:f. :2. «Cr .

. lefl‘
.. ..

“3”ngan

 

yuan

a.
a

1-7 awe ru-

«at!

.1l.L
311...}.
.43.; v5? 3;:

up

 

S"- urn-q

nq

. . 4.41.”.qu .

-I(vull'b|t-[

 

 

 

it‘lfiizggts‘ LIBRARY
Q Michigan State
0200] g f University

 

 

This is to certify that the
dissertation entitled

THE ROLE OF ROOT GROWTH TRAITS IN
RESISTANCE TO THE BIOTIC STRESS, FUSARIUM
ROOT ROT AND THE ABlOTlC STRESS, LOW SOIL
PHOSPHORUS IN COMMON BEAN
(PHASEOLUS VULGARIS L.).

presented by

KAREN ANN CICHY

has been accepted towards fulfillment
of the requirements for the

 

 

 

 

Doctoral degree in Plant Breeding and Genetics
7 4/
/ Major Weéor’s Signature
[22,; / Z 200 C
Date

MSU is an Affirmative Action/Equal Opportunity Institution

 

—~—.—.- -.--.—.-..gJ---.—.-.-.---.-.--.-.-.-.-.-.-.-.—.-.-.-.-o-.-.---n--.-.-.-—.-.-.-.-o-.—-.—

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATEDUE

DAIEDUE

DAIEDUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 p:/CIRC/DateDue.indd-p.1

 

 

THE ROLE OF ROOT GROWTH TRAITS IN RESISTANCE TO THE
BIOTIC STRESS, FUSARIUM ROOT ROT AND THE ABIOTIC STRESS,
LOW SOIL PHOSPHORUS IN COMMON BEAN
(PHASEOLUS VULGARIS L.).

By

Karen Ann Cichy

A DISSERTATION
Submitted to
Michigan State University

In partial fulfillment of the requirements
For the degree of '

DOCTOR OF PHILOSOPHY
Plant Breeding and Genetics Program

2006

ABSTRACT

THE ROLE OF ROOT GROWTH TRAITS IN RESISTANCE TO THE

BIOTIC STRESS, FUSARIUM ROOT ROT AND THE ABIOTIC STRESS,
LOW SOIL PHOSPHORUS IN COMMON BEAN
(PHASEOLUS VULGARIS L.).
BY
Karen Ann Cichy
Genetic and environmental variability for root architecture has been
identified in common bean. The overall objective of this research was to
determine if root architecture traits are an adaptation against an abiotic and a
biotic stress. The stresses compared in this study were the fungal disease
Fusarium root rot caused by F usarium solam' f.sp. phaseolz' (Fsp) and phosphorus
deficient soils, both significantly reduce yield in major bean production regions.
Fusarium root rot is greatly influenced by environmental conditions that

stress plant roots, which has hindered progress in breeding resistant bean
cultivars. The first objective was to determine if Fusarium root rot resistance is
conferred by genesexpressed in the root or shoot and if root architecture plays a
role in resistance. Reciprocal grafting of a resistant (FR266) and susceptible
(Montcalm) bean cultivar was used to study resistance to Fsp in two root grth
environments, one without additional stress (Experiment 1) and one with the
additional stress of a compacted soil layer (Experiment 2). Reciprocal grafting
revealed that root rot incidence was controlled by genes expressed only in the
roots in Experiment 1. Variability for root architecture was present, but it did not

affect root rot incidence. In Experiment 2, root and shoot genotype both dictated

root rot incidence and root traits including root length and root dry weight
appeared to be related to root rot incidence.

Root architecture traits have been shown to be related to P uptake and
influence tolerance to low P soils. The second objective was to study the
relationship between root architecture and P uptake in an Andean recombinant
inbred line (RIL) population developed from a cross between a low P tolerant
(G19833) and susceptible (AND696) bean line. The RILS also varied in plant
growth habit and this was examined in relation to root architecture and yield. The
population was field- grown under both high and low P levels and root architecture
traits, P uptake and seed yield were measured. A linkage map was developed and
a QTL study was conducted to determine which regions of the genome controlled
these traits. Two QTL for root length density were identified that explained
nearly 40% of the phenotypic variation, but root traits were not important for low
P tolerance as measured by P uptake and seed yield. Growth habit influenced
yield differently across soil treatments and indeterminate RILS had higher root
length density than determinate RILS in the high P treatment.

The third objective was to examine the relationship between tolerance to
low P soil and seed P, Fe, and Zn levels and identify QTL controlling these traits
in the population. Iron and Zn are important to human nutrition and interact with
P stored as phytic acid. Variability for seed Fe and Zn levels related to seed P was
detected and QTL for these traits co-localized to linkage group B01 near the

region of the gene that controls determinate plant grth habit.

ACKNOWLEDGEMENTS

I would like to thank my advisor Dr. Sieg Snapp, for her support and interest in my
research. I was fortunate to be part of her lab group, where I gained a holistic view
of agriculture science. I would also like to thank my co-advisor, Dr. James Kelly, his
impeccable knowledge of plant breeding and common bean improvement was
inspiring to me and what I have learned from him has helped me become a better
scientist. I am grateful to the members of my guidance committee: Dr. Mitch
McGrath, for his support in terms of openness to new research ideas and generous
sharing of research equipment, and Dr. Willie Kirk, who helped direct and focus my
research through helpfiJl conversations.

I am very grateful to Dr. Matthew Blair at the International Center for Tropical
Agriculture. He allowed me to conduct research at CIAT and also provided genetic
and laboratory materials that made it possible for me to address thoroughly 3
complex research question. He was also a great mentor.

I would like to thank the members of Dr. Snapp’s lab group, including Kitty Oneil,
Rich Price, and Bill Quackenbush who where a great help in experimental
preparation and were also great to work with. Thanks also to the members of Dr.
Kelly’s lab for their help, including Evan Wright, Halima Awalie, Belinda Roman,
Veronica Vallejo, Jen Wagner and Karolyn Terpestra. I would like to thank the .
many people at CIAT who helped with the field and laboratory work presented here,

_ especially Gina Caldas and Carlos Galeano.

iv

I would like to thank Dr. George Hosfield for being an excellent mentor during my
graduate studies.

I would like to thank my parents and my brothers and sister for their support and
understanding. I am also thankful to my husband, David Mota-Sanchez for his faith
and support. I would also like to thank my daughters Sofia and Liliana for being
wonderful. I would also like to thank the many fi'iends I made while at Michigan

State University who created a wonderfiil environment in which to work and live.

TABLE OF CONTENTS

LIST OF TABLES ............................................................................ xii
LIST OF FIGURES .......................................................... '. ................ xvii

CHAPTER 1: LITERATURE REVIEW: BREEDING A BETTER ROOT

SYSTEM ......................................................................................... I
Introduction .................................................................................. 1
Abiotic Stress: Phosphorus Deficiency .................................................. 3

Plant adaptation to low P soil ..................................................... 5
Phosphorus acquisition efficiency ....................................... 6

Root system architecture ......................................... 7

Root hairs ........................................................... 9

Mycorrhizae ...................................................... 1 1

Root exudates ..................................................... 11

Phosphorus utilization efficiency ....................................... 13
summary ........................................................................... 15
Biotic Stress: Fusarium Root Rot ........................................................ 16
Nature of Disease .................................................................. 17
Environmental Factors ............................................................ 1 8
SoilTemperature..........................................................19
Drought .................................................................... 20
CompactiOn ............................................................... 20
Methods of Control ............................................................... 22

vi

Genetics ............................................................................. 24

Sources of Resistance ................................................... 24
Inheritance of Resistance ................................................ 25
Root Architecture and Disease .................................................. 26
Root and Shoot Growth Habit Coordination ......................... 28
Summary ........................................................................... 29
References .......................................................................... 30
Figures .............................................................................. 40

CHAPTER 2: FUSARIUM ROOT ROT INCIDENCE AND ROOT SYSTEM

ARCHITECTURE IN GRAFTED COMMON BEAN LINES ............................. 43
Abstract ............................................................................. 43
Introduction ........................................................................ 44
Materials and Methods ............................................................ 48

Plant Materials ............................................................ 48
Reciprocal Grafting ...................................................... 48
Experiments .............................. ' ................................. SO
Experiment 1 Non-compacted soil ............................ 50

Experiment 2: Compacted soil layer .......................... 52

Plant Measurements ..................................................... 54

Data Analysis ............................................................. 55
Results .............................................................................. 55
Experiment 1 .............................................................. 55
Experiment 2 .............................................................. 57

vii

Discussion .......................................................................... 58

Conclusions ........................................................................ 60
References........._. ................................................................ 61
Tables ................................................................................ 64
Figures .............................................................................. 72

CHAPTER 3: THE RELATIONSHIP AMONG ROOT ARCHITECTURE

TRAITS, PLANT GROWTH HABIT AND TOLERANCE TO LOW SOIL

PHOSPHORUS LEVELS IN AN ANDEAN BEAN POPULATION .............. 73
Abstract .............................................................................. 73
Introduction ................ ......... 74
Materials and Methods ............................................................ 77
Results ............................................................................... 80

P uptake .................................... ' ................................ 80
P use efficiency ........................................................... 80
Seed P content ............................................................. 81
Yield ........................................................................ 81
Correlations ................................................................ 81
Root traits and P uptake ................................................. 82
Differences by Growth Habit ........................... A ................ 83
Discussion .......................................................................... 85
Conclusions ........................................................................ 87
References. ................................ ......................................... 89
Tables ................................................................................ 92

viii

Figures ........................................................................... 101

CHAPTER 4: IDENTIFICATION OF QTL RELATED TO ROOT

ARCHITECTURE TRAITS AND LOW PHOSPHORUS TOLERANCE IN AN

ANDEAN BEAN POPULATION ..................................................... 107
Abstract ........................................................................... 107
Introduction ..... 108
Materials and Methods ......................................................... 111

Plant Material ............................................................ 111
Field Trials ............................................................... 111
DNA Isolation and Molecular Marker Analysis .................... 112
SSR Markers .................................................... 1 13
RAPD Markers ................................................. 113
AF LP Markers ................................................. 114
Genotypic Data Analysis .............................................. 115
Results and Discussion ......................................................... 116
Linkage Map ............................................................. 1 l6
QTL Identification ................. .. .................................... 117
Root Architecture ............................................... 1 17
P uptake ......................................................... 118
Seed Yield ...................................... ' ................. 119
P use efficiency ................................................. 120
Seed P content ................................................. 121
Conclusions ...................................................................... 121

ix

References ........................................................................ 123
Tables .............................................................................. 126
Figures .............................................................................. 127
CHAPTER 5: QTL ANALYSIS OF SEED NUTRIENT LEVELS IN AN

ANDEAN BEAN POPULATION SEGREGATING FOR TOLERANCE TO

LOW PHOSPHORUS SOILS .......................................................... 137
Abstract ........................................................................... 137
Introduction ...................................................................... l 3 8
Materials and Methods ..................................................... ....140
Results ............................................................................. 143

Variability ............................................................... 139 '
Seed Weight ..................................................... 139
Seed P and Phytic acid ........................................ 144
Seed Fe and Zn ................................................. 144
Correlations .............................................................. 145
Growth Habit ............................................................ 145
QTL Identification.......................... ........................... 146
Seed Weight ..................................................... 146
Seed P and Phytic acid ........................................ 147
Seed Fe and Zn ................................................. 147
Discussion ........................................................................ 148
Conclusions .................................................................... ' ...150
References ........................................................................ 151

Tables .............................................................................. 155
Figures ............................................................................. 162

APPENDIX ................................................................................ 170

xi

LIST OF TABLES

CHAPTER 2:

Table 1. Selected plant and seed characteristics of the common bean cultivars
FR266 and Montcalm ......................................................................... 64

Table 2. Mean response root rot score, shoot dry weight, and root dry weight of
different graft combinations of common bean lines FR266 and Montcalm in the
presence (+ Fus) or absence (-FuS) of F usarium solani f.sp phaseoli inoculum. Data
analyzed across 2 runs of experiment 1 ..................................................... 65

Table 3: Mean response for total root length and average root diameter of ungrafted
Montcalm and FR266 in the presence (+ Fus) or absence (—FuS) of F usarium solani
f.sp phaseolz' inoculum. Data analyzed across 2 runs of experiment 1 .................. 66

Table 4: Mean response and analysis of variance for root traits of different graft
combinations of common bean lines FR266 and Montcalm in the presence of

F usarium solani f.sp phaseolz’ inoculum. Data analyzed across 2 runs of experiment
1 and the category graft combination is separated into root and shoot components for
analysis to determine which components are affecting a response ...................... 67

Table 5: Mean response for root traits of different graft combinations of common
bean lines FR266 and Montcalm in the presence of F usarium solam’ f.sp phaseolz'
inoculum. Data analyzed across 2 runs of experiment 1 and the category graft
combination is separated into root and shoot components for analysis to determine
which components are affecting a response, means are averaged by root genotype
and shoot genotype ............................................................................. 68

Table 6. Mean response of root rot score, shoot dry weight, and root dry weight of
different graft combinations of common bean lines FR266 and Montcalm grown in
containers with a layer of compacted soil in the presence (+ Fus) or absence ( Fus) of
Fusarium solani f. sp phaseoli inoculum, Data analyzed across 1 run (Nov. 2005) of
experiment 2 .................................................................................... 69

Table 7: Mean response for total root length and average root diameter of ungrafted
Montcalm and FR266 grown in containers with a layer of compacted soil in the

presence .(+ Fus) or absence (-Fus) of Fusarium solam‘ f.sp phaseoli inoculum, Data
analyzed across 2 runs of experiment 2 .................................................... 70

Table 8. Mean response and analysis of variance for root traits of different graft
combinations of common bean lines FR266 and Montcahn grown in containers with .

xii

a layer of compacted soil in the presence of F usarz'um solam' f.sp phaseoli inoculum.
Data analyzed across 1 run (Nov. 2005) of experiment 2 and the category graft
combination is separated into root and shoot components for analysis to determine
which components are affecting a response ............................................... 71

CHAPTER 3:

Table 1. Analysis of variance of traits related to phosphorus uptake and use, and seed
yield in a population of 75 recombinant inbred lines from the AND696/G19833
population field grown in Darien, Colombia in 2000 and 2005 under two treatments:
high and low soil phosphorus ................................................................ 92

Table 2. Mean plant growth traits in high (HP) and low (LP) phosphorus soils
conditions for parents AND696 and G19833 and the means and ranges of 75
recombinant inbred lines (RILS) developed from the parents. Means are also
included for the check variety, Carioca. The experiment was planted in 2000 and
2005 in Darien, Colombia. Mean values are of 3 replications. P value indicates
level of significant genotypic differences among the RILS for each trait ............... 93

Table 3. Phenotypic correlations among P uptake, P use efficiency, and seed yield in
a population of 75 recombinant inbred lines from a AND696/G19833 cross grown in
high (HP) or low (LP) soil phosphorus in Darien, Colombia in 2005 .................. 94

Table 4. Analysis of variance of root traits in a population of 75 recombinant inbred
lines from the AND696/G19833 population field grown in Darien, Colombia in 2005
in two environments: high and low soil phosphorus ...................................... 95

Table 5. Mean root growth traits in high (HP) and low (LP) phosphorus soils
conditions for parents AND696 and G19833 and the means and ranges of 75

- recombinant inbred lines (RILS) developed from the parents. Means are also
included for the check variety, Carioca. The traits are from the 2005 field
experiment planted in Darien, Colombia. Mean values are of 3 replications. P value
indicates level of significant genotypic differences among the RILS for each

trait ............................................................................................... 96

Table 6. Phenotypic correlations between root traits and P uptake and seed yield in a
population of 75 recombinant inbred lines developed from a AND696/619833 cross
grown in high (HP) or low (LP) soil phosphorus in Darien, Colombia in 2005 ...... 97

Table 7. Means of root and shoot traits of 75 RILS developed from a

AND696/G19833 cross and grown under high and low soil phosphorus treatments in
Darien, Colombia in 2000 and 2005, grouped and averaged by plant growth habit of
which there were two categories, indeterminate (1nd.) and determinate (Det.) ....... 98

xiii

Table 8. Phenotypic correlations between root traits and P uptake and seed yield in a
population of 75 recombinant inbred lines (RILS) developed from a
AND696/G19833 cross grown under high or low soil phosphorus treatments in
Darien, Colombia in 2005. For the analysis, RILS were grouped according to plant
grth habit of which there were two categories, indeterminate (1nd.) and
determinate (Det.) .............................................................................. 99

Table 9. Phenotypic correlations between P uptake and seed yield and P seed
content and seed yield in a population of 75 recombinant inbred lines (RILS) from a
AND696/G19833 cross grown under high or low soil phosphorus treatments in
Darien, Colombia in 2000 and 2005. For the analysis, RILS were grouped according
to plant grth habit of which there were two categories, indeterminate (1nd.) and
determinate (Det.) ............................................................................. 100

CHAPTER 4:

Table 1. Putative QTL for seed traits identified from 75 recombinant inbred lines
developed from AND696/G19833 cross grown under high (HP) and low (LP) soil
phosphorus conditions in Darien, Colombia in 2000 and 2005 ......................... 126

CHAPTER 5:

Table 1: Analysis of variance of seed traits in a population of 75 recombinant inbred
lines from the AND696/G19833 population field grown in Darien, Colombia in 2000 '
and 2005 in two environments: high and low soil .

phosphorus ..................................................................................... 1 55

Table 2: Seed traits in high (HP) and low (LP) phosphorus soils conditions for
parents AND696 and G19833 and the means and ranges of 75 recombinant inbred
lines (RILS) developed from the parents. The experiment was planted in 2000 and
2005 in Darien, Colombia. Mean values presented (n=3). P value indicates level of
significant genotypic differences among the RILS for each traits ...................... 156

Table 3. Phenotypic correlations among seed traits in a population of 75 recombinant
inbred lines from the AND696/G19833 population field grown in Darien in 2005 in
high and low soil phosphorus. The values above the diagonal line are under low soil
P and the values below the diagonal line are under high soil P ......................... 157

Table 4. Phenotypic correlations among seed traits in a population of 75 recombinant
inbred lines developed fromAND696/G19833 and field grown in Darien in 2000 in
high and low soil phosphorus. The values above the diagonal line are under low soil
P and the values below the diagonal line are under high soil P ........................ 158

xiv

Table 5. Seed traits in a population of 75 recombinant inbred lines developed from
an AND696/G19833 cross and grown under high and low soil phosphorus conditions
in Darien, Colombia in 2000 and 2005, grouped and averaged by plant growth habit
of which there were two categories, indeterminate (1nd.) and determinate (Det.).
Mean values presented (n=3) ............................................................... 159

Table 6. Putative QTL for seed traits identified from 75 recombinant inbred lines
developed from an AND696/G19833 cross grown under high (HP) and low (LP) soil
phosphorus conditions in Darien, Colombia in 2000 and 2005. . . . . . . . .......... 160

APPENDIX:

Table A2-1. Mean response for root dry weight, root length, and average root
diameter in the top soil layer of different grafi combinations of common bean lines
FR266 and Montcalm grown in containers with a layer of compacted soil in the
presence (+ Fus) or absence (-F us) of F usarium solam’ f.sp phaseoli inoculum, Data
analyzed across 1 run (Nov. 2005) ofexperiment 2............. 171

Table A2-2. Mean response for root dry weight, root length, and average root
diameter in the middle soil layer of different graft combinations of common bean
lines FR266 and Montcalm grown in containers with a layer of compacted soil in the
presence (+ Fus) or absence (-Fus) of F usarium solam' f.sp phaseoli inoculum, Data
analyzed across 1 run (Nov. 2005) of experiment 2 ..................................... 172

Table A2-3. Mean response for root dry weight, root length, and average root
diameter in the bottom soil layer of different graft combinations of common bean
lines FR266 and Montcalm grown in containers with a layer of compacted soil in the
presence (+ Fus) or absence (-Fus) ofFusarium solani f.sp phaseoli inoculum, Data
analyzed across 1 run (Nov. 2005) of experiment 2 ..................................... 173

Table A3-1. Bartlett test for variance homogeneity based on growth habit, where
there were two classes, determinate and indeterminate, and 1 degree of freedom.
Tests were conducted for shoot and seed traits measured under low P (LP) and high
P (HP) treatments in the recombinant inbred line population developed from a cross
between common bean lines AND696 and G19833 .................................... 174

Table A3-2. Bartlett test for variance homogeneity based on grth habit, where
there were two classes, determinate and indeterminate, and 1 degree of freedom.
Tests were conducted for roots traits measured under low P (LP) and high P (HP)

treatments in the recombinant inbred line population developed from a cross between
common bean lines AND696 and 619833 ............................................... 175

Table A4-1. Amplified fragment length polymorphisms (AF LP) names and selective
primers used in developing a linkage map of AND696/G19833 ....................... 183

-XV

Table A4-2. Simple sequence repeat (SSR) primer sequence, linkage group and
source information for markers polymorphic between AND696 and G19833 and
used to develop a linkage map of AND696/G19833 .................................... 184

Table A4-3. Random Amplified Polymorphism DNA (RAPD) primer sequence for
markers polymorphic between AND696 and G19833 and used to develop a linkage
map of AND696/G19833 recombinant inbred line population ......................... 188

Table A4-4. Bartlett test for variance homogeneity based on growth habit, where
there were two classes, determinate and indeterminate, and 1 degree of freedom.
Tests were conducted for seed traits measured under low P (LP) and high P (HP)
treatments in the recombinant inbred line population developed from a cross between
common bean lines AND696 and G19833 ................................................ 189

xvi

LIST OF FIGURES

CHAPTER 1:

Figure 1. Responses of Arabidopsis root systems to supply of nutrients P, N, and S.
Figure reproduced from Lopes-Bucio et a1. (2003) ........................................ 40
Figure 2: The Phosphorus Cycle (Potash and Phosphate Institute, 2004). . . . . . . . ......41
Figure 3: Fusarium root rot infection in beans (Schwartz, 2006) .................... _. ..42
CHAPTER 2:

Figure 1. Photos of containers used in Experiment 1 (a) and Experiment 2 (b). The
bulk densities of soil are listed in g/cm3 for each of the 3 layers of containers used in
experiment 2 .................................................................................... 72

CHAPTER 3:

Figure 1. Frequency distribution of root length density and specific root length in 75
recombinant inbred lines (RILS) developed fiom a cross between AND696 and
G19833 planted in Darien, Colombia in 2005 in high (HP) and low (LP) phosphorus
soil. The distribution of RILS in LP is shown on the graphs on the left and on graphs
on the right show distribution under HP. Frequency distributions are separated by
soil P level with LP represented by graphs on the left and HP by graphs on the right.
Within each graph RILS are also separated by grth habit. Arrows represent mean
values of determinate and indeterminate RILS within each treatment. Means are the
average of 3 replications ...................................................................... 101

Figure 2. Frequency distribution of root surface area and average root diameter in 75
recombinant inbred lines (RILS) developed from a cross between AND696 and
G19833 planted in Darien, Colombia in 2005 in high (HP) and low (LP) phosphorus
soil. The distribution of RILS in LP is shown on the graphs on the left and on graphs
on the right show distribution under HP. Frequency distributions are separated by
soil P level with LP represented by graphs on the left and HP by graphs on the right.
Within each graph RILS are also separated by growth habit. Arrows represent mean
values of determinate and indeterminate RILS within each treatment. Means are the
average of 3 replications ...................................................................... 103

xvii

Figure 3. Regression of seed yield under low P by seed yield under high P in the
2005 growing season of the population in 75 recombinant inbred lines developed
from a cross between AND696 and G19833. Horizontal bar represents mean yield
under low P and vertical bar represents mean yield under high P. Note: G19833 yield
values are shown from 2000, not 2005, because seed of this parent was not planted in
2005 ............................................................................................ 105

Figure 4. Regression of seed yield under low P by seed yield under high P in the
2000 growing season of the population in 75 recombinant inbred lines developed
from a cross between AND696 and G19833. Horizontal bar represents mean yield
under low P and vertical bar represents mean yield under high P ...................... 106

CHAPTER 4:

Figure 1. Common bean linkage map of AND696 by 619833 developed from 75
recombinant inbred lines of the F57 population. The map contains RAPD, AF LP,
SSR molecular markers and one phenotypic marker for a total of 6330M on 12

linkage groups. Primer information for markers at each locus can be found in Tables
A4-1, A4-2, and A4—3 of the Appendix ................................................... 127

Figure 2. Frequency distribution of root traits in 75 recombinant inbred lines
developed from a cross between AND696 and G19833 planted in Darien, Colombia
in 2005 in high (HP) and low (LP) phosphorus soil. Arrows represent mean values
of parental genotypes under different treatments (G19833 was not grown in 2005).
Means are the average of 3 replications ................................................... 129

Figure 3: Frequency distribution of shoot phosphorus concentration and phosphorus ‘
uptake in 75 recombinant inbred lines developed from a cross between AND696 and
_ G19833 planted in Darien, Colombia in 2005 in high (HP) and low (LP) phosphorus
soil. Arrows represent mean values of parental genotypes under different treatments
(G19833 was not grown in 2005). Means are the average of 3 replications. . . . . ....131

Figure 4. Frequency distribution of seed phosphorus content and seed yield in 75
recombinant inbred lines developed from a cross between AND696 and G19833

' planted in Darien, Colombia in 2000 and 2005 in high (HP) and low (LP)
phosphorus soil. Arrows represent mean values of parental genotypes under
different treatments (G19833 was not grown in 2005 and AND696 was not grown in
2000). Means are the average of 3 replications .......................................... 132

Figure 5. Frequency distribution of phosphorus use efficiency (defined in this study

. as the amount of seed yield per unit of P taken up by the plant) in 75 recombinant
inbred lines developed from a cross between AND696 and G19833 planted in
Darien, Colombia in 2005 in high (HP) and low (LP) phosphorus soil. Arrows
represent mean values of parental genotypes under different treatments (G19833 was
not grown in 2005). Means are the average of 3 replications ......................... 134

xviii

Figure 6. AND696/G19833 linkage map with QTL locations for seed yield, seed P
content, root length density (RLD), root surface area (surfarea), shoot P
concentration, P uptake, and P use efficiency (PUB). QTL are fiirther identified by
treatment, high P (HP) and low P (LP) and year, 2000 and 2005 ..................... 135

CHAPTER 5:

Fig. 1. Frequency distributions for seed weight and seed P in 75 recombinant inbred
lines (RILS) fi'om the population AND696/G19833 grown in Darien Colombia in
2000 and 2005 under high (HP) and low (LP) soil phosphorus levels. Arrows
represent mean values of parental genotypes under different treatments (G19833 was
not grown in 2005 and AND696 was not grown in 2000). Values are the average of
3 replications. Graphs on the left are for 2005 and those on the right are for

2000 ......... . ................................................................................... 162

Fig. 2. Frequency distributions for seed phytic acid in 75 recombinant inbred lines
(RILS) from the population AND696/G19833 grown in Darien Colombia in 2005
under high (HP) and low (LP) soil phosphorus levels. Arrows represent mean values
of parental genotypes under different treatments (G19833 was not grown in 2005).
Values are the average of 3 replications ................................................... 164

Fig. 3. Frequency distributions for seed Fe and Zn in 75 recombinant inbred lines
(RILS) from the population AND696/G19833 grown in Darien Colombia in 2000
and 2005 under high (HP) and low (LP) soil phosphorus levels. Arrows represent
mean values of parental genotypes under different treatments (G19833 was not
grown in 2005 and AND696 was not grown in 2000). Values are the average of 3
replications. Graphs on the left are for 2005 and those on the right are for

2000 ............................................................................................. 165

Figure 4. AND696/G19833 linkage map with QTL locations for seed P, Fe, and Zn
concentration, percent phytic acid in the seed (PA), and seed weight (seedwt). QTL
are further identified by treatment, high P (HP) and low P (LP) and year, 2000 and
2005 ..... g ....................................................................................... 167

APPENDIX:

Figure A3-l. Frequency distribution of shoot P concentration, P uptake, and P use
efficiency of 75 recombinant inbred lines (RILS) developed from a cross between
AND696 and G19833 planted in Darien, Colombia in 2005 in high (HP) and low
(LP) phosphorus soil. The distribution of RILS in LP is shown on the graphs on the

xix

left and on graphs on the right show distribution under HP. Frequency distributions
are separated by soil P level with LP represented by graphs on the left and HP by
graphs on the right. Within each graph RILS are also separated by growth habit.
Arrows represent mean values of determinate and indeterminate RILS within each
treatment. Means are the average of 3 replications ..................................... 176

Figure A3-2. Frequency distribution of seed P content in 75 recombinant inbred
lines (RILS) developed from a cross between AND696 and G19833 planted in
Darien, Colombia in 2000 and 2005 in high (HP) and low (LP) phosphorus soil. The
distribution of RILS in LP is shown on the graphs on the left and on graphs on the
right Show distribution under HP. Frequency distributions are separated by soil P
level with LP represented by graphs on the left and HP by graphs on the right.
Within each graph RILS are also separated by growth habit. Arrows represent mean
values of determinate and indeterminate RILS within each treatment. Means are the
average of 3 replications ..................................................................... 179

Figure A3-3. Frequency distribution of seed yield in 75 recombinant inbred lines
(RILS) developed from a cross between AND696 and G19833 planted in Darien,
Colombia in 2000 and 2005 in high (HP) and low (LP) phosphorus soil. The
distribution of RILS in LP is shown on the graphs on the left and on graphs on the
right Show distribution under I—IP. Frequency distributions are separated by soil P
level with LP represented by graphs on the left and HP by graphs on the right.
Within each graph RILS are also separated by growth habit. Arrows represent mean
values of determinate and indeterminate RILS within each treatment. Means are the
average of 3 replications ..................................................................... 181

Chapter 1: Literature Review: Breeding a Better Root System

Introduction:

To understand how root systems function as an organ, it is useful to
classify roots based on type and distribution. Roots are often grouped into four
broad classes, the taproot, basal roots, adventitious roots, and lateral roots (Zobel,
1991). The taproot or the primary root is Of embryonic origin, basal roots
originate from the root/shoot transition zone, adventitious roots originate from
non—root tissue, and lateral roots originate from existing roots. Lateral roots can
also be further classified into secondary and tertiary roots depending branching
from the taproot (Fitter, 2002). There are some species specific specialized root
types such as the proteiod root Clusters, which are clusters of densely positioned
lateral roots, found in the Proteaceae family. Additionally, some species have
enlarged, starchy roots that are used as food crops, including carrot (Daucus
carota), beet (Beta vulgaris), and sweet potato (Ipomoea batatas) (Austin, 2002).

Architecture describes the spatial configuration of the root system (Lynch,
1995). Root system architecture is formed by intrinsic developmental and
response to environment type cues. Intrinsic mechanisms are those that are an
essential part of plant development and allow for the characteristic architecture of
a particular root system. Response mechanisms are pathways that determine how
plants respond to external signals (Malamay 2005). The ability of a plant’s root
system to respond to particular enviromnental conditions to improve the survival

in a specific enviromnent is know as plasticity (Sultan, 1987).

Natural variation exists between and within plant species for intrinsic and
response type mechanisms of root architecture development. For example, in a
recombinant inbred line population from two ecotypes of Arabidopsis, two QTL
were identified for lateral root length. One QTL acted intrinsically and the other
was responsive to mannitol concentration (Gerald et al. 2006). Arabidopsis has
been used successfully to identify root system architectural responses to nutrients
N, P, and S as seen in Figure 1.

Both intrinsically and response regulated root system architectural traits
can have a role in stress tolerance, and a growing number of studies measured
genetic variation for root architecture and related that to stress avoidance or yield
gains (Gerald et al., 2006; Laperche et al., 2006; F ita etial., 2006). A study of
cultivated (Lactuca sativa L.) and wild (L. serriola) lettuce found that a QTL for
taproot length co-localized with a QTL for moisture extracted fiom a deep soil
zone (50-100 cm) (Johnson et al., 2000). Regions of the maize genome that
explained genotypic variation for adventitious root mass co-localized with QTL
for grain yield under water stress and non-water stress conditions, such that an
increase in root mass was associated with an increase in yield at four separate
QTL (Tuberosa et al., 2002). Adaptive plasticity for root mass and root length
has also been observed in response to soil moisture, including drought and
flooding stress in two Polygonum species (Bell and Sultan 1999).

The scope of this review is to examine the literature closely for attempts at
breeding better root systems and relate where possible applications to common

bean improvement. Common bean (Phaseolus vulgaris L.) is an excellent

candidate for improved root system architecture. It is an important grain legume
often planted in marginal environments (Wortmann et al., 1998). Common bean is
also very susceptible to a number of stresses.

Common bean roots have a large root length and root length density as
compared to other annual legumes, soybean (Glycine max L.) and field pea
(Pisum sativum L.). Common bean was found to have a smaller median root
depth (0.46m) than soybean (0.56m) and field pea (0.48m) based on the average
of three years of data on a Haplustoll soil (Merrill et al., 2002). Within P.vulgaris,
genetic variability for root architecture traits exist and some of these traits have
been implicated in stress resistance. For example, basal root shallowness is
associated with bean genotypes tolerant to low phosphorus soils (Liao et al.,
2001). The remainder of this literature review will focus on two stresses
detrimental to bean production in many parts of the world, the abiotic stress,
phosphorus deficiency, and the biotic stress, Fusarium root rot. The nature Of each
of these stresses and root system traits were examined along with other factors
that may be useful for genetic control of these ailments.

Abiotic Stress: Phosphorus Deficiency

Phosphorus is an essential mineral for plant metabolic processes, including
photosynthesis, synthesis and breakdown of carbohydrates, and energy transfer.
Phosphorus is also a structural element of nucleic acids and phospholipids.
Phosphorus levels of 0.3-0.5% plant dry weight are required for optimal plant
growth (Marschner, 1995). Insufficient plant phosphorus levels result in reduced

leaf size and number, flower set, and shoot to root ratios (Lynch et al., 1991).

Phosphorus is extremely reactive and exists in the soil as organic P or one
of at least 170 different mineral phosphates. Organic P makes up 20 to 80% of the
soil P, a large fraction of which is phytic acid (inositol hexaphosphate)
(Richardson, 1994). Both mineral and organic P forms found in soils are
insoluble, and therefore there are very low levels of P in the soil solution available -
for plant absorption at any one time (Holford, 1997). For most soils, the amount
of P in solution is insufficient for crop production. There is an equilibrium that
occurs with P in soil solution as plants absorb P (Whitney, 1988) (Figure 2). At
the extreme end of low soil P availability are acid weathered soils, where P is in a
complex with iron and aluminum. Such soils, specifically Andosols, Ultisoils,
and Oxisols, make up 43% of the land area in the tropics. Many acid weathered
soils adsorb or fix P so that 70 to 90% of P fertilizer applied will become part of
these low solubility compounds (Sanchez and Salinas, 1981). Cultivation leads to
mineralization of organic P and mineralization occurs rapidly in tropical soils as
compared to temperate soils (Hedley et al., 1995).

Plants take up phosphorus from the soil solution in the form of
orthophosphate (l-IPO4'2 or HzPO4'1). Thetrelative amount of each anion depends
on soil pH. Acid soils favor the formation of H2P04'l, and alkaline soils favor the
formation of HPO4‘2. Phosphorus uptake by plants roots across the cell boundary
occurs against a steep concentration gradient via an ATP dependent proton A
syrnport (Bucher et al., 2001). Phosphorus uptake is greatest. at the root tip, and
is reduced by soil drying. Phosphorus diffusion is limited in the soil to less than 6

mm, therefore roots must be very close to phosphorus in the soil for uptake to

occur (Whitney, 1988). Root system surface area is very important for P uptake
because the greater the surface area, the greater the potential contact with P anions
(Lynch, 1995).

Phosphorus is translocated from the roots to young leaves in the xylem as
inorganic P. Remobilization of P (in organic or inorganic form) from older leaves
to younger leaves and roots occurs via the phloem and is more prevalent in P
deficient plants (Schachtman et al., 1998).

Phosphorus deficient soils will continue to be a problem in the coming
years, because P is a finite resource and the world’s supply of rock phosphate is.
estimated to expire in 60-80 years (Beebe et al., 2006). Common bean is
especially susceptible to low P soils in part due to the high P requirement
associated with energy-intensive nodulation and N2 fixation. For this reason,
common bean genotypes tolerant to low P soils when N fertilizer is added may
not be as tolerant if dependent on fixation to supply N to the plant (Christiansen
and Graham, 2002). Low P soils are a major constraint to bean production in
regions of Africa and Latin America where farmers lack access to sufficient P
fertilizer (Wortmann et al., 1998).

Plant adaptation to low P soil:

' Nutrient efficiency is the effectiveness with which a plant system utilizes
nutrient inputs to produce outputs. Phosphorus efficiency, has been defined in
numerous ways, here it is considered as the ability to produce plant growth and
yield in relation to the amount of available phosphorus (Lynch and Beebe, 1995).

Tolerance to low P soils, or phosphorus efficiency, can be achieved by two

distinct routes: acquisition efficiency and utilization efficiency. Acquisition
efficiency reflects a plants ability to extract P from the environment. It has been
shown to be related to root system traits that increase the root surface area and
allow capture of more P form the soil, including specific root length, root hair
density and mycorrhizae (Gahoonia and Nielsen, 2003). Root exudates that can
free inorganic P from insoluble P forms are another form of acquisition efficiency
(Marshcner, 1995).

Utilization efficiency is the superior ability of a plant to convert acquired
P into plant biomass and yield. Possible P utilization efficiency mechanisms .
include reduced requirements and modified phenology. Increased seed P
reserves is another form of utilization efficiency that may allow plants to get a
better start in low P soils (Lynch and Beebe, 1995).

Phosphorus acquisition efficiency:

Root system traits that improve phosphorus acquisition efficiency, and act
to capture more P from the soil by increasing the surface area of the root, have
been studied extensively in bean and other crops. Complex root architectures may
have evolved for the capture of immobile nutrients, such as P, from the soil. Even
with greatly reduced lateral branching, Arabidopsis can capture sufficient nitrate
ions (mobile), but require more lateral branching to effectively uptake sufficient
phosphate ions (Fitter et al., 2002).

Genotypic variability for P acquisition efficiency has been identified in
many crops including rice, soybean, and common bean (W issuwa and Ae, 2001a;

Yan et al., l995ab; Li et al., 2005). Landraces of rice showed improved tolerance

to low P soils than improved varieties, suggesting there is still unexplored
variation for P uptake that can be exploited for rice improvement (Wissuwa and
Ae, 2001a). In common bean, wild gennplasm was not more adapted to low P
soils than cultivated genotypes (Araujo et al., 1998). A screening of thousands of
both Mesoamerican landraces and Mesoamerican and Andean improved common '
bean varieties for tolerance to low fertility soil revealed variability within each of
the groups, suggesting that selected genotypes of either type can be useful as
parental material in a low fertility breeding program (Singh et al., 2003).
According to Yan et al. (1995b), Andean common beans are superior to
Mesoamerican beans in low P soils. They often have higher yields, although
Mesoamerican lines are more responsive to P fertilization (Y an et al., 1995b).
Phosphorus deficient soils of the tropics are of two major classes: mineral low P,
including oxisols and utisols, with inorganic bound P, and volcanic low P soils,
mostly andisols, with high organic matter and organic bound P. In a study of 16
Common bean lines from the Andean and Mesoamerican gene pools, large genetic
variation for P efficiency was discovered. The P acquisition efficiency appeared
to be general and not soil specific (Y an et al., 1995a).

Root system architecture

Most studies that relate root architecture to P acquisition efficiency
examine genotypes with diverse plant performance in P-lirniting environments, eg.
media with high and low levels of P and observe root responses. In maize, P-
deficient media promotes larger root systems in P-efficient and P-inefi‘icient

genotypes. However, the P-efficient genotype had a larger root system and higher

total root length than the P-inefficient genotype in low P conditions (Lui et al.,
2004). A plant growth simulation model for rice grown in low P showed that
improving root growth was the best way to enhance tolerance to P deficiency
(Wissuwa, 2003). Low P nutrient solution was associated with altered basal root
growth angle of some common bean genotypes, a change not exhibited with
deficiencies of N, K, Ca, Mg, S or micronutrients (Bonser et al., 1996). Rubio et
a1. (2003) found the shallowness of basal roots of common bean added a
competitive advantage to shoot biomass production as compared to deep basal
roots when P is not evenly distributed in soil depth but instead is found only in the
topsoil. Phosphorus acquisition efficient bean genotypes have more lateral roots
and shallower basal roots under low P conditions as compared to P inefficient
genotypes grown under similar conditions (Lynch and Brown 2001). Low
phosphorus insensitive Arabidopsis mutants did not switch from primary to lateral
root growth as seen in wild type plants. The mutants also showed reduced
expression of genes important in phosphate uptake, including a phosphate
transporter and acid phosphatases (Sanchez-Calderon et al., 2006).

In a study of four common bean lines, efficient lines had more
adventitious roots than inefficient lines. Differences were more pronounced early
in development (22 days after planting) than later in development (43 days after
planting) (Miller et al., 2003). This type of root system modification has
implications for acquisition efficiency because adventitious roots are shallow and
so are located within the P-emiched topsoil. Adventitious rooting can also be

categorized as a P use efficiency mechanism because they tend to have a lower

carbon construction cost than other root types as they are less dense (Lynch and
Ho, 2005).

The next step after the identification of root architecture traits associated
with nutrient efficiency is to understand the genetic basis of these traits, and
determine if the inheritance of a given trait is such that it can be readily
introduced to inefficient germplasm.

Variability for tolerance to low P soils has been identified in the Andean
gene pool (Yan et al., 1995a). In a population of recombinant inbred lines
developed from a P acquisition efficient, large seeded Andean bean,G19833 and a
P acquisition inefficient small seeded Mesoamerican bean DOR364, Liao et a1.
(2004) also found root gravitropism to be correlated with P. use efficiency. These
traits include basal root growth angle and shallow basal root length. Results from
this growth pouch assay were correlated with field responses of bean to low P.
Three QTL for root gravatropism traits all from G19833 detected using'the
growth pouch assay were associated with QTL» for P uptake in low P field
conditions. In the same population of RILS, a QTL for root length and one for
specific root length on two different linkage groups co-localized with QTL for
phosphorus uptake. The uptake QTL and the root QTL each explained about 15%

of phenotypic variation for these traits (Beebe et al., 2006).

Root hairs
Root hairs, or trichoblasts are tubular outgrowths of rhizodennal cells.

Since root hairs are thin, they are better able to explore soil for nutrients with low

mobility in soil such as P. They are able to penetrate cracks in the soil and also
can penetrate moderately compacted soils. Root hairs therefore aid in plant
anchorage. Root exudates are mainly released from root hairs. The growth of the
root system and the movement of root hair zone allow the nutrient availability to
be maintained spatially. (J ungk, 2001). Root hairs have many plasmodesmata
between cells to aid in symplastic transport of nutrients to the vascular tissue.
Root hair characteristics including length, radius, and density are important for P
uptake (Bucher et al., 2001). Root hairs have been shown to be important for P
use efficiency in bean, white clover, cowpea, wheat, and barley.

Yan et al. (2004) conducted a QT L mapping study of root traits associated
with P use efficiency in 86 RILS of an Andean efficient (G19833) by
Mesoamerica inefficient (DOR 364) cross in common bean. The efficient parent
had higher root hair density, average root hair length and total root hair length.
Root hair traits were induced by low P (2 uM) in both cultivars as compared to
high P (1000 uM). H+ exudation and total acid exudation increased in low P as
compared to high P in both parents, and was higher in G19833 than DOR364.

Wang et al. (2004) analyzed root hair traits associated with P use
efficiency in RILS of a wild x cultivated soybean cross. The wild genotype had a
higher shoot biomass and‘P uptake. It also had higher root hair density and total
root hair length per unit root length but lower average root hair length. None of

these traits were however correlated with total P in the plant.

10

Mycorrhizae

Mycorrhizae are symbiotic associations between plant roots and fimgi.
They are widespread in nature. An estimated 83% of dicots and 79% of monocots
have mycorrhizal associations (Marschner, 1995). Vesicular-arbuscular (VAM) is
the only type of mycorrhizae associated with annual crop plants. In this symbiosis,
host plants provide carbohydrates to the fungi and fungi primarily provide
increased access to uptake of mineral nutrients to the host plant, especially for low
mobility ions. Mycorrhizal plants take up 2-3 times more P per unit root length
than non mycorrhizal plants (Tinker et al., 1992). The predominant mechanism
by which VAM increase plant P uptake is by extending the surface area of the
root system (Cardoso and Kuyper, 2006). In addition VAM can acquire inorganic
P from some of the soil pools marginally available to plants, as shown in an
experiment in an Oxisol soil with mycorrhizal and nonmycorrhizal maize
(Cardoso et al., 2006). In the case of the phosphorus efficient bean cultivar
‘Carioca’, mycorrhizal plants were able to acquire more P fi'om the soil than
nonmycorrhizal plants because hyphae had a greater affinity for P in the soil
solution than the plant roots and therefore a greater influx of P occurred. The
beneficial effects of the mycorrhizal symbiosis on plant phosphorus levels were

evident at flowering and mid podfill (deSilveira and Nogueira Cardoso, 2004).
Reot exudates

Root exudates are able to increase phosphorus available to plants for

uptake fi'om the inorganic or organic pools not immediately available for plant

11

uptake (Figure 2). The type and amount of root exudate released by a plant is
genotype dependent. Proteoid roots of white lupin, which are clusters of closely
spaced determinate rootlets are very effective in acquiring P from the soil. They
are known to release citrate and acid phosphatases which release P from organic
and inorganic compounds (Gilbert et al., 1999). Lupin roots when compared to
soybean roots, were able to accumulate 4.8 times more P per unit root length than
soybean, suggesting that where soybean relied on increased root length to capture
more P, lupin used root exudation to increase P availability (Watt and Evans,
2003). A study of exudates of three grain legumes, fava bean (Viciafaba L.),
field pea (Pisum sativum L.) and white lupin (Lupinus albus L.) and the cereal
spring wheat (Triticum aéstivum L.) demonstrated all species’ roots exude acid
phosphatase, but the legumes more so than the wheat. There was also variability
among species as to which soil P pools (inorganic and organic) were accessed
(Nuruzzaman et al., 2006). A purple acid phosphatase gene isolated from
Medicago truncatula was expressed in Arabidopsis roots and the transgenic plants
were able to use organic P in the form of phytate as their sole P source (Xiao et al.,
2006). This study illustrated the potential for P acquisition efficient crop plants to
be developed via mechanisms that make plant unavailable P available. Another
impOrtant point is that plant species and even genotypes have significant
variability for the types and amounts of P solubilizing enzymes they contain and
express.

In common bean, genotypic differences for root exudation of organic acids

(citrate, acetate, and tartrate) also suggest a mechanism for P acquisition

12

efficiency. The large seeded Andean genotypes G19833 and G19839 exude
more organic acids from their roots than Mesoamerican genotypes DOR364 and
G212121. These organic acids are useful to mobilize Al and Fe bound P from
soils for uptake by the plant roots (Shen et al., 2002). Low P was associated with
an increase in acid phosphatase activity in the roots and on the root surface of the
P-efficient genotype, whereas only slight increased acidity was found on the root
surface of the inefficient genotype (Lui et al., 2004).

Root exudation is also soil type dependent. The cation-anion balance of
the plant plays a role in the pH of the rhizosphere, which in turn can effect the P
mobilization in the soil. The value of rhizosphere pH in mobilizing P is soil type
dependent. In a luvisol, where P was mainly bound to Ca, a decrease in
rhizosphere pH enhanced P mobilization, whereas in an oxisol where P was
mainly bound to Al and Fe an increase in rhizosphere pH mobilized more P
(Gahoonia and Nielsen 2004).
Phosphorus utilization efficiency: .

An evaluation of 26 bean lines with varying levels of P efficiency,
showed that the efficient and inefficient lines did not differ in P uptake, but
differed in pods per plant and seeds per pod under low P conditions. At 65 days
after emergence the efficient lines had a greater proportion of total P in pods and
less in stems and leaves than that of the inefficient lines that had more P
comparatively in stems and leaves and less in the pods (Y oungdahl, 1990). A

more in depth study of the P efficient common bean cultivar ‘Calima’ looked at

remobilization of P from stem and leaves in low and high P growing media. The

13

study found that in low P media remobilization occurs earlier (56-70 days after
emergence) than in high P media (70 days after emergence). Under low P
conditions 80% of P stayed in the root system, whereas in high P only 20% stayed
in the roots (Snapp and Lynch, 1996).

P use efficiency can also demonstrate itself in terms of carbon cost. A root
system that requires less energy (i.e. lower root respiration) can increase P uptake
per unit carbon invested. A comparison of P utilization in four common bean
genotypes indicated that efficient and inefficient genotypes have a similar rate of
P absorption per root dry weight. Phosphorus efficient bean genotypes G19839
and G1937 had a significantly lower rate of root respiration than the P inefficient
genotypes DOR364 and Porillo Sintetico. The efficient genotypes had a higher
percentage of their biomass as roots than the P-inefficient genotypes in low P
soils (Nielsen et al., 2001).

A high level of seed P is considered to be a form of P use efficiency as it is
indicative of the effect of P remobilization. In upland rice, the amount of shoot
and root dry weight produced at six weeks of growth was highly correlated with
the amount of phosphorus in the seed, suggesting high seed P as a trait to select
for when breeding cultivars tolerant to low P soils (Hedley et al., 1994). A
greenhouse experiment with four annual legume species Wedicago polymorpha,
T nfolium subterranean T. balanSae, Ornithopus compressus)
indicated under low P fertilizer levels demonstrated that increases in seed P
concentration was associated with increased plant yields (Bolland and Paynther,

1990)

14

Brachiaria species are the most widely planted forage grasses worldwide,
and are well adapted to low P soils. In addition to a root system efficient in P
acquisition, this species also uses P efficiently in the plant, as seen in the cultivar
Mulato. P use efficiency appeared to be related to enhanced sugar catabolism and
increased phosphohydrolase activity in low P conditions, which allows rapid
turnover of P (Nanamori et al., 2004).

The QTL for phosphorus uptake and use efficiency were each found on
chromosomes 2 and 12 of the rice genome, with the P uptake QTL coming from
one parent and the use efficiency coming from the other parent used in the study.
While P uptake was positively correlated with plant dry weight, use efficiency
was negatively correlated. This indicates that P use efficiency may not be a
positive adaptation worth exploring by rice breeders, but may be a response of
plants with poor P uptake (Wissuwa et al., 1998).

Summary '

The use of root properties as selection criteria to improve P use efficiency
through plant breeding requires a trait that is easily identifiable, for which large
genetic variability exists and the mode of inheritance is known (Gahoonia and
Nielsen, 2004). This poses a challenge using root traits in breeding as
characterization is notoriously difficult below ground. Success in breeding rice to
be more tolerant to low P soils was attained through a QTL for variation in P
uptake. The trait considered in that study was simply plant P concentration; no

root system measurements were taken. This QTL was transferred from one

15

cultivar to another and the resulting near isogenic line more than tripled its grain
yield (Wissuwa et al., 2001b).

The same success with breeding for improved tolerance to low P soils has
yet to be realized in common bean, despite the large body of literature regarding
potential mechanisms of P efficiency and genetic control of related traits. One
should keep in mind that seed yield is the bottom line in improving common bean
gennplasm for tolerance to low P soils. Singh et al. (2003) advocated field
selections under general low soil fertility instead of just low soil P. In addition,
and in direct contrast to previous finding by Yan et al. (1995a, 1995b) when field
selections of more than 5000 landraces and improved common bean lines were
conducted based on yield, it was Mesoamerican, not Andean races that were
identified as better adapted to low soil fertility (Singh et al., 2003). The
discrepancies observed most likely arise from what was measured as an indicator

of P efficiency, overall yield or plant P uptake.

Biotic Stress: Fusarium Root Rot

In contrast to the large amount of research that relates low soil phosphorus
tolerance to root architectural traits, there has been limited research that associates
root architecture with tolerance to Fusarium root rot. There have been numerous
studies illustrating that the severity of Fusarium root rot is highly influenCed by
environmental stresses that impact root growth. This is consistent with an

important role for root architecture in determining resistance.

16

Nature of disease:

Root rot, caused by a combination of fungi, iS a major limiting factor in
the production of Phaseolus vulgaris (Abawi and Pastor Corrales, 1990). The
complex of pathogens believed to contribute to root rot include: F usarium solam’
f.sp. phaseoli, Rhizoctom'a solam' Kuhn, Pythium ultimum, Thielavz'opsis basicola
(Berk. & Br.) Ferr. and Aphanomyces euteiches f.sp. phaseoli. In the majority of
dry bean and snap bean producing regions of the United States, including
Michigan, F. solani f.sp. phaseoli (Fsp) is considered the major pathogen of the
complex (Chatterjee 1958; Keenan et al., 1974; Burke and Kraft 1974; Steadman
et al., 1975; Saettler 1982). Yield losses of up to 84% have been attributed to
Fusarium root rot (Keenan et al., 1974).

Susceptibility to Fsp is developmentally regulated in beans. Fsp is not able
to effectively cause infection until the plant develops a root system (10 days after
planting). Earlier in development, considerably fewer spores attach to plant tissue
than in older bean plants. Disease symptoms in younger plants are limited to a
hypersensitive response at the inoculation site by the outermost cell layers
(V ogeli-Lange et al., 1995). Symptoms of Fusarium root rot include red
longitudinal lesions on the hypocotyls, taproot and lateral roots of the bean plants,
followed by root rot. In severe infections complete rotting of a root system is
observed (Schneider and Kelly, 2000). Increased adventitious rooting is also
often a sign of disease infection (Chatterjee 1958; Steadman et al., 1975).
Severity of Fsp increases as the plant develops, and with time it becomes possible

to see symptoms on upper parts of the plant including chlorosis, stunting, and

17

premature defoliation, along with reduction in pod production (Burke and Barker,
1966).

Although Fsp infection is readily visible on the hypocotyl and taproot,
lateral root infection most significantly impacts crop yield. A plant with a healthy
lateral root system can outgrow the detrimental effects of localized hypocotyl and
taproot infection (Burke and Barker 1966). The nature of the infection by Fsp
may be the reason for the importance of lateral roots. Infection is defined as
direct penetration of fungal hyphae into healthy epidermis, wound sites on
hypocotyl or. root tissue, or through hypocotyl stomata. The hyphae spread
extracellularly in the cortex until cell death occurs, at which time hyphae begin to
spread intracellularly. The pathogen is unable to pass the endodermis of the
taproot, but can enter the vascular system of lateral root during their emergence
(Christou and Snyder, 1962) suggesting that large loss of lateral roots leads to the
plant’s inability to transport enough water and nutrients to the shoot, but this may
not be a factor with taproot infection because the vascular system is able to
remain functional. The disease cycle is shown in Figure 3.

Environmental Factors:

Under optimal growing conditions, Fsp is a miner pathogen of beans.
Environmental conditions that increase the stress on plant roots and impede root
growth, including drought, flooding, (Miller and Burke, 1977) soil compaction
(Burke 1968; Miller and Burke, 1985), low soil fertility, low soil temperature
(Burke et al., 1980), and plant competition (Burke, 1965) exacerbate incidence of

Fusarium root rot.

18

Soil temperature

Soil temperature plays an important role in root rot disease incidence caused
by Fsp on common bean. Soil temperatures fluctuate seasonally and are impacted
by air temperature, with the lowest average soil temperatures occurring early and
late in the growing season. Common bean grows best at soil temperatures of 22
to 26°C. Temperatures above 30°C cause an inhibition in root growth by reducing
cell division and root elongation. Temperatures below 18°C also significantly
reduce yield. Chilling temperatures reduces membrane permeability and decrease
the ability of the root system to uptake water. Studies conducted by Reddick
(1917) and Sippell and Hall (1982a) indicated Fsp has a temperature tolerance of
12 to 35°C. Disease caused by Fsp appears to be favored by temperatures in the
range of 16 to 22°C as compared to warmer temperatures (26 to 34°C). At
temperatures above 30 °C there is reduced attachment of firngal spores to the-root
surface which Can decrease amount of infection and disease severity.

Suboptimum soil temperatures for bean grth increase Fusarium root rot
incidence. Suboptimal soil temperatures cause decreased root growth, water
uptake, and nutrient uptake. Such conditions exacerbate root rot caused by Fsp by
limiting root growth which reduces a plant’s ability to compensate for a portion of
the root system malfunctioning from rot. Beans: planted in warm soils are often
able to escape the yield depressing effects of root rot that occur under cold soil
temperatures. There is also an interaction between moisture, temperature, and

root rot severity. Although moist soil is most favorable for germination of resting

Fsp spores, it also favors the growth of the bean seedling and incidence of disease

19

is much greater in dry soils. In dry soils there is a larger temperature range over
which infection occurs as compared to moist soils. (Buerkert and Marschner,
1992). Under low soil temperatures (21°C days 16°C nights) root growth was
more reduced in the presence of Fsp as compared to inoculum free soil (Burke et
al,1980)

Drought

Deep root systems appear to be an important factor in enhancing
resistance to Fusarium root rot under dryland conditions. Fsp spores most widely
inhabit soil at depths of 15-45 cm below ground, as do the majority of bean roots.
. Dryden and Van Alfen (1984) found lower levels ostp infection on roots
growing below 45 cm as compared to those roots above 45 cm. Under dryland
conditions, soil water availability is often greatest below 60 cm, therefore, deep
rooted cultivars may tolerate drought by accessing water 45 to 60 cm below the
soil surface. Intact deep roots which are able to take up water may help plants
avoid Fsp infection (Miller 1985; Burke and Miller, 1983).

I Compaction

Soil compaction is an increase in soil bulk density and a decrease in air
filled pores as a result of force applied to the soil (Allrnaras et al., 1988).
Excessive soil compaction, which impedes root growth and function, may be
observed at a critical penetration resistance of 2 MPa and a DzDo ratio (oxygen
diffusion coefficient in the soil relative to air) of zero at 10% air-filled porosity
(Allrnaras et al.,-1988). Such soil compaction levels are frequently present in

agricultural fields as a result of machinery used for planting, harvest, and other

20

crop maintenance activities (Harverson et al., 2005). Plant root systems respond
to increases in soil bulk density or mechanical impedance by exerting a growth
pressure to get through soil channels that are smaller than the root diameter (Clark
et al., 2003). There is a positive correlation between root diameter and elongation
rate under impeded conditions among different species. In addition cultivar
differences for ability to penetrate soil with high bulk density have been shown in
rice using a wax layer technique in a greenhouse study (Yu et al., 1995).
Excessive soil compaction deprives roots of oxygen and they must undergo
anaerobic respiration, this process produces less than half the energy of aerobic
reSpiratibn and halts root growth and maintenance (Alhnaras et al., 1988).

Grimes et al. (1975) observed in corn and cotton that in the presence of a
compacted layer 30 cm below the soil, roots proliferate above the compacted
layer. Asady et al. (1984) developed a controlled environment method to test bean
root tolerance to compacted soil and showed that it was strongly correlated (R =
0.91) to field tests. Fewer roots were able to penetrate compacted soil layers,
causing an increase in root mass above the compacted layer and an overall
reduction in root biomass.

Excessive soil compaction can slow root tip advance up to 75%, making
interception and infection by soil fimgal pathogens much more likely than in non-
compacted soils (Huisman 1982; Alhnaras et al., 1988). Studies with bean and
Fsp showed that inoculation reduced root ability to penetrate a compacted soil
layer. Fusarium species are generally likely to infect the hypocotyl or more

stationary parts of the root system because they do not germinate quickly and

21

usually exist largely as colonies epiphytic on the root surface (Allrnaras et al.,
198 8). In compacted fields with Fusarium spores present, tillage increased bean
seed yields by 40% and reduce disease severity scores from 3.4 to 1.4 over non-
tilled fields (Harverson et al., 2005).

Bean genotypes that perform well in the presence of Fsp inoculum also
often demonstrate tolerance to other soil stresses. For example the resistant
genotype NY-2114-12 was able to increase root growth above the point of soil
compaction to a greater extent than susceptible cultivars (Miller, 1985). The
resistant snap bean ‘FR266’ also has been shown to have high tolerance to
compacted soil (Silbemagel, 1990). NY 2114-12 also had higher shoot and root
growth under water deficit'as compared to susceptible cultivars (Miller 1985).
Fusarium root rot resistant cultivars outperform susceptible cultivars under each
of the environmental stresses listed above except flooding stress. Resistant
cultivars were unable to overcome disease under low soil oxygen stress (Miller
1985; Burke and Miller, 1983).

Methods of Control:

Root rot severity in beans is related to environmental conditions that limit
root growth. Cultural practices that allow the development of a vigorous root
system are essential to reducing root rot, especially when resistant cultivars are
unavailable. Thus, planting in warm soil and maintaining optimum soil water are
important to establish vigorous root grth which allows plants to better
withstand root rot. Irrigation is especially important in dry years because water

stressed plants are more susceptible to root rot than non water stressed plants.

22

Increasing the spacing between plants within rows and between rows may reduce
root rot severity by reducing competition among root systems for water and
nutrients.

Crop rotation: Crop rotation can be used to reduce Fsp populations in the
soil. Soil population density (cfu/ g) of Fsp is highest in continuous bean
production and rotation can reduce the amount of inoculum in the soil. Fsp is
very immobile in the soil with 14 years of continuous corn, soybean or fallow
having very little Fusarium (Sippell and Hall 1982b). Recommendations are
generally for 4 year rotations to reduce inoculum.

Rotation with small grains and alfalfa add residue to the soils and can
reduce compaction, improve the soil structure and increase water holding capacity
of the soil. However, residues must be incorporated at least one month before
planting or phytotoxicity can result (Burke and Miller, 1983). Organic
amendments can also increase overall microbial population in the soil which in
some cases reduCe habitat and nutrient availability in soil which could reduce
competitiveness of Fsp.

Deep Tillage: Deep tillage of the soil has been very effective to reduce
root rot severity in compacted soils. Both susceptible and resistant cultivars
exhibit decreased root rot severity with deep tillage (Tan and Tu, 1995). Deep
tillage in the presence of Fsp inoculum improves plant yields more so than in non-
infested soil. A study by Miller and Burke (1986) found that susceptible cultivars
showed a larger yield increase fiom tillage than resistant cultivars illustrating the

importance of deep root growth to overcome infection.

23

Chemical Treatments: Seed treatments are generally not effective against
Fusarium root rot (Sippell and Hall, 1982b). Soil fumigants such as chloropicrin
and methyl bromide may suppress F usarium root rot but are economically and
environmentally impractical (Burke and Miller, 1983).

Genetics:

Sources of Resistance:

N203 (P1203958) is considered the best source of resistance to F usarium
root rot in P. vulgaris. Although this accession does not exhibit complete
immunity, it does have a high degree of resistance. N203 is a late maturing,
indeterminate accession collected in Mexico (Wallace and Wilkinson 1965). P.
coccineus, also a primary source of resistance has shown slightly higher levels of
resistance that N203 (Wallace and Wilkinson 1965). Both principle known
sources of resistance, N203 and P. coccineus, have been reported to exhibit higher
levels of resistance in the field as compared to the greenhouse (Wallace and
Wilkinson, 1965; Baggett and Frazier, 1959), suggesting root avoidance or escape
of infection may be an important component to resistance in the field.

The first F usarium root rot resistant dry bean cultivars were released in
1974. They included the pink cultivars ‘Viva’, and ‘Roza’ and the small red
cultivar ‘Rufus’ (Boomstra and Bliss, 1977). Each cultivar derived resistance
from N203 (Burke, 1982). The first snap bean cultivars with resistance to
Fusarium root rot were released in 1978. The source of resistance of these
cultivars (RRR 77) and (RR 83) was not specified (Hagadom and Rand, 1978).

Available genotypes with the highest level of resistance are late maturing with a

24

vine growth habit. It has been difficult to transfer resistance to early maturing
short-vine and bush type beans (Burke and Miller, 1983). ‘FR266’ is a resistant
snap bean with ‘Blue Lake’ pod quality characteristics released in 1987 which
derived resistance from N203 and P. coccineus (Silbemagel, 1987). It is one of
the few genotypes available with Fusarium root rot resistance and a bush type
growth habit.

Inheritance of Resistance:

Bravo et al. (1969) found the resistance from N203 and P. coccineus to be
inherited in a similar manner. They found resistance with either of the parents to
be mostly dominant and controlled by 3 Or more genes (Bravo et al., 1969). Based
on comparisons of the inheritance of resistance in N203 and 2114-12 (resistance
from P. coccineus), Hassan et al. (1971) found N203 to possess 4 resistance genes
and 2114-12 to posses 5-6 resistance genes, with 4 of the resistant genes in 2114—
12 also present in N203. Hassan et al. (1971) used a hypocotyl disease severity
index to rate disease and did not consider root rot. Thus, their results may not be
applicable to ‘root rot’. Hassan et a1 (1971) obtained similar results as Bravo et a1.
(1969), who scored disease index on the roots. Overall there are many conflicting
reports on the number of genes controlling resistance and it is clearly
i . quantitatively inherited. The exact number of genes controlling the trait remains
unclear. Numerous conflicting reports on the inheritance of resistance to Fsp
suggest that growth and environmental conditiOns influence study results

(Boomstra and Bliss 1977).

25

Schneider et al. (2001) observed while studying a population of RILS
derived from a cross between FR266 and Montcalm that there is no hypersensitive
response in the resistant genotype, only a slowing of symptom development. In
the same population QTL were found to be associated with resistance in the field
or in the greenhouse but there was no overlap. In addition, no single QTL
explained more than 15 % of phenotypic variation, indicating that many genes are
involved. Intermediate heritability scores also indicated a high effect on the
environment for this trait. In a recent QTL study by Roman-Aviles and Kelly
(2005) an inbred backcross line population was used to reduce the background
effects and isolate more thoroughly the genetic mechanism for resistance. In this
study a total of 5 QTL on two linkage groups explained 73% of the phenotypic
variation. Again, QTL identified in field were different than those in the
greenhouse. This may be in part because greenhouse plants were scored after two
weeks whereas in the field they were scored after flowering or at maturity,
resulting in selection for early expressed genes for resistance. One QTL was
linked to a marker by Schneider et a1 (2001) that was identified (F R266 x
Montcalm) as important for resistance.

Root Architecture and Disease:

Root pathogen interactions are largely dependent onroot growth, as roots
are typically much more mobile in the soil than fungal pathogens. Growth rates
of root length can serve as an indicator for the potential for root pathogen contact
(Huisman, 1982). Root length also serves to indicate how much area is available

for water and nutrients in the presence of root rot (Burke and Barker, 1966).

26

Genotypic variability for root length has been shown to be associated with
resistance in common bean. Under field conditions, in the presence of Fsp spores,
the root rot resistant genotype FR266 had a root dry Weight twice that of the
susceptible cultivar ‘Montcalm’, along with a greater length and surface area,
indicating an overall larger root system (Roman-Aviles et al., 2004). In addition
average root system diameter was negatively correlated with root rot severity (r2:
0.4) in RILS developed from a cross between Montcalm and FR266. The average
number of adventitous in the top cm was positively correlated with root rot
tolerance (r2=0.6) (Snapp et al., 2003). In a screening of 10 bean cultivars from
different market classes and with varying susceptibility to Fusarium root rot the
three most resistant cultivars also had the most adventitious root mass (Roman-
Aviles et al., 2004). Adventitious roots are commonly considered a host plant
response to compensate for the infection of taproots and lateral roots (Chatterj ee,
1958; Steadman et al., 1975).

A few studies have examined the role of root system architecture in
resistance to root pathogens. Vine decline of muskrnelon (Cucumis melo L.) is
caused by a complex of soil borne fungi, the most prominent fungus,
Monosporascus cannonballus Causes the root rot symptoms. As with Fusarium
root'rot in bean, disease severity is strongly influenced by the environment and no
highly resistant cultivars exist (Dias et al., 2002). Root morphology variability
has been discovered in tolerant and susceptible cultivars infected with the
pathogen. Tolerant cultivars had more vigorous root systems, including a greater

total root length, more root tips, and a greater fine root length than susceptible

27

cultivars (Crosby et al., 2000). The highly resistant wild melon accession Pat81
had a larger root system and more lateral roots that a highly susceptible
commercial cultivar ‘Amarillo Canario’ both in the presence and absence of
fungal inoculum (Dias et al., 2002). Pat8l was used as a parent in a backcross
breeding program, and progeny with larger root systems demonstrated root rot
tolerance (Dias et al., 2004).

Root and Shoot Growth Habit Coordination

Common bean shoot architecture has been classified into four types:
determinate (Type I), indeterminate upright (Type H), indeterminate prostrate
(Type III), and vine (Type IV) (Kelly 2001). Limited observational studies
support a coordinated grth habit of the root and shoot (Lynch and van Beem
1993; Kelly, 1998). ‘Carioca’ a type III plant with highly branched shoot
architecture and little apical dominance also exhibited highly branched root
architecture (Lynch and van Beem 1993). In the case of Fusarium root rot
resistance, it has been much easier to develop resistance in indeterminate type
cultivars than in determinate cultivars. Therefore, if root architecture is important
for resistance to Fusarium root rot, it may be strongly influenced by shoot
architecture traits such as total leaf area. To effectively breed determinate
Fusarium resistant bean cultivars an understanding of root and shoot traits is

required to detennine how to separate them.

28

Summary:

A half century of research has shown that Fusarium root rot severity in
common bean is strongly influenced by environmental conditions that affect plant
root growth. In addition all known sources of resistance provide partial resistance
to infection. These factors of disease have indicated that chemical resistance is not
a reliable selection tool as seen with differences between genetic control of field
and greenhouse resistance. Breeding of cultivars resistant to Fsp is typically done
in combination with breeding for general tolerance to stress.

Only within the last few years have researches began to study the
relationship between root architecture and F usarium root rot. In that time, it has
been shown that indeed, the resistant cultivars do have larger root systems in field
conditions than susceptible cultivars. Much more research needs to be conducted,
however to determine the genetic control of this relationship and what root traits
may serve as selection tools to facilitate breeding efforts that can provide cultivars
that are less susceptible to Fusarium root rot in the presence of environmental

stresses.

29

References:

2004. The Phosphorus Cycle [Online]. Available by Potash and Phosphate
Institute http://msucares.com/crops/soils/phosphorus.html.

Abawi, GS, and MA. Pastor Corrales. 1990. Root rots of beans in Latin
America and Africa: Diagnosis, research methodologies, and management
strategies CIAT, Cali, Colombia.

Allmaras, R.R., J .M. Kraft, and DE. Miller. 1988. Effects of Soil Compaction
and Incorporated Crop Residue on Root Health. Annual Review of
Phytopathology 26:219-243.

Araujo, A.P., M.G. Teixeira, and BL. de Alrneida. 1998. Variability of traits
associated with phosphorus efficiency in wild and cultivated genotypes of
common bean. Plant and Soil 203: 173-182.

Asady, G.H., A.J.M. Smucker, and M.W. Adams. 1985. Seedling Test for the
Quantitative Measurement of Root Tolerances to Compacted Soil. Crop Science
25:802-806.

Austin, DR 2002. Roots as a Source of Food, p. 1025-1043, In Y. Waisel, Eshel,
A., and Kafltafi U., ed. Plant Roots The Hidden Half, 3rd ed. Marcel Dekker, Inc,
New York.

Baggett, J .R., and WA. Frazier. 1959. Disease Resistance in the Runner Bean,
Phaseolus coccineus L. Plant Disease Reporter 43: 137-143.

Beebe, S.E., M. Rojas-Pierce, X.L. Yan, M.W. Blair, F. Pedraza, F. Munoz, J.
Tohme, and JP. Lynch. 2006. Quantitative trait loci for root architecture traits

. correlated with phosphorus acquisition in common bean. Crop Science 46:413-
423.

Bell, D.L., and SE. Sultan. 1999. Dynamic phenotypic plasticity for root growth
in Polygonum: A comparative study. American Journal of Botany 86:807-819.

Bolland, M.D.A., and RH. Paynter. 1990. Increasing Phosphorus Concentration

in Seed of Annual Pasture Legume Species Increases Herbage and Seed Yields.
Plant and Soil 125: 197-205.

Bonser, A.M., J. Lynch, and S. Snapp. 1996. Effect of phosphorus deficiency on
growth angle of basal roots in Phaseolus vulgaris. New Phytologist 132:281-288.

Boomstra, AG, and FA. Bliss. 1977. Inheritance of Resistance to F usarz’um?
Solam' F Sp Phaseoli in Beans (Phaseolus vulgaris L) and Breeding Strategy to

30

Transfer Resistance. Journal Of the American Society for Horticultural Science
102:186-188.

Bravo, A., D.H. Wallace, and Wilkinso.Re. 1969. Inheritance of Resistance to
Fusarium Root Rot of Beans. Phytopathology 59: 1930-&.

Bucher, M., C. Rausch, and P. Daram. 2001. Molecular and biochemical
mechanisms of phosphorus uptake into plants. Journal of Plant Nutrition and Soil
Science 164:209-217.

Buerkert, A., and H. Marschner. 1992. Calcium and Temperature Effects on

Seedling Exudation and Root-Rot Infection of Common Bean on an Acid Sandy
Soil. Plant and Soil 1472293-303.

Burke, D.W. 1968. Root Growth Obstructions and Fusarium Root Rot of Beans.
Phytopathology 58:1575-&.

Burke, D.W. 1982. Registration of Pink Beans Viva, Roza, and Gloria. Crop
Science 22:684-684.

Burke, D.W., and AW. Barker. 1966. Importance of Lateral Roots in Fusarium
Root Rot of Beans. Phytopathology 56:292-&.

Burke, D.W., and J.M. Kraft. 1974. Responses of Beans and Peas to Root
Pathogens Accumulated During Monoculture of Each Crop Species.
Phytopathology 64:546-549.

Burke, D.W., and DE. Miller. 1983. Control of Fusarium Root-Rot with Resistant
Beans and Cultural Management. Plant Disease 67:1312-1317.

Burke, D.W., D.E. Miller, and AW. Barker. 1980. Effects of Soil-Temperature on
Grth of Beans in Relation to Soil Compaction and Fusarium Root-Rot.
Phytopathology 70: 1047-1049.

Cardoso, I.M., and T.W. Kuyper. 2006. Mycorrhizas and tropical soil fertility.
Agriculture Ecosystems & Environment 116172-84.

Cardoso, I.M., C.L. Boddington, B.H. Janssen, O. Oenema, and T.W. Kuyper.
2006. Differential access to phosphorus pools of an oxisol by mycorrhizal and

nonmycorrhizal maize. Communications in Soil Science and Plant Analysis
37:1537-1551. *

Chatterjee, P. 1958. The Bean Root Rot Complex in Idaho. Phytopathology
48:197-200.

31

Christiansen, 1., and PH. Graham. 2002. Variation in di-nitrogen fixation among
Andean bean (Phaseolus vulgaris L.) genotypes grown at low and high levels of
phosphorus supply. Field Crops Research 73:133-142.

Christou, T., and WC. Snyder. 1962. Penetration and Host-Parasite Relationships
of Fusarium—Solani-F-Phaseoli in Bean Plant. Phytopathology 52:219-&.

I Clark, L.J., W.R. Whalley, and PB. Barraclough. 2003. How do roots penetrate
strong soil? Plant and Soil 255:93-104.

Crosby, K., D. Wolff, and M. Miller. 2000. Comparisons of root morphology in
susceptible and tolerant melon cultivars before and after infection by
Monosporascus cannonballus. Hortscience 35 :68 l -683.

da Silveira, A.P.D., and E. Cardoso. 2004. Arbuscular mycorrhiza and kinetic
parameters of phosphorus absorption by bean plants. Scientia Agricola 61 :203-
209. .

Dias, R.D.S., B. Pico, A. Espinos, and F. Nuez. 2004. Resistance to melon vine -
decline derived from Cucumis melo ssp agrestis: genetic analysis of root structure
and root response. Plant Breeding 123:66-72.

Dias, R.D.S., B. Pico, J. Herraiz, A. Espinos, and F. Nuez. 2002. Modifying root .
structure of cultivated muskmelon to improve vine decline resistance. Hortscience
37: 1092-1097.

Dryden, P., and N.K. Vanalfen. 1984. Soil-Moisture, Root-System Density, and
Infection of Roots of Pinto Beans by F usarz'um-Solani 1F Sp Phaseoli under
Dryland Conditions. Phytopathology 74: 132-135.

Fita, A., B. Pico, and F. Nuez. 2006. Implications of the genetics of root structure
in melon breeding. Journal of the American Society for Horticultural Science
131:372-379. .

Fitter, A. 2002. Characteristics and Functions of Root Systems, p. 15-32, In Y.
Waisel, Eshel, A., and Kafltafi U., ed. Plant Roots The Hidden Half, 3rd ed.
Marcel Dekker, Inc., New York.

Fitter, A., L. Williamson, B. Linkohr, and O. Leyser. 2002. Root system
architecture determines fitness in an Arabidopsis mutant in competition for
immobile phosphate ions but not for nitrate ions. Proceedings of the Royal
Society of London Series B-Biological Sciences 269:2017-2022.

Gahoonia, TS, and NE. Nielsen. 2003. Phosphorus (P) uptake and growth of a

root hairless barley mutant (bald root barley, brb) and wild type in low- and high-
P soils. Plant Cell and Environment 26:1759-1766.

32

Gahoonia, TS, and NE. Nielsen. 2004. Root traits as tools for creating .
phosphorus efficient crop varieties. Plant and Soil 260:47-57.

Gerald, J.N.F., M.D. Lehti-Shiu, P.A. Ingram, K.I. Deak, T. Biesiada, and J .E.
Malamy. 2006. Identification of quantitative trait loci that regulate Arabidopsis
root system size and plasticity. Genetics 172:485-498.

Gilbert, G.A., J.D. Knight, C.P. Vance, and D.L. Allan. 1999. Acid phosphatase
activity in phosphorus-deficient white lupin roots. Plant Cell and Environment
22 : 8O 1 -8 10.

Grimes, D.W., R.J. Miller, and PL. Wiley. 1975. Cotton and Corn Root
Development in Two Field Soils of Different Strength Characteristics. Agronomy
Journal 67 :5 19-523.

Hagedom, DJ, and RE. Rand. 1978. Bean Breeding Lines Wisconsin (Rrr) 77
and Wisconsin (Rrr) 83. Hortscience 13:202-202.

Harveson, R.M., J .A. Smith, and W.W. Stroup. 2005. Improving root health and
yield of dry beans in the Nebraska panhandle with a new technique for reducing
soil compaction. Plant Disease 89:279-284.

Hassan, A.A., D.H. Wallace, and Wilkinso.Re. 1971. Genetics and Heritability of
Resistance to F usarium-Solani F Phaseolr' in Beans. Journal of the American
Society for Horticultural Science 96:623-&.

Hedley, M.J., G.J.D. Kirk, and MB. Santos. 1994. Phosphorus Efficiency and the
Forms of Soil-Phosphorus Utilized by Upland Rice Cultivars. Plant and Soil
1 58 : 53-62.

Hedley, M.J., J.J. Mortvedt, N.S. Bolan, and J .K. Syers. 1995. Phosphorus
Fertility Management in Ecosystems, p. 59-92, In H. Tiessen, ed. Phosphorus in
the Global Environment. John Wiley and Sons, New York.

Holford, I.C.R. 1997. Soil phosphorus: its measurement, and

its uptake by plants. Australian Journal of Soil Research 35:227-235.

Huisman, CC. 1982. Interrelations of Root-Growth Dynamics to Epidemiology
of Root-lnvading Fungi. Annual Review of Phytopathology 20:303-327.

Johnson, W.C., LE. Jackson, 0. Ochoa, R. van Wijk, J. Peleman, D.A. St Clair,
and R.W. Michehnore. 2000. Lettuce, a shallow-rooted crop, and Lactuca serriola,
its wild progenitor, differ at QTL determining root architecture and deep soil
water exploitation. Theoretical and Applied Genetics 10111066-1073.

33

Jungk, A. 2001. Root hairs and the acquisition of plant nutrients from soil. Journal
of Plant- Nutrition and Soil Science-Zeitschrift Fur Pflanzenemahrung Und
Bodenkunde 164:121-129.

Keenan, J.G., H.D. Moore, N. Oshima, and LE. Jenkins. 1974. Effect of Bean
Root-Rot on Dryland Pinto Bean Production in Southwestern Colorado. Plant
Disease Reporter 58:890-892.

Kelly, J .D. 1998. Bean roots — A plant breeder’s perspective. Bean Improvement
Cooperative Annual Report 41 :214-215.

Kelly, J .D. 2001. Remaking bean plant architecture for efficient production.
Advances in Agronomy 71 :109-143

Laperche, A., F. Devienne-Barret, O. Maury, J. Le Gouis, and B. Ney. 2006. A .
simplified conceptual model of carbon/nitrogen functioning for QTL analysis of

winter wheat adaptation to nitrogen deficiency. Theoretical and Applied Genetics
1 13:1 131- 1 146.

Li, Y.D., Y.J. Wang, Y.P. Tong, J .G. Gao, J.S. Zhang, and S.Y. Chen. 2005. QTL
mapping of phosphorus deficiency tolerance in soybean (Glycine max L. Merr.).
Euphytica 142:137-142.

Liao, H., G. Rubio, X.L. Yan, A.Q. Cao, K.M. Brown, and JP. Lynch. 2001.
Effect of phosphorus availability on basal root shallowness in common bean.
Plant and Soil 232:69-79.

Liao, H., X. L. Yan, G. Rubio, S. E. Beebe, M. W. Blair, and J. P. Lynch. 2004.
Genetic mapping of basal root gravitropism and phosphorus acquisition efficiency
in common bean. Functional Plant Biology 31 9-.59 970.

Liu, Y., G.H. Mi, R]. Chen, J .H. Zhang, and ES. Zhang. 2004. Rhizosphere
effect and root growth of two maize (Zea mays L.) genotypes with contrasting P
efficiency at low P availability. Plant Science 167:217-223.

Lopez-Bucio, J., A. Cruz-Rarnirez, and L. Herrera-Estrella. 2003. The role of
nutrient availability in regulating root architecture. Current Opinion 1n Plant
Biology 6: 280- 287.

Lynch, J. 1995. Root Architecture and Plant Productivity. Plant Physiology 109. 7-
13.

Lynch, 1., and J .J . Vanbeem. 1993. Growth and Architecture of Seedling Roots of
Common Bean Genotypes. Crop Science 33:1253-1257.

34

Lynch, JP, and SE. Beebe. 1995. Adaptation of Beans (Phaseolus- Vulgaris L)
to Low Phosphorus Availability. Hortscience 30:1165-1171.

Lynch, J .P., and KM. Brown. 2001. Topsoil foraging - an architectural adaptation
of plants to low phosphorus availability. Plant and Soil 237:225-23 7.

Lynch, J .P., and MD. Ho. 2005. Rhizoeconomics: Carbon costs of phosphorus
acquisition. Plant and Soil 269245-56.

Lynch, J., A. Lauchli, and E. Epstein. 1991. Vegetative Growth of the Common
Bean in Response to Phosphorus-Nutrition. Crop Science 3 1 :380-3 87.

Malamy, J.E. 2005. Intrinsic and environmental response pathways that regulate
root system architecture. Plant Cell and Environment 28:67-77.

Marschner, H. 1995. Mineral Nutrition of Higher Plants. 2nd ed. Academic Press,
New York.

Merrill, S.D., D.L. Tanaka, and J .D. Hanson. 2002. Root length grth of eight
crop species in haplustoll soils. Soil Science Society of America Journal 66:913-
923.

Miller, DE, and D.W. Burke. 1977. Effect of Temporary Excessive Wetting on
Soil Aeration and Fusarium Root-Rot of Beans. Plant Disease Reporter 61: 175-
179.

Miller, DE, and D.W. Burke. 1985. Effects of Soil Physical Factors on
Resistance in Beans to Fusarium Root-Rot. Plant Disease 69:324-327.

Miller, DE, and D.W. Burke. 1986. Reduction of F usarirun Root-Rot and
Sclerotinia Wilt in Beans with Irrigation, Tillage, and Bean Genotype. Plant

‘ Disease 70: 163-166.

Miller, OR, I. Ochoa, K.L. Nielsen, D. Beck, and J .P. Lynch. 2003. Genetic
variation for adventitious rooting in response to low phosphorus availability:

potential utility for phosphorus acquisition fiom stratified soils. Functional Plant
Biology 30:973-985.

Nanamori, M., T. Shinano, J. Wasaki, T. Yamarnura, I.M. Rao, and M. Osaki.
2004. Low phosphorus tolerance mechanisms: Phosphorus recycling and
photosynthate partitioning in the tropical forage grass, Brachiaria hybrid cultivar
mulato compared with rice. Plant and Cell Physiology 45:460-469.

Nielsen, K.L., A. Eshel, and JP. Lynch. 2001. The effect of phosphorus

availability on the carbon economy of contrasting common bean (Phaseolus
vulgaris L.) genotypes. Journal of Experimental Botany 52:329-339.

35

Nuruzzarnan, M., H. Lambers, M.D.A. Bolland, and E.J. Veneklaas. 2006.
Distribution of carboxylates and acid phosphatase and depletion of different

phosphorus fractions in the rhizosphere of a cereal and three grain legumes. Plant
and Soil 281:109-120.

Reddick, D. 1917. Effect of soil temperature on the growth of bean plants and on
their susceptibility to a root parasite. American Journal of Botany 4:513-519.

Richardson, AB. 1994. Soil Microorganisms and phosphorus availability, p. 50-
62, In C. Pankhurst, et al., eds. Soil biota management in sustainable farming
systems. CSIRO Publishing, Melbourne.

Roman-Aviles, B., and JD. Kelly. 2005. Identification of quantitative trait loci

conditioning resistance to F usarium root rot in common bean. Crop Science
45:1881-1890. ’

Roman-Aviles, B., S.S. Snapp, and J .D. Kelly. 2004. Assessing root traits

associated with root rot resistance in common bean. Field Crops Research 86: 147-
156.

Rubio, G., H. Liao, X.L. Yan, and JP. Lynch. 2003. Topsoil foraging and its role
in plant competitiveness for phosphorus in common bean. Crop Science 43:598-
607.

Saettler, AW. 1982. Bean Disease and Their Control, In R. D. Robertson, ed. Dry
Bean Production: Principles and Practices. MSU Cooperative Extension, East
Lansing, MI.

Sanchez, RA, and JG. Salinas. 1981. Low-input technology for managing
Oxisols and Utisols in tropical America. Advances in Agronomy 34:280-406.

Sanchez-Calderon, L., J. Lopez-Bucio, A. Chacon-Lopez, A. Gutierrez-Ortega, E.
Hemandez-Abreu, and L. Herrera-Estrella. 2006. Characterization of low
phosphorus insensitive mutants reveals a crosstalk between low phosphorus-
induced determinate root development and the activation of genes involved in the
adaptation of Arabidopsis to phosphorus deficiency. Plant Physiology 140:879-
889.

Schachtrnan, D.P., R.J. Reid, and SM. Ayling. 1998. Phosphorus uptake by plants:
From soil to cell. Plant Physiology 116:447-453.

Schneider, K.A., K.F. Grafton, and JD. Kelly. 2001. QTL analysis of resistance
to Fusarium root rot in bean. Crop Science 41 :535-542.

36

Schwartz, HF. 2006. Root Rot of Dry Beans [Online]. Available by Colorado
State University Cooperative Extension
http://www.ext.colostate.edu/pubs/crops/02938.htrnl (posted June 2006).

Shen, H., X.L. Yan, M. Zhao, S.L. Zheng, and X.R. Wang. 2002. Exudation of
organic acids in common bean as related to mobilization of aluminum- and iron-
bound phosphates. Environmental and Experimental Botany 4811-9.

Silbemagel, M.J. 1987. Fusarium Root Rot-Resistant Snap Bean Breeding Line
Fr-266. Hortscience 22: 1 337-1338.

Silbemagel, M.J., and LI. Mills. 1990. Genetic and Cultural-Control of Fusarium
Root-Rot in Bush Snap Beans. Plant Disease 74:61-66.

Singh, S.P., H. Teran, C.G. Munoz, J.M. Osorno, J .C. Takegami, and M.D.T.
Thung. 2003. Low soil fertility tolerance in landraces and improved common
bean genotypes. Crop Science 43:110-119.

Sippell, D.W., and R. Hall. 1982a. Effects of pathogen species, inoculum
concentration, temperature, and soil moisture on bean root rot and plant growth.
Canadian Journal of Plant Pathology 4:1-7.

Sippell, D.W., and R. Hall. 1982b. Effects of F usarium solani phaseoli, Pythium
ultimum, and F. oxysporum on yield components of white bean. Canadian Journal
of Plant Pathology 4:54-58.

Snapp, SS, and J .P. Lynch. 1996. Phosphorus distribution and remobilization in
bean plants as influenced by phosphorus nutrition. Crop Science 36:929-935.

Snapp, S., W. Kirk, B. Roman-Aviles, and J. Kelly. 2003. Root traits play a role
in integrated management of Fusarium root rot in snap beans. Hortscience 38: l 87-
191.

Steadman, J.R., E.D. Kerr, and RE. Murnm. 1975. Root-Rot of Bean in Nebraska
- Primary Pathogen and Yield Loss Appraisal. Plant Disease Reporter 59:305-308.

Sultan, SE. 1987. Evolutionary Implications of Phenotypic Plasticity in Plants.
Evolutionary Biology 21 :127-178.

Tan, CS, and J .C. Tu. 1995. Tillage Effect on Root-Rot Severity, Growth and
Yield of Beans. Canadian Journal of Plant Science 75:183-186.

Tinker, P.B., M.D. Jones, and D.M. Durall. 1992. A functional comparison of

ecto- and endomycorrhizas, p. 303-310, In D. J. Read, eta1., eds. Mycorrhizas in
Ecosystems. CAB International, Wellingford, UK.

37

Tuberosa, R., M.C. Sanguineti, P. Landi,‘M. Michela Giuliani, S. Salvi, and S.
Conti. 2002. Identification of QTLS for root characteristics in maize grown in
hydroponics and analysis of their overlap with QTLS for grain yield in the field at
two water regimes. Plant Molecular Biology 48:697-712.

Vogelilange, R., U. Mohr, A. Wiemken, M. Duggelin, R. Guggenheim, and T.
Boller. 1995. Developmental Regulation of Susceptibility and Tolerance to
Fusarium Root-Rot in Beans. Botanica Acta 108:387-395.

Wallace, DH, and Wilkinso.Re. 1965. Breeding for Fusarium Root Rot
Resistance in Beans. Phytopathology 55:1227.

Wang, L.D., H. Liao, X.L. Yan, B.C. Zhuang, and Y.S. Dong. 2004. Genetic
variability for root hair traits as related to phosphorus status in soybean. Plant and
Soil 261 :77-84.

Watt, M., and JR. Evans. 2003. Phosphorus acquisition from soil by white lupin
(Lupinus albus L.) and soybean (Glycine max L.), species with contrasting root
development. Plant and Soil 248:271-283.

Whitney, DA. 1988‘. Phosphorus Facts Soil, Plant, and Fertilizer. Extension

Bulletin. Kansas State University Cooperative Extension Service, Manhattan,
Kanas.

Wissuwa, M. 2003. How do plants achieve tolerance to phosphorus deficiency?
Small causes with big effects. Plant Physiology 133:1947-1958.

Wissuwa, M., and N. AC. 2001. Further characterization of two QTLS that
increase phosphorus uptake of rice (Oryza sativa L.) under phosphorus deficiency.
Plant and Soil 237:275-286.

Wissuwa, M., and N. Ae. 2001. Genotypic variation for tolerance to phosphorus
deficiency in rice and the potential for its exploitation in rice improvement. Plant
Breeding 120:43-48.

Wissuwa, M., M. Yano, and N. Ae. 1998. Mapping of QTLS for phosphorus-
deficiency tolerance in rice (Oryza sativa L.). Theoretical and Applied Genetics
97:777-783.

Wortrnann, C.S., R.A. Kirkby, C.A. Eledu, and DJ. Allen. 1998. Atlas
‘ of common bean (Phaseolus vulgaris L.) production in Africa CIAT, Cali,
Colombia.

Xiao, K., H. Katagi, M. Harrison, and Z.Y. Wang. 2006. Improved phosphorus

acquisition and biomass production in Arabidopsis by transgenic expression of a
purple acid phosphatase gene fi'om M. truncatula. Plant Science 170:191-202.

38

Yan, X.L., J .P. Lynch, and SE. Beebe. 1995. Genetic-Variation for Phosphorus
Efficiency of Common Bean in Contrasting Soil Types .1. Vegetative Response.
Crop Science 35:1086-1093.

Yan, X.L., S.E. Beebe, and J .P. Lynch. 1995. Genetic-Variation for Phosphorus
Efficiency of Common Bean in Contrasting Soil Types .2. Yield Response. Crop
Science 35: 1094-1099.

Yan, X.L., H. Liao, M.C. Trull, S.E. Beebe, and JP. Lynch. 2001. Induction of a
major leaf acid phosphatase does not confer adaptation to low phosphorus
availability in common bean. Plant Physiology 125:1901-1911.

l Yan, X.L., H. Liao, S.E. Beebe, M.W. Blair, and JP. Lynch. 2004. QTL mapping
of root hair and acid exudation traits and their relationship to phosphorus uptake
in common bean. Plant and Soil 265:17-29.

Youngdahl, L]. 1990. Differences in Phosphorus Efficiency in Bean Genotypes.
Journal of Plant Nutrition 13:1381-1392.

Yu, L.X., J.D. Ray, J.C. Otoole, and HT. Nguyen. 1995. Use of Wax-Petrolatum
Layers for Screening Rice Root Penetration. Crop Science 35:684-687.

Zobel, R.W. 1991. Root Growth and Development, p. 61-71, In P. B. Keister, ed.
The Rhizosphere and Plant Growth. Kluwer Academic Publishers, Boston.

39

 

 

 

 

 

Figure 1. Responses of Arabidopsis root systems to supply of nutrients P, N, and
S. Figure reproduced from Lopes-Bucio et al. (2003).

40

 

Soil phosphorus cycl HARVEST _

H-Iliiiii'
I.Ill|l'|-li a H‘ :1, lillr‘.;:idl'
ll'lllll ‘l'l‘v

    

 

UHHIIH

‘5. l triuiiili ’ ‘ BIOITIE—‘SS

it, Uptake

A'
, *4,”
I".

‘ Weathering

P fixation

Figure 2: The Phosphonrs Cycle (Potash and Phosphate Institute, 2004)

41

s»

lnltlel leelone
i on root: and

Older lesions
and general
discoloration.

    

Chlamydoepore “ {a Chlamydoeporee
i germinatee end ' V " it" « form In Infected
. penetrates nearby = ‘ m roots and

root and hypocotyl .‘ .

. , ......... .. . _

Figure 3: Fusarium root rot infection in beans (Schwartz, 2006)

42

Chapter 2: Fusarium Root Rot Incidence and Root System Architecture in
Grafted Common Bean Lines
Abstract

There is minimal understanding of the complex genotype by environment
interaction controlling Fusarium root rot (caused by F usarium solani f.sp.
phaseoli) disease expression in common bean (Phaseolus vulgaris L.). Severity
of the disease is increased by environmental factors such as soil compaction that
add additional stress to the plant root system. It has proven difficult to develop
resistant cultivars, especially with a bush growth habit as seen in snap and kidney.
bean market classes. One resistant determinate snap bean line, ‘FR266’, has been
developed. The current study used reciprocal grafting techniques with resistant
‘FR266’ and susceptible ‘Montcalm’ genotypes to determine if the genetic control
of resistance lies in the root genotype. The influence of a compacted layer on the
root and shoot genotype and root rot resistance was studied. Root rot resistance
was found generally to be controlled by the root genotype. However, in the
presence of a compacted layer, the root and the shoot genotype play a part in the
resistance. Root traits were also observed under these conditions and root mass
and root diameter Were shown to be controlled by the root genotype without
compaction but by the root and shoot genotype when a compacted. layer was

present.

43

Introduction

Root rot of common bean (Phaseolus vulgaris L.) is caused by a
combination of fungi and environmental stress and is a major limiting factor in
production (Abawi and Pastor Corrales 1990). The complex of pathogens
believed to contribute to root rot include: F usarium solani f.sp. phaseoli,
Rhizoctonia solani Kuhn, Pythium ultimum, flielaviopsis basicola (Berk. & Br.)
Ferr. and Aphanomyces euteiches f.sp. phaseoli. In the majority of dry bean and
snap bean producing regiOns of the United States, F. solani f.sp. phaseoli (Fsp) is
considered the major pathogen and severe infections have caused yield losses of
up to 84% (Chatterjee, 1958; Keenan et al., 1974; Burke and Kraft, 1974;
Steadman etal., 1975; Saettler, 1982). I

Symptoms Of Fusarium root rot include red longitudinal lesions on the
hypocotyls, taproot and lateral roots of the bean plants which lead to rot of
infected roots. Severity of infection increases as the plant develops, and with time
complete rotting of the root system can occur and above ground symptoms
develop including leaf chlorosis, stunting of stems, and reduction in pod
production (Schneider and Kelly, 2000; Burke and Barker, 1966).

Under optimal growing conditions, Fsp is a minor pathogen of beans.
Environmental conditions such as exocssive soil compaction can cause plant
stress and constrain optimal root development and enhance Fusarium root rot
development (Burke, 1968). Soil compaction is an increase in soil bulk density
and a decrease in air filled pores as a result of force applied to the soil; excessive

soil compaction impedes root grth and function (Alhnaras et al., 1988). Soil

44

compaction is frequently present in agricultural fields as a result of the frequent
passage of heavy machinery over soil used for planting, harvest, and other crop
maintenance activities (Harverson et al., 2005). Excessive soil compaction can
slow root tip advance up to 75%, making interception and infection by soil fungal
pathogens much more likely than in non compacted soils (Huisman, 1982,
Allrnaras et al., 1988). In compacted soil in the presence of inoculum of Fsp,
tillage enhanced bean seed yields by 40% and reduced disease severity scores

from 3.4 to 1.4 (on a scale of 0-4, where 0 indicates no disease and 4 indicates
more than 75% of the root system is rotted) over non tilled fields (Harverson et al.,
.2005)

Bean genotypes with the highest level of Fusarium root rot resistance are
late maturing with an indeterminate plant growth habit. It has been difficult to
transfer resistance to beans with a determinate bush type plant growth habit
(Burke and Miller, 1983). FR266, a resistant snap bean was released in 1987 and
derived resistance from N203 a late manning, indeterminate accession collected
in Mexico (Wallace and Wilkinson, 1965) and Phaseolus coccineus (Silbemagel
1987). FR266 has no hypersensitive response, only a slowing of symptom
development. Schneider et a1 (2001) investigated QTL for resistance in a
population of recombinant inbred lines derived from a cross between FR266 and a
highlysusceptible kidney bean variety ‘Montcalrn’. QTL were identified,
however, no single QTL explained more than 15% of the phenotypic variation
and different QTL were identified in greenhouse as compared to field. tests.

Intermediate heritability scores also indicated a high effect of the environment for

45

this trait. Breeders have typically not relied only on root rot resistance scores
from greenhouse screening alone, but have combined root rot scoring with field
screening for stress tolerance that involved compaction and drought interactions
(Burke and Miller, 1983).

The effect of environmental factors such as compaction (Miller and Burke,
1985) on disease severity and the necessity to develop stress tolerant plants in
general raised the question, what traits besides biochemical resistance may be
responsible for the resistance seen in the cultivar FR266. Characteristics such as
root system architecture (Snapp et al., 2003) may be involved. Under field
conditions, in the presence of Fsp chlarnydospores, FR266 had a root dry weight
twice that of Montcalm, along with a greater length and surface area, indicating
an overall larger root system (Roman-Aviles et al., 2004). The average number of
lateral roots in the top 1.0 cm of soil was positively correlated (r2=0.6) with root
rot tolerance in snap beans (Snapp et al., 2003). Lateral root density per volume
of soil may be predictive of stress tolerance rather than resistance to Fsp and has
been associated with drought tolerance in lettuce (Johnson et al., 2000). FR266
also has been shown to grow well in compacted soil (Silbemagel, 1990).
Previous studies on compaction tolerance in corn. (Zea mays L.) and soybean
(Glycine max L.) have identified cultivar specific variability in the ability of roots
to grow through compacted layers (Busharnuka and Zobel, 1998). These findings
suggest that in compacted soils biochemical resistance to a pathogen is not the

only factor required for a plant to grow well.

46

The difficulty in introducing root rot resistance into Andean beans, has led
to the hypothesis that there are inherent physiological factors in F usarium root rot
resistant bean cultivars that are expressed only under field conditions. Root and
shoot growth is genetically integrated, therefore determining if a trait such as
F usarium root rot resistance is expressed in the root, the shoot, or the whole plant
is difficult to discern in whole plant or hybridization studies, but can be
differentiated by interchanging root and shoot genotypes via grafting (White and
Castillo, 1989; Izquierdo and Hosfield, 1982).

The grafting technique has been demonstrated to be a sound tool for
genetic research in common bean. Research has demonstrated the technique is
free of artifacts by employing the self graft as a control for the technique (White
and Castillo, 1989; Izquierdo and Hosfield, 1982). Using the reciprocal grafting
technique to study drought tolerance in common bean, the root genotype, or more
specifically, genes expressed only in the root system, had a large effect on seed
yield under drought conditions. The shoot genotype had some effect on the trait,
but much less than that of the root (White and Castillo, 1989; White and Castillo,
1992). A study of resistance to the soil borne pathogen, Phialophora gregata
which causes brown stem rot in soybean was also found to be controlled by the
root system (Bachman and Nickell, 1999). Pantalone et a1. (1999) found that the
root system architecture trait “fibrous-like rooting” was controlled by the root
system and enhanced seed biomass when grafted onto some soybean genotypes.

In addition there is evidence that the shoot genotype influences the rOot system

47

phenotype as seen with nodule development and N fixation in soybean (Sheng
and Harper, 1997).

The objectives of this research were to determine a) if the expression of
Fusarium root rot resistance was mediated by root or shoot factors and if specific
root architectural traits were associated with resistance in reciprocally grafted
Fusarium root rot resistant and susceptible bean lines and to evaluate b) how root
rot resistance and root architectural traits were mediated by root and shoot factors
when reciprocally grafted plants were grown with the added stress of soil
compaction.

Material and Methods

Plant material

Two common bean genotypes with differences in susceptibility to.
Fusarium root rot were used in this study. They were the snap bean line ‘FR266’
characterized as root rot resistant (Silbemagel, 1987) andthe kidney bean
‘Montcalm’ characterized as root rot susceptible (Schneider et al., 2001). FR266
was developed by USDA/ARS in Prosser, WA (Silbemagel, 1987). Montcalm is
representative of commercial kidney bean types and was developed at Michigan
State University (Copeland and Erdmann, 1977). Select characteristics of FR266
and MOntcalm are described in Table 1.

Reciprocal Grafting

In all experiments reciprocal grafting was employed to determine if
genetic control of Fusarium root rot resistance originated from the shoot or root

system. Plants were grafted in one of four combinations. The first combination

48

was F R266 shoot grafted onto Montcalm root and the second combination was
Montcalm Shoot grafted onto FR266 root. The third and fourth graft types were
self grafts of FR266 shoot grafted onto FR266 root and Montcalm shoot grafted
onto Montcalm root, respectively The self grafts were used as controls to
determine the effect of grafting on root rot and plant grth and these were
compared to ungrafted FR266 and ungrafted Montcalm plants alSo included in all
experiments.

Seeds were surface sterilized by soaking 2 min in 70% ethanol, followed
by 2 min in 0.5% NaOCl. Seeds were rinsed with distilled water 4 times and
incubated at 18°C in the dark in moist germination paper for 2 days. Seeds were
planted as described below for each experiment. Five days after germination,
seedlings were grafted using the cleft graft technique (Izquierdo and Hosfield,
1982). A razor blade was used to cut through the stem 1.5 cm above the cotyledon.
The bottom 1.5 cm of the stem of the shoot was cut into a wedge shape with a
razor blade by removing a thin slice of the epidermis on 2 sides of the stern. A
longitudinal cut of 1.5 cm was made into the top part of the stern of the root
System. The wedge shaped base of the shoot’s stem was placed into the stem
above the root system. The shoot and root were held together with parafilm.

Grafted plants were kept in a controlled environment at a temperature of
25°C day and a relative humidity of 90%. Plants received 14 hr of light (350-450
mE). Plants were misted with water by hand with a spray bottle every 4 hours
until callus tissue formed (5-7 days) and water could be supplied through the root

system.

49

Experiments:

Two distinct controlled environment plant growth experiments were
conducted and each experiment was repeated twice. All experiments were set up
as randomized complete block designs with two treatments, inoculation with
Fusarium solani spores and no inoculation. There were 6 plant genotypes grown
under each treatment, including each of the 4 graft types described above and
ungrafted plants of FR266 and Montcalm. There were four replications of each
treatment/ genotype combination.

Experiment 1 Non-compacted soil:

Media: Sand harvested from Sandhill Farm, Michigan State University, East

I Lansing, MI was sieved, sterilized via autoclaving, and mixed with perlite (60:40)

by volume.

Containers: Seeds were planted into 72 cell flats. Five day old plants were

grafted as described above. One week following grafting, plants were

transplanted into containers referred to as root observation boxes. These

containers promoted the plant’s root system to grow flat and therefore improved

the ability to harvest roots for measurement. The containers were constructed of
glass panes 21.6 cm by 27.9 cm separated by 5.1 cm wooden strips for a

dimension of 5.1 x 21.6 x 27.9 cm (Figure 1a).

Plant growth conditions: Root observation boxes were completely covered with

aluminum foil to prevent light from interfering with root growth. Individual root
observation boxes were placed together into the growth chamber with other boxes,

each at a 45° angle. Plants were exposed to 14 h of light (350-450 mE) at 28°C

50

and 10 h dark cycle at 25°C. Plants were watered as needed and were fertilized
weekly with ‘/2 strength Hoagland’s solution (Hoagland and Arnon, 1938).
Fusarium spore preparation and inoculation: Fourteen day old plants were
inoculated with 15 ml of 2.0 x 105 conidia ml'l suspension of a mix of 2 isolates of
F. solani f. Sp. phaseoli (Fsp). The Hawks 2B isolate was collected by Schneider
and Kelly (2000) in Presque Isle County, MI and S-602 was received from the I
Penn State Fusarium Research Center, University Park, PA. Potato dextrose agar
(PDA, Invitrogen TM, Life Technologies, MD; 39 g L'1 water) was the media used
to grow the Fusarium colonies. Inoculum discs, 4 mm diameter were cut from
margins of 1- week old cultures grown on PDA and transferred mycelial side
down, to the. center of individual Petri plates (15 m depth x 90 mm diameter)
containing 25 ml of PDA. Plates were stored at room temperature for 2 weeks
prior to inoculum preparation. The inoculum was prepared by scraping mycelium
and conidia from PDA plates into distilled water, then strained through
cheesecloth. The conidia] concentration was adjusted to 2 x 105 rrrl'1 using a
hemocytometer. Fifteen ml of the inoculum was applied to the soil at the base of
the plant.

Harvest: Plants were harvested 32 days after planting and shoots were separated
from roots and soil was washed from the root system with distilled water. The

experiment was carried out twice, in August 2003 and in August 2004.

51

Experiment 2: Compacted soil layer:
Media: The growing media was a Capac loam harvested from the Michigan State
University Horticulture Farm, East Lansing, MI. soil was sieved, autoclaved, air
dried and brought to 18% moisture before adding to containers.
Containers: PVC tubes of 7.6 cm diameter and 0.5 cm wall thickness were used
for plant grth containers based on the design of Asady et al (1985). The
containers consisted of three layers. The top layer was 12.5 cm long, the middle
layer was 2.6 cm long, and the bottom layer was 15 cm long (Figure lb). Soil
was added to the top and bottom layers to achieve a bulk density of 1.1g/cm3. The
middle layer was compacted to a bulk density of 1.7 g/cm3 with a hydraulic
compactor machine. Bulk densities of each layer were determined based on soil
weight. The bulk density of the compacted layer was chosen because the soil
strength at that density reduces common bean root growth (Asady et al., 1985).
The three layers of the container were held together with duct tape.
Fusarium spore preparation and inoculation: The same mix of 2 isolates of F.
solani f. sp. phaseoli (Fsp) were used in this experiment as described in
Experiment 1, Fusarium spore preparation and inoculation. Colonies were first
grown on potato dextrose agar (PDA, Invitrogen TM, Life Technologies, MD; 39 g
L'1 water). After one week of growth, fusariurn inoculum discs, 4 mm diameter
were cut from margins and transferred mycelial side down to sterile pearl millet
seed prepared as follows. Millet seed (1000 ml) was poured in an aluminum pan
(30.5 cm x 25.4 cm x 11.4 cm) mixed with 600 ml of distilled water, covered with

aluminum foil and was kept at room temperature for 12 h. The pan was

52

autoclaved for 4 h and cooled 4 h. Next, 12 inoculum disks (1.3 cm wide) were
evenly spaced over the seed. Two-week old cultures were used to prepare
inoculum. Millet seed was mixed to homogeneity. Mycelium and conidia of a 1
gm sample of the seed was collected in distilled water. The inoculum
concentration of the sample was determined with a hemocytometer. The weight of
seed for a conidia] concentration of 2 x 105 ml'1 was determined. The
inoculum/millet mixture was combined with the soil for the top layer of the
container.

Plant growth conditions: Surface sterilized pregerminated seeds were planted
directly into containers, two seeds per container. After germination, seeds were
thinned to one per container. Grafting, as described above, was conducted
directly in containers. Plants were exposed to 14 h of light (350—450 mE) at 28°C
and 10 h dark cycle at 25°C. Plants were watered as needed. Percent moisture of
the soil was kept at 18% and the amount of water to add to each plant was
determined by container weight. Plants were fertilized weekly with 1/2 strength
Hoagland’s solution (Hoagland and Amon, 1938).

Harvest: Thirty-two day old plants were harvested and shoots were separated
fiom roots. Roots were harvested separately from each of the three layers of the
containers and soil was washed from the roots with distilled water. The
experiment was carried out once in November 2005. In October, 2004 the

experiment was carried out with ungrafted and reciprocal grafts only.

53

Plant Measurements:

Following harvest of plants in each experiment, root rot scores were
measured based on a scale of l to 7 as described by Schneider et al (2000) where
“1: healthy root system with no discoloration of root or hypocotyls tissue and no
reduction in root mass compared to the uninoculated control; 2 =appearance of
small reddish-brown lesions, 0.1-0.2 cm in length, at base of hypocotyls with size
and appearance of root mass normal; 3 = increase in intensity and size and
coalescing of localized root/hypocotyl lesion circling 3 180° of stem with lesions
from 0.5-1cm and 10% to 20% discoloration on roots but no reduction in size of
root mass; 4 =increasing intensity of discoloration and size of hypocotyl lesions
with lesions becoming extended and completely encircling the stem; 5% to 10%
reduction in root mass with 95% of the roots discolored, 5 = increasingly
discolored and extended hypocotyls lesions. 100% of the roots intensely reddish-
brown with a 20% to 50% reduction in root mass, 6 = hypocotyl lesions encircling
stem become more extended (2 cm); root mass is intensely discolored and reduced
from 50% to 80%; 7 = pithy or hollow hypocotyl with very extended lesions,
where root mass is reduced from 80% to 100% and is functionally dead.”

Following root rot scoring, cleaned roots were placed in a transparent
plastic tray (22 x 28 x 6.5 cm) and covered with water. The tray containing roots
was then scanned with a flatbed scanner with a top and bottom light source at 300
dpi to produce a 2 dimensional scanned image. The images were analyzed with
WinRHIZO software (Regent Instruments Inc., Quebec, Canada). With this

program root length, root surface area and average root diameter of the samples

54

was determined. Following image capture, roots and shoots were dried at 65°C
for two days and dry weights were determined.

Data Analysis:

Statistical analysis was conducted with SAS for Windows V8 (SAS
Institute, Cary, NC). The command PROC GLM was used to determine treatment,
genotype and interaction effects. An initial model with 2 runs, 4 replications, 2
treatments, and 6 genotypes, was run for experiment 1 to determine the effect of
Fusarium inoculation and grafting on plant growth. The same model was
developed for experiment 2, but there was 1 run, instead of 2 runs in the model. A
second model was then developed with 2 runs (1 run for experiment 2), 4
replications, 1 treatment (inoculated plants only) and 2 shoot genotypes (FR266 or
Montcalm) and 2 root genotypes (FR266 or Montcalm. Ungrafted plants and
uninoculated plants were excluded from this model. This model was used to
determine the specific effects of each root and shoot genotype in the presence of
Fusarimn inoculum with out confounding results with ungrafted plants. Least
significant differences (LSD) values were used to determine significant
differences among treatments.

Results:

Experiment 1: Consistent with earlier results (Schneider et al., 2001),
ungrafted Montcalm plants grown in the presence of Fsp spores had a
significantly higher root rot score than ungrafted FR266 plants (Table 2). No
significant differences were observed between FR266 and Montcalm for root dry

weight in the presence of Fsp inoculum, which is not unexpected based on length

55

of experiment. Root measurements were taken on 32 day old plants, and at this
time root rot symptoms of root lesions and discoloration were visible, but a longer
grth period would be required to observe differences in grth response.
Similar results have been observed by other researchers (Roman-Aviles et al.,
2004)

Root system architectures of FR266 and Montcalm were characterized by
root weight, length, and diameter. The most striking difference between FR266
and Montcalm root architecture was average root diameter. In both inoculated
and uninoculated plants, FR266 had a greater average root diameter than
Montcalm (Table 3).

The reciprocal grafting technique was used with FR266 and Montcalm
plants to understand genetic control of resistance to Fusarium root rot and root
architecture. Self grafts of each genotype were included and compared to the
ungrafted genotypes as controls to determine if the technique itself impacted
reaction to Fusarium. The ungrafted and grafted plants did not differ in root rot
scores (Table 2). The ungrafted plants exhibited larger root dry weights than their
self grafted counterparts (Table 2).

The comparison of root traits in self grafts and reciprocal grafts in the
presence of Fsp inoculum showed that root rot score is controlled by the root
1 genotype (Table 4) and it was the F R266 root in any combination that exhibited
the lower score (Tables 4 and 5). Root dry weight was also controlled by the root
genotype (Table 4) and the FR266 root in any graft combination produced a larger

root system that the Montcahn root (Table 5). Average root diameter was

56

controlled by the shoot genotype (Table 4). The FR266 shoot in any graft
combination increased the average diameter of the root system (Table 5). For
each of the traits described above, there were no significant root*shoot
interactions, which allowed for the presentation of data averaged over root or
shoot genotype as in Table 5.

Experiment 2: Ungrafted Montcalm plants had higher root rot incidence
than FR266 plants when they were grown in containers with a compacted soil
layer (1.7 g /cm3 ) at 12.5 cm below the surface of the container (Table 6).

Differences in root architecture between FR266 and Montcalm were
observed in the soil below the compacted layer. In the presence of Fsp inoculum,
F R266 plants had greater total root length and average root diameter than
Montcalm plants (Table 7).

Comparison of ungrafted and self grafted plants indicate that grafting did
not affect root rot scores (Table 7). Grafting, however, reduced shoot and root
dry weight of both FR266 and Montcahn plants (Table 7).

Root rot incidence of grafted plants in this experiment was affected not
only by the root genotype as it was in Experiment 1, but was affected by root and
shoot genotype (Table 8). The graft combination FR266 root with Montcalm
shoot had the same root rot incidence as the graft combination Montcalm root
with FR266 shoot.

The root architecture variables root dry weight, root length, and average
root diameter also were influenced not by root or shoot genotype alone, but by the

interaction of root and shoot genotypes (Table 8). The graft combination

57

Montcalm root/F R266 shoot, which had a different reaction to root rot in this
experiment than in Experiment 2, also exhibited greater root dry weight and root
length in the middle compacted and bottom layers of soil than the Montcahn self
graft (Table 8). Ungrafted and self grafted Montcalm plants exhibited reductions
in root length below the compacted soil layer in the inoculated treatment (Tables

7 and 8) not exhibited by the Montcalm root/FR266 shoot grafted plants (Table 8).

The Montcalm root/F R266 shoot graft combination did not, however, have
a larger average root diameter than the Montcalm self graft. Each of the four graft
combinations had consistently similar root diameters in this experiment, except in
the top layer where the Montcalm root/F R266 shoot combination had a lower
average root diameter (Table 8). 9
Discussion:

Experiments 1 and 2 showed that root rot incidence was controlled by
genes expressed only in the root system in the absence of a compacted layer and
by genes expressed throughout the plant in the presence of a compacted layer.
These results demonstrated that different mechanisms of resistance to F sp were at
work under different enviromnental conditions. I looked at a few basic root
architectural traits in each experiment to determine their involvement in F sp
resistance.

One of the potential Fsp resistance mechanisms of FR266 is its thick roots
(Snapp etal., 2003). The reciprocal grafting technique alloWed the testing of the
hypothesis that thicker roots have less root rot incidence than thinner roots. In

Experiment 1, it was found that the root genotype dictated root rot score, whereas

58

the shoot genotype dictated average root diameter. Since the graft combination
Montcalm root/FR266 shoot had thicker roots than the Montcalm self graft, but
did not have reduced root rot incidence, therefore root thickness was not a factor
in root rot resistance under the experimental conditions. It would be interesting to
determine if the same relationship holds true later in plant development when root
rot symptoms are more severe.

In Experiment 2, there were no clear differences in root diameter between
the genotypes. In the compacted layer there was a uniform reduction of average
root diameter of each graft combination, indicating both FR266 and Montcalm
responded in the same way to compaction. Small diameter roots have the ability
to grow through smaller pores in compacted soils than larger diameter roots
(Bennie, 1991). This may be the reason why a reduction in average root diameter
was observed in the compacted layer. Average root diameter was influenced by
the root and shoot genotypes. Gibberellin (GA) may be a possible signal from the
shoot that influenced average root diameter in this study. Gibberellin moves from
shoot to root and vice versa and inhibitors of GA biosynthesis have been shown to
increase root diameter (Tanimoto, 2005).

Total root length was not an indicator of root rot resistance in Experiment
_ 1. In Experiment 2, however, increased root length below the compacted layer
was observed in the resistant genotypes and graft combinations, indicating that
FR266 (in any root or shoot combination) was better able to penetrate the
compacted layer than Montcalm. One of the mechanisms by which roots move

through soils with an increased bulk density and a smaller pore size is by

59

increasing root diameter and displacing soil to push through an area of
compaction (Goodman and Ennos, 1999).

The different interaction of root and shoot signals under varying soil
environmental conditions is a likely reason for the difficulty in developing
common bean lines tolerant to Fusarium root rot.

Conclusions

Root genotype controlled root rot incidence in the absence of compaction,
but with the addition of a compacted soil layer, the interaction of the root and
shoot genotype dictated root rot incidence. In the absence of compaction, root
traits were clearly influenced by root or shoot genotype. In the soil with a
compacted layer, interaction of the root and shoot genotypes was important in

determining root system architecture.

60

References:

Abawi, GS, and MA. Pastor Corrales. 1990. Root rots of beans in Latin
America and Afiica: Diagnosis, research methodologies, and management
strategies CIAT, Cali, Colombia.

Allmaras, R.R., J.M. Kraft, and DE. Miller 1988. Effects of Soil Compaction
and Incorporated Crop Residue on Root Health. Ann. Rev. Phytopathol. 26:219-
243.

Asady, G.H., A.J.M. Smucker, and M.W. Adams. 1985. Seedling Test for the
Quantitative Measurement of Root Tolerances to Compacted Soil. Crop Science
' 25:802-806.

Bennie, ATP. 1991. Growth and Mechanical Impedance. Pages 393-414 in:
Plant Roots, the Hidden Half. Y.Eisel, A. Eisel, and U. Kaflcafi, eds. Marcel
Dekker, Inc., New York.

Bachman, MS. and Nickell, CD. 1999. Use of Reciprocal Grafting to Study
Brown Stem Rot Resistance in Soybean. Phytopathology 89:59-63

Burke, D. W. 1968. Root Growth Obstructions and Fusarium Root Rot of Beans.
Phytopathology 58: 1575- 1576.

Burke, D.W., and A.W. Barker. 1966. Importance of Lateral Roots in Fusarium
Root Rot of Beans. Phytopathology 56:292-294.

Burke, D.W., and J .M. Kraft. 1974. Responses of Beans and Peas to Root
Pathogens Accumulated during Monoculture of Each Crop Species.
Phytopathology 64:546-549.

Burke,.D.W., and DE. Miller. 1983. Control of Fusarium Root Rot with Resistant
Beans and Criltural Management. Plant Disease 67: 1312-1317.

Bushamuka, V.N. and Zobel R.W. 1998. Differential Genotypic and Root Type
Penetration of Compacted Soil Layers. Crop Sci. 38:776-781.

Chatterj ee, P. 1958. The Bean Root Rot Complex in Idaho. Phytopathology
48: 197-200.

Copeland, L. O., and M. H. Erdmann. 1977. Montcalm and Mecosta, halo blight
tolerant kidney bean varieties for Michigan. Ext. Bull. 957. Michigan State
University, East Lansing

Goodman, A.M., and AR. Ennos. 1999. The effects of soil bulk density on the

61

morphology and anchorage mechanics of the root systems of sunflower and maize.
Annals of Botany 83:293-302.

Harveson, R.M., Smith, J .A., and Stroup, W.W. 2005. Improving Root Health
and Yield of Dry Beans in the Nebraska Panhandle with a New Technique for
Reducing Soil Compaction. Plant Dis. 89: 279-284.

Hoagland, D. R. & Amon, D. I. (1938) The Water-Culture Method for Growing
Plants Without Soil (Univ. of California College of Agriculture Agricultural
Experiment Station, Berkeley, CA).

Huisman, CC. 1982. Interrelations of Root Growth Dynamics to Epidemiology
of Root Invading Fungi. Ann. Rev. Phytopathol. 20:303-327.

Izquierdo, J .A., and G.L. Hosfield. 1982. A Simplified Procedure for Making
Cleft Grafts and the Evaluation of Grafting Effects on Common Bean.
Hortscience 17:750-752.

Johnson W.C., L.E. Jackson, 0. Ochoa, R. van Wijk, J. Peleman, D.A. St Clair,
R.W. Michelmore 2000. Lettuce, a shallow-rooted crop, and Lactuca serriola, its
wild progenitor, differ at QTL determining root architecture and deep soil water
exploitation. Theor. App. Genet. 101: 1066-1073.

Keenan, J .G., H.D. Moore, N. Oshima, and LE. Jenkins. 1974. Effect of Bean
Root Rot on Dryland Pinto Bean Production in Southwestern Colorado. Plant
Disease Reporter 58:890-892.

Miller, DE, and D.W. Burke. 1985. Effects of Soil Physical Factors on
Resistance in Beans to Fusarium Root-Rot. Plant Disease 69:324-327.

Pantalone, V.R., Rebetzke, G.J., Burton, J .W., Carter, TE, and Israel, D.W. 1999.
Soybean PI 416937 Root System Contributes to Biomass Accumulation in
Reciprocal Grafts. Agron.J. 91: 840-844.

Roman-Aviles, B., Snapp, SS, and Kelly, J .D. 2004. Assessing root traits
associated with root rot resistance in common bean. Field Crops Res. 86:147-156.

Saettler, A.W. 1982. Bean Disease and Their Control, In (L. S. a. F. Robertson,
R.D., ed. Dry Bean Production: Principles and Practices. MSU Cooperative
Extension, East Lansing.

Schneider, K.A., and J .D. Kelly. 2000. A Greenhouse Screening Protocol for
Fusarium Root Rot in Bean. HortScience 35: 1095-1098.

Schneider, K.A., K.F. Grafton, and J .D. Kelly. 2001. QTL Analysis of Resistance
to Fusarium Root Rot in Bean. Crop Sci. 41 :535-542.

62

Sheng, C. and J .E. Harper 1997. Shoot versus R00t Signal Involvement in
Nodulation'and Vegetative Growth in Wild—Type and Hypemodulating Soybean
Genotypes. Plant Physiol. 113:825-831.

Silbemagel, M.J. 1987. Fusarium Root Rot-resistant Snap Bean Breeding Line
FR266. HortScience 22:1337-1338.

Silbemagel, M.J. and LJ. Mills 1990. Genetic and Cultural Control of Fusarium
Root Rot in Bush Snap Beans. Plant Dis. 74: 61-66

Snapp, S., W. Kirk, B. Roman-Aviles, and J. Kelly. 2003. Root Traits Play a Role
in Integrated Management of F usarium Root Rot in Snap Beans. HortScience In
Press. '

Steadman, J .R., E.D. Kerr, and RF. Murnm. 1975. Root Rot of Bean in Nebraska:
Primary Pathogen and Yield Loss Appraisal. Plant Disease Reporter 59:305-308.

Tanimoto, E. 2005. Regulation of root grth by plant hormones - Roles for
auxin and gibberellin. Critical Reviews in Plant Sciences 24:249-265.

White, J .W. and Castillo, J .A. 1989. Relative Effect of Root and Shoot Genotypes
on Yield of Common Bean under Drought Stress. Crop Sci. 29:360-362.

Wallace, DH, and RE. Wilkinson. 1965. Breeding for Fusarium Root Rot
Resistance in Beans. Phytopathology 55: 1227-1231.

63

Table 1. Selected plant and seed characteristics of the common bean cultivars

 

 

 

FR266 and Montcalm.
Cultivars
FR266 Montcalm
Reaction to resistant susceptible
F usarium solani
Gene pool Andean Andean
Market class Snap Kidney
Plant growth habit determinate determinate
Seed color white red
Seed size medium large
100 seed weiLht 30.7 g 61.3 g

 

64

Table 2. Mean response root rot score, shoot dry weight, and root dry weight of
different graft combinations of common bean lines FR266 and Montcalm in the
presence (+ F us) or absence (-Fus) of F usarium solani f.sp phaseoli inoculum.

Data analyzed across 2 runs of experiment 1.

 

 

 

 

 

Root rot Shoot dry Root dry
score (1-7)j weight (mg) weight (mg)
Graft combination
Root Shoot -Fus +Fus -Fus +Fus -F‘us + Fus
FR266 ungrafted 1.5 2.9 785 834 604 580
FR266 FR266 1.0 2.1 634 718 330 413
FR266 Montcalm 1.4 2.5 726 831 470 526
Montcalm ungrafted 1.6 4.1 770 868 449 530
Montcalm Montcalm 1.8 4.5 816 782 381 371
Montcahn FR266 1.6 4.3 705 710 459 408
LSD 0.051 0.84 172 132
ANOVA

Source df p value

Run (R) 1 0.0001 0.0006 <0.0001
Replication 3 0.8838 0.0649 0.8174
Graft (G) 5 <0.0001 0.0461 <0.0001
Fusarium (F) 1 <0.0001 0.1068 0.3075
G*F 5 0.0018 0.7542 0.3324
R*G 5 0.5961 0.4347 0.1013
R*G*F 5 0.3825 0.1603 0.0264

 

T Root rot score is on a scale of 1-7, where 1 is no root rot and 7 is completely

rotted ( Schneider and Kelly, 2000).

I LSD value to compare any values within root rot score, shoot dry weight, and

root dry weight variables.

65

Table 3: Mean response for total root length and average root diameter of
ungrafted Montcalm and F R266 in the presence (+ Fus) or absence (-Fus) of
F usarr’um solani f.sp phaseoli inoculum. Data analyzed across 2 runs of
experiment 1. '

 

 

 

Genotype Total root Average root
length (cm) diameter (mm)
-Fus +Fus -Fus +Fus

FR266 3475 at 3402 a 0.461 a 0.467 a

Montcalm 3273 a 2990 a 0.417 b 0.399b

 

T= Values that do not share a letter are significantly different as determined by
LSD (0.05). Letters are compared down a column.

66

Table 4: Mean response and analysis of variance for root traits of different graft
combinations of common bean lines FR266 and Montcalm in the presence of

F usarium solani f.sp phaseoli inoculum. Data analyzed across 2 runs of
experiment 1 and the category graft combination is separated into root and shoot
components for analysis to determine which components are affecting a response.

 

Graft Combination

 

 

 

 

 

Root Shoot Root Rot Root dry Root Avg root
Score wt. (mg) length diam.
(1-7)L (cm) me)
FR266 FR266 2.1 at 453 ab 3070 ab 0.45 a
FR266 Montcalm 2.5 a 527 b 3209 b 0.44 a
Montcalm FR266 4.3 b 351 a 2920 ab 0.46 a
Montcalm Montcalm 4.5 b 371 a 2689 c 0.40 b
ANOVA
Source df p value
Run (R) l <.0001 <.0001 0.04 <.0001
Replication 3 0.40 0.95 0.18 0.17
Root 1 <.0001 0.01 0.17 0.23
Shoot 1 0.26 0.23 0.86 0.04
R*Root l 0.82 0.18 0.67 0.29
R*Shoot l 0.26 0.1 1 0.47 0.98
Root*Shoot l 0.82 0.34 0.42 0.13
R*Root* l 0.26 0.75 0.81 0.61
Shoot

 

1' Root rot score is on a scale of 1-7, where 1 is no root rot and 7 is completely
rotted (Schneider and Kelly, 2000).

I Values that do not share a letter are significantly different as determined by
LSD (0.05). Letters are compared down a column.

67

Table 5: Mean response for root traits of different graft combinations of common
bean lines FR266 and Montcalm in the presence of F usarium solani f.sp phaseoli
inoculum. Data analyzed across 2 runs of experiment 1 and the category graft
combination is separated into root and shoot components for analysis to determine
which components are affecting a response, means are‘averaged by root genotype
and shoot genotype.

 

 

 

 

Root 1' Root Rot Root dry f Root Avg. root

Score (1-7) * wt. (mg) length diam.
(cm) (mm)

FR266 2.31 a** 492 a 3144 a 0.446 a

Montcalm 4.38 b 362 'b - 2805 a 0.426 a

Shoot it

FR266 3.19 a 402 a 2991 a 0.455 a

Montcalm 3.50 a 449 a 2949 a 0.418 b

 

T Root heading indicates that means of each variable are averaged over root
genotype.

i Shoot heading indicates that means of each variable are averaged over shoot
genotype. .

* Root rot score is on a scale of 1-7, where 1 is no root rot and 7 is completely
rotted (Schneider and Kelly, 2000).

** Values that do not share a letter are significantly different as determined by
LSD (0.05). Letters are compared down a column; comparisons under ‘Root’ are
separate from those under ‘Shoot’.

68

Table 6. Mean response of root rot score, shoot dry weight, and root dry weight of
different graft combinations of common bean lines FR266 and Montcahn grown
in containers with a layer of compacted soil in the presence (+ Fus) or absence (-

Fus) of F usarium solani f.sp phaseoli inoculum, Data analyzed across 1 run (Nov.
2005) of experiment 2.

 

Root rot Shoot dry Root dry
score (1 -7) 1‘ weight (mg) weight (113)

Graft combination

Root Shoot -Fus +Fus -Fus +Fus -Fus + Fus
FR266 ungrafted 1.0 2.4 1424 1436 350 398
FR266 FR266 1.0 2.3 717 647 129 153
FR266 Montcahn 1.0 2.5 . 710 651 140 144

Montcalm ungrafted 1.0 4.0 1158 932 198 256
Montcalm Montcalm . 1.0 4.5 61 1 321 129 1 15
Montcahn F R266 1.0 2.5 700 692 1 10 129

 

 

 

LSD 0.05:1: 0.74 252 94
ANOVA
Source (11‘ ' p value
Replication 3 0.5330 0.2752 0.8888
Graft (G) 5 <0.0001 <0.0001 <0.0001
Fusarium (F) 1 <0.0001 0.0230 0.2062
G*F 5 <0.0001 0.3361 0.8516

 

T Root rot score is _on a scale of 1-7, where 1 is no root rot and 7 is completely
rotted (Schneider and Kelly, 2000).

1 LSD value to compare any values within root rot score, shoot dry weight, and
root dry weight variables.

69

Table 7: Mean response for total root length and average root diameter of
ungrafted Montcalm and F R266 grown in containers with a layer of compacted
soil in the presence (+ Fus) or absence (-F us) of F usarium solani f.sp phaseoli
inoculum, Data analyzed across 2 runs of experiment 2.

 

 

 

Soil layer Genotype Total root Average root
lengh (cm) diameter (mm)
Top -Fus +Fus -Fus +Fus
FR266 1577 at 677 a 0.454 a 0.43 a
Montcalm 1412 a 525 a 0.419 b 0.42 a
Middle
FR266 181 a 215 a 0.380 a 0.37 a
Montcalm 248 a 180 a 0.344 a 0.37 a
Bottom

FR266 2483 a 4322 a 0.406 a 0.45 a
Montcalm 2141 a 3082 b 0.376 a 0.40 b

TValues that do not share a letter are significantly different as determined by LSD
(0.05). Letters are compared down a column.

70

Table 8. Mean response and analysis of variance for root traits of different graft
combinations of common bean lines FR266 and Montcalm grown in containers
with a layer of compacted soil in the presence of F usarium solani f.sp phaseoli
inocultun. Data analyzed across 1 run (Nov. 2005) of experiment 2 and the
category graft combination is separated into root and shoot components for
analysis to determine which components are affecting a response.

 

Graft combination

 

 

 

 

 

 

 

 

Soil IayerT Root Shoot Root Root Root Avg
Rot dry wt. length root
Score (mg) (cm) diam.
(1-7)l (mm)
Top FR266 FR266 2.25 a * 89 a 461 a 0.44 a
FR266 Montcalm 2.50 a 84 a 326 a 0.46 a
Montcahn FR266 2.50 a 50 b 449 a 0.39 b
Montcahn Montcalm 4.50 b 78 a 436 a 0.46 3
AN OVA Source df p value
Root (R) 1 0.002 0.019 0.481 0.0019
Shoot (S) 1 0.002 0.213 0.300 0.004
R*S 1 0.0113 0.063 0.387 0.053
Middle FR266 FR266 NA 7 ab 84 ab 0.37 a
compacted FR266 Montcalm -- 9 a 94 ab 0.37 a
Montcalm FR266 -- 7 ab 119 a 0.36 a
Montcalm Montcalm -- 5 b 40 b 0.37 3
ANOVA Source (if p value
Root (R) 1 NA 0.07 0.698 0.75
Shoot (S) 1 -- 0.917 0.177 0.95
R*S 1 -- 0.099 0.087 0.77
Bottom FR266 FR266 NA 57 a 1204ab 0.36 a
FR266 Montcalm -- 95 b 1667 a 0.36 a
Montcahn FR266 -- 76 ab 1568 a 0.34 a
Montcalm Montcalm -- 28 c 667 b 0.37 a
ANOVA Source df p value
Root (R) 1 NA 0.018 0.252 0.49
Shoot (S) 1 -- 0.534 0.421 0.33
R*S 1 -- 0.006 0.026 0.07

 

 

 

T Soil layer is defined as top, middle, and bottom. Root traits were measured
separately for each layer.
I Root rot score is on a scale of 1-7, where l is no root rot and 7 is completely
rotted (Schneider and Kelly, 2000).
*Values that do not share a letter are significantly different as determined by LSD
(0.05). Letters are compared down a column.

71

 

Figure 1. Photos of containers used in Experiment 1 (a) and Experiment 2 (b).
The bulk densities of soil are listed in g/cm3 for each of the 3 layers of containers
used in experiment 2.

72

Chapter 3: The Relationship among Root Architecture Traits, Plant Growth
Habit and Tolerance to Low Soil Phosphorus Levels in an Andean Bean
Population
Abstract:

Tolerance to low P soils is a desirable characteristic for common bean line
grown in acid-weathered soils of the tropics. Variability for tolerance to low P
soils exists in the Andean gene pool. The objective of this research was to
identify mechanisms of tolerance to low P soils in a By recombinant inbred line
(RH) population developed fiom two common bean lines in the Andean gene
pool, AND696, an improved line, susceptible to low P soils, with a determinate
grth habit and G19833, a landrace tolerant to low P soils, and an indeterminate
growth habit. The 77 RILs were grown in low and high soil phosphorus levels in a
field in Darien, Colombia. P uptake and rootarchitecture traits were measured in
the population, along with P uptake efficiency and seed P content. Genetic
differences were identified for root architecture traits, including root length
density, specific root length, and average root diameter in the top 16 cm of soil.
Specific root length was weakly correlated with P uptake (r=0.16) in both the high
and low P treatments. P uptake and P use efficiency were correlated with yield.
Plant growth habit affected seed yield, and root length density, but was not related

to P uptake.

73

Introduction:

Common bean (Phaseolus vulgaris L.) is an important grain legume
worldwide, produced on 20 million hectares of land each year. The highest
production and consumption of common bean occurs in Latin America (5.6
million metric tons) and Afi'ica (2.8 million metric tons) (FAOSTAT, 2006).

One of the major constraints to bean production in Afi'ica and Latin
America is low available soil phosphorus (Wortmann et al., 1998). An estimated
43% of the lands in the tropics are acid weathered soils, specifically Andosols,
Ultisoils, and Oxisols, that adsorb or fix P so that 70 to 90% of P fertilizer applied.
reacts with iron or aluminum to form compounds of low solubility (Sanchez and
Salinas, 1981). Common bean is susceptible to low P soils in part due to the high
P requirement associated with energy-intensive nodulation and N2 fixation
(Christiansen and Graham, 2002).

Tolerance to low P soils or P efficiency is defined as the ability to
produce plant growth and yield in relation to the amount of available phosphorus
(Lynch and Beebe, 1995) and can occur by two distinct routes: acquisition
efficiency and utilization efficiency. Acquisition efficiency reflects the plants
ability to extract P fiom the environment. It has been shown to be related root
system traits that increase the root surface area and allow capture of more P from
the soil (Gahoonia and Nielsen, 2003). Utilization efficiency is the superior
' ability of a plant to convert acquired P into plant biomass and yield and is related

to reduced tissue P requirement (Lynch and Beebe, 1995).

74

Genetic variability for tolerance to low P soils has been identified in
common bean (Beebe et al., 1997; Singh et al., 2003). Within the primary gene
pool of common bean there are multiple sites of domestication divided into two
major groups, Mesoamerican (northern Mexico and Central America) and Andean
(South America) (Singh, 2001). The most studied bean lines tolerant to low P
soils have been landraces from the Andean gene pool (Lynch and Beebe, 1995;
Yan et al., 1996; Bonser et al., 1996; Yan et al., 2004; Beebe et al., 2006).

The Andean landraces identified as tolerant to low P soils have been
shown to be efficient in P acquisition from the soil (Liao and Yan, 2001; Liao et
al., 2001). Controlled environment physiology studies have identified specific
plant root traits related to P uptake in these P efficient lines, including increased
lateral and adventitious root number and more shallow basal roots. (Lynch and
Brown, 2001; Miller et al., 2003). QTL studies employing inter-gene pool crosses
have identified regions of the bean genome important for both root grth and P
uptake (Liao et a1, 2004; Beebe et al., 2006; Ochoa et al., 2006). Within a -
recombinant inbred line population developed fi'om a P efficient, Andean bean,
G19833 and a P inefficient, Mesoamerican bean DOR364, QTL were identified
for total root length and specific root length on two different linkage groups that
co-localized with QTL for phosphorus uptake. The uptake QTL and the root QTL
each explained about 15% of phenotypic variation for these traits (Beebe et al.,
2006).

Low P tolerant, Andean landraces perform well in low P soils, but

typically do not yield well in P sufficient soils. They have an indeterminate

75

growth habit, long maturation time, and are day length sensitive (Beebe et al.,
1997). There is question as to whether the features of these landraces that enhance
yields in low P environments are the same features that limit their yield potential
in other environments. Previous observational studies with common bean support
a coordinated grth habit of the root and shoot such that indeterminate plant
types with highly branched shoot architecture and little apical dominance also
exhibited highly branched root architecture (Lynch and van Beem, 1993). The
parental material used in this study had contrasting plant growth habit, and the
population was segregating for grth habit, which allowed the opportunity to
look at the relationship among grth habit, root architecture, and P acquisition
efficiency.

The first objective of this study was to identify the broad mechanisms (P
acquisition and P use efficiency) of tolerance to low P soils in a bean recombinant
inbred line (RIL) population developed from an Andean intra-gene pool cross
between a P efficient, unadapted landrace, G19833 and a P inefficient line,
AND696. Specific root architecture traits were also examined for their
relationship to P acquisition efficiency. The RH. population was grown in both P
deficient and P sufficient soils and plants were evaluated in each environment to
clearly identify traits generally important for plant yield vs. those Specifically
important for adaptationto low P soils.

The second objective of this study was to determine if there were
differences in shoot growth habit associated with differences in root growth and if

they were related to P uptake in common bean.

76

Materials and Methods:

Seventy seven F 5:7 recombinant inbred lines were developed from a cross
between two common bean lines from the Andean gene pool. The initial cross
G19833 x AND696 was advanced to the F 5 generation by single seed descent and
seed of the RILS was increased for field studies. The parent G19833 is a Peruvian
landrace with an indeterminate (Type III) growth habit. G19833 has large yellow
and red mottled seed with an average 100 seed weight of 41 g. G19833 is
relatively unadapted (i.e. low yield potential) but has been identified as tolerant to
low P soils (Y an et al., 199Sab). The parent AND696 is a CIAT improved line
from the race Nueva Granada. It has a determinate grth habit (Type I) and has
large red and white mottled seed with an average 100 seed weight of 51 g.
AND696 has been identified as susceptible to low phosphorus soils (CIAT, 2000).

The RIL population developed from G19833 and AND696 segregated for
plant growth habit, 56 of the lines have a determinate growth habit and 21 lines
have an indeterminate growth habit. Growth habit in common bean is controlled
by a single gene, fin (Norton, 1915). The number of determinate and
indeterminate lines deviated significantly from a 1:1 ratio that is expected for a
single gene trait. This deviation is likely due to the selection of RILs with i
uniform days to maturity, which biased the selection to favor detemrinate lines.

In 2000 and 2005, the 77 RILs, G19833, AND696, and two check
varieties, G4017 (Carioca) and Gl6140 were planted in Darien, Colombia in a
9x9 lattice design with three replications at two soil P levels, low P (45 kg/ha

triple super phosphate) and high P (300 kg/ha triple superphosphate). Phenotypic

77

data was collected on 75 of the 77 RILS. The soil of this site in an Andisol with a
native soil P of 2 mg/kg based on Bray II extraction method. Seed was hand-
planted at 10 cm spacing in 4 row plots. Each row was 3 m in length. The middle
tworows were planted to the genotypes of interest and the outer two rows were
border rows of Mesoamerican tan seeded cultivar BAT 477. The border rows
were included to improve uniformity in soil P availability to the different
genotypes and also aided in identification of genetic material.

During the 2005 growing season, various plant measurements were taken
to elucidate underlying factors involved in differences between the parental
genotypes in tolerance to low soil P. Adventitious root number was counted on
twoplants for each P treatment and in each replication at 21 days after planting.
This was done by excavating entire root systems from the ground. At mid-pod fill
(which ranged from 54 to 67 days after planting) shoots from 50 cm row length
(5-10 plants) were harvested and oven dried at 70°C until free of moisture (about
5 days). Dry weights were recorded and shoots were ground to a fine powder in a
Wiley mill with a 60-mesh screen. Plant tissue samples were Kj eldahl acid
digested and subsequently measured for concentration of P according to the
method of Murphy and Riley (1962) using colorimetric spectrometry.

At mid pod fill, a single root sample was taken per plot using a soil anger
with a diameter of 7.1 cm and to a depth of 16 cm. Samples were taken directly
adjacent to the plant stem. Roots were separated fiom soil and washed with water.
Soil was dried and weighed to determine volume. Cleaned roots were placed in a

transparent plastic tray (22 x 28 x 6.5 cm) and covered with water. The tray

78

containing roots was then scanned with a flat bed scanner with a top and bottom
light source at 300 dpi to produce a 2 dimensional scanned image. The images
were analyzed with WinRHIZO software (Regent Instruments Inc., Quebec,
Canada). Root length, root surface area and average root diameter of the samples
was determined. Following image capture, roots were dried at 65°C for five days
and dry weights were determined.

Seeds reached maturity at 76 to 98 days after planting. At maturity, seeds were
hand—harvested. Seeds were dried and seed yield was determined at 18%
moisture. Seed weight (of 100 seed) was measured at 18% moisture for the 2000
and 2005 plantings. A sub sample of seed fi'om the 2000 and 2005 harvest was
analyzed for total phosphorus concentration. Five grams of seed of each
treatment for each of the three replications were cleaned with distilled water and
dried. Samples were then freeze dried to remove all moisture. Freeze-dried seed
samples were placed into aluminum tubes containing three silver balls.
Mechanical agitation was used to crush seeds into a fine powder. Seed samples
were Kjeldahl acid digested and subsequently measured for concentration of P
according to the method of Murphy and Riley (1962) using colorimetric
spectrometry.

Statistical analysis was conducted with SAS for Windows V8 (SAS
Institute, Cary, NC). The command PROC GLM was used to determine treatment,
genotype and interaction effects. Data was analyzed as a split plot with soil P
level as the whole plot and plant genotype as the split plot. The coMand PROC

CORR was used to determine Pearson correlation coefficients among variables.

79

Effects of plant growth habit on measured variables were determined by means
comparisons of determinate vs. indeterminate growth habit types and differences
were established using Tukey’s test for significance. Bartlett test for variance
homogeneity was conducted for growth habit because there were unequal number

of determinate (56) and indeterminate (21) RILS.

Results:

P uptake: Significant differences in P uptake were observed between
environments such that mean uptake of the RILs was more than 3 times higher in
the high P than the low P environment (Table 1 and Table 2). Genotypic
variability was observed for P uptake among the RIL population (Table 1). P
uptake by the inefficient parent, AND 696 was below the RIL mean in the high P
environment, and similar to the RIL mean in the low P environment (Table 2).
Significant genotype x environment (G x E) interactions were observed for this
trait (Table 1).

P use efficiency: In addition to P uptake efficiency, P use efficiency also
exists as a tolerance mechanism to low P soil. In this study P use efficiency was
defined as the amount of seed yield per unit of P uptake by the plant. There was
variability for P use efficiency between environments and among genotypes. G x
E was not significant (Table 1). Efficient use of P was greater in the low P
environment, but genotypic differences in the population were only observed in

the high P environment (Table 2).

80

Seed P content: Another measure of P efficiency is the content of P in the
seed. There were significant environmental and genetic differences for seed P
content observed, and a G x E effect was present in 2000, but not 2005 (Table 1
and Table 2). The seed P content of AND696 was higher than the mean RIL
content in both the high and low P soil treatments (Table 2).

Yield: Significant differences in seed yield were observed between the
high and low soil P environments in the RIL population (Table 1). The mean
yield in low P was 597 kg/ha and 1345 kg/ha in high P in 2005. Similar yield
differences as those seen in 2005 were also recorded between the environments in
2000 (Table 2). The differences in yield in each environment demonstrate the
effects of limiting P on plant growth in this population. Yield of the check variety,
Carioca was higher than that of the RIL population across environments and years
(Table 2). In both years, there was significant genotypic variability for yield and
in 2005 there was a significant G x E interaction (Table 1).

Correlations: P uptake was significantly correlated with seed yield in
.2005 in each P treatment, and the correlation coefficient was higher in the low P
(1:053) than the high P (r=0.48) treatment (Table 3). This is consistent with P
uptake having an important role in yield detennination in a P limiting
environment. P use efficiency was more strongly correlated with seed yield in the
high P than the low P treatment (Table 3). There was a negative correlation
between P uptake and P use efficiency. This was expected because by definition,

P use efficiency is based on getting the highest yields with the least amount of P

81

uptake. Seed P content was positively correlated with seed yield in 2005, but not
in 2000 (Table 9). i

Root traits and P uptake: A selection of root traits was measured in the
2005 growing season to determine their importance in P uptake. Adventitious
root number was counted at 3 weeks after planting. These roots arise fiom shoot
tissue, and have a lower carbon construction cost than roots arising from root
tissue, as they have low density (Lynch and Ho, 2005). They are generally
shallow in orientation and deploying in the topsoil therefore closer to P enriched
sources of the soil, such as organic matter. Environmental differences were
observed for adventitious root number (Table 5). No genotypic differences were
detected for adventitious root number in this population (Tables 4 and 5).

Mean squares fiom analysis of variance identified significant genotypic
differences in root length density, specific root length, root surface area, and
average root diameter (Table 4). Further inspection of these traits indicated that
in the low P soil treatment, only root length density and root surface area were
different among the RILS in this study (Table 5). These traits were not however
correlated with P uptake. Only specific root length was weakly correlated with P
uptake in the low P soil treatment (Table 6).

i In the high P so'il environment, root length density, root surface area, and
average root diameter showed genotypic variation within the RILs (Table 5). In
this environment there were also genotypic differences in P uptake (Table 2).
There was a weak correlation between specific root length and P uptake and a

negative correlation between average root diameter and P uptake (Table 6).

82

_ The weak correlations presented in this study between root architecture
traits and P uptake are not surprising because roots are extremely plastic,
especially in field environments. However, despite the plastic nature of root
growth it was possible to detect genetic" differences in these traits. The
observation that there was genotypic variability for root architecture traits, but
these traits were not correlated with P uptake shows that these traits should not be
considered adaptive for tolerance to low P soils in this population.

Differences by Growth Habit

In this study, the RIL population segregated for indeterminate and
determinate plant grth habit, this trait is controlled by a single gene, the fin
gene (Koinange et al., 1996; Norton et al., 1915). Plant grth habit has been
shown to be linked to maturity in common bean (Coyne and Schuster, 1.97 4;
White et al., 1991) as observed here. Detenninate RILs in the population reached
maturity on average in 80 to 82 days whereas indeterminate RILs required 88 to
89 days to reach maturity depending of soil P level (Table 7).

There were differences in seed yield in the determinate and indeterminate
RILS in both 2000 and 2005. The trend however was not the same across years.
In 2000, the indeterminate RILS had greater seed yield than the determinate RILS
under low P and no differences in yield under the high P treatment (Table 7).
Detenninate plants have separate vegetative and reproductive stages of
development, whereas indeterminate plants continue to grow vegetatively during
flowering and seed development (Huyghe, 1998). In 2000, the indeterminate

plants yielded higher on average than the determinate plants, perhaps because of

83

the ability to grow longer and acquire nutrients whereas the determinate plants
had switched fully to seed production.

' In 2005 there were no differences is seed yield in low P between the
determinate and indeterminate plants. Yield was however greater in the
determinate plants in the high P treatment (Table 7). Comparison of the 2000 and
2005 yield data suggest the presence of an additional stress present in 2005 that
affected the indeterminate RILs to a greater extent than the determinate RILS.
Such a stress may have become a factor later in the life cycle of the plants because
the indeterminate lines took 6-9 days longer to reach maturity.

In 2005, no significant differences in shoot dry weight or P uptake were
observed for determinate as compared to indeterminate RILs (Table 7). There
were differences observed in two root traits, but these differences were only
present in the high P soil treatment. The indeterminate plants had greater root
length density and root surface area than determinate RILS in the population
(Figures 1 and 2). Specific root length was weakly correlated with P uptake in the
determinate RILS in the low P treatment and with the indeterminate RILS in the
high P treatment (Table 8).

The determinate RILs had a higher P use efficiency in high P than the
indeterminate RILS. In 2005 the seed P content was higher in the indeterminate
RILS in both the high and low P treatments. In 2000 differences in seed P content
were only present in the high P treatment (Table 7). Correfations for seed yield

and seed P content were significant and positive in 2005 but not in 2000 (Table 9).

Since there were unequal number of determinate and indeterminate RILS in the

84

population, Bartlett’s test for variance homogeneity was conducted to identify
bias that may have existed in trait analysis by growth habit. Most traits exhibited
variance homogeneity although there were exceptions, including seed P content
(high P), yield (low P, 2000) and root length density, specific root length, and root
surface area (high P) (Appendix, Tables A3-l; A3-2).

Discussion:

Seed yields were greatly reduced in the low P treatment, and genetic
variability was observed for response to low soil phosphorus. The population also
exhibited genetic differences in P uptake and P use efficiency, although only P
uptake efficiency was important in the low P soil treatment. Tolerance to low soil
P exhibited by the efficient parent G19833 has been shown to be based on its
efficient uptake of P from the soil, which can be attributed to its extensive root
system (Liao et al., 2001; Nielson et al., 2001). Root architecture traits that
increase the area of soil exploration may in turn increase the capture of immobile
P ions by a plant root system (Gahoonia and Nielsen, 2003). Each of the root
traits measured in this study are indicators of explorative capacity and we hoped
to be able to use the RILs to identify those root traits important for P uptake.
None of genetic variability observed for the root architecture traits studied here
factored into improved P uptake efficiency.

Previous studies have, however, identified adventitious root number as
important for P uptake (Miller et al., 2003) and studies of RILs developed by a
cross between an Andean and Mesoamerican line have shown increased numbers

of adventitious roots to be important for P uptake (Ochoa et al., 2006). Root

85

length density and root surface area are both measures of root spacing in a volume
of soil. Specific root length and average root diameter indicate root thickness.
Thinner roots have an advantage in P uptake over thicker roots because they
require less construction cost in the form of carbon, to explore the same area of
soil (Lynch and Brown, 2001). Most plant available P is found in the topsoil,
therefore in this study, measurements on root traits were conducted on roots in the
top 16 cm of soil. Perhaps within the Andean gene pool there is less variability
for root traits than in the Mesoamerican gene pool.

9 Genetic variability was uncovered for P use efficiency in the high P
treatment. This trait was associated with the determinate plants (Table 7), and
may be indicative of the different grth strategies of indeterminate and
determinate plants. In rice (Oryza sativa L.), for example, P use efficiency was
negatively correlated with plant dry weight and found to be an adaptation to low
soil P found in plants inefficient in P uptake (Wissuwa et al., 1998).

Selection of genotypes with greater seed P content may improve plant
growth in low P soils. Nutrients stored in the seed are important for germination
and early plant growth (Liptay and Arevalo, 2000). Increased seed P content
partially compensated for the negative effects of low P scils on early seedling
grth in wheat (Triticum aestivum L.), by increases in shoot dry weight, leaf size,
and root length (De Marco, 1990). A positive correlation between seed P and
yield was observed in 2005. The correlation between these traits was stronger for

the indeterminate than the determinate lines and therefore may be a more

86

important factor when specifically selecting indeterminate lines adapted to low
soil P.

Another important factor that may be related to plant growth habit is
adaptation, because the parents exhibited differences for this trait. The unadapted
nature of G19833 can be seen in Figures 3 and 4, where seed yield in high P is
regressed by seed yield in low P in 2000 and 2005. G19833 did not perform well
in the high P environment (Figure 4). AND696 on the other hand is an improved
line and demonstrated an average yield potential in the high P environment,
although it performed poorly in the low P environment (Figure 3).

Based on the yield regressions, it would appear that selecting a
determinate plant type, would be the best way to maximize yield potential across
environments. In 2005, the top yielding lines across environments possessed a
determinate growth habit (Figure 3), whereas in 2000, the indeterminate RILS on
average performed better than the determinate lines, but numerous determinate
lines performed well in both environments (Figure 4). In a previous study of
Andean beans conducted in three environments in Colombia, determinate lines
were found to have greater yield stability (White et al., 1992). Detenninate lines
are not always more stable and other studies in temperate environments have

shown indetenninate lines to be more stable (Kelly et al., 1987).
Conclusions

Genotypic differences were observed for yield in 2000 and 2005. There

were also genotypic differences for P uptake in‘the low P treatment and P uptake

87

and P use efficiency in the high P treatment. Both P uptake and P use efficiency
were correlated with yield. Genetic variability for root length density and root
surface area was observed among the RILS. These differences did not appear to
improve adaptation to low P soils, however, because they were not correlated with
P uptake.

Growth habit affected seed yield differently between years, and was not
related to P uptake in 2005. Root length density and root surface area were
greater in indeterminate plants in the high P treatment only. Determinate lines
were more P use efficient in high P and indeterminate lines had greater seed P
content in high and low P in 2005 and in low P in 2000. The differences in trait
response in 2000 and 2005 suggest that while the indeterminate lines were (on
average) better adapted to the low P environment than the determinate lines, the

determinate lines were better able to perform across the two environments.

88

References:

Beebe, S., J. Lynch, N. Galwey, J. Tohme, and I. Ochoa. 1997. A geographical
approach to identify phosphorus-efficient genotypes among landraces and wild
ancestors of common bean. Euphytica 95:325-336.

sBeebe, S.E., M. Rojas-Pierce, X.L. Yan, M.W. Blair, F. Pedraza, F. Munoz, J.
Tohme, and J .P. Lynch. 2006. Quantitative trait loci for root architecture traits

correlated with phosphorus acquisition in common bean. Crop Science 46:413-
423.

Bonser, A.M., J. Lynch, and S. Snapp. 1996. Effect of phosphorus deficiency on
growth angle of basal roots in Phaseolus vulgaris. New Phytologist 132:281-288.

Christiansen, 1., and PH. Graham. 2002. Variation in di-nitrogen fixation among
Andean bean (Phaseolus vulgaris L.) genotypes grown at low and high levels of
phosphorus supply. Field Crops Research 732133-142.

CIAT. 2000. Inheritance of low phosphorus tolerance in the Andean population
AND696 x G19833. Bean Improvement for the Tropics Unit Annual Report.
CIAT, Cali, Colombia.

Coyne, DP, and ML. Schuster. 1974. Inheritance and Linkage Relations of
Reaction to Xanthomonas-Phaseoli Smith,Ef Dowson (Common Blight), Stage of
Plant Development and Plant Habit in Phaseolus-Vulgaris L. Euphytica 23: 195-
204.

Demarco, D.G. 1990. Effect of Seed Weight, and Seed Phosphorus and Nitrogen
Concentrations on the Early Growth of Wheat Seedlings. Australian Journal of
Experimental Agriculture 30:545-549.

FAO. 2006. FAOSTAT [Online] http://faostat.fao.org (posted November 14, 2006;
verified November 14).

Gahoonia, TS, and NE. Nielsen. 2003. Phosphorus (P) uptake and growth of a
root hairless barley mutant (bald root barley, brb) and wild type in low- and high-
P soils. Plant Cell and Environment 26:1759-1766.

Huyghe, C. 1998. Genetics and genetic modifications of plant architecture in
grain legumes: a’review. Agronomic 18:383—411.

Kelly, J .D., M.W. Adams, and G.V. Vamer. 1987. Yield Stability of Detenninate

and Indeterminate Dry Bean Cultivars. Theoretical and Applied Genetics 74:516-
521.

89

Koinange, E.M.K., S.P. Singh, and P. Gepts. 1996. Genetic control of the
domestication syndrome in common bean. Crop Science 36:1037-1045.

Liao, H., and X.L. Yan. 2001. Genotypic variation in root morphological
characteristics of common bean in relation to phosphorus efficiency. Acta
Botanica Sinica 43:1161-1166.

Liao, H., G. Rubio, X.L. Yan, A.Q. Cao, K.M. Brown, and J .P. Lynch. 2001.
Effect of phosphorus availability on basal root shallowness in common bean.
Plant and Soil 232:69-79.

Liao, H., X.L. Yan, G. Rubio, S.E. Beebe, M.W. Blair, and J .P. Lynch. 2004.
Genetic mapping of basal root gravitropism and phosphorus acquisition efficiency
in common bean. Functional Plant Biology 31:959-970.

Liptay, A., and AB. Arevalo. 2000. Plant mineral accumulation, use and transport
during the life cycle of plants: A review. Canadian Journal of Plant Science
80:29-38.

Lynch, J ., and J .J . Vanbeem. 1993. Grth and Architecture of Seedling Roots of
Common Bean Genotypes. Crop Science 33:1253-1257.

Lynch, J. P. and S. E. Beebe. 1995. Adaptation of Beans (Phaseolus-Vulgaris L) to
Low Phosphorus Availability. Hortscience 30: 1165- 1171.

Lynch, J .P., and KM. Brown. 2001. Topsoil foraging - an architectural adaptation
of plants to low phosphorus availability. Plant and Soil 237:225-237.

Miller, CR, 1. Ochoa, K.L. Nielsen, D. Beck, and J .P. Lynch. 2003. Genetic
variation for adventitious rooting in response to low phosphorus availability:

potential utility for phosphorus acquisition from stratified soils. Functional Plant
Biology 30:973-985.

Murphy, J ., and JP. Riley. 1962. A modified single solution for determination of
phosphate in natural waters. Anal Chem Acta 27:31-36.

Nielsen, K.L., A. Eshel, and J .P. Lynch. 2001. The effect of phosphorus
availability on the carbon economy of contrasting common bean (Phaseolus
vulgaris L.) genotypes. Journal of Experimental Botany 52:329-339.

Norton, J .B. 1915. Inheritance of habit in the common bean. Am. Nat.
49:547-561.

Ochoa, I..,E M. W. Blair, and J. P. Lynch. 2006. QTL analysis of adventitious root

formation In common bean under contrasting phosphorus availability. Crop
Science 46: 1609- 1621.

90

Park, 80., DP. Coyne, J .M. Bokosi, and J .R. Steadman. 1999. Molecular

markers linked to genes for specific rust resistance and indeterminate growth
habit in common bean. Euphytica 105:133-141.

Sanchez, PA, and J .G. Salinas. 1981. Low-input technology for managing
Oxisols and Utisols in tropical America. Advances in Agronomy 342280-406.

Singh, SP. 2001. Broadening the genetic base of common bean cultivars: A
review. Crop Science 41 :1659-1675.

Singh, S.P., H. Teran, C.G. Munoz, J .M. Osomo, J .C. Takegami, and M.D.T.
Thung. 2003. Low soil fertility tolerance in landraces and improved common
bean genotypes. Crop Science 43:110-119.

White, J .W., J. Komegay, J. Castillo, C.H. Molano, C. Cajiao, and G. Tejada.
1992. Effect of Growth Habit on Yield of Large-Seeded Bush Cultivars of
Common Bean. Field Crops Research 29: 15 1-161.

Wissuwa, M., M. Yano, and N. Ae. 1998. Mapping of QTLs for phosphorus-
deficiency tolerance in rice (Oryza sativa L.). Theoretical and Applied Genetics
97:777-783.

Wortmann, C.S., R.A. Kirkby, C.A. Eledu, and DJ. Allen. 1998. Atlas
of common bean (Phaseolus vulgaris L.) production in Africa CIAT, Cali,
Colombia.

Yan, X.L., J .P. Lynch, and SE. Beebe. 1995. Genetic-Variation for Phosphorus
Efficiency of Common Bean in Contrasting Soil Types .1. Vegetative Response.
Crop Science 35:1086-1093.

Yan, X.L., S.E. Beebe, and J .P. Lynch. 1995. Genetic-Variation for Phosphorus
Efficiency of Common Bean in Contrasting Soil Types .2. Yield Response. Crop
Science 35 : 1 094-1099.

Yan, X.L., J .P. Lynch, and SE. Beebe. 1996. Utilization of phosphorus substrates
by contrasting common bean genotypes. Crop Science 36:936-941.

Yan, X.L., H. Liao, S.E. Beebe, M.W. Blair, and J .P. Lynch. 2004. QTL mapping

of root hair and acid exudation traits and their relationship to phosphorus uptake
in common bean. Plant and Soil 265: 1 7-29.

91

Table 1. Analysis of variance of traits related to phosphorus uptake and use, and
seed yield in a population of 75 recombinant inbred lines from the
AND696/G19833 population field grown in Darien, Colombia in 2000 and 2005
under two treatments: high and low soil phosphorus.

 

Mean squares

 

Source df P P use Days to Seed P Seed yield

 

 

 

 

uptake eff. Maturity content
2005
Genotype (G) 74 263‘" 0.04" 113'" 2092'" 161674'"
Env. (E) 1 120770‘" 526"" 217‘" 324541‘" 62929473”
G x E 74 220‘" 0.02 "S 5.4 "S 420 "S 103183”
2000
Genotype (G) 74 —————————— 55'" 1672'" 465740‘"
Env. (E) 1 ---------- 195*" 205516‘" 88298547”
G x E 74 ---------- 3.2 “s 673‘” 211912 “S

 

“8 indicates not significant

* Indicates significance at P < 0.05.
”Indicates significance at P <0.01.
*** Indicates significance at P<0.001

92

Table 2. Mean plant growth traits in high (HP) and low (LP) phosphorus soils
conditions for parents AND696 and G19833 and the means and ranges of 75
recombinant inbred lines (RILS) developed from the parents. Means are also
included for the check variety, Carioca. The experiment was planted in 2000 and
2005 in Darien, Colombia. Mean values are of 3 replications. P value indicates
level of significant genotypic differences among the RILS for each trait.

 

 

 

Parents Recombinant inbred lines Check

Traits P AND G mean range p Carioca

level 696 19833 value (G4017)
P uptake HP 36.5 92 46.7 25-92 0.0010 70.9
(mg ~plant") LP 14.5 ----- 13.7 6.6-26.1 0.0112 8.9
P use HP 0.33 ----- 0.29 0.11-0.50 0.0016 0.29
efficiency LP 0.27 ----- 0.51 0.25-0.78 0.2787 0.79
(g seed mg P")
Days to HP 77.3 ----- 82.9 77-98 <.0001 86.3
maturity LP 77 .3 ----- 84.2 77—94 <.0001 84.7
2005
Days to HP ----- 93 83.4 77-90 <.0001 86.7
maturity LP ----- 93 84.7 80-93 <.0001 83.3
2000
Seed P HP 198 165 193 146-261 <.0001 l 1 1
content LP 147 ----- 139 97-181 <.0001 87
(mg 100 seed")
2005
Seed P HP ----- 161 160 107-256 <.0001 93
content LP ---------- 1 16 76-151 <.0001 68
(mg 100 seed")
2000
Seed yield HP 1347 ----- 1345 582-1962 <.0001 2096
(kg-ha“) 2005 LP 465 ----- 597 323-944 0.0542 830
Seed yield HP ----- 1080 1516 681-2871 <.0001 3038
(kg ha") 2000 LP ----- 649 642 268-1064 <.0001 1344

 

93

Table 3. Phenotypic correlations among P uptake, P use efficiency, and seed yield
in a population of 75 recombinant inbred lines from a AND696/G19833 cross
grown in high (HP) or low (LP) soil phosphorus in Darien, Colombia in 2005.

 

P P use Seed yield
level efficiency

 

P uptake HP -0.51m 0.48m

0*.

LP E059“ 0.53

ti.

P use HP ------ 0.43
efficiency

it!

LP ------ 0.24

 

*Indicates significance at P < 0.05.
“Indicates significance at P <0.01.
*** Indicates significance at P<0.001.

94

Table 4. Analysis of variance of root traits in a population of 75 recombinant
inbred lines from the AND696/G19833 population field grown in Darien,
Colombia in 2005 in two environments: high and low soil phosphorus.

 

 

 

11 Mean Squares

s Source df Adv. Root Specific Root Average
roots length root surface root

i density lergth area diameter

nGenotype (G) 74 5 “S 0.56"" 21‘ 2210'" 0.008‘"

dEnvironment (E) 1 142‘" 9.06‘" 31 "S 691 "S 0.065‘"

11G x E 74 7 “S 0.31 "s 13 "5 1538 "5 0.006 "5

 

“3 indicates not significant

* Indicates significance at P < 0.05.
“Indicates significance at P <0.01.
*** Indicates significance at P<0.001

95

Table 5. Mean root growth traits in high (HP) and low (LP) phosphorus soils
conditions for parents AND696 and G19833 and the means and ranges of 75
recombinant inbred lines (RILS) developed from the parents. Means are also

included for the check variety, Carioca. The traits are from the 2005 field

experiment planted in Darien, Colombia. Mean values are of 3 replications. P ,
value indicates level of significant genotypic differences among the RILS for each

 

 

 

trait.
Parents Recombinant inbred lines Check
Traits P AND G mean range p value Carioca
level 696 19833 _ (G4017)

Adventitious HP 7.2 8.5 9.9 7-12 0.8931 8.2
roots ( #) LP 11.2 ----- 8.8 6-14 0.5291 9.2
Root length HP 1.35 1.1 1.08 0.57-2.30 0.0008 0.70
density LP 1.59 ----- 1.36 0.59—2.59 0.0281 0.66
(cm-cm3-1)
Specific root HP ’ 11.1 10.7 8.4 3.3-17.6 0.5144 5.7
length LP 9.71 ----- 8.8 4.0-18.7 0.3887 4.4
(cmmg'l)
Root surface HP 87.9 57.3 71.7 33-147.4 0.0040 53.4
area (cmz) LP 603 ----- 74.0 32.2-156.5 0.0289 37.7
Average HP 0.24 0.27 0.33 0.25-0.53 0.0128 0.37
root LP 0.26 ----- 0.31 0.24-0.40 0.3464 0.32
diameter
(mm)

 

96

Table 6. Phenotypic correlations between root traits and P uptake and seed yield

in a population of 75 recombinant inbred lines developed from a

AND696/G19833 cross grown in high (HP) or low (LP) soil phosphorus in Darien,
Colombia in 2005.

 

 

P Adv. Root _ Specific Root Average
level root length root surface root
number density length area diameter

P HP -004 -0.06 0.16" —0.16" -018'"
uptake ..

LP 008 0.001 0.16 0.002 -0.01
Seed HP 0.09 -0.098 -0.04 -0.10 -0.05
yield

LP 0099 0.03 0.10 0.02 -0.022

 

*Indicates significance at P < 0.05.
**Indicates significance at P <0.01.
*** Indicates significance at P<0.001.

97

Table 7. Means of root and shoot traits of 75 RILS developed from a

AND696/G19833 cross and grown under high and low soil phosphorus treatments

in Darien, Colombia in 2000 and 2005, grouped and averaged by plant growth

habit of which there were two categories, indeterminate (1nd.) and determinate

 

 

(Det.).

Traits HigLP Low P

Det. Ind. Bet.

2005
Adventitious roots 9.8 a 8.7 a 8.9 a
H")
Root length density 1.01 b 1.42 a 1.34 a
(cm-cm3-1)
Specific root length 8.2 a 8.8 a 8.9 a
(curing!)
Root surface area 63.5 b 76.9 a 71.3 a
(cm’)
Average root 0.32 a 0.31 a 0.30 a
diameter (mm)
Shoot dry weight 13.7 a 6.6 a 6.3 a
(3 'Plant")
P uptake 46.6 a 13.7 a 13.7 a
(mg ~plant")
P use efficiency 0.30 b 0.47 a 0.52 a
(g-seed mg P")
Days to maturity 80.6 b 89.4 a 82.2 b
Seed P content 186 b 147 a 136 b
(mg 100 seed")
Seed yield 1435 b 573 a 605 a
(kg°ha")
2000

Days to maturity 82 b 88.3 a 83.3 b
Seechontent 153b 112a 117a
(mg 100 seed")
Seed yield (kg-ha") 1551 a 707 a 614 b

 

 

 

 

 

‘1 Significant differences are based on Tukey tests and are at alpha = 0.05. Values
that do not share the same letter are significantly different. Tests should be read

across rows and within each phosphorus treatment level, High P or Low P.

98

Table 8. Phenotypic correlations between root traits and P uptake and seed yield
in a population of 75 recombinant inbred lines (RILS) developed from a
AND696/G19833 cross grown under high or low soil phosphorus treatments in
Darien, Colombia in 2005. For the analysis, RILs were grouped according to
plant grth habit of which there were two categories, indeterminate (1nd.) and

determinate (Det.).

 

 

 

 

 

Growth Adv. . Root Specific Root Average
habit root # length root surface root
density length area diameter
Low P
P Det. -0.10 0.02 0.20*** 0.02 -0.04
uptake
Indet. -0.01 -0.03 -0.08 -0.03 0.05
Seed Det. -0.04 0.13 0.17** -0.10 0.10
yield
Indet. -0.27*** -0.17 -0.05 -0.15 ' 0.19
High P
P , Det. -0.09 -0.12 0.10 -0.17** -0.14* '
uptake
Indet. 0.04 0.03 0.24*** -0.13 -0.27***
Seed Det. 0.15* 0.10 0.14* 0.05 -0.04
yield
Indet. 0.04 -0. 18 0.02 -0. 13 —0.005

 

* Indicates significance at P < 0.05.
**Indicates significance. at P <0.01.
*** Indicates significance at P<0.001

99

Table 9. Phenotypic correlations between P uptake and seed yield and P seed
content and seed yield in a population of 75 recombinant inbred lines (RILS) from
a AND696/G19833 cross grown under high or low soil phosphorus treatments in
Darien, Colombia in 2000 and 2005. For the analysis, RILS were grouped
according to plant grth habit of which there were two categories, indeterminate
(1nd.) and determinate (Det.). '

 

High P-Ind. High P-Det. Low P-Ind. Low P-Det

 

 

 

 

 

 

Seed P uptake (2005)
yield 0.03 0.18** 0.46*** 0.42***
P seed content
2005
0.17 0.30*** 0.42*** 0.30***
2000
-0.06 0.12 0.08 0.08

 

* Indicates significance at P < 0.05.
**Indicates significance at P <0.01.
*** Indicates significance at P<0.001

100

.mnoumomnon m mo. $226 2: 2a 832 .nnonnoob £80.26? «and 03222235 .

fins magnet mo 83? neonn 58232 292.34 ...Een 8.30% 3 consanom 8? 2m mtmm nnfiw some nag? .Ewn 05 no
magnum 43 mm una a2 05 no 23an .3 w2n22n2 AA 6.9» 26— m now 43 Bamanom 2a mnounnwnmwu mononvonm mm noun:
nouns—Ema Beam Emu ofi no 392» no “on“ $2 2: no £3me 2: no 850% mm .3 E mam mo nonnfihmmo 2F .mom agonnmonn
SE 32 one EB 83 a 88 a 2828 gonna a eoeoaa 225 eds 282... oooEop aeo a see eooo_o>oe

Gina men: 3.55 “935882 mu 3 awn»: noon 058% end bane“. fimnfl noon mo nounflbmmo zononwonm ._ 03mm

Gian—953 bane—o finne— «com

md 0N m... . 0... m6 mN . 0N m.—. oé m6

 

 

 

 

 

 

 

 

 

 

 

 

 

F l r I c Q 4 Es - fim‘ S _ o
l N . :fiN‘ l N
r. V I V
T 0 l m
a . 1
l w fi l w
t S . . . t
or
L L ,
. 1 NF . 1 NF
ofimfidhuoonwfi g. 22:55qu § .
. I t E EmEEQouc. D t 3

 

 

d... n:
d 1.2. [or

Aouenbeld

101

or mw vw

 

6358.235 we.
26.55.23 I . ~

 

n_I

 

Aswan—av finned “com “.59on

to?

 

lwv

or mr vr NF Ow

 

 

 

SmEEnouonc. § .
, 22.5.98 U

a.—

 

 

 

 

o v

. . Hm . 0
IN
..,.
to
In
..2
..s
lvw
..e
..e .

 

.283 n 8&2

Kouenber :1

102

0850332 m no 0m20>0 06 0.8 0502. .nnonnmob £000 38?», m5 Begun—003
one 2022200 no 003? n88 E2052 032.24 £3. £3on .3 883mg 003 20 mam nnfiw n80 n55? .332 05 no

232» 43 mm one #2 05 no £32» 3 00800052 .5 an? ~22 m mom \3 @2232 20 0555508 xenon—00¢ .mm Bonn,

nounnwnmfi Bonm Emu 05 no 232w no fins c2 05 no 322w 05 no gonm mm .5 E mam mo noun—95.0% 2E. .mom 0223023
8.: 32 one And an a moon 5. 2628 .eofin a 383 82 no one 232... copies none a Bee eoooeeoe
fig men: 0205 “923882 mu 3 220820 noon 0w20>0 one 02a 00250 noon mo nounnwnmfi mononwnnm .N 0.5me

$53 «0.3 000......» 30% .

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8F 9: cm. 8. 8 8 8 9:. e: on. . 8F 8 8 2.

ac . - . _ .F s O

t N m , . WI t N

.I V . . .I V

t 0 . _ L _ ~ .. 0

t w T 0
fi T2 . i 2
1 NF _ , i a.

. V IL
2055.20.05 I .. 3 2058.200... V///////. l t 2
2055.98 I . . . oeosésoo D g

. t 9 . i S

 

d: _ fl , . . n...

Aouenbelg

103

 

90.55.3005 ems...
2055.900 I

a:

 

Annnv 22255. «00.. 03203»

2.0 .

13.
law

INN

 

IVN

9058.200... §

32.5.0.8 D
d._

s
m
\
w
a
m
\
m
\
m

 

 

__..._..l

 

 

 

 

 

.8388 N 035

Kouenbeld

104

.38 S 0253 “on 003 anon—an 0:: no 0000
003000 moon non ooom 80.0 8530 20 0020., 200» mmmeU .0002 .m nmE 2005 205 508 00n002n2 .80 0820., 05 m .52

.00n: 22% 508 00n002n2 .03 nanownom .mmwanw 05 000% 303000 0020 0 22m 0320.60 00n: 00.53 «539802
E. S non—«Egon 05 mo no0000 mnrsonm moon 20 E m :30 2005. 205 0000 .3 m 32 008» 05005 0000 we nom002w0m .m 0.32m

.72 9: o :02 5? 203 590

ooou. coup . com? 003 cam? Door com com
. h p e .— . —

 

 

 

 

 

— - b b p b. r - F CON
1 com 9
. m.
I 00? U
4. M. .
. . ..W
r Com 0-
e w.
. m..-
r 08 IOI
. M
1 02. d
I )
I I I 4 . . mun
I 1 000 U.
I I .
I w B.
. '7
0005500005 4 I 5

 

0005:0200 I t 82

‘ 105

. .m :30 0005 200A 508 03000500 03 0000.02, 0:0 m 32
000:: 20% 0008 03000090 an. anagram .900 a 0:0 000% 803000 0080 0 800 00020060 00:: 005E 0903:8000."
mu E ”800153 06 .00 80000 wEBSw CO8 08 E m :mE 000:: 2000 0000 3 m 32 00:: 200A 0000 mo 03000Hw0m .0 0.5mm

A 72 9; a :02 £3, 0.0.; 5.20

 

 

 

_ 800 800 88 89 82 80
P _ p p F L 5’ , . . n b
80
4 I .. 80
f. ‘ I f mu
. 08 m.
. U
u 80 .M.
w.
r 80 D.
. .M:
w con 1U...
n W
I . 80 M
.. 80 d
1 82 .W
. u.
u 8: e
j V
. 80F (
0005500005 0. . .
0005:0900 I .180?
u 83

 

 

106

Chapter 4: Identification of QTL related to Root Architecture Traits and
Low Phosphorus Tolerance in an Andean Bean Population
Abstract

Low soil phosphorus is a major constraint to common bean production.
The identification of QTL (quantitative trait loci) for phenotypic traits associated
with tolerance to low phosphorus soils will aid in the development of cultivars
that can perform well in such environments. Root architecture was identified as a
determinate of P uptake in a cross between an Andean tolerant and Mesoamerican
susceptible bean genotype. This experiment was conducted to determine if root
architecture aids in tolerance to low P soils in Andean intra gene pool cross
between a tolerant (G19833) and susceptible (AND696) genotype. A linkage map
was developed from 77 F57 recombinant inbred lines of Gl9833/AND696 using
SSR, AF LP and RADP markers with 12 linkage groups for a total of 633cM.
QTL were identified for root length density (cm/cm3) and root surface area in the
top 16 cm of soil. The QTL were derived fi'om G19833, but did not co-localize
with QTL for P uptake. A QTL for P use efficiency, derived from G19833 was
identified on linkage group B11. QTL were identified for seed yield on linkage
group B1 near the fin gene for determinacy, while G19833 was the source of the
increased effect in the low P environment in 2000, AND696 was responsible for

the yield increase in this location in the high P environment in 2005.

107

Introduction

Quantitative trait loci (QTL) analysis is a powerful tool for understanding
genetic variation and control of complex traits. QTL are regions of the genome
statistically associated with phenotypic variation of a quantitative trait (Doerge,
2002). QTL identification requires phenotypic data for a trait of interest in a
segregating population and a genetic linkage map (Collard et al., 2005). .

. Mapping of QTL permits the identification QTL controlling multiple
traits in numerous genetic backgrounds. This technique also can be used to
dissect epistatic interactions and predict genotypic by environment interactions at
the genetic level (Dekkers and Hospital, 2002). The identification of QTL serves
as a starting point for marker assisted selection (MAS), which offers potential to
improve traits with low heritabilities or difficult to measure traits without a heavy
reliance on phenotypic selection (Collard et al., 2005). QTL mapping has also
been successfully employed as a first step to identify genes underlying phenotypic
variation for a trait of interest via positional cloning (Li et al., 2006).

In common bean (Phaseolus vulgaris L.) QTL studies have been
conducted for the major diseases, insects, and abiotic stress that limit productivity.
To date, QTL analysis for disease resistance has been the most fruitful area of
research, and MAS based on QTL studies for bean golden mosaic virus and
common bacterial blight resistance are currently employed in breeding programs
(Miklas et al., 2006). .

QTLanalysis for abiotic stresses affecting common bean has yet to reach

the level of success of that seen with biotic stress, nonetheless, this tool is

108

currently being used to identify underlying mechanisms of tolerance to low P soils
(Yan et al., 2004; Beebe et al., 2006).

Phosphorus deficiency is one of the most prevalent stresses, especially in
bean growing regions of the tropics. An estimated 60% of bean production in
Latin America and Africa occurs under deficient soil P conditions (Wortman et al.,
1 998).

Tolerance to P deficiency has been identified in bean lines from diverse
genetic backgrounds in both the Andean and Mesoamerican gene pools (Beebe et
al., 1997). A number of Peruvian landraces, including G19839 and G19833, have
been found to grow especially well in low P soils (Y an et al., 1995a; 1995b).
Strong genotype x environment interactions for seed yield, however, necessitate
the use of other indicators of tolerance to low P soils that exhibit higher
heritabilities (Beebe et al., 1997). A number of physiological traits correlated
with tolerance to low P soils have been identified. In G19833, research has
shown P efficiency to be related to greater P uptake from the soil. RoOt system
phenotypes that exhibit shallower basal root angle, greater total root length and
root surface area, and root length of basal roots in top 3 cm have been shown to
enhance P uptake in P limiting environments (Bonser et al., 1996; Liao etal.,
2004; Beebe etal., 2006).

The measurement of root traits in the field is laborious and root growth is
very plastic even with small changes in the soil environment (Snapp et al., 1995),
making selection for plants with desired root traits challenging. The

identification of QTL for root traits that improve P uptake in low P soils is a first

109

step to conducting MAS for P efficiency. Studies (Beebe et al., 2006 and Liao et
al., 2004) using G19833 as the efficient parent in crosses with DOR364, a
Mesoamerican small seeded black bean as the inefficient parent, have identified
QTL for root traits that co-localize with QTL for P uptake efficiency. QTL for
both root length and P uptake in low P soil were found in the same region of
linkage group B4 with r'2 values of 0.21 and 0.13, respectively. Additional QTL
for specific root length and P uptake under the same environmental conditions co-
localized to a region of linkage group B10 and had 12 values of 0.19 and 0.14,
respectively (Beebe et al., 2006). Liao et al. (2004) also identified QTL on
linkage group B4 in the G19833/DOR364 population for percent of basal roots in
the top 3 cm in a growth pouch assay that co-localized with QTL for P uptake in
the field.

The identification of QTL for root architecture traits with QTL for P
uptake efficiency in crosses between the Andean and Mesoamerica gene pools
raises the question if similar mechanisms for P efficiency would be observed in an
Andean x Andean cross using the same P efficient parent, G19833. The first
objective of this study was to construct a linkage map of two Andean common
bean lines with different tolerance to low P soils. The second objective was to
identify QTL related to tolerance to low P soils using an Andean common bean

recombinant inbred line population.

110

Materials and Methods:

Plant Material

The common bean genotypes G19833 and AND696 from the Andean gene
pool were used as parents to develop 77 F57 recombinant inbred lines (RILS).
The initial cross was advanced to the F5 generation by single seed descent.
G19833 is a Peruvian landrace with an indeterminate grth habit and large
yellow and red mottled seed with an average 100 seed weight of 41 g. G19833 has
also been identified as tolerant to low P soils (Yan et al., 1995ab). AND696 is a
CIAT improved line from the race Nueva Granada with a determinate growth
habit and large red and white mottled seed with an average 100 seed weight of
51 g. AND696 has been identified as susceptible to low phosphorus soils (CLAT,
2000). The RILs developed fiom G19833 and AND696 segregated for
determinacy, with 56 determinate and 21 indeterminate lines. Growth habit in
common bean is controlled by a single gene, fin (Norton, 1915). The number of
determinate and indeterminate lines deviated significantly from a 1:1 ratio that is
expected for a single gene trait. This deviation is likely due to the selection of
RILS with uniform days to maturity, which biased the selection to favor
determinate lines.

Field Trials

In 2000 and 2005, the 77 RILS, G19833, AND696, and two check
varieties, G4017 (Carioca) and G16140 were planted in Darien, Colombia in a
9x9 lattice design with three replications at two soil P levels, low P (45 kg/ha

triple super phosphate) and high P (300 kg/ha triple superphosphate). Phenotypic

111

data was collected on 75 of the 77 RILS. The soil of this site in an Andisol with a
native soil P of 2 mg/kg based on bray II extraction method. Seed was hand-
planted in 4 row plots where each row was 3 m long. During the 2005
growing season, various plant measurements were taken to elucidate underlying
factors involved in the parental genotypes differences in tolerance to low soil P.
Plant measurements include adventitious root number at 3 weeks after planting,
shoot dry weight and total P concentration at mid pod (which ranged fi'om 54 to
67 days after planting), root length and weight to a depth of 16 cm below ground
at mid pod fill (Chapter 3, Materials and Methods).

Seeds reached maturity at 76 to 98 days after planting, as determined by
seed dryness. At maturity, seeds were hand-harvested. Seeds were dried and seed
yield was determined at 18% moisture. Seed weight of 100 seed was measured at
18% moisture for the 2000 and 2005 plantings. A sub sample of seed from the
2000 and 2005 harvest was acid digested and analyzed for total P concentration
according to the method of Murphy and Riley via colorimetric spectrometry
(1962)

DNA Isolation and Molecular Marker Analysis

Plant tissue was harvested from 5 plants of greenhouse grown seed of
G19833, AND696 and 77 F57 recombinant inbred lines. Total DNA was extracted
from plant tissue with 24:1 Chloroform: Isoarnyl alcohol. The rniniprep
procedure was carried out according to the method of Edwards et al. (1991).

DNA was quantified with a fluorometer (Hoefer DyNA Quant 200, San Francisco,

CA). Molecular markers screened in the study for polymorphisms between the

112

parents (319833 and AND696 included 125 simple sequence repeats (SSR)
(Metais et al., 2002; Blair et al., 2003; Gaitan-Solis et al., 2002; Yu et al., 2000;
Caixeta et al., 2005), 50 Random amplified polymorphic DNA (RAPD)
(Integrated DNA Technologies, Inc., Coralville, IA) and 41 amplified fragment
length polymorphisms (AF LP) primer combinations.

SSR Markers

Amplification was performed with 2 pl of DNA diluted to 20 ng°pl", 0.2
pl of primer, 0.15pl of Taq polymerase, 1.2pl (2.5mM) MgClz , 1.2pl (10x)
PCR buffer, and 0.12 pl of a 10mM mix of dNTPs. PCR was conducted in a 96
well FTC-100 Programmable Thermal Controller (MJ Research, Inc., Waltham,
MA). The therrnocycler was programmed for 1 cycle of 5 minutes at 94°C,
followed by 30 cycle of 1 minute at 94°C, 1 minute at 47°C, and 1 minute at 72°C,
and a final extension step at 72°C for 5 minutes. Double stranded DNA was
denatured for 5 minutes at 94°C. PCR amplification products were separated on
4% polyacryilimide gels and DNA bands were visualized with silver nitrate
according to the procedure of Blair et al. (2003).

RAPD markers

Amplification was performed with 5 pl of DNA diluted to 10 ng°pl'l, 1 pl
of primer (Integrated DNA Technologies,-Inc., Coralville, LA), 0.3pl pf Taq
polymerase, 25 pl (2.5mM)MgC12 , 25pl (10x) PCR buffer, and 0.5pl of a
10mM mix of dNTPs. PCR was conducted in a 96 well FTC-100 Programmable
Thermal Controller (MJ Research, Inc., Waltham, MA). The thermocycler was

programmed for 2 cycles of 1 minute at 91°C, 15 seconds at 42°C, and 1 minute

113

and 10 seconds at 72°C, followed by 38 cycles of 15 seconds at 91°C, 15 seconds
at 42°C, and 1 minute and 10 seconds at 72°C, and a final extension step at 72°C
for 5 minutes. A total of 20 pl of PCR product were loaded ontol .5% agarose
(diluted with 0.5x TBE buffer) gels with ethidium bromide and ran in
electrophoresis chambers with 1.5L 0.5x TBE buffer for 3 hours at 110 volts.
PCR amplification products were visualized under UV light. RAPD markers
were named according to the primer used and the molecular weight of the band in
kilobases.

AFLP Markers

AF LP reactions were performed according to the procedure described by
Vos et al. (1995), using a commercially available kit (AFLP analysis System I,
Invitrogen Corporation, Carlsbad, California) and following the manufacturer’s
instructions. The digestion of 500 ng of DNA was performed with 2 pl
EcoRI/MseI (1.25U/pl), incubated for two hours at 37°C. The restriction
fragments were ligated to enzyme adapters using 1pl of T4 DNA ligase (1 U/ pl)
incubated for two hours at 20°C. The prearnplification (primer +1 base) was
performed with 2.5 p1 of DNA diluted 1:10 from the digestion-ligation product,
20 pl preamplification primer mix, 2.5 10x PCR plus Mg and 0.15 pl of DNA
polymerase was added. PCR conditions were 20 cycles of 94°C for 30 seconds,
56°C for 60 seconds, and 72°C for 60 seconds. The PCR product was run in 1%
agarose gel and diluted 1:50. The amplification (primer +3 bases) was performed
with 5 pl of DNA's dilution 1:50 and 5 pl of Mix I (primers and dNTP's) and 10 p1

Mix II (10x buffer, MgCl and DNA polymerase). The PCR conditions were one

114

cycle at 94°C for 30 seconds, 65°C for 60 seconds, and 72°C for 60 seconds, then
the annealing temperature was lowered 0.7°C per cycle for 13 cycles and 23
cycles at 94°C for 30 seconds, 56°C for 60 seconds, and 72°C for 60 seconds.
PCR amplification products were separated on 4% polyacryilimide gels and DNA
bands were visualized with silver nitrate according to the procedure of Blair et al.
(2003). The AF LPs were named by combining the last 2 selective bases ligated to
EcoRl with the last two selective bases ligated to Msel in the amplification step.

Genotypic Data Analysis

Linkage analysis was conducted with the software J oinMap 3.0 for
Windows (Van Ooijen and Voonips, 2002) set to Kosambi’s map fiinction.
Kosambi’s mapping function assumes the existence of interference that is
negatively related to recombination frequency. Analysis was conducted with
marker data of 160 molecular markers segregating in the population of 77
recombinant inbred lines. Mapping parameters were set to recombination
frequency smaller than 0.300 and a LOD score larger than 3.0. Parameters were
relaxed to a recombination frequency of smaller than 0.45 and a LOD score
greater than 2.0 to join unlinked markers and to join fragmented linkage groups.
LOD (log of odds) scores are = log (Ll/Lo), where L1 is the likelihood for the
alternative hypothesis and L0 is the likelihood of the null hypothesis. A LOD
score of 3 means the alternative hypothesis is 1000 times more likely than the null
hypothesis. Linkage groups were identified according to location of SSR markers
(Blair et al., 2003) and were named according to the core bean linkage map

(Freyre et al., 1998).

115

QTL analysis was conducted with the genetic map developed with the
J oinMap program and with the phenotypic means for each RIL collected from the
field study. The computer software program Windows QT L cartographer version
2.5 (Wang et al., 2006) was used to identify QTL for root length density, root
surface area, shoot P concentration, P uptake, seed yield and seed P content. The
Composite Interval Mapping (CIM) feature set to a window size of 100M and
with forward and backward regression model was used to identify QTL. CIM
involves the use of maximum likelihood estimates and linear regression to
identify QTL within marker intervals. Significant QTL were considered by
defining the LOD score at p=0.01 afier 1000 permutation tests.
Results and Discussion:

Linkage Map

DNA polymorphism levels with the molecular markers used in this study
were low to moderate between the parent lines 619833 and AND696. Of the 125
SSR markers screened, all of which were developed for use in common bean,
only 26.4% were polymorphic between the parent lines. Low polymorphism
levels are often seen in intra gene pool crosses such as this one between two
Andean genotypes. A recent study of SSR polymorphism levels compared 129
markers in 44 common bean genotypes fi'om both Andean and Mesoamerican
gene pools, and in the study, the average inter gene pool polymorphism level
(Andean x Mesoamerican) was 59.6% and intra gene pool polymorphisms were

37.9% (Blair et al., 2006).

116

The 103 locus linkage map developed from the SSR, RAPD, and AF LP
markers screened in the G19833/AND696 RIL population spaned 633 cM.
Common bean has 11 linkage groups corresponding to the genome’s 11 '
chromosomes. In this study, 12 linkage groups were identified, 10 of which were
able to be identified and named according to the bean consensus map, based on
the placement of SSR markers (Freyre et al., 1998; Blair et al., 2003) (Figure 1).
Linkage group BS is missing from the map, although the identities of Groups A
and D, (which are comprised solely of AF LP markers), have yet to be determined
in relation to the consensus map.

QTL identification

Using composite interval mapping, 21 QTL were identified for 8 traits in
12 marker intervals on 7 linkage groups. One-third of the QT L were clustered on
linkage grOups B1 and affected more than one trait. An additional 19% of the
QTL were located on linkage group B6 and also affected multiple traits.
Individual QTL explained 12 to 45% of the phenotypic variation, and total
phenotypic variation explained for any one trait was 14 to 60% (Table 1).

Root Architecture

Frequency distribution graphs for four root traits measured at mid pod fill
are shown in Figure 2. Means of RILS and parental genotypes for each trait are
available in Chapter 3, Table 4. The mean root length density of the RILS was
greater in the low P environment (1.36 cm/cm3) than in the high P environment
(1.08 cm/cm3). QTL identified for root length density in the high P environment

were distinct from the QTL identified for the same trait in the low P environment

117

(Table 1, Figure 6). There were no significant differences for root surface area
based on P treatment. However, QTL for this trait were only detected for the high
P environment. The QTL found for root surface area in high P mapped to the
same linkage groups as the QTL for root length density in high P. The increased
effect of each of the root traits was derived fiom G19833 (Table 1). A QTL for
specific root length in greenhouse grown RILS from the population
G19833/DOR364 was identified in the same region of linkage group B1 (Beebe et
al., 2006) as QTL for root length density and root surface area under high P
described here. Additional QTL for root traits in the G19833/DOR364 population
were identified in the same region of B3 as the QTL for root length density in low
soil P identified in this study. They include a QTL for taproot length in
greenhouse grown RILS (Beebe et al., 2006) and a QTL for taproot root hair _
. length in solution culture grown plants (Yan et al., 2004). None of these QTL,
however, overlap with QTL for P uptake.

P uptake

Shoot P concentratiOn and P uptake distribution among the RILs was
affected by the P level of the soil (Figure 3; Chapter 3, Table 2). The range of
values for P uptake was greater in the high P than the low P soil (Figure 3),
suggesting luxury consumption of P occurred in some RILS in the high P
treatment. QTL were detected on B8 for shoot P concentration and P uptake in
the low P environment (Table 1; Figure 6). A QTL for P uptake in the high P
environment was found on Bl 1. The increased effect of each of these traits was

conferred by AND696. No QTL for P uptake were identified with the 2005 field

118

data where G19833 increased P uptake. This result is contrary to results found
with the G19833/DOR364 population where QTL for p uptake from G19833 were
identified (Beebe et al., 2006; Yan et al., 2004). This result, however, may be
specific to the 2005 field season, as suggested by the 2000 field season data
where a QTL was identified for yield in the low P environment derived from
G19833 (Table 1).

Seed Yield

Frequency distribution graphs for seed yield in 2005 and 2000 show the
large variability for yield across RILS and soil P levels (Figure 4, bottom; Chapter
3, Table 2). In the low P treatment in 2005, AND696 yielded less than the mean
of the RILs (465 and 597 kg-ha'l, respectively). In the high P treatment in 2005,
AND696 had the same yield as the mean of the RILs (1347 kg-ha‘l). One QTL
was detected for seed yield in 2005 which explained 27% of the phenotypic
variation. It was observed only in the high P treatment, and AND696 conferred
the increased effect on yield (Table 1). In 2000, one QTL was found for yield in
low P and one was found for yield in high P, explaining 17 and 45% of the
phenotypic variation, respectively. Each QTL was derived from G19833 (Table 1).
The QTL identified for yield under the high P treatment in 2005 and that for yield
under the low P treatment in 2000 both map to the same region of B1 and the
increased effects on yield for these QTL were derived from different parents .
(Table 1; Figure 6). The region of Bl where these QTL are located is the same
region where the single gene (fin) responsible for deterrninacy is located

(Koinange et al., 1996), as the AND696/G19833 population segregated for

119

determinacy. The efficient parent, G19833 exhibits indeterminate plant growth
and the inefficient parent, AND696, exhibits determinate plant growth. These
differences were associated with differences in plant maturity in the population,
such that the indeterminate plants averaged 5 days longer to reach maturity
(Chapter 3, Table 7). Linkage group B1 has previously been shown to be
important in common bean domestication, carrying both the fin gene and the de
gene for photoperiod sensitivity. QTL for domestication syndrome, including
earliness and seed size traits have been identified on this linkage group, near these
genes (Koinange et al., 1996). The importance of B1 near the fin gene suggests
either a pleiotrophic effect of this gene on yield, or linkage of this gene with other
genes important for yield. The observation that this region of the bean genome
had an opposite effect on seed yield under high and low soil P levels may be an
indication that growth habit itself is a key mechanism to improve tolerance to low
P soils, perhaps by increasing days to maturity and allowing more time for P
uptake from the soil. The cluster of domestication genes on B1 have pleiotrophic
effects on other traits in common bean, including leafhopper resistance (Murray et
al., 2004).

P use efficiency

P use efficiency (PUE) was defined in this study as the amount of seed
yield per unit of P taken up by the plant. In 2005, PUE was found to be higher in
the low P environment than in the high P environment (Figure 5; Chapter 3, Table
2). AND696 was at the low end of the distribution for this trait under low P, with

an efficiency of 0.27 compared to the mean of the RILs at 0.51 (Figure 5). A

120

QTL for PUB was detected under high P on El] that was positively contributed
by G19833 (Table l). PUE was not an important mechanism for tolerance to low
soil P in the G19833/DOR364 population (Beebe et al., 2006). PUB may be more
important for tolerance to low P soils in intra gene pool crosses like
G19833/AND696 compared to inter gene pool crosses. Preliminary QTL studies
with the P efficient bean line G21212 identified PUE as a factor in tolerance to
low P soils (Miklas et al., 2006).

Seed P content

P content of the seed may serve as an indicator of tolerance to low P soils.
In P limiting environments, greater levels of P in the seed may aid early plant
growth (Lynch and Beebe, 1995). In the low P soil treatment, seeds contained
less P than in the high P environments in both 2000 and 2005 (Figure 4, top;
Chapter 3, Table 2). P content was correlated with seed yield (1:030 p value
<0.0001) in 2005 in the low P environment, but was not correlated with yield in
any other environment/year combination. QTL identified for this trait were
identified on B1, B2 and B6 across environments and years, and in all cases an
increase in seed P content was derived from G19833 (Table 1, Figure 6).
Although the QTL for seed P content were consistent across years and
environments, based on phenotypic correlation with seed yield it does not appear
to be a valuable trait for selection to improve tolerance to low P soils.
Conclusions:

QTL were identified for root growth traits including root length density

and root surface area. These QTL were derived from the P efficient parent,

121

G19833, and co-localized with QTL for root traits identified in previous studies.
The QTL for root traits identified here did not, however, co-locate with QTL for P
uptake, and no QTL for P uptake detected in the current study were derived from
G19833. -

QTL for yield in low P 2000 and high P 2005 mapped to the same region
of the genome (B1), but were derived from different parents. Plant grth habit
appeared to play an important role in determining yield in the low P environment
in this population. QTL for root length density, root surface area, and seed yield
were also found in the same area of B1. The clustering of QTL in this region
suggests pleitropic effects of the fin gene on a diversity of plant traits, and raises
questions to the value of Selecting QTL in this region of the genome, to improve
tolerance to low P soils. This assertion is supported by regression of yield data by
soil P treatment, that identify determinate lines that yield better than indeterminate
lines under both soil P levels (Chapter 3, Figures 3 and 4).

Genetic differences in P use efficiency were identified in this p0pulation.

QTL across years and environments were identified for seed P content.

122

References:

Beebe, S., J. Lynch, N. Galwey, J. Tohme, and I. Ochoa. 1997. A geographical
approach to identify phosphorus-efficient genotypes among landraces and wild
ancestors of common bean. Euphytica 95:325-336.

Beebe, S.E., M. Rojas-Pierce, X.L. Yan, M.W. Blair, F. Pedraza, F. Munoz, J.
Tohme, and J .P. Lynch. 2006. Quantitative trait loci for root architecture traits

correlated with phosphorus acquisition in common bean. Crop Science 46:413-
423.

Blair, M.W., M.C. Giraldo, H.F. Buendia, E. Tovar, M.C. Duque, and SE. Beebe.
2006. Microsatellite marker diversity in common bean (Phaseolus vulgaris L.).
Theoretical and Applied Genetics 113:100-109.

Blair, M.W., F. Pedraza, H.F. Buendia, E. Gaitan-Solis, S.E. Beebe, P. Gepts, and
. J. Tohme. 2003. Development of a genome-wide anchored microsatellite map for

common bean (Phaseolus vulgaris L.). Theoretical and Applied Genetics
107:1362-1374.

Bonser, A.M., J. Lynch, and S. Snapp. 1996. Effect of phosphorus deficiency’on
grth angle of basal roots in Phaseolus vulgaris. New Phytologist 132:281-288.

Caixeta, ET, A. Borem, and J .D. Kelly. 2005. Development of microsatellite
markers based on BAC common bean clones. Crop Breeding and Applied
Biotechnology 5:125-133.

CIAT. 2000. Inheritance of low phosphorus tolerance in the Andean population
AND696 x G19833. Bean Improvement for the Tropics Unit Annual Report.
CIAT, Cali, Colombia.

Collard, B.C.Y., M.Z.Z. Jahufer, J .B. Brouwer, and B.C.K. Pang. 2005. An
introduction to markers, quantitative trait loci (QTL) mapping and marker-

assisted selection for crop improvement: The basic concepts. Euphytica 142: 169-
196.

Dekkers, J .C.M., and F. Hospital. 2002. The use of molecular genetics in the
improvement of agricultural populations. Nature Reviews Genetics 3:22-32.

Doerge, R.W. 2002. Mapping and analysis of quantitative trait loci in
experimental populations. Nature Reviews Genetics 3:43-52.

Edwards, K., C. J ohnstone, and C. Thompson. 1991. A Simple and Rapid Method

for the Preparation of Plant Genomic DNA for Pcr Analysis. Nucleic Acids
' Research 19:1349-1349.

123

Freyre, R., P.W. Skroch, V. Geffroy, A.F. Adam-Blondon, A. Shinnoharnadali,
W.C. Johnson, V. Llaca, R.O. Nodari, P.A. Pereira, S.M. Tsai, J. Tohme, M. Dron,
J. Nienhuis, C.E. Vallejos, and P. Gepts. 1998. Towards an integrated linkage

map of common bean. 4. Development of a core linkage map and alignment of

RF LP maps. Theoretical and Applied Genetics 97:847-856.

Koinange, E.M.K., S.P. Singh, and P. Gepts. 1996. Genetic control of the
domestication syndrome in common bean. Crop Science 36:1037-1045.

Li, C.B., A.L. Zhou, and T. Sang. 2006. Rice domestication by reducing
shattering. Science 31 1 :1936-1939.

Liao, H., X.L. Yan, G. Rubio, S.E. Beebe, M.W. Blair, and J .P. Lynch. 2004.
Genetic mapping of basal root gravitropism and phosphorus acquisition efficiency
in common bean. Functional Plant Biology 31:959-970.

Lynch, J .P., and SE. Beebe. 1995. Adaptation of Beans (Phaseolus-Vulgaris L) to
Low Phosphorus Availability. Hortscience 30:1165-1171.

Metais, I., B. Harnon, R. J alouzot, and D. Peltier. 2002. Structure and level of
genetic diversity in various bean types evidenced with microsatellite markers

isolated from a genomic enriched library. Theoretical and Applied Genetics
104:1346-1352.

Miklas, P.N., J .D..Kelly, S.E. Beebe, and M.W. Blair. 2006. Common bean
breeding for resistance against biotic and abiotic stresses: From classical to MAS
breeding. Euphytica 147:105-131.

Murphy, J ., and J .P. Riley. 1962. A modified single solution for determination of
phosphate in natural waters. Anal Chem Acta 27:31-36.

Murray, J .D., T.E. Michaels, C. Cardona, A.W. Schaafsma, K.P. Pauls, 2004.
Quantitative trait loci for leafhopper (Empoascafabae and Empoascakraemeri)

resistance and seed weight in the common bean. Plant Breeding 123:474—479.

Norton, J .B. 1915. Inheritance of habit in the common bean. Am. Nat.
49:547-561.

Plant Research International BV. 2002. JoinMap. Release 3.0. Plant Research
International BV.

Snapp, S., R. Koide, and J. Lynch. 1995. Exploitation of localized phosphorus-
patches by common bean roots. Plant and Soil 177:211-218.

124

Statistical Genetics, North Carolina State University, USA. 2006. Windows QTL
Cartographer. Release 2. 5. Statistical Genetics, North Carolina State University,
USA.

Vos, R, R. Hogers, M. Bleeker, M. Reijans, T. Vandelee, M. Homes, A. Frijters,
J. Pot, J. Peleman, M. Kuiper, and M. Zabeau. 1995. Aflp - a New Technique for
DNA-Fingerprinting. Nucleic Acids Research 23:4407-4414.

Wortmann, C.S., R.A. Kirkby, C.A. Eledu, and DJ. Allen. 1998. Atlas
of common bean (Phaseolus vulgaris L.) production in Africa CIAT, Cali,
Colombia.

Yan, X.L., J .P. Lynch, and SE. Beebe. 1995. Genetic-Variation for Phosphorus
Efficiency of Common Bean in Contrasting Soil Types .1. Vegetative Response.
Crop Science 35:1086-1093.

Yan, X.L., S.E. Beebe, and J .P. Lynch. 1995. Genetic-Variation for Phosphorus
Efficiency of Common Bean in Contrasting Soil Types .2. Yield Response. Crop
Science 35:1094-1099.

Yan, X. L., H. Liao, S. E. Beebe, M. W. Blair, and J. P. Lynch. 2004. QTL mapping
of root hair and acid exudation traits and their relationship to phosphorus uptake
in common bean. Plant and Soil 265: 17- 29.

Yu, K., 8.]. Park, V. Poysa, and P. Gepts. 2000. Integration of simple sequence

repeat (SSR) markers into a molecular linkage map of common bean (Phaseolus
vulgaris L.). Journal of Heredity 91:429-434.

125

Table 1. Putative QTL for seed traits identified from 75 recombinant inbred lines
developed from AND696/G19833 cross grown under high (HP) and low (LP) soil
phosphorus conditions in Darien, Colombia in 2000 and 2005.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Traits Linkage Nearest LOD 1?.2 AdditivityF
group markerT scoreI CIM¥
Root length density
LP 2005 B3 AGl 3.1* 0.14 -0.17
HP—2005 B1 AGT A01 4.7 0.18 -0.15
D AGTA02 4.5 0.20 -0.15
Root surface area
HP 2005 B1 fin 5.6 0.21 -12.1
D AGTA02 4.2 0.24 -12.0
Shoot P
LP 2005 B8 CTT A04 3.5 0.17 0.08
P uLtake .
LP 2005 . B8 012.2200A 2.9* 0.12 1.54
HP 2005 B11 ACACO4 3.4 0.15 4.8
P use efficiency
HP 2005 B11 R4.1400A 3.1* 0.12 -0.03
Seed yield
HP 2005 B1 AGTAOI 6.4 0.27 155
LP 2000 B1 ATA4 3.5* 0.17 -83
HP 2000 B6 BM170 4.3 . 0.45 -301
Seed P content
LP 2005 B1 . fin 8.0 0.28 -9.6
B2 BM164 4.9 0.19 -7.2
B6 GCTC03 4.5 0.13 -6.0
HP 2005 B1 fin 9.0 0.27 -l3.9
B6 GCTC03 6.8 0.19 -10.9
LP 2000 B2 BM164 3.7 ‘ 0.13 -5.0
HP 2000 - B1 ATA4 6.0 0.18 -12.6
B2 ' BM164 , 5.5 0.15 -10.8
B6 GCTC03 3.4 0.09 -8.4

1' Primer information for each marker can be found in Tables A4-1, A4-2, and

A4-3 of the Appendix.

I LOD: Log ofodds

¥ Proportion of the phenotypic variance explained by QTL at test site using CIM
(composite interval mapping).

F Effects of substituting a single allele from one parent to another. Positive values
indicate that allelic contribution is from AND696 and negative from G19833.

* Indicates that the LOD score fell below the cutoff range of 1000 permutations
at p = 0.05.

126

#00593. 05 .00 ntv< 0:0 $-30 . 70¢. 00308 E 0:00.“. 00 :00 0:03 0000 00 0:00.00E :00 8.008003 00% 000.0% 0mg
N: :0 2020 .00 038 0 00.0 00:08 0:30:00: 0:0 0:0 0:00:08 00300—05 Mmm .0302 fig 0:00:00 :08 00:. 0000300:
30 20 0e 800 85: .850882 00 :80 0002300 00020 .3 00002.... 0e .08 0000:. .080 8.0an ._ 200.:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

00.20 0.0..
.0020 0.00.
.850 ..00 .0920 ..00
.0050 0.0:
. 351-1... 0000.00.60“...
00:20 0.0 080.00. 0 00 08.5.0 .117 0.00 808.0 \ / ...00
09204. /D ~00 .
(80.92 0.3 @033 . adv . .
8520/1\000 _ 00:20 /t\00..
.Sfii .. 0.0.. . 8020 \i/ 0.0..
080.00 0.00 .0220 \ / 0.00 .
(and—.2 0.8 ”min ...om
~008< 0.00 . . .
.0<<0< _ H 0.0 .250 . 0.00 .3 a 00
.00<0< 0.0. 0280 0.0 8.20 0.: .
8200 ..0. 0850 00 .. 22 0 0.
.350 0.0 820.. JR .. . =0 1\ 0.0.
0. .0< 08.20 0.0 09200 0 0 .0502 Url 0.0
0020 0.. . .0980E 03.20 . . 8<<0< \ 0..
00.20 0.0 2.20 0 0 0000.: 0. 02.2 0 0 3:20 o 0 0.0.20 ml 0.0
. 08.02. 0.0.
_ 0o<t0 0..
.0050 0.0

30 000 30 80 «00 .00

127

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0:20 ..00
8.25 0. .0 .
~0<<<0 Mm“ 0.00 0800?. /D 0. .0
.8005 It 0.0. .850 0.0.. 08.05 /1 0.0..
530 l ...3. . (82 N;— \l/ 0.0.0
00500 \tfl 0.04 0.0.2 0. .0 8200 0 3.
30<0< 0.0.. .0080 II\ 0.0.0 83:0 0.0
00290 / \ 0.00 0.80.3 0.00 «0050 it! 0.00
th<w Mum ~0<t0 Uni. 0.00 .
$0 003.00 H/ 0.00 . .
. 0005.0 0... 6050 \ / ..00 .032 \ 0.0. . Fowfiw Mm"
8500 0.0. .0054. o. .. . .9Eo A 0.0. .02 VIA ..0. .
Noni-<0 ad roh<mv< \ / 0.6? ROCHE 0 mr 3&t0 N.w
. . 00500 0 N. 800 0.0 0 a
.0200 0 . $0.20 0.0
. .0200 0.0 . .
00502. 0 o .9200 .. 00<t0 0.0 08.20 0.0 0. :20 0.0 <02. 0.: 0.0
005.00 0.0 .250. 0.0
. .250 0 0
«000.0. 0.0
a . < _. ..m . c ..m mom mam

@0080 . 8:0:

128

003000.300 m mo 0w000>0 05 0.00 0:002 Amoom S :3on won 003 mmwm 3% 000080005 E000b% .0025 009.3000» 300.000
00 0030.. 088 080202 0305 .08 03000800 Ed 32 000 SE 00.0 a 0000 0. 0.08200 .0005 a 0300.0 0000.0
200 30% 0.003009 0.080 0 88.”. 0000300000 00:: 00.5.: 02980000 m0 3 0:00. 88 mo nova—p.520 00000000....“ .N 0.00mi

Ava 83 A 053.23
.0000. .09 00.0000 . £0000 £000.. .80
av or 3. N_. or o v MN 9N . m._. 0.0 m6

    

 

 

   

g

[or

\\\\\\\\\\\\\\\\\\\\‘

   

    

.\\\\\\\\\\\\\\\\\\\\\\\\\\\\‘ a)

   

&\\\\\\\\\\\\\\\\\\\\\\\\\\V

 

a: 00002.. q . _ # H
.0: 0000.0 ‘ -0. n... 000nz< -0. m
0.. 00002< - 00 ‘ .u0
0
-00 a: 00002... 0: 0000.0 . - 00
a... I .-00 . -00

 

 

r00 . . -00

129

Figure 2 (cont’d).

  

0.40 0.45 0.50

0.35

 

 

 

 
  

 

8
o.
00:!
_|L
CO
83. 33
CE3 0'
I l I l'-l ‘ l 'l
{-‘o’ 8 (“0’ 8 2‘3 53 ‘° °
0
$2
3
O
9.‘
n.
:I:
8 8
(D- '-
c:
z
<
—’ 8
tL—> 8
.1
§ .
g 8
< L
I I'I I I I‘ I
:3 8 a a :2 2 *0 °
Aauenbau

 

130

Average root diameter
(mm)

Root Surface Area
(cmz)

00.00.003.000 m .00 0w000>0 00. 000 000002 .308

00. 030% .00 003 mmweg 000000000000 00000th 00.00: 00050000m 00000000 .00 0030.0 0000.0 0000000000 030.004 ...00 0000030000
Ed 30. 00.0 ENE. .33 00. moom 3 0.0.8200 .0055 00. 0000.03 mmwmfiu .000 mace/2 0003000— 00000 0 800.0 0000—2000
00.0: 00.500 00000050000000 m0 0. 00300 0000000000 .0000 0000000000000 0000000000 00000 .00 0000055030 50000000. ”m 000mE

    
      

..-mx 9
A EmifiEV
T . .0000 n. .0020
00.00000 . .
00 00 0. 00 00 00 00 00 0. 0 0.0 0.0 0.0 0.0 0.0. 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0..
0
\ \\\\00 r 0
\ \\\\\ N
0 0“““\
““0““ . 0
““0““ -0 .
00.00006 ““0““ . 0
x... ..
xxxx .
“0““ -0 n... 00002.... 0
““\\ .
\\\\ o.
\.\\\
“0“ -0.
\\\ .
0““ 0.
0“ -0.
. .0 r ..
“0 -
n... 00002< I 0.
mm . 0.
00.000020. “0 -0.
0“ . fl 0.
I m“ -0. 0... 0000.0 00
a... .0 . 0100002...
0... § m. .-00 . 00
T00

 

Aouanbeu

131

cow 3N ONN 8N our Dow

01'
fig

Doom

0.... mmmeO

1

.\\\\

.9:

.\\\\\\‘

k\\\\\\\\\\\\\\\\\\\\\\\\\V

000000000000 m .00 0m000>0 000 000 000002 .808 000 039% 000 003
000972 0000 moom 000 0300» 0000 003 300000 0000000000 00.000.00.00 00000000 000000000w 0000000000 00200, 000000 0000000000 03000<
.08 0000000000 00.0 30. 000 €0.00 00.... a 0000 000 0000 0. 00080.00 00.000 00 02.000 000000 000 000072 083.00 00000
0 8000 0000005000 00000 0000000 000000080000 00. 000 200% 00000 00000 00000000 0000000000 00000 .00 000000000000 000000000000 .0 000w00

00.0000. 000.000 000.08 0 0000

ON

k\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\‘

F

cow

r.-

on

P

 

;\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

0..0

 

 

000. 0.0 000 000 00. 00. 00. 00. 0..0.
P. h I

 

co
co
0)
P
(D

a

  

$0.

I .
—> k\\\\\\\\ to W
W
E
E
. 0\\\\\

0.... 00002... 0
n... 00002<

mooN

I l I
N O

l
V

T

l
(D

V

I
Q

Aouenbau

Wop
MN?
r3
..0.
..0.
H.00
..00
new

 

~

132

2000

A‘\‘

G19833 HP

—>
I““\\‘

s\\\\\\\\\\\\\\\\\\\\‘
0‘\\\\\\\\\\\\\\\\\\\\\\\\\\‘

u\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\‘

l G19833 LP

L\\\\\\\\\\\\\““\‘\\‘\\\\\\\\\\‘\V

L\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ 1—

s\\\\\\\\\\\\\\\\\\\\\\\\\\\‘
s\\\\\\\\\\\\\\\\\\\\‘

m

 

400 600' 800 1000 1200 1400 1600 1800 2000 2200 2400

h

 

 

U I I ' T

T
(D V

T
N O

16-
14-
12-
1o-
8

1800

 
  
   

2005

AN0696 HP
1400 1600

1200

000

1

800

AN0696 LP
600

955.

  

‘400

16
14
12

O

0 (D V N O

AouenbeJd

. Figure 4 (cont’d).

133

Seed yield (kg-ha")

.8053:qu m mo ommbaw 05 Pa .382 .Amoom S :3on «on 33 @339 35880.: Sacha

noun: anabonom 333m mo 83S, Qua 38298 955 48m mEoanoag. Ed 32 23 Ag :mE S moon E «58200
.559 a 353 288 93 8324 80an 3.20 a 88 833% m8: 855 “55988“ mu a 983 05 E a: :33
A no :5. Ba 20% coca mo 2525 08 8 33m £5 E @0303 homomolmuo om: aonmmoam mo nousawammu 5893.5 .m Pawwm

Arm 9: 33.9
3:205» .3: n.

Pd

[x
o
<0
G
‘0.
o
V
o
:0
o
N
0

ad

      

   

 

    

Q

\‘r--
\\\\\V

.\\\\\\\\\ *
\\\\\\\\\

M\\- -
k\\\

n... 883 .

\
mm

  

 

AouanbeJd

3%
§
I

 

a: 883 q

134

L

.88 25 88. 5% 25 Ed 5 32

Ba Ag m :3: “magnet. >9 @328? .853 0.8 HBO .AMDB @5850 8: m 98 0M3? m dosmbnoonoo m 80% .Amoafludmv no.8
8555 82 646 £33 eme— 58 52:8 m 83 5E» Ba 5 3882 db 5? 5a owflfi 23665372 5 25mm

 

5.:va

50(00

 

moOF0<

 

mmem

 

r06.

 

RLD-LP2005

flow

N65

0.3

96¢

NNN

82% /1\ P. :
«9200 \1/ mm

 

 

m w<h<

 

mam

od

SeedPoontent-LPZOOO
Seechontent-HPZOOO
Seechontent-LP2005

mmzzm

53.30

50((0

we _. .2 m
mo._.<<0

vow—2m

vo...<<0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

to:
9.85
.8
. «2.56 38
to . m85</m\m.8
3m 38b<\ /«.$
5
. 3
Brim/Kane .1
\ 8~2m\I/m.8
o R
m 5 5 m 82m v.8
H m m 5 55 o
... 2 m w 455 m
S. m m u w m WW;
n 4.... 4m a H mew # 21P< D.W..
m m m u R po<5<\llo.o
. mm W s 3<<o<l \3
S m .
00 d m Buzm\ (3

3m

PUE-HP2005

 

mo<<<o .
NO(((0 ./

.885 [m
Fo<<00

 

8500 \
30<0<

No._.<@< /

 

Puptake-HP2005

~o<torr
(83.x. \ n
So<00.\

 

 

no<t0

l
//|\ // \\
:02ch wozvnam.

$1888 3$3$$$

 

:m

0.0

Pu aka-LP2005
ShootP-LP2005

 

 

 

 

«88?. / \ a“ B
825$ / \ 0.9.
(80:02 I o 9.
3.200 \ / 34
8.2.6 «.5
Soto. a. 2
3205‘ «.2
35.5 a.»
80:20 3.
:35 v.0
32$ 3
3500 . .3.
8t<0 a.»
<83. o5
mam

SeedPoontent-HP2000

Surfarea-HP2005

Seechontent-HP2005

 

 

 

 

 

 

 

 

 

 

m 8500 S.
m mm ~o<h0< 9o
R .
a
5.2m 3.0
m
1..
m 8050 98
m
W
2:6 2.
mom

.5580 0 2&3

136

Chapter 5: QTL analysis of Seed Nutrient Levels in an Andean Bean
Population Segregating for Tolerance to Low Phosphorus Soils
Abstract:

Iron and zinc are essential micronutrients for human health that are often
found in insufficient quantities in the diet, and biofortification of seed crops has
been undertaken as a strategy to reduce micronutrient malnutrition.
Bioavailability of Fe and Zn is reduced by seed phytate, the major form of
phosphorus in the seed. Seed P levels also play a role in tolerance to low P soils,
a major constraint of common bean production in the tropics. The objective of
this study was to identify environmental and genetic variability for seed Fe, Zn
and P concentrations in an F 57 recombinant inbred line (RIL) population
developed from Andean common bean lines (AND696/(319833) with different
tolerance to low P soils. Quantitative trait loci (QTL) analysis was conducted
based on phenotypic data across two years and two environments (high and low
soil P levels) and a previously described linkage map of AND696/G19833.
Significant environmental and genetic variability for Fe, Zn, and P concentrations
was identified population. Seed P levels were positively correlated with seed Fe
and Zn levels. QTL for seed P levels explaining 17 to 55% of the total
phenotypic variation depending on year and environment were identified across 4
linkage groups. QTL for Fe levels were identified across 5 linkage groups and
QTL for Zn levels were identified on 2 linkage groups QTL for both Fe and Zn
levels clustered on linkage groups B1 and B6 and also co-localized with QTL for

seed P concentration on linkage group B1.

137

Introduction:

Micronutrient malnutrition is a major global health concern (Kennedy et
al., 2003). More than half the population of the developing world receives
inadequate micronutrient levels in their diets (Bouis, 2003). Two of the most
widespread micronutrient deficiencies are that of Fe and Zn. Iron deficiency
affects 3.7 billion peOple (Welch, 2002) and an estimated 49% of the human
population is at risk for inadequate zinc in their diet (International Zinc
Association, 2000).

Through the international, interdisciplinary research program HarvestPlus,
a proposal was developed to alleviate micronutrient malnutrition by increasing the
content of Fe and Zn in staple crops (Harvest Plus, 2004). Common bean
(Phaseolus vulgaris L.), an important legume staple in much of the developing
world, is one of the crops targeted for biofortification. Bean has long been
recognized as a nutritionally valuable food for human nutrition. Although a cup of
bean supplies 25% of the recommended daily allowance of iron and 15% RDA of
zinc, the potential exists to develop varieties with two to three times the Fe and Zn
content. Genetic variability for seed Fe concentration is 34 to >100 ug/g and 21 to
54pg/g for seed Zn concentration (Beebe et al., 2000).

Increased mineral content in the diet does not guarantee increased
micronutrient status for the consumer. There are numerous compounds in plant-
based diets that reduce micronutrient absorption by the body. Among these,
phytate is the major determinate of Fe and Zn bioavailability in the diet (Lopez et

al., 2002). Phytate (myo-inositol 1,2,3,4,5,6 hexakisphosphate) is the major form

138

of phosphorus in cereal and legume seeds (Raboy, 1990). Phytic acid has 6

‘ phosphate groups that can covalently bind to cations, especially K+, Ca2+, Mg“,
Fe”, and Zn2+ (Tsao, 1997). Under human physiological conditions, some
phytate complexes are insoluble, thus decreasing the bioavailability of phosphorus
and cations, especially iron and zinc, stored as a part of the complex (Schlemmer
et al., 1995). Numerous studies, using a variety of experimental techniques, have
shown a correlation between high phytate diets and limited Fe and Zn absorbance
in the gastrointestinal tract of humans and animals (Saha et al., 1994; House et a1,
1982; Tumlund et al., 1984; Hunt et al., 1998; Zhou et al., 1992).

‘ On one hand, P stored in legume seeds (in the form of phytic acid) can
reduce micronutrient absorption by humans, but on the other hand, P is an
element essential to plant growth that is often limited in the environment. In fact,
P deficiency is one of the most widespread abiotic stresses affecting common
bean production (W ortmann et al., 1998). Soil P deficiency often occurs in the
same regions of the world plagued by micronutrient deficiencies in humans.
Screening of common bean germplasm for ability to grown in low soil P
conditions has resulted in the discovery of bean lines tolerant to low P soils
(Lynch and Beebe, 1995; Yan etal., 1995 a,b; Beebe et al., 1997). QTL studies
conducted with P efficient lines resulted in the identification of regions of the
bean genome that play a role in tolerance to low P soils, and plant phenotypes
associated with those regions (Beebe et al., 2006; Ochoa et al., 2006; Yan et al.,

2004; Liao et al., 2004).

139

Since biofortification of common bean for Fe and Zn is underway and
phosphorus plays such an important role in plant and human nutrition, an
increased understanding of the genetics of tolerance to low P soils together with
the genetics of seed P, Fe, and Zn concentration will be beneficial in the
development of stress resistant and micronutrient fortified crops.

The objectives of this research were first, to measure seed P, Fe, Zn, and
phytic acid levels in a population of recombinant inbred lines developed from
common bean line G19833 tolerance to P soil and AND696, a line susceptible to
low P soil. The second objective was to examine the relationship among these
nutrients in the seed and tolerance to low P soils. The third objective was to
identify QTL'for seed nutrient levels in the RIL population.

Materials and Methods:

Seventy seven F57 recombinant inbred lines were developed from a cross
between two common bean lines from the Andean gene pool. The initial cross
619833 x AND696 was advanced to the F59 generation by single seed descent
and seed of the RILs was increased for field studies. The parent 619833 is a
Peruvian landrace with an indeterminate (Type III) growth habit. G19833 has
large yellow and red mottled seed with an average 100 seed weight of 41 g.
G19833 has also been identified as tolerant to low P soils (Y an et al., 1995a; Yan
et al., 1995b). The parent AND696 is a CIAT improved line fiom the race Nueva
Granada. It has a determinate growth habit (Type I) and has large red and white
mottled seed with an average 100 seed weight of 51g. AND696 has been

identified as susceptible to low phosphorus soils (CIAT, 2000).

140

In 2000 and 2005, the 77 RILS, G19833, AND696, and two check
varieties, G4017 (Carioca) and G16140 were planted in Darien, Colombia in a
9x9 lattice design with three replications at two soil P levels, low P (45 kg/ha
triple super phosphate) and high P (300 kg/ha triple superphosphate). Phenotypic
data was collected on 75 of the 77 RILS. The soil of this site in an Andisol with a
native soil P of 2 mg/kg based on bray II extraction method. Seed was hand-
planted at 10 cm spacing in 4 row plots. Each row was 3 m in length. The middle
two rows were planted to the genotypes of interest and the outer two rows were
border rows of Mesoamerican cultivar BAT 477.

Seeds reached maturity at 76 to 98 days after planting. At maturity, seeds
were hand-harvested. Seeds were dried and seed yield was determined at 18%
moisture. Weight of 100 seed was measured at 18% moisture. A sub sample of
seed was analyzed for P, Fe, and Zn concentration. Five grams of seed of each
treatment and replication were cleaned with distilled water and dried. They were
then freeze dried to remove all moisture. Freeze-dried samples were ground to a
fine powder. Total seed P analysis was conducted according to Murphy and Riley
(1962). Iron and Zn were extracted from seed with a Nitric-percloric (2:1)
digestion according to the method of Benton and Jones (1989). Atomic
absorption spectrometry was subsequently used to measure Fe and Zn
concentration of the seed samples.

Phytic acid was measured in seed from the 2005 season by high performance
liquid chromatography (HPLC) using a modified version of Graf and Dintis

(1982). Sample preparation for PA analysis in bean seed was conducted

141

following the procedure of Lehrfeld (1989). PA was extracted from the samples
by the addition of 20ml 0.65M HCl (trace element grade) to 0.500 g of sample.
The acidified samples were mechanically agitated at 250rpm for 2 hours at 21°C.
Samples were centrifuged at 4,000 rpm for 15 minutes. The supemant was
collected and diluted 1:5 (v/v) with HPLC grade water. A 3 m1 aliquot of the
diluted sample (15 m1 total) was passed through a Bond Elut strong anion
exchange column (V arian, Walnut Creek, CA) for purification. The column was
washed once with 5 ml 0.14 M HCl, followed by a 10 ml wash with 0.5M HCL.
Phytic acid was eluted from the column with 3m12M HCl. The eluted fraction
containing PA was freeze-dried and dissolved in SmM sodium acetate. The
dissolved sample was filtered through a 2pm filter. PA levels in each sample
were quantified by HPLC with a refractive index detector. The column used for
analysis was a Waters Symmetry C18 column (3.9mm x 150mm) (Waters,
Milford, MA) heated to a temperature of 40°C. Sodium acetate (SmM) was used
as the solvent at a flow rate of 1.4 ml min'l. Phytic acid dodecasodium salt from
corn (Z. mays L.) (Sigma, St. Louis, MO) was used as a standard to determine PA
concentration. Data was converted to concentration on a dry weight basis.
Statistical analysis was conducted with SAS for Windows V8 (SAS
Institute, Cary, NC). The command PROC GLM was used to determine treatment,
genotype and interaction effects. Data was analyzed as a split plot with soil P
level as the whole plot and plant genotype as the Split plot. The command PROC
CORR was used to determine Pearson correlation coefficients among variables.

Effects of plant growth habit on measured variables were determinedby means

142

comparisons of determinate vs. indeterminate growth habit types and differences
were established using Tukey’s test for significance. Bartlett test for variance
homogeneity was conducted for growth habit because there were unequal number
of determinate (56) and indeterminate (21) RILS.

A linkage map of the G19833/AND696 was developed using JoinMap
J oinMap 3.0 for Windows (Van Ooijen and Voorrips, 2002) as described in the
Materials and Methods section of Chapter 4 in this thesis. QTL analysis was
conducted with the genetic map developed with the J oinMap program and with
the phenotypic means for each RIL collected from the field study. The computer
software program Windows QTL cartographer version 2.5 (Wang et al., 2006)
was used to identify QTL for seed weight, seed P, Fe, Zn, and phytic acid
concentrations. The Composite Interval Mapping (CIM) feature of Windows
QTL cartographer employing the forward and backward regression model set
tolOcM window width was used to identify QTL. CIM analysis is based on
maximum likelihood estimates and linear regression to identify QTL within
marker intervals. Significant QTL were considered by defining the LOD score at
p=0.01 after 1000 permutation tests.
Results:

Variability:

Seed weight: Seed weight was influenced by genotype and environment
in the RIL population in both 2000 and 2005. There were no significant genotype
x environment interactions (GxE) (Table 1). The mean 100 seed weight of the

RILS was 2-3 g higher in the high P environment than the low P environment in

143

each year (Table 2). The ranges for seed weight overlapped between the
environments, in each year (Figure 1). AND696 had larger seeds than 619833 in
the high P environment (51.4 and 41.4 g per 100 seed, respectively) (Figure 1).

Seed P and phytic acid: Genotype and environment had a significant
effect on seed P levels of the RILS in 2000 and 2005. In each year there was a
significant GxE for this trait (Table 1). Concentration of seed P was higher in the
high P than the low P environment each year (Table 2). The frequency
distributions of seed P showed more variability for this trait in the high P
environment than the low P environment (Figure 1). 619833 had higher seed P
concentration in the high P environment than AND696 (Figure 1). Percent phytic
acid, which was only measured in 2005, was significantly affected by soil P
environment (Table 1). There was only a genotype effect for this trait observed in
the low P environment (Table 2). In the high P soil environment, AND696 was at
the low extreme of the RILS, at 1.46 %, whereas the range for the RILS ranged
from 1.33 to 2.85 % (Figure 2).

Seed Fe and Zn: The concentrations of Fe and Zn in the seeds were
significantly affected by genotype and environment. There was no significant 6 x
E for either of these traits (Table 1). Iron levels were higher in the high P
environment, whereas Zn levels were higher in the low P environment (Table 2).
The inhibition of Zn uptake by high levels of soil P has been observed previously
in common bean (Singh et al., 1988) and the phenomena is known as P induced

Zn deficiency. There was greater variability for Fe levels in the population that

144

for Zn levels (Figure 3). The range of Fe levels was larger in 2005 than in 2000
(Figure 3).

Correlations: Correlations with seed weight and the other seed traits were
not consistent across years. In 2000, seed weight was positively correlated with
seed Fe and Zn in both high and low P treatments (Table 4). In 2005, seed weight .
was only positively correlated with seed Zn in the high P treatment (Table 3).

In each year and each P environment, seed P was positively correlated
with seed Fe and Zn concentrations (Tables 3 and 4). Phytic acid % was also
correlated with seed Fe and Zn in high and low P environments, but not as
strongly as total seed P (Table 4). In 2005, seed Fe and Zn concentrations were
weakly and significantly correlated with P uptake in the low P treatment, I = 0.17
and 0.12 respectively. In the high P treatment, only seed Zn was correlated with P
uptake (r = 0.32).

Growth Habit: Seed weight, P, Fe, Zn, and phytic acid were each
influenced by plant growth habit (Table 5). The indeterminate RILS had larger
seed weights (49.6 g in high P and 46.2 g in low P) on average than the
determinate RILs (46.8 g in high P and 43.7 g in low P) and also had greater
concentrations of P, Fe, and Zn. There was an exception to this trend observed in
2000, under low P conditions where the determinate lines had higher seed P
concentrations (Table 5). The differences in seed P concentration among years in
may relate to differences in seed yield by growth habit in 2000 and 2005. In 2000
the average yield of the indeterminate lines was 707 kg-ha'1 under low P

conditions and that of the determinate lines was 614 kg-ha'l. In 2005, there were

145

no differences in yield in the low P treatment based on growth habit. (data shown
in Chapter 3). Since there were unequal number of determinate and indeterminate
RILs in the population, Bartlett’s test for variance homogeneity was conducted to
identify bias that may have existed in trait analysis by growth habit. Most traits
exhibited variance homogeneity although there were exceptions, including seed
weight (high P, 2005), seed phytic acid (low P, 2005) and seed Zn (low P, 2005)
(Appendix, Tables A4-4).

QTL identification:

Using composite interval mapping, 32 QTL were identified for 5 seed
traits in 15 marker intervals on 8 linkage groups. Twenty-eight percent of the
QTL were clustered on linkage groups B1 and affected more than one trait.
Tweleve percent of the QTL were clustered on linkage groups B3. An additional
31% of the QTL were located on linkage group B6 and also affected multiple
traits. Individual QTL explained 10 to 35% of the phenotypic variation, and total
phenotypic variation explained for any one trait was 10 to 61% (Table 1).

Seed weight: A QTL for seed size was identified on linkage group B6 in
8 each year and environment where the increased effect was contributed by 619833
(Table 6, Figure 4). This QTL explained 21 to 27% of the phenotypic variability
in 2005, depending on environment and only 13 to 15% of the variability in 2000.
An additional QTL for seed size was found in 2005 on B3 where the increased
effect was contributed by AND696 (Table 6, Figure 4). QTL for seed size have
previously been identified on B3 and B6 in a population developed from bean

lines PC50 and XAN159 segregating for white mold resistance (Park et al., 2001).

146

A QTL for seed size, in the same region of B3 as that identified by Park et a1.
(2001) has also been found in the BAT 93/Jalo EEP558 population (BIC, 2006)

Seed P and phytic acid: QTL for seed P were found on B1 near the gene
for plant grth habit. QTL found for seed P concentration on Bl explained 17
to 28% of the phenotypic variation for this trait depending on year and soil P level
(Table 6, Figure 4). In 2000 and 2005 in the high P environment the increased
effect was contributed by 619833, whereas in low P 2000 the increased effect
was contributed by AND696.

A QTL for phytic acid on B8 mapped to same region as a QTL for total
seed P in high P, 2005, where the increased effect was contributed by AND696
(Table 6, Figure 4). The QTL for % phytic acid explained only 10% of the
phenotypic variation and the QTL for total seed P explained 20% of the
phenotypic variation. .

Seed Fe and Zn: QTL for both seed Fe and Zn were identified in the
identical regions of B1 and B6 (Table 6) across years and environments. QTL
found on B1 were near the fin gene, and also overlap with QTL for seed P
concentration. 619833 was the source of the QTL identified for Fe and Zn
concentration in the seed on B1. Those QTL found for seed Fe and Zn
concentration on linkage group B6 explained 15 to 26% of the phenotypic
variation. They also co-localized with QTL for seed weight. AND696 was the
source of the increased effect of each of these QTL, conversely, 619833 was the
source of the QTL for seed size identified in this region of linkage group B6

(Table 6).

147

Discussion:

Increasing the Fe and Zn concentration of seed crops is a major goal that
bridges agriculture and human nutrition. In common bean, significant natural
variability exists for seed Fe and Zn, making such a goal achievable through plant
breeding (Beebe et al., 2000). Within the AND696/619833 RIL population, Fe
levels ranged from 38 to 79 mg kg1 and Zn levels ranged from 16 to 33 mg kg".
Understanding the genotype x environment interactions of these traits is essential
to developing nutritious seeds for different environmental conditions. In this
study, 6 x E was not significant for Fe or Zn seed concentration, suggesting that
breeding for increased levels of these elements is possible under diverse soil P
environments. -

Mutational studies in plants have shown that many genes effect
micronutrient accumulation and many remain to be discovered. QTL analysis
will allow gene discovery to move forward for genes underlying natural variation
in seed Fe and Zn levels (Ghandilyan et al., 2006). Overlapping QTL for Fe and
Zn concentration were identified on two linkage groups in this study. This
_ demonstrates the possibility of breeding for these traits simultaneously.

QTL for increased seed Fe and Zn concentration were located on the same
region as a QTL for seed P concentration. These QTL were all contributed by ‘
619833, which was the parent exhibiting tolerance to low P soils. This region of
the genome (linkage group B1) may contain genes that improve tolerance to low
P soils, as well, according to yield QTL data (Chapter 4, Table 1). Further

research is required to determine the consequences of simultaneously selecting

148

increased seed Fe, Zn, and P levels on the bioavailability of these micronutrients
in humans upon consumption.

In this study, only a single QTL for seed phytic acid was detected (Table
6). This QTL was not in the same region as QTL for seed Fe or Zn concentrations.
In Arabidopsis, QTL for seed phytic acid levels have been separate from those for
seed Fe and Zn levels suggesting it is possible to increase Fe and Zn levels
without increasing phytate levels as well (Bentsink et al., 2003; Ghandilyan et al.,
2006)

Since many of the QTL detected in this study for each of the traits co-
localized it is interesting to consider the physiological role of the genes
underlying these QTL. Could these genes be involved in seed transport or in
uptake of nutrients by the plant roots? Iron uptake by pea (Pisum sativum) has
been shown to increase during seed filling as compared to earlier in plant
development. During this time, a constant source of micronutrients in required in
the xylem stream for xylem-'to-phloem exchange to occur, and for micronutrients
to arrive in the seed (Grusak et al., 1999).

QTL for seed Fe, Zn, and P on B1 co-localized with the fin gene (Figure
4). This is the single gene in bean responsible for deterrninacy (Norton, 1915).
The QTL for increased levels of each of these elements was conferred by the
indeterminate parent (61983 3). That indeterminate plants generally continue root
growth beyond flowering whereas determinate plants do not (Huyghe, 1998),
which may explain why this region of the genome influenced these traits for

increased nutrients in the seed.

149

Conclusions:

Significant variability for seed weight, P, Fe and Zn concentrations was
present in the AND696/619833 population. Seed P concentration was correlated
with seed Fe and Zn concentrations. QTL for these traits were found to co
localize on linkage group B1. Seed Fe and Zn concentrations were also correlated
and QTL for these traits co-localized not only to linkage group Bl, but to B6 as
well. The absence of GxE interactions for seed Fe and Zn levels is promising in
terms of future development of lines with increased levels of these nutrients when

planted in diverse environments.

150

References:

Beebe S, Gonzalez A, and J. Rengifo 2000. Research on trace minerals in the
common bean. Food Nutr. Bull. 21: 387-391

Beebe, S., J. Lynch, N. Galwey, J. Tohme, and I. Ochoa. 1997. A geographical
approach to identify phosphorus-efficient genotypes among landraces and wild
ancestors of common bean. Euphytica 95:325-336.

Beebe, S.E., M. Rojas-Pierce, X.L. Yan, M.W. Blair, F. Pedraza, F. Munoz, J.
Tohme, and J .P. Lynch. 2006. Quantitative trait loci for root architecture traits

correlated with phosphorus acquisition in common bean. Crop Science 46:413-
423.

Bean Improvement Cooperative (BIC) 2006. Bean Core Map 2006 [Online]
http://www.css.msu.edu/bic/Genetics.cfin (posted 2006; verified November 14).

Benton, J .B., and J. Jones. 1989. Plant analysis techniques Benton-J ones
Laboratories, Georgia.

Bentsink, L., K. Yuan, M. Koornneef, and D. Vreugdenhil. 2003. The genetics of
phytate and phosphate accumulation in seeds and leaves of Arabidopsis thaliana,
using natural variation. Theoretical and Applied Genetics 106: 1234-1243.

Bouis, HE. 2003. Micronutrient fortification of plants through plant breeding: can

it improve nutrition in man at low cost? Proceedings of the Nutrition Society
62:403-41 1.

CIAT. 2000. Inheritance of low phosphorus tolerance in the Andean population
AND696 x 619833. Bean Improvement for the Tropics Unit Annual Report.
CIAT, Cali, Colombia.

Ghandilyan, A., D. Vreugdenhil, and M.G.M. Aarts. 2006. Progress in the genetic
understanding of plant iron and zinc nutrition. Physiologia Plantarum 126:407-
417. '

Graf, E., and FR. Dintzis. 1982. Determination of Phytic Acid in Foods by High-
Perforrnance Liquid-Chromatography. Journal of Agricultural and Food
Chemistry 30: 1094-1097.

Grusak, M.A., J .N. Pearson, and E. Marentes. 1999. The physiology of
micronutrient homeostasis in field crops. Field Crops Research 60:41-56.

Harvest Plus 2004 Bioforified Beans Available:
http://www.harvestplus.org/pubs.html

151

House, W.A., R.M. Welch, and DR. Vancampen. 1982. Effect of Phytic Acid on
the Absorption, Distribution, and Endogenous Excretion of Zinc in Rats. Journal
of Nutrition 112:941-953.

Hunt, J .R., L.A. Matthys, and L.K. Johnson. 1998. Zinc absorption, mineral
balance, and blood lipids in women consuming controlled lactoovovegetarian and
omnivorous diets for 8 wk. American Journal of Clinical Nutrition 67:421-430.

Huyghe, C. 1998. Genetics and genetic modifications of plant architecture in
grain legumes: a review. Agronomic 18:383-41 1.

International Zinc Association 2000. Zinc and Human Health. International Zinc
Association. Brussels, Belgium.

Koinange, E.M.K., S.P. Singh, and P. Gepts. 1996. Genetic control of the
domestication syndrome in common bean. Crop Science 36: 1037-1045.

Lehrfeld, J. 1989. High-Performance Liquid-Chromatography Analysis of Phytic
Acid on a Ph-Stable, Macroporous Polymer Column. Cereal Chemistry 66:510-
5 1'5.

Liao, H., X.L. Yam, 6. Rubio, S.E. Beebe, M.W. Blair, and J.P. Lynch. 2004.
Genetic mapping of basal root gravitropism and phosphorus acquisition efficiency
in common bean. Functional Plant Biology 31 :959-970.

Lynch, J .P., and SE. Beebe. 1995. Adaptation of Beans (Phaseolus vulgaris L) to
Low Phosphorus Availability. Hortscience 30:1165-1171.

Murphy, J ., and J .P. Riley. 1962. A modified single solution for determination of
phosphate in natural waters. Anal Chem Acta 27:31-36.

Norton, J .B. 1915. Inheritance of habit in the common bean. Am. Nat.
49:547-561.

Ochoa, I.E., M.W. Blair, and J .P. Lynch. 2006. QTL analysis of adventitious root
formation in common bean under contrasting phosphorus availability. Crop
Science 46:1609-1621.

Park, S.O., D.P. Coyne, J .R. Steadman, P.W. Skroch, 2001. Mapping of QTL for
resistance to white mold diseases in common bean. Crop Sci41: 1253-1262.

Raboy V. 1990. The biochemistry and genetics of phytic acid synthesis. In: Morre

DJ, Boss W, Loewus FA (eds) Inositol metabolism in Plants. Alan R. Liss, New
York, pp 52-73.

152

Saha, P.R., C.M. Weaver, and AC. Mason. 1994. Mineral Bioavailability in Rats
fi'om Intrinsically Labeled Whole Wheat-Flour of Various Phytate Levels. Journal
of Agricultural and Food Chemistry 42:2531-2535.

Schlemmer, U., K. Muller, and K.D. J any. 1995. The Degradation of Phytic Acid
in Legumes Prepared by Different Methods. European Journal of Clinical
Nutrition 49:S207-S210.

Tsao, G.T., Y.Z. Zheng, J. Lu, and CS. Gong. 1997. Adsorption of heavy metal
ions by immobilized phytic acid. Applied Biochemistry and Biotechnology 63-
51731-741.

Tumlund, J.R., J.C. King, W.R. Keyes, B. Gong, and MC. Michel. 1984. A
Stable Isotope Study of Zinc-Absorption in Young Men - Effects of Phytate and
Alpha-Cellulose. American Journal of Clinical Nutrition 40:1071-1077.

Plant Research International BV. 2002. JoinMap. Release 3.0. Plant Research
International BV.

Singh, J. P., Kararnanos, RE, and J .W.B. Stewart, 1988. The Mechanism of
Phosphorus-Induced Zinc Deficiency in Bean. Canadian Journal of Soil Science.
68: 345-358.

Statistical Genetics, North Carolina State University, USA. 2006. Windows QTL
Cartographer. Release 2.5. Statistical Genetics, North Carolina State University,
USA.

Welch, RM. 2002. The impact of mineral nutrients in food crops on global
human health. Plant and Soil 247:83-90.

Wortmann, C.S., R.A. Kirkby, C.A. Eledu, and DJ. Allen. 1998. Atlas
of common bean(Phaseolus vulgaris L.) production in Africa CIAT, Cali,
Colombia.

Yan, X.L., J .P. Lynch, and SE. Beebe. 1995. Genetic-Variation for Phosphorus
Efficiency of Common Bean in Contrasting Soil Types .1 .V Vegetative Response.
Crop Science 35: 1086-1093.

Yan, X.L., S.E. Beebe, and J .P. Lynch. 1995. Genetic-Variation for Phosphorus
Efficiency of Cormnon Bean in Contrasting Soil Types .2. Yield Response. Crop
Science 35: 1094-1099.

Yan, X.L., H. Liao, S.E. Beebe, M.W. Blair, and J .P. Lynch. 2004. QTL mapping

of root hair and acid exudation traits and their relationship to phosphorus uptake
in common bean. Plant and Soil 265:17-29.

153

Zhou, J .R., E.J. Fordyce, V. Raboy, D.B. Dickinson, M.S. Wong, R.A. Burns, an
J .W. Erdman. 1992. Reduction of Phytic Acid in Soybean Products lrnproves
Zinc Bioavailability in Rats. Journal of Nutrition 122:2466-2473.

154

Table 1: Analysis of variance of seed traits in a population of 75 recombinant
inbred lines from the AND696/619833 population field grown in Darien,
Colombia in 2000 and 2005 in two environments: high and low soil phosphorus.

 

Mean Squares

 

 

 

Source df Seed Seed P Seed Seed Seed
weight Fe Zn phytic
acid
2005
Genotype (G) 74 127‘" 0.29"" 205’" 26“" 0.11 "5
Environment (E) 1 1094"" 95.13‘" 3404*" 943‘" 15.49"”
G x E 74 7.3 “S 0.10‘ 36 "S 7 "S 0.10 “5
2000
Genotype (G) 74 104"" 0.26"” 133‘" 20*" -----
Environment (E) 1 540*" 72.07"" 459‘" 862"" -----
G x E 74 13 "S 0.18"” 43 "5 6 “5 -----

 

"3 indicates not significant

* Indicates significance at P < 0.05.
**Indicates significance at P <0.01.
*** Indicates significance at P<0.001

155

Table 2: Seed traits in high (HP) and low (LP) phosphorus soils conditions for
parents AND696 and 619833 and the means and ranges of 75 recombinant inbred
lines developed from the parents. The experiment was planted in 2000 and 2005
in Darien, Colombia. Mean values presented (n=3). P value indicates level of
significant genotypic differences among the RILS for each trait.

 

 

 

 

 

 

 

Parents Recombinant Inbred Lines

Traits P AND696 619833 mean range P value
level
2005
Seed weight HP 51.4 41.4 47.6 39.5-59.9 <.0001
(g .100 seed") LP 44.7 ------- 44.5 33.3-56.5 <.0001
Seed P HP 3.84 4.00 4.05 3.56-4.85 <.0001
(g kg") LP 3.28 ------- 3.14 2.63-3.51 <.0001
Seed Fe HP 62.5 61.6 58.3 39.02-78.75 <.0001
(mg -kg") LP 48.6 ------- 52.8 37.92-67.66 <.0001
Seed Zn HP 20.2 21.0 22.7 18.74-30.35 0.1550
(mg kg") LP 26.3 ------- 25.6 20.18-32.83 <.0001
Seed phytic HP 1.46 ------- 1.77 1.33-2.85 0.2214
acid (%) LP 1.26 ------- 1.38 1.09-2.46 0.0272
2000

Seed weight HP ------ 34 44.94 35.5-56.75 <.0001
(g -100 seed") LP ------------ 42.78 30555 <.0001
Seed P HP ------ 4.77 3.54 2.84-4.98 <.0001
(g -kg") LP ------------ 2.70 2.36-3.04 <.0001
Seed Fe HP ------ 78.45 64.97 53.3-76.9 <.0001
(mg -kg") LP ------------ 63.01 53.1-75.6 <.0001
Seed Zn HP ------ 25.43 20.60 16.3-27.4 <.0001
(mg -kg'1) LP ------------ 23.59 18.9-29 <.0001

 

156

Table 3. Phenotypic correlations among seed traits in a population of 75
recombinant inbred lines from the AND696/G19833 population field grown in
Darien in 2005 in high and low soil phosphorus. The values above the diagonal
line are under low soil P and the values below the diagonal line are under high

 

 

soil P.

LP Seed P Seed Seed Seed Fe Seed Zn

------------ phytic weight

HP acid
Seed P ---------- 0.35'" -0.05 0.53'" 0.50'"
Seed phytic 0.27‘" ---------- 0.07 0.16" 0.26""
acid
Seed weight 0.04 -0.09 ----------- 0.10 0.10
Seed Fe 031‘" 0.16" 0.00 ........... 053"”
Seed Zn 0.48"" 0.10 0.23"" 0.31‘” ...........

 

* Indicates significance at P < 0.05.
**Indicates significance at P <0.01.
*** Indicates significance at P<0.001

157

Table 4. Phenotypic correlations among seed traits in a population of 75
recombinant inbred lines developed fromAND696/Gl9833 and field grown in
Darien in 2000 in high and low soil phosphorus). The values above the diagonal
line are under low soil P and the values below the diagonal line are under high

 

 

soil P.
LP Seed P Seed Seed Fe Seed Zn
------------ weight

HP
Seed P ---------- -0.13' 0.18" 0.28""
Seed weight 026'" ----------- 0,14" 0,14"
Seed Fe 0.39"" 0.15" ........... 0,17‘"
Seed Zn 0.65‘" 0.26‘" 0.45‘" ...........

 

* Indicates significance at P < 0.05.
**Indicates significance at P <0.01.
*** Indicates significance at P<0.001

158

Table 5. Seed traits in a population of 75 recombinant inbred lines developed
from an AND696/G19833 cross and grown under high and low soil phosphorus
conditions in Darien, Colombia in 2000 and 2005, grouped and averaged by plant
growth habit of which there were two categories, indeterminate (1nd.) and
determinate (Det.). Mean values presented (n=3).

 

 

 

 

 

 

Traits High P Low P

Ind. Det. Ind. Bet.

2005
Seed weight 49.6 a)‘ 46.8 b 46.2 a 43.7 b
(g 100 seed")
Seed P 4.23 a 3.97 b 3.2 a 3.1 b
(g 'kg")
Seed Fe 63.3 a 56.1 b 55.7 a 51.4 b
(mg 158")
Seed Zn 24.22 a 21.96 b 27.4 a 24.7 b
(mg 'kg")
Seed phytic 1.78 a 1.76 a 1.46 a 1.34 b
acid (°/o)
2000

Seed weight 46.8 a 44.2 b 43.6 a 42.4 a
(g .100 seed'l)
Seed P 3.7 a 3.5 b 2.6 a 2.8 b
(g 458")
Seed Fe 67.4 a 64.0 b 63.4 a 62.8 a
(mg 158") ‘
Seed Zn 22.3 a 19.9 b 24.4 a 23.2 b
(mg kg“)

 

1' Significant differences are based on Tukey tests and are at alpha = 0.05. Values
that do not share the same letter are significantly different. Tests should be read
across rows and within each phosphorus treatment level, High P or Low P.

159

Table 6. Putative QTL for seed traits identified from 75 recombinant inbred lines
developed from an AND696/619833 cross grown under high (HP) and low (LP)
soil phosphorus conditions in Darien, Colombia in 2000 and 2005.

 

Seed traits Linkage Nearest LOD if AdditivityF

 

 

 

 

 

 

 

 

 

 

goup marker score? CIMI
Seed wt.
LP 2005 B3 AGTC03 5.6 0.18 2.1
B6 GCTC03 6.7 0.21 -2.4
HP 2005 B3 AGTC03 8.4 0.31 2.7
B6 BM170 7.3 0.27 -2.6
LP 2000 B3 AGTC03 2.7* 0.16 1.9
B6 GCTC03 3.8 0.15 -1 .9
HP 2000 B2 BM164 3.9 0.35 -2.8
B6 GCTC03 3.8 0.13 -1.8
Seed P
LP 2005 B3 AGTC03 4.3 0.22 0.09
HP 2005 B1 AGTAOI 4.3 0.17 -0.14
B4 CGTCOS 3.6 0.18 0.13
B8 16.550A 4.3 0.20 0.14
LP 2000 B1 AGTAOI 4.3 0.17 0.08
HP 2000 B1 fin 6.9 0.28 -0.22
B2 BM143 3.1 0.10 -O.13
A ACTC06 3.6 0.14 -0.15
Phytic acid
LP 2005 B8 . 16.550A 2.3* 0.10 0.06
Seed Fe
LP 2005 B1 BMle 5.4 ' 0.18 -2.4
B6 BM170 6.9 0.26 2.7
HP 2005 B1 AGTAOI 8.1 0.24 -3.9
B6 BM170 5.3 0.15 2.8
B9 ATA9 5.1 0.14 2.8
A GGATOZ 3.3 0.08 22
LP 2000 BS M12.1600A 4.0 0.15 1.86
HP 2000 B1 BM165 2.6* 0.10 -2.1
B6 BM170 4.5 0.22 3.2
Seed Zn
LP 2005 B1 fin 7.5 0.18 -l .2
B6 BM170 8.2 0.24 1.3
HP 2005 B1 fin 6.8 0.31 -1.3
LP 2000 B1 BMle 28* 0.10 -0.7
B6 BM170 4.1 0.16 0.9
HP 2000 Bl fin 7.8 0.35 -1.5

 

160

Table 6 (cont’d).

T LOD: Log of odds

1 Proportion of the phenotypic variance explained by QTL at test site using CIM
(composite interval mapping).

F Effects of substituting a single allele from one parent to another. Positive values
indicate that allelic contribution is from AND696 and negative from G19833.

. Indicates that the LOD score fell below the cutoff range of 1000 permutations
at p = 0.05.

161

doom mom one 5m: 65 no 805 was meow 8m 98 a3 65 no 3&5 .maougnmfl m «o 09985 Ba Ba mos—3w, .Aooom a :3on
«on an? 80% Ba moon 3 85on “on ma? «.meva 38:58.: wannabe BEE mogoonow 3:38 no menu? :38 “concave
6382 .222 636888 :8 Ed 32 28 86 SE Home. 88 ea 88 a 638260 88o a gem «366689.54

nonmfipaom 05 Bob Am: won: “.053 “mafia—noon: mu 3 m Boom was 3395 voom .8.“ mnouaflbmmu zonoswoum A .mwm

b- 38 0o? wv Emmoawoom

8 8 mm . , 8 me 3 mm . om

 

   
     

 

8 mm Om mw ow mm
b—
vv Vesvv\\\‘\‘\.\\ o o
:“ieeeew:
“Scenery .. e
\\ \\\\\\
\\\\\\\\\ .
\\\\\\\\\
\\\\\\\\\
\\.\\ \\\\
\ \\\\\\\ ..v v
m “mm““mmnxmmmmeo.
“was“ . _
\.\\\\ i
\\\\ o o
a“:
“m““ 61385
““8“ S e
:m
.e e, . 61885.
a“
m“ .2 . 2
“e
\ \
m“
. m -e 6.. 882.4 a
\K Y
nil
d._§\\\n -2 . z
. 3888

Emu ooom

81m )0 Aouenbeid

162

9m

 

11 989.0

m6

06

md

0.”

Wm

9. me. my m 86m

ON

9.

NF

3.

m:

9.

ON

o6

 

m6

m1 mmmmww

.QI mach—24‘

of ad

_

n3. omen—24¢ A

9m

md

.383 2

ON

     

3

NF

3.

9.

3

ON

ma

81mm Aouenbeig

163

.mnoamoanou

m mo owns? 65 can 835» .Amoom E 850% “on we? mmwofiwv flfigmob Enact—o .893 mambonom 332mm mo mos?» SEE
68662 $65. .865 £8888 :8 So 36. an 83 fie see: 38 a 63828 85a a 52» 5298372

couflzmoq ofi 89a Gina won: 333 “afloEooB mm E Eon cumin ween .8.“ mnousfibmmo >325on .N .wmm.

g .266 ease 8%

Nu . oN a; m... V... N4 oé

 

“a... woooz<

S

 

‘cocovmo

.\\\\\\\\\\\\\\\\\

 

NP

3.

91m )0 Aouanbeid

or

O.———>

 

.— moooz<

  

or

61'
64$

Emu moow

on.

164

1G19833 HP

 

 

 

E 0- a”
(u ...I I
0 § . L\\\
g \I . .m
8 § , \\\\\\\\
n\\\\\\\\\\\
t\“\\\‘\‘\““‘
.\\\\\\\\\\\\\\w
.\\\\\\\\\\\\\\\\\\i
l\\‘\\\\‘\\\\\\\\\\‘\‘\‘\\\\‘\“\‘\“‘\““
L‘\\\‘\\\\\\\\‘\\\\\\\\\\\\\\‘\‘\\\‘\\‘\“\“\‘\\‘
.\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\v
s\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\‘
t\\\\\\\\\\\\\\\\\\\\\\‘
\\\\\\\\
. h\“\\\‘
L\\\‘\\\
m.
b
I I I I I I I r
V N 0 w (D V N C
1- 1- ‘-
3
(B
'0
IO
8 CL
(0
m 8
to
2 5
o n\\
—' \\\\\\\\\\\\\\\‘
———.
n\\\\\\\\\\
u\\\\\\\\\\\\\\\\\\\\\\‘
s\\\\\\\\\\\\\\\\\\‘
.\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
n\\\\\\\\\\\\\\\\\\\\\\‘
A‘\\\\\‘\“\\\\“‘\‘\\“\\“‘\\‘\‘\“\““‘
.\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\~
0- L\\\\\\\\\\\\\\‘
_I ——. s\\\\\\\\\\\\\\\\\\‘
8 u\\\\\\\\\\\\\\\\\\\\\\\\\‘
(D
D n\\\\\\\
. E L\‘\
n\\\
I I I I I ' I ' I ' 1
V N O Q ‘0 V N O
F 1-

mm )0 Aouanbeid

'165

8

In
N

o
N

«‘3

8

'75

70

ID
0

8

B

8

9

3

Seed Fe (mg-kg‘l)

Fig. 3. Frequency distributions for seed Fe and Zn in 75 recombinant inbred lines (RILS) from the population

AND696/G19833 grown in Darien Colombia in 2000 and 2005 under high (HP) and 10w (LP) soil phosphorus levels. Arrows
represent mean values of parental genotypes under different treatments (G19833 was not grown in 2005 and AND696 was not
grown in 2000) Values are the average of 3 rephcations Graphs on the left are for 2005 and those on the right are for 2000

Fig. 3 (cont’d).

 

G19833 HP
28

k\\\\\\\\ b

k\\\\\\\\\\\\\\\\\V
—-—>

k\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

26

N\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

24

k\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\‘
;\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\V

22

k\\\\\\\\\\\\\\V
k\\\\\\V

‘20

.\\‘

18

 

16

32

30

28

    
 

AN0696 LP
24

     
 

1

 

G19833 HP 1
18

AND696 HP I

   

16'

20
18
4
2
0

CD? N000
P

91m )0 Aouanbeig

166

‘ Seed Zn (mg-kg'l)

 

.88 Ba 88 .48» Ba 66 m as 28 SE m 8.3 .6688: E 8882 865 on 4.8 .AEeeeeo Emcee 88 ea .28
88 65 a 28 23% 688 .88888 6N es... .8...“ .m $8 é 8882 H8 55 one emfié 23666872 .e 883

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3:5 to:
5.0.20 . 32
F.oo<<o :8
mote. 4.8
We.“ 351112.“. mooeow/D\e.8
m M”. 825 3.6. I v8.6.4 \1/ v.8
em _
meg/[\ome
m oomzm\I/m.mv
m
h 82m ..8
m .. m 5
$25 at . . .
m mm m m m mm mwm m... 2+.“ m"
. .l ll‘l
mW P M m 7». m m Hm M 35.91113
. P mfl m F 3<<o<UI\e..
82% co m H 22% Glad
F 7

Non . won

167

um :20

mfim

 

9mm

 

 

 

o.o

 

mm...

mm
W‘s ms
mos
.mwmm
.mmu.:

mom

mo<<0< / N.Nm

 

 

 

 

 

 

  

<8m.mx< one
MOF<<O / \1 QM?
88>“. 1 9%
6 Em \1 / mmm
38.82 . no»
5320 new
85.00 to
8050 to
m 325 m.»
mu. m. .8633 8:5
6.. ., . «58.8: o.»
82n>d ~.~
8<to 3
805.0 o o
com

.6883 e 88E

168

 

 

 

 

 

 

 

 

 

 

 

 

 

322m ..mm
_ . «885/ 9%
..w. 8050 we. m <om%wwwn /m mm“
MW 22 m.:. m .3P<0o\ /m.3.
M F 8 P00 n.3, 8425 2m
8050 \II 0.3
8E6 5m
8<<0< 5.2 5 .
80P0< 0.: . m Sotw Mm"
8055. o. 3 Es. F m. 85.0
. Bani VIAomr 5 m 35.6 m.»
5 80.20 m m w H. 8880 mg
m 8500 1 1 mg . m P 20% to
m 55.00 S . . . «.835 P.
w 5<<00 mm 3 Em. o o A 8500 me.
. D. .
P 8.200 0 o .820 m.»
<83. ed
< . 8m . 8m

.9283 #4 DEE

169

APPENDIX

170

 

Table A2-1. Mean response for root dry weight, root length, and average root
diameter in the top soil layer of different graft combinations of common bean
lines FR266 and Montcalm grown in containers with a layer of compacted soil in
the presence (+ Fus) or absence (-Fus) of F usarium solani f.sp phaseoli inoculum,
Data analyzed across 1 run (Nov. 2005) of experiment 2.

 

 

 

 

 

 

Top Soil Layer Root dry Root length Average root
. weight (mg) (cm) diameter
(mm)
Graft combination

Root Shoot -Fus +Fus -Fus +Fus -Fus + Fus

FR266 ungrafted 155 158 1088 676 0.41 0.43

FR266 FR266 84 89 765 461 0.40 0.44

FR266 Montcalm 81 84 660 326 0.41 0.46

Montcalm ungrafted 107 99 1050 525 0.37 0.42

Montcalm Montcahn 84 78 989 436 0.39 0.46

Montcalm FR266 88 50 965 449 0.39 0.39
LSD 0.051‘ 17 182 0.02

AN OVA

Source df p value

Replication 3 0.4575 0.1083 0.9271

Graft (G) 5 <0.0001 0.1899 0.0057

Fusarium (F) 1 0.4369 <0.0001 <0.0001 '

G*F 5 0.6759 0.9385 0.1356

 

T LSD value to compare any values within root dry weight, root length, and
average root diameter variables: 9

171

Table A2-2. Mean response for root dry weight, root length, and average root
diameter in the middle soil layer of different graft combinations of common bean
lines FR266 and Montcalm grown in containers with a layer of compacted soil in
the presence (+ Fus) or absence (-Fus) of F usarium solani f.sp phaseoli inoculum,
Data analyzed across 1 run (Nov. 2005) of experiment 2.

 

 

 

 

 

 

Middle (compacted) Soil Root dry Root length Average root
Layer weight (mg) (cm) diameter
(mm)
Graft combination

Root Shoot -Fus +Fus -Fus +Fus -Fus + Fus

FR266 ungrafted 20 24 247 215 0.38 0.37

FR266 FR266 4 7 123 84 0.33 0.37

FR266 Montcahn 8 9 133 94 0.39 0.37

Montcalm ungrafted 22 15 261 180 0.37 0.37

Montcalm Montcalm 7 5 146 40 0.38 0.37

Montcalm FR266 4 7 100 1 19 0.45 0.36
LSD 0.051 3 54 0.02

ANOVA

Source df p value

Replication 3 0.9691 0.7316 0.1015

Graft (G) - 5 <0.0001 0.0089 0.1682

Fusarium (F) 1 0.9411 0.0947 0.2063

G*F 5 0.2655 0.8398 0.0487

 

1' LSD value tocompare any values within root dry weight, root length, and
average root diameter variables.

172

Table A2-3. Mean response for root dry weight, root length, and average mm

diameter in the bottom soil layer of different graft combinations of common bean
lines FR266 and Montcalm grown in containers with a layer of compacted soil in
the presence (+ Fus) or absence (-Fus) of F usarium solani f.sp phaseoli inoculum,

Data analyzed across 1 run (Nov. 2005) of experiment 2.

 

 

 

 

 

Bottom Soil Layer Root dry Root length Average root
weight (mg) (cm) diameter
(mm)
Graft combination

Root Shoot -Fus +Fus -Fus +Fus -Fus + Fus

FR266 ungrafted 175 216 3698 4322 0.36 0.40

FR266 FR266 41 57 1306 1204 0.35 0.36

FR266 Montcahn 69 95 1700 1667 0.35 0.36

Montcalm ungrafted 140 141 3639 3082 0.36 0.35

Montcalm Montcahn 38 28 1026 667 0.33 0.37

Montcalm FR266 36 76 1554 1568 0.34 0.34
LSD 0.05T 31 347 0.01

ANOVA

Source df p value

Replication 3 0.9101 0.6953 0.5281

Graft (G) . 5 <0.0001 <0.0001 0.0307

Fusarium (F) 1 0.2533 0.5803 0.0633

G*F 5 0.8542 0.3902 0.1465

 

T LSD value to compare any values within root dry weight, root length, and

average root diameter variables.

173

Table A3-1. Bartlett test for variance homogeneity based on growth habit, where
there were two classes, determinate and indeterminate, and 1 degree of freedom.
Tests were conducted for shoot and seed traits measured under low P (LP) and
high P (HP) treatments in the recombinant inbred line population developed from
a cross between common bean lines AND696 and G19833.

 

 

Traits P Chi-square Pr>Chi-
level Square
P uptake HP 1.06 0.3032
LP 0.49 0.4851
P use efficiency HP 0.01 0.9206
LP 0.36 0.5468
Days to HP 10.35 0.0013*
maturity 2005 LP 0.03 0.8631 '
Days to HP 1.58 0.2081
maturity 2000 LP 0.06 0.8094
Seed P content HP 9.56 0.0020*

' 2005 LP 0.03 0.8561
Seed P content HP 4.98 0.0257*
2000 LP 2.76 0.0967
Seed yield HP 1.13 0.2888
2005 LP 1.07 0.3015
Seed yield HP 0.76 0.3847

2000 LP 6.07 0.0137*

* Variance is not homogeneous between grth habit classes

174

Table A3-2. Bartlett test for variance homogeneity based on growth habit, where
there were two classes, determinate and indeterminate, and 1 degree of freedom.
Tests were conducted for roots traits measured under low P (LP) and high P (HP)
treatments in the recombinant inbred line population developed from a cross
between common bean lines AND696 and G19833.

 

 

 

Traits P Chi-square Pr>Chi-
Ievel Sguare
Adventitious HP 0. 12 0.7266
roots LP 2.37 0.1234
Root length HP 8.62 0.0033*
density LP 1.86 0.1723
Specific root HP 16.31 <0.0001*
length LP 0.63 0.4259
Root surface HP 7.98 0.0047*
area LP 1.37 0.2418
Average root HP 0.21 0.6431
diameter LP 0.1 1 0.735 8

* Variance is not homogeneous between growth habit classes

175

 

of md

905E902. £0 .
0005:0200 I

n_I

 

9m

0000002000 m 00 0w000>0 05 000 00.002 00080000 3000 03:3 32 00058000003
90 0000300000 m0 0030., G008 00000500 $50.90. .030: £30% .3 000000000 0&0 000 a 300& 0000 0.355 .2.—mt 05

00 300% .3 Am 0:0 “0.— 08 :0 300% .3 030000500 .3 at? 30.0— m :00 .3 000000000 000 mucusabmfi 3000005 mm 00::
009350.30. 30:0 Ewu 05 00 300% 00 00.0 $2 0.5 00 300% 06 :0 E9000 3 .5 3 32 .00 00003506 25. .200 030000030
80 32 08 EB :03 a 88 a 038250 .8000 a 8.53 0820 05 000072. 80303 so... a .80 802950 300
00:: 0033 0003080000 mm 00 350$». 00: m 20 .0030: m 000000000000 m 000% m0 00003506 mofiwcohm .72 000me

 

A733 .0000 A 000.5
00 0.0

. . o
- 0
u v
- 0
u 0
I 2
1 NF

 

I3.

o.v md

905E000... § .

905E050 U
n...

fioi

 

 

 

    

.\\\\\\\\\\\‘

.\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

 

 
     

    

- \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

  

 

.\\\\\‘ N

  

\\\\\\\\\\\\\\\\\‘

 

 

Aouanbeid

176

Figure A3-l (cont’d).

 

 

 

 

 
   
   
   
 

 

 

 

   

 

 

 

 

‘ .9
3 m
m CE
5 E
0- E
O
I m 3
H
o '0
D E
is?
—.
0
~ co
. a
.. o
,_
l f I ' I ' I ' l ' l 1 I ' I
<0 0 N o co <0 0 N o
1- !- 1- ‘—
0 [ o
.9 T6 "
m .5
.E g -
a g-
S
\‘ .
D §
3 - s
=
—
.\\\\\\\\\\\\\\\ 8
.\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\‘
—
_. .
L .\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
.\\\\\\\\\\\\\\\\\\\\\\\\\\\\\V
L
o
‘ ‘-
l'l"l'l"l'f
to v N o a co v N o -
v- 1- 1- .

Aouanbaig

177

 

P uptake (mg-plant'l)

000550.005 ..
0.055.900 I

A...

 

 

«.0

Nd

to

 

Awn— wfi . 0000. My. bun—2050 00: may.

 

 

 

 

 

 

 

 

 

19.
T.
INF
r

. h 0.0 . . .
. -N

. .4

7. .l@

a. a ‘ ...

IS.

low

V

0005:5000... § _ .. 2

0005:5000 U . u on

n: . .-NN

 

50.303 Tm< 0.5mm

Muenbeid

178

 

. 00000000000 m .00 0w000>0 000 000 00002 8080.00 0000 00000.3 05

0000080000008 0000 00000800000 .00 002000 0008 000000000 0300.00. .0300 00.50% .3 000000000 0000 000 . 01mm 000% 0000 000003
.0030 000 00 00000w .3 0: 000 0.000 000 00 00000».~ .3 0080000000 00 0003 0.60. 0 :00 .3 000000000 000 0000008000000000000
.000 0000: 0000000000000 3000 Bat 05 00 00000w 00 000 .000 000 00. 0.000% 000 00 03000 00 0A 8 01:00 .00 0000300000

200 .08 08808.00 Ed .32 05 EB 0000 a 080 0a 080 a 038200 .5000 a 80003 00020 08 000078

0003000 00000 0 800.0 000000.000 A305 000.: 000000 0000380000 mu 8 0000000 0 0000 .00 000.000.00.00 0000000000 .N-m< 00080

0000 00.08... 000.005 050:8 0 080

 

 

 

 

 

 

 

 

 

 

8N OVN ONN 8N on _. cow 0!. ON w 00 _. omN OVN CNN .OON on P 09. 0.3. ON _. 8P
- b — r 0 . 0 — u _ p _ p n b p .
b o \ o
d 0 c S c .

I N I N

.l V I Q

l 0 I m

l O u I Q
T 2 g u 2
a m. r 00
I VF I 3.

v . 055500000:— \\\\\\\x .
0005000000 I . .

n... r / - 2

 

 

n_I :00

Aouanbaig

179

00050000000. 0%,
2055.200 I

n:...

 

 

 

080 0.08... 02.000 050:8 0 080

 

 

CNN OON . our 09
_

03.
—

 

- D D b L — L

00050000000. §
00.050.00.00 B

n...

 

§
8

 

 

R\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\V

 

——>
' I
O
P

 

 

L _ -2

 

 

 

Aouanban

, .5080 0.00. 200E .

180

00000000000 m .00 0w000>0 000 000 00002 0000000000 0000 000003 05

0000000000000 0000 00000000000 .00 00000., 00000 000000000 0300.0. .0000 00.50% .3 000000000 0000 00 01mm 0000» 0000 00000.5

.00w00 000 00 00000w .3 mm 000 000 000 00 00000w .3 00000000000 00 003 00.000 0 :00 .3 000000000 00 0000000000000 000000000

.00.“ 0000: 0000000000000 3000 Emu 000 00 300% 00 000 02 000 00 00000» 00.0 00 03000 00 A: 00 01mm .00 0000000000000

000. .08 30000800 00.0 32 08 E00 0000 a 0000 000 0000 00 038200 .8000 a 08020 000000 0.0. 000072

000.500 00000 0 000.0 000000.000 0.00% 00.00 000000 00000000000000 mm 00 200% 0000 .00 0000000000 000000000 .062 000mm
0000 0.-0... 00 0.000 080 w

OOON OOO_. O09. OOE. OONw 82 com OOO OO?
..P._._.....

     

 

0000 0000 0000 003, 0000 0000 000 000 000

  

 

  

 

      

 

 

 

 

 

 

 

. _ q... M Io 0. \vav IO
w. m m m .0. _... m 0 .0... = m m m m m
, ... L ... .. T. .. . \ \ \ \ \
. ..,, .0 .., I “ “ “ “ “
0 u 0 0 0 0 0
. .0 u . 0 0 0 0. m u
g 0 N 0 “ 0 0 N
t. ..u z ; m m m m
.e m . E . M
_. 0 0 0
0 ; w m 0
s ..v m x. 0 I0.
0... . _ 0
. N
0
m
. I0 \ l0
0
g 0
IQ In
00050000000. a :00 00055000000. § . u 00
0005500000 I T 0005:0900 D
a: 0..

IN—. IN?

' AouenbaJd

181

2500

 

 

 

     
   
   
   

 

 

 

 

 

 

 

 

 

 

 

0
H
3 m
2 IE
0 O
o E 8
I 5%.
——>
—-—-—> g
E
I ' I ' I ' I ' I 1 I ' I
N O G) (D V N O
I- I-
O
P 8
N
0
H D
.03 to
g .E
o. E 9 - §
.J .9 d) N
0 1:
D 5
§ .
U § r
0
Q
“G .\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ g I
H l—
: .\\\\\\\\\\\\\\\\\\\\’
8 .\\\\\\\\\\\\\\\\\\\\\\\\\\\
V .\\\\\\\\\\\\\\\\\\\\'
m —-’
2' ' §
L
D a
.3 ' T I I I I I I I ' I fl
, I14 53 o co <0 0 N o

Aouanber

182

Seed yield (kg 113'“) 2000

 

 

Table A4-1. Amplified fragment length polymorphisms (AF LP) names and
selective primers used in developing a linkage map of AND696/G19833.

 

 

AFLP name Selective Selective Polymorphism
Jrimer 1T primer 21 number.
ACAA E-AAC M-CAA 5
ACAC E-AAC M-CAC 4
ACAT E-AAC M-CAT 3
ACTA E-AAC M-CTA 2
ACTC E-AAC M-CT C 6
ACTG E—AAC M-CTG 3
ACTT E-AAC M-CTT 2
AGAC E-AAG M-CAC 1
AGAT E-AAG M-CAT 6
AGTA E-AAG M-CTA 2
AGTC E-AAG M-CTT 3
CAAA E-ACA M-CAA 3
CAAC E-ACA M-CAC 5
CAAG E-ACA M-CAG 1
CAAT E-ACA M-CAT 5
CATC E-ACA M-CTC 2
CATT E-ACA M-CTT 3
CCAA' E-ACC ' M-CAA 3
CCAC E-ACC M-CAC 1
CCTA E-ACC M-CTA 2
CCTT E-ACC M-CTT 1
‘ CGAA E-ACG M-CAA 3
CGAC E-ACG M-CAC l
CGTA E-ACG M-CTA 3
CGTC E-ACG M-CTC 5
CGTT E-ACG M-CTT 1
CTAA E-ACT M-CAA 4
CTAC E-ACT M-CAC 1
CTTA E-ACT M-CTA 6
CTTC E-ACT M-CTC 1
CTTT E-ACT M-CTT l
GCAC B-AGC M-CAC 1
GCAT E-AGC M-CAT 5
GCTC E-AGC M-CTC 4
GGAA E-AGG M-CAA 3
GGAG E-AGG M—CAG 2
GGAT E-AGG M-CAT 3
GGTA E-AGG M4CTA 3

 

TEcoRl adapter (E) ligated to 3 selective nucleotides (i.e. AAC).
* Msel adapter (M) ligated to 3 selective nucleotides (i.e. CAA).

'in AND696/619833 population with selective primer combination 1 and 2.

183

Table A4-2. Simple sequence repeat (SSR) primer sequence, linkage group and
source information for markers polymorphic between AND696 and G19833 and
used to develop a linkage map of AND696/G19833.

 

 

SSR PrimerT Primer Sequence Linkage Source*
group;
AGl F CAT GCA GAG GAA GCA B03 2
GAG TG
R GAG CGT CGT CGT TTC
GAT
ATA] 3 ‘F unpublished B03 5
R
ATA4 F unpublished B01 5
R
ATA9 F unpublished B09 5
R
BM114 F AGC CTG GTG AAA TGC B09 2
TCA TAG (unlinked)
R CAT GCT TGT TGC CTA
ACT CTC T
BM137 F CCG TAT CCG AGC ACC B06 2
GTA AC '
R CGC TTA CTC ACT GTA
CGC ACG
BM14O F TGC ACA ACA CAC ATT B04 2
TAG TGA C
R CCT ACC AAG ATT GAT
TTA T GG G
BM143 F GGG AAA TGA ACA GAG B02 2
GAA A
R ATG TTG GGA ACT TTT
AGT GTG
BM154 F TCT TGC GAC CGA GCT B09 2
TCT CC
R CTG AAT CTG GGA ACG
ATG ACC AG
BM156 F CTT GTT CCA CCT CCC B02 2
ATC ATA GC
R TGC TTG CAT CTC AGC
CAG AAT C
BM159 F GGT GCT GTT GCT GCT B03 2
GTT AT
R GGG AGA TGT GGT AAG

ATA ATG AAA

184

 

Table A4-2 (cont’d).

 

SSR

BM160
BM161
BM164
BM165
BM167
BM170
BMZOO
BM211
BM53

BMd-l

PrimerT Primer Sequence

F

R

CGT GCT TGG CGA ATA
GCT TTG

CGC GGT TCT GAT CGT
GAC TTC

TGC AAA GGG TTG AAA
GTT GAG AG

TTC CAA TGC ACC AGA
CAT TCC

CCA CCA CAA GGA GAA
GCA AC

ACC ATT CAG GCC GAT
ACT CC

TCA AAT CCC ACA CAT
GAT CG

TTC TTT CAT TCA TAT
TAT TCC GTT CA

TCC TCA ATA CTA CAT
CGT GTG ACC

CCT GGT GTA ACC CTC
GTA ACA G

AGC CAG GTG CAA GAC
CTT AG

AGA TAG GGA GCT GGT
GGT AGC

TGG TGG TTG TTA TGG
GAG AAG

ATT TGT CTC TGT CTA
TTC CCA C

ATA CCC ACA TGC ACA
AGT TTG G

CCA CCA TGT GCT CAT
GAA GAT

AAC TAA CCT CAT ACG
ACA TGA AA

AAT GCT TGC ACT AGG
GAG TT

CAA ATC GCA ACA CCT
CAC AA

GTC GGA GCC ATC ATC
TGT TT

185

Linkage

group;
BO7

B04
B02
unlinked

(1301)

B02 .
(unlinked)
B06

B01

B08

B01

B03
(unlinked)

Source*

2

 

Table A4-2 (cont’d).

 

 

SSR Primer’r Primer Sequence Linkage Source*
group:
BMd-lO F GCT CAC GTA CGA B01 3
GTT GAA TCT CAG
R ATC TGA GAG CAG
CGA CAT GGT AG
BMd-3 F TGT TTC TTC CTT ATG unlinked 3
GTT AGG TTG (BO7)
R GTA TCC TCC GAT
CAA ATT CAC CT
BMd—42 F TCA TAG AAG ATT B10 3
TGT GGA AGC A
R TGA GAC ACG TAC
GAG GCT GTA T
Gats9l F GAG TGC GGA AGC B02 2
GAG TAG AG
R TCC GTG TTC CTC TGT
CTG TG
MEI F unpublished B09 5
R
PVct002 F TTA GAC TTT CAA B08 4
ACA TTC AC
R GAT ACT ACT TAA
ATG AGG AAC A
PVttc002 F ATA TCT TAC AGC CAT B08 4
TAC ATT C
R CTC ATC ACC CAG
TCA CCT _
PVcttOOl F CCA ACC ACA TTC 'ITC B04 1
CCT ACG TC
R GCG AGG CAG TTA
TCT TTA GGA GTG
PVagOOl F CAA TCC TCT CTC TCT B11 1
CAT TTC CAA TC
R GAC CTT GAA GTC
GGT GTC GTT T
PVatOO3 F ACC TAG AGC CTA B04 1
ATC CTT CTG CGT
R .
PVgaatOOl F AAG GAT GGG TTC B04 1
CGT GCT TG
R CAC GGT ACA CGA

AAC CAT GCT ATC

186

 

Table A4—2 (cont’d).

 

 

SSR PrimerT Primer Sequence Linkage Source*
MP1
PVat007 F AGT TAA ATT ATA CGA B09 1
GGT TAG CCT AAA TC
R CAT TCC CTT CAC ACA
TTC ACC G

 

1' Primer orientation: forward (F); reverse (R)
iBean linkage group location of SSR in the consensus map (Freyre et al., 1998),
if different than where it mapped to in AND696/G19833 population, that location
is noted in parenthesis.
* 1 :Yu et al., 2000. Journal of Heredity 91(6):429-434.
2 : Gaitan—Solis etal., 2002. Crop Science 42:2128-2136.
3 : Blair et al. 2003. Theoretical Applied Genetics 107:1362-1374.
4 : Caixeta et al., 2005. Crop Breeding and Applied Biotechnology 5:125-133.
5: Metais et al., 2002. Theoretical Applied Genetics 104:1346-1352.
(full citations available in Chapter 4, References)

187

 

Table A4-3. Random Amplified Polymorphism DNA (RAPD) primer sequence
for markers polymorphic between AND696 and G19833 and used to develop a
linkage map of AND696/G1 9833 recombinant inbred line population.

 

 

RAPD name Primer sequence Polymorphism
numberT
AA-19 5 ’-TGAGGCGTGT-3 ’ 1
AK-6 5’-TCACGTCCCT-3’ 2
H-l2 5’-ACGCGCATGT-3’ 2
L-4 5’-GACTGCACAC-3’ I
I-6 5’-AAGGCGGCAG-3’ 1
M-2 S’-ACAACGCCTC-3’ 1
N—12 S’-CACAGACACC-3’ 1
R-4 5’-CCCGTAGCAC-3’ 1
0-12 5’-CAGTGCTGTG-3’ 2

 

lNumber of polymorphisms with this primer in the AND696/G19833
recombinant inbred line population.

188

 

Table A4-4. Bartlett test for variance homogeneity based on growth habit, where
there were two classes, determinate and indeterminate, and 1 degree of freedom.
Tests were conducted for seed traits measured under low P (LP) and high P (HP)
treatments in the recombinant inbred line population developed from a cross
between common bean lines AND696 and G19833.

 

 

 

 

 

Traits P Chi-square Pr>Chi-
level Square
2005
Seed weight HP 13.25 0.0003*
LP 1.02 0.3135
Seed P HP 2.66 0.1029
LP 1.17 0.2799
Seed Fe HP 1.45 0.2290
LP 1.51 0.2190
Seed Zn HP 1.47 0.2261
LP 12.60 0.0004*
Seed phytic HP 0.55 0.4598
acid LP 22.95 <0.0001*
2000
Seed weight HP 1.57 0.2095
LP 0.07 0.7903
Seed P HP 6.44 0.0112*
LP 0.31 0.5781
Seed Fe HP 0.06 0.8004
LP 0.88 0.3484
Seed Zn HP 0.96 0.3260
LP 1.00 0.3173

 

* Variance is not homogeneous between grth habit classes

189

 

   

\Ilflllllllgflllllllllfllfll111111))!