S 0.01, respectively.

 

98
cell line derived from a crown gall tumor induced by wild-type Agrobacrerium nunefaciens. If
the habituation response in soybeans and sugarbeet is similar, hormone-autonomous sugarbeet
callus may contain elevated levels of IAA, particularly when maintained on BA. Stimulation
of regeneration with TIBA supports this possibility.

Endogenous IAA levels in nonorganogenic habituated sugarbeet callus have been
compared to those of organogenic habituated and nonorganogenic auxin-dependent calli.
Organogenic habituated callus derived from the original self-regenerating line of DeGreef and
Jacobs (1979) had less than one-sixth the endogenous IAA as the auxin-dependent callus
(Kevers et al.. 1981b), while the two nonorganogenic cell lines contained similar levels of IAA
(Kevers et al.. 1981a). In these studies, each callus type was represented by a single long term
cell line. and all three were derived from the same original cell line. Considering the
magnitude of genetic variation for in vitro behavior in sugarbeet (Doley and Sarmders, 1989;
Saunders and Shin, 1986; Chapter 1), substantiation of the physiological differences between
auxin-dependent and hormone-autonomous calli should include primary calli of a number of
W-

Genetic differences in long term regeneration have been reported in maize (Kama and
Hodges, 1986), pea (Malmberg. 1979) and tomato (Locy, 1983). In this research. sugarbeet
genotypes differed in their long term regeneration proﬁles on maintenance media as well as in
their response to the various challenge media. Genetic variation for shoot regeneration
longevity highlights the need for careful choice of genotype for procedures involving honnone-
autonomous sugarbeet callus.

The observed genotypic differences in long term regeneration could also be explained
by the same tolerance hypothesis described above. The short lag time and proliﬁc
regeneration characteristic of explants of EL 45/2-108 is thought to be due to low levels of
endogenous auxin and/or sensitivity to BA (Doley and Saunders, 1989). The variation in

sensitivity observed in explants could also carry through to callus derived from these explants.

99
If EL 45/2-108 is more sensitive to cytokinin. it might take longer to develop the tolerance
thought to result in the rapid decline in regeneration in other genotypes. In contrast, genotypes
like REL-1 may contain higher levels of endogenous auxin. but not too high to interact
favorably with the exogenous cytokinin to regenerate shoots from primary callus. Tolerance to
BA may develop rapidly in these genotypes due to rapid increases in endogenous IAA, and
genetic variation for long term regeneration may reﬂect genetic variation in IAA physiology in
habituated calli.

The long term regeneration response of BL 45/2-108 was excellent. but it cannot be
recommended for general use in procedures requiring callus subculture. The extremely high
shoot to callus ratio made working with callus of this genotype very difﬁcult and time
consuming. Additionally. since the majority of the regenerant shoots were vitreous, the
number of quality shoots was not much greater than other genotypes. Maintenance of BL
45/2-108 calli on hormone-free medium suppressed shoot regeneration to a more manageable
level, representing one Option for in vitro manipulation of this genotype. EL 45/2 germplasm
might be useful in efforts to obtain regeneration from protoplasts.

Although shoot regeneration of REL-1 and FC 607-0.20 declined sharply after 6 wk
in culture, shoot regeneration of both genotypes was enhanced by challenge with B3 medium.
Both genotypes had enhanced rates of shoot regeneration and shoot number when maintained
on BlTl, but REL-1 was generally more responsive to maintenance on TIBA. Results from
Exp 4 demonstrated that the effect of TIBA on shoot regeneration was only realized in the
presence of BA. Thus, TIBA acts synergistically with BA to produce an effect much greater
than that of either one alone. Hormone-autonomous sugarbeet callus is capable of sustained
growth on hormone-free medium, but shoot regeneration is generally dependent on cytokinin.
If TIBA is acting as an anti-auxin in our system, its failure to stimulate regeneration without
BA ﬁts the proposed auxin/cytokinin regeneration model. TIBA has been shown to prevent a

buildup in endogenous IAA in tobacco petiole explants (Cassells et al., 1982). Similarly, as

100
sugarbeet calli age, TIBA may reduce a buildup in endogenous auxin. resulting in an auxin
level that can interact favorably with B1 medium.

Calli were also maintained in a non-regenerating state of competence on hormone-free
medium (N180), and then induced to regenerate by transfer to media containing BA. In this
case, removal of the exogenous cytokinin supply increased the effective auxin/cytokinin ratio
to a level that was more conducive to root formation than to shoot production. Although the
calli were hormone-autonomous. their morphogenetic response was manipulated with
exogenous hormone regimes. When calli maintained on M80 in Exp 2 were challenged to
regenerate on B3, highly signiﬁcant increases in shoot regeneration and shoot number per
callus were realized (Tables 5 and 6). In contrast, regeneration by calli maintained on Bl was
enhanced, but not signiﬁcantly, by the B3 challenges. This suggests that the maintenance
medium may inﬂuence the physiological state of the calli, and thus inﬂuence the response to a
given challenge medium. In this research, I have used the term ’challenge medium’ in
preference to ’regeneration medium’ since it is clear that the medium required to Optimize
regeneration in subcultured calli depends on the genotype as well as the maintenance medium.
Whereas BB was capable of greatly enhancing regeneration in calli maintained on M80, higher

levels of BA may be needed to optimize regeneration in calli maintained on B1.

GENERAL DISCUSSION

The proposed model for shoot regeneration from hormone-autonomous sugarbeet callus
involves complementation of endogenous auxin physiology with exogenously supplied
cytokinin. Results from all three chapters supported this hypothesis which is compatible with
the auxin/cytokinin ratio model of Skoog and Miller (1957). The speciﬁc underlying
mechanisms, which were outside the scope of this research. remain unclear. In Chapter 1. a
great range of variation was encountered when leaf disks of 78 genotypes were incubated on
B1 medium. The leaf disks of the various genotypes tested were presumed to differ in either
levels of, or sensitivity to, endogenous auxin and/or sensitivity to the exogenously supplied
cytokinin Differential response of genotypes to a range of BA concentrations in Chapter 2
provided evidence that the leaf disk response could be manipulated by varying the dose of
exogenous cytokinin. Similarly, manipulation of shoot regeneration with BA and TIBA in
Chapter 3 was explained by temporal as well as genetic variation in the endogenous auxin
physiology and/or sensitivity to exogenous cytokinin of the subcultured callus.

An implicit assumption of the proposed model is that cell growth and division cannot
occur in the complete absence of auxin and cytokinin. That is. both types of plant growth
regulators must be present in at least minute quantities to stimulate callus initiation. Can plant
cells divide in an environment devoid of auxin? I presume not. and therefore must assume
that the hormone-autonomous sugarbeet callus initiated on B1 medium contains IAA, and must
have the capacity to synthesize IAA. Meins (1982) has shown that habituated cell lines
produce the substances for which they are habituated, but no data are available for hormone-
autonomous sugarbeet callus.

Genetic variation for in vitro chronology was a curious aspect of this shoot

101

102
regeneration system. Genotypes varied not only for the frequency of callus initiation. but also
forthetimerequiredtoinitiatethecallus. Ihavehypothesizedthatthisisthetimerequired
for one or more cells in the explant to attain auxin-autonomy, however, there is no direct
evidence in support of this. If the hypothesis is correct. there is genetic variation in sugarbeet
for the probability that a cell will become auxin habituated.

'lhe BA gradient experiment in Chapter 2 demonstrated that levels of BA as low as
0.1 mg/L could signiﬁcantly increase the frequency of callus initiation as well as reduce the
time to callus. Kinetin has been shown to maintain endogenous IAA levels in tobacco callus.
thereby replacing the auxin requirement of the callus (Syono and Furuya, 1972). Perhaps the
BA in our medium stimulates IAA synthesis in the explant, increasing the probability and
speed of callus initiation. Also working with sugarbeet, Van Geyt and Jacobs (1985) reported
that cytokinin habituation occurs spontaneously, while auxin habituation requires induction.
We know that initiation of hormone autonomous sugarbeet callus does not require exogenous
growth regulators (Doley and Saunders. 1989), and BA appears to stimulate induction of auxin
autonomy (Chapter 2). The BA-induced auxin-autonomous callus may spontaneously become
cytokinin habituated. In soybean, Wyndaele et al. (1988) found that callus habituated for both
auxin and cytokinin could be derived from an auxin-habituated cell line, but not ﬂour a
cytokinin-habituated cell line.

Lag time. the other parameter of in vitro chronology, also was affected by genetic
variation. This may be a reflection of variation for sensitivity to the exogenous cytokinin.
The tendency for proliﬁc regenerators, such as plants of EL 45/2, to have short lag times
supported this possibility. Since the procedure is one-step, I cannot separate the physiology of
the callus from that of the explant. The response of subcultured calli in Chapter 3 provided
evidence that the physiology of the primary callus is at least partially maintained in the
absence of the explant.

The results of Chapter 1 suggested that the ability to initiate callus and the ability to

regenerate shoots from that callus may be independently inherited. If callus initiation is

103
dependent on auxin physiology and shoot regeneration is dependent on cytokinin physiology,
there may be a hormonal basis for the postulated independence. Independence of these two in
vitroparametershasbeenreportedinotherspecies,andcouldbeaunifyingaspectofplant
tissue culture.
RECOMMENDATIONS
M

Even with an array of different media, one-step regeneration may not be achievable
across the entire range of germplasm. Thus for some genotypes, a two-step procedure may be
the only alternative. Subculture of fresh callus of recalciuant genotypes to B3 or B3Tl may
be helpful. Subculture of fresh callus to liquid M80 or B1. followed by plating out on B1T1
or B3T1 might be a method of maximizing the number of regenerants obtainable in a limited
period of time. Possibly, immersing the cells in liquid medium results in an increased
probability of becoming determined for shoot regeneration.

Further testing of some of the hypotheses advanced in this research could include
measuring endogenous levels of growth regulators in leaf disks of genotypes diverse for their
in vitro behavior. These measurements should be done before culture, periodically after the
explants have been plated, and after callus has been initiated. Methodology for detection and
quantiﬁcation of endogenous IAA have been published by Cassells et aL (1982) and Syono
and Furuya (1972).

With regard to vitreousness, some possible remedies (GA,, phloridzin, lower BA
concentrations) were tested in the leaf disk system. The results were not encouraging, except
possibly the use of B5 salts in place of MS. The many hypotheses as to the causes of
vitreousness deal with shoot cultures and plantlets. Perhaps callus initiation and shoot
regeneration are so far removed from normal growth and development that the remedies for
vitreous shoot cultures are not applicable. It has been suggested that hormone-autonomous
sugarbeet callus is vitreous by nature (Crevecoeur et al., 1987), but there is certainly genetic

variation for the quality of the regenerant shoots.

104
Mgtm x Mgdium Interaction

Some suggestions for further research on G x M interaction in sugarbeet were
presented at the end of Chapter 2. These included use of ABA, GA,, TIBA, and ethylene
antagarists. Another possibility is 2,4-dinitrophenol which is mentioned by Atanassov
(1986b). He cites work of Slavova (unpublished) showing that this compound enhanced
regeneration in sugarbeet. The ultimate goal of the G x M interaction work is to develop
protocols, if possible, to obtain shoot regeneration from all members of the sugarbeet
germplasm pool. As mentioned above, attainment of this goal may involve utilization of a
two-step procedure for some germplasm. Although they appear somewhat inhibitory in the
callus initiation step (Doley, unpublished; Chapter 2), TIBA and GA3 may be quite useful in
two-step procedures.

Long-Tarn Reggneration

In my research, various maintenance schemes were employed to evaluate their effects
on continuous regeneration, while only a few different media were explored as challenges.
Maintenance of large numbers of calli on M80 or B1 followed by subculture to media with
various combinations of BA, TIBA, GA,. etc may lead to formulation of a ’best’ challenge
medium, which might be worthy of the title ’regeneration medium’. Maintenance on low
levels of BA (e.g. B.1) might have an advantage in prolonging responsiveness to BA.

Maintenance of calli on Bl with increasing levels of TIBA at each subculture might
provide further support for the tolerance hypothesis. Maintenance of calli on B 1T1 may only
postpone an inevitable decline in shoot regeneration. Calli on B1T1 had enhanced
regeneration for about two subcultures and then started to decline at a rate similar to the B1
calli. The BlTl calli thus exhibited the response of younger calli.

Another unexplored area of long term regeneration is the effects of light. Preliminary
studies suggested that transfer of calli from the dark to the light stimulated regeneration
(Doley, unpublished; Galatowitsch and Smith, 1990). Experiments to evaluate the effect of
light should involve transfer of dark-maintained calli to both dark and lighted environments.

105
The importance of long term regenerability is partly determined by the nature of the
protocol to be used. For procedures such as somatic cell selection with acute exposure,
sugarbeet genotypes with sufﬁcient longevity of regeneration are available (e.g. REL-l). but
cultural modiﬁcations might still be able to enhance the probability of obtaining a mutant
regenerant. On the other hand, the available longevity, except in the case of EL 45/2, may be
insufﬁcient for the regeneration of sugarbeet plants from protoplasts.

LIST 01“ REFERENCES

LIST 01“ REFERENCES

Atanassov AI (1986a) Sugar beet (Beta vulgaris L.). In: Bajaj YPS (ed) Biotechnology in
agriculture and forestry vol 2: Crops 1. Springer-Verlag, Berlin, pp 462-470

Atanassov AI (1986b) Sugar beet. In: Evans DA et al (eds) Handbook of plant cell culture. vol
4. MacMillan. New York pp 652-680

Agache S, Bachelier B, de Buyser J, Henry Y, Snape J (1989) Genetic analysis of anther
culture response in wheat using aneuploid, chromosome substitution and translocation
lines. Theor Appl Genet 7727-11

Aharoni N , Yang SF (1983) Auxin-induced ethylene production as related to auxin metabolism
in leaf discs of tobacco and sugar beet. Plant Physiol (Bethesda) 73:598-604

Ahloowalia BS (1983) Spectrum of variation in somaclones of triploid ryegrass. Crop Sci
23:1141-1147

Armstrong CL, Green CE (1985) Establishment and maintenance of friable, embryogenic
maize callus and the involvement of L—proline. Planta (Berl) 164:207-214

Armstrong CL, Phillips RL (1988) Genetic and cytogenetic variation in plants regenerated
from organogenic and friable. embryogenic tissue cultures of maize. Crop Sci 28:363-
369

Balzan R (1978) Karyotype instability in tissue cultures derived from the mesocotyl of Zea
mays seedlings. Caryologia 31:75-87

Bayliss MW (1975) The effects of growth in vitro on the chromosome complement of Daucus
carota (L.) suspension cultures. Chromosoma (Berl) 51:401-411

Beckert M, Qing CM (1984) Results of a diallel trial and a breeding experiment for in vitro
aptitude in maize. Theor Appl Genet 68:247-251

Behki RM, Lesley SM (1976) In vitro plant regeneration from leaf explants of Lycopersicon
esculenrum (tomato). Can J Bot 54:2409-2414

Bhat SR. Ford-Lloyd BV, Callow JA (1985) Isolation of protoplasts and regeneration of callus
from suspension cultures of cultivated beets. Plant Cell Rep 4:348-350

Bhat SR, Ford-Lloyd BV, Callow JA (1986) Isolation and culture of mesophyll prot0plasts of
garden, fodder and sugar beets using a nurse culture system: Callus formation and
organogenesis. J Plant Physiol 124:419-423

Bingham ET, Hurley LV, Kaatz DM, Saunders JW (1975) Breeding alfalfa which regenerates

106

107
from callus tissue in culture. Crop Sci 15:719-721

Bossoutrot D, I-Iosemans D (1985) Gynogenesis in Beta vulgaris L.: From in vitro culture of
unpollinated ovules to the production of doubled haploid plants in soil. Plant Cell Rep
4:300-303

Buiatti M. Baroncelli S, Bennici A, Pagliai M, Tesi R (1974) Genetics of growth and
differentiation in vitro of Brassr‘ca oleracea var. bonytls. II. An in vitro and in viva
analysis of a diallel cross. Z Pﬁanzenzuecht 72:269-274

Carmi A, Heuer B (1981) The role of roots in the control of bean shoot growth. Ann Bot
(Lond) 48:519-527

Cassells AC, Morrish FM (1987) Variation in adventitious regenerants of Begonia rex Putz.
’Lucille Closon’ as a consequence of cell ontogeny, callus ageing and frequency of
callus sub-culture. Sci Hortic (Amst) 32:135-143

Cassells AC, Long RD, Mousdale DMA (1982) Endogenous IAA and morphogenesis in
tobacco petiole cultures. Physiol Plant 56:507-512

Chandler SF, Dodds JH (1983) Adventitious shoot initiation in serially subcultured callus
cultures of Solanum laciniatum. Z Pﬂanzenphysiol 111:115-121

Chandler SF, Vasil IX (1984) Optimization of plant regeneration from long term embryogenic
callus cultures of Pemrr'setum purpurcrtm Schum. (Napier grass). J Plant Physiol
1 17: 147-156

Chi G-L. Pua EC (1989) Ethylene inhibitors enhanced de novo shoot regeneration from
cotyledons of Brassica canpestrr’s ssp. chinensis (Chinese cabbage) in vitro. Plant Sci
64:243-250

Coleman GD, Ernst SO (1989) In vitro shoot regeneration of Populus deltoides: effect of
cytokinin and genotype. Plant Cell Rep 8:459-462

Crevecoeur M, Kevers C, Greppin H, Gaspar T (1987) A comparative biochemical and
cytological characterization of normal and habituated sugarbeet calli. Biol Plant
(Prague) 29: 1-6

Deaton WR, Metz 86, Armstrong TA, Mascia PN (1987) Genetic analysis of the anther-
culture response of three spring wheat crosses. Theor Appl Genet 74:334-338

Detrez C, Sangwan RS, Sangwan-Noneel BS (1989) Phenotypic and karyotypic status of Beta
vulgaris plants regenerated from direct organogenesis in petiole culture. Theor Appl
Genet 77 :462-468

Detrez C, Tetu T, Sangwan RS, Sangwan-Noneel BS (1988) Direct organogenesis from petiole
and thin cell layer explants in sugar beet cultured in vitro. J Exp Bot 39:917-926

De Greef W, Jacobs M (1979) In vitro culture of the sugarbeet: Description of a cell line with
high regeneration capacity. Plant Sci Lett 17:55-61

Doctrinal M, Sangwan RS. Sangwan-Norieel BS (1989) In vitro gynogenesis in Beta vulgaris
L.: Effects of plant growth regulators, temperature, genotypes and season. Plant cell
Tissue Organ Cult 17:1-12

108

Doley WP, Saunders JW (1989) Hormone-free medium will support callus production and
subsequent shoot regeneration ﬁom whole plant leaf explants in some sugarbeet (Beta
vulgaris L.) populations. Plant Cell Rep 8:222-225

Dunwell JM (1981) Inﬂuence of genotype and environment on growth of barley embryos in
vitro. Ann Bot (Lond) 48:535-542

Errgelke AL, Harnzi HQ, Skoog F (1973) Cytokinin-gibberellin regulation of shoot
development and leaf form in tobacco plantlets. Am J Bot 60:491-495

Evans DA, Gamborg CL (1982) Chromosome stability of cell suspension cultures of Nicotiana
spp. Plant Cell Rep 1:104-107

Felsenburg T, Feldman M, Galun E (1987) Aneuploid and alloplasmic lines as tools for the
study of nuclear and cytoplasmic control of culture ability and regeneration of scutellar
calli from common wheat. Theor Appl Genet 74:802-810

Fischer HE (1989) Origin of the ’Weisse Schlesische Riibe’ (white Silesian beet) and
resynthesis of sugar beet. Euphytica 41:75-80

Frankenberger EA, Hasegawa PM, Tigchelaar EC (1981) Diallel analysis of shoot-forming
capacity among selected tomato genotypes. Z Pﬂanzenphysiol 102:233-242

Franklin CI, Mott RL, Vuke TM (1989) Stable ploidy levels in long-term callus cultures of
loblolly pine. Plant Cell Rep 8:101-104

Freytag AH, Anand SC, Rao-Arelli AP, Owens LD (1988) An improved medium for
adventitious shoot formation and callus induction in Beta vulgaris L. in vitro. Plant
Cell Rep 7:30-34

Galatowitsch MW, Smith GA (1990) Regeneration from unfertilized ovule callus of sugarbeet
(Beta vulgaris L.). Can J Plant Sci 70:83-89

Gladfelter 1U, Phillips GC (1987) De novo shoot organogenesis of Pinus eldarr'ca Medw. in
viuo I. Reproducible regeneration from long-term callus cultures. Plant Cell Rep
6: 163-166

Goska M (1985) Sugar beet haploids obtained in the in vitro culture. Bull Pol Acad Sci Biol
33:1-6

Hanzel JJ, Miller JP, Brinkman MA, Fendos E (1985) Genotype and media effects on callus
formation and regeneration in bariey. Crop Sci 25:27-31

Hecker RJ, Helrnerick RH (1985) Sugar-beet breeding in the United States. In: Russell GE
(ed) Progress in plant breeding. Butterworths, London pp 37-61

Heyser JW, Nabors MW (1982) Long term plant regeneration, somatic embryogenesis and
green spot formation in secondary oat (Avena sativa) callus. Z Pﬂanzenphysiol
107: 153-160

Heyser JW, Dykes TA, DeMott KJ, Nabors MW (1983) High frequency, long term
regeneration of rice from callus culture. Plant Sci Lett 29:175-182

109

Heyser JW, Nabors MW, MacKinnon C. Dykes TA, DeMott KJ, Kautzman DC, Mujeeb-Kazi
A (1985) Long-term, high-frequency plant regeneration and the induction of somatic
embryogenesis in callus cultures of wheat (Triticum aestivum L.). Z Pﬂanzenzuecht
94:218-233

Hooker MP, Nabors MW (1977) Callus initiation, growth, and organogenesis in sugarbeet
(Beta vulgaris L.). 2 Pﬂanzenphysiol 84:237-246

Hosemans D, Bossoutrot D (1983) Induction of haploid plants from in vitro culture of
unpollinated beet ovules (Beta vulgaris L.). Z Pﬂanzenzuecht 91:74-77

Horsch RB, Fry JE, Hoffrnann NL, Eichholtz D, Rogers 86, Fraley RT (1985) A simple and
general method for transferring genes into plants. Science (Wash DC) 227:1229-1231

Huizing HJ, van der Molen W, Krens FA (1988) Investigation into chromosomal stability of
in-vitro growing cells from a transformed monosomic addition plant of beet. Euphytica
8:177-184

Hussey G, Gunn HV (1984) Plant production in pea (Pisum sativum L. cvs. Puget and Upton)
from long-term callus with superﬁcial meristems. Plant Sci Lett 37:143-148

Hussey G, Hepher A (1978) Clonal propagation of sugar beet plants and the formation of
polyploids by tissue culture. Ann Bot (Lond) 42:477-479

Izhar 8, Power JB (1977) Genetical studies with petunia leaf protoplasts. I. Genetic variation
to speciﬁc growth hormones and possible genetic control of stages of protoplast
development in culture. Plant Sci Lett 8:375-383

J arl C, Bornman CH (1986) Observations on genotypic variation in Beta vulgaris (sugar beet)
tissues cultured in vitro. Hereditas 105:55-59

Jarret RL, Hasegawa PM, Bressan RA (1981) Gibberellic acid regulation of adventitious shoot
formation from tuber discs of potato. In Vitro (Rockville) 17:825-830

Joarder OI, Joarder NH, Dale PJ (1986) In vitro response of leaf tissues from [.0le
multiﬂorum - a comparison with leaf segment position, leaf age and in vivo mitotic
activity. Theor Appl Genet 73:286-291

Kaleikau EK, Sears RG, Gill B8 (1989a) Monosomic analysis of tissue culture response in
wheat (Triticum aestivum L.). Theor Appl Genet 78:625-632

Kaleikau EK, Sears RG, Gill BS (1989b) Control of tissue culture response in wheat (Triticum
aestivum L.). Theor Appl Genet 78:783-787

Kamo KK, Hodges TX (1986) Establishment and characterization of long-term embryogenic
maize callus and cell suspension cultures. Plant Sci 45:111-117

Kasperbauer MJ, Buckner RC, Bush LP (1979) Tissue culture of annual ryegrass x tall fescue
Fl hybrids: callus establishment and plant regeneration. Crop Sci 19:457-460

Kavi Kishor PB, Reddy GM (19863) Regeneration of plants from long-term cultures of Oryza
sativa L. Plant Cell Rep 5:391-393

110

Kavi Kishor PB, Reddy GM (1986b) Retention and revival of regenerating ability by osmotic
adjustment in long-term cultures of four varieties of rice. J Plant Physiol 126:49-54

Kevers C. Coumans M, De Greef W. Hoﬁnger M, Gaspar T (1981a) Habituation in sugarbeet
callus: auxin content. auxin protectors, peroxidase pattern and inhibitors. Physiol Plant
51:281-286

Kevers C, Coumans M, De Greef W, Jacobs M, Gaspar T (1981b) Organogenesis in habituated
sugarbeet callus: auxin content and protectors, peroxidase pattern and inhibitors. Z
Pﬂanzenphysiol 101:79-87

Keyes GJ, Bingham ET (1979) Heterosis and ploidy effects on the growth of alfalfa callus.
Crop Sci 19:473-476

Keyes GJ, Deaton WR, Collins GB, Legg PD (1981) Hybrid vigor in callus tissue cultures and
seedlings of Nicotiana tabaeum L. J Hered 72:172-174

Koornneef M, Hanhart CJ, Martinelli L (1987) A genetic analysis of cell culture traits in
tomato. Theor Appl Genet 74:633-641

Koornneef M, Hanhart C. Jongsma M, Toma I, Weide R, Zabel P, Hille J (1986) Breeding of
a tomato genotype readily accessible to genetic manipulation. Plant Sci 45 :201-208

Krens FA, Jamar D (1989) The role of explant souree and culture conditions on callus
induction and shoot regeneration in sugarbeet (Beta vulgaris L.). J Plant Physiol
134:651-655

Krens FA, Zijlstra C. van der Molen W. Jamar D, Huizing HJ (1988) Transformation and
regeneration in sugar beet (Beta vulgaris L.) induced by ’shooter’ mutants of
Agrobacterium truncfacr’em'. Euphytica 8:185-194

Lange W, De Bock TSM, Van Geyt JPC, Oleo M (1988) Monosomic additions in beet (Beta
vulgaris) carrying extra chromosomes of B. procumbens. 2. Effects of the alien
chromosomes on in vivo and in vitro plant development. Theor Appl Genet 76:656-
664

Larkin PJ, Scowcroft WR (1981) Somaclonal variation - a novel source of variability from cell
cultures for plant improvement. Theor Appl Genet 60:197-214

Lazar MD, Baenziger PS, Schaeffer GW (1984) Combining abilities and heritability of callus
formation and plantlet regeneration in wheat (Triticum aestivum L.) anther cultures.
Theor Appl Genet 68:131-134

Lee M, Phillips RL (1988) The chromosomal basis of somaclonal Variation. Ann Rev Plant
Physiol Plant Mol Biol 39:413-437

Locy RD (1983) Callus formation and organogenesis by explants of six Lycopersr’con species.
Can J Bot 61:1072-1079

Lowe K, Taylor DB, Ryan P, Paterson KE (1985) Plant regeneration via organogenesis and
embryogenesis in the maize inbred line B73. Plant Sci 41:125-132

Lustinec J, Horak J (1970) Induced regeneration of plants in tissue cultures of Brassr'ca

l l 1
oleracea. Experientia (Basel) 26:919-920

Majewska-Sawka A, Jassern B (1988) The comparison of in vitro morphogenetic abilities in
the Corolltnae Section of the genus Beta. Arch Zuechtungsforsch 18:285-293

Malmberg RL (1979) Regeneration of whole plants from callus culture of diverse genetic lines
of Pr'smn sativum L. Planta (Berl) 146:243-244

Mathias RJ, Fukui K (1986) The effect of speciﬁc chromosome and cytoplasm substitutions on
the tissue culture response of wheat (Triticum aestivum) callus. Theor Appl Genet
71:797-800

Mathias RJ, Simpson ES (1986) The interaction of genotype and culture medium on the tissue
culture responses of wheat (Triticum aestivum L. em. thell) callus. Plant Cell Tissue
Organ (hilt 7:31-37

Mathias RJ, Fukui K, Law CN (1986) CytOplasmic effects on the tissue culture response of
wheat (Triticum aestivum) callus. Theor Appl Genet 72:70-75

McCoy TJ, Phillips RL, Rines HW (1982) Cytogenetic analysis of plants regenerated from oat
(Avena sativa) tissue cultures: High frequency of partial chromosome loss. Can J
Genet Cytol 24:37-50

Meijer EGM (1984) Some aspects of long-term tissue culture of Srylosanthes guyanensr‘s
(Aub1.) Sw. (Leguminosae). J Plant Physiol 117:131-135

Meins F Jr (1982) Habituation of cultured plant cells. In: Kahl G, Schell JS (eds) Molecular
biology of plant tumors. Academic Press, New York. pp 3-31

Meins F Jr (1983) Heritable variation in plant cell culture. Ann Rev Plant Physiol 34:327-346

Meyer-Teuter H, Reinert J (1973) Correlation between rate of cell division and loss of
embryogenesis in longtenn tissue cultures. Protoplasma 78:273-283

Miah MAA, Earle ED, Khush GS (1985) Inheritance of callus formation ability in anther
cultures of rice, Oryza sativa L. Theor Appl Genet 70:113-116

Miedema P (1982) A tissue culture technique for vegetative propagation and low temperature
preservation of Beta vulgaris. Euphytica 31:635-643

Miedema P (1984) The effects of growth regulators on vitriﬁcation in shoot, cultures of Beta
vulgaris. Acta Bot Neerl 33:375

Miedema P, Groot PJ, Zuidgeest JHM (1980) Vegetative propagation of Beta vulgarr‘s by leaf
cuttings. Euphytica 29:425-432

Mikarni T, Sudoh R, Nagao E, Kinoshita T (1989) Genotypic variation in the in vitro
morphogenesis from leaf explants of Beta vulgaris L. and B. maritima L. Euphytica
40:271-273

Mo LH, von Arnold S, Lagercrantz U (1989) Morphogenic and genetic stability in longterrn
embryogenic cultures and somatic embryos of Norway spruce (Picea abies {L} Karst).
Plant Cell Rep 8:375-378

112

Mok MC, Mok DWS, Armstrong DJ (1978) Differential cytokinin structure-activity
relationships in Phaseolus. Plant Physiol (Bethesda) 61:72-75

Murashige T (1961) Suppression of shoot formation in cultured tobacco cells by gibberellic
acid. Science (Wash DC) 134:280

Murashige T (1964) Analysis of the inhibition of organ formation in tobacco tissue culture by
gibberellin. Physiol Plant 17:636-643

Murashige T (1974) Plant propagation through tissue cultures. Ann Rev Plant Physiol 25:135-
166

Muraslrige T, Skoog F (1962) A revised medium for rapid growth and bioassays with tobacco
tissue cultures. Physiol Plant 15:473-497

Nabors MW, Heyser JW, Dykes TA, DeMott KJ (1983) Long-duration, high-frequency plant
regeneration from cereal tissue cultures. Planta (Berl) 157:385-391

Nadolska-Orczyk A, Malepszy S (1989) In vitro culture of Cucumr‘s sativus L. 7. Genes
controlling plant regeneration. Theor Appl Genet 78:836-840

Nagamine T, Catty JP, Ford-Lloyd BV (1989a) Phenotypic polymorphism and allele
differentiation of isozymes in fodder beet, multigenn sugar beet and monogenn sugar
beet. Theor Appl Genet 77:711-720

Nagamine T, Todd GA, McCann KP, Newbury HJ, Ford-Lloyd BV (1989b) Use of restriction
fragment length polymorphism to ﬁngerprint beets at the genotype and species levels.
Theor Appl Genet 78:847-851

Nagarajan P, McKenzie JS, Walton PD (1986) Embryogenesis and plant regeneration of
Medicago spp. in tissue culture. Plant Cell Rep 5:77-80

Nagl W, Pfeifer M (1988) Karyological stability and microevolution of a cell suspension
culture of Brachycome dichromosomatica (Asteraceae). Plant Syst and Evol 158:133-
139

Narasimhulu SB, ChOpra VL, Prakash S (1989) The inﬂuence of cytoplasmic differences on
shoot morphogenesis in Brassr'ca carinata A. Br. Euphytica 40:241-243

Niedz RP, Smith 88, Dunbar KB, Stephens Cl‘, Murakishi HH (1989) Factors inﬂuencing
shoot regeneration from cotyledonary explants of Cucumr‘s melo. Plant Cell Tissue
Organ Cult 18:313-319

Oelck MM, Schieder O (1983) Genotypic differences in some legume species affecting the
redifferentiation ability from callus to plants. 2 Pﬂanzenzuecht 91:312-321

Orshinsky BR, Tomes DT (1985) Effect of long-term culture and low temperature incubation
on plant regeneration from a callus line of birdsfoot trefoil (Lotus corniculatus L.). J
Plant Physiol 1192389-397

Owen FV (1945) Cytoplasmically inherited male-sterility in sugar beets. J Agric Res 71:423-
440

113
Ozawa K, Komamine A (1989) Establishment of a system of high-frequency embryogenesis
from long-term cell suspension cultures of rice (Oryza sativa L.). Theor Appl Genet
77:205-211

Petolino JF, Jones AM, Thompson SA (1988) Selection for increased anther culture response
in maize. Theor Appl Genet 76:157-159

Powell W, Caligari PDS (1987) The in vitro genetics of barley (Hordeum vulgare L.):
detection and analysis of reciprocal differences for culture response to 2.4-
dichlorophenoxyacetic acid. Heredity 59:293-299

Powell W, Dunwell JM (1987) In vitro genetics of barley (Hordeum vulgare L.): Response of
immature embryos to 2,4-dichlor0phenoxyacetic acid. Heredity 58:75-80

Quimio CA, Zapata FJ (1990) Diallel analysis of callus induction and green-plant regeneration
in rice anther culture. Crop Sci 30:188-192

Reisch B, Bingham ET (1980) The genetic control of bud formation from callus cultures of
diploid alfalfa. Plant Sci Lett 20:71-77

Ritchie GA, Short KC, Davey MR (1989) In vitro shoot regeneration from callus, leaf axils
and petioles of sugar beet (Beta vulgaris L.). J Exp Bot 40:277-283

Rogozinska J, Goska M (1978) Induction of differentiation and plant formation in isolated
sugar beet leaves. Bull Pol Acad Sci Ser Sci Biol 26:343-345

Rogozinska JH, Goska M, Kuzdowicz A (1977) Induction of plants from anthers of Beta
vulgaris cultured in vitro. Acta Soc Bot Pol 46:471-479

Romagosa I, Hecker RJ, Tsuchiya T, Lasa JM (1986) Primary trisomics in sugarbeet. 1.
Isolation and morphological characterization. Crop Sci 26:243-249

Roustan JP, Latche A, Fallot I (1989) Stimulation of Daueus carota somatic embryogenesis by
inhibitors of ethylene synthesis: cobalt and nickel. Plant Cell Rep 8:182-185

Sacristan MD (1981) Regeneration of plants from long-term callus cultures of haploid Brassica
napus. Z Pﬂanzenzuecht 86:248-253

SAS Institute Inc (1985) SAS user’s guide:Statistics, version 5 edition; Cary, NC

Satterthwaite FE (1946) An approximate distribution of estimates of variance components.
Biometrics 2:110-114

Saunders JW (1982) A ﬂexible in vitro shoot culture propagation system for sugarbeet that
includes rapid ﬂoral induction of ramets. Crop Sci 22:1102-1105

Saunders JW, Bingham ET (1972) Production of alfalfa plants from callus tissue. Crop Sci
12:804-808

Saunders JW, Bingham ET (1975) Growth regulator effects on bud initiation in callus cultures
of Medicago sativa. Amer J Bot 62:850—855

Saunders JW, Daub ME (1984) Shoot regeneration from hormone-autonomous callus from

114

shoot cultures of several sugarbeet (Beta vulgarr‘s L.) genotypes. Plant Sci Lett 34:219-
223

Saunders JW, Doley WP (1986) One step shoot regeneration from callus of whole plant leaf
explants of sugarbeet lines and a somaclonal variant for in vitro behavior. J Plant
Physiol 124:473-479

Saunders JW, Shin K (1986) Gennplasm and physiologic effects on induction of high-
frequency hormone autonomous callus and subsequent shoot regeneration in sugarbeet.
Crop Sci 26:1240-1245

Savitsky VF (1950) Monogerm sugar beets in the United States. Proc Amer Soc Sugar Beet
Technol 6:156-159

Seitz MH, Stalker HT, Green CC (1987) Genetic variation for regenerative response in
immature leaﬂet cultures of the cultivated peanut, Arachis hypogaea. Plant Breed
98: 104-1 10

Sen 8, Newton RJ, Fong F, Neuman P (1989) Abscisic acid: a role in shoot enhancement from
loblolly pine (Pinus taeda L.) cotyledon explants. Plant Cell Rep 8:191-194

Singh BD (1975) Evolution of dominant karyotypes in Haplopappus gracilr's cells cultured in
vitro. Caryologia 28:29-37

Skirvin RM (1978) Natural and induced variation in tissue culture. Euphytica 27:241-266

Skoog F, Miller CO (1957) Chemical regulation of growth and organ formation in plant tissues
cultured in vitro. Symp Soc Exp Biol 11:118-131

Skvirsky RC, Hanson MR, Ausubel FM (1984) Intraspeciﬁc genetic variation in cytokinin-
controlled shoot morphogenesis from tissue explants of Petunia hybrida. Plant Sci Lett
35:237-246

Smed E, Van Geyt JPC, Oleo M (1989) Genetical control and linkage relationships of isozyme
markers in sugar beet (B. vulgaris L.). 1. Isocitrate dehydrogenase, adenylate kinase,
phosphoglucornutase, glucose phosphate isomerase and cathodal peroxidase. Theor
Appl Genet 78:97-104

Smith GA (1987) Sugar beet. In: Fehr WR (ed) Principles of cultivar development, vol 2 crop
species. Macmillan, New York, pp 577-625

Smith SM, Street HE (1974) The decline of embryogenic potential as callus and suspension
cultures of carrot (Daucus carota L.) are serially subcultured. Ann Bot (Lond) 38:223-
241

Snyder FW (1974) Comparative leaf area and dry matter accumulation by maize and
sugarbeet. Crop Sci 14:529-533

Songstad DD, Drmcan DR, Widholrn JM (1988) Effect of l-aminocyclopropane-l-carboxylic
acid, silver nitrate, and norbomadiene on plant regeneration ﬁom maize callus cultures.
Plant Cell Rep 7:262-265

Stavarek SJ, Croughan TP, Rains DW (1980) Regeneration of plants from long-term cultures

115
of alfalfa cells. Plant Sci Lett 19:253-261

Steel RGD, Torrie JH (1980) Principles and procedures of statistics - a biometrical approach,
2nd ed. McGraw-Hill. New York

Steen P, Kcirner B, D’Halluin K, Pedersen HC (1986) Variability in plants of sugar beet (Beta
vulgaris L.) regenerated from callus, cell suspension and protoplasts. In: Horn W et al
(eds) Genetic Manipulation in Plant Breeding, de Gruyter, Beriin pp 633-635

Stiebelirrg B, Neumann K-H (1987) Identiﬁcation and concentration of endogenous cytokinins
in carrots (Daucus carota L.) as inﬂuenced by development and a circadian rhythm. J
Plant Physiol 127:111-121

Suresh Kumar A, Reddy TP, Reddy GM (1985) Genetic analysis of certain in vitro and in vivo
parameters in pigeonpea (Cajanus cajan L.). Theor Appl Genet 70:151-156

Sutter E, Langhans RW (1981) Abnormalities in Chrysanthemum regenerated from long term
cultures. Ann Bot (Lond) 48:559-568

Syono K, Furuya T (1972) Effects of cytokinins on the auxin requirement and auxin content of
tobacco calluses. Plant Cell Physiol 132843-856

Szabados L, Gaggero C (1985) Callus formation from protoplasts of a sugarbeet cell
suspension culture. Plant Cell Rep 4:195-198

Szakacs E, Kovacs G, Pauk J, Barnabas B (1988) Substitution analysis of callus induction and
plant regeneration from anther culture in wheat (Triticum aestivum L.). Plant Cell Rep
7 :127-129

Templeton-Somers KM, Collins WW (1986) Heritability of regeneration in tissue cultures of
sweet potato (Ipomoea batatas L.). Theor Appl Genet 71:835-841

Tetu T, Sangwan RS, Sangwan-Noneel BS (1987) Hormonal control of organogenesis and
somatic embryogenesis in Beta vulgaris callus. J Exp Bot 38:506-517

Tomes DT, Smith 08 (1985) The effect of parental genotype on initiation of embryogenic
callus from elite maize (Zea mays L.) germplasm. Theor Appl Genet 70:505-509

Torrey JG (1967) Morphogenesis in relation to chromosomal constitution in long-term plant
tissue cultures. Physiol Plant 20:265-275

USDA (1988) Agricultural statistics 1988. US Government Printing Ofﬁce, Washington DC,
pp 75-77

Van Geyt JPC, Jacobs M (1985) Suspension culture of sugarbeet (Beta vulgaris L.). Induction
and habituation of dedifferentiated and self-regenerating cell lines. Plant Cell Rep
4:66-69

Van Geyt JPCF, Smed E (1984) Polymorphism of some marker enzymes of the sugarbeet
(Beta vulgaris L.) investigated by polyacrylamide gel electrophoresis and starch gel
electrophoresis. Z Pﬂanzenzuecht 92:295-308

Van Geyt JPC, Oleo M, Lange W, De Bock TSM (1988) Monosomic additions in beet (Beta

116

vulgaris) carrying extra chromosomes of Beta procumbens. 1. Identiﬁcation of the
alien chromosomes with the help of isozyme markers. Theor Appl Genet 76:577-586

Van Geyt J, Speckrnann GJ, D’Halluin K, Jacobs M (1987) In vitro induction of haploid plants
from unpollinated ovules and ovaries of the sugarbeet (Beta vulgaris L.). Theor Appl
Genet 73:920-925

Vapper M, Kallak H (1986) Karyotypic differentiation of long-term callus cultures of Crepr‘s
capillaris. Biol Plant (Prague) 28:417-423

Vardi A, Hutchinson DJ, Galun E (1986) A protOplast-to-tree system in Microcr‘trus based on
prot0plasts derived from a sustained embryogenic callus. Plant Cell Rep 5:412-414

Weigel RC Jr, Hughes KW (1985) Long term regeneration by somatic embryogenesis in barley
(Hordeum vulgare L.) tissue cultures derived ﬁom apical meristem explants. Plant
Cell Tissue Organ Cult 5:151-162

Welander T (1974) Callus and root formation in explants of Beta vulgaris. Physiol Plant
32:305-307

Wenck AR, Conger BV, Trigiano RN, Sams CE (1988) Inhibition of somatic embryogenesis in
orchardgrass by endogenous cytokinins. Plant Physiol (Bethesda) 88:990—992

White DWR (1984) Plant regeneration from long-term suspension cultures of white clover.
Planta (Berl) 162:1-7

Wyndaele R, Christiansen J, Horseele R, Rudelsheirn P, Van Onckelen H (1988) Functional
correlation between endogenous phytohorrnone levels and hormone autotrophy of
transformed and habituated soybean cell lines. Plant Cell Physiol 29:1095-1101

Yu NIH (1989) Callus induction and differentiation from leaf explants of different species of
the genus Beta. Crop Sci 29:205-209

Zelcer A, Soferrnan O, Izhar S (1984) An in vitro screening of tomato genotypes exhibiting
efﬁcient shoot regeneration. J Plant Physiol 115:211-215