Q
Wins-rig;

.. , -....-
'1; “Fig—4 -
ti.“

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mm [W mm‘mmmn'nm: .

1293 10681 0868

 

 

 

 

 

 

AI

This is to certify that the

dissertation entitled

GENETICS OF NICOTIANA SOMATIC HYBRIDS

 

presented by

Jean Louise Roberts

has been accepted towards fulfillment

of the requirements for
Crop and Soil Sciences
Ph . D . degree in and Genetics Program

 

   

 

 

Major professor

Date 8/12/82

 

MS U is an Affirmative Action/Equal Opportunity Institution 0-12771

-\i"

RETURNING MATERIALS:
MSU Place in book drop to

remove this checkout from
w your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

 

 

 

 

 

 

 

 

 

THE GENETICS OF NICOTIANA SOMATIC HYBRIDS

 

BY

Jean Louise Roberts

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Crop and Soil Sciences
and
Genetics Program

1982

ABSTRACT

THE GENETICS OF NICOTIANA SOMATIC HYBRIDS

 

by

Jean Louise Roberts

Somatic hybridization provides one approach to a better
understanding of the cytoplasmic genetics of higher plants.
Protoplast fusion experiments combined homozygous nuclear
albino, cytoplasm-substituted cells with nuclear wild-type,

plastid albino cells. Nicotiana tabacum with ”white

 

seedling" or "Sulfur" nuclear albino markers, y.

megalosiphon or E. bigelovii cytoplasms, and three different

 

plastid albino genotypes were used. Complementation between
different parental nuclear and cytoplasmic genomes permitted
the selection of eight green fusion products during culture
regeneration. The resulting somatic hybrid plants were
tested for nuclear, mitochondrial and plastid markers.
Nuclear inheritance was examined by progeny tests for
the nuclear albino marker and by ploidy determinations. The
different mitochondrial genotypes were distinguished by a
cytoplasmic male-sterile character. The albino/green visual

marker, ribulose bisphosphate carboxylase/oxygenase

Jean Louise Roberts
isoelectric points, and tentoxin reactions marked the
different plastid genomes. Results showed that these eight
somatic hybrids include four different combinations of the
marked nuclear, plastid and mitochondrial genomes.

The hybrid nuclear genotypes combined the parental
nuclei in six plants; the albino parental nucleus was
completely lost from two hybrid plants. The mitochondrial
male-sterile marker was inherited independently from the
plastid markers in these hybrids. Two plants had
male-fertile anatomy, and six were male-sterile; no new
floral phenotypes occurred. No recombination between the
three plastid characters was detected in the eight green
plants, but one variegated somatic hybrid plant was
recovered.

Additional protoplast fusion experiments selected for
recombination between different plastid albinos in somatic
hybrids. No plastid recombinants were found.

‘ The genotypes employed in these experiments permitted
fusion product selection by complementation between nuclear
and cytoplasmic albino markers and somatic hybrid analysis
for three different cell genomes. Independent assortment of
plastids and mitochondria in these hybrids resulted in new
plant types with plastids and mitochondria derived from
different Nicotiana species. These results contribute to
our understanding of the genetic consequences of cytoplasmic

hybridization in higher plants.

TABLE OF CONTENTS

 

Page

LIST OF TABLES iv
LIST OF FIGURES v
I. INTRODUCTION 1
1.1. Prologue 1
1.2. Concepts of organelle genetics 2
1.2.1. Introduction 2

1.2.2. Meiotic inheritance 2

1.2.3. Mitotic inheritance 3

1.2.4. Molecular aspects 3

1.2.5. Differentiation 5

1.3. Mitochondrial genetics 5
1.3.1. Yeast 5

1.3.2. Mitochondrial genetics of animals 11

1.3.3. Plant mitochondria 13

1.4. Chloroplast genetics 19
1.4.1. Introduction 19

1.4.2. Chlamydomonas 20

1.4.3. Higher plant chloroplast genetics 26

1.5. Experimental introduction 35
1.5.1. Theoretical argument 35

ii

iii
1.5.2. Somatic hybrids
1.5.3. Outstanding published examples
II. MATERIALS AND METHODS
11.1. Experimental design
11.2. Construction of parental material
11.2.1. Seed sources
11.2.2. Nuclear albino plants
11.2.3. Cytoplasmic albino plants
11.2.4. Nuclear albino shoot cultures
11.2.5. Cytoplasmic albino shoot cultures
11.3. Protoplast fusion
11.3.1. Protoplast isolation and culture
11.3.2. Protoplast regeneration
11.3.3. Protoplast fusion technique
11.3.4. Fusion experiments
11.4. Analysis of somatic hybrids
III. RESULTS
111.1. Recovery of complementing fusions
111.2. Failure to recover selected recombinant
fusions
111.3. Analysis of hybrid plants
111.3.1. Introduction
111.3.2. Macrosc0pic leaf characters
111.3.3. Floral characters
111.3.4. Nuclear genes
111.3.5. Plastid characters
111.4. Summary and conclusions

LIST OF REFERENCES

36
41
45
45
50
50
54
57
58
64
67
67
72
74
76
79
86
86
89

92
92
98
103
107
118
124
131

Table

Table
Table

Table

Table

Table

Table

Table

Table

2.
3.
4.

6.
7.

LIST OF TABLES

Initial seed stocks: sources, references,
descriptions

Tissue culture media
Protoplast culture media

Somatic hybrids recovered by visual
selection

Cytoplasmic parentage, floral phenotype, and
suggested mitochondrial genotype of the
somatic hybrid plants

Progeny test expectations

Results of progeny tests for nuclear markers
in somatic hybrids

Results of analysis of somatic hybrids for
plastid markers

Summary of somatic hybrid analysis for

nuclear, plastid and mitochondrial
characters

iv

Page
52

60
70
93

105

109

110

125

126

Figure

Figure

Figure

Figure

Figure

Figure

Figure
Figure
Figure
Figure
Figure
Figure

Figure

Figure

Figure

Figure

5.
6.

7.
8.
9.
10.
11.

LIST OF FIGURES

Somatic hybridization plan using
complementation between nuclear and
cytoplasmic albino markers to select the
somatic hybrid

Somatic hybridization plan designed to
select for ctDNA recombination

Crossing plan generating homozygous nuclear
albino cytoplasmic substitution lines

Density gradient centrifugation to collect
albino protoplasts

Albino protoplasts growing in medium D3

UV-fluorescent micrograph of a fusion
product

Somatic hybrid #1 and its parents
Somatic hybrid #3 and parental genotypes
Somatic hybrid #4 and parental genotypes
Somatic hybrid #5 and its parents
Selection of somatic hybrid #6

Hybrid #1 potted in soil

Somatic sectoring for the nuclear marker
in a plant of somatic hybrid #5

Variegated plantlet from cultures of
somatic hybrid #5

Propagation of variegated plantlet from
cultures of somatic hybrid #3

Variegated plant of somatic hybrid #3
potted in soil

Page

48

49

56

71

71
77

94
94
95
95
96
96
97

101

101

102

vi

Figure 17. Somatic hybrid #1 displaying the cms floral 102

phenotype

Figure 18. Gel isoelectric focusing assay for 121
RuBPCase LS plastid genotypes of hybrids
#1 and #5

Figure 19. Gel electrophoretic assay for plastid 122

genotype

INTRODUCTION

1.1. Prologue

This thesis discusses the theory and methods of
cytoplasmic genetics and describes somatic hybridization
experiments applying those methods to plant cells. These
experiments were designed to yield both new techniques and
information, not only to extend the formal knowledge of
plant genetics, but also to make useful genetic variation
accessible to plant breeders. Plastid genes influence
economically important characters, such as chemical and
disease resistance and photosynthetic efficiency;
mitochondrial genes are implicated in cytoplasmic male
sterility. Techniques to manipulate these genes could
contribute to crop improvement.

Chapter 1 of this thesis examines the extent of our
present-day understanding of organelle genetics. Chapter 11
details the methods used to obtain Nicotiana somatic hybrids
with marked nuclear, plastid and mitochondrial genomes.
Chapter III presents the results of analysis of the hybrid

plants.

2

1.2. Concepts of organelle genetics

1.2.1. Introduction

This survey of cytoplasmic genetics will place into
perspective our current understanding of cytoplasmic
genetics in plants. Several principles and techniques of
cytoplasmic genetics differ from those of other branches of
genetics. Genes outside the cell's nucleus have distinct
non-Mendelian patterns of inheritance due to their
independence from the cell processes and cytoskeletal
machinery which govern the replication and distribution of
chromosomal genes in meiosis and mitosis. Cytoplasmic
genomes can vary greatly in copy number relative to nuclear
genes, giving rise to quantitative genetic phenomena which
require population analysis. This discussion will cover
only mitochondria and chloroplasts, ignoring other heritable
non-nuclear phenomena such as viruses, symbionts and
maternal effects. The DNAs of mitochondria and chloroplasts
behave genetically as distinct populations of prokaryotes,

c00perating with their host cells in eukaryotic organisms.

1.2.2. Meiotic inheritance

The meiotic and mitotic genetics of cytoplasmic genomes

follow the bulk distribution of cytoplasm in cell fusions

3
and divisions. Thus, the number of genome copies each
gamete effectively contributes to the zygote determines the
inheritance of mitochondria and chloroplasts in the
organism's sexual cycle. Differences in reciprocal crosses
can result if gametes differ in size or if one parental
organelle DNA is less competitive in replication. For these
reasons, heterogametic species frequently exhibit maternal
bias when organelle traits are followed through sexual

crosses.

1.2.3. Mitotic inheritance

Mitosis distributes cytoplasm to cells irregularly.
Random effects mediated by the symmetry of cell division,
the vigor of cyt0plasmic mixing, the number of genome
copies, and any bias in replication cause genetic
segregation of organelle DNAs in mitotic cell divisions.
Thus, a cell containing a heterogeneous population of
mitochondria or chloroplasts can beget different cell
lineages which are uniform for organelle genotype. Somatic
segregation is a phenomenon typical of cytoplasmic

genetics.

1.2.4. Molecular aspects

Mitochondria and chloroplasts sequester special

metabolic functions into membrane-delimited compartments.

4
The DNAs maintained within these organelles encode parts of
enzyme complexes essential for respiration and
photosynthesis, respectively, as well as parts of the
translation machinery needed to express those structural
genes in situ. Both mitochondria and chloroplasts also
compartmentalize other biochemical functions, thereby
contributing to the efficiency and regulation of cell
metabolism. Physical fractionation in yitrg separates both
the biochemistry contained within the organelle and its
genetic system from those of nucleus and cytoplasm.

The apparatus for replication and genetic expression
encoded in organelle genomes shares several characteristics
with prokaryotic systems, notably DNA circularity, polyteny
and membrane attachment, and ribosome size, structure and
sensitivity to antibacterial antibiotics. While suggestive
of a disparate evolutionary origin for nuclear and organelle
genomes, these features have also been widely exploited in
the study of organelle genetics and biochemistry. Physical
differences, antibiotic sensitivity, and the use of
organisms which can tolerate organelle dysfunction have
permitted in gigg dissection of nuclear and organellar
contributions to metabolic traits. It is generally observed
that the nuclear and organelle genomes cooperate to produce

the organelle phenotype.

1.2.5. Differentiation

The physical separation of organelle DNAs into
different membrane-bounded compartments can limit their
genetic interaction. Eukaryotic differentiation involves
organelle differentation, with accompanying changes in the
DNA copy number within the organelle, the number of
organelles per cell, physical characteristics of the
membranes, and in the level of expression of organelle
genes. This phenomenon of organelle differentiation affects
the study of chloroplast and mitochondrial traits.

The biology of organelles dictates that inheritance
studies of mitochondria and chloroplasts draw from
population genetics to describe quantitative effects, while
using molecular approaches to the specialized biochemistry
of these subcellular compartments. Examples of the
phenomena and techniques of organelle genetics in
unicellular eukaryotic models and in higher animals and

plants are detailed below.

1.3. Mitochondrial genetics

1.3.1. Yeast

 

The paradigm of mitochondrial genetics is the baker's

yeast, Saccharomyces cerevisiae. The combination of

6
historical precedence (l), facultative organelle function,
and active genetic recombination have made yeast
mitochondrial DNA (mtDNA) the best understood organelle
genome. Though some features are not typical of all
organelles, the study of yeast mitochondrial genetics
contributes much to our understanding of organelle DNA
inheritance and expression.

Yeast are homogametic. Each gamete of normal genotype
and physiology contributes roughly the same population size
of mitochondria to the zygote. When populations descended
from a mass mating are examined for organelle markers,
biparental inheritance and recombination of the markers are
observed. This evidence implies that mitochondria must
interact and their DNAs associate. Recombination is so
frequent that an unambiguous genetic map cannot be
constructed by this method. In contrast, pedigree studies
of mitotic progeny from single zygotes show rapid
segregation and fixation of markers (2). The most important
form of exchange may be nonreciprocal gene conversion
(3,4).

The mitochondrial DNA of yeast is circular, with a size
of 50 megadaltons, corresponding to 76 kilobase pairs (2,5).
Though a haploid cell contains about 50 copies of mtDNA
within about 20 organelles (4), it behaves genetically as if
about ten mtDNAs were replicating (2,3). Several factors
may contribute to this apparent discrepancy in copy number.

In yeast, mitosis by budding gives the daughter cells

unequal amounts of cytoplasm. Genetic segregation in
mitosis could be accelerated by incomplete cytoplasmic
mixing, asymmetric cell division, segregation by organelle
units, replicative competition, and by mitochondrial gene
conversion. These would limit the duration of the
heterozygous condition and thereby lower recombination and
mutation frequencies in pedigree studies (2,4). A limited
number of replication enzymes or DNA-membrane attachment
sites may be available (4). It has also been suggested that
recombination occurs only when mitochondria are
dedifferentiated for a brief period in the zygote (2), and
that the state of differentiation of the mitochondria,
primarily due to demand for aerobic respiration, affects the
number of mtDNA molecules (6).

Yeast are facultative anaerobes; their mitochondrial
genomes are studied with respiratory defective mutants,
antibiotic resistant mutants, and with petites. Petites are
a class of respiratory defectives with a characteristic high
frequency of occurrence spontaneously or following
mutagenesis (2,3,5). They have been shown to contain
deletions and amplifications of mtDNA. In some petites, a
few sequences are retained and amplified to whole genome
size (5); the amplified tandem repeat may be either nonsense
or coding sequences. Other petites have no detectable
mitochondrial DNA. Both types of petites develop no
functional mitochondrial ribosomes and cannot express their

mtDNA complement. The phenomenon generating and tolerating

8
these deletions is not known to exist in any other organelle
genetic system.

Deletion mutagenesis to the petite condition by
treatment with intercalating acridine dyes permits mtDNA
mapping by co-retention of sequences. This approach yielded
a reliable circular map of mtDNA (5). Petites containing
mtDNA must be crossed to a respiratory-sufficient genotype
to express the retained sequences. This provides another
example of organelle differentiation affecting the study of
organelle genes. Complementation mapping of mitochondrial
genes was possible in zygotes due to mitochondrial fusion
(5). When progeny from petite crosses are examined, it is
evident that some petites have a strong advantage in mtDNA
replication suggesting intracellular selection (4).

Genes definitely located on mtDNA can be enumerated.
Mitochondrial ribosomal RNAs are mtDNA-encoded. Yeast
mitochondria lack the SS rRNA (5). variation in these loci
confers useful resistance markers for the two antibiotics
erythromycin and chloramphenicol (5). All 26 transfer RNAs
found in mitochondria are mtDNA-encoded (5). Mutants in
these display a lack of protein synthesis but have a
complete ribosomal complement. One mtDNA-encoded
polypeptide in yeast is a single ribosomal protein; its
mutant phenotype is variable. The rest of the ribosomal
proteins are synthesized outside the organelle and are

encoded in the nucleus (5).

9
The yeast mitochondrial translation system is less
prokaryotic in structure than in most organelles.
Heterologous associations of ribosome subunits or messages

plus ribosomes from E. coli and s. cerevisiae will not

 

synthesize protein, though translation apparati of g. coli
and Chlamydomonas chloroplast components will function

together in vitro (5). Too, g. cerevisiae mitochondrial

 

rRNA is processed from a transcript containing an intron
(5). Different tRNA anticodons are used (5). These
features distinguish yeast mitochondria from other organelle
genetic systems and may reflect a more advanced, eukaryotic
trend in evolutionary adaptation for this organelle.

Respiratory defective mutants with normal mitochondrial
DNA and ribosomes map to seven mtDNA loci with known gene
products comprising one out of the seven subunits of the
cytochrome b complex, and three out of seven peptides of the
cytochrome oxidase assembly (5). Mutants in these loci have
atypical cytochrome absorbance spectra. The chemiosmotic
transducer, Coupling Factor Fl, includes three subunits
encoded in mtDNA and seven encoded in the nucleus. Mutants
in these genes are resistant to ionophores or to the
antibiotic oligomycin (5). These hydrophobic proteins may
require synthesis close to the site of assembly of the
membrane-bound enzyme complexes; they would be unable to
cross the mitochondrial membranes as do other

nuclear-encoded, cytoplasmically translated components (7).

10

Expression of the mitochondrial genes is mediated by
nuclear genes. In addition to the remaining components of
the above-mentioned multimeric complexes of ribosomes and
the electron transport system, mitochondrial DNA and RNA
polymerases are nuclear-encoded (2). The mode of regulation
of expression of cooperating nuclear and mitochondrial genes
in producing organelle phenotype is as yet undiscovered.

MtDNA must contain a replication origin sequence (5).
This origin is identifiable in petites with a strong mtDNA
replication advantage. Physical techniques show that such
successfully competitive petites have many copies of the
replication sequence, whereas wild-type yeast have several
such loci (8). One other known genetic locus, “omega“,
confers bias of mitochondrial gene transmission in crosses
and is probably an initiation site for gene conversion (5).
The generality of this gene conversion bias factor in
organelle systems is unproven (9).

Four additional hydrophobic polypeptides with unknown
function are encoded by mtDNA (10). There may be other
coding sequences as yet undiscovered in yeast mtDNA.

The population genetics of yeast mtDNA transmission has
drawn from phage genetics to describe a panmictic pool of J)
replicating, recombining DNAs (4,9). Concepts of random
drift (4,11) and competition (9) derive from evolutionary
population genetics. The molecular aspects of copy number
and recombination mechanism are obscured by the effects of

organelle differentiation, asymmetric cell division, and

ll
prolific gene conversion during active recombination. A
quantitative model to adequately describe all known features

has not been completely worked out.

1.3.2. Mitochondrial genetics of animals

Animal cells cannot dispense with mitochondrial
function. The unavailability of facultative fermenters
limits genetic studies to antibiotic resistance markers or
physical characterizations of the mitochondrial genome.
Transmission genetics of animal mitochondria has been
restricted to studies of maternal transmission and
evolutionary mtDNA sequence divergence in nature (12-14),
and to transfers of antibiotic resistances (15,16) and mtDNA
restriction patterns by animal cell fusions (17). In
contrast, the molecular study of animal mitochondria has
proven richly productive.

Human mitochondrial DNA is among the smallest of
organelle genomes, being a 16.5 kilobase pairs long circle
(18), of 10 megadaltons molecular weight (2). A laborious
collaboration recently produced the complete nucleotide
sequence of human mtDNA from Hela and placental preparations
(19). This remarkable piece of work depended on both
molecular biological and information processing advances of
the past decade.

Restriction fragments of human mtDNA, cloned in

E. coli, provided templates for sequencing by the Sanger

12
dideoxy chain termination method; a computer program
correlated the results (19). The mtDNA sequence obtained
was compared with the 5' and 3' terminal sequences of
mitochondrial polyA RNAs produced in giyg, and with known
amino acid sequences of mitochondrial proteins, to clarify
transcription origins and termini (20,21). The results
describe a highly evolved, compact genome with unusual
features contributing to its economy of maintenance. New
genes were discovered, and previously postulated functions
were assigned loci (19).

The replication origin of this mtDNA and its
transcription and translation origins are now known. Both
strands of mtDNA are completely transcribed with a single
promoter each, which have also been identified.
Posttranscriptional processing to mRNA is apparently
mediated by immediately adjacent tRNA sequences. The
previously known mtDNA structural genes for three cytochrome
c oxidase subunits, one cytochrome b subunit, and F1 ATPase
subunits can now be precisely located. Eight additional
previously unknown protein-encoding genes were discovered in
the mtDNA sequence; transcripts of most of these could then
be identified in mitochondrial polyA-mRNA preparations.
Economy in transcription and translation also results from
the use of only two undersized rRNAs (no SS rRNA), a lack of
mRNA leader sequences, the synthesis of translation termini
by posttranscriptional polyadenylation of transcripts, and

by the sufficiency of only 22 tRNAs. Remarkably, the

13
'universal' genetic code proves to not apply to human
mitochondria. These organelles employ different translation
initiation and termination codons.

Physical characterization of animal mtDNA suggests that
base substitution plays a much greater role in its evolution
than does rearrangement. Very little physical rearrangement
of mtDNAs is found when distantly related species are
compared (13).

Some features of animal mitochondria clearly
distinguish them from prokaryotic systems. Mitochondrial
ribosomes are structually different from, and share little
sequence homology with, bacterial ribosomes (19).
Translation depends upon some different codons. The
tRNA-mediated posttranscriptional processing of
mitochondrial mRNA has no precedent in prokaryotes. Unlike
the situation in yeast, economy of code in animal mtDNA
evidently permits no introns (19). Evolutionists might
speculate profitably on the processes by which loss and
modification of organelle genomes could yield the existing
differences and similarities between prokaryoes and the

various organellar systems.

1.3.3. Plant mitochondria

 

Plant mitochondria present a somewhat different

experimental picture. Unlike yeast, plant cells appear to

14
be obligate aerobes. There are presently no antibiotic
resistance markers for use for ig_gigg meiotic or mitotic
genetic analysis of the organelle. Additionally, because
plant cells enclose two separate organelle systems, it
becomes essential to separate the contributions of
mitochondria from those of chloroplasts in determining plant
characters. For these reasons, physical analysis of plant
mtDNA and in gitrg studies on mitochondrial protein
synthesis have supplied almost all the genetic information
currently available about this organelle.

In contrast to animal and fungal mitochondria, each
plant mitochondrion can enclose different size classes of
circular DNAs which seem to represent various rearrangements
of a basic mitochondrial genome, with the possible addition
of episomal or plasmid-like genetic elements (22,23). It is
unclear whether all DNA size classes contain the same
sequences; different mtDNA species could coexist and undergo
mutual rearrangements. One piece of evidence suggestive of
mtDNA rearrangement will be discussed below.

Evolutionary distance between plant species is
reflected in mtDNA sequence rearrangements, rather than base
substitutions as is the case in animals (24). The mtDNA of
plants is much larger than the minimal animal mtDNA, and its
isolation in different size classes makes it difficult to
assign a molecular weight to the basic mitochondrial genome
by analysis of restriction digests (22). Estimates of mtDNA

size by these methods range from 70 megadaltons (25) to 320

15
megadaltons (24). Measurements of reassociation kinetics
showed that in different cucurbit species, mitochondrial DNA
ranges from 220 to 1600 megadaltons in sequence complexity,
with only a small repetitive component (26). The origin and
function of this mtDNA sequence abundance in plants, and its
arrangement on variable plant mtDNA molecular sizes, are
unknown. Regardless of a species' mtDNA size, each plant
cell contains 100 to 150 mtDNA copies (26). There are cases
of mtDNA replication surges accompanying tissue
differentiation.

The plant mitochondrial DNA encodes all three
mitochondrial rRNAs, including a SS rRNA (25). A comparison
of proteins synthesized in yitgg by plant mitochondria with
those made by isolated animal and fungal mitochondria showed
that plants produce polypeptides corresponding in size to
the known mtDNA-encoded gene products in yeast and man:
three cytochrome c oxidase subunits, one cytochrome b
subunit, and at least two Fl ATPase subunits (25). However,
no enzyme activity has actually been identified for these
plant mitochondrial products.

Cytoplasmic male sterility (cms) is one very
significant whole-plant character which has been correlated
with mitochondrial inheritance. The phenomenon is widely
used by seedsmen, but significant advances in understanding

have been gained only in Zea and Nicotiana. Expression of

 

cms is tissue-specific; it requires floral development.

This effect of plant differentiation on expression of

16
organelle-determined characters has hampered the study of
cms. For complete coverage of this topic, the reader is
referred to an excellent monograph (27).

For purposes of this discussion, cytoplasmic male
sterility should be distinguished from viral effects, which
are mechanically transmissible, and from nuclear genes
conferring male sterility, which follow Mendelian genetics.
Nuclear genes contribute to the expression of
cytoplasmically-inherited male sterility; in grain crop seed
production, each cms genome is combined with nuclear genes
conferring restoration of fertility (22). Nuclear
background can also influence the frequency of
cytoplasmically-inherited reversion to fertility (22).
Because the primary genetic determinant for the cms
character is strictly maternally inherited, it seems that
there is no male transmission of the relevant cytoplasmic
genome. The physiology of male organ or gamete failure
varies with the cytoPlasmic genome. The contribution of any
specific mtDNA sequence to the physiological or
developmental breakdown of male gametogenesis is not yet
known for any case.

Linkage of cms with disease susceptibility caused a
renewed effort in plant mitochondrial genetics. In corn,
Egg gays, a cytoplasm-linked susceptibility to Southern corn
leaf blight contributed to a 1970 epidemic and severe crop
loss because one single male-sterile cytoplasm was

extensively used in seed production (22,27,28). The blight

l7

fungus, Helminthosporium maydis (Bipolaris maydis), produces

 

a toxin which causes breakdown of the mitochondria from the
disease-susceptible 'T' male-sterile cytoplasm (22). Only
the mitochondria are affected; plastids from the same
cytoplasm are unaffected by the toxin (29).

Physical characterization showed that mtDNAs of
different sterile and fertile corn cytoplasms differ more
than do their chloroplast DNAs (22). Further examination
revealed that the genetically variable '8' male-sterile
cytoplasm contains episome-like small mitochondrial DNAs in
addition to the larger mtDNAs typical of corn. These small
DNAs have palindromic repeat termini; they disappear from
native isolations of mtDNA concurrently with plant reversion
to cytoplasmically-inherited male fertility (30). The time
required to observe this event is about a plant generation.
These episomal sequences are present in the fertile corn
mtDNA, but their presence on separate mtDNA molecules in the
sterile genotype apparently affects male gamete development
(30). The rearrangements found in the fertile 'S'
revertants' mtDNA, along with apparent loss of their
episomal mtDNA, are suggestive of transposition and
nonhomologous recombination mechanisms (22).

Corn plants with male-sterile 'C' and 'T' cytoplasms
and their nuclear-gene restored fertile counterparts
translate some different proteins on mitochondrial ribosomes
(31). This alteration of mitochondrial translation products

(presumably representing mtDNA transcripts) suggests nuclear

18
regulation of mitochondrial gene expression in this system.
All this experimental evidence strongly correlates mtDNA
characteristics with the male-sterile trait in corn. Due to
obligate maternal transmission of cytoplasmic male
sterility, it is still difficult to separate mitochondrial
influence from any cryptic plastid contribution to the
character, and causality is not yet proven in the
ems-mitochondria system.

The linkage of disease susceptibility to male sterility
has also been used to explore mitotic variation of corn
mitochondrial inheritance. Efforts to obtain male-sterile,
blight—resistant plants from tissue cultures selected for
toxin resistance showed that though variation could be
generated during tissue culture growth, the linkage was
difficult to break. Toxin-resistant variants were also
male-fertile (29). Toxin-sensitive lines which were fertile
could be recovered (32), but the desirable recombinant is
still elusive.

In plant populations descended from cells containing

heterologous mixtures of Nicotiana cytoplasms derived from

 

cell fusions, mitotic recombination of mtDNAs, as inferred
from restriction fragment changes, occurred concurrently
with modification of floral cms phenotype (33,34).

The phenomenon of plant cytoplasmic male sterility and
nuclear restoration of fertility offers an interesting
example of the interaction of nuclear and organellar genetic

systems. It is widely observed that a recurrent

l9
backcrossing program between different species within the
same genus uncovers incompatibility between the cytoplasmic
genome of one species and the nuclear genome of the other.
A frequent nonlethal result is cytoplasmic male sterility.
For hybrid breeding programs, nuclear genes which
specifically restore male fertility to the cytoplasm can be
obtained from the cytoplasmic parent species. Parents
contributing cytoplasmic male sterility and nuclear restorer
genes have been identified in many crops (cf. 35-39). In

Nicotiana, the nuclear restorer genes are located in the

 

nucleolus organizing regions (40). How ribosomal RNA
transcription in the nucleus could affect mitochondrial gene
expression is unknown. It would be interesting to determine
whether nuclear genes can affect mitochondrial DNA stability
and recombination in plants.

At present, cytoplasmic male sterility is the only
phenotype correlated with plant mitochondrial DNA.
Selectable biochemical markers would advance the study of

plant mtDNA genetics.
1.4. Chlorgplast genetics

1.4.1. Introduction

 

The rest of this discussion concerns chloroplasts.

Like mitochondria, chloroplasts are double-membrane-bounded

20
and can be represented genetically by many chloroplast DNAs
per plastid and many plastids per cell. Both types of
organelles demonstrate autonomous transcription and
translation. Both organelles grow by membrane accretion,
but have not yet demonstrated heritable characteristics with
regard to this phenomenon. Because there are no eukaryotes
with plastids but without mitochondria, it becomes necessary
to discriminate plastid from mitochondrial characteristics.
These distinctions depend primarily on the identity of
chloroplasts as the photosynthetic compartment, enclosing
both the light and dark primary processes of photosynthesis.
Facultative photoheterotrophy facilitates the genetic
analysis of chlorOplasts. Chloroplast-encoded characters

will be discussed in some detail later.

1.4.2. Chlamydomonas

Chlamydomonas rheinhardtii provides the most
instructive model of chloroplast genetics. 1n the early
19508 Sager chose this green alga as the simplest organism
in which to study chloroplast genetics because this
eukaryotic microorganism is unicellular, has a single
chloroplast, can be grown on defined media and handled by
bacteriological techniques, and has a simple sexual cycle
(2). Chlamydomonas is homogametic. with gametes of the same
size, and heterothallic, with two nuclear gene-determined

mating types. Following gamete fusion, the zygote undergoes

21
meiosis to produce four haploid products, or grows as a

vegetative diploid. Chlamydomonas is a facultative

 

heterotroph; acetate in the medium can substitute for
photosynthetically fixed carbon.

Selection for resistance to antibacterial antibiotics
yielded the first selectable markers with a demonstrably
cytoplasmic pattern of inheritance: maternal bias in
crosses and segregation in mitotic progeny. Some
photosynthetically impaired mutants follow the same
cytoplasmic pattern of inheritance. These therefore could
be separated from Mendelian markers of similar phenotype.
Following several such markers in crosses, evidence was
obtained for linkage and recombination between them. The

Chlamydomonas chloroplast system deserves review because it

 

includes all the necessary elements for genetic analysis:
genetic variation, formal meiotic and mitotic genetics,
active recombination, and population phenomena. A healthy
controversy persists over analysis of the system. For a
complete treatment of the topic, the reader is referred to
several contrasting reviews (2,3,41).

The chloroplast DNA (ctDNA) of Chlamydomonas is a
circular molecule of molecular weight 200 megadaltons (2).
It can be distinguished from nuclear DNA by its
characteristic density. There may be 25 to 50 of these DNAs
in a cell's single plastid (3). Plastid DNA replicates at a

different time from nuclear DNA in synchronized cultures;

22
cytoplasmic mutagenesis is most effective at the time of
ctDNA replication (42).

Mutant genetic markers include resistance to
antibiotics such as streptomycin, spectinomycin,
erythromycin and kanamycin. These variants have altered
chloroplast-encoded ribosomal components (43,44). Because
the plastid ribosomes are assembled from many different
components encoded in either plastid or nucleus, isolation
of antibiotic resistant phenotypes yields nuclear mutants as
well (45). Acetate-requiring photosynthetically deficient
mutants can also be mapped to chloroplast or nuclear genes,
although photosynthetic biochemistry has not been analyzed
genetically as often in Chlamydomonas as in the higher
plants. Chloroplast protein synthesis is necessary for
nuclear DNA replication (46).

Chlamydomonas usually exhibits uniparental inheritance

 

of ctDNA from the mating type plus (mt+) parent in sexual
crosses (2). Sager discovered that UV-irradiation of the
mt+ gametes alters the outcome of a cross to yield a
biparental pattern of inheritance. The UV effect is
photorepairable. The radiation presumably prevents the
nucleus (the larger target, and the site of the mt+ gene) of
the 'female' gamete from expressing a ctDNA-specific
endonuclease. In the normal course of events, methylation
of cytosine residues occurs in mt+ ctDNA, does not occur in
mating type minus (mt-) ctDNA, and is followed by enzymatic

destruction of mt- ctDNA in the zygote, while mt+ ctDNA is

23
protected by its pattern of methylation (47). Sager has
discovered a mutant locus, mat-l, in 'paternal' mt- gametes
which causes methylation and protection of mt- ctDNA and
subsequent biparental inheritance of ctDNA markers (47). In
the life cycle of the alga, the signal for activation of the
maternal restriction-modification system is apparently
aggregation of the gametes' flagella; protoplast fusion in

Chlamydomonas results in biparental inheritance of

 

chloroplast markers (48). When gametes fuse, their plastids
also fuse in the zygote, as seen in electron micrographs
(49). Thus the membranes of the organelles do not prevent
organelle fusion and molecular exchange between heterologous
ctDNAs in this system.

To examine the results of a ctDNA-marked cross, the
investigator must grow mitotic populations from haploid or
diploid progeny to observe segregation of ctDNA copies.
Thus, analysis of meiotic events depends on mitotic events.
The four haploid meiotic progeny of a zygote can be
separated on a petri dish, their daughter cells spread, and
so on until a mitotic pedigree has developed.

Alternatively, mass cultures of haploid or diploid
vegetative progeny can be grown up and tested for frequency
of ctDNA genotypes remaining after several generations.
These techniques give evidence for both reciprocal and
nonreciprocal recombination between parental ctDNAs. The
frequency of these events depends on the cell's

physiological state and is affected by growth conditions and

24
gamete pretreatments (50). Apparently the inheritance of
ctDNA sequences in crosses is a function of DNA copy number,
nitrogen starvation, modification-restriction enzyme status,
and segregation mode in subsequent cell divisions. The
picture is not clear enough for there to exist an undisputed
mathematical description of genetic exchange and
distribution processes.

In studies on mutagenesis and on segregation
frequencies in vegetative diploids, an attempt can be made
to correlate ctDNA copy number and c0py distribution in
mitosis with the rate of appearance of selectable markers.
These studies suggested that the chloroplast genome can
segregate without genetic exchange, because marker
appearance is delayed and rarer than target theory predicts
(51,52). A likely hypothesis proposes that the multiple
copies of ctDNA within each plastid are physically
associated in a small number of nucleoids, dependent on a
membrane attachment point. There may be little exchange
between nucleoids.

Mapping of chloroplast markers relative to each other

in Chlamydomonas uses several different techniques:

 

cosegregation, rate of segregation, and standard
recombination mapping in crosses (3,53,54). Unfortunately,
no deletion mapping is possible; all mutants retain ctDNA.
However, different authors obtain different map distances
and, rarely, different map orders for the same markers

(3,53,54). An attachment point is felt important to all

25
models of ctDNA exchange and distribution (3,53,54). Map
function has been hypothesized to correspond to a diploid
model (41), or to a complex papulational model (3). Any
realistic model must account for the number of ctDNAs per
cell and possible asymmetric distribution to daughter cells
of attachment points and nucleoids. The populational model,
as with the work on yeast mitochondria, is drawn from phage
genetics, postulating a panmictic pool and depending on
observation of progeny genotype 'output frequencies' and
inference about 'input frequencies' in the original zygote
(3).

Chlamydomonas does possess several mitochondria; its
mtDNA is present in about 40 copies, and is small as in
vertebrates, weighing 16 megadaltons (3). There has been
little success in its genetic analysis, due to obligate
aerobic requirements of the alga.

Much remains to be clarified in the Chlamydomonas

 

organelle genetic system. The regulation of
nuclear-chloroplast cooperation, the characteristics of
mitochondria, and a clear model of ctDNA exchange and
distribution remain to be established. Although the single
plastid suggested simplicity at the outset of these
investigations, this is nevertheless a complex genetic

system.

26

1.4.3. Higher plant chloroplast genetics

An even more complex organismal and cellular biology
influences the genetics of higher plant chloroplasts.
Sexual transmission of this organelle occurs in the context
of floral and seedling development; plastid numbers per cell
and their state of differentiation change with growth and
development of different plant tissues. There are numerous
examples of nuclear genetic effects on plastid inheritance.

Many plant species exhibit strict maternal inheritance
of plastids in reciprocal sexual crosses. In these species,
plastids are excluded from the generative cell of pollen,
and the plastids of the egg contribute all the chloroplast
genomes of the progeny (55). Such species have been used to
study the cell lineage of female organogenesis (56) and
nuclear gene—induced plastid mutagenesis (S7). The simplest
plastid phenotype to follow is a cell-autonomous albino; all
early work used this single character to study chloroplast
genetics.

In some plant genera, notably Pelargonium and

 

Oenothera, plastids are inherited from both parents in

 

sexual crosses. Here, sexual hybridization experiments
illustrate the quantitative effects of maternal nuclear
genotype, intracellular competition between plastids,
asymmetric cell divisions in early embryo development, and
hybrid nuclear genotype on plastid sexual inheritance. When

an egg cell is fertilized by the male gamete, any

27
genetically marked plastids contributed by the pollen parent
are introduced into the cytoplasm of the egg, which has been
conditioned by the nucleus of the maternal genotype.
Pre-existing biochemical conditions in the egg can favor
replication or survival of one chloroplast genome over the

other. Thus, in Pelargonium, female nuclear genetic

 

background can influence the apparent competitive ability of
a pollen-introduced mutant plastid genotype (58). The
absence of some expected classes of variegated progeny may
be the result of asymmetric cell divisions early in embryo
development. In angiosperms, the bulk of zygote cytoplasm
is sequestered into the suspensor cell and thus eliminated
from the developing seedling's cell lineage (49).
Consequently, a chloroplast genotype present in miniscule
proportion in the zygote has a smaller than proportionate
probability of appearing in the progeny seedlings.

In Oenothera, interspecific crosses cause albino and

 

virescent variegation in leaves of the progeny. This
photosynthetic dysfunction presumably reflects a breakdown
of cooperation between heterologous chloroplast and nuclear
genotypes; the variegation results from somatic segregation
of the two parental plastid genotypes in the hybrid.
Consistent differences in competitive ability of different
chloroplast genomes permitted their ranking on the basis of
apparent replication rates; nuclear genotype clearly
affected plastid intracellular replicative competition

(59,60).

28
Plastid fusion has not been observed in higher plants
(49). No evidence has been obtained for genetic
recombination between marked parental plastids in

Oenothera or Pelargonium sexual progeny. Thus a primary

 

genetic phenomenon of paramount importance in all other
genetic systems is striking by its absence in the higher
plant plastid. We do not know today whether plastid DNA
interaction is prevented by membrane phenomena which prevent
organelle fusion, or is simply not observed due to a lack of
appropriate genetic markers for chloroplast DNA.

The random process of somatic segregation of plastids
is modulated by the multiplicity of plastid genomes and
organelles. The rate of plastid segregation in cell
divisions depends upon the number of plastids and their mode
of distribution in mitosis, and on the number of ctDNAs per
plastid and their distribution in chloroplast budding.
Michaelis (61) derived a rate function for segregation based
on an average plastid number, but this model needs
modification as both plastid number and ctDNA multiplicity
fluctuate throughout the 100 or so cell divisions in the
plant life cycle. The situation becomes yet more complex if
a ctDNA lesion impairs molecular replication, plastid
budding, or cell division (62).

Genetic studies to date have depended on plastid
variants which survive the test of molecular, organellar and
cellular fitness, but are photoheterotrophic. Plant cells

are facultative photoheterotrophs. The mosaic organism can

29
tolerate partial loss of photosynthetic function because it
mimics normal developmental pathways, in which nongreen
tissues such as stem and petal parenchyma and roots depend
on translocated photosynthate.

States of plastid differentiation are as numerous as
tissues. Chromoplasts, amyloplasts and chloroplasts provide
striking examples of organelle differentiation. The
carotenoid biosynthetic pathway, which occurs within v/
Chromoplasts, is apparently specified completely by nuclear
genes, as is starch synthesis in amyloplasts (49). Only in
chloroplast development has extensive ctDNA participation in
organelle phenotype been demonstrated. These differentiated
plastid types, and the undifferentiated proplastid, are all
interconvertible in giyg (63). The mode of regulation of
plastid differentiation and multiplicity is unknown.

Chloroplast DNA multiplicity, organelle number and size
vary with plant cell differentiation. The DNA content of a
given plastid varies directly with its size (64), and
plastid size varies directly with nuclear ploidy (64).
Chloroplast number per cell ranges from a high of 650 in a
sugar beet cotyledon cell (64) to zero in a haploid alfalfa
stomate -- the latter is a rare case of terminal
differentiation (65). Chloroplasts range in diameter from a
single thylakoid, 0.5 micron in an apical meristem, to 5
microns in a leaf cell (66).

Additional genome substructure affects the analysis of

ctDNA inheritance. The plastid DNA is clustered in a few

30
regions of the organelle. These nucleoids number from one
to several per plastid, varying proportionally to organelle
size; they may contain more than twenty DNA molecules each.
Their distribution in organelle division could be asymmetric
and membrane-dependent (49,67). Chloroplast DNA can be
isolated associated with membrane fragments. More study is
needed on these attachment points, their mode of division
and distribution and their participation in ctDNA
replication processes. We do not presently understand the
role in plastid genetics of different multiplicities at the
different hierarchical levels of plastid genome
organization. The high number of ctDNA molecules is
believed to be a conservative force constraining sequence
divergence (49), but the fact of plastid DNA evolution shows
that some nucleotide sequence change is possible despite
these features.

Physical analysis of ctDNA in yitrg has supplied
information not available through transmission genetics
(68). CtDNA is present in higher plants as a fairly
homogeneous population of 100 megadalton circles. A single,
usually inverted, repeat sequence includes some of the ctDNA
ribosomal cistrons. Restriction site mapping yielded a
circular map, on which the few known sequences for
plastid-encoded proteins and RNAs can be located (68).
Special features of higher plant ctDNA include the presence
of about twenty ribonucleotide residues and the appearance

in native preparations of ctDNA oligomers. The latter

31
observation suggests that homologous recombination events
may occur between ctDNAs in giyg (68). Chloroplast DNA
polymerase and RNA polymerase are different from the
corresponding nuclear enzymes; they are probably
nuclear-encoded, as in other organelle systems (2). The
origins of replication of ctDNA have not been mapped.

After allowing for the ribosomal cistrons, the ctDNA
should still encode about 100 proteins. Considering this
wealth of sequence diversity, a relative paucity of genes
have been described. Most of these genes require some form
of differentiation to be expressed; most cooperate with
nuclear-encoded gene products in chloroplast function. The
known loci and their variants will be briefly summarized.

All the ribosomal RNAs and transfer RNAs needed for
message translation have been mapped. Loci have been
described accounting for the 238, 168, SS, and 4.58 rRNAs,
and a sufficient number of tRNAs to translate the normal
genetic code (68). Chloroplast rRNAs share extensive
homology with those of E. 2211 (69). Variants in the
plastid rRNAs are not known, but variants have been found in
chloroplast-encoded ribosomal proteins. A
maternally-inherited streptomycin-resistant tobacco mutant
has altered plastid ribosomal protein patterns (70). No
variants are known in plastid tRNAs; perhaps the cell could
not survive mistranslation caused by a suppressor-type

chloroplast tRNA mutation.

32

The most thoroughly studied plastid gene encodes the
large subunit of ribulose bisphosphate carboxylase/oxygenase
(RuBPCase). The holoenzyme performs the primary dark
reaction of photosynthesis, fixing gaseous C02 into
carbohydrate in the chloroplast stroma. It also initiates
photorespiratory catabolism of fixed carbon when 02 serves
as its substrate. Eight identical large subunits of the
enzyme, which include the active site, combine in plastids
with eight small subunits, the latter having regulatory
properties. Early evidence for maternal inheritance of
electrophoretic variants of the large subunit (LS) came from

interspecific hybridization in Nicotiana (71). Maternally-

 

inherited species-specific variants of the LS have indicated
the maternal parentage in ancestral hybridization of

allopolyploid species in several genera (Nicotiana, 72;

 

Triticum, 73). The sequence coding for the LS has been
cloned from ctDNA (74). It appears as a single copy on the
plastid DNA (68). Regulation of RuBPCase expression during
plastid deveIOpment and of its relative efficiency in
different genotypes has received attention because of its
importance in photosynthesis. 1n C4 plants, the different
leaf cells typical of Kranz anatomy exhibit differential
expression of the gene; bundle sheath cells do, and
mesophyll cells do not, transcribe RuBPCase LS mRNA (75),
although they contain the same ctDNAs (76).

The expression of the RuBPCase LS gene is coordinately

regulated with production of small subunits (SS) from

33
nuclear genes. The SS gene has also been cloned (77,78).
The small subunit is translated as a precursor with a
'transit sequence' of 44 hydrophobic amino acid residues at
the N-terminus needed for transport from the cytoplasm,
across both plastid membranes, and into the chloroplast
stroma (79). Some nuclear ploidy series of grasses show
altered RuBPCase enzymatic properties (80, 81); whether the
nuclear gene responsible for this effect encodes the SS or
is some other locus remains to be established.

A second ctDNA locus currently receiving attention
encodes a thylakoid lamellar protein. This gene product was
originally identified by comparing plastid translation
products of etioplasts and greening chloroplasts (82). The
protein appeared to be synthesized as a 34.5-kilodalton
precursor and posttranslationally processed to a
32-kilodalton protein within the chloroplast (83). Recent
work identified this ctDNA sequence as the gene responsible
for maternally-inherited triazine herbicide resistance in
weeds of field crops in Europe and Canada (84-87). Thus a
second chloroplast gene has been identified which
contributes to agronomic performance.

A third group of plastid genes includes polypeptides of
known function which are synthesized in isolated
chloroplasts but which have not yet been mapped to a ctDNA
sequence. Three subunits of the multimeric membrane ATPase,
coupling factor 1 (CFl) are encoded on plastid genes by the

criteria of synthesis on chloroplast ribosomes (88) and

34
maternal inheritance (92). Species-specific electrophoretic
variation in these polypeptides was correlated with
maternally-inherited resistance or sensitivity to a
CFl-binding fungal toxin, tentoxin, in Nicotiana (89-92).
This provides a third example of agricultural relevance of
plastid genes. Another plastid protein identified by
isolated organelle synthesis is the apoprotein of cytochrome
f (68).

Maternally-inherited mutants deficient in chloroplast
development have been collected by plant geneticists over
the years. It is reasonable to assume these represent ctDNA
lesions, but so far physical studies of their DNA have found
no restriction map differences, suggesting only that no
large rearrangements have occurred in the ctDNA of the
pleiotropic mutants. A primary genetic defect in their
photosynthetic biochemistry is hard to assign because
assembly of entire compartments within the plastid is
impaired. Genetic loss of any component of a structural
assembly in the chloroplast could result in apparent
deficiency of all the proteins and pigments of the complex,
both nuclear and ctDNA-encoded. Another unassigned,
apparently temperature-sensitive plastid genotype has been
briefly described (93-95).

As mentioned above, ctDNA is big enough to code for 100
proteins. At least 90 unidentified polypeptides are labeled
in isolated organelle protein synthesis, but we presently

know nothing about their functions in the plastid or their

35
respective code locations on ctDNA (68). The function of a

large proportion of ctDNA remains to be described.

1.5. Experimental introduction

 

1.5.1. Theoretical argument

Classical genetic analysis consists of genetic marker
selection, sexual hybridization and progeny analysis.
Cytoplasmic genes exhibit distinct behavior in such an
analysis, showing aberrant inheritance in the organismal
sexual cycle and marker segregation in mitotic cell
divisions. What constitutes sexual hybridization of an
organellar genome? The DNA molecules of heterologous
parental organelles must interact to produce progeny cell
lines with observably recombined sequences. The absence of
genetic recombination phenomena in higher plant chloroplasts
seriously limits growth of our knowledge of this genome.
Within a single genotype, plastid genes have been identified
by compartmentation of their transcription and translation
or by detailed characterization in gitrg of extracted
nucleic acids. Given a pair of alleles at a single ctDNA
locus, an observer could detect sexual inheritance or
mitotic segregation, but not ctDNA recombination. Some
ctDNA loci have been identified by comparing markers between
species and observing their transmission genetics; in

Oenothera, different species' plastids can be combined into

36
the same cell by sexual hybridization, but the heterologous
chloroplasts exhibit no spontaneous organellar
hybridization.

Thus there is no readily apparent technique for
combining genetic markers of interest into the same
functional plastid genome ig_yigg. If such a phenomenon
were available, one might construct a genotype with more
selectable markers in order to better examine details of
ctDNA genetic behavior. It is possible that heterologous
chloroplast fusion and DNA recombination do occur
spontaneously in gigg, but at a low frequency or within
specific small ctDNA regions. If that were the case, we
could not observe it because of the paucity of known genetic
markers in any experimental system. It might be that for
lack of a well-marked plastid genetic map, we cannot
recognize recombination between ctDNAs. The ability to
construct recombinant plastid genotypes would signify more
than advances in fine-structure mapping; it also would
permit the creation of useful ctDNA recombinants for

agricultural plant breeding.

1.5.2. Somatic hybrids

Construction of cytoplasmic genotypes in plants is most
readily approached via the route of protoplast manipulation.
Plant protoplasts, viable single cells enzymatically

stripped of cell walls, have been widely used during the

37
past decade for physiological and genetic studies. Somatic
hybrids produced by fusion of two different protoplasts or
uptake of subcellular particles by protoplasts have been the
object of intensive and competitive scientific activity.
Only outstanding examples of these techniques and
applications will be covered here, as the literature on this
topic is voluminous. For excellent reviews of methods and
results in cytoplasmic hybridization by protoplast
manipulation, the reader is referred to the references
(96-99).

Whether the parentage of the somatic hybrid combines
intergeneric or interspecific genomes, the operational
problems include selection of the hybrid against a
background of unfused and homotypic fused cells. An optimal
result would also include subsequent regeneration of the
hybrid into a sexually mature plant for further meiotic
inheritance tests and for analysis of genes which require
tissue differentiation for their expression. In 1974, there
was only one such hybrid plant reported (100). By 1981,
nearly twenty such successful hybrids had been reported
(99-103).

Intergeneric fusion-generated hybrids were laboratory
curiosities which survived through only a few cell divisions
(104-106) until the recent announcements of successful plant
regeneration from two different intergeneric somatic
hybrids. One of the these fusions combined potato and

tomato genomes (107), and the other used Arabidopsis and

38
Brassica (108). One outcome of this procedure is
illustrated by cell and plant lines containing bits of one
parental genotype introgressed into the other in the hybrid,
as in a fusion reported between carrot and parsley (109).
Inter- and intraspecific plant cell fusions, especially

in the Solanaceae, have been much more successful than

 

intergeneric experiments. The first report of a plant
somatic hybrid regenerated into a sexually competent plant
was the work of Carlson (110). In this instance, media
requirements of the somatic hybrid, less stringent than
those of its parental species, permitted selection of an

interspecific Nicotiana hybrid in vitro. The next

 

publication of successful regeneration of a plant derived
from a protoplast fusion described a photoautotrophic hybrid
within N. tabacum, selected by complementation between
nuclear albino genes in the parents (111). This technique

has been used elsewhere in the Solanaceae; nuclear albino

 

complementation permitted selection and regeneration of a

green somatic hybrid in Datura innoxia (112). By a

 

combination of these techniques, somatic hybridization of
sexually incompatible Petunia species was possible (113).
The transfer of disease resistance between sexually
incompatible Nicotiana species by protoplast fusion
illustrates the relevance of this technique for plant
breeding (114).

As success at somatic hybrid selection and regeneration

increases, analysis of these plants can become more

39‘

sophisticated. For thorough reviews of examples and
analysis, see the references (98,99). The genetic analysis
of plants arising from cell fusion can use markers from any
of the cell's genomes: nuclear, plastid or mitochondrial.
These may behave independently and require different
analytical techniques. One example of nuclear gene
inheritance in a somatic cell fusion is provided by the
albino complementation scheme mentioned above to select
somatic hybrids. This result implies additive fusion of
nuclear genomes. When this is the case, proof of meiotic
transmission of both parental genotypes from the hybrid is
desirable. A

A second trait receiving attention in somatic hybrids
is cytoplasmic male sterility (cms). The available evidence
links cms with mitochondrial DNA (section 1.3.3). Transfer
of cytoplasmic male sterility by protoplast fusion has been
reported for inter- and intraspecific hybrids in Petunia

(115,116) and in Nicotiana (117,118). In the most molecular

 

of these studies, the generation of different floral
morphologies in the somatic hybrids was correlated with the
appearance of novel, presumably recombinant mtDNA
restriction fragments (34).

Chloroplast segregation in somatic hybrids can be
followed by any of the few markers known on ctDNA (section
1.4.3): electrophoretic variants of the RuBPCase LS, ctDNA
restriction maps, tentoxin resistance, drug resistance, and

albinism. Unfortunately only the latter two markers are

40
selectable in early states of hybrid cell growth. Some
studies of somatic hybrids examined the plastid genotype in
regenerated plants. Sexual progeny of the first Nicotiana
hybrid produced by Carlson (110) were shown to have the
RuBPCase LS of only one parental chloroplast type (119).
When that work was later repeated (120), some
first-generation hybrid plants of each parental RuBPCase LS
genotype were found (121). In the historic wide
hybridization between potato and tomato by protoplast
fusion, analysis of RuBPCase LS showed some plants with
potato, and some with tomato, LS electrofocusing patterns

(107). A similar result was obtained later when Nicotiana

 

tabacum and N. rustica were the parents of a somatic hybrid;
some plants of each parental chloroplast type were recovered
(122). In another recent study, inheritance of a
streptomycin resistance marker was correlated with ctDNA

restriction patterns in plants derived from a Nicotiana

 

interspecific fusion experiment (101).

From these results, the expectation developed that
heterologous plastids in a cell fusion would always rapidly
segregate into cytoplasmically homogeneous lines; the
heterozygous cytoplasmic condition would be short-lived.
This, if true, would necessarily limit the chance of plastid
genome interaction. However, this is not always the case,

as shown below.

41

1.5.3. Outstanding published examples

 

The most complex genetic combinations of nuclear and
cytoplasmic markers in a cell fusion are possible in

Nicotiana, due both to the availability of a series of

 

cytoplasmic male-sterile alloplasmic substitution lines
(27), and to the ready regenerability from culture of this
genus. Several published studies examined all three

possible genomes in Nicotiana somatic hybrid plants.

 

In the French study already mentioned (34), Protoplasts
of a wild-type tobacco variety were fused with those of a
male-sterile N. debneyi cytoplasmic substitution line. No
selection for hybrids was imposed. The regenerated plants
were scored for morphological characters intermediate
between the parents; these alone represented the parental
nuclear contributions to the hybrid. After plants were
grown, the ctDNA restriction map of each individual was
shown to be that of one or the other parent (123). The
cytoplasmic male-sterile character segregated independently
from the ctDNA and was correlated with mtDNA (34). Their
failure to cosegregate implies a lack of linkage between cms
and ctDNA within a system without observable plastid
recombination, and lends support to the assignment of cms as
a mtDNA character.

A Soviet scientist, Yury Gleba, fused plastid-linked
albino with nuclear albino cells in two types of experiments

(124). In combinations within N. tabacum, he obtained a

42

high proportion of variegated plants. Using a male-sterile
N. debneyi cytoplasmic substitution line homozygous for a
nuclear albino marker and a N. tabacum plastid albino
genotype, Gleba recovered and maintained plastid variegation
in fusion plants (124). He correlated the color phenotype
of the leaves with the RuBPCase LS electrophoretic markers;
the variegated plants showed a combination of parental LS
genotypes. The observation of variegated plants implies
that somatic segregation of the heterologous plastids in a
cell fusion need not prevent recovery of the heterozygous
cytoplasmic state, provided that visual plastid markers
permit its identification. Furthermore, the cosegregation
of the albino character with the LS trait implies a stable
linkage between these two genes. The plastid variegation
was transmitted to sexual progeny of the somatic hybrids
(124). Unfortunately, meiotic transmission of the nuclear /'
marker was not tested and could have confused sexual progeny
test results. Finally, all the plants which reached sexual
maturity, including the variegated individuals, were
male-sterile. Again, cms apparently did not cosegregate
with plastid markers.

In the most recent reported thorough analysis of

Nicotiana somatic hybrids, both inter- and intraspecific

 

combinations could give variegating plastid markers in the
fusion plants (102,103). Plastid albino tobacco mutant
cells were fused with a nitrate reductase-deficient nuclear

tobacco mutant (102). The variegated plant recovered could

43
not be tested for any ctDNA markers other than the
green/albino allele pair. When the nitrate
reductase-deficient cells were fused with protoplasts of a
N; suaveolens cytoplasmic substitution line, some somatic
hybrid plants exhibited a mixture of parental chloroplast
markers for RuBPCase LS electrophoretic pattern and tentoxin
reaction (103). In the latter case, no visual marker
discriminated between parental plastids. Here too, it was
observed that male sterility and plastid markers do not
cosegregate in fusion products.

These experiments overcome sexual barriers to
cytoplasmic hybridization by somatic cell fusion. Other
relevant techniques include protoplast uptake of subcellular
particles, and pre-treatment of protoplast fusion partners
to selectively inactivate genomes. Protoplasts can take up
isolated nuclei (125,126), chromosomes (127), or plastids
(128,129). In none of these experiments was a transplanted
organelle shown to function in a regenerated plant.
Subprotoplasts containing subsets of the cell's several
genomes may be used for partial cell fusions in the future
(96). Because whole protoplast fusions have been more
successful than organelle uptake, techniques have emerged to
selectively inactivate parental genomes in fusions between
intact protoplasts, both in the fusion product and in the
surrounding population of unfused and self-fused parental
cells. X-irradiation pretreatment has been used to select

against one parental cell type and to uniparentally

44
eliminate nuclei in a cell fusion (117,130,131).
x-irradiation of one parental protoplast genotype was held
responsible for introgression of a few genes into the genome
of the other parent in the somatic hybrid (109). Iodoacetic
acid inactivation of one parental cell line was used to
select against that parent in a fusion experiment (131).
These techniques have proven successful in narrowing the
range of possible outcomes in protoplast fusions and
selections.

The preceding discussion of organelle genetics suggests
that the study of organelle behavior in plant somatic
hybrids could contribute experimental methods to manipulate
those genomes for agronomic benefit. That premise led to
the design of a well-marked cellular hybridization in
Nicotiana. Experiments performed according to this design
and their results are covered in the following chapters of

this thesis.

MATERIALS AND METHODS

11.1. Experimental design

 

Cell fusions within the genus Nicotiana have genetic

 

consequences for nuclear, mitochondrial and chloroplast
genomes (see section 1.5.2). To explore those consequences,
a series of protoplast fusion experiments which included
contrasting genetic markers from all three of these cell
genomes was designed. The different chloroplast genomes
paired together in these somatic hybrids were specially
chosen to detect and select for ctDNA recombination.

Genetic marker combinations which were constructed used
variants collected and kindly supplied by tobacco breeders
and cytogeneticists. These initial seed stocks were of
three types: nuclear albinos, cytoplasmic albinos, and

male-sterile Nicotiana cytoplasmic substitution lines, all

 

three within N. tabacum nuclear backgrounds. From these
stocks, cytoplasmic male-sterile, nuclear albino homozygous
genotypes were constructed by sexual crossing. Protoplasts
of these cytoplasmically-marked, nuclear albino genotypes

were fused with cytoplasmic albino N. tabacum protoplasts;

45

46
the genotypic construction of these somatic hybrids is shown
in Figure l.

Protoplast mixtures were treated to induce cell fusion,
cultured, and regenerated into plants. Green individuals
were visually selected; these were expected on the basis of
complementation between nuclear and cytoplasmic albinos in
the somatic hybrids. The cytoplasmic albino parent
contributed a nuclear gene for plastid greening to the
hybrid, while the nuclear albino contributed
cytoplasm-substituted plastids which were genetically
capable of greening in an N. tabacum nuclear background.
Both cytoplasmic and nuclear albino parental genotypes and
their homotypic fusion products could be selected against
visually.

Green and variegated plants were recovered from these
fusion preparations; they were analyzed for nuclear and
cytoplasmic genotypes. Besides the obvious use of these
genes in the selection of the somatic hybrid, each genome
was analyzed separately, as described below.

Meiotic transmission of nuclear albino alleles was
tested in sexual progeny of the fusion plants. Visual
observation also contributed information about additive
morphological traits in the somatic hybrid. Ploidy of
fusion plants gave information about the fate of the nuclei
in the original cell fusions.

Because the nuclear albino, cms parent contributed a

different species' cytoplasm from the N. tabacum,

47
cytoplasmic albino parent, both mitochondrial and plastid
genomes of the protoplast fusion parents differed in
species-specific characters. The male-sterile character
contributed to the fusion by the cytoplasmic substitution
line, nuclear albino parent was followed by observation of
the floral morphology in sexually mature fusion plants.
This marker can be taken as an indicator of mitochondrial
inheritance (see section 1.3.3).

Three plastid markers were studied: the green/albino
character, the electrophoretic focusing pattern of Fraction
1 protein, which differs between the cytoplasms used, and
tentoxin reaction, representing ctDNA-encoded Coupling
Factor 1 subunit differences (see section 1.4.3). Other
plastid markers in this system of contrasting species'
cytoplasms are the ctDNA restriction endonuclease digestion
patterns. These were beyond the scope of this investigation
and their use was deferred until after the experiments
reported in this thesis.

Recovery of green and variegated hybrid plants in a
system with greater than one ctDNA marker permits
examination of ctDNA markers for linkage. A different
protoplast fusion experiment was designed to directly select
for ctDNA recombination. In this experiment, protoplasts of
two cytoplasm-linked albinos were fused; these parents

differed as one was from a N. megalosiphon cytoplasmic

 

substitution line, and the other was in N. tabacum

cytoplasm. Both parents were in N. tabacum nuclear

48

Figure 1. Somatic hybridization plan using complementation
between nuclear and cytoplasmic albino markers to select the
somatic hybrid

 

Nuclear albino,

 

 

 

 

 

 

cytoplasm-substituted parent Cytoplasmic albino parent
NUCLEUS: N. tabacum NUCLEUS: N. tabacum
Nuclear marker: Nuclear marker:
homozygous albino ws/ws homozygous wild-type +/+
CYTOPLASM: N. megalosiphon CYTOPLASM: N. tabacum
Chloroplast markers: Chloroplast markers:
RuBPCase LS: megalosiphon RuBPCase LS: tabacum
albinism: green albinism: albino
tentoxin: sensitive tentoxin: resistant
Mitochondrial marker: Mitochondrial marker:
male fertility: sterile male fertility: fertile

Protoplast fusion product
NUCLEUS: tetraploid N. tabacum

Nuclear marker:
heterozygous ws/ws/+/+

CYTOPLASM: mixed for all markers
including albino and green

 

49

Figure 2. Somatic hybridization plan designed to select for
ctDNA recombination

 

 

 

 

 

 

Cytoplasm-substituted, N. tabacum cytoplasm,
cytoplasmic albinogparent - gytoplasmic albino parent
NUCLEUS: N. tabacum +/+ NUCLEUS: N. tabacum +/+
CYTOPLASM: N. megalosiphon CYTOPLASM: N. tabacum
Chloroplast markers: Chloroplast markers:
albinism: albino l albinism: albino 2
RuBPCase LS: N. megalosiphon RuBPCase LS: N. tabacum
tentoxin: sensitive tentoxin: resistant
Mitochondrial marker: Mitochondrial marker:
male fertility: sterile male fertility: fertile

Protoplast fusion product
NUCLEUS: tetraploid N. tabacum +/+/+/+

CYTOPLASM: mixed for all markers
including albino l and albino 2

Select for green ctDNA recombinants

50
backgrounds. Visual selection for green regenerants
attempted to employ positive selection for ctDNA
recombination. The genetic construction of this somatic
hybrid is diagrammed in Figure 2.
These experiments permit examination of the genetic

consequences of somatic hybridization in Nicotiana for

 

nuclear, plastid and mitochondrial genomes. Special
attention is paid to the inheritance of plastid characters.
This chapter describes the development of materials and
techniques to perform the somatic cell hybridizations in
Figures 1 and 2. The results of these experiments are

discussed in Chapter 111 of this thesis.

11.2. Construction of parental materials

11.2.1. Seed sources

 

P. Carlson obtained seed of cytoplasmic male-sterile
tobaccos in the varietal background cv. 'Hicks Broadleaf'
from L. Burk, USDA-ARS, Oxford, N.C. These original stocks
will be referred to as Hicks cms. The cytoplasmic

substitutions from N. megalosiphon, N. bigelovii,

 

 

N. suaveolens, and N. undulata into N. tabacum were

supplied, as well as the Nicotiana tabacum cv. 'Hicks

 

Broadleaf' (N. tabacum cytoplasm) maintainer line. These

genotypes were grown in the greenhouse and their floral

51
morphologies were confirmed to be as published (see Table
1).

Nuclear albino markers were obtained from L. Burk by
P. Carlson. Selfed seed from heterozygotes of these
genotypes were obtained: 'Sulfur', Su/+ in cv. 'John
Williams Broadleaf', 'white seedling', ws/+ in white burley
background, and 'yellow seedling', ys/+ in an undetermined
varietal background.

This author obtained cytoplasmic albino markers from
several sources. S. Wildman, UCLA, supplied a cytoplasmic
albino mutant (herein called SamV) in the background cv.
'Turkish Samsun'. D. Gerstel, NCSU, contributed Seltman's
V in the background of a coral-flowered Red Russian variety.
L. Burk supplied DPl (herein called wvc) in cv. 'Xanthi nc',
NC95V in NC 95 background, C298Vl in mixed background, and
MSBZlV. These are spontaneously-occurring mutants from
breeder's stocks and are otherwise normal, male-fertile N.
tabacum, except MSBZlV. The latter is a cytoplasmic albino
which R. Chen isolated in N. megalosiphon-cytoplasm
substituted N. tabacum cv. 'Burley 21', and is male-sterile.
All of these cytoplasmic albinos were supplied as selfed
seed from variegating (mixed wild-type and albino plastids)
plants, with the exception of MSB21V, which requires
pollination by a male-fertile Hurley 21 N. tabacum
maintainer line.

This information is summarized in Table l.

52

ocsoumxomo asocxcs
o>fimmooou Monaco:

mwauso oufin3 ca
o>wwmoomu uwoaosc

.mmoaomoum .3.bs.>o

cmmamm +\ms
.mcfiaoomm 3OHHwa.

camawm+\ms
.masaomom muons.

nmuamm +\sm

 

 

 

ucacflsoeaamm .ummflusc Immac xusm .q .usmasm.

«mumxug HQOHUn—Z

mafia Hocwmucfims

maauumuumams Ammae . .mxoam. .>o saunas» .m

muonucc owoamuom Ammav : Amundsen: .mv

muonucm ofioumsmaum Ammav : Amcmao>msm .mv

mumaucm madam lands . luaaomacamanesam .mc

mangoes eaoaauma .mmflc . laa>oamman .mc

mumanaa eaouasuaum Ammac . Iconmwmoammms .mc
oz .cuomxo .mm<

IdeD .xnnm .A ”man mxofim

onamHmommn mozmmmmmm mumoom ammm mmmaozmw

 

 

 

 

 

 

 

 

mcofiumwuomoo .moocouomou .moousOm

"mxooum

comm HmauacH .H manna

53

.cm«mmsm pom. .>o ca
ocfioao oMEmmHaouho

.csmfimm
nmfixuns. .>o CH
ocfinao ofiEmmamouho

oawuuouloama .>H~mmz
now mead nocwnucfimz

muonucw ofioumsmwum
oawuoumloans
Emmamouhu
conmwmoammms .m
.Hm hoauam. .>o cm
ocfioao Gasmoamoumo

 

mnuw> nuw3 omuofloommm
ocfioau owammamouho

.mmoz. .>o cw
cownao owfimmadoumo

.oc «nucmx. .>o ca
ocfinao oflsmmamouao

 

ZOHBmHmUme

Ahmav

Awma .mmav

Avmav

Avmav

Avmav

Acmav

 

muzmmmhmm

DmUZ
Hmumhww .D > m.flflfiunm
dQUD
GMECHMS .m >Emm
.HN %0Hhfim. o>0
.- gnvflflflu oz
3 >Hwa2
: H>mmNU
: >mmUZ
xunm .4 Aamnv o>3

 

"nonwoam ofiEmmHQoumo

 

mUMDOm Dmmm mmwfiozmw

 

.AU.ucoov 4H OHQMB

54

All plants were grown in pots in the greenhouse under
ambient illumination. Genetic stocks were maintained as
perennials and propagated by shoot cuttings or by cutting
back existing plants, or treated as annuals and reproduced
by self or cross seed when appropriate. All
cross-pollinations were made by hand. Pollinations were
easily made, especially with the cms genotypes which often
had exserted stigmas. Careful tagging identified the many
different crosses performed (Section 11.2.2). Pollen was
stored in gelatin capsules in dessicators in the freezer
when necessary. Approximately 30 days were required for
seed maturation in the greenhouse. Seeds were planted in
the laboratory under lights and selected seedlings

transplanted to the greenhouse when needed.

11.2.2. Nuclear albino plants

The Hicks cms lines and the nuclear albino stock lines
(the latter as seed from selfed heterozygotes) were planted
in the greenhouse; Su/+ individuals were identified visually
as their chlorophyll levels are lower than those of
wild-type plants (138). The following discussion of sexual
crosses follows the convention that the female parent is
written first; the symbol N denotes a self-pollination.

Five to seven wild-type individuals from the ws/+ a and
from the ys/+ I populations were grown to maturity and

allowed to self-pollinate. That selfed seed was planted in

55

soil as a progeny test to identify the heterozygous
individuals. Plants showing albino seedlings among their
progeny were identified as nuclear albino heterozygotes,
retained, and used as pollen parents during the transfer of
the nuclear albino markers into the Hicks cms lines.

Transfer of the nuclear albino markers into the Hicks
cms lines was accomplished during two sexual generations of
the tobacco genotypes used. In the first generation, the
Hicks cms lines were pollinated with pollen from the nuclear
albino heterozygotes. The resulting seed was planted and
comprised the second generation. Five to seven progeny in
the case of the recessive nuclear markers, or three plants
in the case of Su/+, were grown to maturity from each of the
eighteen combinations of Hicks cms lines and nuclear albino
markers. The nuclear albino heterozygotes identified in the
preceding generation were again used to pollinate these
plants. This progeny test was necessary to identify albino
heterozygotes among the cms plants in the case of the two
recessive markers. Heterozygous cytoplasm-substituted
plants thus identified were used to generate a quantity of
seed from the cross: Hicks cms nuclear albino heterozygote
X nuclear albino heterozygote. This crossing scheme gave
25% nuclear albino seedlings among the 100% cytoplasmic
substitution progeny. This plan was followed with all five
cytoplasmic substitution lines plus the maintainer line as
females, using the two recessive nuclear albino

heterozygotes ws/+ and ys/+ and the semidominant nuclear

.sm MoguME ocwoac ucocaaoo
(«Eon on» sufi3 no: moo .mm was at muoxngs cognac Hoodoo: m>Hmmooou 03» on» soak
hummmoooc mums mummy mammoum .mucmuom mass on mumxuofi ocwoan umoaosc mounu Han mo
mmuommuououon nufi3 can .mucounm mamsom mm mafia Monmoucwos yawn» can mocfid mac nxofim
m>flm Ham sums oosoaaom was swam 038m one .coflumuumoaaw mo monomusm now cacao one
mommuocmm ouomauououon ocfioam Hoodooc +\m3 can mammamouao conmfimoaomoa .mq man mxowm «

 

 

56

 

mammamouho conmwmoaoamfi xmv
u3\m3 man mxofim «\H m

I
_

+\m3 x flammHQODMO
conmfimoammos .zo

\\\\\+\m3 map mxofim

mamoofi>wocw +\m3 mEo mxofim

mommaucoofi umou hammoum

«+\m3 x moHMEom mm mmouo
can mammoua h I m 30mm N

+
_

+\m3 x AEmMHQouMO
conmwmoHnmoE .zv
+\+ mac mxowm

 

 

mamsowbwocw +\m3
mmfiwwucoofi umou mommoum
an agenda A u m

«c3oum comm a +\m3 H

 

 

and: EmmB szwomm m0 mmOMU quzmm 20H8€mmzm0

omega
cofiuouflumosm OMEmmHmoumo ocfioam nmmHosc msomhuoaonlmaaumumcom «cmHm,ucflmmouo .m ousmflm

57
marker Su/+ as male parents. No progeny test was needed to
identify the Su/+ heterozygotes. No double mutants were

constructed. The crossing scheme is diagrammed in Figure 3.

11.2.3. Cytoplasmic albino plants

The seed from variegating cytoplasmic albino plants, as
obtained from other researchers, was planted in the
greenhouse and variegated individuals were selected.
Construction of pure albino lines required exploitation of
the known developmental pathways of tobacco. The following
discussion employs the convention that the three ontogenetic
tissue layers in a plant apical or axillary meristem are
written, in the order describing descent from the surface to
the pith, as: L1 (epidermis), L11, L111 (pith).

Variegated plants were pruned to encourage the growth
of -WW and -WG sectors. This was accomplished by observing
the color pattern on leaves and cutting the stem above any
leaf showing a quarter of its area or more of the
appropriate pattern. GWW and WWW chimeras are completely
albino in leaf phenotype, as the epidermis in tobacco is
colorless. They can be distinguished one from the other by
occasional tiny green sectors which occur in a GWW leaf when
the epidermis supports a rare periclinal mitosis and inserts
green tissue into the L11 during development (139). -WG
chimeras are white around the periphery of the leaf and

green in the midrib region, although the pattern created by

58
the flexible leaf ontogeny is somewhat variable (140). When
a leaf showed a quarter of more of its area in either of
these chimeral patterns, the stem above it was cut off. The
axillary shoot which grew out above the subtending
conspicuously patterned leaf usually showed a chimeral
pattern approximating that of the leaf. Then further
pruning of this new mericlinal shoot could be performed to
stimulate growth of an appropriately located axillary bud,
for example one associated with an entirely chimeral leaf
arising from the chimeral sector. When such -WW or -WG
shoots were obtained they were used for further isolation of
the albino genotypes.

Seed capsules were collected from naturally
self-pollinated -W- chimeras of all five cytoplasmic
albinos, except in the case of the MSBZlV genotype, in which
case a Hurley 21 maintainer line was the pollinator. The
LII gives rise to the gametes in tobacco; this was
originally demonstrated with the wvc genotype by Burk (139).
Such seed capsules gave 100% albino progeny; any exceptions

caused the seed lot to be discarded.

11.2.4. Nuclear albino shoot cultures

 

Seed from crosses generating homozygous albino Hicks
cms lines was surface-sterilized and planted £2 vitro. The
protocol for this operation was as follows. A small

quantity of seed was wrapped in a Miracloth bag and tied

59
with a rubber band. This package was placed into a 10%
(v/v) Chlorox solution (equivalent to 0.525% hypochlorite),
containing 0.01% SLS (sodium lauryl sulfate, technical
grade), for five minutes with stirring. The package was
opened and fastened across the mouth of a flask. Several
hundred ml of 95% ethanol, followed by 500 ml of sterile
distilled water, were poured through the seeds. The seeds
were then picked off the Miracloth with a forceps and
aseptically planted on solid medium R5 (See Table 2). Seed
was germinated in the light and the homozygous albino
individuals were transferred to fresh RS plates. Thereafter
they were grown as plantlets in the light on R5 in deep
petri dishes (Falcon 1005, 2 X 10 cm) and transferred
approximately monthly as shoot cuttings.

Sterile plantlets from ws/ws homozygous nuclear albino
Hicks cms lines showed multiple, small green leaf sectors.
To study the genetic stability of the ws marker 22 giggg,
white areas and green sectors of the sterile plantlets were
excised and planted on regeneration medium (0.3/10, see
Table 2) and plantlets derived from them were examined for
pigmentation. Early regeneration stage buds were always
white. Usually all expanded leaves from such regenerants
had the same low proportion of green to white tissue as did
the original seedling-derived homozygotes. Some plants
derived from green sectors showed increased proportions of
green tissue but such plants never survived transfer to soil

culture. Predominantly albino plants derived from white

60

 

 

 

 

 

Table 2. Tissue culture media*
R5 3/.3 0.3/10
Shoot culture Callus growth Shoot regeneration
Hormones:
Indoleacetic
acid 0 3 mg/L 0.3 mg/L
(l7 uM) (1.7 uM)
Kinetin 0 0.3 mg/L 0
(1.4 uM)
2-isopentenyl
adenine 0 0 10 mg/L
(49 uM)
Vitamins:
Pyridoxine
HCl 0.5 mg/L 0 0
Nicotinic
acid 0.5 mg/L 0 0

 

All three media contained the following,
liter:

‘NH4NO3

KNO3

CaC12.2H20
MgSO4.7H20
KH2P04
MnSOq.H20
ZnSO4.7H20

H3803
KI

CuSO4.5320
CoC12.6H20
NazEDTA
FeSO4.7H20
Thiamine
Inositol
Sucrose

Agar

1650
1900
440
370
170
16.8
10.6
6.2
0.83
0.25
0.025
0.025
37.3
27.8
1.0
100

301000

9,000

pH before autoclaving 6.0

in milligrams per

 

* Modified from Murashige and Skoog (147).

61
regions of Hicks cms ws/ws plants were again subjected to
regeneration from white leaf tissue and the resulting plants
were used for subculture on medium R5 and for protoplast
experiments. Control experiments with protoplasts from this
material never produced any green plants. It is assumed
that green sectors on ws/ws plantlets represent a
transient epigenetic escape from the albino condition, and
that the white seedling homozygotes are genetically stable
in mitosis. Such ws/ws homozygotes have never been studied
through meiosis.

'Yellow seedling' homozygous individuals (Hicks cms
ys/ys) never grew beyond the cotyledon stage on R5 medium.
Callus was sometimes produced on the seedlings. It is
possible that a nutritionally supplemented medium would
support the growth of this albino genotype as sterile
plants, but this was not explored experimentally. This
marker was never used for protoplast experiments.

Su/Su homozygotes are known to produce revertant
sectors which are genetically stable. Such sectors can be
distinguished by coloration and can be produced by several
different genetic mechanisms (141). Macroscopic sectors of
dark green color on Su/Su plantlets were avoided when
subculturing or making protoplasts from shoot cultures.

All six Hicks lines including the maintainer line were
cultured as homozygous albino plantlets; all had the same
performance Np giggg with respect to plantlet growth rate

and frequency of sectoring. It was necessary to choose

62
several genotypes for protOplast experiments. Preliminary
experiments in our laboratory had shown that the

N. megalosiphon derivative, the N. bigelovii derivative, and

 

the N. undulata substitution grew more rapidly than the
others as callus on medium 3/.3 (see Table 2) and
regenerated most readily on a high-cytokinin tobacco
regeneration medium (142). In contrast, the N. suaveolens
and the N. plumbaginifolia derivatives grew poorly and
regenerated little in standard media.

A survey of the evolutionary relationships and known
cytoplasmic characters within the sixty-member genus

Nicotiana showed that N. megalosiphon and N. suaveolens

 

are within the same taxonomic section Suaveolentes (143). A
series of tentoxin reaction studies in sexual hybrids showed
that these two species probably share cytoplasmic parentage

(92). Because the N. megalosiphon cytoplasmic substitution

 

performed better than the N. suaveolens derivative in tissue
culture regeneration assays (142), the N. megalosiphon
derivative was chosen to represent this cytoplasm to
contrast with tobacco in protoplast fusion experiments.

This fusion parent pair also provides a clear parallel to
the protoplast fusions between N. tabacum and

N. megalosiphon-cytoplasm substituted cytoplasmic albinos.

 

The sensitive tentoxin reaction of N. megalosiphon and

 

N. suaveolens differs from the resistant reaction of

N. tabacum; however, Coupling Factor 1 subunits from these

three species could not be differentiated on focusing

63
electrophoretic gels (91). They do differ in their RuBPCase
LS electrophoretic pattern (91,144).

Nicotiana undulata cytoplasm, though arising in a
different taxonomic section from N. tabacum, shares the
character of tentoxin resistance with tobacco (145). This
cytoplasmic trait caused it to be dropped from the list of
protoplast fusion parents.

The N. plumbaginifolia cytoplasmic substitution derives

 

from the same taxonomic section Alatae as does the putative
cytoplasmic parent of the ancestral hybridization which led
to the present-day allopolyploid species N. tabacum

(72,92,143). The N. plumbaginifolia cytoplasm differs from

 

N. tabacum in its reaction to tentoxin; N. plumbaginifolia

 

is sensitive to the CF I-binding toxin, whereas N. tabacum
is resistant (92). However, the N. plumbaginifolia
cytoplasmic substitution line regenerated with less

efficiency than did the N. megalosiphon, N. bigelovii and

 

 

N. undulata derivatives in preliminary studies (142). Its

 

taxonomic relationship to N. tabacum and its poor tissue
culture performance led to the rejection of the

N. plumbaginifolia cytoplasm derivative for protoplast
work.

The N. bigelovii cytoplasm is distinct from N. tabacum

 

and its precursors by its position in a different taxonomic
section; chromosome pairing in F1 hybrids between section
Genuinae and section Bigeloviana is virtually nil,

suggesting a relatively distant ancestral divergence (143).

64

N. bigelovii has a sensitive reaction to tentoxin,

 

contrasting with the resistant N. tabacum response (145).
Its RuBPCase LS electrophoretic pattern is distinct from

both N. tabacum and N. megalosiphon (144). The N. bigelovii

 

cytoplasmic substitution line responded with vigorous
regeneration in standard tobacco media (142). The

N. bigelovii cytoplasmic derivative was chosen as a second

 

genetic contrast to the N. tabacum cytoplasm in planned

protoplast fusions.

11.2.5. Cytoplasmic albino shoot cultures

Seed from -W- chimeras of cytoplasmic albinos was
surface-sterilized and planted on R5 medium as described
above. Vigorous sterile cytoplasmic albino plantlets could
be obtained more rapidly with other explants, as described
below.

Pith from -WW genotypes was excised aseptically from
stem segments. The stem was surface-sterilized in 10%
Chlorox, 0.01% SLS for five minutes and immersed briefly in
95% ethanol, followed by a rinse in sterile distilled water.
The pith was dissected out with a scalpel and placed onto
tobacco regeneration medium (0.3/10, see Table 2). The
plantlets obtained were observed carefully for several
months of transfer on medium R5 to confirm their pure albino
status. Any cultures showing wild-type green, chimeral or

variegated appearance were discarded.

65

Some cytoplasmic albino cultures were started from leaf
explants. Leaves were surface-sterilized by a five-minute
dip with stirring into 10% v/v Chlorox with 0.01% w/v SLS,
followed by a brief immersion into 95% ethanol and extensive
rinsing in sterile distilled water. Leaf discs were cut
with a cork borer (0.5 cm) from -WW leaves or from the white
margin of -WG leaves, where no green L111 tissue would be
present. These albino leaf discs were planted aseptically
on regeneration medium (0.3/10, see Table 2). The first
crop of plants which are regenerated from such explants,
appearing within six weeks without intervention of extensive
callus growth, are usually derived from the LII or L111 and
do not represent epidermal derivatives (146). Such plants
were subcultured for at least six transfers on medium R5 to
confirm their cytoplasmic albino homogeneity; again, any
variegating, chimeral, or wild-type lines were discarded.

The cytoplasmic albino genotypes, when grown as
plantlets in sterile culture, developed some green
pigmentation although expanded leaves could readily be
distinguished from the wild-type. The growth rate of
cytoplasmic albino plants ipyyiggg was less than that of
wild-type, green sterile plantlets. The different
cytoplasmic albino markers exhibited different severity of
pigment reduction; in order from the most green to the most
bleached under laboratory conditions, they ranked:
Seltman's V - MSBZlV - SamV - NC95V - wvc. To confirm the

albino cytoplasmic marker in these tissue cultures as being

66
homogeneous, a number of tests were performed. Plants were
carefully observed to attempt to detect rearrangements from
a green LI; no evidence was found for wild-type epidermis,
or the chimera was discarded. Adventitious roots from
explanted shoot cuttings were subjected to callus induction
on medium 3/.3 and the callus was transferred to
regeneration medium (see Table 2). This would give plants
derived from the L111 of the original sterile albino plants.
All such plants were albino, or the stock culture was
discarded.

C298Vl was not completely albino as isolated Np yipgg.
Additionally, it appeared to carry a virus contaminant.
Samples of the sterile plants were ground in a mortar and
pestle and rub-inoculated with carborundum onto the leaves
of N. tabacum cv. 'Xanthi nc' plants. The test showed an
easily transmissible virus but gave systemic symptoms on the
host which gives local lesions with tobacco mosaic virus
(TMV). TMV therefore was ruled out. The virus was
transmissible in ordinary culture manipulations and
contaminated other albino stock cultures. C298Vl was not
used any further after this discovery and sterile lines of
it were discarded. NC95V was not used for protoplast
experiments; sufficient different cytoplasmic albino lines

were available without it.

67

11.3. Protoplast fusion
11.3.1. Protoplast isolation and culture

Preliminary experiments determined a protocol for
isolation of leaf protoplasts from sterile plantlets of
these albino tobacco genotypes. The tender shoot culture
leaves were cut sterilely into strips several millimeters
wide and incubated with slow rotation in a mixture of
commercially available cell-wall degrading enzymes. This
crude enzyme solution had first been cleared by
prefiltration or centrifugation, and sterilized by passage
through an 0.45 micron Nalgene filter. Observations showed
that a mixture of 3% w/v Cellulysin and 0.9% w/v Macerozyme
(both from Calbiochem) would release the protoplasts within
five hours. A milder treatment with 1% Cellulysin and 0.1%
Macerozyme would release the protoplasts within a 12 to 16
hour period. This latter overnight enzyme incubation was
used throughout the fusion experiments. An osmoticum of 0.6
N mannitol was required to protect the protoplasts from
plasmolysis and bursting during preparation and handling.
Buffers (0.1% w/v MES and 50 mM glycine, pH 5.7) were
withdrawn from the enzyme incubation without apparent ill
effect on the resulting protoplast preparations. Leaf
debris was removed in a filtration step using Miracloth.

Sterile albino tobacco leaf protoplasts thus prepared

were abundant and of varying size; many were large and

68
highly vacuolate, making use of a hemocytometer
impracticable. Albino cells of all sizes proved to be
relatively light in density and were impossible to sediment
with a centrifuge. In drop culture, they floated throughout
the medium. A flotation method was devised to wash and
collect these cells.

In preliminary experiments, albino protoplasts could be
collected by flotation in 25% w/v sucrose, but these cells
never grew when plated on culture medium. The iso—osmotic
density gradient described by Harms and Potrykus for
protoplast isolation (148,149) proved to be of too low
viscosity to be practical for room temperature work; the
polyethylene glycol (PEG) and dextran gradient described by
Kanai and Edwards for protoplast flotation (150) was
considered too damaging to these cells. PEG is applied
specifically as a fusogen in the genetic experiments
described in section 11.3.4. A discontinuous density
gradient of Ficoll, a sucrose polymer of low osmotic
strength, was designed to collect the protoplasts and to
separate them from the enzymes. The optimal version of this
flotation step, as used for genetic experiments, is as
follows: Protoplast media were made up to 25%, 8% and 2%
w/v Ficoll (400DL, Sigma) and filter-sterilized (0.45 micron
Nalgene). The enzyme-protoplast mixture was combined 1:1
with the 25% Ficoll solution and layered under 6 ml of 8%
and 1 ml of 2% Ficoll in a 50 m1 centrifuge tube.

Centrifugation at SOxG for 45 minutes at room temperature in

69
a swinging bucket rotor floated most of the albino
protoplasts to the 8%-2% or the 2%-air interface, where they
were collected with a wide-bore Pasteur pipette (Blood bank
dropper, A.H. Thomas). The brown color characteristic of
the crude enzyme mixture remained in the lowest layer of the
discontinuous density gradient. Protoplasts could be
collected in this manner with pOpulation densities averaging
106 cells per ml (see Figure 4).

Following the publications of Binding (151) and
Schieder (152), V47 medium was extensively tested for
culture of these albino tobacco protoplasts (see Table 3).
In a series of experiments, protoplasts were prepared and
plated on V47 salts plus varying sugars and hormones
(J media, Table 3). Growth or division of protoplasts was
seen only once in many trials with cells isolated and plated
on these media. Albino tobacco protoplasts could not be
successfully cultured on media with Binding's V47 salts.

An adaptation of Nagata and Takebe's tobacco protoplast
culture medium (153) with the complex additives of Kao and
Michayluk (154) had been devised by R. Malmberg in our
laboratory for culture of tobacco protoplasts (155). This
medium D3 (see Table 3) proved successful for culture of
albino tobacco leaf protoplasts. All genotypes showed cell
wall formation and cell divison within a week after
isolation and plating in D3 (see Figure 5). To quantify
cell density and plating effectiveness, measured aliquots of

dilutions of the protoplasts as isolated were plated on thin

70
Table 3. Protoplast culture media

V47 Salts(151) N+T Salts(153)

 

 

 

 

mg/L mg/L
NH4NO3 280 825
KNO3 1466 950
CaC12.2H20 735 220
M9804.H20 986 1233
KH2PO4 68.5 680
MnSO4.7H20 3.89 168
ZNSO4.7H20 1.44 10.6
K1 2.49 0.83
H3803 0 6.2
NaMoO4.2H20 0.097 0.25
CuSO4.5320 0.015 0.025
COC12.6H20 0.010 0.030
NazEDTA 37.3 37.3
FeSO4.7H20 27.8 27.8
J Series 23(155)

Thiamine HCl 1.0 1.0
Nicotinic acid 0.5 0
Pyridoxine HCl 0.5 0
Inositol 100 100
Hormones(153) 1.5 - 1.0 1.0

2',4'-D 3.0 3.0

NAA 1.0 1.0

6-BA
Casamino acids

vitamin-free(154)* 0 250*

Coconut water

deproteinized(154)* 0 20 ml/L*
Sugars
Sucrose varied 17 g.
Glucose varied 0
Mannitol varied 82 g.
Agar
pH before autoclaving 6.0 6.2

 

* Filter-sterilized

Figure 4. Densityigradient centrifugation to collect albino
protoplasts. ch cells in enzyme before centrifugation, at
left, were collected at the top of the gradient after
centrifugation, right.

Figure 5. Albinogprotoplasts growing in medium D3. MSBZlV
cells formed a cell wall (asymmetric cells) and divided
(two-cell figure).

 

71

 

Figure 4.

 

Figure 5.

72
layers of solid culture medium in gridded petri dishes
(Falcon 3030, 2 mm grid) and their dead, surviving, and
dividing numbers were counted over several grid squares at
200x magnification in an inverted microscope. Some
protoplasts floated in the remaining liquid medium and
counting on several focal planes was necessary. Most cell
death occurred within 24 hours, so all counts were made
after 24 hours of culture. Cell density ranged from
105 - 106 cells per ml, and plating efficiency ranged up
to 10% in these experiments. Plating at cell densities less
than 104 per ml caused culture death.

Good preparations of nuclear and cytoplasmic albino
protoplasts were prepared from sterile shoot culture leaves
using the overnight enzyme incubation period, Ficoll
flotation wash, and plating over D3 medium solidified with
0.5% agar in 60x15 mm Petri dishes. Cultures were incubated
in dim light at high population densities (105 cells per
m1). Cell divisions (Figure 5) occurred by 7 days of

culture.

11.3.2. Protoplast regeneration

 

The growth of albino protoplast cultures beyond their
first few divisions in the high-osmoticum D3 medium was
investigated in preliminary experiments. After two weeks on
D3, the soft protoplast aggregates were spooned and spread

sterilely on a series of media with varying hormone

73
supplements. A month later, these plates were subcultured
onto regeneration medium (0.3/10, Table 2) with varying
sucrose content. Results showed that exposure to the high
cytokinin hormone additions of medium 0.3/10 in the first
protoplast transfer had a strong effect in promoting early
regeneration from such cultures. Other hormone variables
had inconclusive results in these tests. Many cultures were
lost due to inviability or contamination of the protoplast
preparation. Lowered sucrose during plant regeneration
lowered the yield of regenerated colonies. Plantlets
appeared from the 75th day after protoplast isolation and
more frequently thereafter in the successful cultures.
Regenerated plants could be grown as shoot cultures on
medium RS (Table 2), but not all could be recovered.
Culture performance was extremely variable with respect to
growth and regeneration per petri dish.

These experiments showed that all albino genotypes
could be regenerated from protoplasts isolated and plated on
D3, transferred to 0.3/10 (0.5% agar), and passaged
repeatedly on medium 0.3/10 thereafter. No wild-type plants
were ever regenerated from the albino protoplast cultures in
these experiments. An estimated maximum of 5 x 103
plantlets per original plate of 5x105 protoplasts could be
recovered after 5 - 6 months. By this time the culture

would occupy about 100 petri dishes.

74

11.3.3. Protoplast fusion technigue

The high pH - high Ca+2 treatment used to induce
cell fusion in early reports (156) failed in preliminary
experiments on albino protoplasts. Fusion treatments with
polyethylene glycol (PEG) induced cell aggregation but
resulted inra high frequency of cell death. The number of
cells per aggregate varied with cell density during the PEG
treatment. A high density of 106 protoplasts per m1
caused high multiple aggregates which always died. Low
density protoplasts (103 - 104 per ml) failed to make
cell contact during PEG treatment; fusion frequency was very
low. Intermediate densities, about 105 cells per ml,
yielded dead cells, unfused cells, binary aggregates and
some high multiple-cell aggregates.

Osmotic measurements were made in an attempt to
rationalize performance of the cells in different media and
fusion treatments. Readings on a thermocouple psychrometer
(Wescor 33-T) compared sucrose and mannitol solutions, raw
and autoclaved, with PEG (Baker 4000) solutions and culture
media. Small effects due to sugar autoclaving and media
salts, each equivalent to an increment of 0.1 N manitol,
were noted, as well as information relating the osmotic
strengths of the PEG and mannitol solutions. A fusion
protocol was devised on the premise that cell fusion would
be aided by slight cell plasmolysis, followed by relief of

this osmotic stress during removal of the PEG.

75

Cells prepared in D3 (Table 3) were mixed 1:1 with 56%
w/v PEG (Baker 4000) made up in 50 mM Ca(NO3)2 and 50 mM
glycine, pH 5.7. The resulting 28% PEG - half D3 solution
caused observable two-cell aggregations and the development
of flat cell appression surfaces within five minutes.
Higher cell multiples were aggregated in longer exposure
times, depending on cell density. Increasing proportion of
cell death occurred with longer PEG exposures. The cell
preparation was diluted with D3 medium after ten minutes;
this would lower the osmotic strength of the solution. Cell
appressions were still visible after this step.

A discontinuous Ficoll gradient similar to that used
after the enzyme treatment (section 11.3.1) was devised in
order to separate the cells from the damaging PEG solution
and to collect them in sufficient density for successful
plating. As finally established, this Ficoll wash of the
protoplast fusion preparation was performed as follows.

The fusion preparation was mixed 1:1 with 25% w/v
Ficoll in D3 and overlayered with 8% w/v Ficoll in medium
D3. Gentle centrifugation at SOxG for 30 minutes floated
the protoplasts to the 8% - air interface. Protoplasts
could be recovered at a density of 105 - 106 per ml. A
maximum of 20% of the PEG-treated cells could be recovered
in this way. These cells included appressed cells in about
the same proportion as observed during the PEG treatment.

These PEG-exposed, Ficoll-floated cells could be

76
successfully cultured on D3 followed by 0.3/10, as described
for untreated albino protoplasts in section 11.3.2.

No obvious fusion figures had been observed other than
the two-cell appression type. A microscopic assay for cell
fusion was devised. One-daywold protoplast fusion cultures
were sampled, mixed on a slide with a little 0.1% acridine
orange in water, and observed with UV-fluorescence in a
high—power compound microscope. Fusion preparations were
examined after 24 hours, before any wall formation or cell
division were observed. Preparations showed about 1%
binucleate protoplasts (Figure 6). These figures were
interpreted as surviving cell fusions. The 1% fusion rate
was considered a worthwhile compromise between the
population dependence of protoplast growth and the high
proportion of cell death in more rigorous PEG treatments.
This fusion protocol was used for genetic experiments with

the albino tobacco protoplasts.

11.3.4. Fusion experiments

 

Protoplast fusion experiments combined different albino
tobacco genotypes using the methods optimized in the
preceding sections. In each experiment, two albino leaf
protoplast cultures were enzymatically isolated and
collected by flotation in a Ficoll gradient according to the
procedure outlined in section 11.3.1. The two cultures were

diluted with medium D3 to 5 X 105 viable cells per m1 as

77

 

 

 

Figure 6. UV-fluorescent micrograpg of a fusion product.
Acridine orange-stained, PEG—treated cultures 24 hours after
plating showed 1% binucleate cells; debris is visible.

 

78
suggested by a quick visual estimate of cell population
density and viability, mixed 1:1, treated with PEG, and
collected by Ficoll flotation as described in section
11.3.3. This preparation was then cultured on D3 in the
light for two weeks, and subsequently repeatedly subcultured
at monthly intervals on medium 0.3/10 in the light. This
sequence yielded regenerated plantlets (section 11.3.2).
Within each experiment, each plated one-milliliter aliquot
of PEG-treated cells was numbered. All tissues and plants
subcultured from that plate carried the same identifying
number. Control plates of unfused cultures of each albino
were also prepared and subcultured from each fusion
experiment.

Cultures derived from fusion treatments combining
nuclear and cytoplasmic albino cells, or different
cytoplasmic albinos, were grown under lights at 1500 lux.
Wild-type green callus and green regenerated plants were
visually selected during repeated passages of the fusion
cultures on medium 0.3/10 in the light. Recovered plants
were subcultured onto medium R5 (Table 2) until shoot
cultures were verified as to phenotype and until they had
rooted. The plants were then washed of agar, transplanted
into soil with a gradual acclimatization to low humidity,

and grown up in the greenhouse.

79

11.4. Analysis of somatic hybrids

 

The fusion experiments combined contrasting genetic
characters representing the nuclear, mitochondrial and
plastid genomes (section 11.1). Genetic analysis following
recovery of green plants used leaf, root and floral tissues.
The basis for selection, leaf greening ability, was the
first character examined. Leaf variegation and sectoring
were noted. Leaf material was consumed in the RuBPCase LS
electrophoretic assay discussed below. Roots were used for
chromosome number determinations. Floral development
permitted scoring of anther morphology and the collection of
genetic data through sexual crosses with the fusion plants.

Nuclear inheritance in the fusion plants was studied in
several tests. The chromosome number of the somatic hybrids
was examined in the compound light microscope using root
tips pretreated with 0.1% w/v colchicine in water for five
hours in the dark, fixed overnight in 3 parts ethanol: 1
part glacial acetic acid (v/v) in the cold, hydrolyzed in 1N
HCl at 60° C. for 30 minutes, and stained with acetocarmine,
propiocarmine, or Feulgen stain (157). Pollen mother cell
meiotic squashes were made for plants exhibiting normal
anther development. Floral buds of various sizes were
collected into 6 parts ethanol: 3 chloroform: 1 glacial
acetic acid (v/v/v), fixed overnight at room temperature,
transferred to 70% ethanol, and stained with acetoorcein

(158). Chromosome counts were made on meiotic figures under

80
the microscope and from photographs and projected slides of
such figures. For a quick test of nuclear ploidy in the
somatic hybrids, their stomatal size was compared to haploid
and diploid tobacco controls. Lower epidermal strips were
peeled from expanded leaves, mounted in water, and measured
for stomatal length and width under the microscope.

Meiotic transmission of nuclear albino markers in the
somatic hybrid was studied by crossing the fusion plants as
females to a nuclear albino heterozygote pollen parent.
Progeny seedlings were grown in the laboratory under lights
and scored for albino characters. Female fertility and seed
viability in these crosses with fusion plants contributed
information on their nuclear euploidy.

Cytoplasmic male sterility, which probably represents a
mitochondrial DNA character (section 1.3.3), was analyzed.
Fusion plant floral morphology was examined and compared
with parental phenotype (Table 1). Repeated floral flushes
were observed from shoot cuttings of the fusion plants to
attempt to detect somatic variation in floral characters.

Plastid markers included pigmentation, a leaf protein
electrophoretic phenotype, and seedling tentoxin'reaction.
Regenerated plants were observed for color, variegation, and
sectoring behavior. Determination of plastid contribution
to RuBPCase electrophoretic pattern was extensively
investigated. Isolation of the protein was attempted
according to various published procedures (159-161). Leaves

were homogenized in the cold with a blendor or mortar and

81
pestle in buffer solutions containing 0.1-0.2 M NaCl, l-S mM
EDTA, 0-10 mM MgC12, 10-50 mM Tris HCl pH 7.4, 0.2% w/v
insoluble polyvinylpolypyrrolidone (PVPP, Sigma), and 10-80
mM mercaptoethanol. 3-10 mg/ml dithiothreitol (DTT)
replaced mercaptoethanol in some of these experiments. The
homogenate was squeezed through cheesecloth and clarified by
centrifugation at 17-20 kG for up to 30 minutes. The
supernatant was passed through a Sephadex 625 column and
eluted with the same buffer minus DTT and PVPP. The column
retarded phenolics; the protein front was collected. This
eluate was concentrated under N2 in an Amicon pressure
ultrafiltration system and dialyzed against NaCl-free
buffer. Repeated trials with this method showed that
Fraction 1 protein from genotypes with substituted
cytoplasms could not be precipitated by withdrawing the salt
during dialysis of the leaf extract.

Rabbit antiserum to spinach Fraction 1 protein (the
gift of Dr. N. Tolbert) was used in attempts to collect the
various tobacco Fraction 1 proteins by immune precipitation.
As in published procedures (162,163), leaf extracts prepared
as described above were mixed in varying proportions with
antiserum, held for 30 minutes at 4° C., and in some
experiments held overnight at 4° C. These preliminary
experiments were also unsuccessful at collecting the protein

of the different Nicotiana genotypes by precipitation. A

 

published procedure for collection of spinach Fraction 1

protein by selective precipitation of leaf extracts with PEG

82
and Mg+2 ion (164) did not work for these tobacco
extracts. Salt fractionation experiments showed that
collecting the protein precipitate from the leaf extract
between 35% and 50% of saturation with ammonium sulfate,
(NH4)ZSO4 (AS) gave appreciable purification of
Fraction 1 protein, as determined by band intensification in
SDS gel electrophoresis of the redissolved isolate.

Leaf proteins collected between 35% and 50% of
saturation in AS could be resuspended in extraction buffer
to a maximum protein concentration of 10 mg/ml. The
apparent relative intensity of the pI bands of interest
varied with the preparation. Addition of Mg+2 and
HCO3' ions, the enzyme's cofactors, to retard endogenous
proteolytic attack (165) resulted in protein insolubility.
Storage of protein preparations in the freezer always caused
progressive insolubility. The carboxymethylation procedure
of Kung (166) did not yield reliable results in preliminary
experiments; problems were encountered with the generation
of extraneous bands and with dilution in the column
treatment after carboxymethylation.

Procedures for dissociating electrofocusing gels were
adapted from published sources (166) and recommendations by
coworkers (167). Ampholyte source and concentration,
buffers and running conditions were varied in attempts to
optimize band separation and focusing in the RuBPCase LS

region. Methyl red, spinach RuBPCase (Sigma) and

83
commercially available acetylated cytochromes (U.S.
Biochemicals) were used as p1 markers in some gels.

1n the final version of this assay, protein
preparations were isolated in the cold as follows. Five
grams of deveined leaf tissue were scissored, frozen and
pulverized in liquid nitrogen, thawed to 4° C., and ground
to paste in a mortar and pestle with 0.5 g PVPP and 50 mg
DTT in 10 m1 of buffer containing 0.1 M NaCl, 1 mM EDTA, and
10 mM Tris HCl, pH 8.0. This paste was squeezed through
four layers of cheesecloth and centrifuged at 17 kc for 20
minutes. The supernatant was made up to 35% AS by the
addition of saturated ammonium sulfate, pH 8.0, and allowed
to stand 30 minutes at 4° C. After centrifugation at 12 k6
for 15 minutes, the supernatant was made up to 50% AS; this
precipitate was again collected by centrifugation and was
redissolved in 0.2 ml of the extraction buffer. This crude
protein isolate was dialyzed against the extraction buffer
for one hour to reduce AS concentration, and applied to the
gel. With this procedure, four different protein samples
could be prepared in 8 hours.

Dissociating electrofocusing polyacrylamide gels were
cast of 6.5% acrylamide (36 acrylamide: 1 bis by weight)
with 8 M urea and 2% v/v Triton X-100 (reagents from
BioRad). Ampholytes totaled 5% by volume, 3:1:1 (pH
S-8:4-6.5:3-10) (Sigma or BioRad) and 1.5 mg/ml each
arginine, asparagine and lysine HCl (all from Sigma). Gels

were polymerized with Temed and (NH4)2802 (BioRad).

84
Gels were prerun at 200-300 V for an hour between the upper
(anode) buffer of 0.4% v/v triethanolamine and the lower
(cathode) buffer of 0.2% H2804. After protein loading,
gels were run overnight with cooling at voltages between
200-700 V and with current decreasing from 15 to 5 mA per
gel. Gel field exposure totalled 4000-6000 volt-hours.

Gels were washed in 20% v/v methanol, 5% v/v glacial
acetic acid to fix and to prevent stain precipitation. They
were stained for two hours in the same solution containing
0.1% w/v Coomassie brilliant blue R and 0.5% w/v
CuSO4.SH20, and destained in the same wash solution.

Gels were photographed with transillumination to heighten
contrast.

Another plastid-determined character analyzed was the
tentoxin reaction of seedlings from the somatic hybrids
(section 1.4.3). Following published techniques (92,103),
100-200 seeds of each genotype were placed in petri dishes
on filter paper moistened with 20 ug/ml tentoxin (Sigma), or
with water as a control, and germinated in the light.
Remnant seed from progeny tests of the fusion plants was
tested alongside seed from green plants with the parental
cytoplasmic genotypes as a control. The dishes were
rewatered as necessary with the apprOpriate solution. After
germination, seedlings were scored for yellow,
toxin-sensitive, or green, toxin-resistant cotyledons. Toxin

reaction was unambiguous under these conditions.

85
The final chapter of this thesis details the recovery
of somatic hybrids from protoplast fusion experiments using
the design of Figure l, the failure of the selection for
ctDNA recombinants planned in Figure 2, and the analysis of
the hybrid plants obtained for nuclear, plastid and

mitochondrial genomes.

RESULTS

111.1. Recovery of complementing fusions

Mixtures of nuclear and cytoplasmic albino protoplasts
were subjected to bulk fusion and regeneration treatments.
Cells of the N. megalosiphon cytoplasmic substitution
homozygous for the nuclear albino markers we or Su, or the
N. bigelovii cytoplasmic substitution homozygous for ws,
were fused in varying combinations with cytoplasmic albino
N. tabacum genotypes Seltman's V, wvc, or SamV. A total of
thirteen different fusion experiments were prepared and
plated. Each fusion experiment produced four to six plates
of medium D3, each with a one-milliliter aliquot of
fusion-treated cells. Each D3 plate was numbered on the day
of plating. All subcultures and plants descended from that
original plate were identified by that same number. These
cultures were passaged on medium 0.3/10 in the light. Green
plants were selected among regenerating plantlets.

Complementation between parental nuclear and
cytoplasmic albino genomes was expected to yield green
fusion—derived plants (see Figure l). The first green plant

recovered was plated during the study comparing culture

86

87
regeneration media. This first plant was potted in soil and
transferred to the greenhouse six months after the date of
that protoplast fusion experiment. A subsequent series of
fusion experiments using the newly developed regeneration
protocol produced a series of green plantlets which were
found from four months to a year after the protoplast fusion
plating date. If green plantlets were not recovered after
five months in culture, green callus sectors were visually
selected and propagated on regeneration medium. This callus
selection proceure also yielded green regenerants. All
green plantlets were grown for several passages as shoot
cultures on medium R5, permitting confirmation of their
phenotype during shoot multiplication and rooting. Green
plants grown to seven to fifteen internodes were
transplanted to soil and grown up to sexual maturity in the
greenhouse. These putative somatic hybrids have been
maintained and propagated by shoot cuttings to the present
day. The eight different numbered groups of green plants
recovered from these fusion experiments are described in
section 111.3.

Thirteen protoplast fusion experiments were plated;
seven perished due to culture contamination or inviability.
Six experiments produced growing cultures on four to five
plates of D3 per experiment. Monthly subculture of these
numbered plates on regeneration medium for up to a year led
to as many as one hundred plates of tissue bearing any one

identification number at a given time. During regeneration

88
and selection of the series of fusion experiments, as many
as a thousand petri dishes of fusion-treated materials
required monthly subculture, making this a labor-intensive
stage of the experiment.

Green plants and callus were visually selected at every
stage of regeneration and shoot culture from the fusion
experiments. A high proportion of false positive
selections either lost their green phenotype or failed to
survive culture. The visual selection system employed had
limited effectiveness. The high volume of cultures
propagated during regeneration was necessary due to
inadequate discrimination between mutant and wild-type
phenotypes during regeneration. Shoot cultures of selected
plantlets could be scored more successfully than could
regenerating buds. Expanded leaf color and growth rate of
plantlets subcultured on R5 provided the final tissue
culture selection.

Markers varied in their interference with the visual
selection. The cytoplasmic albinos all had dark green
regenerating shoot apices. This permitted no discrimination
of fusion plants from one of the parents during the early
regeneration stage. Experiments using the Su marker showed
an overall yellow-green color of callus and regenerants.
Callus selection proved helpful in fusion experiments using
the Su marker. The difficulties experienced with the visual
selection system apparently are due to the requirement for

leaf cell differentiation for phenotypic expression of the

89
hybrid genotype. This led to a delay in the exertion of
selection pressure, resulting in excessive populations. A
positive biochemical selection for a tissue culture
phenotype could perform better in this situation.

Each successful fusion experiment produced green
plants from one or two numbered cultures. Up to a dozen
green plants were recovered within any one successful
numbered culture. These plants derived from the same
initial protoplast fusion culture carried the same
identification number; they might represent either
independent fusion events, or the descendants of only one
fused cell. The final and most rigorous selection step was
selection for photoautotrophy during plant transfer to the
greenhouse. The time required from the date of plating of
the protoplast fusion preparations to recovery of sexual
offspring from the mature green plants was about 12 to 18

months.

111.2. Failure to recover selected recombinant fusions

In other fusion experiments, prot0plasts of the
cytoplasmic albino N. tabacum genotypes wvc or SamV were
mixed and PEG—treated with protoplasts of the cytoplasmic
albino, N; megalosiphon cytOplasm-substituted genotype
MSB21V. These cell fusion preparations were then subjected
to regeneration treatment under lights. It was predicted

that green plants visually selected in these cultures would

90
represent the products of genetic recombination between the
two different plastid genotypes in somatic hybrid cells
(see Figure 2).

Seven fusion experiments combined the two types of
cytoplasmic albino cells. Some of these experiments were
performed while techniques were being developed; some were
made during the series of fusion experiments already
mentioned, and some were prepared later. As found in the
other fusion experiments, over half of the cultures perished
due to inviability or contamination. Plates from the three
experiments which survived were repeatedly subcultured on
regeneration medium in the light. A prolonged attempt to
select green plants from these cultures yielded none. The
cytoplasmic albino genotypes employed were difficult to
discriminate from the wild-type during the regeneration
treatment; this added to the difficulty of selection. Green
callus selection was attempted in some of these experiments.
Selected plantlets always reverted to albino phenotype on
subculture in medium R5.

The final fusion experiment combining different
cytoplasmic albino cells was followed quantitatively to
measure these negative results. Four months after
protoplast fusion and plating on D3, 15,000 plantlets
regenerated from 0.3/10 were scored as parental cytoplasmic
albino phenotype and discarded. A small number of these
plantlets had also been grown on RS. At the fifth month of

subculture an additional 10,000 plantlets were scored

91
negative. These figures were estimated by counting the
number of plates and counting representative plates. This
particular experiment was terminated after 5 months; earlier
experiments had been carried under selective conditions for
as long as 8 months.

Although these numbers are from only one experiment,
they may be taken as roughly representative of the
performance of any one growing protoplast fusion culture
under these growth conditions. Results of these selections
suggest that the frequency of ctDNA recombination in somatic
hybrids, if it does occur at all, is too low to detect among
the 75,000 plantlets regenerated from all three viable
fusion preparations. An examination of the experiments
reported in section 111.1 shows that six viable fusion
preparations gave recovery of eight distinct putative
somatic hybrids. Taking the number of plantlets scored in
the experiment described above as representative of all such
fusion experiments, this would be equivalent to eight
somatic hybrids recovered per 150,000 plantlets examined.
The number of plantlets scored in these attempts to select
ctDNA recombinants is of roughly the same order of
magnitude. By analogy with these complementation
experiments, five or six somatic hybrids should have been
present in the cultures derived from fusion of two different
cytoplasmic albinos. CtDNA recombination cannot be frequent
in the somatic hybrid cells (as has been observed before,

see section 1.5.3). An improved selection scheme for

92
recovery of a product of ctDNA recombination in somatic
hybrids, without the examination of a greater volume of
material, will require either early selection for the
somatic hybrids, or the use of some as yet unknown causative

agent to increase the frequency of the recombination event.

111.3. Analysis of hybrid plants

 

111.3.1. Introduction

 

Table 4 introduces the hybrid plants recovered by
visual selection. These plants appear to represent a
variety of possible outcomes of the somatic hybridization
process. Green plants resulted from several fusion

experiments using the N. megalosiphon cytoplasm ws/ws

 

nuclear albino parent. A variegated plant was recovered

using the N. bigelovii cytoplasm, but not with the

 

N. megalosiphon cytoplasm parent. The fusions utilizing the
Sn marker gave two very different results. One plant seems
to have completely lost one parental nucleus and the other
cytoplasm; the other plant shows obvious sectoring for the
hybrid nuclear marker and gave some evidence for cytoplasmic
sectoring. Contrasting results were also obtained for
inheritance of the cms character in this series of
experiments; both fertile and sterile anatomies were found.
The numbered cultures yielding more than one green plant

gave an opportunity to examine these individuals for clonal

93

 

«Emwamoumo ww>oHomwo .2 .Am.

.oocofiucos omwzuonuo nomad: mamuoconm mEo .oafiumumnmamz ««

.oouomowum> .> «mawuquIona .mE u> m.cmEuHom .uaow

 

«Emmamouho conaflmoanmoa .z .sz

"anewuow>ouoo¢ «

 

muouoom on Don H¢ mxfia
Hfi wxwa
H* mwa

umoa ocean >
muouowm mood nomaom

oaauummumaom
can Naoumcm we

omum>oomu panda >

oawuoumumamm uso
hfioumcm we

Hw mwa

muouumm NQOH Ganja?

 

«comauocosm

 

 

 

o>3 + m3\m3xzv a mom ma m
. . = H . «ma . a
o>3 + m3\m3sz a aha m m
g a g b mma . m
o>3 + nm\om.zv N omH m v
o>3 + m3\m31mv N mHH v m
I I I m 00 I N
uaom + m3\m3xzv m no N a
mucuam .Idmml .02 .oz
mucoumm mo nonadz oumam .umxm canowm
«newuooaom Hmnmfi> No oouo>ooou mowuown vaunEOm .v manna

Figure 7. Somatic hybrid #1 and its parents. A shoot
culture of plant #1 is shown at top; the protoplast fusion
parents were N. megalosiphon cytoplasm-substituted ws/ws,
lower left, and plastid albino Seltman's V, lower right.

Figure 8. Somatic hyrid #3 and parental genotypes. Shoot
culture of #3 is shown at top; N. bigelovii cytoplasm-
substituted ws/ws Hicks line at lower left, and wvc plastid
albino at lower right, were protoplast fusion parents.

 

94

l3101‘siteraxms'su sum-=3

  

 

 

Figure 8.

Figure 9. Somatic hybrid #4 and parental genotypes. Two
shoot cultures of plant #4 are shown at top; it was derived
from fusion of protoplasts of N. megalosiphon cytoplasm,
Su/Su plants, lower left, and wvc plastid albino, lower
right.

 

 

Figure 10. Somatic hybrid #5 and its parents. Hybrid #5,
selected from the upper left plate, and its N. megalosiphon
cytoplasm, Su/Su (lower left) and wvc (lower right) parents.
Another regeneration plate is shown at the upper right.

 

 

95

 

Figure 9.

 

\
l
l
I
I
it-

Figure 10.

Figure 11. Selection of somatic hybrid #6. The culture
regeneration plate at upper left yielded somatic hybrid #6;
upper right plate has no hybrids. The parental N.
megalosiphon cytoplasm, ws/ws parent is at the lower right;
wvc is at the lower left.

 

Figure 12. Hybrid #1 potted in soil. The plant is
autotrophic; small sectors are visible.

96

 

 

 

Figure 11.

 

 

 

 

Figure 12.

97

 

 

Figure 13. nggtic sectorigg for the nuclear marker in a
plant of soggtic hybrid #5. Single events caused the
appearance of +/+ and Su/Su sectors on Su/Su/+/+ plants from
the somatic hybrid plant family #5.

98
origin. These developments and further analysis are

discussed below.

111.3.2. Macroscopic leaf characters

Hybrid plants were selected based on leaf color. The
final ig_yi5£g selection stage is illustrated in Figures 7
through 11 for five plants. These photographs show dark
green (or medium green, in the case of Su/+) plantlets and
regenerating callus contrasting with the background of
parental phenotype tissues in plants. The first plant
recovered to the greenhouse is shown in Figure 12 as potted
in soil. This plant #1 showed small white leaf sectors.
These sectors were excised, sterilized and explanted onto
regeneration medium in an attempt to recover the sector
genotype as plantlets for further study. In almost all
cases, green plantlets grew out of other layers of the
explanted leaf lamina. It was not possible to rescue these
sectors in this way. Cell lineage mapping was attempted by
carefully recording the position and size of white leaf
sectors on plants of the #1 family for a total of over 200
leaf nodes scored. These records show no relation of the
rather random leaf sector pattern to phyllotaxy, to floral
anatomy or to meiotic test results. The white sectors on
these plants were always relatively small and never assumed
dominance in a meristem. Similar sectoring behavior was

observed on other plants (#2,6,7) derived from the fusion

99
of the same parental genotypes, and at a lower frequency in
their sexual progeny. Sectors of different leaf textures,
some with restricted growth, were also observed. Leaf
sectoring was especially prominent on the plants of group
#5, regenerated from a fusion experiment using the
semidominant nuclear marker Su. Figure 13 illustrates a
larger than usual display of Sulfur phenotype sectoring on a
plant from group #5. Whole plant wild-type pigment
phenotypes were also recovered in this group.

In the analysis of somatic sectoring in nuclear and
cytoplasmic heterozygotes, one must consider the different
consequences of nuclear and organellar segregation events on
the mitotic pedigree displayed in the plant shoot. The
nuclear marker is theoretically present in two doses in the
tetraploid fusion plant genome (see Figure l). Segregation
for the dominant or semidominant markers could occur
following mitotic recombination between homologous or
homeologous chromosomes, or following a nondisjunction event
in mitosis. More phenotypes can be distinguished with the
semidominant marker, but in any case these events lead to
contiguous single-phenotype sectors of unique origin. Twin
spots of contrasting phenotype may be visible in the case of
the semidominant marker, but both arise from a single
genetic event. Such events can recur elsewhere in the
plant, but each occurrence yields a single-event macroscopic
clone (see Figure 13). This pattern created by nuclear

segregation contrasts visibly with the pattern generated by

100
the many segregating copies of organelle genomes. Organelle
segregation leads to a multiple-event pattern in which pure
cell lineages appear to arise Ngyggyg in many closely
related cell lineages. When revealed by a color phenotype,
this multiple-event pattern is accurately termed
variegation.

Two different plantlets showing this variegated
phenotype were recovered during ig'yiggg selection of
somatic hybrids; one was found in cultures of hybrid #3, and
the other in cultures of hybrid #5. Their appearance as
variegated selections is illustrated in Figure 14 which
shows the plantlet from culture #5. These two plantlets
were propagated as shoot cuttings. Early somatic
segregation yielded mostly albino tissue in the case of #5.
Axillary buds subtending variegated leaves of the shoot
cultures were explanted onto shoot culture medium to
propagate the variegated phenotype. This process is
demonstrated with the plantlet from hybrid #3 in Figure 15.
Of these two variegated somatic hybrid selections, only one
survived subculture and potting into soil. The variegated
plant from family #5 was lost to contamination and to
segregation before the potting stage. Figure 16 shows the
recovered variegated plant from hybrid #3 as potted in the
greenhouse; the multiple-event pattern is evident.

After a year of growth in soil, several additional
variegated shoots and one albino shoot have developed from

this plant but neither of these slower-growing phenotypes

Figure 14. Variegated plantlet from cultures of somatic
hybrid #5. This plantlet was later lost.

 

 

Figure 15. Propagation of variegated plantlet from cultures
of somatic hybrid #3. The axillary bud grew into a
variegated plant.

 

 

 

ll

Figure 14.

 

Figure 15.

101

 

 

 

Figure 16. Variegated plant of somatic hybrid #3 potted in
soil. The multiple-event plastid variegation pattern is
evident; 113 was the original protoplast plate number.

 

 

Figure 17. Somatic hybrid #1 displaying the cms floral
phenotype. The stigmatoid anther development is typical of
the N. megalosiphon Hicks cms line.

 

 

 

102

 

Figure 16.

 

 

 

 

Figure 17.

103
has yet flowered. All floral and meiotic data are from the
first, green plant grown from this culture. The variegated
and albino shoots arise from growth which mitotic records
show coming from the crown of the original selected green
plantlet. This observation makes sense considering that the
first few leaf nodes developed from the regenerating
meristem should contain residual mixed cytoplasm. Future
efforts involving hybrid plant cytoplasms might profitably
concentrate on this region of the regenerated plant anatomy.
It was also possible to propagate the variegated shoot
culture of #3 by explants of the finely variegated leaves
onto regeneration medium, but this technique yielded few

variegated plantlets.

111.3.3. Floral characters

The onset of flowering in these plants permitted
meiotic analysis for the nuclear albino marker and
observation of male and female fertility. Design of the
crosses testing nuclear marker transmission was affected by
the hybrids' floral anatomies and by their anticipated
nuclear genotypes.

The floral phenotypes of the parents and fusion plants

are as follows. Plants with the N. megalosiphon cytoplasmic

 

substitution have accessory stigma and style development
below the ovary, in a position in the floral whorl which

indicates they replace the anthers. Plants with the N.

104

bigelovii cytoplasmic substitution have petaloid floret

 

development from the same position. Environmental effects
cause an additional phenotype, variable petal dissection,

in the N. bigelovii cytoplasmic substitution line. Normal

 

fertile N. tabacum are borne freestanding on slim filaments
attached between the gynoecium and petal base. Table 5
presents the results of observations on the floral
phenotypes of the somatic hybrids recovered. All fusion
plant phenotypes fall clearly into the parental male-fertile
or male-sterile anatomical classes. Both male-sterile and
male-fertile phenotypes were recovered from fusion

experiments using the N. megalosiphon cytoplasmic parent

 

with N. tabacum. The only plant arising from a fusion

combining the N. bigelovii and N. tabacum cytoplasms, #3, is

 

male-fertile as with the N. tabacum cytoplasmic phenotype.

Plants exhibiting male-sterile floral development often
showed further development of the accessory female
structures into partially complete ovaries attached at the
base of the primary gynoecium. Variably extreme pistil
hypertrophy and splitting impeded pollination of the
male-sterile fusion plants. One plant, #6, abscised most
flowers prematurely; seedling data could not be collected
from this plant. Seed set on some fusion plants had poor
germination (see section 111.3.4).

Although anatomically male-fertile, plant #3 did not
set seed when self-pollinated. It never showed the split

corolla typical of its N. bigelovii cytoplasmic parent.

105

“mumsucm HmEuocv

acumen» .m_+
Ammonucm owoamuomv

Savanna .z muonucm Hmauoz m wfi>oaomfin .z

Amuogucm Hmauoc.
.5033 .m +
Amumsucm ofioumfimwumv I

Edomomu .m muocucm HmEhoz v conmwmoammmfi .z

 

Amumnucm assuage
savanna .m +

 

 

 

 

 

muonucm Awumnucm ofioumEmNDmv
conawmoammms .m ofloumfiofium m.h.w.m.N.H conmfimoaoums .m
ucoscmfimmm oQMuoconml w ucmam moamuocmsm HoHon
Hofiuoconoouflz DGMHQ acamcm cofimom can Emmamoumo ucwuom

 

madman panama ofiumfiom may no ommuocmm Howuoconoouws
ooummmmsm can .ommuoconm Hmuoam .mmoucoumm ONEmnHmouho .m manna

106
Plant #4 exhibited both normal anther development and
complete self-fertility. Acetocarmine staining showed that
the pollen produced on plant #3 averaged less than 25%
viable compared to tobacco controls, while plant #4 had a
relative pollen viability of 91%. Self-fertilized seed was
used for seedling assays on plant #4. All other fusion
plants displayed the stigmatoid anther cms phenotype of

their N. megalosiphon cytoplasmic parent. Figure 17 shows

 

the cms floral phenotype of a plant from somatic hybrid
family #1.

It is noteworthy that no variation in floral phenotype
was found in any plant or group of plants regenerated from
the same original protoplast fusion plate. For example, the
seven members of fusion plant group #5 had strongly similar
stigmatoid anther development. The correspondence of the
fusion plant floral anatomies with the parental cytoplasmic
phenotypes, and the evidence correlating mtDNA with the cms
character (see section 1.3.3), prompt the conclusion that
the respective parental mitochondria are represented
unaltered in fusion plants of similar phenotype. However,
this analysis assays only one putative mitochondrial trait.
Unfortunately, this is the only known whole-plant
mitochondrial character. Parental mitochondrial
constitutions in the fusion plants are assigned based on
floral phenotype in Table 5, with the caveat that analysis
of only one character cannot rule out genetic interaction

between parental mitochondria in these somatic hybrids.

107

111.3.4. Nuclear genes

The fate of the parental nuclei in the somatic hybrids
was investigated by progeny tests and ploidy determinations
on the fusion plants. Meiotic transmission of the parental
wild-type and mutant nuclear albino alleles was investigated
by pollinating the fusion plants with pollen from a diploid
tobacco plant heterozygous for the appropriate nuclear
albino allele. Seedlings germinated from at least four
successful crosses were scored for the albino character.
These crosses can be written: fusion plants #l,2,3,6,7,8,
(females) x (male) ws/+; and, plant #5 (females) x (male)
Su/+. The expected nuclear genetic constitution of the
fusion plants was a tetraploid heterozygote, receiving two
mutant alleles from the nuclear albino parent and two
wild-type alleles from the cytoplasmic parent (see Figure
1). True male fertility was not anticipated in the fusion
plants due to expected high ploidy and to the use of the cms
marker. Plant #4, which was male-fertile and self-fertile,
produced abundant viable selfed seed which was used in this
progeny test.

The expected genetic ratios resulting from this cross
can be calculated. Table 6 presents the expected percentage
of the lethal homozygous albino class, calculated assuming
maximal or minimal exchange at synapsis with euploid
disjunction in meiosis. Progeny expectations for hexaploid

and oct0ploid female plants are included in Table 6 because

108

of the possibility of regenerating plants from multiple cell
fusions or from endopolyploidized cell colonies. The effect
of possible aneuploidy in the fusion plants on these
predicted progeny percentages in unknown. Any chromosome
structural rearrangements in the fusion plants should
decrease the recovery of the lethal albino class; high
ploidy, aneuploidy, and rearrangements all should reduce
female and male fertility. Although the we marker is used
as an example in Table 6, the calculated theoretical
expectations are valid for Su or any other selectable
marker.

The results of the progeny tests are shown in Table 7.
The analysis of these data focuses on the frequency of the
homozygous lethal class of progeny. The top half of Table 7
lists the observed frequency of the albino lethal class of
progeny seedlings from test crosses made on fusions using
the recessive nuclear marker ws. The observed frequency of
this progeny class varied from 0% to 10% for different
somatic hybrids. Plants #1,2,3 and #7 clearly show meiotic
transmission from the fusion plants of nuclear albino and
wild—type alleles originating from both somatic hybrid
parental nuclei. Plant #6 was too infertile for any
conclusions about its nuclear genotype to be drawn from the
progeny tests, other than the judgment of aneuploidy.
Progeny test results for plant #8 suggest that this somatic
hybrid has lost the parental nuclear albino alleles. Based

on overlap between the predictions of Table 6 and the

109

.msooa noumufiocfi may we cofiufimom mos co mcfiocmmoo .cBonm mocan>
03D on» coozumo mHHmm anemone Hmsuoa ocfioan mo ommucouuoa Hosuoo one ea

.haco GHQmeo cm on com:
ma msooa m3 use .maco moaoouoo mammoum ofioamno «+\m3 .ucouma coHHom «

 

 

 

 

mm.m uuuuuuuuuuuuuuuuu wv.a m3 m3 m3 ms + + + + "zm
mm.m~ uuuuuuuuuuuuuuuuu wo.oN m3 m3 m3 ms + + uzG
m>.N uuuuuuuuuuuuuuuuu wo.o m3 m3 + + + + uzG
w>.oa uuuuuuuuuuuuuuuuu mm.m m3 m3 + + “zv
uo>o mcwmmouo scafixmz III um>o mcwmmouo Edaficfiz mommuocom oaoawm

summanm “mammoua Hmouma ocfioam ucooumm

 

 

4mc0wuuuooaxo ammo acomoum .o manna

110

.ovmucmouoa m on oouuo>coo can .oaowm mcwaommm condom oonQNo
wo mumswumo an no coca mo woow>fio .ooucoam can cocoons» manmmmo you
omumcfisuom mmcfiaooom mo nooaoc mmnum>m mm omumasoano ma auNkuuom maoamm ««

.3oaaom Hmcuoa .w «oHn8w> couumIonHom .GM «compo

 

 

 

 

 

 

 

unmwa .04 «comma xumo .wa «moans .3 «cmoum .w ”anewumfl>muooo omhuoconm «
whm ac.o o o o ammo o mama om
mm «m.HH we Ham omm «m m
o.m o.m mm oca mma mp hIm . :
v.m m.¢a ow moH oma mcH oIm a x
h.m N.ma mv Hod OMH mm mIm - .
v.N m.¢ ea mm HmH mv oIm : .
o.m v.m m NN ¢N h mIm . a
m.NI_ m.NH oN mm mm ov NIm : :
m.v+ H.ma an NNH «ma «m film +\sm cm
H N M m w 0» on 09
am No. o.o H omov m s x
mm m.o >.N moa mean > s s
H IOI O Nfi @ c 8
Ne m.o m.H mm shoN m a n
ma A.Hl m.h Hod moaN N a 2
mm 4.H+ ao.o~ Has Hmwa H +\m3 m3
«amuwaauuom mN Manama m 3 o ucmHm vacuum noxuufi
oaoawm «noouoom mmcfiaooom godmsm coaaom unoaosz

 

 

 

mowunaz caumEOm cw muoxnua unmaooc you mumou anemone mo muasmom .h wanna

lll
observed values of %W : 28E in Table 7, nuclear genotypes of
the somatic hybrids can be assigned. By this criterion,
hybrid plants from families #1 and #2 follow the expectation
for tetraploids, and hybrids #3 and #7 could be either 6N or
8N. Plant #8 cannot have any of the hybrid nuclear
genotypes listed in Table 6; its high fertility suggests a
diploid +/+ nuclear genotype.

The lower half or Table 7 lists progeny test results
for crosses on fusion plants made using the Sn nuclear
marker. Plant #5 represents an example of multiple
recoveries of a single somatic hybrid. Seven
photoautotrophic plants were regenerated from a single
protoplast fusion plate; they were analyzed separately in
this progeny test as part of the assessment of the
difference or similarity between these plants. Because this
marker is semidominant, intermediate classes of progeny were
counted.

The range of percentages of recovered homozygous lethal
progeny seedlings varied from 4.6% to 16.4% in plants of
group #5. Plants #5-1 and #5-2, the first two listed in the
table, had a wild-type, dark green leaf phenotype but showed
a higher average percentage of homozygous progeny than did
the other five plants which had heterozygous phenotype,
yellow-green leaves. There is overlap between the observed
%Y : 28E from the plants in family #5, but they are
significantly different by the Chi square test (df=6,

P<.005). Plants #5-2 and #5-7 follow the expectations for a

112
tetraploid segregation, while the %Y i 28E values for the
other plants in this family fall variously into the 4N, 6N,
or 8N expectation ranges, or between them.

Intermediate phenotypic progeny classes were examined
in the case of this semidominant nuclear marker. A
tetraploid female parent would be expected to produce four
classes of offspring in a ratio varying from l:5:S:1 to
3:1l:ll:3, with minimal or maximal crossing-over,
respectively, in meiosis. Gradations of leaf color were
observed in the seedling populations, but environmental
variation made scoring of the seedling phenotypes difficult.
The intermediate class seedling results in Table 7 have a
high degree of uncertainty.

The progeny test results for somatic hybrid family #5
reported in Table 7 lack the symmetry of these predicted
ratios; mutant progeny classes are underrepresented. The
proportion of homozygotes can be tested to compensate for
this asymmetry. The observed value of 31.8% exceeds the
range predicted for the tetraploid, but is closest to the
expectation calculated with maximal crossing over. Plants
in family #5 with ploidies greater than 4N explain the
failure of the tetraploid model to predict the number of
progeny classes and their ratios.

All seven plants within somatic hybrid family #5
produced some albino offspring in this test, indicating
meiotic transmission of both parental marker alleles. All

displayed a consistent cms phenotype. All showed leaf

113
sectoring and a reduced female fertility characteristic of
aneuploids. Although these plants gave different results in
this progeny test, they could represent different aneuploid
outcomes of a single nuclear fusion followed by chromosome
rearrangement, loss and endopolyploidization.

Plant #4 is unique in the fusion plant analysis. It
was regenerated from a different original plate, but from
the same experimental protoplast preparation, as plant group
#5. The parental genotypes for this fusion were these: N.
Nggglosiphon cytoplasm-substituted nuclear albino homozygote
Su/Su, and N. tabacum nuclear wild-type +/+ cytoplasmic
albino. Plant #4 differs strikingly from all the plants in
group #5. Although constructed using the semidominant
marker Su, somatic hybrid plant #4 showed wild-type, dark
green leaves without any somatic sectoring. Its unexpected
male fertility and self-fertility led to the use of selfed
seed for progeny tests. Systemic fungal contamination in
the seed capsules and germinating seed populations lowered
effective seedling yield from plant #4, but enough data were
collected to give a good female fertility index and a
significant rating of 0.0% homozygous lethal progeny for
this plant. Pollinating plant #4 with pollen from a Su/+
male parent would be more directly comparable to the other
crosses discussed here; still, the lack of any phenotypic
expression of the semidominant nuclear marker in the
regenerated plant or its selfed offspring is sufficient

evidence to conclude that the parental Su allele has been

114
completely lost from this plant's nuclear genotype.
Further, the conclusion that this plant must be a euploid,
diploid individual can be drawn from its male and female
fertility.

A rough assessment of the somatic hybrids' nuclear
euploidy can be made based on a ranking of the plant
fertility ratings of Table 7. Plants with 80% or better
relative female fertility include #4, #7 and #8. As argued
above, the complete loss of the Su marker and good male and
female fertility of plant #4 indicate a diploid, +/+
genotypic assignment for that plant. Though male-sterile,
plant #8 also did not transmit the nuclear albino marker to
its progeny. Its single albino seedling among four thousand
might have resulted from a Ngyggyg dominant mutation, or
from a paternal haploid. Plant #8 thus could also be a +/+
diploid. Plant #7, however, showed appreciable meiotic
transmission of the nuclear albino marker. With a
homozygous seedling progeny percentage of 2.7%, it could be
hexaploid or octoploid with respect to the ws locus. Its
female fertility indicates euploidy, but other tests would
be necessary to resolve its nuclear genotype.

Only one plant had intermediate female fertility in
this test. The 42% seedling yield of plant #3 is the sole
value in the 20%-70% range of female fertility in these
results. The low but significant proportion of albino

seedlings from plant #3 supports a prediction that it is

115
probably hexaploid or octoploid with respect to the ws
locus.

Low female fertilities in the range 6%-15% characterize
plants #1, #2 and #5. These three plant families all showed
a 2%-15% range of homozygous albino progeny. Plant #6 gave
such poor seedling yield that it could not be scored in any
seedling tests. The degree of female infertility in the
fusion plants may be taken as an indication of their
relative aneuploidy. Thus, plant #6 is severely aneuploid,
contrasting with plants #4, #7 and #8, which are euploid by
this criterion.

The diploid chromosome number for tobacco is 2N=48.
These somatic hybrids were expected to combine the nuclei
from two diploid parents. Chromosome number changes
following fusion could arise by incomplete addition of the
parental nuclei, or during the tissue culture phase of the
experiment. Thirty root tip mitotic chomosome counts from
cells of somatic hybrid #1 showed an average chromosome
number of 2N=83, with a range from 69 to 97. Plant #1 is
therefore subtetraploid by this assay. It was difficult
to obtain an unobstructed spread of the many small tobacco
chromosomes in these somatic hybrids and chromosomes were
often overlayered. Several counts each were made on a
minimum of five cell division figures, however all
chromosome counts reported here probably err on the low
side. Meiotic pollen mother cell squashes of plant #3

showed that at meiosis II, N averaged 62, with counts

116
ranging from 37 to 82. This average figure suggests that
plant #3 may have a ploidy between 5N=120 and 6N=l44.
Recall that segregation of the albino marker in seedlings
from this plant (Table 7) indicated a 6N or 8N segregation;
5N was not calculated. It is possible, too, for chromosome
loss to occur during aneuploid plant growth. Meiosis 11
counts from plant #4 showed an average of N=25 chromosomes.
Meiosis 1 counts of plant #4 ranged lower, averaging 40 for
the diploid value. This also serves as a reminder that
meiotic disjunction can act to regularize the apparent
ploidy of an aneuploid somatic hybrid's progeny.

When stomatal size was examined, haploid plants
averaged 25 by 20 microns in stomatal length and breath;
diploid control plants had measurements of 35-40 by 25—30
microns. Stomata from plant #7 averaged 50 by 40 microns,
significantly larger than the value obtained for the diploid
controls. Stomata from plant #3 averaged 57 by 42 microns,
suggesting an even higher ploidy for this plant. Stomata
from plants #4 and #8 averaged roughly the same measurements
as the diploid control, supporting their assignment as 2N,
+/+ nuclear genotypes.

It appears from this analysis that two nuclear
genotypic results of the cell fusion procedure can be
recovered with this system of markers: additive nuclear
outcomes of various ploidies and proportions, as in plants
#l,2,3,5,6 and 7, and wild type nuclei alone, as in plants

#4 and #8. Results discussed in this section illustrate the

117
different classes of nuclear genotypic results in the
somatic hybrids. Plants #1 and #2 represent additive
nuclear fusion to give roughly tetraploid nuclei with
genotype ws/ws/+/+. Plants #3,5 and 7 represent other
outcomes of nuclear addition with increased ploidy and
variable marker dosage. Though plant #6 was severely
infertile, it is included in this group on the basis of leaf
sectoring for the nuclear marker. Only plant #7 of this
group had good female fertility. A conclusion that nuclear
genomes partially or completely combined in the somatic
hybrid is supported by findings of somatic sectoring,
maternal transmission of the nuclear albino marker, and
raised ploidy in plants #l,2,3,5,6 and 7. In contrast,
plants #4 and #8 appear to retain only one parental nucleus;
both are approximately diploid, fully female-fertile, and
did not transmit the nuclear marker in meiotic progeny
tests. Neither showed sectoring for the marker allele.

Thus it appears that the consequence of somatic
hybridization for the nuclear genome are variable. These
results suggest that the initial event of nuclear fusion may
or may not occur, and that it can be followed by subsequent
changes in the hybrid nucleus. This analysis of the nuclear
outcome in somatic hybrids could be improved with a
different ploidy design or by the inclusion of a greater

number of markers and individuals.

118

111.3.5. Plastid characters

Following scoring of leaf and floral characters and
progeny tests, the somatic hybrids were subjected to
electrophoretic analysis and seedling assays for two
plastid-encoded chloroplast proteins, with the objective of
determining the hybrids' plastid genotypes.

Leaf protein preparations enriched for RuBPCase were
isolated from the fusion plants and subjected to
electrofocusing on dissociating gels. This procedure is
described in detail in section 11.4. These experiments
showed that crystallization of the enzyme by lowering salt
concentration did not work for the cytoplasm-substituted
genotypes; poor results were also obtained in preliminary
attempts at specific immune precipitation of the protein. A
crude differential ammonium sulfate precipitate gave better
results but problems were encountered with loss,
modification and progressive insolubilization of the
electrofocusable LS polypeptides. Gel conditions were
progressively modified to optimize the display of the LS
bands of interest.

During this stage of hybrid analysis, 8 isolation
techniques were used in a total of 24 protein isolation
experiments. The protein preparations were analyzed in 28
gel runs yielding over 40 stained polyacrylamide gels.
Isoelectric point standards (methyl red, spinach RuBPCase,

and acetylated cytochromes) proved that the p1 bands of

119
interest, the RuBPCase LS triplet, lie in the p1 region
5 - 6. Control experiments verified that the LS pattern
displayed by the Seltman's V or wvc cytoplasmic albino
genotypes in variegated or chimeral leaves is the same as
in other, green N. tabacum varieties. Other controls
confirmed the phenotypic differences between the LS triplet

patterns of the N. megalosiphon and N. bigelovii cytoplasmic

 

substitution lines and the N. tabacum cytoplasmic
genotypes.

A typical good electrophoretic gel displayed the
parental RuBPCase LS polypeptides as a triplet in the p1

region 5 - 6. The N. megalosiphon triplet is displaced

 

toward a more basic p1, and the N. bigelovii triplet is
displaced toward a more acidic p1. All genotypes have
N. tabacum nuclear genotypes, encoding the N. tabacum SS, a
doublet in the acidic p1 region 3.5 - 4.0 (see Figure 19).
With this procedure, the p1 region 3 - 6 is relatively
uncontaminated by other stainable entities. Other leaf
proteins in the crude isolate, and their degradation and
crosslinking products, stain in other regions of the gels.
Protein samples of the parental cytoplasmic genotypes were
isolated and run concurrently as standards with each
experimental comparison of fusion plant genotypes. The
relative concentration of the p1 bands of interest in
different isolates, and the appearance of shadow bands,
smearing and background staining in the gels, varied

considerably from experiment to experiment. Plant stresses

120
from insects, insecticides, temperature extremes, plant age
and aneuploidy contributed to the deterioration of
extractable phenotypic gel patterns. Associated protease
and phenolic oxidase activities hamper analysis of this
abundant plastid genotypic marker.

All electrophoretic analyses of green leaves from
fusion plants displayed parental phenotypes; no unusual
results were obtained. All had the N. tabacum SS doublet.
None showed the N. tabacum LS triplet, which was linked to
the albino plastid trait in the parental genotype. The
N. megalosiphon plastid marker was observed in LS triplet
electrofocusing patterns from plants #l,4,6,7 and #8, and
from all plants of family #5. White leaf sectors of plant
#1 also gave the N. Nggalosiphon pattern. No successful
displays of LS phenotype were obtained with samples from
plant #2 or from variegated leaf material of plant #3.
Green leaf tissue from plant #3 gave its parental

N. bigelovii LS electrophoretic pattern. Figure 18 shows a

 

typical gel. This particular gel resulted in a

N. megalosiphon LS phenotypic assignment for the plastids of
somatic hybrids #1 and #5 when extracts of these two plants
were compared with the parental genotypes. Figure 19 shows
an idealized representation of the results of all fusion
plant analyses for the RuBPCase LS electrophoretic plastid
marker. This figure combines the results of many different
gels. The results of this analysis for RuBPCase LS in the

fusion plants are included in Table 8, but related tests of

121

 

 

 

 

Figure 18. Gel isoelectric focusing assay for RuBPCase LS
plastid genotypes of hybrids #1 and #5. This example shows
the N. tabacum (T) and N. megalosiphon cytoplasm—substituted
(M) parents electrophoresed next to extracts from somatic

hybrids #1 and #5, which resemble the M parent. The A has
no meaning here.

122

.uoxume
mzo ofiummam m can scan: .mcuouuwm ma mummfioo cu mmamfiom Hmucmumm ooammcoam

}

can mums mucmam cane»: Scum mmamsmm cwououm .mm was msoaooc savanna xm
m o>nn om>Ho>cN moawuocmm Ham .ommummom mo aficoosm Hanan .mm can .uacnoom

mound .mn «mafia mxoflm cofiuouaumocm ofiammamouho wfl>oHomwn .m .m “mafia mxowm

coflusuwumosm ONEmmHmouao commfimoammos .z .2 «Hoodoo» .z .B umc0wumw>ounn¢

 

 

m h m m w m H B 2 m
mQHmmMm UHB<ZOm mmmNBOzmw Adazmmflm

 

«oawuocom cwumMHm you wanna ofiuouonmouuooam How .mH ousmwm

v

HQ

mm

ma

123
another plastid marker, tentoxin resistance, should be
discussed first.

Remnant seed from progeny tests of the somatic hybrid
plants was assayed for tentoxin reaction and compared to
parental genotypes with and without the toxin treatment.
Techniques are detailed in section 11.4. Resistance or
sensitivity of the seedlings to tentoxin is a phenotypic
marker for species-specific, ctDNA-encoded subunits of
plastid Coupling Factor I; tobacco is resistant, whereas
these related species' cytoplasms are sensitive to the
fungal toxin. All genotypes were green when germinated
without toxin, except where nuclear marker segregation gave
a mixture of mutant and wild-type phenotypes. Seed of

toxin-sensitive N. megalosiphon and N. bigelovii cytoplasmic

 

substitution lines gave yellow seedling leaves when
germinated in the toxin, while two N. tabacum varieties gave
green seedlings in the toxin treatment. Readings on some
fusion plant progeny seed lots were made more difficult by
their characteristic poor germination or by fungal
contaminants. However, a reading could be made in all cases
where seed germinated. Plant #6 gave such poor seed yield
that it could not be included in this test (Table 7). Seed
germination from plant #2 was sparse but sufficient for a
limited toxin sensitivity assay. Nuclear marker segregation
was evident in seed lots from plant #5 in the water control,
and in both control and toxin treatments of seedlings from a

self—pollinated Su/+ N. tabacum. All fusion plants tested

124
gave sensitive seedling responses to the toxin, indicating

the presence of N. megalosiphon or N. bigelovii plastids.

 

 

Loss of the N. tabacum ctDNA-encoded resistance was
correlated with loss of the linked albino plastid marker of
the parental cytoplasmic genotype.

Table 8 summarizes the results of analysis for three
plastid characters in the somatic hybrids: color, RuBPCase
LS, and tentoxin reaction. From the results abbreviated in
the table, it is evident that the parental plastid genotypes
were recovered intact, without any recombination between the
three plastid markers tested. The albino N. tabacum plastid
appeared to be lost entirely, with the single exception of
variegated segregants from plant #3; these exceptional
plants have not yet been analyzed for any other genetic
markers. Two plants each could not be scored for one
marker. The overall picture permits confident assignment of

the somatic hybrids' plastid genotypic outcomes.

111.4. Summary and conclusions

Somatic hybridization experiments combining contrasting
nuclear, plastid and mitochondrial markers yielded the array
of hybrid plant genotypic outcomes shown in Table 9.
Selection for albino complementation in these experiments
resulted in the recovery of eight somatic hybrids. From
zero to two hybrids were obtained per successful protoplast

fusion experiment. When more than one plant was regenerated

.mowmohc ommoEOm unamo on» no maomcomoa
ocu mom v manna om mmmom .ummu umcm mow sumo oc ouoomocm mocmoo one
.mm>oaoawo .m .m «cocmmmoaoaos .m .2 «Summon» .m .9 .chNumN>omno¢ «

 

 

 

125

 

 

 

 

z mbmummcom z comma m
2 m>mmmmcmm z cooma n
2 III 2 comma m
z m>wmmmcom z comma m
2 m>mummcom z comma v
m o>mummcmm m comma m
2 o>mummcom III comma N
z o>mmmmcom z comma a

"mommohm

m o>mummcom m comma mm>onamo .m

z mbmummcom z comma cocmmmoaoaoa .m

B ucommmmom a ocmnam mmmmmmw .m

”mucomom

mzmzzmmmmc gm mm 35.83. a g

mzozmw DHEm¢Am ZHNOBsz

 

«mmmxmoa omummHu mam mommowc mmmoEOm mo mammaoco mo muacmom .m manna

126

.momommoco mam ocu co MHoHom comma mm mocmummoccm Hommococmoume ocm
«mmoumomoco mmommomoaco oomcu co momma mm mocommmoccm omummam «ammo
Momoam oco auwamumom .mumou Mcoaomm co momma mm oocmummmccm mooaocz
.Ecmmomu .m.ocmoao wmmmmmm .9 «ocmoam mooaocc coucmmumochEmmaaouao
Hm>onamo .m .m “ocmoao mooaocc omucumumochEmnHaomho
cocmmmoaoaoa .m..z «oouom>omooo omm mmmwmocma ammo Houcomca one «

 

 

 

 

 

z z a m

z 2 9+2 a

z 2 9+2 G

2 2 9+2 m

e z a v

a m a+m m

z 2 9+: m

z 2 9+2 m
mammmmm emacSOm

m m m m

z z z z

a a a a
«cmmazomuoamz .mmmmmmm .mmmmmmm mmmmmmm

 

mmomoomoco Hammococooums
oco mammoam .mooHocc mow mammaocm ommoac ommoEOm mo monEcm .a manna

 

127
from the same protoplast plate, these plants all gave
similar nuclear, plastid and mitochondrial results. These
results suggest that under these experimental conditions,
only one fusion product was recovered per plate; the plants
are grouped accordingly. When different petri dishes plated
within the same protoplast fusion experiment are compared
(Table 4), the recovery of different genetic outcomes in
plants #4 and #5 contrasts with the recovery of similar
outcomes in plants #1 and #2, and in plants #6 and #7. The
eight somatic hybrid genotypes displayed in Table 9 are the
products of independent fusion events by virtue of their
separate plating at the one-celled stage. The types
recovered represent several different possible outcomes of
the cell fusion experiments combining marked nuclear,
plastid and mitochondrial genomes.

Somatic hybrids were selected by complementation
between a nuclear albino marker in one parent and a plastid
albino marker in the other cell parent to yield green
plants. The nuclear albino parent contributed
N. megalosiphon wild-type plastids and cms mitochondria, and
the plastid albino parent contributed a wild-type nucleus
and male-fertile mitochondria. Due to selection for green
regenerants, any cell hybrids retaining only an albino
parental nuclear genome, or only the albino plastid genome,
were lost. The mitochondrial marker was unselected, and
both types were recovered in all possible somatic hybrid

cell types. Within the limits imposed by the selection

128
strategy, all possible combinations of nuclear, plastid and
mitochondrial genotypes are represented in the somatic
hybrids.

A large group of hybrid plants appear to retain a mixed
nuclear genome in company with N. megalosiphon plastid and
mitochondrial markers. Plants #l,2,5,6 and 7 all fall into
this class. The green tissue of hybrid plant #3 also shows
a mixture of parental nuclear contributions, but combines
the N. tabacum male-fertile mitochondrial marker with the
other parent's N. bigelovii plastids. Plant #8 retains the

N. megalosiphon plastids and mitochondria, but appears to

 

transmit only the other parent's nuclear markers. Plant #4

shows a combination of N. megalosiphon plastid markers with

 

N. tabacum mitochondria, but only one parental nucleus.
These four hybrid groups represent all the possible
autotrophic assortments of the three genomes examined. We
must conclude that under these experimental conditions,
independent segregation of the nuclei, plastids and
mitochondria leads to recovery of all classes of somatic
segregants.

Some green genotypes might be argued to represent
escapes from the nuclear albino stocks by somatic genetic
exchange or mutation. However, the hybrid plants with
different species' plastids and mitochondria can only
represent segregants from cell fusions; in fact, such
plants cannot be constructed in any other way at present. A

reasonable conclusion is that plants with homologous

129
plastids and mitochondria obtained in these experiments
result, as do other classes of somatic hybrids, from
independent mitotic segregation of the three genomes in
somatic hybrids.

No evidence was obtained for plastid recombination,
either between selected markers, or between unselected
markers in confirmed somatic hybrids. Selection against one
plastid species mitigates against its recovery, except in
the form of variegation; this proved difficult to obtain,
and the sole example brought to greenhouse culture was not
extensively analyzed. These variegated shoots from somatic
hybrid #3 provide material for further work, as there is a
selectable plastid recombinant seedling phenotype between
the albino, tentoxin-resistant N. tabacum plastid and the

green, toxin-sensitive N. bigelovii plastid. The visual

 

marker facilitates vegetative propagation of the mixed
plastid condition, and seed from variegated shoots can be
collected when this plant flowers.

A higher cell fusion percentage and a more efficient
selection scheme might permit a quantitative analysis of
the behavior of different cytoplasmic genotypes in such
experiments; however, Table 9 contains too few plants to
draw any conclusions about the compatibility or segregation
rates of the cytoplasmic genomes involved. The analysis of
cell genome inheritance in somatic hybridization experiments
like those reported here could be improved after discovery

of more and better mtDNA markers, and by the inclusion of

130
more nuclear loci in the analysis scheme. High nuclear
ploidy in the resultant somatic hybrid genotypes could also
be avoided with a different experimental design. Where high
ploidy results, as with these experiments, one can proceed
to use a pollinator which causes maternal haploids to obtain

reduced ploidy in the next generation. Nicotiana africana

 

has been reported to produce maternal haploids on tobacco
(168).

These experiments created an assortment of plant cell
organellar mixtures by ig‘yiggg manipulation. As described
in Chapter I, we presently know little about plant
cytoplasmic genetics. Improved techniques for obtaining,
maintaining and deliberately rearranging plant cell
component mixtures will improve our ability to manipulate
plant plastid and mitochondrial genomes. Benefits of such
progress can be anticipated in both the genetic and

agronomic arenas.

LI ST OF REFERENCES

10.

11.

LIST OF REFERENCES

Ephrussi, B. et a1. 1949. Action de l'acriflavine sur
les levures.l. La mutation 'petite colonie'. Ann. Inst.
Pasteur Paris 76:351-367, cited in Sager, reference 2.

Sager, R. 1972. Cytoplasmic Genes and Organelles.
Academic Press, New York. 405 pp.

Gillham, N. 1974. Genetic analysis of the chloroplast
and mitochondrial genomes. Annual Review of Genetics
8:347-391.

Birky, C. 1975. Mitochondrial genetics in fungi and
ciliates. pp. 182-224 in Genetics and Biogenesis of
Mitochondria and Chloroplasts. C. Birkey et al, eds.
Ohio State University Press.

Borst, P. and L. Grivell. 1978. The mitochondrial
genome of yeast. Cell 15:705-723.

Meyer, J. and P. Whittaker. 1977. Respiratory
repression and stability of the mitochondrial genome.
Molec. Gen. Genet. 151:333-342.

Chua, N.-H. and G. Schmidt. 1979. Transport of
proteins into mitochondria and chloroplasts. J. Cell
Biology 81:461-483.

Blanc, H. and B. Dujon. 1980. Replicator regions of
the yeast mitochondrial DNA responsible for
suppressiveness. Proc. Nat. Acad. Sci. USA
77:3942-3946.

Birky, C. 1976. The inheritance of genes in

mitochondria and chloroplasts. Bioscience 26:26-33. V
Schatz, G. and T. Mason. 1974. The biosynthesis of
mitochondrial proteins. Annual Review of Biochemistry
43:51-87.

Thrailkill, K. and C. Birky. 1980. Intracellular /

population genetics: evidence for random drift of
mitochondrial allele frequencies in Saccharomyces
cerevisiae and Schizosaccharomyces pombe. Genetics 96:
237-262.

 

131

12.

13.

14.

15.

16.

l7.

18.

19.

20.

21.

22.

23.

24.

132

Avise, J. et a1. 1979. Mitochondrial DNA clones and
matriarchal phylogeny within and among geographic
populations of the pocket gopher, Geomys pinetis.
Proc. Nat. Acad. Sci. USA 76:6694-6698.

 

Brown, W. 1980. Polymorphism in mitochondrial DNA of
humans as revealed by restriction endonuclease
analysis. Proc. Nat. Acad. Sci. USA 77:3605-3609.

Ferris, S. et al. 1981. Extensive polymorphism in the
mitochondrial DNA of apes. Proc. Nat. Acad. Sci. USA
78:6319-6323.

Mitchell, C. and G. Attardi. 1978. Cytoplasmic
transfer of chloramphenicol resistance in a human cell
line. Somatic Cell Genetics 4:737-744.

Wilson, J. et a1. 1978. Polyethylene-glycol-mediated
cybrid formation: high-efficiency techniques and
cybrid formation without enucleation. Somatic Cell
Genetics 4:745-752.

DeFrancesco, L. et a1. 1980. Uniparental propagation
of mitochondrial DNA in mouse-human cell hybrids.
Proc. Nat. Acad. Sci. USA 77:4079-4083.

Borst, P. and L. Grivell. 1981. Small is beautiful -
portrait of a mitochondrial genome. Nature 290:443.

Anderson, S. et a1. 1981. Sequence and organization of
the human mitochondrial genome. Nature 290:457-465.

Montoya, J. et a1. 1981. Distinctive features of the
5'-terminal sequences of the human mitochondrial mRNAs.
Nature 290:465-470.

Ojala, D. et a1. 1981. tRNA punctuation model of RNA
processing in human mitochondria. Nature 290:470-474.

Levings, C. and D. Pring. 1979. Mitochondrial DNA of
higher plants and genetic engineering. pp. 205-222 in
Genetic Engineering Vol. 1, J. Setlow and A.
Hollaender, eds. Plenum, New York.

Dale, R. 1981. Sequence homology among different size
classes of plant mtDNAs. Proc. Nat. Acad. Sci. USA
78:4453-4457.

Sederoff, R. et a1. 1981. Evolution of DNA sequence
organization in mitochondrial genomes of Zea. Proc.
Nat. Acad. Sci. USA 78:5953-5957.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

133

Leaver, C. and P. Pope. 1977. Biosynthesis of plant
mitochondrial proteins. pp. 213-237 in Nucleic Acids
and Protein Synthesis in Plants, L. Bogorad and J.
Weil, eds., Plenum.

Ward, B. et a1. 1981. The mitochondrial genome is
large and variable in a family of plants
(Cucurbitaceae). Cell 25:793-803.

Gerstel, D. 1980. Cytoplasmic Male Sterility in
Nicotiana (A Review). North Carolina Agricultural V
Research Service Tech. Bull. No. 263. 31 pp.

Burton, G. and W. Hanna. 1976. Ethidium bromide
induced cytoplasmic male sterility in pearl millet.
Crop Science 16:731-732.

Gengenbach, B. et al. 1977. Inheritance of selected
pathotoxin resistance in maize plants regenerated from
cell cultures. Proc. Nat. Acad. Sci. USA
74:5113-5117.

Levings, C. et a1. 1980. Cytoplasmic reversion of
cms-S in maize: association with a transpositional
event. Science 209:1021-1023.

Forde, B. and C. Leaver. 1980. Nuclear and cytoplasmic
genes controlling synthesis of variant mitochondrial
polypeptides in male-sterile maize. Proc. Nat. Acad.
Sci. USA 77:418-422.

Brettel, T. et a1. 1980. Reversion of Texas
male-sterile cytOplasm maize in culture to give
fertile, T-toxin resistant plants. Theor. Appl. Genet.
58:55-58.

Belliard. G. et a1. 1980. Caracterisation moleculaire
des cytoplasmes de cybrides obtenus par fusion de
protoplastes de Tabac. Physiologie Vegetale 18:392
(abstract).

Belliard. G. et al. 1979. Mitochondrial recombination
in cyt0plasmic hybrids of Nicotiana tabacum by
protoplast fusion. Nature 281:401-403.

 

Ahokas, H. 1980. Cytoplasmic male sterility in barley.
Part 7. Nuclear genes for restoration. Theor. Appl.
Genet. 57:193-202.

Meyer, V. 1975. Male sterility from Gossypium
harkenessii. J. Heredity 66:23-27, cited in Lee, J.
Cotton. pp. 313-325 in Hybridization of Crop Plants,
W. Fehr and H. Hadley, eds. American Society of
Agronomy, 1980. 765 pp.

 

37.

38.

39.

4o.

41.

42.

43.

44.

45.

46.

47.

48.

134

Mukai, Y. and K. Tsunewaki. 1979. Basic studies on
hybrid wheat breeding. VIII. A new male
sterility-fertility restoration system in common wheat
utilizing the cytoplasm of Aegilops kotschyi and Ag.
variabilis. Theor. Appl. Genet. 54:153-160.

 

Allan, R. 1980. Wheat. pp. 709-731 in Hybridization of
Crop Plants. W. Fehr and H. Hadley, eds. American
Society of Agronomy, 765 pp.

Izhar, S. 1978. Cytoplasmic male sterility in petunia.
III. Genetic control of microsporogenesis and male
fertility restoration. J. Heredity 69:22-26.

Gerstel, D. et a1. 1980. Cytogenetics of flower
modification of a cytoplasmic male-sterile tobacco.
Genetics 96:223-235.

Sager, R. 1976. The circular diploid model of
chloroplast DNA in Chlamydomonas. pp. 295-303 in
Genetics and Biogenesis of Chloroplasts and
Mitochondria, Th. Bucher et al, eds., Elsevier-North
Holland.

 

Chiang, K.-S. and N. Sueoka 1967. Replication of
chloroplast DNA in Chlamydomonas rheinhardi during
vegetative cell cycle: its mode and regulation. Proc.
Nat. Acad. Sci. USA 57:1506-1513.

 

Schlanger. G. and R. Sager. 1974. Localization of five
antibiotic resistances at the subunit level in
chloroplast ribosomes of Chlamydomonas. Proc. Nat.
Acad. Sci. USA 71:1715-1719.

 

Ohta, N. et al. 1975. Identification of a chloroplast
ribosomal protein altered by a chloroplast mutation in
Chlamydomonas. J. Biol. Chem. 250:3655-3659.

Bogorad, L. et a1. 1977. The genetics of the
chloroplast ribosome in Chlamydomonas reinhardi. pp.
135-153 in Nucleic Acids and Protein Synthesis in
Plants. L. Bogorad and J. Weil, eds, Plenum.

 

Blamire, J. et a1. 1974. Regulation of nuclear DNA
replication by the chloroplast in Chlamydomonas. Proc.
Nat. Acad. Sci. USA 71:2867-2871.

 

Sager, R. et al. 1981. The mat-1 gene in Chlamydomonas
regulates DNA methylation during gametogenesis. Cell
24:41-47.

 

Adams, G. 1981. Chloroplast gene transmission in
Chlamydomonas rheinhardtii following polyethylene
glycol induced cell fusion. Genetics 97:sl
(abstract).

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

135

Kirk, J. and R. Tilney-Bassett. 1978. The Plastids.
Elsevier. 960 pp.

Sears, B. et al. 1980. The effect of gametogenesis
regimes on the chloroplast genetic system of
Chlamydomonas rheinhardtii. Genetics 96:95-114.

 

Lee, R. and G. Haughn. 1980. Induction and segregation
of chloroplast mutations in vegetative cell cultures of
Chlamydomonas rheinhardtii. Genetics 96:79-94.

 

VanWinkle-Swift, K. 1978. A model for the vegetative
segregation of multiple copies of the chloroplast genes
in Chlamydomonas. Genetics 88:8103 (abstract).

Sager, R. and Z. Ramanis. 1976. Chloroplast genetics
of Chlamydomonas. I. Allelic segregation ratios.
Genetics 83:303-321.

Sager, R. and Z. Ramanis. 1976. Chloroplast genetics
of Chlamydomonas. II. Mapping by cosegregation
frequency analysis. Genetics 83:323-340.

Possingham, J. 1980. Plastid replication and
development in the life cycle of higher plants. Ann.
Rev. Plant Physiol. 31:113-129.

Stewart, R. and L. Burk. 1970. Independence of tissues
derived from apical layers on ontogeny of the tobacco
leaf and ovary. Amer. J. Bot 57:1010-1016.

Walbot, V. and E. Coe. 1979. Nuclear gene iojap
conditions a programmed change to ribosome-less
plastids in Egg may . Proc. Nat. Acad. Sci. USA 76:
2760-2764.

Tilney-Bassett, R. 1973. The control of plastid
inheritance in Pelargonium II. Heredity 30:1-13.

V

Stubbe, W. 1964. Origin and continuity of plastids.
Cell Differentiation 2:65-81. V

Kirk, J. and R. Tilney-Bassett. 1967. The Plastids.
Freeman, London. 608 pp.

Michaelis, P. 1967. The investigation of plasmone
segregation by the pattern-analysis. Nucleus 10:
1-14.

Michaelis, P. 1967. The segregation of plastids as an
example of plasmone analysis. Nucleus 10:111-127.

Thomson, W. and J. Whatley. 1980. Development of
nongreen plastids. Ann. Rev. Plant Physiol. 31:
375-394.

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

136

Butterfass, T. 1980. The continuity of plastids and
the differentiation of plastid populations. pp. 29-43 V
in Chloroplasts, J. Reinert, ed. Springer.

Butterfass, T. 1969. Die plastidenverteilung bei der
mitose der schliesszellen von haploiden schwedenklee
(Trifolium hybridum L.). Planta 84:230-234.

Wildman, S. and C. Jope. 1977. Origin of chloroplasts
the size of single grana in apical meristems. pp.
385-401 in Plant and Cell Physiology, special edition
No. 3, Photosynthetic Organelles. S. Miyachi et al,
eds. Jap. Soc. Plant Physiol.

Herrmann, R. and J. Possingham. 1980. Plastid DNA -
the plastome. pp. 45-96 in ref. 64.

Bedbrook, J. and R. Kolodner. 1979. The structure of
chloroplast DNA. Ann. Rev. Plant Physiol. 30:593-620.

Schwarz, Z. and H. Kossel. 1980. The primary structure
of 168 rDNA from Zea mays chloroplast is homologous to
N. coli 16$ rRNA. Nature 283:739-742.

Yurina, N. et a1. 1978. An altered chloroplast
ribosomal protein in a streptomycin resistant tobacco
mutant. Theor. Appl. Genet. 52:125-128.

Kung, S.-D. 1976. Tobacco Fraction 1 protein: a
unique genetic marker. Science 191:429-434.

Gray, J. et a1. 1974. Origin of Nicotiana tabacum L.
detected by polypeptide composition of Fraction 1
protein. Nature 252:226-227.

Chen, K. et a1. 1975. Fraction 1 protein and the
origin of polyploid wheats. Science 190:1304-1306.

Gatenby, A. et a1. 1981. Expression in E. coli of
maize and wheat chloroplast genes for large subunit of
ribulose bisphosphate carboxylase. Nature
291:117-121.

 

Link, G. et a1. 1978. Differential expression of the
gene for the large subunit of ribulose bisphosphate

carboxylase in maize leaf cell types. Cell
15:725-731.

walbot, V. 1977. The dimorphic chloroplasts of the C4
plant Panicum maximum contain identical genomes. Cell
11:729-737.

Bedbrook, J. et a1. 1980. Molecular cloning and
sequencing of cDNA encoding the precursor to the small
subunit of chloroplast ribulose-1,S-bisphosphate
carboxylase. Nature 287:692-697.

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

137

Broglie, R. et a1. 1981. Cloned DNA sequences
complementary to mRNAs encoding precursors to the small
subunit of ribulose-l,5-bisphosphate carboxylase and a
chlorophyll a/b binding polypeptide. Proc. Nat. Acad.
Sci. USA 78:7304-7308.

Schmidt, G. et a1. 1979. NHz-terminal amino acid
sequences of precursor and mature forms of the
ribulose-1,5-bisphosphate carboxylase small subunit
from Chlamydomonas reinhardtii. J. Cell Biol.
83:615-622.

 

Garrett, M. 1978. Control of photorespiration at RuBP
carboxylase/oxygenase level in ryegrass cultivars.
Nature 274:913-915.

Rathnam, C. and R. Chollet. 1980. Photosynthetic and
photorespiratory carbon metabolism in mesophyll
protoplasts and chloroplasts isolated from isogenic
diploid and tetraploid cultivars of ryegrass (Lolium
perenne L.). Plant Physiol. 65:489-494.

Grebanier, A. et al. 1979. Comparison of the molecular
weights of proteins synthesized by isolated
chloroplasts with those which appear during greening in
Ngg_may . Plant Physiol. 63:436-439.

Grebanier, A. et a1. 1978. Membrane proteins
synthesized but not processed by isolated maize
chloroplasts. J. Cell Biol. 78:734-746.

Pfister, K. et al. 1981. Photoaffinity labeling of an
herbicide receptor protein in chloroplast membranes.
Proc. Nat. Acad. Sci. USA 78:981-985.

Darr, S. et a1. 1981. Uniparental inheritance of a
chloroplast photosystem II polypeptide controlling
herbicide binding. Biochim. Biophys. Acta
634:219-228.

Pfister, K. et a1. 1979. Modification of herbicide
binding to photosystem II in two biotypes of Senecio
vulgaris L. Plant Physiol. 64:995-999.

Steinback, K. et al. 1981. Identification of the
triazine receptor protein as a chloroplast gene
product. Proc. Nat. Acad. Sci. USA 78:7463-7467.

Bouthyett, P.-Y. and A. Jagendorf. 1978. The site of
synthesis of pea chloroplast coupling factor 1. Plant
and Cell Physiol. 19:1169-1174.

Steele, J. et a1. 1976. Chloroplast coupling factor 1:
a species-specific receptor for tentoxin. Proc. Nat.
Acad. Sci. USA 73:2245-2248.

90.

91.

92.

93.

94.

95.

96.

97.

98.

99.

100.

101.

138

Selman, B. and R. Durbin. 1978. Evidence for a
catalytic function of the coupling factor 1 protein
reconstituted with chloroplast thylakoid membranes.
Biochim. Biophys. Acta 502:29-37.

Kwanyuen, P. and S. Wildman. 1978. Evidence that
genetic information for chloroplast coupling factor is
shared by nuclear and chloroplast DNA. Biochim.
Biophys. Acta 502:269-275.

Burk, L. and R. Durbin. 1978. Reaction of Nicotiana
species to tentoxin. A new technique for identifying
the cytoplasmic parent. J. Hered. 69:117-120.

 

Matzinger, D. and E. Wernsman. 1973. Extranuclear
temperature-sensitivity in Nicotiana tabacum L. Proc.
Nat. Acad. Sci. USA 70:108-110.

 

Nessler, C. and E. Wernsman. 1980. Ultrastructural
observations of extranuclear temperature-sensitive
lethality in Nicotiana tabacum L. Bot. Gaz. 141:9-14.

Nessler, C. et al. 1980. Physiological observations of
extranuclear temperature-sensitive lethality in
Nicotiana tabacum L. Z. Pflanzenphysiol. 99:27-35.

Binding, H. 1977. Subprotoplasts and organelle
transplantation. pp. 789-805 in Plant Cell and Tissue
Culture, W. Sharp et al., eds. Ohio State University
Press.

Mastrangelo, I. 1979. Protoplast fusion and organelle
tranfer. pp. 65-73 in Nicotiana: Procedures for
Experimental Use. R. Durbin, ed. USDA Technical Bull.
No. 1586.

 

Bonnett, H. et a1. 1980. The analysis of organelle
behavior and genetics following protoplast fusion or
organelle incorporation. pp. 137-152 in Genetic
Improvement of Crops: Emergent Techniques. 1.
Rubenstein et al., eds. University of Minnesota Press.

Chaleff, R. 1981. Genetics of Higher Plants: v/
Applications of Cell Culture. Cambridge University
Press. 184 pp.

Chaleff, R. and P. Carlson. 1974. Somatic cell
genetics of higher plants. Ann. Rev. Genetics
8:267-278.

Menczel, L. et al. 1981. Streptomycin resistant and
sensitive somatic hybrids of Nicotiana tabacum +
Nicotiana knightiana: correlation of resistance to N.

 

tabacum plastids. Theor. Appl. Genet. 59:191-195.

139

102. Glimelius, K. and H. Bonnett. 1981. Somatic
hybridization in Nicotiana: restoration of
photoautotrophy to an albino mutant with defective
plastids. Planta 153:497-503.

 

103. Glimelius, K. et al. 1981. Somatic hybridization in
Nicotiana: segregation of organellar traits among
hybrid and cybrid plants. Planta 153:504-510.

 

104. Kao, K. et al. 1974. Plant protoplast fusion and
growth of intergeneric hybrid cells. Planta
120:215-227.

105. Gamborg, O. 1977. Somatic cell hybridization by
protoplast fusion and morphogenesis. pp. 287-301 in
Plant Tissue Culture and its Bio-technological
Application, W. Barz et al., eds. Springer.

106. Binding, H. 1976. Somatic hybridization experiments in
Solanaceous species. Molec. Gen. Genet. 144:171-175.

107. Melchers, G. et a1. 1978. Somatic hybrid plants of
potato and tomato regenerated from fused protoplasts.
Carlsberg Res. Commun. 43:203-218.

108. Gleba, Y. and F. Hoffmann. 1980. 'Arabidobrassica': a.‘//
novel plant obtained by protoplast fusion. Planta
149:112-117.

109. Dudits, D. et a1. 1980. Intergeneric gene transfer \/
mediated by plant protoplast fusion. Molec. Gen.
Genet. 179:283-288.

110. Carlson, P. 1973. The use of protoplasts for genetic
research. Proc. Nat. Acad. Sci. USA 70:598-602.

111. Melchers, G. and G. Labib. 1974. Somatic hybridization
of plants by fusion of protoplasts. 1. Selection of
light resistant hybrids of 'haploid' light sensitive
varieties of tobacco. Molec. Gen. Genet. 135:277-294.

112. Schieder, O. 1977. Hybridization experiments with
protoplasts from chlorophyll-deficient mutants of some
Solanaceous species. Planta 137:253-257.

113. Power, J. et al. 1980. Somatic hybridization of
sexually incompatible petunias: Petunia parodii,
Petunia parviflora. Theor. Appl. Genet. 57:1-4.

114. Evans, D. et a1. 1981. Disease resistance:
incorporation into sexually incompatible somatic
hybrids of the genus Nicotiana. Science 213:907-909.

115.

116.

117.

118.

119.

120.

121.

122.

123.

124.

125.

Izhar,

140

S. and Y. Tabib. 1980.
Heteroplasmic state in somatic

Petunia.

Part 2.

57:241-245.

Somatic hybridization in

hybrids followed by cytoplasmic segregation into male

sterile and male fertile lines. Theor. Appl. Genet.

Bergounioux-Bunniset, C. and C. Perennes. 1980.

Transfert de facteurs cytoplasmiques de la fertilite
male entre 2 lignees de Petunia hybrida par fusion de
protoplastes.

Zelcer, A. et al. 1978.

Plant Sci.

 

Lett. 19:143-149.

Interspecific transfer of

cytoplasmic male sterility by fusion between
protoplasts of normal Nicotiana sylvestris and X-ray

irradiated protoplasts of male-sterile N. tabacum.

 

Pflanzenphysiol. 90:397-407.

Belliard, G.

et a1. 1977.

Z.

Cytoplasmic hybridization in

Nicotiana by protoplast fusion and transfer of

 

cytoplasmic male sterility and fertility factors.

PP-

237-242 in Interspecific Hybridization in Plant
Breeding, Proc. 8th Congr. EUCARPIA, E. Sanchez-Mange

and F. Garcia-Olmedo, eds. Univ.

Kung,

S. D.

et a1. 1975.

genus Nicotiana.

Smith,

Chen,

 

H.

K. et a1. 1977.

et a1. 1976.
protoplast fusion in Nicotiana.
extension.

Politec.

Madrid.

Polypeptide composition of
Fraction 1 protein from parasexual hybrid plants in the

Science 187:353-355.

 

J. Heredity 67:123-128.

74:5109-5112.

Douglas, G. et a1. 1981.
Nicotiana rustica and N. tabacum.

Proc.

 

Nat. Acad. Sci.

IV.

Interspecific hybridization by
Confirmation and

Chloroplast DNA distribution in
parasexual hybrids as shown by polypeptide composition v/‘
of Fraction 1 protein.

USA

Somatic hybridization between

Analysis of

nuclear and chloroplast genome expression in somatic

hybrids.

Belliard, G. and G.
cytoplasmique par fusion de protOplastes chez Nicotiana

tabacum.

Gleba,

Y. 1979.
plants as studied by somatic cell hybridization.

Can.

J.

Bot. 59:1509-1513.

Pelletier.

1978.

Hybridization

 

Physiol. Veget. 16:441-448.

Nonchromosomal inheritance in higher

PP-

775-788 in Plant Cell and Tissue Culture, Principles
W. Sharp et al., eds. Ohio State

and Applications.

University Press.

Potrykus,

nuclei into protoplasts of higher plants.

I. and F. Hoffman.

Pflanzenphysiol.

69:287-289.

1973.

Transplantation of
Z.

\/i

/

126.

127.

128.

129.

130.

131.

132.

133.

134.

135.

136.

137.
138.

139.

141

Lorz, H. and I. Potrykus. 1978. Investigations on the
transfer of isolated nuclei into plant protoplasts.
Theor. Appl. Genet. 53:251-256.

Griesbach, R. et a1. 1982.
chromosomes by tobacco protoplasts.
press.

The uptake of isolated lily
J. Heredity, in

Potrykus, I. 1973. Transplantation of chloroplasts
into protoplasts of Petunia. Z. Pflanzenphysiol.
70:364-366.

Landgren, C. and H. Bonnett. 1979. The culture of
albino tobacco protoplasts treated with polyethylene
glycol to induce chloroplast incorporation. Plant Sci.
Lett. 16:23-30.

Aviv, D. et a1. 1980. Progeny analysis of the
interspecific somatic hybrids: Nicotiana tabacum (cms)
+ Nicotiana sylvestris with respect to nuclear and
chloroplast markers. Theor. Appl. Genet. 56:145-150.

 

Sidorov, V. et a1. 1981. Chloroplast transfer in
Nicotiana based on metabolic complementation between
irradiated and iodoacetate treated protoplasts. Planta
152:341-345.

Burk, L. 1960.
interspecific tobacco hybrids.

Male-sterile flower anomalies in
J. Heredity 51:27-31.

1964.
Tobacco Sci.

A dominant aurea
8:101-104.

and H. Menser.
in tobacco.

Burk, L.
mutation

Burk, L. Personal communication.
Wildman, S. et al. 1973. Maternal inheritance,
cytology and macromolecular composition of defective
chloroplasts in a variegated mutant of Nicotiana
tabacum. Planta 113:293-312.

 

Wong-Staal, F. and S. Wildman. 1973. Identification of
a mutation in chloroplast DNA correlated with formation
of defective chloroplasts in a variegated mutant of
Nicotiana tabacum. Planta 113:313-326.

Gerstel, D. Personal communication.
Koivuniemi, P. et al. 1981. Characterization of the
thylakoid membranes of the tobacco aurea mutant Su/su

and of three green revertant plants. Planta
151:40-47.

Burk, L. et al. 1964. Histogenesis and genetics of a
plastid-controlled chlorophyll variegation in tobacco.
Amer. Jour. Botany 51:713-724.

140.

141.

142.

143 .

144.
145.

146.
147.

148.

149.

150.

151.

152.

153.

tabacum.

142

Stewart, R. and H. Dermen. 1975. Flexibility in
ontogeny as shown by the contribution of the shoot
apical layers to leaves of periclinal chimeras. Amer.
Jour. Botany 62:935-947.

Koivuniemi, P. 1980. Isolation and characterization of
revertants of a photosynthetic mutant of Nicotiana
Ph.D. thesis, Michigan State University, E.
Lansing, USA.

 

Floyd, 3., J. Roberts, M. Peterson and P. Carlson.
Unpublished results.

The Genus Nicotiana. Chronica
536 pp.

Goodspeed, T. 1954.
Botanica, Waltham, Mass.
Chen, K. Personal communication.

Durbin, R. and T. Uchytil. 1977.
insensitivity to tentoxin. Phytopath.

A survey of plant
67:602-603.
Wang, S. Unpublished results.
Murashige, T. and F. Skoog. 1962. A revised medium for

rapid growth and bio assays with tobacco tissue
cultures. Physiol. Plant. 15:473-496.

Harms, C. and I. Potrykus. 1978. Fractionation of
plant protoplast types by iso-osmotic density gradient
centrifugation. Theor. Appl. Genet. 53:57-63.

Harms, C. and I. Potrykus. 1978. Enrichment for
heterokaryocytes by the use of iso-osmotic density
gradients after plant protoplast fusion. Theor. Appl.
Genet. 53:49-55.

Kanai, R. and G. Edwards. 1973. Purification of
enzymatically isolated mesophyll protoplasts from C3,
C4, and crassulacean acid metabolism plants using an
aqueous dextran-polyethylene glycol two-phase system.
Plant Physiol. 52:484-490.

Binding, H. 1974. Regeneration von haploiden und
diploiden pflanzen aus protoplasten von Petunia hybrida
L. Z. Pflanzenphysiol. 74:327-356.

 

Schieder, O. 1977. Attempts in regeneration of

mesophyll protoplasts of haploid and diploid wild type
lines, and those of chlorophyll-deficient strains from
different Solanaceae. Z. Pflanzenphysiol. 84:275-281.

Nagata, T. and I. Takebe. 1971. Plating of isolated
tobacco mesophyll protoplasts on agar medium. Planta
99:12-20.

154.

155.

156.

157.

158.

159.

160.

161.

162.

163.

164.

165.

143

Kao, K. and M. Michayluk. 1975. Nutritional
requirements for growth of Vicia hajastana cells and
protoplasts at a very low papulation density in liquid
media. Planta 126:105-110.

Meeks, J. et a1. 1978. Uptake of auxotrophic cells of
a heterocyst-forming cyanobacterium by tobacco
protoplasts, and the fate of their association. Planta
139:55-60.

Constabel, F. 1978. Development of protoplast fusion
products, heterokaryocytes, and hybrid cells. pp.
141-149 in Frontiers of Plant Tissue Culture 1978, T.
Thorpe, ed. University of Calgary, Canada.

Berlyn, G. and J. Miksche. 1976. Botanical
Microtechnique and Cytochemistry. Iowa State
University Press, Ames.

Collins, G. 1979. Cytogenetic techniques. pp. 17-22
in Nicotiana, Procedures for Experimental Use. R.
Durbin, ed. USDA Tech. Bull. 1586. USDA-SEA.

Chan, P. et al. 1972. Crystalline Fraction 1 protein:
preparation in large yield. Science 176:1145-1146.

Sakano, K. et a1. 1973. Absence of carbohydrate in
crystalline Fraction 1 protein isolated from tobacco
leaves. Biochim. Biophys. Acta 329:339-341.

Koivuniemi, P. et al. 1980. Ribulose-l,S-bisphosphate
carboxylase/oxygenase and polyphenol oxidase in the
tobacco mutant Su/su and three green revertant plants.
Plant Physiol. 65:828-833.

Gray, J. and R. Kekwick. 1974. An immunological
investigation of the structure and function of ribulose
1,5-bisphosphate carboxylase. Eur. J. Biochem.
44:481-489.

Uchimiya, H. et al. 1979. A micro electrofocusing
method for determining the large and small subunit
polypeptide composition of Fraction 1 proteins. Plant
Sci. Lett. 14:387-394.

Hall, N. and N. Tolbert. 1978. A rapid procedure for
the isolation of ribulose bisphosphate

carboxylase/oxygenase from spinach leaves. FEBS Lett.
96:167-169.

Strobaek, S. and G. Gibbons. 1976. Ribulose-l,
S-bisphosphate carboxylase from barley (Hordeum
vulgare). Isolation, characterization and peptide
mapping studies of the subunits. Carlsberg Res.
Commun. 41:57-72.

166.

167.
168.

tabacum L. from seed. Science 206:585.

144
Kung, S. et a1. 1974. Multiple peptide composition of
the large and small subunits of Nicotiana tabacum
Fraction I protein ascertained by fingerprinting and
electrofocusing. Biochim. Biophys. Acta 365:138-147.
Malmberg, R. Personal communication.

Burk, L., et al. 1979. Maternal haploids of Nicotiana

 

“‘mm1m?