CYTOGENETIC STUDIES OF COMMON BEAN (Phaseolus vulgaris) AND
THE SCARLET RUNNER BEAM (Phaseolus coccineus) ~

BY

SIMON S. CHENG

A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA

IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE -
DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

Elizabeth, my wife
Dr. Tien-Ho, my father
Dr. Su-Su, my mother

ACKNOWLEDGMENTS

The author expresses his sincere gratitude to Dr. Mark J. Bassett,
the chairman of his graduate committee, for his unusual ability to
"Kai-chiau" (open the doors of thinking), and his patience as an educa-
tor. The author also expresses his thanks to Dr. Kenneth H. Quesenberry
and his group of cytogenetic co-workers for the orientation and facil-
ities generously provided. Thanks are extended to the faculty members,
staff and all the graduate students of the Vegetable Crops Department
for the good times experienced during the author's pursuit of knowledge
at the University of Florida.

Special thanks are extended to the Brazilian Ministry of
Education and the Escola Superior de Agricultura de Lavras for financial
support during the author's training.

Finally, thanks are extended to Elizabeth, his wife, for her
patience duri ng hi s ordeal .

ii i

TABLE OF CONTENTS

Page

ACKNOWLEDGMENTS Hi

ABSTRACT vi

INTRODUCTION 1

PART I

MEIOSIS IN COMMON BEAN (Phaseolus vulgaris)

INTRODUCTION 4

LITERATURE REVIEW 6

MATERIALS AND METHODS 9

RESULTS AND DISCUSSION 12

Slide Preparation 12

Meiotic Sequence 14

SUMMARY 28

PART II

CYTOGENETIC ANALYSIS OF INTERSPECIFIC HYBRIDS BETWEEN
COMMON BEAN (Phaseol us vulgaris) AND THE
SCARLET RUNNER BEAN (Phaseolus cocci neus)

INTRODUCTION 30

LITERATURE REVIEW 32

MATERIALS AND METHODS 39

RESULTS AND DISCUSSION 41

Pollen Abortion Analysis 44

Gross Chromosome Homology Between the Genomes 45

Anaphase I Analysis 46

Anaphase II Analysis 51

Unequal Distribution of Chromosomes at Anaphase I . . . 54

Normal Pollen Abortion 55

iv

TABLE OF CONTENTS (Continued)

PART II (Continued) Page

RESULTS AND DISCUSSION (Continued)

F2 Seed Development 55

Aneuploid Progeny in Interspecific F^ 56

CONCLUSIONS 57

PART III

CHROMOSOME IDENTIFICATION AT THE DIPLOTENE STAGE OF
ME I OS IS IN SCARLET RUNNER BEAN (Phaseolus coccineus)

INTRODUCTION 59

LITERATURE REVIEW 62

MATERIALS AND METHODS 66

RESULTS AND DISCUSSION 68

CONCLUSION 85

SUMMARY 86

REFERENCES 87

BIOGRAPHICAL SKETCH 90

v

Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the Requirements

for the Degree of Doctor of Philosophy

CYTOGENETIC STUDIES OF COMMON BEAN (Phaseolus vulgaris) AND
THE SCARLET RUNNER BEAN (Phaseol us cocci neus)

By

Simon S. Cheng
August 1979

Chairman: Mark J. Bassett

Major Department: Vegetable Crops

Sixty photomicrographs of the meiotic sequence of common bean
(Phaseolus vulgaris) were presented to discuss the peculiarities of
meiotic configurations in bean meiosis. The advantages and limitations
of cytogenetic research in bean were discussed and compared with the
meiotic configurations observed in other plants, especially tomato and
maize.

A detailed quantitative analysis of chromosomal behavior during
the meiotic process in the pollen mother cells (PMCs) of interspecific
hybrids from vulgaris x P^ cocci neus crosses was conducted to find
the cause of the high degree (59-79%) of pollen sterility in F-j plants.
Contrary to the reported homology between these two genomes, 16 to 25%
of the PMCs were found in anaphase I and II to have either one chromo-
some bridge or occasionally two chromosome bridges involving different
pairs of chromosomes. Unequal distribution of chromosomes at anaphase I
(12:10) was also observed. The analytical data indicated that at least
two large paracentric inversions differentiated the two species in the
hybrids studied. The high crossover rate within these inverted segments

vi

of the chromosomes causes nearly half of the pollen abortion rate
observed in the interspecific hybrids studied.

An easy method of staining prophase chromosomes in scarlet
runner bean (Phaseolus coccineus) was developed by applying a small
quantity of 45% iron propionic carmine for a longer than conventional
treatment time. By using this staining procedure, it was found that
the 11 pairs of chromosomes could be identified in the diplotene stage
by means of characterise c banding patterns and the relative chromo-
some lengths. The 11 pairs of chromosomes were presented individually
with photomicrographs, interpretive drawings, and an idiogram. The
most suitable techniques for flower bud collection, slide preparation,
and photomicrographic illustration of meiosis were developed and pre-

sented in detail.

INTRODUCTION

Common bean blight is a very serious bacterial disease in tropical
regions where beans are grown in the rainy season under high tempera-
tures. No effective control measures for this disease are available
once the region is infested. No common bean variety is reported to
contain resistant genes against tropical strains of this bacterial dis-
ease. However, genotypes of Phaseolus coccineus have been selected at
the Mayaguez Institute of Tropical Agriculture in Puerto Rico for blight
resistance in an environment favorable for bacterial infestation (Dr.
Vakili, personal communication). However, the scarlet runner bean lacks
many of the essential qualities of the common bean (Phaseolus vulgaris).

The transfer of genes for resistance from P. coccineus to
P. vulgaris is a major objective of the bean breeding program at the
University of Florida at Gainesville. However, despite the reported
chromosome homology between these two species, the interspecific F-j
hybrid plants are usually semi-sterile. The genetic study of the pro-
genies is difficult because of selective embryo abortion, leading to
selection against certain genotypes. Therefore, a cytogenetic investi-
gation was conducted to search for abnormalities during meiosis in the
F-| hybrids to increase basic knowledge of Phaseol us cytogenetics.

Early attempts to stain and identify bean chromosomes (14,24)
failed because the prophase of mitosis did not offer sufficient differ-
ences in chromosome morphology. An effective staining technique is

1

2

needed to enable researchers to examine early prophase chromosomes in
bean species (24). If the prophase stages of meiosis in bean could
be adequately observed, it might be possible to identify the 11 chromo-
somes of bean cytologically.

ME I OS I S IN COMMON BEAN (Phaseolus vulgaris)

INTRODUCTION

The meiotic sequence in maize (27) and in tomato (2) has been
published for analytical purposes. However, the meiotic sequence in
common bean and other Phaseolus species has not been published with
proper photomicrograph and idiogram illustration. Publications cover-
ing bean cytogenetic analysis are scarce partly because of the diffi-
culties in staining chromosomes in the prophase stage (24).

The importance of meiotic configurations is presented by the
following arguments:

1. Knowing the whole sequence in advance is very helpful in
planning a cytogenetic research program and determining which stage to
investigate in order to get the desired data. The desired stages may
not be observable or may not be sufficiently clear to fulfill the exper-
imental purpose.

2. When the cytogenetic data are in disagreement with reports
of work with other species, differences in meiotic sequence may help to
interpret the discrepancies.

3. Detailed knowledge of the meiotic sequence enables one to
judge the suitability for cytogenetic research offered by the species
in question. By comparing the meiotic configurations of maize with
those of beans, it is known readily that the chromosomes of maize exhibit
greater detail than those of beans. Bean chromosomes are so small at
metaphase that the fragmentation produced by chromosome aberrations can

4

5

not be detected. Such fragments were essential to the precise cyto-
genetic analysis made by McClintock in maize (21). Double chromosome
bridges involving one chromosome pair are easily detected in maize but
not in beans.

The presentation of meiotic configurations in beans helps
researchers to understand the advantages and the limitations of cyto-
genetic studies in this species, avoiding useless speculation.

In this paper, we present meiotic configurations in bean PMCs
from interphase to pollen mitosis in photomicrographs taken at 500 X and
more often at 1000 X, using the phase contrast system in most of the
pictures. Discussions were made of the meiotic sequence itself, of
staining procedure innovations, and of the problems encountered in
interpreting the PMCs observed.

LITERATURE REVIEW

The classical scheme of meiotic sequence was established around
1930. The order of interphase - prophase (leptotene - zygotene -
pachytene - diplotene - diakinesis) - prometaphase I - metaphase I -
anaphase I - telophase I - interkinesis - prophase II - metaphase II -
anaphase II - telophase II - tetrad - pollen grain was followed by
Rhoades in his presentation of meiosis in maize (27). A brief discus-
sion of the observations of each stage was also presented in his paper.
However, molecular genetics began to be developed after Wastson and
Crick published the physical structure of DNA in 1953. This was a long
time after the publication by Rhoades (27). Therefore, some of his
interpretations and conclusions did not agree with reports of meiotic
events studied at the molecular level, especially his interpretation of
the structure and function of the nucleolus. Nevertheless, his meiotic
illustrations were very helpful, and many textbooks adopted them for use
in their chapters describing meiosis.

Brown (2) studied the function of euchromatin and heterochro-
matin in chromosome #2 of tomato during meiosis. He tried to determine
whether these two kinds of chromatin are different in the timing of
pairing, chiasma location and condensation rate. He found that chro-
matic zones (heterochromatin zones) pair more slowly than centromeric
and achromatic zones (euchromatin zones) from zygotene to pachytene.

Only 30% of the chiasmata terminal ize in the long euchromatin arm at

6

7

diakinesis while the average terminalization percent of ail tomato
chromosome arms is 70%. In the short heterochromatic arm of chromo-
some #2, chiasmata also occur at a much lower frequency. He concluded
that the heterochromatic zones at both sides of the centromere deter-
mine the chromosome morphology at metaphase.

The tomato chromosome map was established by Barton (1), using
the distinct heterochromatic structure at pachytene on both sides of
the centromere to differentiate the 12 chromosomes of tomato. The heter-
ochromatic zones are so specific that they compensated the dimensional
error produced through camera lucida drawings made during chromosome
identification.

The pachytene banding technique was used by Krishnan and De (14)
in an attempt to establish a chromosome idiogram for the Phaseolus
aureus genome. Unfortunately, the heterochromatic bodies in bean were
not as distinct as in tomato. In addition to experimental error, their
chromosome diagrams made at pachytene and metaphase led to many problems
of interpretation.

The classical scheme of meiosis was established on the basis of
PMCs observed under the microscope. Two of the events are challenged
today although the challengers do not have the type of direct proof to
change the "conventional wisdom" represented by the classical scheme.

One of the challenges is that chromosome pairing occurs at
premeiotic mitosis. Therefore the pairing at zygotene in the classical
scheme is pre-determined. The formation of a synaptonemal complex is
not a pairing process but rather the production of a device for crossing
over. The indirect proof comes from fungi, animals and plants (3) where

8

chromosome pairing is observed before leptotene stage. The challenge,
however, does not alter the scheme of meiosis because at early prophase
or interphase cytogenetic techniques do not have the necessary power of

r

resolution to settle the question directly.

Another challenge is more serious, and an alternative scheme
has been proposed to substitute for the classical one. It was discov-
ered by Moens (23) that there is a meiotic time sequence in the tomato
anther. The PMCs located at the basal part of the anther divide first
and those at the tip divide later. Thus in a longitudinal section, the
PMCs show the meiotic sequence from the tip to the basal part in the
order of occurrence. In this time sequence it was found that below the
pachytene zone, instead of the expected diplotene cells, the zone is
filled with PMCs having diffuse chromatins similar to the leptotene
and zygotene chromosome structure. Below this zone, the chromosomes
begin to condense again into diplotene, diakinesis and the remainder
of the classical scheme. Does zygotene come after pachytene is the
challenge. There is a question about the adequacy of the interpretation
given to the meiotic sequence in the classical scheme. Until direct
proof can settle the matter, the challenge provides an alternative view

of the matter.

MATERIALS AND METHODS

The Phased us vulgaris varieties La Vega, Great Northern #1 sel .
27, and a breeding line, 6-19, were used for this study. The flower
buds were harvested in the morning at about 9:30 a.m. and fixed in
Carnoy's fixative containing 6:3:1 by volume of ethanol (95%) :chloroform:
glacial acetic acid. After 24 hours of fixation, the buds were trans-
ferred to 70% ethanol for preservation. A dissecting microscope was
used to extract the anthers from flower buds between 2.5 and 3.0 mm in
length. The extracted anthers were put in an excess of 45% iron pro-
pionic carmine (2.0 ml for every 50 anthers) for one hour to stain the
chromosomes of all the stages except prophase. To observe prophase
chromosomes, a batch of 50 anthers was placed in 0.5 ml of the same
stain for one hour. During this period, the stain dries to 1/3 of the
original volume, which creates a stain concentration gradient. These
higher concentrations are needed to adequately stain prophase chromosomes.

Bean chromosomes are much smaller than maize chromosomes. In
order to have better chromosome spread, a more flattened PMC preparation
is required. In fact, mounting technique is a critical factor in bean
cytogenetics. From our experience of preparing hundreds of slides, the
best mounting procedure is described in detail below.

After staining, the anthers in the porcelain well were trans-
ferred to the slide, using a disposable pasteur glass pipet 14.60 cm
long. The anthers were sucked in and carried with the stain solution

9

10

in the narrow region of the pipet near the tip. If the anthers are
sucked into the broad chamber of the pipet, they cannot be discharged
easi ly.

After the anthers were transferred onto the slide, the next
step is to check the quantity of stain. Squashing the anthers in an
excessive amount of stain will expel many of the PMCs when the cover-
glass is put on. Therefore, the quantity of stain should be controlled
to the amount just enough to flow when the slide is tilted. Any excess
dye should be removed by absorption with an absorbent material such as
Kimwipes or filter paper.

Squashing the anthers with an ultrafine tweezer #5 should be
performed until no visible tissue pieces are present. Generally it
takes two minutes to reach this stage. Too much squashing will damage
the larger PMCs which are so precious for bean cytogenetic studies.

In order not to spread the PMCs out of the range of the coverglass,
the squashing is done with all the anthers concentrated at the middle
of the slide. The larger fragments of debris should be eliminated
as much as possible before the dye dries excessively. Coagulation of
squashed materials will occur if the stain concentration is increased
too much by evaporation or if a dirty tweezer containing traces of dried
materials is used for squashing. When squashing is complete, the drop-
let containing the squashed material is then spread out over an area
nearly equivalent to the size of the coverglass. However, the squashed
material should not be spread to the perimeter of the coverglass because
the PMCs will be lost during blotting. Another reason to restrict

11

the squashed material is that the edge of the coverglass will be sealed
with sticky wax. Therefore, all the PMCs near the edge are not avail-
able for observation. A 2-mm border around the coverglass should be
left free of the squashed materials.

After the coverglass is put on the squashed materials, the slide
is blotted between two pieces of filter paper to get rid of excessive
stain and small debris. Big PMCs are flattened during this step and
remain under the coverglass.

After blotting, the coverglass is sealed with sticky wax while
maintaining a high compression force applied on the coverglass with a
cork stopper. These operations are done simultaneously. One hand
presses the cork stopper on the coverglass and the other hand seals the
coverglass with a curved dissecting needle and hot wax. The high com-
pression force helps to make the PMCs flat. Additional stain is placed
under the coverglass when air bubbles or air space are present. The
ultrafine tweezer #5 is good for this "filling" work. A good slide
should not have air bubbles sealed within, because they will shorten
the life of a temporary slide by absorbing the moisture from the PMCs.

Fresh slides were used for photomicrographs using 500 X and
1000 X magnification. The phase contrast system was used to photograph
most of the stages. Kodakchrome slide film with ASA 25 and black and
white film Panatomic-X with ASA 32 were used for photomicrographs.

Because of the unavailability of PMCs containing chromosome
bridges and laggards in P. vulgaris, the PMCs from hybrids of P^_ vuj_gari_s
x P. coccineus were used to make photomicrographs (Figs. 28, 29, 41, 48)

of such irregularities.

RESULTS AND DISCUSSION

Slide Preparation

The preliminary problems that a cytogeneticist encounters
before he begins to work on a certain species of plant are:
a) determining the time of occurrence of meiosis in the anthers during
a 24-hour day, b) how to obtain large PMCs with their chromosomes well
spread, and c) how to obtain maximum contrast between chromosomes and
cytoplasm through the process of slide preparation.

In cytogenetic work, photomicrography is the principal tool
for presenting the results of the research. Good quality pictures can
only be obtained from good quality slides. Even for a cytogeneticist
with long experience, the above three problems should be solved for
every new species before he starts to work on any problem.

The duration of meiosis varies from plant to plant. In Orni tho-
galum virens (Lilliaceae), it was reported that it takes 4 days to
complete a meiotic cycle at 18°C (4). However, in most of annual plant
species the duration of meiosis is not so long. In these plants, most
cytogeneticists harvest the flower buds in the morning between 9:00 a.m.
and 11:00 a.m. Since it is easy to obtain dividing PMCs at this time,
the duration of meiosis in a plant species is generally neglected by
the researchers.

No information about the duration of meiosis in beans was
available. Therefore a survey was performed on 10/1/78 to determine

12

13

the time of meiosis in this species. Buds were sampled every 30 min-"
utes from 8:00 a.m. to 2:00 p.m. It was found that the major peak of
meiotic activity ocurred between 9:00 a.m. and 11:00 a.m. The buds
collected before or after this period contained very few PMCs in
metaphase and anaphase. The results indicate that bean meiosis does
not last more than 2 hours. The collection of buds before sunset
(5:00 p.m. ) also contained PMCs in intense meiotic activity. It seems
that bean meiosis has more than one activity peak per day. Some PMCs
divide in the morning and others divide in the afternoon. It is
assumed that meiosis is a function of solar activity in this species.

During our work, we discovered that the size of a bean PMC is
not constant throughout the flowering period. The larger PMCs were
found in anthers collected at the beginning of anthesis and the size
diminished as flowering progressed. The large PMCs observed were
almost twice the size of small PMCs. However, the size of their chromo-
somes in meiotic phases is not proportional to the size of PMCs.

Larger PMCs are better for observation of prophase chromosomes because
they provide more space for the longer chromosomes and reduce overlap-
ping. However, for photomicrographic purposes, smaller PMCs are desir-
able for those stages after metaphase II because the chromosomes are in
widely separated groups in these stages. It is difficult to photograph
the whole PMC under 1000 X magnification if the size of PMC is too
large. A PMC can be photographed at 500 X magnification, but the chromo-
some detail is not as good as at 1000 X. Photographs containing the
whole PMC are important for the study of chromosome number, especially
for aneuploid study. Under 1000 X or higher magnification, the loss

14

from the field of view of some portion of a large PMC together with
one or more chromosomes can be misinterpreted as an indication of
aneuploidy.

The preparation of prophase chromosomes in beans required special
attention in bud bleaching. Prophase chromosomes can only be stained
adequately when the cytoplasm of a PMC is thoroughly bleached by exces-
sive amounts of fixati ve agents. When too small amounts of fixative
are used to fix the bud clusters, the fixative solution becomes very
dark with the breakdown of chlorophyll. The dark solution also darkens
the cytoplasm of the PMCs, affecting the quality of the preparations.

Meiotic Sequence

1. Interphase is shown in Figs. 1 and 2, showing the same
nucleus at 500 X and 1000 X phase contrast. The genetic materials are
seen to be contained in the nuclear envelope. Those chromatic strips
are heterochromatin bands of the chromosomes. They are in a condensed
structure. The achromatic zones represent euchromatin. According to
Comings (5), this kind of genetic material is less condensed than

heterochromatin.

15

Figures 1 through 15. Meiotic sequence of bean (Phaseolus vulgaris)
from interphase to diplotene. (Figures are discussed in
the text. )

16

2. Pre-leptotene is presented in Figs. 3 and 4, showing the
same cell at 500 X and 1000 X phase contrast. In this stage, the
heterochromatin zones are apparently smaller than those of interphase.
Heterochromatin is reported to duplicate after the duplication of
euchromatin (3). The heterochromatic zones are probably in the process
of decondensation and duplication. Biochemical data reported by Hotta
et al . (10) showed that DNA replication is not completed until the
pachytene stage. They showed that any inhibition of DNA synthesis
before this stage produces damaged chromosomes. They refused to attrib-
ute the late synthesis (after leptotene) of DNA to the repair
mechanism which mends broken chromosome ends during crossing over. In
the classical scheme, DNA duplication is thought to be accomplished
before leptotene.

3. Leptotene is presented in Figs. 5 and 6, both at 1000 X

phase contrast. The nucleolus is the only distinct body. The volume of the
nucleus is greatly increased, and the chromatin threads cannot be seen
clearly in this stage.

4. Zygotene is presented in Figs. 7 through 11 in which the
homologous chromosomes synapse gradually. According to Brown (3),
each chromosome synthesizes a "lateral element," a proteinaceous core
structure along the side to be synapsed later. When the lateral elements
of the two homologous chromosome segments are put in juxtaposition,

a proteinaceous "central element" is formed parallel to the orien-
tation of the chromosome segment. Transversal fibrils are formed to
connect lateral and central elements. Thus, the "zippering" of the two
homologous segments is accomplished. The whole structure is called

17

a synaptonemal complex. The lateral element is thought to recognize
the homolog at juxtaposition for synapsis. The lateral element has
a width of 400 A° along side the chromosome segment. The central ele-
ment is 160 A° wide. The transversal fibrils are about 500 A° long
at each side. Thus, the distance between the two synapsed chromosome
segments is about 2000 A°.

Under a light microscope, 2000 A0 is the minimum dimension that
can be seen. Therefore, the synaptonemal complex between the two
chromosomes is seen as a thin line. Its fine structure, as mentioned
above, was discovered through electron microscopy.

The synaptonemal complex is directly involved in crossover
events. Chromosomes that pair without a synaptonemal complex
(as in mitotic pairing, or due to deletion or mutation) have fewer
chiasmata. The protein subunits for assembling the synaptonemal complex
are synthesized only at this stage during meiosis.

5. Pachytene is presented in Fig. 12 in which the two homo-
logous chromosomes are completely united by the synaptonemal complex.
Unlike tomato pachytene, the pachytene chromosomes of beans do not have
distinct and thick heterochromatic regions at both sides of the
centromere. The configurations of P. vulgaris from mid-zygotene to
pachytene show that the heterochromatic bodies are well distributed
along the chromosome arms. The structure of P. vulgaris chromosomes
has no similarity with that of P, aureus as reported by Krishnan and
De (14). During the prophase stage before diakinesis, most PMCs in
P. vulgaris present only one spherical nucleolus. However, 3.4% of
the 115 PMCs studied contain one or more extra nucleolus-like bodies

18

attached at different chromosomes as shown in Figs. 8, 17 and 34. In
P. coccineus we found that 18.2% of the 235 PMCs observed contain these
extranucleolar bodies. In the hybrid of P^ vul gari s x P^_ coccineus,
we found that 22.1% of the 285 PMCs observed contain extra nucleolar
bodies. The hybrid possesses the sum of the frequencies of the extra-
nucleolar bodies observed in the two parental species. The extra-
nucleolar bodies are smaller than the normal nucleolus although the
shape and chromatic nature are similar.

The nucleolus is an active transcribing section of the chromosome
during prophase. It is composed of puffed DNA coils and the transcribed
ribosomal RNA molecules. The ribosomal RNA molecules diffuse from the
nucleolus to the cytoplasm to combine with protein components, forming
ribosomes which are the site of protein synthesis (translation)(3) . In the
meiotic prophase of amphibians, many small extranucleoli not associ-
ated with chromosomes are seen floating within the nucleus. They are
synthesized through a special DNA multiplication process called gene
amplification in which copies of DNA from the chromosomes are released
into the nucleus for a quick transcription of ribosomal RNA in order to
translate a large quantity of proteins in amphibian egg cells (3).

Multiple nucleoli are also found in polyploid plants (3).

The presence of extranucleoli in some fraction of bean PMCs seems to
have evolutionary significance. It looks as if genes located in the
primary nucleolus may be duplicated at sites in other chromosomes, and
these genes are normally activated only in a fraction of the PMCs. One
may speculate that such redundancy of genetic information provides a
"safety factor" against the possibility of mutations or deletions of
the principal gene locations in the primary nucleolar chromosome.

19

Figures 16 through 30. Meiotic sequence of bean (Phaseolus vulgaris)

from diplotene to telephase I. Figures discussed in the text;
Figs. 28 and 29 were taken from hybrids of P. vulgaris x
P. cocci neus.

20

At the pachytene stage, inversion loops, addition-deficiency
loops, and translocation pairing can be observed if such chromosomal
aberrations exist in an aberration heterozygote.

When a pachytene chromosome is in a tilted position, it is
difficult to measure its absolute length and structure. Measurement of
chromosome length in pachytene has limited value in bean cytogenetic
analysis and is probably unfeasible due to excessive overlapping.

Although metaphase chromosomes of many species of animals and
plants are easily identified, pachytene morphology is used in maize
(27) and tomato (1) for chromosome identification. Their pachytene
chromosomes possess distinct physical markers. Unfortunately, in
beans such markers are not conspicuous enough to be used for chromo-
some identification.

There are ambiguous situations in pachytene interpretation when
some sections of the chromosome pair remain open as rings. It is dif-
ficult to tell if the failure of synapsis is due to an inversion, an
addition, a deficiency, or simply a delay in the synaptic process.
Therefore, pachytene observation is a supplemental approach rather than
a major one in the study of translocation, inversion, or addition and
deficiency.

6. Diplotene is presented in Figs. 13 through 19 in which the
chromosomes are condensing and separating. Chiasmata appear at the
location of crossovers and move to the ends of the chromosomes through
the process of terminalization. The centromeres are paired during
this stage (Fig. 14). In beans, this is probably an ideal stage for
chromosome identification because the chromosomes are more spread.

21

The characteristic banding patterns are still present. The two
homologous chromosomes are located side by side, which facilitates
comparison by observing details of the banding sequence. Any pair of
chromosomes which does not have any crossovers does not have any points
of contact, and the members of the pair may even be widely separated.

7. Diakinesis is presented in Figs. 20 and 21. At early
diakinesis the chromosomes are thick, with chiasmata terminal i zed at
the chromosome ends. Many chromosome pairs appear as rings because
the centromeres are no longer synapsed, and the chiasmata in both arms
are terminal i zed. Some chiasmata are not yet terminal ized at this
stage.

Late in this stage, the nucleolus appears to be reduced in size.
This is the stage of maximum chromosome condensation.

When there is insufficient homology between the two related
chromosomes, the univalents are often widely separated at diakinesis.

8. Prophase chromosomes are the same as metaphase I chromosomes.
However, they have not migrated to the equatorial plate yet. The
chromosomes in Fig. 22 may be at this stage. They could also be in
another orientation of metaphase I chromosomes.

9. Metaphase I chromosomes are present in Figs. 23 through

26 in which the chromosomes are located at the equatorial plate. Each
centromere faces each pole. The inconspicuous spindle fibers begin to
pull the chromosomes to each pole. It is difficult to identify bean
chromosomes at this stage because most of the chromosomes are not in
longitudinal orientation (Fig. 23). Some pairs of chromosomes may
separate earlier than others. The event (Fig. 26) is called
"precocious separation."

22

10. Anaphase I is presented in Figs. 27 through 29 in which
the chromosomes are migrating to each pole. A proteinaceous matrix
begins to connect the chromosomes into a spherical mass. Certain
types of crossover events occurring within the inversion loop (3,21)
can give rise to the chromosome bridges shown in this stage (Figs. 28
and 29);when chromosome bridges appear in high frequency, pollen ste-
rility is also high. The quantitative relationship between the bridge
frequency and the pollen abortion ratio is discussed by McClintock (21)
and Brown (3) .

The acentric fragment produced during chromosome bridge forma-
tion cannot be detected in most of bean PMCs (Figs. 28 and 29). The
inability to detect acentric fragments greatly reduces the power of
bean cytogenetic analysis to resolve important questions.

11. Telophase I is presented in Figs. 30 and 31, in which the
chromosomes are organized into a spherical mass. Euchromatin zones
begin to diffuse the chromatin into invisible fibrils. Then the two
daughter nuclei are formed.

Unlike maize PMCs, bean PMCs do not form a cell wall between
the two daughter cells at the end of the first division. Therefore,
the chromosome bridges in beans are not cut by the cell wall at this
stage. Many bridges are left intact at metaphase II.

12. Interkinesis is presented in Fig. 32. The nucleolus
reappears in each daughter nucleus, and the heterochromatin zones are
thick and more compact than those at interphase before meiosis. The
transcription of RNA in the euchromatic region is active in this stage.

23

Figures 31 through 45. Meiotic sequence of bean (Phaseol us vul garis )
from telophase I to anaphase II. Figures discussed in the
text; Fig. 41 was taken from a hybrid of P. vulgaris x
P. cocci neus.

24

13. Prophase II is presented in Figs. 33 through 38. This
process basically repeats the prophase I process mentioned above, except
that no DNA replication occurs at early prophase II. The daughter
nucleus in prophase II is much smaller than that of prophase I. There-
fore, their chromosomes are not easy to identify although details of
chromosome morphology are sharp and clear in some PMCs.

14. Metaphase II is presented in Figs. 39 through 42. The
metaphase chromosomes are spread on two equatorial plates in the two
daughter cells of the PMC.

Although they appear as two groups of chromosomes at the polar
regions, similar to that of an anaphase I PMC configuration, they do
not have a proteinaceous matrix connection. They are not subject to
the pulling force of the spindle fibers which fold the anaphase I
chromosomes. They are relaxed on the equatorial plates waiting for
further division. The doubleness of the chromosomes can be seen
in Fig. 42.

If the chromosome bridge does not break at anaphase I, it will
show up in this stage as a narrow thread with fragments near the bridge
(Fig. 41). If the bridge breaks, many laggard chromosomes are seen
floating between the two groups of chromosomes.

15. Anaphase II is presented in Figs. 43 through 46 in which
the chromosomes split and migrate to the four poles within the PMC.

Chromosome bridges (Fig. 48) appear in this stage if there have
been two crossovers during the first division, one occurring within the
inversion loop and the other in the proximal region, involving three
of the four chromatids. Many laggard chromosomes also appear in this stage

25

56

; V', V 'mV5

« *v. ;*y ..

*5 • ■ s \ \ - > v» *

'4^,

'X

• • 7

* $9

. % -..v

>• .

- '. /*v * JT > •

y-<«' v— r \>y- .
V* ^V*, \>-v

f **,r « T.<C

\ > 'v- V

H

! *

«

i-il

■*

>■* ' ' i?

/• -K

: O fV.

S 58

V **

. N '\

> * '

■yy .

v ®a,o ^ *?. ■

Skate iMw?

*» - ®?3£,©%‘ia"-

<,& A, »»

‘i «• S? ©tfiaML,,

Figures

46 through 60. Meiotic sequence of bean (Phaseolus vulgaris)
from anaphase II to pollen mitosis. Figures discussed in
the text; Fig. 48 was taken from a hybrid of P. vulqaris
x P. coccineus. 2

26

16. Telophase II is presented in Fig. 47 in which the chromo-
somes of each pole are connected by the proteinaceous matrix. Four
nuclei are formed within the PMC without the formation of cell walls to
separate them.

If a chromosome bridge breaks late in this stage, the nucleus
containing the broken chromosome will show a protrusion toward the oppo-
site nucleus. Since no cell wall is formed, many bridges remained intact.

17. The tetrad stage is presented in Figs. 49 through 51. The
four nuclei within the PMC develop from outside inward to separate the
PMC into four equal parts, each part having a pyramid shape (Fig. 51)
and containing a nucleus. The heterochromatin regions and the nucleolus
can be seen in the nuclei.

18. The pollen stage is presented in Figs . 52 through 60. The
tetrad cells change from their pyramid shape into a spherical shape
within the PMC envelope (Fig. 52). The pollen grains are released with
the rupture of the envelope (Fig. 53). Each pollen grain presents

4 exines (pollen tube exits) at the four corners.

The nucleus of a pollen grain undergoes two mitotic divisions
(Figs. 54 through 60), producing three nuclei in the pollen. One nucleus
functions in pollen as the tube nucleus, one fertilizes the egg nucleus
to form an embryo, and the other fertilizes the two polar bodies to form
a triploid endosperm. Accordi ng to Weinstei n (33) the endosperm of bean
develops first during seed development, and all of its materials, are
absorbed by the two cotyledons during the later stages of seed formation.

During pollen mitosis, the nucleus contains only 11 chromosomes.

In many nuclei, they are well spread. Nevertheless, it is difficult to

27

use pollen cells for chromosome identification because the chromosomes
are very small. Frequently, the pollen nucleus cannot, be seen clearly
because of the presence of starch grains in the pollen.

In Fig. 60, pollen grains with one, two and three nuclei are

present in the pollen population. However, on the day of anthesis,

the pollen nuclei are not visible after staining with iron propionic

carmine for 10 minutes. Mature pollen appears turgid and spherical in
shape. The shrunken pol len and empty pollen can easily be distinguished
from normal pollen filled with cytoplasm. Pollen analysis for semi-
sterility is an important part of cytogenetic analysis.

SUMMARY

A series of photomicrographs illustrating the entire sequence of
meiosis in common bean (Phaseolus vulgaris) was presented to discuss
the role of each stage in the cytogenetic analysis of beans. The pecu-
liarities of each stage were presented and their implications in bean
cytogenetic analysis were discussed. The study of pachytene PMCs
reported here is far from complete. Nevertheless, it is clear that
heterochromatin is well distributed throughout most of the chromosomes.
This contrasts markedly with tomato chromosomes in which the hetero-
chromatin is located in the proximal regions on both sides of the centro-
meres while the distal regions are largely achromatic.

In the common bean, the identification of the individual
chromosomes can best be achieved at diplotene because of the distinc-
tive banding patterns. Although chromosome bridges resulting from
inversions can be seen at anaphase I and anaphse II, the double bridges
in one chromosome and acentric fragments reported by McClintock (21)
in maize are not observable in common bean.

28

CYTOGENETIC ANALYSIS OF INTERSPECIFIC HYBRIDS BETWEEN
COMMON BEAN (Phased us vulqaris) AND THE
SCARLET RUNNER BEAN (Phaseolus coccineus)

INTRODUCTION

High pollen abortion has been observed previously (16,17) in
the F-j plants from crosses between Phaseolus vulgaris and Phaseolus
coccineus . However, the meiotic process of the F-j plants was reported
to be normal and without any evidence of chromosome aberration. The
two genomes were reported to have identical chromosome structure on the
basis of Giemsa staining of root tip chromosomes (24). This banding
technique is presently the only means of identifying the 11 chromosomes
of beans (25). There is still no satisfactory explanation of the mechan-
ism of pollen abortion in the hybrid plants (16,17,18,30).

The only report about abnormal meiosis in F-| plants was published
by Xolocotzi et al . (34) in Mexico. In natural interspecific hybrids
of these two species, 25 to 50% of pollen mother cells (PMCs) presented
a chromosome bridge in anaphase I and anaphse II. Two univalents in
prophase stage were observed occasionally. Although such high meiotic
irregularity was present, these authors did not report pollen viability
or seed development in their paper. The partial restoration of pollen
fertility in amphi diploids derived from these two species reported by
Smartt and Haq (30) implied the existence of gross chromosome struc-
tural differences between these two genomes, besides the presence of
genic sterility.

Evidence of interspecific barriers between P. vulgaris and P.
coccineus is observed frequently in such problems as difficulties in

30

31

obtaining F-j seeds (11,16), the failure of F-j plants to develop, the
failure of F-j plants to flower (13), high sterility in F^ and advanced
generations, and the loss of desirable genes from P. coccineus (19,32).
Although environmental conditions may contribute to pollen abortion,
no previous study relates inversions with the high percentage of pollen
abortion in interspecific hybrid beans.

In this paper, we have searched for meiotic aberrations to
explain the high pollen abortion observed and have challenged previous
reports that the chromosome structure of these two species is probably
free from large aberrations.

LITERATURE REVIEW

The genetics of hybrids between F\_ vul gari s and cocci neus
was studied thoroughly by Lamprecht during the 1940s (16,17,18,19).

In his studies of the inheritance of more than a hundred characters
differentiating these two species, he concluded that only two char-
acters can neither be transferred from P. cocci neus to P. vulgaris
nor vice versa. The two characters are the cotyledon position and
stigma orientation. P. vulgaris has an epigeous cotyledon and an
internal stigma position. P. coccineus, on the other hand, has a
hypogeous cotyledon and an external stigma position. The former spe-
cies is self-pollinated while the latter is cross-pollinated by
insects. These two characters ("interspecific genes" according to
Lamprecht) are thought to be specific for each species. It was pos-
tulated that some unspecified type of isolation mechanism prevents the
incorporation of species specific genes into another species.

Lamprecht observed chromosome pairing in the hybrids from
these two species and found the 11 bivalents at diakinesis and meta-
phase I to be normal. Two univalents occur only occasionally. In his
reports Lamprecht stressed that the meiotic process of the hybrid runs
a normal course without detectable aberration. However, he did not
mention how many PMCs he analyzed at anaphase I, anaphase II or meta-
phase II. His publications indicate that he scanned many cells at
diakinesis and metaphase I and found them all normal. Therefore, the

32

33

examination of the later stages probably was not conducted in detail
before he ruled out the possibility of any structural differences
between the two sets of chromosomes.

In order to explain the occurrence of a 60-80% pollen abortion
rate, he set up the hypothesis that specific genes of P. cocci nues
cannot be replicated in P. vulgaris cytoplasm. During the reproductive
process in the hybrid there is selective suppression of replication of
certain alleles from P. coccineus. These specific alleles from
coccineus are thus eliminated gradually as the inbreeding process
approaches homozygosity. He called this hypothesis "mutation of inter-
specific genes." This hypothesis involves selective replication of DNA
(copy choice) which is inconsistent with current molecular genetic
principles .

Mok and Mok (24) tried to detect the structural differences
between the chromosomes of P. vulgaris and P. coccineus through Giemsa
staining at the mitotic prophase stages. They reported that no differ-
ence could be detected using this technique. These results are in
agreement with the opinion of Lamprecht that no chromosome structural
differences exist between these two species. They also used this tech-
nique to identify monosomic chrosomes in P. vulgaris (25).

Chromosome structural differences within P. vulgaris were
reported by Honma (9). He found that at least two inverted segments are
present in different chromosomes in crosses between Blue Lake pole bean
and Contender. The supporting evidence was the repeated observation of
two chromosome bridges present simultaneously in different pairs of chro
mosomes in the same PMC. Honma suggested that these inversions may

34

explain his failure to transfer pole Blue Lake pod characters into the
bush bean Contender.

An inheritance study of cotyledon position in hybrids between
these two species was conducted later by Wall and York (32). They found
that the inheritance of this character is quantitative and that both
the hypogeous cotyledon and external stigma gene frequencies decrease
progressively with successive inbreeding generations.

Innes (11) reported the complete failure to backcross the inter-
specific hybrid onto the P. cocci neus parent used as female. His breed-
ing program encountered many difficul ties because he used P. cocci neus
as female. Lamprecht (16,19) had reported earlier that P. vulgaris
must be used as female parent for interspecific crosses.

Kedar and Bemis (13) reported arrested development among some
fraction of interspecific plants. The behavior of these dwarfs
varies from dying in the seedling stage to developing a weak and sterile
plant. Two different physiological blockage factors were found in dif-
ferent hybrids. When the two factors are genetically combined into one
genotype, the resulting seedling is arrested in development with the
cotyledon energy only partially used. The seedling dies within a few
weeks with the cotyledons largely unabsorbed by the plant.

A detailed analysis of chromosome inversion in maize was
published by McClintock in 1938 (21) using an inversion in chromosome
#4 for quantitative study. Since the acentric fragment produced by
crossing over within the inversion loop is visible in the cytoplasm
during meiosis and pollen mitosis, she was able to observe the fate of
these fragments and conclude that such a fragment is randomly distributed

35

into the four spores derived from a PMC. From counts of sporocytes
with bridge configurations at the second division, she estimated that
75% of the anther spores contain the normal chromosome #4 while 25%
contain the fragmented chromosome. The frequency of a normal spore end-
ing up with an acentric fragment was determined to be 7.5% while the
frequency of an aberrant spore (carrying a fragmented chromosome #4)
getting an acentric fragment was 26%. By counting the bridged chromo-
some frequencies in the pollen grains during the mitosis and interphase
stages, she found that most of the pollen in mitotic phase did not have
bridge protrusion and that those pollen cells with bridge protrusion
were still in interphase. She concluded that pollen containing a frag-
mented chromosome exhibits delayed pollen mitosis. During the anaphase
stage of pollen mitosis, she observed that the bridge frequency was
the same as the frequencies of chromosome bridges during anaphase I and
anaphase II in meiosis. Thus, she concluded that the chromosome bridges
observed in pollen mitosis are derived from chromosome bridges in ana-
phase I and II which broke apart. The broken ends of the fragmented
chromosome and its newly replicated copy are joined together during
pollen mitosis. A bridge is formed when the centromeres are pulled to
the poles.

In beans, the cytogenetic analysis is not as precise as in
maize because the chromosomes are much smaller in most meiotic stages
and their fragments are not detectable in most of the PMCs. However,
the mechanism of chromosome bridge formation and its consequences in
maize also apply in beans.

36

In her diagram of chromosome bridge formation, she showed the
types of crossovers that produce:

1. Single bridge at anaphase I with one acentric fragment.

2. Double bridge with two acentric fragments at anaphase I.

3. Single bridge at anaphase II.

4. Double bridge at anaphase II.

A more complete mechanism of chromosome bridge formation was
presented by Brown (3). He cited five types of crossover events that
give rise to an anaphase I bridge:

1. Single exchange in the inversion loop.

2. Three-strand double in the loop.

3. Two or four-strand doubles with one in the loop and one in
the proximal region.

4. Three-strand double with one in the loop and one in the
distal region.

5. Four-strand double in the loop (double bridges).

Only one type of crossover gives rise to an anaphase II bridge, viz.,
the three-strand double with one crossover in the loop and one in the
proximal region.

Chromosome bridges can be formed with a higher number of cross
overs. The frequency of such cases is so low that it is negligible.
One or two crossover events within a chromosome are usually taken into
consideration in analytical data.

Chromosome bridge formation produces pollen abortion. During
bridge formation an acentric fragment, which carries the duplicate dis
tal sections and the whole inversion loop, is lost in the cytoplasm.

37

Therefore, both of the two chromosomes lack the distal section after
bridge breakage. The random breaking point along the chromosome bridge
also produces duplication and deficiency in the fragmented chromosome.

Thus the occurrence of an anaphase I or an anaphase II bridge is respon-
sible for the abortion of two of the four pollen grains formed from
a PMC. Occasionally the aberrant spore can be functional, but this
requires that the acentric fragment happens to be in its nucleus (28).

In maize, the chromosome bridges are broken either by the
pulling force of the spindle fibers or by the formation of cellular
walls between the daughter cells.

The position of an inverted segment can be determined by
recombination frequencies, using the inverted segment as a gene to
recombine with gene markers along the chromosome in a series of three-
point test crosses. The method is illustrated in maize by Morgan (26).

The change of chromosome structure has been thought to be a
driving force of the evolution of species. In the Phaseol us genus,

Sarbhoy (29) reported that P. vulgaris is probably one of the most
primitive species of the genus because most of its chromosomes have
either a median or submedian primary constriction, while the more advanced
species have either a submedian or a subterminal primary constriction.

The chromosome breakage and fusion events during the evolutionary process
damage many genes at or near the breaking points, and this results in
small and undetectable differences among the species of the same genus.

The interspecific hybrid may show some degree of sterility of this nature,
resulting from cryptic hybridity (29). Another type of hybrid

38

sterility between species is produced by major structural differences,
e.g., translocation, inversion, addition and deficiency.

Crosses involving advanced Phaseolus species, such as P.
aureus x P_^ trilobus (6) , or F\_ lunatus x P^ polystachyus (8) , have a
higher degree of sterility. Major structural differences between these
species were detected through a high number of pachytene loops, univa-
lents and multivalents at diakinesis and metaphase, chromosome bridges
at anaphase I and anaphase II, and a high pollen abortion rate.

MATERIALS AND METHODS

Two P. cocci neus lines (Pc-37 and Pc-50) and three P. vulgaris
lines (6-19, Great Northern #1 sel. 27, and La Vega) were used in inter-
specific hybrid crosses, using the P. vulgaris lines as female parents.
The two P. cocci neus pollinator lines were obtained from Dr. Nader Vakili
at the Mayaguez Institute of Tropical Agriculture in Puerto Rico. These
lines were derived from an open-pollinated population which was mass
selected for multiple disease resistance, including high resistance to
common bacterial blight.

For pollen abortion analyses, flower and bud materials were
harvested in the morning, i.e., both freshly opened flowers and buds
due to open in approximately 24 hours. The anthers were stained with
45% iron propionic carmine for 20 minutes before squashing on slides.

The unstained, transparent, and shrunken pollen grains were classified
as aborted. The round, fully developed, and well-stained grains were
classified as normal.

For cytogenetic analysis, the F-| and parental lines were
planted in late August 1978 at Gainesville, Florida. During October,
the flower buds were harvested and fixed in Carnoy's fixative containing
6:3:1 by volume of ethanol (95%) :chloroform:glacial acetic acid. After
24 hours, the buds were transferred to 70% ethanol for preservation.

For meiotic study, anthers were extracted from buds in the range of 2.5
to 3.0 mm in length and were stained in 45% iron propionic carmine for
45 to 60 minutes before squashing on slides.

39

40

The PMCs were classified individually according to their stage
of meiosis and the presence of abnormalities. Although diakinesis and
metaphase I data were taken, only anaphase I and anaphase II data were
used for quantitative analysis. Metaphase II configurations were espe-
cially emphasized because the chromosomes are well dispersed in the
PMCs and because they give valuable information about chromosome aberra-
tions and their consequences.

The mature F 2 seeds were harvested from dried pods on several F-|
plants planted in the spring (March) of 1978. They "over-summered" in
spite of die-back of the above ground plant parts during mid-summer.
These F-j plants regenerated from the roots in the fall and flowered
early enough to produce mature seeds before frost killed them early in
December. The harvested seeds were dried and weighed individually in
order to study seed development.

RESULTS AND DISCUSSION

Pollen Abortion Analysis

As presented in Table 1, the interspecific hybrids had 58 to 79%
of non-stai nabl e pollen grains while their parental lines had less than
21. Very little difference was detected between the pollen sterility
on the day of flowering and one day before flowering. The range of
pollen sterility observed in this work is in accord with the reports of
Lamprecht (16) and Smartt and Haq (30). The 60 to 80% of pollen steril-
ity observed is thus typical of hybrids between these two species.

Undivided PMCs four times the size of a normal pollen grain
were frequently seen among the pollen population of F-j plants. Such
undivided PMCs were not observed among the pollen grains of parental
lines. The interspecific hybridization probably interfered to some
degree with the meiotic mechanism during pollen formation. Such a phe-
nomenon has been reported in interspecific crosses in All i urn (12).

While the stainable pollen grains of the parental lines are
uniform in size, those from F-j hybrids are heterogeneous (Fig. 61-A2).
The differences in size may contribute to different degrees of viabil-
ity and vigor. If viability and vigor are a function of size, this
could lead to differential zygotic frequencies and non-Mendel ian ratios
for genes linked to loci controlling pollen characteristics.

41

42

Table 1. Pollen staining analysis at anthesis and one

day before anthesis* of P. vulgaris, P. cocci neus
(Pc) and their hybrids.

At

Anthesis

P. cocc.
Pc 3 7

P. vulg.
6-19

Fi

L.V. x

Pc 3 7

Fi

G.N. x
Pc 37

Fi

6-19 x
Pc 3 7

Total pollen

633

755

1103

586

892

Aborted pollen

13

5

825

342

589

% abortion

2.05

0.66

74.80

58.36

66.03

One day before

Anthesis

Total pollen

651

768

1359

650

664

Aborted pollen

10

6

1073

418

470

l abortion

1.53

0.78

79.32

64.30

70.78

★

Anthers were stained for 20 minutes in propionic carmine
prior to squashing on slides.

Figure 61. Pollen grains, meiotic configurations of interspecific

hybrids between vul gari s and P^_ cocci neus , and a

histogram of individual seed weights of interspecific

F2 seeds.

A-l: Normal pollen grains from P. cocci neus.

A-2: Pollen grains from a hybrid plant showing empty pollen
and stained pollen grains of different sizes.

B: A normal anaphase I PMC with two groups of chromosomes

arranged in spherical form around each pole (1000 X
phase contrast) .

C: A chromosome bridge at anaphase I in a PMC from an F-,

plant (1000 X phase contrast).

D: Anaphase I PMC from an F] plant showing two chromosome

bridges joining different pairs of chromosomes (1000 X
phase contrast).

E: A normal metaphase II PMC from P. vulgaris showing

chromosomes spread on two different planes and a sharp

chromosome image (1000 X phase contrast).

F: A metaphase II PMC from an F] plant showing chromosome

bridge-laggards (1000 X phase contrast).

G: An anaphase II PMC from an F] plant showing a chromosome

bridge in one half and a chromosome laggard in the other

half (1000 X).

H: An anaphase II PMC from an F^ plant showing a chromosome

bridge in one half (500 X).

I: A metaphase II PMC from an F] plant showing 10-12 chromo-

some distribution (500 X).

J: A histogram of individual seed weights (mg) of seeds.

44

45

Gross Chromosome Homology Between the Genomes

P. vulgaris and P. coccineus chromosomes share a considerable
degree of homology. In diakinesis and metaphase I PMCs, 11 pairs of
chromosomes are normally observed in the hybrids of the two species.

Two univalents are seen only occasionally, which was also reported by
Xolocotzi et al . (34). The low frequency of univalents was also
observed in both parental lines. These two univalents appear to be the
result of precocious separation of a certain pair of chromosomes.
According to the literature, chiasmata are responsible for close pair-
ing at diakinesis and metaphase I. The frequent appearance of precocious
separation in one pair of chromosomes probably indicates a low fre-
quency of crossing over in this pair. In interspecific hybrids, asynap-
sis may occur when structural differences are located along the chromo-
somes, causing difficulties in pairing. Since the two univalents are
also observed at low frequency in the parental lines, the frequency of
occurrence of univalents in those interspecific hybrids is not suffi-
cient to indicate structural differences between the two species in the
chromosome involved. In other interspecific hybrids of the Phaseolus
genus where gross structural differences were reported (6,8), univa-
lents at metaphase I were observed in more than one chromosome and at
high frequency.

Although our data on prophase observations cannot detect struc-
tural differences of chromosomes, it cannot be used to prove structural
homology either. Not all types of structural differences give rise to
univalents. Lamprecht (16,18) was not justified in assuming complete
structural homology on the basis of prophase observations of the

46

interspecific hybrids of these two species. However, our data indicate
that F\_ vulgaris and F\_ cocci neus have a closer relationship than that
between F\_ aureus and P^ tri lobus (6) or between F\_ 1 unatus and P.
polystachyus (8) .

Anaphase I Analysis

The analysis of chromosome inversions was well illustrated by
McClintock (21) and by Smith (31). The data from anaphase I and
anaphase II are used together to count crossing over events which
occurred within the inversion segment. A chromosome bridge occurs
either at anaphase I or at anaphase II depending on the type of cross-
over events involved. According to Brown (3), there are five types of
crossover events that produce an anaphase I bridge, but only one type
that causes an anaphase II bridge. Their common requirement is that
at least one crossover occurs in the inversion loop.

The anaphase I data from our material (Table 2) indicate 15.6
to 25.6% of the observed PMCs have one chromosome bridge, or occasion-
ally they have two bridges in the same PMC involving different pairs of
chromosomes (Figs. 51 -C and -D). The cross with Pc-50 used as the male
parent has a much higher bridge frequency than the crosses with Pc-37
used as male parent. The use of different P. vulgaris lines as female
parents did not result in different bridge frequencies.

The high bridge frequencies observed and the simultaneous
presence of two bridges at different pairs of chromosomes indicate that
there are at least two large inverted segments located in different
pairs of chromosomes. These inversions are paracentric.

47

Table 2. Anaphase I bridge frequencies' in' P. vulqaris,
P. coccineus and their hybrids.

Species

Total
cel 1 s

1

bridqe

2

bridqes

%

bri dqes

6-19 (vulgaris)

102

0

0

0.0

L.V. (vulgaris)

50

0

0

0.0

Pc37 (coccineus)

116

1

0

0.8

F-j 6-19 x Pc37

143

21

1

16.8

F1 L.V. x PC37

146

21

1

15.6

F1 6-19 x Pc50

117

18

6

25.6

48

In the intraspecific hybrid of P. vulgaris between the Blue
Lake pole bean and the Contender snap bean varieties, Honma (9) also
observed two chromosome bridges each occurring in a different pair of
chromosomes. His data indicated that two large inverted segments are
present, differentiating the chromosomes of the two bean varieties.

At the present stage of bean cytogenetics, it is impossible to tell
which chromosomes contain these inversions. Also, we do not know
whether or not the two paracentric inversions in the present work are
identical to the two paracentric inversions reported by Honma. How-
ever, the latter question could be answered by crossing line Pc-50 to
Blue Lake pole bean and Contender. If the two inversions are identical,
one cross should give two bridges in some PMCs and a much higher fre-
quency of bridge-laggards similar to the present work. The other cross
should not give any bridges.

The number and size of inversions and the chiasma frequency
within the inversion loops all contribute to bridge frequencies in
inversion heterozygotes, according to Sarbhoy (29). In our material
precise information on the location and number of inverted segments
cannot be obtained because P. vulgaris and P. cocci neus lack chromo-
some and linkage maps. The technique for detecting the location of
inverted segments used by Morgan (26) cannot be applied at the present
stage.

The quantitative relationship between bridge frequency and
pollen abortion ratio was presented by Brown (3) and by Rhoades and
Dempsey (28), using maize (Zea mays) as an example. In maize, when one

bridge is present at anaphase I, two of the four pollen grains from that

49

PMC are expected to abort because of chromatid deficiencies resulting
from a crossover within the inversion loop. The other two chromatids
remain intact. They join the polar groups of chromosomes together with
the broken chromatids when the bridge is broken. Thus, in maize the
expected percentage of pollen abortion due to anaphase I events is half
of the bridge percentage observed in anaphase I. For example, if 10% of
the PMCs are observed to have a bridge at this stage, they are respon-
sible for S7o pollen abortion in the pollen population of that plant.

The above calculation is based on two assumptions. The first
is that the chromosome bridge breaks at anaphase I. The second is
that the two bridged chromosomes join the polar groups after bridge
breakage. The present study indicated that neither of these two assump-
tions is true in bean PMCs to the extent observed in maize. In the
hundreds of PMCs studied it was noted that most of the bridges did not
break at anaphase I. After telophase I, interphase, and prophase II,
the bridges are still intact in metaphase II cells (Fig. 61-F).
Furthermore, the bridge frequencies observed in clusters of metaphase II
PMCs were always similar to those observed in anaphase I PMCs. Unlike
the PMCs in maize, bean PMCs do not form a cellular wall to separate the
daughter cells at telophase I. The formation of a cellular wall is
responsible for cutting the chromosome bridge in maize at telophase I,
but this does not happen in bean PMCs. A few bridges do break as a
result of stretching in anaphse I. Consequently, the four chromatids
of the bridged chromosomes become laggards in the middle metaphase II
cells (Fig. 61-F). From the above observations we infer that none of
the four chromatids will join the polar group when there is bridge

50

formation in the bean species studied. Furthermore, we conclude that
in beans, the percentage of chromosome bridges in anaphase I is directly
responsible for an equivalent percentage of abortion among pollen grains.

In our material, chromosome bridges are responsible for 15.6, 16.8, and
25.6% pollen abortion in the three interspecific hybrids studied.

Since a small percentage of chromosome bridges break at anaphase I,
the bridge frequency in metaphase II is always slightly lower than that
of anaphase I. Since only the anaphase I bridge frequency is directly
related to the pollen abortion ratio, it is necessary for the observer
to distinguish the chromosome configurations of anaphase I and metaphase
II in bean species. Any failure to distinguish between these two meiotic
phases will underestimate the bridge frequency and pollen abortion ratio.
Since the existence of chromosome bridges at the metaphase II stage has
important implications in the cytogenetic analysis of bean, three major
criteria are suggested here for discriminating between anaphase I cells
and metaphase II cells (Fig. 61-B,C,E, and G).

1. In anaphase I, the chromosomes cluster around two points
(poles), forming two spherical masses. In metaphase II, the chromo-
somes spread on two different planes (equatorial plates).

2. The chromosomes of anaphase I are dragged by spindle fibers.
Although the tiny bean chromosomes do not show V or J shapes, they
become folded and thicker as force is applied. Chromosomes at the
metaphase II stage have just passed prophase II. Metaphase II chromo-
somes are relaxed and are not deformed by any dragging force. There-
fore, metaphase II chromosomes have a sharp chromosome edge, in contrast
to the folded and fuzzy-edged chromosomes of anaphase I.

51

3. Anaphase I chromosomes tend to uncoil or "melt" into
interkinesis. Many proteinaceous ligands begin to connect the chromo-
somes at each pole. The metaphase II chromosomes tend to split into
two chromatids and no ligand is formed.

Anaphase II Analysis

The meiotic aberrations observed as chromosome bridges and
laggards in anaphase II (Table 3 and Fig. 61 -G and H) are identical to
those observed in anaphase I of the hybrids, La Vega x Pc-37 and 6-19
x Pc-50 (Table 2). They are 16.0 and 25.7%, respectively. The equal
frequencies of bridge-laggards in these two phases were previously
reported by Xolocotzi et al . (34). The mechanics of anaphase II bridge
formation was discussed by McClintock (21) and by Brown (3). The pres-
ent data confirm the equal bridge-1 aggards frequencies of anaphase I
and anaphase II PMCs.

According to Brown (3), there are five types of pachytene
configurations that give rise to anaphase I bridges: a single exchange

or a three-strand double crossover within the inversion loop, a two- or
four-strand double crossover with one crossover in the inversion loop
and another in the proximal region, a three-strand double crossover
with one in the inversion loop and another in the distal region, and a
four-strand double crossover in the inversion loop. Only one pachytene
configuration gives rise to anaphase II bridges. That is the three-
strand double crossover with one in the inversion loop and another in
the proximal region.

In this work, anaphase II bridge-laggard frequencies were
observed as 16.0 and 25.7% in two of the hybrids. Identical bridge-

52

Table 3. Anaphase II bridge-laggard frequencies of

F\_ vulgaris,. cocci neus and their hybrids.

Species

Total

cells

1

laqgard

1

bridqe

%

abnormal

6-19 (vulgaris)

45

0

0

0.0

L.V. (vulgaris)

40

0

0

0.0

Pc37 (coccineus)

44

0

0

0.0

F-j 6-19 x PC37

63

4

2

9.5

F-j L.V. x PC37

131

13

8

16.0

F1 6-19 x PC50

66

7

10

25.7

53

laggard frequencies were observed for anaphase I PMCs. Our difficulty
is that we cannot explain the high frequencies of bridge-laggard events
observed in anaphase II which required a particular type of double
crossover event involving three strands for its occurrence.

It seems probable that the frequency of the four pachytene
crossover events leading to anaphase I bridges (3) is greater than the
frequency of three-strand double crossovers which produce bridges in
anaphase II. In maize the expectation that anaphase I bridges should
be higher than anaphase II bridges has been confirmed by experimental
observations (21,28). However, our observations and those of Xolocotzi
et al . (34) show that the bridge frequencies of anaphase I and II in
bean are equal. The lower bridge-laggard frequency observed in anaphase
II PMCs of the hybrid 6-19 x Pc-37 is probably caused by sampling error.

In bean plants, cytogenetic analysis cannot be done as precisely
as in maize (21) because of the difficulty in detecting tiny acentric
fragments and double bridges involving two chromatids of the same
chromosome. Bean chromosomes are too small for these events to be
observable. Also, bean chromosomes cannot be distinguished in pachytene
phase. Thus, we have no way to determine the frequency of the various
types of crossover events leading to bridges and fragments when an
inversion is present.

According to the literature (3,28), the PMCs of maize with an
anaphase II birdge are responsible for 50% of pollen sterility in pollen

54

derived from those PMCs. Thus, 8.0 and 12.8% pollen sterility (1/2 x
16% and 25.7%) in the bean hybrids La Vega x Pc-37 and 6-19 x Pc-50,
respectively, are the direct result of anaphase II bridge-laggards
events.

Unequal Distribution of Chromosomes at Anaphase I

It is difficult to find a sufficient number of PMCs with
chromosomes well spread (without overlapping or connection) for chromo-
some distribution analysis. In such analysis, metaphase II cells are
more suitable than anaphase I cells. Only 10 PMCs in this category
were observed. Two of these PMCs showed 10 chromosomes at one pole and
12 chromosomes at the other (Fig. 61-1). The sample was too small to
provide a reliable estimate of the frequency of this aberration. How-
ever, it is certain that some fraction of the sterile pollen is due to
unequal distribution of the chromosomes at anaphase I stage. The cause
of this irregularity could be either asynapsis or both centromeres being
dragged to one pole at anaphase I. Perhaps due to "spindle fiber fail-
ure" at one pole, both centromeres of the bridged chromosome pair were
dragged entirely to one pole intact. In fact, one of the two PMCs in
this analysis showed two chromosomes firmly connected among the group of
12 chromosomes.

A 10:12 distribution resulting from asynapsis is less likely
because the frequency of two univalents observed in diakinesis was too
low. It is more likely that this 10:12 distribution was produced by the
breakage or stretching of one of the spindle fibers attached to a bridged
chromosome in view of such high frequencies of chromosome bridges in

55

anaphase I. In this case, dragging the bridged chromosome to one pole
may cause at least 50% pollen abortion in the tetrad, depending upon
the behavior of the connected chromosomes during anaphase II. It may
form a chromosome bridge if the connected chromatids are pulled to
opposite poles.

Normal Pollen Abortion

In normal plants pollen abortion can occur as high as 5% (26).

We observed 2% in P. cocci neus and less than 1% in P. vulgaris. This
fraction should be considered in the makeup of pollen sterility observed
in those interspecific hybrids studied.

Seed Development

The mature seeds harvested from plants of the Great Northern
#1 sel . 27 x Pc-37 hybrid showed a wide distribution in terms of seed
development (Fig. 61 -J ) . Only 30% of the 155 seeds were fully developed.
Another 30% of the seeds were shrunken, with each individual seed weight
in this group less than 200 mg. The remaining 40% of the seeds were
intermediate in development. The present data on F-| meiotic analysis
do not show any direct relationship between chromosome aberrations and
the causes of underdevelopment of seeds.

56

Aneuploid Progeny in Interspecific Fp

The observation of metaphase II cells with bridge-laggard and
with unequal distribution of chromosomes at two poles raises the possi-
bility that aneuploid progeny may be produced by chromosome inversions
in bean species if the abnormal megaspores can survive. Since at least
two inversions in different pairs of chromosomes are present in these
two species, many types of aneuploids can be obtained, including mono-
somies and trisomics.

CONCLUSIONS

The above analytical data indicate conclusively that the genomes
of P. vulgaris and P. coccineus cannot be considered as identical in
chromosome structure. In the materials studied, at least two chromo-
some inversions located in different pairs of chromosomes and associated
with high crossing over rates account for nearly half of the pollen
abortion observed in the interspecific hybrids. If we consider addi-
tional percentages of pollen abortion due to unequal distribution of
chromosomes at anaphase I and the occurrence of normal pollen abortion,
the total fraction of pollen abortion caused by gene unbalance may be
less than half of the observed percentages. These data also help to
explain the results obtained by Smartt and Haq (30) in which the pollen
fertility was improved from 21% in the interspecific hybrids to 42% in
the amphidiploids, and reaching 76% by further selection with inbreed-
ing. The residual 25% of pollen sterility was probably due to homozygous
genic imbalances which could not be removed by selection.

57

CHROMOSOME IDENTIFICATION AT THE DIPLOTENE STAGE OF
ME I OS IS IN SCARLET RUNNER BEAN (Phaseolus coccineus)

INTRODUCTION

The identification of each chromosome of the genome of a plant
species is a very important step in the development of cytogenetic knowl-
edge about that species. Once individual chromosomes can be identified
on the basis of their morphology, it becomes much easier and requires
less time to perform additional cytogenetic studies, e.g., the associa-
tion of genes with a particular chromosome, the detection of structural
differentiation within or between species, and developing aneuploid
and translocation stocks.

In maize, different knob positions and centromere location on
each of the 10 chromosomes were used to identify each chromosome (20,27).
In the tomato plant, centromeric constitutive heterochromatin zones, are
present in every one of the 12 chromosomes. The banding patterns of such
centromeric zones, the ratio between the lengths of the two arms of the
chromosomes, and the length of distal euchromatin are used for tomato
chromosome identification (1,22). The chromosome map and linkage map for
maize and tomato have been worked out thoroughly, and extensive prac-
tical application of this knowledge has been made.

Beans ( Phaseol us vul gari s and Phaseol us coccineus) are important
field crops throughout the world. However, their chromosome maps have
not been established. Because of the unavailability of a chromosome map,
a complete genetic linkage map would be difficult to construct through
the use of aneuploid stocks.

59

60

A bean chromosome map has not been developed previously because
the metaphase chromosomes are too small to be identified and the pachy-
tene chromosomes are difficult to stain. Pachytene chromosomes of these
species are very long and overlap each other extensively in a flattened
PMC. Thus under conventional staining technique, their morphology cannot
be studied. The Giemsa banding pattern of mitotic prophase chromosomes
proposed by Mok and Mok (24) is difficult to use because the patterns
are so similar.

Several attempts to establish chromosome morphology in Phaseolus
aureus, Phaseolus mungo and tetraploid Phaseol us species in pachytene
and mitotic reproductive phases do not correlate with each other (14).

The chromosome condensation from the pachytene stage to metaphase in
tomato was analyzed carefully by Brown in 1949 (2). He observed that
the heterochromatic zones of pachytene chromosomes condense much less
than euchromatic zones during chromosome condensation and that euchro-
matic zones contribute very little to the chromosome length at metaphase.
These observations are still in close agreement with the up-to-date
model of chromosome structure presented by Comings (5). The discrepan-
cies between mitotic and meiotic chromosome maps in P. aureus cannot be
corrected using Brown's condensation principles (2). The trouble lies
in the technical difficulties of studying bean pachytene and metaphase
stages. The photomicrographs of P. aureus pachytene cells do not show
distinct centromeric heterochromatin zones similar to those found in
tomato pachytene (1). On the contrary, the chromatic bodies are well
distributed along the chromosome arms which are neglected as euchromatin.
Therefore, the metaphase map showed too much heterochromatin complement
as compared with the pachytene map.

61

Besides the difficulties described above, a quantitative error
in pachytene measurement is produced by camera lucida drawings which
neglect the tilting and folding-under effects. The pachytene chromo-
somes are so entangled that it is difficult to obtain a precise measure-
ment of chromosome length.

The metaphase chromosomes of Phaseolus are very small. Slight
tilting can give rise to considerable measurement error. In this stage
no banding pattern can be observed for chromosome identification.

The above considerations led to the idea of chromosome identi-
fication in the diplotene stage in beans. Because the chromosomes in
this stage are shorter than those of pachytene stage, most of them are
free from overlapping. They also have characteristic banding patterns
for identification. The two chromosomes are paired but not synapsed.

The presence of two slightly separated homologues increases the contrast
and reliability of the banding patterns.

In this work a new staining procedure was developed to study the
morphology and banding patterns of the diplotene stage chromosomes of
Phaseolus coccineus, which is a closely related species of the common
bean (Phaseolus vulgaris).

LITERATURE REVIEW

The subject of chromosome structure and banding patterns has
been reviewed by Comings (5). According to his paper, there are three
kinds of banding structures in a chromosome:

1. Centromeric heterochromatin is a dark staining zone located
on both sides of the centromeres. In tomato pachytenes, this structure
is seen in all 12 chromosomes (1,22).

2. Intercalary heterochromatin is distributed along the chromo-
some arms. These segments are observed as distinct dark staining bodies
in pachytene chromosomes.

3. Euchromatin is weakly stained or non-stainable zones in the
chromosome arms.

Comings also presented the model of chromatin packing in mamma-
lian chromosomes based on current knowledge from DNA research. The model
describes the structure of small and large chromomeres, banding bodies,
chromosome spiral ization, and condensed chromosome morphology.

According to this author, the traditional cytogenetic stains
generally give G band patterns on the chromosomes. In the staining
process, the alcohol and acetic acid fixation of dividing cells removes
protein molecules attached to the chromosome and increases the binding
of dye to the chromosome. The distinct chromatic zones observed in
prophase chromosmes are directly correlated to the tightly packed heter-
ochromatin bodies which contain large amounts of DNA and nucleoprotein

62

63

to bind large quantities of dye. Euchromatic zones are composed of
loosely packed chromatin containing a small amount of genetic material.

Barton (1) established definitively the pachytene morphology
of the tomato genome by camera lucida drawings. He identified each
chromosome by total length, short arm length, long arm length, arm
ratio and the chromatic and achromatic regions on both short and long
arms. Since camera lucida drawings record a three dimensional object
(a cell in pachytene stage) as a two dimensional diagram, the length of
the chromatin involved in a third dimension (depth) is either reduced
or lost completely. Therefore, pachytene chromosome maps made by dif-
ferent observers always present discrepancies. However, the centro-
meric heterochromatins of tomato pachytene chromosomes are so distinct
that there is no difficulty in identifying the different chromosomes.

Menzel (22) failed to detect any structural difference in
pachytene between Lycopersicon esculentum and Sol anum lycopersicoides
by comparing pachytene banding patterns of parental and interspecific
F-j hybrid plants. Nevertheless, the amphidiploid data showed that a
great portion of pollen sterility in the interspecific F-| was due to
chromosome structural differences. This paper clearly demonstrates the
difficulty in interpreting pachytene configurations. In spite of the
presence of many pachytene loops, she could not distinguish whether
the loops were the result of an inversion, an addition, a deficiency, or
just asynapsis. Two pachytene segments with the same length can appear
totally unequal when they are arranged in loops with tilt orientation.
Thus, despite the fact that pachytene analysis is useful for chromosome
identification in tomato and other species, its precision in detecting

64

structural differences is inferior to analysis of anaphase and other
meiotic stages.

Krishnan and De (7,14,15) tried to establish the morphology of
individual chromosomes of Phaseol us species at pachytene and metaphase
stages in order to study chromosome differentiation among the species
of this genus. Unfortunately, the pachytene configurations found in
P. aureus do not have distinct centromeric heterochromatin regions for
identification. The chromatic bands in F\_ aureus pachytenes are more
or less distributed along the chromosomes. The centromeric regions do
not have a distinctive morphology like those in tomato pachytene. There-
fore, the chromosome model built according to pachytene observations
does not correspond with the model derived from metaphase cells. In
one paper, they were unable to identify four of the chromosomes (7).
However, their pachytene studies led them to conclude there were chromo-
some aberrations such as translocations, duplications, and deficiencies
distinguishing P^_ aureus and P^ mungo.

The Giemsa banding model of bean prophase chromosomes proposed
by Mok and Mok (24) gave less detail and led to more ambiguity than
pachytene morphology. The chromosomes are too short and the banding too
broad. The short chromosomes simply provide too few differentiating
regions for identification. There is no possibility to study structural
differences between bean genomes with this banding technique.

Brown (2) studied the condensation process of tomato chromosome
#2, the nucleolar chromosome, from pachytene to metaphase. He concluded
that the heterochromatin zones at both sides of centromere at pachytene
contribute to most of the chromosome length at metaphase. The longer

65

euchromatic distal regions make a negligible contribution to chromo-
some length at this stage. He attributed all the condensation to
euchromatin packing, while the heterochromatin bodies do not contract
at all. This hypothesis is in close agreement with Comings' model
of chromosome packing (5).

MATERIAL AND METHODS

The bud samples from scarlet runner beans (P. cocci neus) used
in this study were collected at the beginning of the flowering period.
The flower buds were harvested at 9:30 a.m. for several days in May
1979. The buds were immersed immediately in Carnoy's fixative solution
containing 6:3:1 of ethanol (95%) :chloroform:glacial acetic acid by
volume and treated for 24 hours. They were transferred to 70% alcohol
for preservation.

Anthers were extracted from selected flower buds which ranged
in length from 2.5 to 3.0 mm, using a stereo dissecting microscope.
About 50 anthers were put in the well of a porcelain dish for staining.
Three drops (0.5 ml) of 45% iron propionic carmine were added to stain
the PMCs within the anthers for 1.0 to 1.5 hours. During this period,
the stain dries to 1/3 of the original volume. The anthers were then
transferred to a slide for squashing, using an ultrafine dissecting
tweezer (A. Dumont & Fils #5). Temporary slides were sealed with Kerr
Sticky Wax. The slides were good for one week. After that period, the
cells began to shrink and debris accumulated around the cell surface.
Therefore, all the photomicrographs were taken during the first 3 days
after slide preparation, using a phase contrast system under 1000 X
magnification. High contrast black and white films, Panatomic X with
ASA 32, and color slide film, Kodakchrome with ASA 25, were used for
photomicrographs. The slides were searched for diplotene cells having

66

67

chromosomes with favorable orientation. The morphology of the 11 chro-
mosomes was studied from at least five photographs of the same chromo-
some in the most favorable orientation in five different diplotene cells.
When several pairs of diplotene chromosomes were present in favorable
orientation in one cell, their relative lengths were measured in order
to arrange the 11 chromosomes in approximate rank order from longest
to shortest. For measuring the relative lengths of the chromosomes, the
color slides were projected to a screen 15 feet from the projector.

The apparent length of each chromosome was measured, and from these data
the relative lengths of the 11 chromosomes were calculated. Finally, an
idiogram was built according to the relative length and specific banding
patterns of each chromosome. Numbers were assigned to the 11 chromo-
somes in the rank order, 1 being the longest and 11 being the shortest.

RESULTS AND DISCUSSION

Overlapped and entangled chromosomes in the prophase stage
have been the major obstacle in chromosome identification in bean and
other species. The problem was partially solved by searching for
large PMCs which provide more space for chromosome distribution, reduc-
ing overlapping and entanglement. From searching many meiotic slides for
PMCs, it was learned that there are considerable differences in the
size of PMCs within a plant. The large PMCs can be twice as big as
small PMCs. The large PMCs are produced almost exclusively at the
beginning of flowering. The PMCs collected during the middle or late
parts of the flowering period were usually small. Therefore, the mate-
rials for prophase chromosome identification were all harvested before
and during anthesis as early in the flowering stage as possible.

Another problem encountered in prophase chromosome identification
is the difficulty in obtaining a high contrast between the chromosome
and the cytoplasmic background. It was discovered that the fixation
process has important effects on the contrast problem. In order to have
lightly stained cytoplasm after staining, the flower buds should be
thoroughly bleached during the fixation process by using an excessive
quantity of fixative. When only a small amount of fixative is used,
the fixative solution becomes very dark with the breakdown of chlor-
ophyll from the bud clusters. The dark solution tends to darken the
cytoplasm and reduces the contrast between the cytoplasm and chromosomes.

68

69

In order to obtain PMCs with clear cytoplasm, at least 100 ml of fixa-
tive should be used for the fixation of 50 bud clusters.

Besides the cell size and fixation problems, the conventional
chromosome staining procedure was found to be inadequate for staining
prophase chromosomes in bean for identification. The standard staining
solution with 45% iron propionic carmine was adequate for staining
chromosomes at diakinesis, metaphase and anaphase. Unfortunately, it
did not stain prophase chromosomes satisfactorily. To find out if the
inadequate staining of prophase chromosomes was due to stain concen-
tration, an experiment was performed, using three treatments:

1. Constant concentration of 45% iron propionic carmine in
which the anthers were immersed in a porcelain well sealed with a cover-
glass and wet cheese cloth for 2 hours.

2. Small concentration gradient in which the excessive stain
(1.5 ml) was put in a porcelain well with anthers for 1.5 hours without
covering the well.

3. A steep concentration gradient in which 3 drops (0.5 ml) of
the stain were put in the well with anthers for 1.5 hours without cover-
i ng the wel 1 .

After squashing anthers from each treatment and mounting the
slides, the darkly stained prophase chromosomes with high contrast were
found only in the third treatment. The leptotene, pachytene and diplo-
tene chromosomes with their characteristic dark banding patterns could
be seen distinctly against the pale pink cytoplasm. The PMCs from the
first two treatments did not show clear chromosome threads. Therefore,
it was concluded that a steep concentration gradient is necessary for

70

the carmine to stain prophase chromosomes from ieptotene to diplotene
stage. The technique was adopted to study the diplotene morphology in
this work and to complete the photomicrographic illustration of the
meiotic sequence in Part I. This new staining technique is adequate
for staining meiotic chromosomes at all the stages from interphase to
pollen mitosis. Therefore, it increases the number of observable PMCs
in each slide.

The banding patterns in diplotene are very specific for each
chromosome. Using the new staining procedure, the chromosomes show
dark heterochromatic bands and clear euchromatic zones. The hetero-
chromatic zones are different in thickness along the chromosome arms.
Therefore, by the different combination of thickness and band length, it
is possible to distinguish one chromosome from another. The relative
length is useful to separate the long chromosomes from the short chro-
mosomes, although this character is not sufficiently accurate for
identifying chromosomes of similar length. It is fortunate that no two
diplotene chromosomes possess similar banding patterns, which might
confound the identification work. The individual chromosome morphology
and its banding pattern were unique for each chromosome. -

For each chromosome (Figs. 62 through 72) there is presented
a series of photomicrographs, A. B, C. etc., showing only that chromo-
some. Each of the lettered photomicrographs has a drawing directly
below it. These photos were cut out of a print originally including an
entire diplotene PMC. At the bottom of each figure (Figs. 62 through
72) there is an idiogram for that chromosome. None of the idiograms has
a centromere because it was impossible to locate the centromere position

71

in any of the chromosomes in the diplotene PMCs studied. The figures
are presented in the order of chromosome length, #1 being the longest.
Chromosomes #2 through #6 are indistinguishable in length but are
shorter than #1. Chromosomes #7 and #8 are similar in length but are
shorter than #2 through #6. Chromosomes #9, #10, and #11 are each
shorter than the previous ones, and as a group they are distinctly
shorter than chromosomes #1 through #8.

72

4* V", 1A

IB

* " 1C

r _ *

. ID

*-•

V

1 E

IF

friy ,t<e '

iV- -• '

Figure 62. Chromosome #1 of Phased us cocci neus in the diplotene
stage of meiosis.

Chromosome #1 (Fig. 62) is the longest diplotene chromosome
in most of the PMCs. It is easily identified by its long achromatic
arm extending about half of the chromosome length with two lightly
stained bodies separated by equal distances and a dark telomere.

The other half is darkly banded. In this region three major bands are
present. The longest band is located near the achromatic half, the
second band in the terminal region, and a minor band between the two
longer bands. Thus, this chromosome is half dark and half achromatic
with a dark telomere at the achromatic end.

73

Figure 63. Chromosome #2 of Phaseolus cocci neus in the diplotene stage
of meiosis.

Chromosone #2 (Fig. 63) is as long as chromosome #1. It has
three bands nearly equal in length, two of them at the ends and one
in the middle region of the chromosome. Each band occupies slightly
less than 1/3 of the chromosome. They are all composed of four dark-
staining bodies (Fig. 63-C). The dark-staining bodies are very dark
in the band at one end, less dark in the middle band, and light in the
band at the other end. Two achromatic zones separated the three bands.

* '<a^ •

3B

. / *

V" 3D

Figure 64. Chromosome #3 of Phaseolus coccineus in the diplotene stage
of meiosi s .

Chromosome #3 (Fig. 64) has a dark telomere at each end, one is
more distinct than the other. An achromatic zone located in the middle
region occupies nearly 1/5 of chromosome length in which two lightly
stained bodies sometimes appear. Near the darker telomere, a band is
constituted of 3 dark chromatic bodies and a light chromatic body. The
other band near the lighter telomere is about 1/3 longer than the previ-
ous band and is composed of four dark chromatic bodies and two lighter

chromatic bodies.

75

Figure 65. Chromosome #4 of Phaseolus cocci neus in the diplotene stage
of meiosis.

Chromosome #4 (Fig. 65) has two dark bands located in the middle
region of the chromosome. The band located at the center of the chro-
mosome is longer than the submedian band. Near this shorter band,
there is an achromatic region as long as this band. A dark telomere
appears at the end of this achormatic zone. At the other arm beyond
the long band, it is all achromatic with three to four lightly stained
chromatic bodies spread along the arm. One of these bodies is a telomere.

76

Figure 66. Chromosome #5 of Phaseolus coccineus in the diplotene stage
of meiosis.

Chromosome #5 (Fig. 66) has a very long achromatic region in
the middle of the chromosome occupying 3/5 of the chromosome length.

Two dark bands, one longer than the other, are located at both ends.

This is a chromosome contrary to the theory of centromeric heterochro-
matin reported by Krishnan and De (15) in Phaseolus aureus and by Barton
(1) in tomato. In this chromosome, the whole central region, where the
centromere is supposed to be located, is all achromatic, with 5 to 6
inconspicuous bodies distributed along this region. This chromosome
is frequently misinterpreted as two separate chromosomes, sometimes
as late as the early diakinesis stage.

77

Figure 67. Chromosome #6 of Phaseolus cocci neus in the diplotene stage
of meiosis.

Chromosome #6 (Fig. 67) is also as long as the previous chromo-
somes, #2 through #5. It contains the nucleolar organizer near one end
At the opposite end, a very dark telomere is present. Near this telo-
mere there is a long dark band. Two smaller dark bands are located in
the long arm near the nucleolus. The short arm does not have a dark
telomere. Only two small bands are located near this achromatic telo-
mere. This nucleolar chromosome is the easiest among the H chromosomes
to be identified. Its morphology is also the most difficult to observe
because of the nuceolar mass.

78

*

• ’

7D

\

V-'

I

Figure 68. Chromosome #7 of Phaseolus coccineus in the diplotene stage
of meiosis .

Chromosome #7 (Fig. 68) is apparently shorter than the previous
chromosomes. The chromatic bands are all located in one arm occupying
less than half of the chromosome. The banding region is composed of
three continuous chromatic lobes, and the telomere is a lighter lobe.
More than half of this chromosome is achromatic with a lightly stained
telomere. A few inconspicuous bodies are distributed along the
achromatic region.

79

8 A

• r «

►>* ...<t £

' 8D

. *

v

Figure 69. Chromosome #8 of Phaseolus coccineus in the diplotene stage
of meiosis.

Chromosome #8 (Fig. 69) is as long as chromosome #7. It con-
tains four segments of dark banding, two at the ends and two around
the center. These two bands are different in length. The three
achromatic segments that separate the banding regions into four sec-
tions are sometimes not very distinct (Fig. 69-B). The two telomeres,
which are part of the terminal bands, are very distinct.

80

9 A

Figure 70. Chromosome #9 of Phased us coccineus in the diplotene stage
of meiosis.

Chromosome #9 (Fig. 70) has one band at each end. The longer
band occupies about 7/10 of the chromosome and the shorter band occu-
pies 2/10. The achromatic region that separates the two bands is
about 1/10 of the chromosome length. The long band is composed
of at least four chromatic lobes. The one near the achromatic zone is
sometimes separated from the other three lobes (Fig. 70-A).

81

10 A

«-v

¥ <. 0 t

«V

10 B

10 D

*V

10 F

«f9

Figure 71. Chromosome #10 of Phaseolus coccineus in the diplotene stage
of meiosis .

Chromosome #10 (Fig. 71) is a short chromosome containing three
chromatic bands separated by two achromatic zones. The three bands
are different in length. The longest band is located at one end and
the shortest band at the other end. The middle band is intermediate
in length. The approximate length ratio of these three bands is 3:2:1.

82

11 A

o<£>

. HB

* 1 \C

v

*•

UD

Figure 72. Chromosome #11 of Phaseolus cocci neus in the diplotene stage
of meiosis.

Chromosome #11 (Fig. 72) has four dark chromatic bodies at one
end which are sometimes connected into two similar lobes. The band
region occupies about half of the chromosome. The other half of the
chromosome is composed of a thin achromatic thread with a distinct
telomere at the end. A small nucleolar-like body is frequently asso-
ciated with the chromatic end of this chromosome.

83

Figure 73. A Phaseolus coccineus PMC at the diplotene stage of meiosis.
Nine chromosomes are identified by number. Chromosomes #8 and

#9 are not identifiable in this PMC.

84

Attempts to match the present diplotene model with Mok and
Mok's somatic prophase model (24) failed because of a lack of structural
similarity. Therefore, a new numeric order similar to that used in
maize and tomato was adopted for this model. Among the 11 chromosomes
of Mok and Mok's model, four of them have an overall banding pattern
similar to the present model although their relative lengths are dif-
ferent. They are: chromosome F, similar to chromosome #9; chromosome

H, similar to chromosome #5; chromosome J, similar to chromosome #11;
and chromosome K, similar to chromosome #4. In order to differentiate
all the chromosomes from all others, a stage with abundant morphological
detail is necessary. Mok and Mok's model is ambiguous because few dif-
ferentiated regions are evident in somatic prophase.

The morphology of the 11 chromosomes of Phaseolus cocci neus
presented above is useful for chromosome identification in translocation
and aneuploid studies. It may not be an effective tool for detecting
chromosome structural abberations unless the segment involved in the
abberation is very large and includes different banding regions.

CONCLUSION

The new staining procedure with iron propionic carmine shows
that diplotene chromosomes of Phaseolus cocci neus have unique banding
patterns which can be used for identification of each chromosome.
Their banding patterns are presented in idiograms and photomicro-
graphs in this paper. Their potential for application in other cyto-
genetic studies is discussed.

85

SUMMARY

The new staining procedure with 45% iron propionic carmine shows
that diplotene chromosomes of Phased us cocci neus have unique morphol-
ogy and banding patterns. The relative length of the chromosomes is
also useful for identification of each chromosome. The banding patterns
and relative lengths are presented in an idiogram and photomicrographs
in this paper. The chromosome models presented in this work will prob-
ably be most useful in identifying which chromosomes are involved in
trisomic and translocation heterozygotes, either directly or by elim-
ination of the chromosomes not involved in the anomalous configuration.

86

REFERENCES

1. Barton, D. W. 1950. Pachytene morphology of the tomato chromo-
some complement. Amer. J. Bot. 37:639-643.

2. Brown, S. W. 1949. The structure and meiotic behavior of the
differentiated chromosomes of tomato. Genetics 34:437-461.

3. Brown, W. V. 1972. Inversions p. 196-206. _In Textbook of
Cytogenetics. The C. V. Mosby Company, St. Louis.

4. Church, K.,and D. E. Wimber. 1971. Meiosis in Ornithogalum

vi rens (Lilliaceae) II. Univalent production by preprophase cold
treatment. Expt. Cell Res. 64:119-124.

5. Comings, D. E. 1978. Mechanisms of chromosome banding and impli-
cations for chromosome structure. Ann. Rev. Genet. 12:25-46.

6. Dana, S. 1966. Species cross between Phaseolus aureus Roxb and
Phaseolus trilobus Aid. Cytologia 31:176-187.

7. De, D. N., and R. Krishnan. 1966. Cytological studies of the
hybrid Phaseol us aureus x Phaseol us mungo. Genetica 37:588-600.

8. Dhaliwal, A. S., L. H. Pollard and A. P. Lorz. 1962. Cytological
behavior of an F-j species cross (Phaseolus 1 unatus L. var. Fordhook
Phaseolus polystachyus L.). Cytologia 27:369-473.

9. Honma, S. 1968. Inversion in the chromosomes of Phaseolus
vulgaris. Cytologia 33:78-81.

10. Hotta, Y., M. Ito and H. Stern. 1966. Synthesis of DNA during

meiosis. Proc. Nat. Acad. Sci . 56:1184-1191.

11. Innes, N. L. 1975. Plant Breeding p. 23-33. Rep. Natn. Veg.

Res. Stn. for 1974.

12. Joshi,S.,and S. S. Raghuvanski. 1967. Loss of nuclear capacity
to undergo division in certain PMCs of Allium. Cytologia 32:
421-425.

13. Kedar, N. (Kammermann), and W. P. Bemis. 1960. Hybridization
between two species of Phaseol us separated by physiological and
morphological blocks. Proc. Amer. Soc. Hort. Sci. 76:397-401.

87

88

14. Krishnan, R. , and D. N. De. 1965. Studies on pachytene and somatic
chromosomes of Phaseolus aureus. The Nucleus 8:7-16.

15. . 1970. Pachytene chromosomes and origin of a tetra-

ploid species of Phaseol us . Cytologia. 35:501-512.

16. Lamprecht, H. 1941. Die Artgranze zwischen Phaseolus vulgaris L.
und mul ti florus Lam. Hereditas 27:51-175. (English summary) .

17. . 1944. Die geni sch-pl asmatische Grundlage der Artbar-

riere. Agri Hort. Genet. 2:75-141.

18. . 1945. Intra- and interspecific genes. Agri Hort.

Genet. 3:45-60.

19. . 1948. Zur Ldsunq des Artproblems. Agri Hort. Genet.

6:87-141“

20. McClintock, B. 1929. Chromosome morphology in Zea mays. Science
69:629.

21. . 1938. The fusion of broken ends of sister half-

chromatids following chromatid breakage at meiotic anaphases.

Mo. Agr. Expt. Sta. Res. Bull. 290:1-48.

22. Menzel , M. 1962. Pachytene chromosomes of the intergeneric hybrid
Lycopersicon esculentum x Solanum lycopersi coi des . Amer. J. Bot.
49:605-615.

23. Moens, P. B. 1964. A new interpretation of meiotic prophase in
Lycopersicon esculentum (tomato). Chromosoma 15:231-242.

24. Mok, D. W. S., and M. C. Mok. 1976. A modified Giemsa technique
for identifying bean chromosomes. J. Hered. 67:187-188.

25. . 1977. Monosomies in common bean, Phaseolus vulgaris.

Theor. Appl . Genet. 49:145-149.

26. Morgan, D. T. , Jr. 1950. A cytogenetic study of inversion in
Zea mays. Genetics 35:153-171.

27. Rhoades, M. M. 1950. Meiosis in maize. J. Hered. 41:49-67.

28. , and Ellen Dempsey. 1953. Cytogenetic studies of

deficient-duplicate chromosomes derived from inversion hereto-
zygotes in maize. Amer. J. Bot. 40:405-424.

29. Sarbhoy, R. K. 1977. Cytogenetical studies in the Genus Phaseolus
Linn. III. Evolution in the genus Phaseolus. Cytologia 42:

401-413.

89

30. Smartt, J., and Nazmul Haq. 1972. Fertility and segregation of
the amphi diploid Phaseol us vulgaris L. x cocci neus L. and its
behavior in backcross. Euphytica 21:496-501.

31. Smith, L. 1941. An inversion, a' reciprocal translocation,
trisomics and tetraploids in Barley. J. Agr. Res. 63:741-750.

32. Wall,J.R. , and T. L. York. 1957. Inheritance of seedling
cotyledon position in Phaseolus species. J. Hered. 48:71-74.

33. Weinstein, A. I. 1926. Cytological studies on Phaseolus vulgaris.
Amer. J. Bot. 13:248-263.

34. Xolocotzi, E. F., S. M. Colin and C. Prywer. 1959. El origen de
Phaseolus cocci neus L. Darwinianus HDZ, X and Miranda C, subspecies
Nova. Revista de la Soc. Mexicana de Historia Natural 20: 99-121.

BIOGRAPHICAL SKETCH

Simon Suhwen Cheng was born February 4, 1944, in Tai-Chung,
Taiwan, Republic of China. He graduated from the First Provincial
High School in that city in 1962, and in 1967 he received a B.S.
degree in agronomy from the National Taiwan University, Taipei, Taiwan.
He emigrated to Brazil in 1968 and received an M.S. degree in horti-
culture from the Universidade Federal de Viqosa, Viqosa, Minas Gerais,
Brazil, in 1972. Currently, he is an Assistant Professor in the
Department of Agriculture in Escola Superior de Agricultura de Lavras,
Lavras, Minas Gerais, Brazil.

He married Elizabeth Ying Chu who gave him three children:
Henrique, Isabela and Spencer.

90

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.

TH . (R caaq —

Mark J. Bassett, Chairman

Associate Professor of Vegetable Crops

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.

Chesley B. Hall
Professor of Vegetable Crops

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.

Larkin C. Hannah

Associate Professor of Vegetable Crops

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.

Rex L. Smith
Professor of Agronomy

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of schlarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.

Kenneth H. Ouesenberry
Assistant Professor of Agronomy

This dissertation was submitted to the Graduate Faculty of the College
of Agriculture and to the Graduate Council and was accepted as partial
fulfillment of the requirements for the degree of Doctor of Philosophy.

August 1979

Dean, Graduate School