«new. .u ‘ . . .
fiéflufi:

x

“1.3:?
J II‘VDJ}1

w fig. 1...,
13.nmmqn2rdwmwv ‘

.
:. M 1.th»...9.§
.. a.

.mu! XV?
aha?
y. . .a
, n. u 2...! u... ,
d)
.
3 M . rs...
. . ‘ . it... 2 I. .
JQ¢F.$H~ J. .9
k};

 

:4.
a.

a...

m. s

 

 

 

 

E...
.31:

, .tiy
[2.4.2.1 .. a. A
,5 1
1):: 3.;
1

 

 

 

.1 f:
I 5113:.

.
1 .. I: s

4% mafia

 

9;)

.‘\
-J
k u

it

3mm

LIBRARIES
GAN STATE UNIVERSITY
EIXSTHLIANSING, MICH 48824-1048

This is to certify that the
dissertation entitled

GENETICS OF APHANOMYCES DISEASE RESISTANCE IN
SUGARBEET (BETA VULGARIS), AFLP MAPPING AND QTL
ANALYSES

presented by
YI YU

has been accepted towards fuifillment
of the requirements for the

Ph.D. degree in Plant Breedingand Genetics

 

/,/76/Z/(/57%

Major Professor’s Signature

 

”7545/ 2- 3:. 2M5
I

Date

MSU is an Affirmative Action/Equal Opportunity Institution

- T..- -._.-—-—.—..‘--

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.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/ClRC/DateDue.p65—p.15

GENETICS OF APHANOMYCES DISEASE RESISTANCE IN
SUGARBEET (BETA VULGARIS), AFLP MAPPING AND QTL ANALYSES

By

Yi Yu

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

2004

ABSTRACT

GENETICS OF APHANOMYCES DISEASE RESISTANCE IN SUGARBEET

(BETA VULGARIS), AFLP MAPPING AND QTL ANALYSES

By

Yi Yu

Aphanomyces cochlioides causes darnping-off and root-rot diseases in sugarbeet. As a
seedling disease, Aphanomyces severely reduces stand establishment in the fields.
Since there is no effective chemical control, resistance breeding has been considered
the best approach to overcome this problem. All domestic varieties of sugarbeet are
more or less susceptible to this disease, and the genetic basis of reduced susceptibility
is poorly understood.

Evaluation of Aphanomyces resistance has been problematic for resistance
breeding, since no reliable disease screening method has been available. Five different
inoculation methods were attempted in this research, and a novel disease evaluation
system known as the “box inoculation” was developed, which could effectively reduce
experimental error. Active zoospores were found to play a major role in the infection
process. By using active zoospores as the inoculum, resistant and susceptible varieties
could be reliably discriminated. In addition, submersion of seedlings in water
facilitated resistance evaluation by creating anoxic conditions, which are often

associated with severe symptoms in the field. Twenty sugarbeet accessions were

tested for disease resistance. None was immune, but obvious differences in relative
susceptibility were observed. Lab test results were shown to be consistent with field
data, implying the feasibility to use lab test results to predict relative field
performance.

Crosses were made between Aphanomyces susceptible genotype C869 (sugarbeet)
and resistant wild beets (Beta vulgaris ssp. maritima). F1 hybrid plants were
challenged with zoospores of Aphanomyces. Survival percentages demonstrated
different susceptibility among different individuals within the same cross. A single
resistant F 1 hybrid was self pollinated to produce a segregating F2 population. From
145 F2 plants, an AFLP (amplification fragment length polymorphism) genetic linkage
map was constructed, which contained 163 markers on 9 linkage groups. The total
map length was 507.1 CM, averaging 3.1 cM for marker intervals.

Since resistance evaluation is destructive, Aphanomyces resistance of each F2
plant was estimated by the average performance of self-pollinated F 3 families in a
progeny test. Composite interval mapping analysis of Aphanomyces disease resistance
identified one major QTL (quantitative trait locus) on linkage group 9 (p < 0.01), and
one minor QTL on linkage group 2 (p < 0.05). The model combining both QTLs
explained 63.2% variation of relative AUDPC (Area Under Disease Progress Curve,
rAUDPC). Resistance introduced from wild beet (Beta vulgaris ssp maritima,
P1540625) was heritable, and the broad sense heritability of rAUDPCs was estimated

to be 40.1%.

To all scientists fighting for the surviving of human being

iv

ACKNOWLEDGMENTS

I sincerely appreciate my major advisor, J. Mitchell McGrath for his insightfulness
and instructions on research, for his encouragement when l was frustrated by the
experiments, for his cultivation of my independence on thinking, observing,
exploring and concluding, for financial support and great help on scholarship
application, especially for his warrnhearted help on my English improvement. His
working enthusiasm and responsiveness will inspire me all my life. Because of
his continuous support and patience, I was finally able to complete the tough

project I chose, Aphanomyces disease resistance.

I also appreciate my dissertation committee members, James D Kelly,
Rebecca Grumet, and Ray Hammerschmidt for valuable discussion and

suggestions, for their patience and careful reading of this manuscript.

David Johnson provided Aphanomyces isolates; he also generously helped
on Aphanomyces zoospore preparation. His deep knowledge about
Aphanomyces cochlioides is greatly appreciated. Without his suggestions, I may

have had to give up this project.

I thank Cathy Derrico and Peter Hudy for instructions on some lab
techniques, Xiaofeng Wang and Huangying Qin for so many years’ experience

sharing and encouragement.

I also thank David Douches, George Hosfield, Mike Thomashow, Suleiman

Bughrara, Joe Saunders, and John M Halloin for using their equipments,

Dechun Wang for personal help.

Ken Sink, Doug Buhler and Darlene Johnson provided administrative help.
Tim Duckert provided field assistance. Susan helped with lab work. Benildo,
Daniel and Suba shared experimental skills and criticized the first manuscript

with terrific comments. I am thankful to all of them.

The Plant Breeding and Genetics Graduate Program at Michigan State
University provided fellowship support. The Graduate School provided the

Dissertation Completion Fellowship.

Thanks to Cong and Hao, Yuzhi and Dongsheng, Yinhui and Kai, Qiong and
Weiqing, Guoliang, Hua, Baolin, Ji, Qin, Yonggang, Xiaoming and others, who
made my life in East Lansing more comfortable, and to many unsung heroes who

helped to make the completion of my PhD degree possible.

Finally, I thank my parents for their love and support, and my wife Mingyao

Li for help on manuscript and sharing of my burdens.

vi

TABLE OF CONTENTS

List of tables .................................................................................... xi
List of figures .................................................................................... xii
Key to abbreviations .......................................................................... xiii
Overview .......................................................................................... 1
References ....................................................................................... 6
Chapter 1

A novel approach for screening sugarbeet (Beta vulgan's) seedlings for reaction
to seedling disease caused by Aphanomyces cochlioides

Abstract ................................................................................... 7
Introduction .............................................................................. 9
Materials and Methods ............................................................... 12
Biological materials ............................................................... 12
Zoospore preparation ............................................................ 13
Plant preparation and Inoculation ............................................. 13
Field tests ........................................................................... 1 5
Results .................................................................................... 16
Discussion ............................................................................... 22
References .............................................................................. 26

vii

Chapter 2

Development of a genetic map to detect genes involved with Aphanomyces
resistance and some agronomic traits

Abstract .................................................................................... 29
Introduction ............................................................................... 3O
Amplified Fragment Length Polymorphism (AF LP) ......................... 30
Genetic map of sugarbeet ......................................................... 32
Agronomic traits ...................................................................... 32
Monogerm .............................................................................. 32
Germination time ...................................................................... 33
Genetic male sterility (GMS) ...................................................... 34
Materials and Methods ................................................................... 35
Mapping population ................................................................. 35
AFLP ................................................................................... 35
Data collection and analyses ..................................................... 37
Agronomic trait analyses .......................................................... 38
Results ....................................................................................... 39
Genetic map of sugarbeet by wild beet ....................................... 39
Pstl and EcoRI comparison in AFLP ........................................... 39
Distorted segregation of markers ................................................ 42

Accessions’ phylogenetic relationship on Aphanomyces disease
resistance ........................................................................ 43

viii

Monogenn/Multigerm trait analysis ................................................ 44

Germination time trait analysis .................................................... 47
Genetic Male sterility ............................................................... 48
Discussion .................................................................................. 50
SAMPL markers ...................................................................... 50
Genetic map construction .......................................................... 52
Phylogenetic relationship of the accessions ................................... 55
References ................................................................................. 57
Chapter 3

Genetic analysis of Aphanomyces resistance in a sugarbeet x wild beet cross

Abstract ...................................................................................... 61
Introduction ................................................................................. 62
Simple statistical analysis ........................................................... 62
Interval mapping ....................................................................... 63
Composite Interval Mapping (CIM) ................................................ 63
AFLP map based QTL analysis ................................................... 64
Marker assisted selection (MAS) .................................................. 65
Mapping population ................................................................... 66
Aphanomyces disease resistance ................................................ 67
Materials and Methods .................................................................. 69
Results ....................................................................................... 72
Discussion .................................................................................. 77

ix

QTL analyses .......................................................................... 77

Aphanomyces resistance ........................................................... 78

Marker Assisted Selection (MAS) ................................................. 81
References ................................................................................... 84
Summary ........................................................................................... 89
Appendix I AFLP technique .................................................................. 92
Appendix II Primers used in sugar beet AFLP ........................................... 98
Appendix III Additional CIM results for QTL analyses of rAUDPCs ................. 99

Table 2.1

Table 2.2

Table 2.3

Table 2.4

Table 2.5

Table A1

Table A2

Table A3

Table A4

Table A5

Table A6

Table A.7

Table A8

LIST OF TABLES

Marker nomenclature ............................................................ 36
Summary of Pstl, EcoRI and SAMPL primer results ..................... 41
AFLP marker Chi-square test for monogerrny ............................. 46
AFLP marker Chi-square test for germination time ....................... 48
AFLP marker Chi-square test for genetic male-sterility .................. 49
EcoRI or Pstl adapter preparation for AFLP ............................... 92
M39] adapter preparation for AFLP .......................................... 92
Digestion reaction (RL I) for AFLP and SAMPL .......................... 93
Digestion reaction (RL II) for AF LP and SAMPL .......................... 94
Preamplification reaction for AFLP ........................................... 94
Amplification reaction for AFLP ............................................... 95
PCR program for AFLP pre-amplification ................................... 95
PCR program for AFLP amplification ........................................ 96

xi

Figure 1.1

Figure 1.2

Figure 1.3

Figure 2.1

Figure 2.2

Figure 3.1.

Figure 3.2

Figure A.1

Figure A.2

LIST OF FIGURES

Mean percent of Aphanomyces seedling disease reaction scores of
eight sugarbeet germplasm lines in the water box inoculation test

...................................................................................... 18
Area under the disease progress curve (AUDPC) scores for two
sugarbeet germplasm lines inoculated with Aphanomyces
cochlioides in the water box inoculation test ............................ 19

Correlation of Aphanomyces disease reaction for eight germplasm
releases in the water box inoculation test and field evaluation 21

Genetic linkage map of sugarbeet (0869) X wild beet (PI540625)
...................................................................................... 40

Phylogenetic relationship of wild beet, breeding lines, and mapping

population for Aphanomyces resistance ................................. 45
Mapmaker/QTL analyses on rAUDPCs for Aphanomyces resistance
in sugarbeet ..................................................................... 75

CIM results for rAUDPCs in response to Aphanomyces infection in
sugarbeet ......................................................................... 76

Additional CIM results for rAUDPCs in response to Aphanomyces
infection in sugarbeet (1) .................................................... 99

Additional CIM results for rAUDPCs in response to Aphanomyces
infection in sugarbeet (2) ................................................... 100

xii

AFLP:

APS:
bp:
CIM:
cM:

CMS:

dATP:

DNA:

GMS:

kb:
LOD:
MAS:

Mbp:

PAU P:

PCR:
PVP:
QTL:

RIL:

KEY TO ABBREVIATIONS

Amplified Fragment Length Polymorphism
Ammonium Persulfate

Base pair

Composite Interval Mapping

centiMorgan

Cytoplasmic Male Sterility
2'-deoxyAdenosine 5'-TriPhosphate, sodium salt
Deoxyribose Nucleic Acid

Genetic Male Sterility

Kilo-base pairs

Log of the Odds

Marker Assisted Selection

Mega base pairs

Phylogenetic Analysis Using Parsimony
Polymerase Chain Reaction
Polyvinylpyrrolidone
Quantitative Trait Locus

Recombinant Inbred Line

SAMPL: Selective Amplification of Microsatellite Polymorphism Locus

STS:

Sequence Tagged Site

TEMED: N,N,N’,N’-Tetramethylethylene-diamine

xiii

Genetics of Aphanomyces disease resistance in sugarbeet (Beta

vulgaris), AFLP mapping and QTL analyses

OVERVIEW

Sugarbeet (Beta vulgaris ssp. vulgaris) is an important cash crop in the U.S., contributing
over 1.2 billion dollars annually to the US. economy. F orty-seven percent of the US.
sugar consumed is produced from sugarbeets (USDA 1997). Pulp and molasses, by-
products of sugar extraction, are often used as livestock feed. Compared to other major
crops, sugarbeet has had a short cultivation history of about 200 years. Originated from
white fodder beet, sugarbeet is currently grown in more than 40 countries. In the U.S.,
there are five major sugarbeet growing regions: 1) Red River Valley region (Minnesota
and North Dakota), 2) Idaho and Washington, 3) "Intermountain" region (Montana,
Wyoming, Nebraska and Colorado), 4) Western region (California), and 5) Great Lakes

region (Michigan and Ohio).

Sugarbeet is classified in the order C aryophyllale, family Chenopodiaceae, and
genus Beta. Plants in C aryophyllale are chemotaxonomically distinct from other dicots,
because they contain betalain pigments, a unique class of alkaloid pigments, instead of
anthocyanin. Family C henopodiaceae includes B. vulgaris (Swiss chard, red table beet,
fodder beet, etc.), spinach (Spinacia oleracea) and a number of common weeds
[Iambsquarters (Chenopodium album), Russian thistle (Salsola kali), and kochia (Kochia

scoparia)]. Genus Beta comprises 12 species in four sections: 1) Beta, containing

cultivated sugarbeet, and wild beet, B. vulgaris ssp. maritima; H) Corollinae; 1H) Nanae
and IV) Procumbentes (Letschert et al. 1994). The section Beta is comprised fiom three
species: B. vulgaris, B. macrocarpa, and B. patula. Beta vulgaris is further divided into
three subspecies: Beta vulgaris ssp. vulgaris, ssp. maritima, and ssp. adanensis.
Subspecies maritima is also called as wild beet, sea beet or weed beet, and has often been
used in sugarbeet breeding programs as a source of resistance genes (Lewellen and
Whitney 1993; Letschert et al. 1994). Its potential breeding value includes aphid and

BYV (beet yellow virus) resistance (Dale et al. 1985).

Haploids of section Beta have nine chromosomes, and plants have varying ploidy
levels from diploid to pentaploid based on x=9. Most cultivated sugarbeets in the US.
are diploid. In Europe, triploid beets are often used for production, due to its potential to

produce higher sucrose yield.

Wild populations of B. vulgaris consist of annual, biennial, and perennial types.
Annuality is genetically determined by a dominant gene B (Boudry et al. 1994) and linked
to red hypocotyl color (Schumacher et al. 1997). All sugarbeets are biennial. During the
first year of growth, sugarbeets form enlarged roots, which can be harvested for sugar
production. After vemalization either in the field or in cold room, plants will bolt, flower
and set seeds. Sugarbeet has small wind-pollinated flowers borne on spike-like
indeterminant inflorescences, with one single fruit per flower. Sugarbeet flowers do not
have petals, and flowers are perfect, including a tricarpellate pistil surrounded by five
stamens and five sepals. Flowers may cluster on one common receptacle (utricle), and

form a single seed ball with multiple fruits, called a multi germ seed. Most commercial

varieties have monogerm seeds (one fruit per receptacle). Cross-pollination is the major
pollination type in sugarbeet, due to the lack of synchrony between pollen release and
receptiveness of stigma (Cooke and Scott 1993). In addition, most germplasm is self-

incompatible.

The genetic variation of sugarbeet is thought to be limited since it was selected from
white fodder beet. Artificial hybridizations with B. procumbens, B. webbiana, and B.
vulgaris ssp. maritima have been made to improve disease resistance. Public breeding
programs were initiated by USDA in the early 20th century after the devastating outbreak
of curly top disease. However, due to the open pollination habit and frequent
incorporation of different germplasm, the pedigree of different sugarbeet accessions is
difficult to elucidate. Several significant accomplishments in sugarbeet breeding in the
US. include breeding for curly top resistance, Cercospora resistance, partial
Aphanomyces resistance, the discovery and deployment of cytoplasmic male sterility
(CMS, O-type maintainer line) and the introduction of monogerm seed. Curly top
resistance was mainly selected from German germplasm ‘Klein E’ (Lewellen 1992).

C ercospora leaf-spot resistance derives from wild beet, B. vulgaris ssp. maritima, and an
elite progeny of resistance breeding is US. 201 (Lewellen 1992). Resistance to
Aphanomyces was primarily developed in East Lansing, MI, and several moderate
resistant breeding lines, for instance, US-H20, EL48, and SP6822, were derived from

selections made under high disease pressure.

Discovery of CMS plant has greatly facilitated sugarbeet breeding, since it

eliminates the necessity of hand emasculation and makes it possible to generate large

amount of genetically identical hybrid seeds for production. Three-way crosses have

been the dominant in hybrid cultivars in the US.

The use of monogerm sugarbeet seed has greatly reduced the efforts needed to thin
sugarbeet seedlings. Monogerm seed was developed primarily through the work of V. F.
Savitsky. In 1950-52, he developed two monogerm lines, SLC 101 and SLC 107, and
found that the monogerm trait is controlled by a single recessive gene (m).

Sugarbeet is affected by a number of diseases, such as curly top, rhizomania,
Rhizoctonia, Aphanomyces, and Pythium (Cooke and Scott 1993). In the Great Lakes
region, sugarbeets are attacked by several fungal diseases, including leaf spot caused by
Cercospora beticola, root rot caused by Rhizoctonia solani, and black leg caused by
Aphanomyces cochlioides. Cultivars and germplasm lines have been bred with varying
degrees of resistance to each of these diseases. Of these three diseases, least progress has
been made in the understanding of Aphanomyces disease, because of the lack of highly
resistant sources, and the difficulty to create uniform Aphanomyces epidemics for lab or
field screening.

Breeding for resistance to Aphanomyces is important, as it reduces not only
production costs, but also the potential pollution caused by fungicide application.
However, in domestic varieties, the resistance resources are limited. From 1950 to 1955,
Afanasiev (1956) tested several hundred varieties, and found no resistance to
Aphanomyces. Schneider and Gaskill (1961) used greenhouse test to evaluate the
resistance of 217 foreign introduction lines of sugarbeet. None of the introductions were
highly resistant, and most were more susceptible than the moderately resistant line-

US401. Limited progress on selection for partial resistance to Aphanomyces was made by

C. L. Schneider in East Lansing. The developed germplasm exhibiting partial resistance is
related to variety U8216, the probable contributor of Aphanomyces resistance
(Bockstahler et al. 1950). However, the genetic mechanism of the resistance is poorly

understood (Campbell and Bugbee 1993).

This research focused on (1) exploring new approaches for Aphanomyces disease-
resistance evaluation, (2) introducing a new resistance source into domestic varieties, and
(3) determining the genetic basis of Aphanomyces resistance, including mapping the
quantitative trait loci (QTLs) responsible for the resistance. Specifically, an F2
population of sugarbeet by wild beet was constructed and progeny of each F2 individual
was tested for Aphanomyces resistance. Amplified Fragment Length Polymorphism
(AFLP) protocol was used to generate markers for genetic map construction and QTL
analyses. Software MapMaker and QTL Cartographer were used to identify the number,

location and effect of potential QTLs related to resistance.

REFERENCES

Afanasiev MM (1956) Resistance of inbred varieties of sugar beets to Aphanomyces,
Rhizoctonia, and F usarium root rots. J Am Soc of Sugar Beet Tech 9: 178-179

Bockstahler HW, Hogaboam G], and Schneider CL (1950) Further studies on the
inheritance of black root resistance in sugar beets. Proc Am Soc Sugar Beet Technol
pp 104-107

Boudry P, Wieber R, Saumitou-Laprade P, Pillen K, Van Dijk H, and Jung C (1994)
Identification of RFLP markers closely linked to the bolting gene B and their

significance for the study of the annual habit in beets (Beta vulgaris L.). Theor Appl
Genet 88: 852-858

Campbell LG and Bugbee WM (1993) Pre-breeding for root-rot resistance. J Sugar Beet
Res 30: 241-251

Cooke DA and Scott RK (1993) The Sugar Beet Crop. Chapman and Hall. 675 pp

Dale MFB, Ford-Lloyd BV, and Arnold MH (1985) Variation in some agronomically
important characters in a germplasm collection of beet (Beta vulgaris L.). Euphytica
34: 449-455

Letschert JPW, Lange W, Frese L, and Van Den Berg RG (1994) Taxonomy of Beta
Section Beta. J Sugar Beet Res 31: 69-85

Lewellen RT (1992) Use of plant introductions to improve populations and hybrids of
sugarbeet. In ‘Use of Plant Introductions in Cultivar Development (part 2)’. Crop
Science Society of America. pp. 117-136

Lewellen RT and Whitney ED (1993) Registration of germplasm lines developed from
composite crosses of sugarbeet x Beta maritima. Crop Sci 33: 882-883

Schneider CL and Gaskill JO (1961) Tests of foreign introductions of Beta vulgaris L. for

resistance to Aphanomyces cochlioides Drechs. and Rhizoctonia solani Kuehn. J Am
Soc Sugar Beet Tech 11: 656-660

Schumacher K, Schondelmaier J, Barzen E, Steinriicken G, Borchardt D, Weber WE,
Jung C, and Salamini F (1997) Combining different linkage maps in sugarbeet (Beta
vulgaris L.) to make one map. Plant Breed 116: 23-38

USDA (United States Department of Agriculture) (1997) Agricultural Statistics. US.
Government Printing Office

Chapter 1

A novel approach for screening sugarbeet (Beta vulgaris) seedlings for
reaction to seedling disease caused by Aphanomyces cochlioides

Abstract

Breeding for resistance to sugarbeet Aphanomyces seedling disease caused by
Aphanomyces cochlioides has been problematic, with high variability and uncertainty
associated with field scores and those obtained with soil-based methods conducted in the
greenhouse. The genetic control of seedling resistance has not been reported, and
developing a screening method with sufficient sensitivity to reveal genetic variability in
breeding and genetic populations was the goal of this work. A relatively rapid and
inexpensive laboratory method was designed, tested, and evaluated, and results were
correlated to germplasm lines with known reaction of seedling disease and field
resistance. Success in meeting these initial criteria required reexamination of the disease
infection process. Principal innovations were in the production and delivery of inoculum,
and the preparation of seedlings prior to evaluation. Active zoospores of Aphanomyces
cochlioides were used as the inocula, and delivered to seedlings held in a shallow pan
containing sufficient water to cover roots and hypocotyls to simulate flooded field
conditions. A brief (1 day) hydrogen peroxide pre-treatment of seeds prior to
germination was used to induce seedling vigor and to reduce contamination of tests by
seed-borne organisms. Internal controls were included with each replication to reduce
experimental scoring uncertainties. Compared to previous tests, soil-less inoculation

provided a direct estimation of Aphanomyces seedling disease reaction and results were

more consistent with field resistance, which was reflected by a reasonable significant

correlation (R2 = 65.5%, P = 0.015) between water box inoculation and field evaluations.

Introduction

Seedling disease of sugarbeets caused by Aphanomyces cochlioides (Drechs.) has been a
long-term, nearly ubiquitous, post-emergence damping-off problem that sporadically can
seriously reduce stand establishment and thus indirectly reduce yield. The disease nearly
crippled the beet sugar industry in Michigan, USA in 19408 (Schneider 1979). The
causal organism also causes a chronic root rot (Schneider and Whitney 1986). Seedling
disease symptoms include discoloration of hypocotyls and necrosis. Disease intensity is
most severe under high moisture and warm temperatures, and seedlings are seldom
infected below 15°C (Windels and Jones 1989). Chemical controls normally are not
effective for either the seedling or chronic phases of Aphanomyces infection, but may
augment genetic resistance. Metalaxyl, a common and effective fungicide seed treatment
against Pythium damping-off, has limited effect on Aphanomyces (Payne and Williams
1990). Hymexazol (Tachigarin, 3-hydroxy-5-methyl-isoxazole) is commercially applied
as a seed treatment to control Aphanomyces seedling disease, but its efficacy varies, and it
is relatively expensive, toxic to seedlings, and ineffective past the seedling stage (Payne

and Williams 1990, Byford and Prince 1976, Heijbroekm and Huijbregts 1995).

Partial genetic resistance to Aphanomyces seedling disease has been available for
over 50 years (Bockstahler et a1. 1950), however inheritance of resistance is described
incompletely. Most if not all Aphanomyces tolerance deployed in commercial hybrids is
derived from selections obtained in the 1940’s (Henderson and Bockstahler 1946,
Bockstahler et al. 1950). Breeding sugarbeet for resistance to Aphanomyces at both the

seedling and adult plant stages remains an important breeding objective. Host plant

resistance was first selected in naturally infested fields, and was measured by harvest
yield as an indirect measure of the persistence of sugarbeet stands after infection earlier in
the season. Subsequently, greenhouse evaluations were found to generally reflect field
performance (Schneider 1954, 1959) and assist in selection (Coe and Schneider 1966),
however correlations were inconsistent (Schneider and Hogaboam 1983) presumably due
to the difficult to control variables such as effective inoculum concentration and

variability in soil conditions (e. g. soil type, moisture).

Aphanomyces oospores survive several years in soil and debris of infected weed
hosts or sugarbeet (Kirk et al. 2001). Under suitable conditions oospores release
zoospores that swim limited distances (<50 cm) in the soil water column. Zoospores are
sensitive to the environment, and rapidly encyst under adverse conditions such as
agitation, light, and presence of polyvalent cations. Cysts may either form zoospores or
produce germ tubes that are capable of directly penetrating plant tissues (Cerenius and
Soderhall 1985). In infected tissue, mycelia form sporangia and release zoospores into
the soil profile or form oospores within the infected plant tissues. Both oospores and
zoospores have been used for artificial inoculation of sugarbeet seedlings (Coe and
Schneider 1966, Schneider 1954, 1978, Schneider and Hogaboam 1983). As inocula for
resistance selection, oospores have advantages of being easily produced and can be
maintained in long-term storage, however their disadvantage is that the timing of
inoculation of very young seedlings (< two weeks old) introduces a time lag between
oospore germination, zoospore release, infection, ramification, and development of
symptoms. Zoospores have an advantage of being abundant and easily produced, but the

disadvantage of rapidly encysting in adverse conditions, effectively reducing the

10

concentration of active zoospores presented to the seedling.

The method described here was developed for the purpose of determining whether
Aphanomyces seedling disease tolerance is heritable, and if so, how many genes may be
contributing to tolerance, as well as a test that would discriminate between resistance
genes already present in sugarbeet and those that might be available in the wider
germplasm pool. As a method, the protocol may be of general interest and the results of
its development are presented here. Specific criteria used included the ability to rapidly
screen the possible youngest seedlings, suitability for scoring relatively large genetic
populations in a reasonable time frame, and the demonstration of a reasonable correlation

with field results.

11

Materials and Methods
Biological materials

Aphanomyces cochlioides was isolated from a diseased sugarbeet seedling from Saginaw,
MI (designated Ach-15-8-4; courtesy of D. Johnson, Michigan State University) and was
used throughout these experiments. Ach-15-8-4 incited disease development to the same
extent as two other independent isolates tested (designated Ach-R6-4 and Ach-Bl from
the same source) using equivalent inoculum concentrations (however these three isolates

differed in their ability to produce zoospores, data not shown).

A range of sugarbeet varieties and germplasm lines were tested, ranging from highly
susceptible to moderately resistant. Susceptible lines used were a commercial hybrid
‘Edda’ (KWS SAAT AG, Einbeck, Germany) and a penultimate breeding population
leading to the USDA-ARS germplasm release C869 (Lewellen 2003). Moderately
resistant materials included: EL48 (Saunders et al. 2003), EL50 (Saunders et al. 1999b),
US-H20 (Coe and Hogaboam 1971a), SP7622 (the pollinator parent of US-H20, Coe and
Hogaboam 1971b), and ACH185 (American Crystal Sugar Company, Moorhead, MN).

A series of eight moderately resistant germplasm lines sharing a common parentage
through a smooth-root (SR) architecture breeding program were also tested (in order of
increasing relative resistance): SR93 (Saunders 2000), EL0204 (McGrath and Lewellen,
USDA-ARS germplasm release), SR96 and SR97 (McGrath 2003), SR94 (Saunders et al.
1999a), SR95 (Saunders et al. 2000a), SR87 (Saunders et al. 2000b), and SR80 (C.
Theurer, USDA-ARS germplasm release). Finally, three experimental F1 hybrid

populations derived from single plant hybrids of a cross between C869 (or SP7622 in the

12

case of Y1 1-33) and a Beta vulgaris ssp. maritima P1540625 with reported high

resistance to Aphanomyces chronic root rot (Rush 1994).

Zoospore preparation

Mycelia were grown in corn meal broth supplemented with 0.18 mM rifampicin. One
day prior to inoculation, mycelia were washed twice with sterile deionized water,
resuspended in SP solution (85 uM NaCl, 13 uM KCl), and incubated at room
temperature the dark without shaking. The concentration of zoospores released into
solution was estimated using a hemocytometer, and then diluted to 100 zoospores/ml with

SP solution immediately prior to inoculation.

Plant preparation and inoculation

Sugarbeet seeds were surface sterilized (75% ethanol for 1 min, followed by two 15 min
incubations in 0.79% sodium hypochlorite, seeds were then rinsed three times for 5 min
each with deionized water). Surface sterilized seeds were transferred to 88 mM (0.3%)
hydrogen peroxide for 24 hr (50 ml of solution per 100 seeds in 125 m1 Erlenmeyer flask
with shaking at 100 rpm), and then incubated in water under the same conditions until
radicle protrusion (about 2 to 6 days). The water was replaced every 2 d. High quality
seedlots typically had germinated (radical length > 0.5 cm) enough seedlings for
inoculation by 2 (1 post immersion (McGrath et al. 2000). Germinated seedlings were
transferred to hypochlorite-sterilized polystyrene drawer organizers (23 x 15 x 5 cm,
Rubbermaid Co, part number JBI-2916), which were divided into 6 compartments (each

ca. 7.5 x 7 .5 cm) using 5 mm diameter plastic rods as spacers, with spacers arranged so

13

that fluid could flow freely within the box. The seedlings were maintained partially
submerged in water as a condition that may simulate flooded field condition that
exacerbates Aphanomyces seedling disease symptoms in the field. Inoculation was a brief
exposure to Aphanomyces zoospores (3 hr), which allowed nearly synchronous infection
of a batch of seedlings. Zoospores were diluted in an empirically determined hypotonic
solution that reduced the proportion of encysted zoospores during the inoculation (data
not shown), thus allowing zoospores to actively swim and seek seedling roots and
hypocotyls. Ten seedlings per accession were placed in one of the six compartments. .
All three replicates were distributed to different boxes. Seedlings were placed such that
their cotyledons rested on the spacers and avoided contact with the solution. Seedlings
were incubated in the boxes in 90 ml of water at least 1 (1 prior to inoculation to allow for
additional elongation (generally to 2 cm of hypocotyl plus root length). At inoculation,
90 ml of SP solution containing zoospores (100 zoospores/ml) was added, and inoculated
seedlings were incubated in the dark for three hours at 24°C to allow zoospores to find
and attach to seedling roots and hypocotyls, after which the inoculum was replaced with
90 m1 de-ionized water. Boxes were covered with plastic film (e. g. Saran wrap) to
maintain humidity and limit desiccation, and incubated at 24°C in the light. Disease

response was recorded 2 d and 3 (1 post inoculation.

The following criteria were used to assess disease reaction: (i) disease symptoms
(e. g. brown, water-soaked appearance) present on more than two thirds of the hypocotyl
and root surface were considered diseased and ascribed a value = 1 (one): (ii) few or no
symptoms observed (< 10% of root/hypocotyl surface) were considered resistant and

given a value = 0 (zero): and (iii) all other cases were considered as partially diseased

l4

(value = 0.5). The Area Under The Disease Progress Curve (AUDPC, Shaner and Finney

1977) was calculated as follows:

It
AUDPC = 201' —t,°_1)(x,- +xl-_1)/2

i=1
where ‘xi’ was the averaged disease severity of each tested accession at the ith recording,

‘t’ was the time in days, and ‘n’ was the total number of recordings for each tested
sample. Each box contained two reference accessions (EL50 and ACH185) to monitor

consistency of disease scoring within and between different inoculation boxes.

Field tests

Three replicate 3.75 in plots were evaluated twice during the growing season (July 16 and
August 16, 2001; 1 = full stand, 9 = dead) and visual appearance of root rot at harvest
(October 9, 2001; l = no visible lesions or tip rot, 9 = dead) in the Betaseed, Inc.
Aphanomyces nursery at Shakopee, MN. The initial reading strictly assessed field
emergence, while the second and final evaluations were for disease reaction assessment.
At least two years of Aphanomyces disease field-testing at the same location were done
for each SR line over the period from 1997 to 2001 (with the exception of one year for
EL0204). Aphanomyces disease evaluation of SR germplasm as a distinct group was only
conducted in 2001, and correlations with the 2001 test alone were reported here.
Information on SR germplasm Aphanomyces disease reactions was provided in their

registration notices, and was in general concordance with the results reported here.

15

Results

Inoculation of seedling roots and hypocotyls with relatively low concentrations of
zoospores in solution enabled direct scoring of symptoms on roots and hypocotyls, an
advantage over soil-based methods where indirect measure of wilting is observed and
scored in later disease stages. Seeds were disinfected and soaked in hydrogen peroxide
for one day prior to germination, which appears to have an effect on promoting seedling
vigor (McGrath et al. 2000, De los Reyes and McGrath 2003). Symptoms were clearly

visible as water-soaked lesions on roots and hypocotyls within two days after inoculation.

Zoospore production was investigated in detail, with many combinations of isolates
and media tested (data not shown). Generating and maintaining abundant viable
zoospores for artificial inoculation is essential, and zoospore production effects in vitro
included age of mycelia (<1 week old was optimal), monovalent ion concentration (120
mg/L NaCl was optimal), pH (6.5 to 7.0 optimum), and temperature within a range of 20
to 25 C (Schneider 1963, Fowles 1976). Growth of mycelia in corn meal broth was found
to be satisfactory. After resuspension in deionized water, zoospores were released,
however >80% encysted within 2 hours. Resuspension in reverse osmosis purified water
provided greater longevity to released zoospores, however the decay in motility was
variable, presumably due to slight variations in water quality. Among several different
concentrations of salts tested, deionized water plus NaCl (85 uM) and KCl (13 uM) was
found acceptable for production and maintenance of zoospores, and >50% zoospore
activity was evident two days after resuspension. Concentrations of NaCl alone ranging

from 0.1 to 100 mM (in a lO-fold dilution series) were tested, and zoospore encystment

l6

ensued more rapidly with increasing salt concentration (data not shown).

Reaction of commercial hybrids and USDA germplasm releases to Aphanomyces
seedling disease are generally known through public and private variety trials. A range of
germplasm differing with respect to Aphanomyces seedling reaction was chosen to
examine whether this inoculation scheme would discriminate the known range of
reaction, and if so, whether expected relationships would be evident. Of particular
interest was the comparison of hybrid US-H20 with its seed parent lineage exemplified by
EL48 and its pollen parent lineage represented as SP7622 (Figure 1.1). EL48 typifies the
seed parent lineage derived from germplasm developed for the Western US. regions
where Aphanomyces seedling disease has not been breeding priority, whereas SP7622 has
been derived from highly Aphanomyces tolerant germplasm. Although EL48 has
undergone an additional round of field selection for Aphanomyces tolerance from the
actual US-H20 seed parent E1A4/ EL45, the intermediate disease reaction of US-H20
relative to the parents is an expected result. Similarly, both C869 and Edda showed
expected high susceptibility (Figure 1.1). C869 is a germplasm release strictly developed
for Western US. conditions and is highly vigorous, and Edda is a vigorous Aphanomyces
susceptible check variety used in Eastern US. disease nurseries. Test lines, YO3-3 84 and
Y03-388, each derived from crosses between C869 and a highly tolerant wild beet
accession from the coast of France (P1540625), also showed reasonable levels of
tolerance (Figure 1.1). Results from this approach to testing Aphanomyces reaction were

the first to conform to expectations, and were pursued further.

17

 

C
II 30‘
2
i
c 20-
’5
ca
or
E- 10-
i

 

 

coco Edda EL48 roam usmo Y1‘I-33 Yos-saa SP7622
Gerrnplasm Accession

Figure 1.1 Mean percent oprhanomyces seedling disease reaction scores of eight
sugarbeet germplasm lines in water box inoculation test. Data from third day post
inoculation. Score 0 is defined as no or few (<10% of seedling surface) visible lesions.
Error bars are standard errors of the mean.

Two moderately resistant accessions, EL50 and ACH185, were tested extensively to
ascertain the reproducibility of the water box inoculation approach. Four experiments
were done, each with 10 plants of each accession per replicate water inoculation box.
Each of the four experiments had 19, 22, 13, and 6 replicates for experiments 1, 2, 3, and
4, respectively, and was conducted in conjunction with a phenotypic screen of a
population segregating for Aphanomyces seedling disease reaction (data not shown).
With the exception of the first experiment, differences in replicate means for either
variety were not significant (Figure 1.2). For Experiment 1, mean disease reaction was
less than other experiments with both accessions. Two explanations could be offered for
this discontinuity. Zoospores diluted to the final concentration were observed to quickly
settle towards the bottom of the beaker from which inocula were drawn, likely resulting

in less inoculum delivered in experiment 1 relative to the other experiments. The effect

18

of differing inoculum levels was not specifically tested here. The first experiment was
performed at ambient temperature in the laboratory (23°C i 2°C), whereas better
temperature control was afforded in a growth room (24°C i 02°C), where the remaining

experiments were conducted.

 

 

 

 

 

 

 

 

2.07
FL _1_ _r_
15—
O
D.
g
<
0.5-
Test 1 2 a 4 1 2 3 4

ACH1 85 EL50

Figure 1.2 Area under the disease progress curve (AUDPC) scores for two sugarbeet
germplasm lines inoculated with Aphanomyces cochlioides in water box inoculation test.
Error bars are standard errors of the mean.

A series of recent SR germplasm releases (USDA-ARS) were tested with the water
box inoculation approach and compared with field results taken August 16, 2001 at the
Betaseed, Inc. Aphanomyces disease nursery in Shakopee, MN (Figure 1.3). Correlation
of field test scores and water box inoculation scores in this test and at this time was R2:
65.5% (p = 0.01). At this level of resolution, water box inoculation of seedlings appeared
to sufficiently reflect field performance. These results should be interpreted with caution,
however, since at harvest on October 9, 2001, R2 for the same comparison was 29.1% (p
= 0.17). It should be noted that field scores are not combined over the season into a

single score due to uncertainty in their relationships, and scores are not generally

combined over years for the same reason. Four of these accessions from Figure 1.3
(SR87, SR94, SR95, and SR93) were rated over three field seasons (1997, 1998, 2001) in
the same nursery. With the exception of SR93, excellent prediction of field performance
was evident by taking the average of ratings at harvest (R2 = 99.7%, p = 0.03). However,
with the addition of SR93 in the analyses, little or no significance to the trend was seen,
either due to an over-estimation of the score in the water box experiments or under-
estimated field scores, both for unknown reasons. It should be noted that less than a
decade of field testing for Aphanomyces disease in sugarbeet has occurred at any one site,
and field host-pathogen interactions have not been corroborated by genetic analyses of
host resistance, and many questions remain that could begin to be addressed using
unrelated disease assessment approaches. Part of the problem in deducing disease
reactions with Aphanomyces in sugarbeet, as with many other diseases, has been the lack
of simple, reproducible experimental tests with which to build a knowledge base to
extend and confirm field observations. The goal of this work was not to develop a
predictor of field performance per se, but rather to develop a method that would allow an
approach to dissecting the genetic control of the interaction between sugarbeet seedlings
and the causal organism of Aphanomyces seedling disease that, in some degree, reflected

field performance.

20

 

 

 

 

 

4
R2 = 65.5% .
P = 0.015 SR93
a, 3.54
L-
O
O
rn
:2
.9
u' 3 4
2.5 V 77 I I
1.5 1.6 1.7

1 1.1 1.2 1.3 1.4
Box inoculation (AUDPC)

Figure 1.3 Correlation of Aphanomyces disease reaction for eight germplasm releases in

water box inoculation test and field evaluation (taken on August 16, 2001 at Shakopee,
MN).

21

Discussion

Two motivations have driven the research reported here. One is the need to introgress
additional sources of Aphanomyces seedling disease tolerance into the cultivated
sugarbeet gene pool, and second is the need to demonstrate more precisely the genetic
control of Aphanomyces seedling disease tolerance. For the latter to be successful,
methods were sought that would allow screening of segregating populations for genetic
analyses on limited numbers of individuals that are typically available for selection in

early-generation breeding populations.

Laboratory approaches for disease resistance evaluation of young sugarbeet
seedlings were explored, and the most reasonable approach is reported here. Both soil-
based and soil-less methods were evaluated, and various methods of inoculation (e. g.
detached leaf and hypocotyl assays, direct cotyledon inoculation) were tested. Most of
these methods failed to satisfactorily meet our criteria, that is, poor correlations with field
results were obtained (data not shown). Rapid encystment of zoospores when stressed
and their subsequent failure to infect seedlings when used as inocula for soil-based
assays, for instance, or the rapid ramification of detached tissues directly infected with
mycelia or high zoospore concentrations, and the subsequent inability to sufficiently
discriminate between tolerance and susceptibility, may have led in part to a lack of
correlations with field results. For instance, in zoospore-inoculated soil, the highly
susceptible variety Edda showed a highly resistant disease reaction and C869 performed

similar to US-H20 (data not shown).

Aphanomyces cochlioides is a weakly competitive pathogen in natural conditions

22

(Williams and Asher 1996), which may explain some of the variability observed under
field conditions. Surface sterilization of seeds prior to inoculation in water boxes
appeared to be desirable to minimize potential contamination with more aggressive seed-
bome organisms. Unlike soil-based inoculations, contamination would have been readily
apparent in water boxes. However, contamination was not observed, perhaps due in part
to the fi'equent changes of solutions during the course of preparing seedlings for
inoculation. The use of active zoospores to inoculate seedlings growing in water
facilitated the control of inoculum concentration. Zoospores interacted directly with the
plant without interference or encystment. When suspended in an isotonic solution,
zoospores were active during the inoculation period, insuring uniformly infected
seedlings, and the inocula were removed after three-hour incubation, avoiding continuous

zoospore infection as one other possible interference in other types of assays.

Anecdotal reports suggested seedling vigor is a component of Aphanomyces
seedling disease tolerance, and investigations into sugarbeet seedling vigor have
suggested a method to normalize seedling vigor differences between otherwise viable
seedlots using a hydrogen peroxide treatment during germination (McGrath et al. 2000,
de los Reyes and McGrath 2003). For example, the germination rate of EL48 increased
from 48% in water to 75% in hydrogen peroxide. Inoculation of these seedlings with
zoospores solution suggested that seedling disease severity was not affected, or perhaps
slightly increased, by hydrogen peroxide germination treatment (20% with disease score
of 1 in water, vs. 30% in hydrogen peroxide, scored two days post-inoculation).
Hydrogen peroxide treatments do not appear to factor into the outcome of disease

reaction. The induction of seedling vigor in otherwise low vigor seedlots such as

23

ACH185 may bear on the ability to discriminate between disease reactions in very young
seedlings, in addition to any role that hydrogen peroxide treatment may have in sterilizing

the seed surface.

Results obtained here on two-week old seedlings suggest that Aphanomyces seedling
disease resistance exists in sugarbeet, and the inheritance of this resistance should be
accessible using this water box inoculation approach. Further, these results suggest that
such genes are active during the earlier stages of sugarbeet stand establishment. The
nature of resistance is not clear. All accessions tested in the water box inoculation
approach showed the majority of seedlings to be infected, and the exclusion of infection
as a mechanism of resistance does not appear as a possible resistance mechanism. Rather
the outcome of infection appears to be subject to an unknown interaction between
infected and uninfected tissue in the seedlings, or some other mechanism that would limit
the grth of the pathogen relative to the growth of the seedling. In any case, the

conditions of box inoculation are likely more severe than those encountered in the field.

It is expected that water box inoculations will provide a high uniform selection
pressure for breeding improved Aphanomyces seedling disease reaction in sugarbeet. If
not, the technique may allow better observation of the early stages of infection than is
currently available. Field performance of sugarbeets growing under Aphanomyces
disease pressure is highly variable, at least early in the season (Beale et al. 2002). Field
and greenhouse selections have been successful in breeding resistance to Aphanomyces
diseases (Bockstahler 1950, Coe and Hogaboam 1971b). Deploying additional sources of

resistance to Aphanomyces seedling disease, if identified, will require additional

24

screening methods in order to increase the efficiency of selection, particularly in early
breeding generations where sources of resistance may be present at low frequency in wild
species, and insufficient resources limit application of greenhouse and field screening
approaches. The approach described here will allow for the direct analyses of
Aphanomyces seedling disease interaction with its host, and perhaps allow for analyses of
the events that appear to limit economic loss to growers through expression of genetic

resistance.

25

REFERENCES

Beale JW, Windels CE, and Kinkel LL (2002) Spatial distribution of Aphanomyces
cochlioides and root rot in sugarbeet fields. Plant Disease 86: 547-551

Bockstahler HW, Hogaboam GI, and Schneider CL (1950) Further studies on the
inheritance of black root resistance in sugarbeets. Proc. Am. Soc. Sugarbeet Tech. 6:
104-107

Byford WJ and Prince J (1976) Experiments with fungicides to control Aphanomyces
cochlioides in agricultural soils in England. Trans. Br. Mycol. Soc. 65: 159-162

Cerenius L and Soderhall K (1985) Repeated zoospore emergence as a possible
adaptation to parasitism in Aphanomyces . Experimental Mycology 9: 259-263

Coe GE and Hogaboam GJ (1971a) Registration of US H20 sugarbeet. Crop Sci 11: 942

Coe GE and Hogaboam GJ (1971b) Registration of sugarbeet parental line SP 6322-0.
Crop Sci 11: 947

Coe GE and Schneider CL (1966) Selecting sugarbeet seedlings for resistance to
Aphanomyces cochlioides. Journal of the American Society of Sugarbeet
Technologists 14[2]: 164-167

Fowles B (1976) Factors affecting growth and reproduction in selected species of
Aphanomyces. Mycologia 68: 1221-1232

De los Reyes BG and McGrath J M (2003) Cultivar-specific seedling vigor and expression
of a putative oxalate oxidase germin-like protein in sugarbeet (Beta vulgaris L.).
Theoretical and Applied Genetics 107: 54-61

Heijbroekm W and Huijbregts AWM (1995) Fungicides and insecticides applied to
pelleted sugar-beet seeds—H. Control of pathogenic fungi in soil. Crop Protection 14:
363-366

Henderson RW and Bockstahler HW (1946) Reaction of sugarbeet strains to
Aphanomyces cochlioides. Proc Am Soc Sugarbeet Tech, 273-245

Kirk PM, Cannon PF, David JC and Stalpers J A (2001) Ainsworth & Bisby's Dictionary
of the Fungi. 9th ed., CAB International, Wallingford, UK.

Lewellen RT (2003) Registration of rhizomania resistant, monogerm populations C869
and C869CMS. Crop Science: in press

McGrath JM, Derrico C, Morales M, Copeland L0, and Christenson DR (2000)

26

Germination of sugarbeet (Beta vulgaris) seed submerged in hydrogen peroxide and
water as a means to discriminate cultivar and seedlot vigor. Seed Science and
Technology 28: 607-620

McGrath JM (2003) Registration of SR96 and SR97. Crop Sci: in press

Payne PA and Williams GE (1990) Hymexazol treatment of sugar-beet seed to control
seedling disease caused by Pythium spp. and Aphanomyces cochlioides. Crop
Protection 9: 371-377

Rush C (1994) USDA, ARS, National Genetic Resources Program. Gerrnplasm
Resources Information Network - (GRIN). [Online Database] National Germplasm

Resources Laboratory, Beltsville, Maryland. Available: http://www.ars-grin.gov/cgi-
bin/npgs/html/eval.p1?269 (07 June 2003)

Saunders JW, McGrath JM, Halloin J M, and Theurer J C (1999a) SR94 Sugarbeet. Crop
Sci392297

Saunders J W, Theurer J C, and Halloin JM (1999b) Registration of EL50 monogerm
sugarbeet with resistance to Cercospora leaf spot and Aphanomyces blackroot. Crop
Sci 39: 883

Saunders JW (2000) SR93 Sugarbeet. Crop Sci 40: 304

Saunders JW, McGrath JM, Halloin J M, and Theurer JC (2000a) SR95 Sugarbeet. Crop
Sci 40: 1205

Saunders J W, McGrath J M, Theurer J C, and Halloin J M (2000b) SR87 Sugarbeet. Crop
Sci4021834

Saunders JW, Halloin JM, and McGrath JM (2003) Registration of EL52 and EL48
Sugarbeet Gerrnplasm. Crop Sci 43: 744-745

Schneider CL (1954) Methods of inoculating sugarbeets with Aphanomyces cochlioides
Drechs. J Am Soc Sugarbeet Tech 8: 247-251

Schneider CL (1959) Field inoculation of sugarbeets with Aphanomyces cochlioides
Drechs. J Am Soc Sugarbeet Tech 10: 647-650

Schneider CL (1963) Cultural and environmental requirements for production of
zoospores by Aphanomyces cochlioides in vitro. J Am Soc Sugarbeet Tech 12: 597-
602

Schneider CL (1978) Use of oospore inoculum of Aphanomyces cochlioides to initiate
blackroot disease in sugarbeet seedlings. J Am Soc Sugarbeet Tech 20: 55-62

Schneider CL (1979) The present need for black root control. Proc 20th Regional

27

Meeting, Eastern U.S., Am. Soc. Sugarbeet Tech. 79-80.

Schneider CL and Hogaboam GJ (1983) Evaluation of sugarbeet breeding lines in
greenhouse test for resistance to Aphanomyces cochlioides. J Am Soc Sugarbeet Tech
22: 101-107

Schneider CL and Whitney ED (1986) Black Root. p. 17. In E.D. Whitney, and J .E.
Duffus (ed.) Compendium of Beet Diseases and Insects. The American
Phytopathological Society, The American Phytopathological Society.

Shaner G and F inney RE (1977) The effect of nitrogen fertilization on the expression of
slow-mildewing resistance in Knox wheat. Phytopathology. 67:1051-1056

Williams GE and Asher M] C (1996) Selection of rhizobacteria for the control of Pythium
ultimum and Aphanomyces cochlioides on sugar-beet seedlings. Crop Protection 15:
479-486

Windels CE and Jones RK (1989) Seedling and root diseases of sugarbeets. Univ
Minnesota Ext Serv AG-FO—3702. 8 pp

28

Chapter 2

Development of a genetic map to detect genes involved in Aphanomyces
resistance and some agronomic traits

Abstract
To identify possible QTLs related to Aphanomyces disease resistance, an F 2 population
was derived from a cross of susceptible sugarbeet by wild beet, which has been reported
with high levels of resistance. A genetic map based on this population was constructed
with 139 AF LP and 24 Selective Amplification of Microsatellite Polymorphism Loci
(SAMPL) markers, and the final map comprised 9 linkage groups, and had a total length
of 507.1 cM. Both PstI and EcoRI restriction enzymes were used to select AF LP
markers, and their effect on map construction was compared. Marker distortion among
different types of markers was detected. Phylogenetic analyses showed that several
Aphanomyces susceptible accessions were genetically closely related. Several
economically important agronomic traits, such as monogerm/multigerm, male-sterility
and germination time were investigated, and molecular markers linked to each trait were

identified.

29

Introduction
The development of new techniques has dramatically accelerated the progress of
biological research, especially applications of PCR (Mullis et al. 1986) and AF LP (Vos et
a1. 1995) to plant genetic analyses. With these advancements, generating large amount of

reproducible DNA markers has become feasible and efficient.

Amplified Fragment Length Polymorphism (AFLP)
Amplified Fragment Length Polymorphism (AF LP), a PCR-based genotyping method,
has been widely used to detect polymorphism in both prokaryotic and eukaryotic
organisms. The systematic description of AFLP and its application was published in
1995 (V05 et al. 1995). Before the utilization of fluorescence labeling, most AF LP used
P33 dATP to display DNA fragments. Fluorescent chemicals have been used to replace
radioisotopes for prime labeling, which can be detected by laser scanner in automated
sequencing system, such as the ABI3 77 or Li-Cor 4200. Automation has markedly
reduced labor costs and increased the number of markers scored in a given time.
Consequently, AF LP has become one of the best choices for molecular genetic mapping.
The study of disease resistance in crops has been greatly accelerated by the
application of AF LP technique in identifying DNA markers linked to resistance genes.
AF LP markers closely linked to the RI gene for late blight resistance were identified in
potato (Meksem et al. 1995). Thomas et al. (1995) found AF LP markers tightly linked to
Cf-9 gene for resistance to C ladosporium fulvum in tomato, and the time input was much
less than transposon tagging methods. Five AF LP QTL markers had been identified for

partial resistance to Puccinia hordei in barley (Qi et al. 2000). Voorn'ps et al. (1997)

30

demonstrated two AF LP loci (pb-3 and pb-4) linked to clubroot disease resistance in a
double-haploid population of Brassica oleracea. Cnops et al. (1996) obtained 17 AF LP
markers related to mutation locus trnI in Arabidopsis. AF LP markers adjacent to Mla
(powdery mildew) resistance gene cluster were identified by Wei et al. (1999).

In sugarbeet, several maps using AF LP have been reported (Schondelmaier et al.
1996; Schafer-Pregl et al. 1999; Setiawan et al. 2000). The first AF LP map of sugarbeet
constructed by Schondelmaier et al. (1996) was integrated into an existing RFLP map.
The PCR amplification yielded 19 to 40 polymorphic bands per primer combination for
an F2 population. Schondelmaier et al. (1996) showed AF LP markers were distributed
rather evenly throughout nine linkage groups. Hansen et al. (1999) found that AF LP was
highly reproducible (97.6%) in sugarbeet, with low genotyping error (0.2%). Using 226
markers (182 AF LPs and 44 RFLPs), Setiawan et al. (2000) constructed a map (744 cM)
of nine linkage groups.

Because of high sensitivity to C (cytosine) methylation on restriction sites,
restriction enzyme Pstl has been often used to reduce band complexity in AFLP, and may
exhibit preference to cut in gene-rich regions, which is valuable for large and complex
plant genomes like pine (Paglia and Morgante 1998). In soybean, Pstl restriction tends to
generate better distribution and less clustering of AF LP markers than EcoRI (Young et al.
1999), and similar results were observed in maize (Vuylsteke et al. 1999). Thus, both

EcoRI and Pstl were used in this study.

31

Genetic map of sugarbeet

Due to the large number of useful molecular markers detected in the past decade, genetic
map construction is more convenient than ever. Compared to other crops, sugarbeet has a
relatively small genome, which was estimated to be 758 Mbp (Arumuganathan and Earle
1991)

In sugarbeet, the resistance to Aphanomyces disease has been studied for decades,
however, little progress has been made in understanding resistance at genetic level.
Current knowledge is that the resistance is moderate and heritable. No genetic map has
been previously reported for the resistance. In order to gain insight into the genetic
components and location of resistance QTLs (quantitative trait loci), a genetic map from a

resistance-segregating population is necessary.

Agronomic traits

In addition to Aphanomyces resistance, several agronomic traits were evaluated, including
qualitative traits, such as monogerm/multigerm, genetic male sterility, and a quantitative
trait, germination time that was grouped into six ordered categories. These traits were

analyzed with logistic regression.

Monogerm

Monogerm is important for sugarbeet as it can facilitate precision planting and reduce
production costs. In the U.S., most of monogerm materials used in sugarbeet breeding
programs can be traced back to a single monogerm plant in Michigan Hybrid 18 (Savitsky

1952).

32

Multigerm is dominant over monogerm, and the four multi germ alleles are named
M, M], MBr and M2 (Smith 1980). However, dominance is not complete, as monogerm
flowers are observed in F1 plants of multigerm x monogerm cross. Based on selfing
information, monogerm was shown to be controlled by one recessive gene ‘m’ (Savitsky
1952). Sadeghian and Khodaii (1998) reported the heritability of monogermy to be 58%,
which was mainly due to additive effects. Laporte et al. (1998) reported that monogerm

site was located very close to restorer gene for CMS.

Germination time

Seeds of many species do not germinate immediately after dispersal, but instead may
remain dormant for a certain period, and the causes and consequences of this variation of
dormancy and germination remain unclear. Ecker et al. (1994) found that genes with
additive effects conferring fast germination were correlated with seed dormancy in
Lisianthus (Eustoma grandiflorum). In coconut, Herran et al. (2000) used AF LP and
other markers to identify 6 QTLs responsible for early germination.

Sugarbeet germination under different environmental conditions has been
extensively tested, and environmental factors such as moisture, osmotic pressure,
temperature, sowing depth, physical properties of soils, chemical inhibitors and
activators, and disease pressure are found to significantly affect seed germination
(Gummerson 1986, Rush and Vaughn 1993, McGrath et al. 2000, Ghoulam and Fares
2001). However, genetic components for germination time are unclear. Reciprocal
differences in germination (P < 0.05) had been observed in sugarbeet, and accounted for

about 50% of total genetic variance, indicating the involvement of cytoplasmic genes

33

(Sadeghian and Khodaii 1998). Germination time varied with variety and seed lot, which
implied genetic role (McGrath et al. 2000). With a novel, simple, and reliable laboratory
assay developed in this lab, some genes were shown expressed under abiotic stress but
not in standard tests. One of these genes, gerrnin-analog (a putative oxalate oxidase) was
induced to high levels in US-H20 (a variety with good emergence) but at low or
undetectable levels in ACH185 (a variety with poor emergence) (De los Reyes and

McGrath 2003).

Genetic male sterility (GMS)

Owen first reported the existence of male-sterility in sugarbeets, which is known as O
type of cytoplasmic male sterility (CMS). He also reported the genetic (or genie,
Mendelian, nuclear) male sterility (GMS, Owen 1952). Two recessive loci a1, a2,
responsible for GMS in sugarbeet, are different from CMS restorer genes x and 2. Gene
a1 is not linked to either gene R (red hypocotyl) or m (monogerm) (Owen, 1952). Gene
a2 confers complete male sterility in recessive homozygous plants, but heterozygous
individuals are fertile (Mglinets et al. 1998).

GMS has been rarely used in commercial breeding, and is much less studied than
CMS. However, GMS is valuable for sugarbeet breeding, since GMS segregants can be
used to ensure cross-fertilization in experimental crosses. This research intended to
obtain preliminary information on markers linked to genetic male sterility gene in

sugarbeet.

34

Materials and Methods

Mapping population
The population (Y03-384) selected for genetic mapping was derived from a cross of
sugarbeet (C869) by wild beet (P1540625). An F2 population (total 145 plants) from a
single F1 plant was developed in the green house at Michigan State University, East
Lansing in year 2000. Leaf tissues from each F2 plant were collected for DNA extraction.
In addition to 145 F2 individuals from mapping population, three plants from
maternal parent C869, three plants from paternal wild beet parent (P1540625), two plants
from Y04-99129 (F2 of C869 x P1546409), two plants from Y11-33 (F2 of SP6822 x
P1640625), plants from Aphanomyces susceptible accessions (two Beta16AB, two
EDDA), several Aphanomyces resistant accessions (one EL48 plant, two SP6822 plants,
one US-H20 plant, and two ACH185 plants) were used to assess the genetic relationships

among these accessions.

AF LP

AF LP protocol described by Vos et al. (1995) was followed with some modifications (see
Appendix I). The major change was using infrared dye (1RD) labeled primer, and product
detection was done with a Li-Cor NEN 4200 instrument, which accommodates two
different reactions labeled with different colored dyes in a single lane (e. g.
IRD700/IRD800, Table 2.1). Both Pstl and EcoRI restriction enzymes were used in this
mapping study in order to investigate the distribution and differences of EcoRI and Pstl
markers in sugarbeet. Primers containing three selective nucleotides were chosen based

on their GC content, melting temperature, and hairpin information using software

35

PRIMER4 (PREMIER Biosoft International, Palo Alto, CA). Primer sequences used are
listed in Appendix H. The primers were synthesized either by MWG (High Point, NC), or
GIBCO BRL (Rockville, MD).

Three randomly selected F2 plants were examined to evaluate total bands number
and polymorphism for each tested primer combination. A subset of 11 primer
combinations displaying high polymorphism was chosen from these tests, and used for all

subsequent AF LP/SAMPL marker generation (Table 2.1).

Table 2.1. Marker nomenclature

 

 

 

Primer Type Primer Primer 1 Label Primer 2
combination
name

AFLP ACOA Pstl + AC* IRD700 MseI i‘+ ACA
ACAA EcoRI + ACA IRD700 MseI + ACA
AGOA Pstl + AG IRD800 MseI + AGC
AGOC Pstl + AG IRD800 MseI + CAT
AGCC EcoRI + AGC IRD700 MseI + CAG
CAOA Pstl + CA IRD700 MseI + AGC
CATA EcoRI + CAT IRD700 MseI + ACC
CTCC EcoRI + CTC IRD700 MseI + CAG
TOCA Pstl + TC 1RD800 MseI + CAT
TOCT Pstl + TC IRD800 MseI + CTT

SAMPL SL1 C SAMPL] A IRD700 MseI + CGG

 

Note: * : Primer with core sequence for Pstl (CAGTCTACGAGTGCAG), or EcoRI
(GACTGCGTACCAATTC) plus two or three selective nucleotides.

A: SAMPLl sequence is CACACACACACACACTATAT.

# : MseI primer core sequence is GATGAGTCCTGAGTAA .

36

Markers were named by their primer combinations (Table 2.1), plus their estimated
length. The size of marker was estimated by comparing to DNA size standards. For
instance, CAOA383 had primer combination PstI+CA/MseI+AGC, the size of marker was

estimated 383 bp.

Data collection and analyses

Real-time 1RD labeled AF LP data was collected and recorded during electrophoresis, and
saved as a TIFF image. AF LP TIFF images were analyzed with Gene ImagIR 4.03
software (LiCor). Since genotyping errors greatly inflate estimates of map distance, only
markers with un-ambiguous scores were used for data analysis. The results were analyzed
with PAUP (phylogenetic analysis using parsimony). PAUP files for all 11 AF LP
combinations were combined, and analyzed with Treecon® (Van de Peer and De Wachter
1994), and a phylogenetic tree was constructed. Data was then sorted based on
polymorphism in the mapping population. Only polymorphic band information was
imported into MAPMAKER/EXP 3.0. First, high stringent grouping command “group 11
25” was executed in order to reduce map error (Setiawan et al. 2000). After all unlinked
markers were removed, command “group 3 25” was executed. After the best order was
set, other markers were added using either “build” or “try” command. Finally the
command “ripple” was applied repeatedly to obtain the highest likelihood value possible.
Marker order identified by Kosambi function was used, and the corresponding map in

Haldane cM was built to make it comparable with previously reported maps.

37

Agronomic trait analyses

All agronomic traits were first analyzed with single marker association analysis. All
mapped markers showing significant association to a scored trait were analyzed
separately. Map interval showing significant associations with two or more contiguous
marker was considered to define one putative trait locus.

Binomial and multinomial traits were analyzed with logistic regression (PROC
LOGISTIC procedures, in SAS V8). Only markers showing significance in the
association analyses were used in logistic regression. Markers were defined in the class
statement, and stepwise selection was used in PROC LOGISTIC analyses. Once
significant markers were identified, their interactions were added to the model and re-
tested with PROC LOGISTIC to see if the interactions were significant. All non-
significant interactions were removed from the model. After individual markers were
selected, their effect and interactions were evaluated with “rsq” selection in the logistic
regression. Markers with P value greater than 0.05 were removed from the model, and

the remaining markers constituted the final logistic model.

38

Results

Genetic map of sugarbeet by wild beet

The final map contained 163 molecular markers distributed on 9 linkage groups (LGs),
for a total length of 507.1 cM (Figure 2.1 and Table 2.2). Intervals between markers
ranged from 0 to 31.6 cM. The interval average was 3.30 cM, or 4.39 cM (total 115
intervals) when the 38 intervals with 0 cM distance were excluded. If the first step
(“group 11 25”) was not used, the total map length was 712.5 cM, and linkage groups 3A
(LG3A 8.30M) and 3B (LG3B 43.3cM) were linked as one group. This indicated the
genetic map was incomplete, and more markers were needed to complete the map. The
summary of the map is shown in Table 2.2. Twenty-nine markers were umnapped, and
eachrprimer combination had similar unmapped percentages. Linkage group 4 (LG4) had
25 markers, the largest number among 9 linkage groups. Linkage group 6 (LG6) had the
fewest number of markers, six. Other linkage groups were close to the average of 16
markers per group. No obvious bias of primer combination to linkage group was
observed. On average, mapped markers per primer combination for EcoRI and Pstl
primers was 18.8 and 10.6, respectively. In comparison, SAMPL markers had the highest

number of mapped markers, 24, per combination (Table 2.2).

PstI and EcoRI comparison in AFLP

The sugarbeet genome is relatively small, and the number of AF LP bands per primer
combination is manageable on a single gel. Much fewer DNA bands were observed using
Pstl than EcoRI enzyme (Table 2.2). On average, each Pstl primer combination yielded

12.8 polymorphic markers, while each EcoRI primer combination yielded 21.3

39

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

     
   

 

 

 

 

 

 

 

 

 

 

   
   

 

 

 

 

 

 

   

   

   
 
  
 
   
  

 

 

 

 

171.19) ‘ incosr 0.0 4004329 0.0 SL10501
. 000190 4.9 - 7007322 2.3 4004297
8.3 :I [ 0474082 7.5 -\ /— 4004321 2.3 A / 0474230
9.0 A r’ 0700168 8.3 -I.L/ f 0404357 3.1 —\‘*"’/_ 0404494
10.8 '\\1\ 77- 7007188 9.9 -\Q\ )/- 0700498 3.1 7, \ 4000170
11.5 —\\\,1 /, )- 0404202 11.4 —-\\\:;/) 0700181 3.1 ) r“ \ 4000384
12.3 N), /;,- 4044170 11.5 —.~\-,_,/_ /- 0474428 3.2 J \— 0474098
:3; 91:24-— ecmmlg 12.0 QL/{f 4044123 4.0 0474308
. N _.. \ 12.2 “‘44:” 4044078 8.2 4004083
12.8 fimx‘: 7004178 12.7 7E\\— 4004149 8.2 3x /— 4044315
13.2 J/,/“\ 4044252 13.4 _/ \\ 4044215 8.3 4004888
13.2 4044209 14.9 SL1C103
14.5 J/H ‘— 5L1c129 17.9 51.10100
17.7 4004535
:21 2. 113‘” 0. L833
. fl . 114 . ACOAMT
26.4 WFQMV/ CTCCBIO LQ5 2.5 /T\ AGOCIOG
g; “I" T“ 4000318 3 70070120882 3'; 1*; 4000121
8.7 $H¢ 4000293 11.8 —rj‘“\ 5L1c109
7.2 «i=0, s1.10085 13.8 -/ 4004225
L 11.7 \ 4000292 25.4 \ 4044111
195 _. /— SL10313 27.0 \\ /- 0404211
0.0 — 51.10418 20.7 :~\ 0404348 27.1 §TC 4000083
22.8 —\ /— 0474191 20.7 “II /, ~ 4000103 27.1 : \ 0700294
.._ . . ,._ _/ 1’ '
a: I , [22:8 :1: 3.87.1882: :7; x9283
40.2 -\\ I /_ 0474285 23.4 —\‘t;/— 0700209 28.7 / 0474322
40.3 “ii. )4")- 0474147 24.3 7,) 0700142 28.7 J 0474304
40.3 -\ ,- ,4— 4000187 24.4 {amp -— 4000254 43.3 4004157
44.4 «- ' 7— 7004070 24.9 J/ \_ 0404097
44.5 —\;. . 4000198 28.4 J/“*\_ 4044129
48.8 ——~ 4044493 30.8 —' 0474183 L 6
53.9 '7 «— 0404383 35.5 \_ 0474124 9!
2:: 5.. 32 ./ sass:
. L} IN.“ .
54.7 J47 NF 0700401 19.3 -\ /— s1.10112
58.2 J r ‘\\‘- 7004319 23.1 ., , 4044137
582 :/// \_ .\- 404.4572 0.0 4000309 23.8 34 a}; 4004101
57.9 j \ SLICITI 8.0 51.10185 23.9 ‘ J” 4004188
73.3 ‘— SL10225 37.6 )— TOCT076
95.7 4044094 41.3 )— 0474159
103.8 —\ .. 4/ 7004297 42.1 [,7/7— 4004501 L 9
1072 \,.=..,.,/— 4000174 42.4 ,7),- 0474451
107.9 {4+ .‘\— 0700158 44.9 Ir- 0404335 0.0 5L1co72
110.9 -/ H \— 0404129 47.1 . “L!" ACAA344 4; \.4L . $410305
119.4 —/ "-\— 0700200 59.0 f’ ,— 4000195 4. ~ to.“ SL10228
127.0 , 4000272 65.8 —\\ ”:4 44 /, 0474130 4.8 4;? \II‘T‘“ SLIC345
‘7’ 66.9 45), ,,/),.~ 0474175 4.8 +7 ,0, ‘I-li- suczo?
88.9 ~31; )’)f/r- 4044191 63 '4), i. ‘7 0100069
L67 87.0 J71] ',.,:;;;~ 7004081 8.3 J. / \\' 4000150
. .. 87.0 “I 31’ 2- 0700509 10.0 J ' CATA438
0.0 0474383 87.9 __,._..z._ 4004072 10.1 0474140
20.4 -\ SL1C318 87.9 _, , .\“— 7007150 18.1 J 175k 0474091
_, ,- x _ __/.L g _
23.8 —\ )— 4004281 73.8 “/1 \\ 4044339 34.5 4,1,4 \‘ 0404221
25.1 —«\ /)— 0700188 75 4 J4” \\\— 4044334 34 5 +1 r- 1\ - CTCC251
25.8 “I“ ///r- 4044258 780 J4 \\— 7007200 35.1 J?— i \ .‘\\— 5L1c197
27.2 1‘34) )r— 4044233 804 J/ 0700524 35:8 J// “‘\— 0404501
27.3 +£2.1- C. 174-— 40442 78 89.3 J' 0404489 38 5 J/ I: 4004190
27.7 —\>- u- 4004138 102.3 ——-, , SL10294 38.8 J 0474133
28.9 ~/ - 4044284 " 41.0 J 4000070
35.4 — /— 7007145 82.4 V 0474084
41.8 -—. /- 7004348
44.3 -.g~ ,- 4044183
45.2 —j:i{-- 4044281
48.2 J. “>~— 4044537
48.2 J x , 4044373
50.8 J 9 4004083
59.7 —— 51.10317

 

Figure 2.1. Genetic linkage map of sugarbeet (C869) X wild beet (P1540625).
Number on the left side of chromosome shows the genetic distance (0M). Marker name is
to the right

40

ja§<w .0 Room Sun: 002 0:0 00 0006:: LoxemE .0000 00 09300800 .00 000030000 0306 009.0: 000 E .0005: 09.5.2 “0
0000:3800 SEE 0:0 5:0 .._n=2<w .00

0:00 203 00020050000 .92 .Eoom 000 0050000 00000 80 0000050800 65:0 :00. :0 620 29:05. r

 

 

41

00— on 000 mm 000 55 0.500 N2 300.
om e M: 0 S 8.2 2 8.8. 8098.5
ON 0 0.0 w N0 0 v.N0 f 00.0
5.0 N w.: 00 m0 3 m.No_ NN wOA
5.0 N w.: 00 0.0 m 5.0m 50 504
00 m N; 0 0.N N QMN 0 0010
5.0 N N.m 5 0.00 w 0.0m 50 00.0
3 m 02 m0 5.: a 5N_ N 004
msm 0 0.5 0 5.00 w fine 2 mm:
m.m 0 0.5 0 N.m 0 0w 2 484qu
5.0 N 0.5 0 0.0 m 0.50 3 N04
5.0 N 0.0 w 5.: m w.0N E 00.0

0:00.80 .4qu 000200 $00M 80800 900.40 @400 Boo—BE macaw

.0>< 00 .02 .0>< 00 .02 0.03.. 00 .02 £984 00 .02 omega

3:50.. 0080.5 405m 000 Eoom .0340 00 5.88006 .N.N 0300.

polymorphic markers. A higher percentage of Pstl markers (16.9%) was not mapped,
compared to 11.8% of EcoRl. Pstl markers had fewer counts in the categories of 0, 5 or 6
markers per 10 0M, and EcoRI markers had fewer counts in the categories of 1, 2, 3, and
4 markers per 10 0M, which indicated that Pstl markers tended to have a more even
distribution than EcoRI markers on the sugarbeet genetic map. This result was consistent

with previous reports in maize (Vuylsteke et al. 1999) and soybean (Young et al. 1999).

Distorted segregation of markers
The segregation of AF LP markers was determined based on a Chi-square ( 12 ) test, and

some markers showed highly distorted segregation ratios. Some distorted segregation
ratios favored the undetected alleles (absence), and some favored existent alleles. Overall,
forty-five out of 192 (23.4%) markers were segregation-distorted. For 29 unmapped

AF LP/SAMPL markers, 11 (37.9%) had significant segregation distortion, compared to
the percentage of 21.5% for the mapped markers. The Chi-square test showed that 18
markers had segregation distortion at 0.01 < P < 0.05 level, which were distributed on all
linkage groups except LG6 and LG9. Five markers were identified with distortion level
at 0.001 < P < 0.01. Nine markers out of 22 extremely distorted markers (distortion level
of P < 0.001) were not mapped, a percentage of 40.9%, which was much higher than
average, indicating distortion might interfere with mapping. For the 13 mapped markers
with P < 0.001 distortion, seven were located on LG3B, and two on LG8. All distorted

markers on LG7 were in a contiguous 5-marker segment, where all four markers had

42

paternal : maternal ratio less than 3: 1, which might be related to lethal or male-sterile or
self-incompatible gene from the wild beet parent.

The most significantly distorted linkage group was LG3B, where 11 out of 15
(73.3%) markers displayed segregation distortion. The distorted markers were evenly
distributed on LG3B, and all of these markers were present in the paternal parent except
one marker, AGOCIZI. All distorted markers on LG3B had a distortion in favor of
maternal allele. The gene locus might not only cause segregation distortion, but also
interfere with the linkage map construction. LG3A and LG3B could be linked with loose
LOD value (LOD = 3), however, under stringent criterion the influence of distortion on

LOD value separated these two short linkage groups.

Accessions’ phylogenetic relationship on Aphanomyces disease resistance

In addition to the mapping population and the parental plants, breeding accessions were
tested simultaneously, and their genetic relationship is shown in Figure 2.2. The
phylogenetic analysis showed that all individuals from the same accession were grouped
together. All plants from the mapping population were clustered together and separated
from other accessions analyzed. Four replicates of a control sample were tightly clustered
(data not shown). Wild beet was distant from domestic varieties, indicating potential
germplasm diversity resource for sugarbeet improvement was present in wild beet. F2
plants of two other crosses of sugarbeet by wild beet, Y04 (C869 x P1546409), and Y1 1
(SP6822 x P1640625), were positioned between wild beet and domestic varieties,

indicating partial introgression of wild germplasm.

43

An interesting relation concerning the Aphanomyces resistance in sugarbeet varieties
was observed in phylogenetic tree. Two commonly recognized Aphanomyces checks,
Edda and Betal6AB were genetically similar, suggesting they might originate from
similar germplasm sources. Two breeding lines with moderate Aphanomyces resistance,
SP6822 and EL48, were also tightly clustered. The maternal parent of mapping
population, C869 was distantly related from the Midwestern germplasm, such as EL48,
ACH185 and Betal6AB, and was separated from other sugarbeet accessions in the

phylogenetic tree.

Monogerm/Multigerm trait analysis

The association analysis for the monogerm trait showed relationship with 5 linkage
groups (Table 2.3). Logistic regression identified three markers significantly associated
with monogermy. Two markers, CATA383 and AGCC150 (located on LG7 and LG9,
respectively) were positively related to multigerm (data not shown), and the other,
SL1C185 (on LG8) appeared to be a dominant monogerm locus (or multigerm inhibitor),

where the presence of the marker resulted in fewer multigerm plants. The influence of

 

00:30.62 8350:0003... 000 00030000 wEaaaE 0:0 .80: 3000000 .0000 0:3 00 029003200 000003300 .N.N 8:me

.80 0.48%: :13

 

0:? 38.3: 5412.
885.. Sin.
5281:; nine

TI

 

 

 

 

I-u-d

 

Na N95 4 000000 881:; .410.

 

 

 

 

 

95m are.
.08 3.13 _ _

8.05888 58:8 9.0.
mouths—0:04.904 90:05 313 1

9:20.... 3 1|].

 

 

 

 

a: :0... xiii?

 

 

 

005.0002 oszD Rina
8055002043 3.5 N E

 

 

 

33% an:

 

 

 

30.5 81:. _ .

 

 

 

830 $13

 

0:220 .0580: 008 $13
$8 :. ma

 

23.4; 313
Nm nmccvoa x Nuwcmm Ndwv1__> 001mm

'1'

 

88 3418»

 

 

 

 

 

 

 

 

 

 

 

45

Table 2.3. AFLP marker Chi-square test for monogermy.

 

 

Marker Linkage Group Position on LG 12 P-value
ACAA170 LGl 7 4.27 0.0387*
SL1C129 LGl 13 4.86 0.0274*
AGOA321 LG2 3 4.58 0.0322*
CAOA357 LG2 4 5.14 0.0233*
CATA428 LG2 7 3.93 0.0472*
CTCC2OO LG4 24 4.27 0.0387*
CATA383 LG7 1 31.79 O***
ACOA261 LG7 3 6.85 0.0088**
CTCC166 LG7 4 6.95 0.0084**
ACAA258 LG7 5 5.83 0.0157*
ACAA233 LG7 6 6.51 0.0107*
ACAA276 LG7 7 6.51 0.0107*
AGOA136 LG7 8 6.32 0.0119*
ACAA264 LG7 9 4.65 0.031*
SL1C317 LG7 17 3.93 0.0472*
AGOC309 LG8 1 6.63 0.01*
SL1C185 LG8 2 6.32 0.0119*
SL1C294 LG8 22 4.58 0.0322*
SL1CO72 LG9 1 10.04 0.0015**
SL1C305 LG9 2 7.94 0.0048**
SL1C226 LG9 3 9.42 0.0021**
SL1C345 LG9 4 8.6 0.0033**
SL1C207 LG9 5 8.6 0.0033**
CTCCO69 LG9 6 7.55 0.006**
AGCC150 LG9 7 11.15 0.0008***
CATA438 LG9 8 6.21 0.0127*
CATA14O LG9 9 ' 5.62 0.0177*

 

Note: Significance levels: *2 0.01 < P < 0.05. “: 0.001 < P < 0.01. “*z P < 0.001
Position on LG: indicating the position of marker on linkage group

12 : Chi-square test value

CATA383 was very significant, with P <0.001 in all three types of Chi-square test. This

marker was probably tightly linked to the identified recessive ‘m’ locus, since this locus

accounted for most of the genetic variation. Altogether, these three loci accounted for

46

Fit-v.2

31.2% of the variation, and the Max-rescaled R2 was 38.3%. The P value for the model
was < 0.0001. The odds ratio for CATA383 was 16.21 (95% confidence interval, CI:
(6.34, 41.45)), for AGCC150 was 4.80 (95% CI: (1.828, 12617)) and for SL1C185 was
0.21 (95% CI: (0.06, 0.81)). The concordance rate was 74.0%, and discordance rate was
10.5%. If all three categories other than “All multigerm” were combined, it would result
in a ratio of 103: 37 (multigerm : monogerm), which was very close to 3:1, suggesting

that one major gene controlled the monogerm trait with the influence of modifying loci.

Germination time trait analysis

Germination time, measured by the time that radicle grows out of seed coat, is an
important factor of emergence. Varieties with short germination time normally have
better emergence. In association analysis, four QTLs (on LG3A, LG6, LG7, and LG9)
appeared to be correlated with germination time (Table 2.4). LG9 had a locus with P
value less than 0.01. Logistic analysis identified a model of four markers without
interaction. The model was significant with P value less than 0.0001, and it explained
19.5% of the variance (Max-rescaled R2 = 0.203). Three markers, SL1C318, CAOA494
and CATA084 (on LG7, LG3A and LG9, respectively) seemed to retard germination,
while ACAA137 (on LG6) accelerated germination. The odds ratio for CAOA494 was
0.407 (95% CI: (0.169, 0.983)), for SL1C318 was 0.323 (95% CI: (0.16, 0.651)), for
CATA084 was 0.449 (95% CI: (0.235, 0.857)); however, for ACAA137, odds ratio was
2.470 (95% CI: (1.187, 5.139)). The concordance rate was 62.0%, and the discordance
rate was 25.4%, indicating environmental factors also played an important role in

germination time.

47

 

Table 2.4. AFLP marker Chi-square test for germination time.

 

 

Marker Linkage group Position on LG 12 P-value

ACOA297 LG3A 2 13.78 0.0322*
CAOA494 LG3A 4 12.98 0.0432*
AGOA063 LG3A 9 12.81 0.046*
ACAA137 LG6 4 15.47 0.0169*
AGOAlOl LG6 5 12.81 0.046*
AGOA186 LG6 6 12.81 0.046*
SL1C318 LG7 2 14.13 0.0281*
ACAA261 LG7 13 16.21 0.0126*
AGCC150 LG9 7 13.22 0.0397*
CATA438 LG9 8 15.64 0.0158*
CATA14O LG9 9 16.28 0.0123*
CATA084 LG9 18 17.61 0.0073**

 

Note: Significance levels: *: 0.01 < P < 0.05. **: 0.001 < P < 0.01. "*z P < 0.001
Position on LG: indicating the position of marker on linkage group

12 : Chi-square test value

Genetic Male sterility
Genetic male sterile (GMS) was characterized by complete microspore abortion; anthers
were thin and black, and no viable pollen was observed. GMS of sugarbeet was
correlated with loci on LG2, LG3B, LG8 and LG9, and it exhibited an extremely strong
correlation with loci on LG2 and LG3B (P < 0.001 and P < 0.01, respectively, Table 2.5).
Logistic regression analyses identified three markers significantly correlated to
GMS. No interaction among markers was found, and the P value for the model was <
0.0001, with 16.1% of the variation explained (Max-rescaled R2 = 18.1%). Two markers,
SL1C103 (on LG2) and SL1C318 (on LG7) were positively correlated to male-sterile
when present. The odds ratio was 0.324 (95% CI: (0.140, 0.751)) and 0.315 (95% CI:

(0.137, 0.726)), respectively. The presence of marker AGOA225 (on LG3B) was

48

positively correlated to male-fertile plants. The odds ratio was 2.446 (95% CI: (1.209,

4.950)). The concordance rate was 62.3%, and the discordance rate was 22.1%.

Table 2.5. AFLP marker Chi-square test for genetic male-sterility.

 

 

Marker Linkage group Position on LG [2 P-value
TOCT188 LGl 5 11.95 0.0177*
TOCT322 LG2 2 13.27 0.01*
AGOA321 LG2 3 13.63 0.0086**
CAOA357 LG2 4 14.11 0.0069**
CTCC498 LG2 5 10.27 0.036*
ACAA123 LG2 8 14.53 0.0058**
ACAAO78 LG2 9 10.27 0.036*
ACOA149 LG2 10 14.75 0.0052**
ACAA215 LG2 11 15.02 0.0046**
SL1C103 LG2 12 19.19 0.0007***
AGOC106 LG3B 2 13.50 0.0091**
AGCC504 LG3B 3 15.78 0.0033**
SL1C109 LG3B 5 16.07 0.0029**
AGOA225 LG3B 6 17.47 0.0016**
ACAA111 LG3B 7 11.74 0.0194*
CAOA211 LG3B 8 11.89 0.0181*
AGOC083 LG3B 9 11.44 0.022*
CTCC294 LG3B 10 11.44 0.022*
AGOA266 LG3B 11 11.44 0.022*
CTCC097 LG3B 12 11.89 0.0181*
CATA322 LG3B 13 11.89 0.0181*
CATA304 LG3B 14 11.02 0.0262*
AGOA157 LG3B 15 9.53 0.0491*
SL1C318 LG7 2 9.71 0.0456*
ACOA501 LG8 5 11.54 0.0211*
CAOA335 LG8 7 11.00 0.0265*
CATA175 LG8 11 10.35 0.0348*
SL1C072 LG9 1 11.13 0.0251*
CATA438 LG9 8 11.71 0.0196*
CATA140 LG9 9 10.72 0.0299*

 

Note: Significance levels: ‘2 0.01 < P < 0.05. **: 0.001 < P < 0.01. ***: P < 0.001
Position on LG: indicating the position of marker on linkage group

12 : Chi-square test value

49

Discussion
Genetic mapping is an important step in QTL analysis. The results of this study provided
further evidence that AF LP was an important and reliable tool for genotyping. SAMPL
markers were also explored, and were found to be very useful. With a total of 163
AFLP/SAMPL markers, a genetic map of 507.1 cM was constructed with an average
interval of 3.30 cM. This provided the basis for QTL analysis of Aphanomyces disease

resistance, which will be discussed in chapter 3.

SAMPL markers

Selective amplification of microsatellite polymorphism loci, SAMPL (Paglia et al. 1998),
combines the advantage of AFLP and SSR. In animals, (AC/GT)n sequence is abundant,
but it is rarely observed in plants where AT repeats are common (Morgante and Olivieri
1993). However, Rae et al. (2000) reported that CA repeats were most common in
sugarbeet. Two SAMPL primers were tested in our experiments, which were SAMPLI
(CACACACACACACACTATAT) and SAMPL2 (GTGTGTGTGTGTGTGATAT).
Under the same conditions, SAMPL2 generated far fewer bands than SAMPLl. Among
the seven primer combinations tested for SAMPL2 primer, the highest number of bands
was 20, and SAMPL2 bands were weaker in intensity than those of SAMPL] DNA
fragments. Consequently, SAMPL2 was not used for genotyping in this study. In the
mapping population, 40 polymorphic bands were detected from one SAMPL] primer
combination. High polymorphism of SAMPL (83.3%) was partially due to the high
variability of simple sequence repeats. The surprisingly even distribution was observed

for mapped SAMPL markers. Except for one cluster on LG9 where 5 SAMPL markers

50

were adjacent to each other, all other 19 markers were relatively evenly distributed on all
9 linkage groups (Figure 2.1). SAMPLI markers were normally present in no more than
three locations on each linkage group. The majority of SAMPL markers were mapped to
locations near the centromere or telomere regions. Schmidt and HeslopHarrison (1996)
reported that centromeric regions of sugarbeet chromosomes contained arrays of
microsatellite sequences. For instance, microsatellites based on CAC or GATA repeats
were localized predominantly to centromeres, and TA repeats were localized near rRNA
genes. The SAMPLl primer contained exactly 7.5 CA repeats, followed by 2.5 TA
repeats, thus its composition might correlate the amplification products to centromeres or
rRNA genes. The SAMPL] sequence was within centromere, telomere or secondary
constriction regions in the genome, where short oli go repeats are highly abundant. Future
work to isolate these specific marker DNAs, followed by either sequencing or using them
as probes for in-situ hybridization, would better reveal the identity of SAMPLl markers.
Strong polymorphism demonstrated by SAMPL] indicated that SAMPL markers
could be efficiently applied to genetic mapping, especially for intra-species mapping
where polymorphisms might be less abundant. SAMPL marker results exhibited higher
polymorphism (83.3%) of amplified bands than the best AF LP primer combination
(EcoRI+CTC/MseI+CAG, which had 66.7% polymorphic bands). However, SAMPL
markers had higher unmapped marker percentage (20%), compared to Pstl’s 16.9%
(average) and EcoRI’s 11.8% (average) (Table 2.2). This might be related to more PCR
error when using simple sequence repeat primers. Roder et al. (1998) demonstrated that
microsatellites were not physically clustered in certain regions of wheat chromosomes,

implying the usefulness of SAMPL markers to cover highly repetitive regions. Overall,

51

SAMPL amplification was useful and efficient for selecting large number of molecular

markers.

Genetic map construction
Sugarbeet has nine pairs of chromosomes (2n = 2x = 18), thus nine linkage groups were
expected. However, one linkage group (LG3) was broken under stringent mapping
conditions, i.e. using LOD = 11 to remove any markers not tightly linked. Consequently,
the total map length summed up to 507 .1 cM, and the map was incomplete. If the map
construction was processed without the stringent filtering step (Le. “group 11 25”), a map
with nine linkage groups was obtained with a total length of 712.5 cM. The difference of
more than 200 cM indicated the importance of marker selection for map construction.
Different mapping populations may result in map length difference in that different
populations have different recombination frequencies, on which the genetic map is
constructed. Another influencing factor is the quality of marker and mapping procedures.
Schafer-Pregl et al. (1999) reported the longest map of sugarbeet, with a total of 1119
cM, which contained nine linkage groups. They used a very loose criterion, a LOD score
as low as 2.0 for grouping. Pillen et al. (1992) built a map of nine linkage groups (789
cM) using 115 markers. Barzen et al. (1995) generated a map of 815 cM with nine
linkage groups using 298 RF LP and RAPD markers. Schumacher et al. (1997) used 49
common markers to integrate maps from two populations, with a combined map length of
688 cM in 9 linkage groups. Nilsson et al. (1999) reported a map length of 567 cM.

Setiawan et al. (2000) used stringent conditions when constructing their map, and

52

obtained a map length of 744 cM. In comparison, the map length of this report (507.1
cM) is relatively short, and more markers are necessary for map saturation.

Most previous genetic maps of sugarbeet were constructed from crosses between
cultivated sugarbeets, where genetic variation was more limited than in this work. The
mapping population developed here contained germplasm from wild beet, B. vulgaris ssp.
maritima, which increased the genetic diversity of the mapping population. Wide
genomic differences increase the number of polymorphic markers, which may reveal
more recombination, thus, a longer map was expected. However, the genetic
recombination might be compromised due to the reduced homology of different genomic
regions, and this might counteract the recombination frequency gained by increased
polymorphism.

In this study, Pstl markers had slightly more even distribution than EcoRI markers.
However, a higher percentage of Pstl markers (16.9%) was not mapped, compared to the
11.8% of EcoRI. Similar phenomena have been observed in other species. For instance,
percentage of unmapped Pstl markers in onion was even much higher (43%), versus 12%
of un-mapped EcoRI markers (Van Heusden et al. 2000).

For 29 unmapped AFLP/SAMPL markers, 11 (37.9%) had significant segregation
distortion, compared to the percentage of 21.5% for the mapped markers. This distortion
percentage was a little higher than 15% reported by Pillen et al. (1993). Since the
mapping population was constructed from a cross between sugarbeet and wild beet,
slightly higher distortion level was not unusual. In addition, parental parents had GMS
and self-incompatible genes inside. If markers were linked to sterile or self-incompatible

genes, then those markers would be under-represented.

53

Nullisomic lines, commonly used in wheat chromosome identification, are not
available in sugarbeet due to the critical role of each chromosome in the genome.
Schondehnaier and Jung (1997) used RFLPs to correlate all 9 linkage groups with 9
different trisomic lines of sugarbeet. With related trisomic information, genetic linkage
groups can be assigned to specific chromosomes. The map constructed here was built
with AF LP and SAMPL markers only, thus its relationship with physical chromosome
map was unclear. If anchor RF LP markers were integrated into the map, or the specific
AF LP/SAMPL markers were linked to certain chromosomes, each linkage group could be
assigned to a specific chromosome.

An important application of a genetic map is for map-based cloning of specific
genes. Once markers co-segregating with QTLs were identified, the markers could be
physically located on the chromosome. Sugarbeet genome is 0.8 pg/C, which is about
758 Mbp (Arumuganathan and Earle 1991). Assuming the genetic map to be 700 cM,
each cM corresponds to about 1 Mbp. Thomas et al. (1995) used a high density AF LP
map (42, 000 markers) to locate two markers tightly co-segregating with C]? resistance
gene in tomato. These two markers were found to span a distance of 15.5 kb, which
covered the entire Cf9 gene.

Comparative mapping examines the genomic location of different genes or markers
across species or genera. Even though chromosome crossovers occur frequently between
or within species, related species are often found to share similar order of markers
(Feuillet and Keller 1999). Using common DNA markers across different genetic maps
makes comparative mapping feasible and valuable. Small contiguous regions of genes

(contigs) are often found to be conserved across different species, thus providing valuable

54

information for new research in one species by applying knowledge from a better—studied
species. Kowalski et al. (1994) compared genetic maps of Arabidopsis and Brassica,
which revealed 11 conserved regions between the two species that accounted for 24.6%
of Arabidopsis thaliana genome and 29.9% of the Brassica oleracea. Cavell et al. (1998)
compared a 7.5 Mbp region of A. thaliana with B. napus. The same order of markers was
conserved among 4 out of 6 duplicated copies in Brassica. The other 2 copies had a large
inversion mutation. Grant et al. (2000) discovered that many regions of synteny between
Arabidopsis and soybean were similar or identical. QTLs might exist in similar order
among similar species, or even across genera (Paterson et al. 1995). Homologous genes
from different species had been found to have similar fimctions (Lagercrantz et al. 1996).
Thus, research progress made in one species can be extended to other species.

Genetic linkage maps also provide insights into chromosomal organization and are
useful in map-based evolutionary studies. In general, genetic maps, together with physical
maps, will markedly enrich our understanding of gene functions and interactions at the

genome level.

Phylogenetic relationship of the accessions

The phylogenetic tree deduced from AF LP results confirmed previous results about the
relationship and diversity in US. sugarbeet germplasm, where RAPD markers were used
(McGrath et al. 1999). The wild beet was considerably distant from domestic accessions;
therefore, introducing Beta vulgaris ssp. maritima into sugarbeet can enhance germplasm
diversity of domestic varieties. Discrepancies were detected between AF LP and RAPD

phylogenetic trees. For instance, EL48 was distantly related to SP6822 in the RAPD tree,

55

whereas in the AF LP tree, EL48 was closely related to SP6822. This might be because
fewer domestic sugarbeet accessions were used in the AF LP map, and also might be due
to the instability of RAPD amplification. The phylogenetic analysis confirmed the

importance of introducing wild beet to diversify domestic germplasm.

56

REFERENCES

Ammuganathan K and Earle ED (1991) Nuclear DNA content of some important plant
species. Plant Molecular Biology Reporter 9: 21 1-215

Barzen E, Mechelke W, Ritter E, Schultekappert E, and Salamini F (1995) An extended
map of the sugar-beet genome containing RFLP and RAPD loci. Theor Appl Genet
90: 189-193

Cavell AC, Lydiate DJ, Parkin IAP, Dean C, and Trick M (1998) Collinearity between a
30-centimorgan segment of Arabidopsis thaliana chromosome 4 and duplicated
regions within the Brassica napus genome. Genome 41: 62-69

Cnops G, Denboer B, Gerats A, Vanmontagu M, and Vanlijsebettens M (1996)
Chromosome landing at the Arabidopsis tornad01 locus using an AF LP-based
strategy. Mol Gen Genet 253: 32-41

De los Reyes BG and McGrath JM (2003) Cultivar—specific seedling vigor and expression
of a putative oxalate oxidase germin-like protein in sugarbeet (Beta vulgaris L.).
Theor Appl Genet 107: 54-61

Ecker R, Barzilay A, and Osherenko E (1994) The genetic relations between length of
time to germination and seed dormancy in Lisianthus (Eustoma grandiflorum).
Euphytica 80: 125-128

Feuillet C and Keller B (1999) High gene density is conserved at syntenic loci of small
and large grass genomes. Proc Natl Acad Sci U S A 96: 8265-8270

Ghoulam C and Fares K (2001) Effect of salinity on seed germination and early seedling
growth of sugarbeet (Beta vulgaris L.). Seed Science and Technology 29: 357-364

Grant D, Cregan P, and Shoemaker RC (2000) Genome organization in dicots: genome
duplication in Arabidopsis and synteny between soybean and Arabidopsis. Proc Natl
Acad Sci U S A 97: 4168-4173

Gummerson RJ (1986) The effect of constant temperatures and osmotic potentials on the
germination of sugar-beet. Journal of Experimental Botany 37: 729-741

Hansen M, Kraft T, Christiansson M, and Nilsson NO (1999) Evaluation of AFLP in
Beta. Theor Appl Genet 98: 845-852

Herran A, Estioko L, Becker D, Rodriguez MJ B, Rohde W, and Ritter E (2000) Linkage
mapping and QTL analysis in coconut (Cocos nucifera L.). Theor Appl Gen 101:
292-300

57

 

Kowalski SP, Lan TH, F eldmann KA, and Paterson AH (1994) Comparative mapping of
Arabidopsis thaliana and Brassica oleracea chromosomes reveals islands of
conserved organization. Genetics 138: 499-510

Lagercrantz U, Putterill J, Coupland G, and Lydiate D (1996) Comparative mapping in
Arabidopsis and Brassica, fine scale genome collinearity and congruence of genes
controlling flowering time. Plant J 9: 13-20

Laporte V, Merdinoglu D, Saumitou-Laprade P, Butterlin G, Vernet P, and Cuguen J
(1998) Identification and mapping of RAPD and RFLP markers linked to a fertility
restorer gene for a new source of cytoplasmic male sterility in Beta vulgaris ssp. l
maritima. Theor Appl Gen 96: 989-996 E

McGrath J M, Derrico CA, Morales M, Copeland L0, and Christenson DR
(2000) Germination of sugarbeet (Beta vulgaris L.) seed submerged in hydrogen _
peroxide and water as a means to discriminate cultivar and seedlot vigor. Seed '3
Science and Technology 28: 607-620

 

McGrath JM, Derrico CA, and Yu Y (1999) Genetic diversity in selected, historical US
sugarbeet germplasm and Beta vulgaris ssp maritima. T heor Appl Genet 98: 968-976

Meksem K, Leister D, Peleman J, Zabeau M, Salamini F, and Gebhardt C (1995) A
hi gh-resolution map of the vicinity of the r1 locus on chromosome-v of potato based
on RF LP and AFLP markers. Mol Gen Genet 249: 74-81

Mglinets AV, Veprev SG, and Plyasova NK (1998) The alpha(3) mutation determining
male sterility in sugarbeet Beta vulgaris L. Genetika 34: 304-307

Morgante M and Olivieri AM (1993) PCR-amplified microsatellites as markers in plant
genetics. Plant J 3: 175-182

Mullis KB, Faloona F, Scharf SJ, Saiki RK, Horn GT, and Erlich H (1986) Specific
enzymatic amplification of DNA in vitro: The polymerase chain reaction. Cold Spring
Harbor Symp Quant Biol 51: 263-273

Nilsson NO, Hansen M, Panagopoulos AH, Tuvesson S, Ehlde M, Christiansson M,
Rading IM, Rissler M, and Kraft T (1999) QTL analysis of Cercospora leaf spot
resistance in sugarbeet. Plant Breeding 118: 327-334

Owen FV (1952) Mendelian male sterility in sugarbeets. Proc Am Soc Sugarbeet
Technol. 7: 371-376

Paglia G and Morgante M (1998) PCR-based multiplex DNA fingerprinting techniques
for the analysis of conifer genomes. Molecular Breeding 4: 173-177

Paglia GP, Olivieri AM, and Morgante M (1998) Towards second-generation STS
(sequence-tagged sites) linkage maps in conifers: a genetic map of Norway spruce

58

 

(Picea abies K). Mol Gen Genet 258: 466-478

Paterson AH, Lin YR, Li ZK, Schertz KF, Doebley J F, Pinson SRM, Liu SC, Stansel J W,
and Irvine J E (1995) Convergent domestication of cereal crops by independent
mutations at corresponding genetic-loci. Science 269: 1714-1718

Pillen K, Steinrucken G, Herrrnann RG, and Jung C (1993) An extended linkage map of
sugar-beet (Beta vulgaris L) including 9 putative lethal genes and the restorer gene-x.
Plant Breeding 111: 265-272

Pillen K, Steinrucken G, Wricke G, Hernnann RG, and Jung C (1992) A linkage map of
sugar-beet (Beta vulgaris L). Theor Appl Genet 84: 129-135

ll"-

Qi XQ, F ufa F, Sijtsma D, Niks RE, Lindhout P, and Stam P (2000) The evidence for
abundance of QTLs for partial resistance to Puccinia hordei on the barley genome.
Molecular Breeding 6: 1-9

F
a
'5

Rae SJ, Aldam C, Dominguez I, Hoebrechts M, Barnes SR, and Edwards KJ (2000)
Development and incorporation of microsatellite markers into the linkage map of
sugarbeet (Beta vulgaris spp.). Theor Appl Genet 100: 1240-1248

Roder MS, Korzun V, Gill BS, and Ganal MW (1998) The physical mapping of
microsatellite markers in wheat. Genome 41: 278-283

Rush CM and Vaughn KM (1993) Effect of irrigation, soil matric potential, and seed
priming on sugar-beet seed-germination and damping-off caused by Aphanomyces-
cochlioides. Phytopathology 83: 202-206

Sadeghian SY and Khodaii H (1998) Diallel cross analysis of seed germination traits in
sugarbeet. Euphytica 103: 259-263

Savitsky VF (1952) A genetic study of monogerm and multigerm characters in beets.
Proc Am Soc Sugarbeet Technol 7: 331-338

Schafer—Pregl R, Borchardt DC, Barzen E, Glass C, Mechelke W, Seitzer JF, and
Salamini F (1999) Localization of QTLs for tolerance to Cercospora beticola on
sugarbeet linkage groups. Theor Appl Genet 99: 829-836

Schmidt T and HeslopHarrison J S (1996) The physical and genomic organization of
microsatellites in sugarbeet. Proc Natl Acad Sci U S A 93: 8761-8765

Schondelmaier J and Jung C (1997) Chromosomal assignment of the nine linkage groups
of sugarbeet (Beta vulgaris L.) using primary trisomics. Theor Appl Genet 95: 590-
596

Schondelmaier J, Steinrucken G, and Jung C (1996) Integration of AF LP markers into a
linkage map of sugarbeet (Beta vulgaris L). Plant Breeding 115: 231-237

59

Schumacher K, Schondelmaier J, Barzen E, Steinrucken G, Borchardt D, Weber WE, and
Salamini CJF (1997) Combining different linkage maps in sugarbeet (Beta vulgaris
L) to make one map. Plant Breeding 116: 23-38

Setiawan A, Koch G, Barnes SR, and Jung C (2000) Mapping Quantitative Trait Loci
(QTLs) for resistance to Cercospora leaf spot disease (Cercospora beticola Sacc.) in
sugarbeet (Beta vulgaris L.). Theor Appl Genet 100: 1176-1182

Smith GA (1980) Sugarbeet. In Hybridization of Crop Plants (Edited by Fehr WR and
Hadley HH) Published by American Society of Agronomy-Crop Society of America.
Wisconsin, USA pp 601- 616

Thomas CM, Vos P, Zabeau M, Jones DA, Norcott KA, Chadwick BP, and Jones JDG
(1995) Identification of amplified restriction fragment polymorphism (AF LP) markers

tightly linked to the tomato cf-9 gene for resistance to Cladosporiumfidvum. Plant J
8: 785-794

Van de Peer Y, and De Wachter R (1994) TREECON for Windows: a software package
for the construction and drawing of evolutionary trees for the Microsoft Windows
environment. Comput Applic Biosci 10: 569-570

Van Heusden AW, Van Ooijen J W, Vrielink-Van Ginkel R, Verbeek WHJ, Wietsma
WA, and Kik C (2000) A genetic map of an interspecific cross in Allium based on
Amplified Fragment Length Polymorphism (AF LP TM) markers. Theor Appl Genet
100: 1 18-126

Voorrips RE, Jongerius MC, and Kanne HJ (1997) Mapping of two genes for resistance
to clubroot (Plasmodiophora Brassicae) in a population of doubled haploid lines of
Brassica oleracea by means of RFLP and AF LP markers. Theor Appl Genet 94: 75-
82

Vos P, Hogers R, Bleeker M, Reijans M, Vandelee T, Homes M, Frijters A, Pot J,
Peleman J, Kuiper M, and Zabeau M (1995) AFLP - a new technique for DNA-
fingerprinting. Nucleic Acids Res 23: 4407-4414

Vuylsteke M, Mank R, Antonise R, Bastiaans E, Senior ML, Stuber CW, Melchinger AE,
Lubberstedt T, Xia XC, Stam P, Zabeau M, and Kuiper M (1999) Two high-density
AF LP (R) linkage maps of Zea mays L.: analysis of distribution of AF LP markers.
Theor Appl Genet 99: 921-935

Wei F S, Gobelman-Wemer K, Morroll SM, Kurth J, Mao L, Wing R, Leister D, Schulze-
Lefert P, and Wise RP (1999) The mIa (powdery mildew) resistance cluster is
associated with three NBS-LRR gene families and suppressed recombination within a
240-kb DNA interval on chromosome 53 (lbs) of barley. Genetics 153: 1929-1948

Young WP, Schupp JM, and Keim P (1999) DNA methylation and AF LP marker
distribution in the soybean genome. Theor Appl Genet 99: 785-790

60

Chapter 3

Genetic analysis of Aphanomyces resistance in a sugarbeet x wild beet cross

Abstract
Aphanomyces disease has been a limitation to sugarbeet yield for decades. Different
management approaches have been suggested, such as early sowing, limited irrigation,
biological and chemical control with very limited improvement. Genetic resistance is the
most promising long-term strategy to provide protection and minimize adverse
environmental effects. Disease resistance was studied in an F2 population derived from a
cross between sugarbeet (C 869 with no known Aphanomyces resistance), and wild beet
(Beta vulgaris ssp. maritima, with reported high resistance). An AF LP/SAMPL genetic
map was constructed based on this population. Statistical analyses suggested two QTLs
on Linkage Groups 9 and 2, controlled the disease resistance. Both QTLs were significant
at the 5% level, and together explained 63.2% of the variance in relative AUDPC (Area
Under Disease Progress Curve, rAUDPC) scores. The identified QTLs should be useful

in breeding for Aphanomyces resistance in sugarbeet.

61

 

Introduction

Quantitative trait locus (QTL) mapping approaches include simple statistical analysis
(Edwards et al. 1987); interval mapping (IM) with maximum likelihood approach
(Lander and Botstein 1989); interval mapping using regression analysis (Haley and Knott
1992); composite interval mapping (CIM, Jansen 1993; Zeng 1993); and multiple interval

mapping (MIM, Kao et a1. 1999).

Simple statistical analysis

Simple statistical analysis can be accomplished by ANOVA (analysis of variance)
analysis, t-test, simple linear regression and association analysis. In association analysis,
no map information is required. Each marker is tested for expected genetic segregation
with Chi-square test, and markers significantly related to the trait can be identified.
However, no information on the relationship between markers is available. In addition,
the location, contribution, and effect (i. e. additive, dominant or recessive) of the markers
remain unknown. Other limitations include: 1) effects of QTLs are underestimated,
because if a QTL is located between markers, recombination between a QTL and marker
will reduce the strength of the association, 2) the analysis can not distinguish small effect
tightly linked to a QTL from large effect loosely linked to a QTL (Lander and Botstein
1989, Zeng 1994). Though simple statistical analysis is less powerful than QTL-
specialized approaches, the analysis is quick and simple, and almost all significant QTLs

can be detected by this method.

62

 

Interval mapping

Interval mapping (IM), proposed by Lander and Botstein (1989), is a method that
estimates the probability of a QTL being located between two markers (interval). IM first
used maximum likelihood approach, and more recently linear regression has been used to
attain similar results (Haley and Knott 1992, Martinez and Cumow 1992). Compared to
simple statistical analysis, 1M effectively detects QTLs, while constraining the scoring of
potential false positives. It also estimates phenotypic effects of QTLs and localizes QTLs
in the context of a linkage map (Lander and Botstein 1989).

The process of IM assumes that a QTL is located between two markers, and the
position is calculated at certain interval (e. g. 2 cM) starting from a flanking marker. The
maximized LOD scores are plotted against corresponding positions, either on or between
markers. If the peak LOD score exceeds a predetermined or permutation estimated
threshold, a putative QTL is proposed. Wang and Paterson (1994) concluded that QTLs

with large effects were more likely to be identified by this method.

Composite Interval Mapping (CIM)

Zeng (1993, 1994) and Jansen (1993, 1994) independently proposed using a combination
of multiple-regression with interval mapping, termed composite interval mapping. CIM
overcomes one limitation of 1M, 1’. e. marker information other than that localized to the
target interval markers is excluded. CIM first determines a target interval and then sets
partial regression coefficients for other selected background-markers (known as co-
factors, which can minimize the effect of QTLs in other regions of the genome)

simultaneously. When the effect in the target region is statistically significant, 3 QTL is

63

suggested. Theoretically, the genetic variance caused by QTLs other than the target are
minimized by the regression coefficients of markers outside of QTL effect window. As a
result, the background variance is reduced, and the statistical power to detect QTLs is
improved. The advantages of CIM include: 1) simplifying search processes for QTLs, 2)
improving QTL mapping precision due to reduced interference from other QTLs, and 3)
increasing the efficiency of QTL mapping (Zeng 1993, 1994). However, the number of
background markers is difficult to select. If background markers are too few, then CIM
results are similar to IM. Consequently, Jansen and Stam (1994) suggested selecting
background markers using stepwise regression, which was applied in the analyses
described in this chapter. However, CIM has its limitations: analysis can be affected by
uneven distribution of markers in the genome, and it is difficult to estimate the genetic
effect of linked QTLs. In addition, using closely-linked markers as background may

reduce statistical power (Zeng et al. 1999).

AFLP map based QTL analysis

With the AF LP mapping approach, no genetic background information is required.
Under the facilitation of software, AF LP marker information can be quickly converted
into a genetic linkage map. After QTL analyses, chromosome regions influencing the
trait can be identified, which serve as a platform for in-depth study. In principle, any
traits of interest, including quantitative traits that might be difficult to evaluate with other
molecular tools, can be efficiently evaluated with AF LP mapping and QTL analyses.
Nandi et al. (1997) found one major QTL together with several minor QTLs responsible

for submergence tolerance in rice. Marques et al. (1999) identified QTLs related to

mortality, adventitious rooting, sprouting and other traits in Eucalyptus. Bai et al. (1999)
used AFLP to identify one major QTL, which explained 53% of the total variation for
scab resistance in wheat. QTLs conferring resistance to late blight (Phytophthora
infestans) have been reported in potato (Collins et al. 1999), and resistance is
significantly correlated to plant vigor. In barley, QTLs conferring partial resistance to
leaf rust were reported by Qi et al. (1999).

AF LP has been successfully applied for detection of markers tightly linked to many
agronomic traits in beets. Two AF LP markers linked to the bolting gene (B) were
identified in Beta vulgaris ssp. maritima (Hansen et al. 2001). QTLs controlling
Cercospora resistance have been identified in sugarbeet. Nilsson et a1. ( 1999) reported 5
QTLs on 4 linkage groups related to Cercospora resistance, which accounted for 63% of
the total variation. Setiawan et al. (2000) identified 4 QTLs for Cercospora resistance
using a different mapping population. QTL analyses of sugar content and yield were

reported by Weber et al. (2000).

Marker assisted selection (MAS)

One important application of QTL analysis is marker-assisted selection (MAS), which is
useful for breeding and selection of quantitative traits (Lee 1995, Lande and Thompson
1990). Lee (1995) indicated that MAS might improve genetic gain by increasing gain
percentage and lowering evaluation costs. MAS may be useful by allowing selection in
non-target environments. Van Berloo and Stam (1998) showed that MAS had greater
advantage over traditional phenotypic selection when the heritability of the traits of

interest was relatively low (i.e., 0.1 to 0.3). Van Berloo and Stam (1999) compared MAS

65

1'3, _ w. 's-fij,

with conventional breeding methods to select for early or late flowering time in
Arabidopsis thaliana. They found phenotypic selection was less successful than MAS,
and RILs (recombinant inbred lines) selected by MAS had the highest number of
favorable QTL alleles.

Marker-assisted selection may be helpful for the introgression of useful genes from
wild germplasm (Tanksley and Nelson 1996; Bemacchi et al. 1998a,b), since wild
germplasm may carry genes that positively affect agronomic traits. Selecting markers
tightly linked to a particular trait will reduce the interference from non-targeted genomic
regions with undesirable effects (e.g. linkage drag). Thus, specific chromosome regions
can be efficiently introgressed into cultivated varieties with fewer selection cycles, and

the locus can be rapidly fixed in improved breeding lines. MAS has been used

extensively for disease resistance breeding. For instance, MAS has been used in selection

for soybean cyst nematode (SCN) resistance (Cregan et al. 1999). One SSR marker
Satt309, located 1-2 cM away from the gene rhgl for resistance to SCN, was used to
develop SCN resistant lines. DNA markers linked to a major QTL for powdery mildew
resistance greatly improved selection efficiency in grape (Dalbo et a1. 2000). MAS has

largely decreased the time and effort involved when compared to phenotypic selection.

Mapping population

F2 populations are efficient because of the large variation expressed, which renders more
statistical power to detect potential QTLs. On the contrary, backcross (BC) populations
are simpler, easier to analyze, however, contain less genetic information (e. g. fewer

recombination events) for QTL detection. Thus, F2 populations are more informative

66

‘cmfi

than BC, and fewer progenies are required to detect QTLs in F2 populations (Lander and
Botstein 1989). Recombinant inbred line (RIL) populations are also more informative
than BC populations, as the population variation is fixed among individuals. Furthermore,
using RILs can effectively reduce environmental variance. However, time and effort to

construct RILs might compromise their advantages.

Aphanomyces disease resistance

” ”ll

 

When genetic resistance follows the gene-for—gene interaction (e. g. qualitative resistance), .- ..
significant distinction between resistance and susceptibility are observed. In the case of

quantitative resistance, where multiple genes control resistance, complete separation of

resistance and susceptibility classes is difficult. QTL analysis for quantitative traits can

be performed, and potential trait-associated markers can be identified for use in plant

breeding.

Aphanomyces disease is mainly considered a seedling disease in sugarbeet, and also
causes a chronic root-rot disease at later stages. Domestic sugarbeet varieties have low to
moderate resistance against this pathogen. Resistance breeding has been undertaken for
more than fifty years with limited progress. The causal agent of the disease is
Aphanomyces cochlioides, a member of the oomycete family of plant pathogens. Among
many other pathogens that attack sugarbeet, Aphanomyces was the most destructive in
Red River Valley area (MN and ND), the major sugarbeet growing region in the US.
(Windels and Nabben-Schindler 1996). Under favorable conditions, A. cochlioides could

infect entire 2-5 week old sugarbeet field (Mckeen 1949).

67

Control of Aphanomyces has been attempted by biological, chemical, agrondmic and
breeding methods, but currently no effective methods have been developed to control this
disease (Payne and Williams 1990; Whipps et a1. 1993; Rush and Vaughn 1993).
Resistance breeding has been fundamental to control Aphanomyces in sugarbeet.
Bockstahler et al. (1950) showed resistance to Aphanomyces was heritable and dominant.
Resistance gained in his breeding program appears to have been selected from one or a

few germplasm sources, and the genetic background of resistance appeared to be narrow.

W.
.I- '5

Higher levels of resistance from wild beets have been described by Rush (USDA 1998),
which was crossed to domestic accessions. in this study. One problem in resistance
breeding is that no reliable screening method is available for resistance evaluation. Many
reports on Aphanomyces resistance are based on varieties’ yield performance in the fields
with Aphanomyces epidemics, but reproducibility is low due to interference from many
environmental factors (Downie et a1. 1952). Greenhouse tests for resistance were
reported (e.g. Schneider and Hogaboam 1983), but soil effect made the resistance
estimation unstable. Under laboratory conditions, box inoculation for Aphanomyces has
been successfully developed_(Chapter 1). Box inoculation has shown more consistent
evaluation of resistance in sugarbeet germplasm (Yu and McGrath, unpublished data).
With this new screening method, disease resistance of an F2 population segregating for
Aphanomyces resistance was evaluated. A genetic map was generated using AF LP

markers, which served to identify QTLs related to the disease resistance.

68

Materials and Methods
The female parent of the mapping population, C869, is a vigorous monogerm breeding
line, which carries the Mendelian dominant gene for self-fertility (Sf), and segregates for a
recessive gene conferring male sterility (ms). C869, bred in Salinas, CA (courtesy of Dr.
Robert Lewellen), has resistance to rhizomania, but has no known resistance to
Aphanomyces. Wild beet (P1540625, Beta vulgaris ssp. maritima) is multigerm, self-
sterile, with strong resistance to Aphanomyces (USDA 1998), and has been used as a F
pollen donor in the cross for the mapping population. The resistance source was I: -_
introduced to cultivated breeding line by crossing the wild beet to C869. The F1
progenies were shown to be segregating for Aphanomyces resistance, and a portion of
each F2 population derived from single self-fertilized plant was pre-screened with
Aphanomyces pot-inoculation test to assess disease reaction. One relatively resistant F2

population, Y03-384, was chosen and 145 F2 plants were used for genetic mapping.

Screening for Aphanomyces resistance destroys F2 individuals, so progeny tests (F 3
from selfed F2 plant) were used to estimate the average resistance of each F2 individual.
The evaluation method used was box-inoculation, which was described in Chapter 1. All
fertile plants with selfed seeds were examined for Aphanomyces resistance, only those
that yielded sufficient quantity of seeds for at least two replications were used in the QTL

analyses, a total number of 50. Eight additional samples had only one replication.

Resistance data was processed as follows: Disease symptoms were evaluated on day
3 (x1) and day 4 (x2) post inoculation (dpi). If the seedling was severely infected (e. g.

covered more than 2/3 of stem length), it’s given a score of 1. Few or no disease symptom

69

was assigned a score of 0. All other infected seedlings were scored as 0.5. The overall
disease severity was calculated via an Area Under Disease Progress Curve (AUDPC,

Shaner and Finney 197 7).

AUDPC= ZKxi +xi-l)/2](ti ‘ti-i)

i=1

where x, is disease severity at ith evaluation, and t, is the time for ith evaluation.

Since experimental conditions varied from box to box and from experiment to

 

experiment, the relative AUDPCs (rAUDPCs) against control variety, ACH185, within
the same box were calculated, and the average of rAUDPCs of F3 plants derived from

each F2 individual was used for QTL analyses.

rAUDPC = AUDPCF3 / AUDPCACH.85

Fluorescent AF LP (fAF LP), a semi-automated genotyping technique, was used to
generate AF LP and SAMPL markers for the mapping population, and AFLP/SAMPL
markers linked to Aphanomyces disease resistance were investigated.

Resistance was studied first by association analysis to find markers significantly
correlated to with Aphanomyces reaction. MAPMAKER/QTL and QTL Cartographer
(Basten et al. 1994; Basten et al. 2001) were then used for QTL analyses. The software
developed by Zeng’s group, QTL Cartographer, can effectively handle CIM (Zeng 1993),
and was used in QTL analyses in this research. QTLs were identified with CIM, with the
walking distance set to 0.5 cM, using model 6 with default values, and stepwise

regression. Threshold values were determined using at least 1000 permutations

70

(Churchill and Doerge 1994) (P value was set 0.05 initially, if the peak was above

threshold, then P=0.01 was tested with 5000 permutations).

 

7l

Results
A sugarbeet population segregating for Aphanomyces disease resistance was generated for
QTL analysis. Based on experimental results, all tested sugarbeets and wild beets had
limited resistance, especially under high disease pressure. Wild beets had better
performance than resistant lines of sugarbeet, suggesting wild beets might contribute a

new genetic resource to enhance Aphanomyces resistance in sugarbeet germplasm.

No interactions between the Aphanomyces resistance genes and environmental
factors were assumed. Relative AUDPCs (rAUDPCs) for 172 F3 test entries (from
selfed-F 2, each entry contained 10 to 12 seedlings) ranged from 0.25 to 2.25, with a mean
of 0.83 and a variance of 0.077. The majority of F2s had estimated rAUDPCs from 0.65
to 0.97 (mean of F3 samples from each F2 individual). The distribution of rAUDPCs in
F 3 populations was normal. The female parent C869 (eight replications) had a mean
value of 0.96, the variance was estimated at 0.046. Due to limited seed quantity and poor
emergence, the wild beet accession, P1540625 was only tested twice (two test entries)
with a mean of 0.82 (the smaller the value, the better disease resistance), and a variance
of 0.048. Since C869 was genetically more homogenous than wild beet, variance from
C869 was used to estimate the environmental variation. Thus, broad sense heritability for

rAUDPC was estimated by:

 

 

h2 =3 ,3 V” - V5 ~ 0.0771 — 0.0462

B = 40.1%
Vp VF2 0.0771

72

 

The narrow sense heritability was estimated as:

vpz (mp2 — 777p)2 = (0.8335 — (0.8152 + 0.9636)/2)2 = 0.0031
v. =v.,.vD . 0.0771- 0.0462 - 0.0031 = 0.0278

h2 _ V, ~ 0.0278

_—~ =36.1°/
N V, 0.0771 °

 

Linear regression of rAUDPCs with each marker identified 12 markers from two
linkage groups exhibiting significant correlation (5% level) to rAUDPCs (e.g. SL1C305,
SL1C226, SL1C345, SL1C207, CTCC069, AGCC150, CATA438, CATA140,
CATA091 and AGOC070 on LG9, and ACAA123, ACOA149 on LG2, from Chapter 2).
Aphanomyces resistance data was then analyzed with MAPMAKER/QTL, and one QTL
(AcrI : Aphanomyces gochlioz'des resistance 1) was identified after using “seq [all]”,
“scan” and “show peaks” commands with LOD score threshold 3.0 (Figure 3.1). This
QTL was located between CATA091 and CAOA221 (14 cM from CATA091, and 4.3 cM
from CAOA221). This single QTL explained 49.3% of the variance, and the LOD score
was 4.38. The acrI locus was considered as a recessive locus (Figure 3.1), in that the
recessive model explained almost same amount of variation as did the unconstrained
mode. Another candidate locus (ach, Figure 3.1) was also identified, near marker
CATA428P on LG2, and it explained 18.2% variance with LOD score of 2.13. The
model explained 63.2% rAUDPC variance with a LOD score 6.95, when two QTLs were
combined together.

QTL Cartographer CIM analysis results for F2 rAUDPCs (Figure 3.2 and Appendix
IH) identified two QTLs, one on LG9 (P < 0.01), and the other on LG2 (P < 0.05). The

QTLs were located at similar positions as of those in MAPMAKER/QTL analyses, and

73

 

the LOD score peak value of acr] was about 5.4. For acr2, the peak LOD value was 2.4
(Figure 3.2). No additional QTLs were identified.

Other evidence for the heritability of Aphanomyces resistance was that one F2 plant
survived the pot inoculation, and its progeny had average rAUDPC of 0.66, much better

than F2 average and both parents (data not shown).

74

 

Hooncmwzm E 8:328. emeafioegmw 5.2 mUmDD/t so mowing AHObov—mEam—Z ._.m Semi

1 1 1 03:25 I “Swag—How
use... o>mmmoooofi I 000a
” mUEmZmO
m m m m mmmm m mm m « mMum u m u mzms m m m mmm m

 

 

 

T 9n lllllllllllllllllllllll

r 3

 

 

28m Q04

 

88w DOA

75

a . LOD score Linkage group 2

1 - P=0.05

3 237 ‘/

 

 

 

 

 

 

179
0 d A A A [I A {A A A A 1PM
0 2 1 5 a 10 12 11 15
LOD score Linkage group 9
6
5 I

4.15 P=0.0\ A

52.4
A A .011

 

Figure 3.2. CIM results for rAUDPCs in response to Aphanomyces infection in sugarbeet.
A: indicates a marker, in the order described in the map (Figure 2.1)

76

Discussion

QTL analyses

Setiawan et al. (2000) compared three QTL software, MAPMAKER/QTL, QTL
Cartographer and PLABQTL, for QTL analyses of Cercospora resistance in sugarbeet.
MAPMAKER/QTL and QTL Cartographer use maximum likelihood estimation, while
PLABQTL uses regression approach. Similar QTL results were obtained with different
software. MAPMAKER/QTL was less powerful than other two, since it used the 1M
algorithm only.

In this study, interactions between environmental factors and genetic components
were not considered, since the entire population was cultivated under the same
conditions.

Both Mapmaker/QTL and QTL Cartographer generated similar LOD profiles for
Aphanomyces disease resistance. Mapmaker/QTL estimates genetic effect with four
models: free, additive, dominant and recessive models. However, it does not generate
threshold values for QTL prediction, which is available in QTL Cartographer. The
mapping method used in Mapmaker/QTL is 1M, while in QTL Cartographer, CIM is
available, which is beneficial to gaining more information.

QTL mapping can localize QTLs to several cMs or even ch to facilitate high
resolution mapping. The physical size of such regions is several hundreds to thousands of
kilo-basepairs (kbs), depending on species. If smaller regions must be defined, linkage
disequilibrium approaches should be applied, which can reduce QTL location range to

hundreds or even dozens of kbs, ideal for map-based cloning.

77

Aphanomyces resistance

Disease resistance can be grouped into two categories: 1) monogenic resistance, which
can render complete resistance against certain pathogen strains. 2) quantitative
resistance, which confers resistance to many races and is generally the consequence of
coordination of multiple resistant genes.

From Aphanomyces resistance analyses, narrow sense heritability of rAUDPC was
estimated to be 36.1%, and the broad sense heritability was 40.1%. However, the
estimation of variance was not very accurate as the environmental variance was evaluated
from V p] only instead of an average of VP], Vp2 and VF], due to the limitation of seed
quantities.

In addition to Cercospora resistance mapped with AF LP markers in sugarbeet,
resistance to rhizomania was mapped with RFLP markers (Barzen et al. 1992), and the
major resistance gene Rr] was located on linkage group IV. Barzen et al. (1992) also
identified monogerm gene m on linkage group IX, and the closest marker was 4.2 cM
away. Four genes related to nematode resistance were reported by Heller et al. (1996),
and located to the same positions through RF LP mapping in different F2 populations.
Bockstahler et al. (1950) observed that hybrids were more resistant to Aphanomyces than
the susceptible parent, and inbred plants tended to show either greater resistance or
greater susceptibility. Thus, they concluded that resistance was dominant.

The distribution of F2 rAUDPCs suggested that resistance to Aphanomyces derived
from wild beet was heritable, and the maternal parent C869 might be tolerant, since C869

was vigorous breeding line. Vigor might play a role in disease tolerance at initial disease

78

 

developing stage. The F2 population had many individuals with resistance levels higher
than both parents, which suggested that transgressive segregation might play a role in
resistance. The QTL on LG9 fit well with recessive effect, which was quite different
from the traditional dominance expectation. Recessive genes responsible for disease
resistance have been reported frequently. For instance, recessive xa13 conferred

resistance to bacterial blight in rice (Sanchez et al. 1999), er-I in Pisum was responsible
for resistance to powdery mildew (Timmennan et al. 1994), bc-Iz gene provided

resistance to bean common mosaic virus in common bean (Miklas et al. 2000), and the
edrl mutant conditioned resistance to powdery mildew in Arabidopsis (Frye and Innes
1998). Results from this study suggested the mapped QTL might be a new resistance
resource, which was probably different from previously reported dominant resistance
gene in sugarbeet.

The broad and narrow sense heritability estimates for rAUDPC, which reflected
Aphanomyces resistance, were 40.1% and 36.1%, respectively. The values were
moderate, which might result from small test entry size (each test entry contained only 10
to 12 seedlings). The AUDPC calculated was based on one entry not on each individual
seedling, thus a portion of residual variance was removed from the model, which resulted
in smaller total variance. Experimental error was reduced by comparing the true
resistance value to a common accession as control, and rAUDPC was used for QTL
analyses. Difference across experiments and inoculation boxes was minimized.
Estimated rAUDPC of each F2 plant was based on the average of at least two replications
(derived F3 seedlings), and the use of average greatly reduced environmental variation.

Addition of eight more data points (i. e. those F2 progenies with only one rAUDPC test

79

 

entry) did not change the LOD score pattern significantly. In contrast, it slightly reduced
the peak value of QTLs, presumably because the increased error variance cost more than
the gain of degrees of freedom.

Disease resistance often consists of two to five QTLs (reviewed by Young 1996).
Resistance genes or analogs might form clusters in plant genome (Shen et al. 1998,
Botella et al. 1997). Different QTLs might be detected with different QTL programs,
however, in our case, QT L Cartographer and MAPMAKER/QTL both identified the same 1

loci at similar positions (one locus each on LG2 and LG9), indicating that truly

 

 

 

significant QTLs were probably the same.

Two linkage groups, LG2 and LG9, were significantly correlated to Aphanomyces
disease susceptibility in sugarbeet. The most significant locus (acrI) was located within
an 18.3 cM interval on LG9, thus more markers were needed to cover the gap, and to
improve the location precision of the QTL. This single QTL explained 49.3% of the
rAUDPC variance, and the heritability was estimated to be 36%. Even though only one
QTL was identified on LG9, it could not preclude the possibility of multiple linked QTLs
on that linkage group, since the LOD profile of LG9 had a relatively broad maximum
(Figure 3.2). If alleles were fixed in the tested population, QTLs responsible for
significant variation within and between populations would not be detected. Thus, the
possibility of other QTLs undetected for Aphanomyces resistance in the mapping
population could not be excluded.

Overall, the heritability of Aphanomyces resistance seemed to be moderate, the
recessive locus on LG9 seemed to be important for resistance, and the QTLs identified

might be suitable for marker assisted breeding. Selection for resistance could be

80

effective, since genetic effect accounted for more than one third of rAUDPC variation.
(Note: the estimated heritability was for relative resistance, not absolute resistance value.)
Resistance genes are quite often found clustered in plant genomes (reviewed by
Harnmondkosack and Jones 1997). In lettuce, Dm5/8, DmlO and recessive gene plr for
downy mildew resistance, and Tu gene for turnip mosaic virus, were found located in a
6.4cM cluster (Witsenboer et al. 1995). Van der Voort et al. (2000) reported a resistance
gene cluster, which contained viral resistance (Rx2, Nb), Phytophthora resistance R],
nematode resistance Gpa and Grpl in potato. If similar phenomena existed in sugarbeet,
more resistance genes would be located to several clusters. Two QTLs on different
linkage groups were identified for Aphanomyces resistance in this study, however,
whether these loci co-localized with resistance genes for other sugarbeet diseases needed

further investigation.

Marker Assisted Selection (MAS)
Marker-assisted selection can be applied to indirectly select traits that are difficult or
expensive to evaluate. Markers co-segregating with resistance traits can greatly facilitate
the breeding process. Multiple QTL markers can be useful for complex traits, which are
normally controlled by polygenes. In addition to genetic information about individuals,
breeders can identify elite individuals for crossing, and further enhance selection
efficiency in breeding.

Once useful AF LP markers are identified, they can be converted into STS (Sequence
Tagged Site) markers. With STS marker analysis, the cost of detection can be

significantly reduced, and is applicable to large population with much less effort. For

81

instance, Meksem et al. (2001) converted AF LP markers linked to soybean cyst nematode
resistance (Rhg4) into STS markers to facilitate high throughput genotyping for marker
assisted breeding. von Stackelberg et al. (2003) converted AF LP markers linked to def
locus in pea into STS markers. Guo et al. (2003) validated the converted STS markers
related to F usarium head blight resistance in wheat. AF LP fragments mapped in one

population can serve as chromosome-specific markers if they also segregate in another

population. The transferability of AF LP markers has been demonstrated in potato (Li et [‘1‘
al. 1998). Thus, MAS will facilitate the selection of specific trait among different L_
breeding populations.

Due to the recombination between QTL and markers, using flanking markers on
both sides of a QTL may improve the precision of MAS in that it reduces recombination
interference. For a small genetic distance, such as 10 cM, the occurrence of double
crossovers can be ignored. Thus, with the presence of flanking markers, the linked QTL
is unlikely to be lost or mis-genotyped, and the time needed to fix major effect QTL in a
breeding population can be significantly reduced.

MAS is valuable for prediction, for instance, genetic male sterility. If a marker is
tightly linked to male-sterility, selection can be made at seedling stage ahead of
flowering, which is valuable for breeding purposes. MAS also makes it more efficient to

simultaneously select multiple traits, assuming that markers tightly linked to
corresponding major QTLs are available. Combined with traditional selection, MAS is
able to increase rate of genetic gain and reduce the interference from environmental

Variations.

82

In summary, factors that should be considered for DNA markers application in MAS
include: 1) Trait with moderate heritability would be difficult to study without the use of
linked molecular markers. 2) Markers should be tightly linked to the trait. 3) Ideally, two
flanking markers should be used for each QTL. 4) Markers for multiple QTL loci should
be considered for each trait. 5) Whenever possible, multiple QTLs should be used to
select multiple traits simultaneously.

Current MAS applications in breeding mainly focus on one or two traits, such as
disease resistance. With the progress of quantitative genetic research in crops, more and
more QTLs related to complex agronomic traits will be identified. MAS will evolve into
a complex genome-wide selection for a series of agronomic traits, and thus greatly

improve efficiency and reliability of breeding programs.

83

 

REFERENCES

Bai GH, Kolb FL, Shaner G, and Domier LL (1999) Amplified Fragment Length
Polymorphism markers linked to a major quantitative trait locus controlling scab
resistance in wheat. Phytopathology 89: 343-348

Barzen E, Mechelke W, Ritter E, Seitzer J F, and Salamini F (1992) RFLP markers for
sugar-beet breeding - chromosomal linkage maps and location of major genes for
rhizomania resistance, monogermy and hypocotyl color. Plant J 2: 601-611

Basten CJ, Weir BS, and Zeng ZB (1994) Zmap-a QTL Cartographer. In “Proceedings
of the 5th World Congress on Genetics Applied to Livestock Production: Computing
Strategies and Software”, Vol. 22. pp 65-66. Guelph, Ontario, Canada

Basten CJ, Weir BS, and Zeng ZB (2001) QTL Cartographer, Version 1.30. Department
of Statistics, North Carolina State University, Raleigh, NC.

Bemacchi D, Beck-Bunn T, Emmatty D, Eshed Y, Inai S, Lopez J, Petiard V, Sayama H,
Uhlig J, Zamir D, and Tanksley S (1998a) Advanced back-cross QTL analysis of
tomato. ii. Evaluation of near-isogenic lines carrying single-donor introgressions for
desirable wild QTL-alleles derived from Lycopersicon hirsutum and L
pimpinellifolium. Theor Appl Gen 97: 1191 -l 196

Bemacchi D, Beck-Bunn T, Eshed Y, Lopez J, Petiard V, Uhlig J, Zamir D, and Tanksley
S (1998b) Advanced backcross QTL analysis in tomato. 1. Identification of QTLs for

traits of agronomic importance from Lycopersicon hirsutum. Theor Appl Gen 97:
3 81 -3 97

Bockstahler HW, Hogaboam G], and Schneider CL (1950) Further studies on the
inheritance of black root resistance in sugarbeets. Proc Am Soc Sugarbeet Technol pp
104-107

Botella MA, Coleman MJ, Hughes DE, Nishimura MT, Jones JDG, and Somerville SC
(1997) Map positions of 47 Arabidopsis sequences with sequence similarity to disease
resistance genes. Plant J 12: 1197-1211

Churchill GA and Doerge RW (1994) Empirical threshold values for quantitative trait
mapping. Genetics 138: 963-971

Collins A, Milboume D, Ramsay L, Meyer R, Chatot-Balandras C, Oberhagemann P, De
Jong W, Gebhardt C, Bonnel E, and Waugh R (1999) QT L for field resistance to late

blight in potato are strongly correlated with maturity and vigour. Molecular Breeding
5: 387-398

Cregan PB, Mudge J, Fickus EW, Danesh D, Denny R, and Young ND ( 1999) Two

84

 

simple sequence repeat markers to select for soybean cyst nematode resistance
conditioned by the rth locus. Theor Appl Gen 99: 811-818

Dalbo MA, Ye GN, Weeden NF, Steinkellner H, Sefc KM, and Reisch B1 (2000) A gene
controlling sex in grapevines placed on a molecular marker-based genetic map.
Genome 43: 333-340

Downie AR, Doxtator CW, Swink JF, Ogeden D, Bowman RL, and Tanner JC (1952)
Reaction of sugarbeet strains to Aphanomyces cochlioides at three different locations.
Proc Am Soc Sugarbeet Technol. 7: 393-395

Edwards MD, Stuber CW, and Wendel JF (1987) Molecular—marker-facilitated
investigations of quantitative-trait loci in maize .1. numbers, genomic distribution and 11-
types of gene-action. Genetics 116: 113-125

 

Frye CA and Innes RW (1998) An Arabidopsis mutant with enhanced resistance to ,.
powdery mildew. Plant Cell 10: 947-956

Guo PG, Bai GH, and Shaner GE (2003) AFLP and STS tagging of a major QTL for
Fusarium head blight resistance in wheat. Theor Appl Gen 106: 1011-1017

Haley CS and Knott SA (1992) A simple regression method for mapping quantitative
trait loci in line crosses using flanking markers. Heredity 69: 315-324

Hammondkosack KB and Jones JDG (1997) Plant disease resistance genes. Annual
Review of Plant Physiology and Plant Molecular Biology 48: 575-607

Hansen M, Kraft T, Ganestarn S, Sall T, and Nilsson NO (2001) Linkage disequilibrium
mapping of the bolting gene in sea beet using AF LP markers. Genet Res 77: 61-66

Heller R, Schondelmaier J, Steinrucken G, and Jung C (1996) Genetic localization of
four genes for nematode (Heterodera schachtii Schm) resistance in sugarbeet (Beta
vulgaris L). Theor Appl Genet 92: 991-997

Jansen RC (1993) Interval mapping of multiple quantitative trait loci. Genetics 135:
205-211

Jansen RC (1994) Controlling the type-i and type-ii errors in mapping quantitative trait
loci. Genetics 138: 871-881

Jansen RC and Starn P (1994) High-resolution of quantitative traits into multiple loci via
interval mapping. Genetics 136: 1447-1455

Kao CH, Zeng ZB, and Teasdale RD (1999) Multiple interval mapping for quantitative
trait loci. Genetics 152: 1203-1216

Lande R and Thompson R (1990) Efficiency of marker-assisted selection in the

85

improvement of quantitative traits. Genetics 124: 743-756

Lander ES and Botstein D (1989) Mapping Mendelian factors underlying quantitative
traits using RFLP linkage maps. Genetics 121: 185-199

Lee M (1995) DNA markers and plant breeding programs. Advances in Agronomy. 55:
265-344

Li X, Van Eck HJ, Van Der Voort J, Huigen DJ, Starn P, and Jacobsen E (1998)
Autotetraploids and genetic mapping using common AF LP markers: the r2 allele

conferring resistance to Phytophthora infestans mapped on potato chromosome 4.
Theor Appl Gen 96: 1121-1128

Marques CM, Vasquez-K001 J, Carocha VJ, Ferreira JG, O'malley DM, Liu BH, and
Sederoff R (1999) Genetic dissection of vegetative propagation traits in Eucalyptus
tereticornis and E globulus. Theor Appl Gen 99: 936-946

 

Martinez O and Cumow RN (1992) Estimating the locations and the sizes of the effects
of quantitative trait loci using flanking markers. Theor Appl Gen 85: 480-488

McKeen WE (1949) A study of sugarbeet rootrot in southern Ontario. Can J Res 27:
284-311

Meksem K, Ruben E, Hyten D, Triwitayakom K, and Lightfoot DA (2001) Conversion of
AF LP bands into high-throughput DNA markers. Mol Genet Genomics 265: 207-214

Miklas PN, Larsen RC, Riley R, and Kelly JD (2000) Potential marker-assisted selection
for bc-1(2) resistance to bean common mosaic potyvirus in common bean. Euphytica
1 16: 21 1-219

Nandi S, Subudhi PK, Senadhira D, Manigbas NL, Senmandi S, and Huang N (1997)
Mapping QTLs for submergence tolerance in rice by AF LP analysis and selective
genotyping. Mol Gen Gen255: 1-8

Nilsson NO, Hansen M, Panagopoulos AH, Tuvesson S, Ehlde M, Christiansson M,
Rading IM, Rissler M, and Kraft T (1999) QTL analysis of Cercospora leaf spot
resistance in sugarbeet. Plant Breeding 118: 327-334

Payne PA and Williams GE (1990) Hymexazol treatment of sugar-beet seed to control
seedling disease caused by Pythium spp and Aphanomyces-cochlioides. Crop
Protection 9: 371-377

Qi X, J iang G, Chen W, Niks RE, Starn P, and Lindhout P (1999) Isolate-specific QTLs
for partial resistance to Puccinia hordei in barley. Theor Appl Gen 99: 877-884

Rush CM and Vaughn KM (1993) Effect of irrigation, soil matric potential, and seed
priming on sugar-beet seed-germination and damping-off caused by Aphanomyces-

86

cochlioides. Phytopathology 83: 202-206

Sanchez AC, Ilag LL, Yang D, Brar DS, Ausubel F, Khush GS, Yano M, Saskai T, Li Z,
and Huang N (1999) Genetic and physical mapping of xa13, a recessive bacterial
blight resistance gene in rice. Theor Appl Gen 98: 1022-1028

Schneider CL and Hogaboam GJ (1983) Evaluation of sugarbeet breeding lines in
greenhouse test for resistance to Aphanomyces cochlioides. J Am Soc Sugarbeet Tech
22: 101-107

Setiawan A, Koch G, Barnes SR, and Jung C (2000) Mapping quantitative trait loci
(QTLs) for resistance to Cercospora leaf spot disease (Cercospora beticola Sacc.) in
sugarbeet (Beta vulgaris L.). Theor Appl Gen 100: 1176-1182

Shaner G and F inney RE (1977) The effect of nitrogen fertilization on the expression of
slow-mildewing resistance in Knox wheat. Phytopathology 67: 1051-1056

Shen KA, Meyers BC, Islam-Faridi MN, Chin DB, Stelly DM, and Michelmore RW
(1998) Resistance gene candidates identified by PCR with degenerate oligonucleotide

primers map to clusters of resistance genes in lettuce. Mol Plant Microbe Interact l 1:
815-823

Tanksley SD and Nelson J C (1996) Advanced backcross QTL analysis: a method for the
simultaneous discovery and transfer of valuable QTLs from unadapted germplasm into
elite breeding lines. Theor Appl Gen '92: 191-203

Timmerrnan GM, Frew TJ, Weeden NF, Miller AL, and Goulden DS (1994) Linkage
analysis of er-I , a recessive Pisum sativum gene for resistance to powdery mildew
fungus (Elysiphe pisi DC). Theor Appl Gen 88: 1050-1055

USDA (1998) ARS, National Genetic Resources Program. Germplasm Resources
Information Network - (GRIN). [Online Database] National Germplasm Resources
Laboratory, Beltsville, Maryland. Available: http://www.ars-grin.gov/var/apache/cgi-
bin/npgs/html/close.pl?49074+1

Van Berloo R and Stam P (1998) Marker-assisted selection in autogarnous RIL
populations: a simulation study. Theor Appl Gen 96: 147-154

Van Berloo R and Stam P (1999) Comparison between marker-assisted selection and

phenotypical selection in a set of Arabidopsis thaliana recombinant inbred lines.
Theor Appl Gen 98: 113-118 ‘

Van Der Voort JR, Van Der Vossen E, Bakker E, Overmars H, Van Zandroort P, Hutten
R, Lankhorst RK, and Bakker J (2000) Two additive QTLs conferring broad-spectrum

resistance in potato to Globodera pallida are localized on resistance gene clusters.
Theor Appl Gen 101: 1122-1130

87

von Stackelberg M, Lindemann S, Menke M, Riesselmann S, and J acobsen HJ (2003)
Identification of AF LP and STS markers closely linked to the def locus in pea. Theor
Appl Gen 106: 1293-1299

Wang GL and Paterson AH (1994) Assessment of DNA pooling strategies for mapping
of QTLs. Theor Appl Gen 88: 355-361

Weber WE, Borchardt DC, and Koch G (2000) Marker analysis for quantitative traits in
sugarbeet. Plant Breeding 119(2): 97-106

Windels CE and Nabben-Schindler DJ (1996) Limitations of a greenhouse assay for

determining potential of Aphanomyces root rot in sugarbeet fields. J Sugarbeet Res
33: 1-13

Whipps JM, Mcquilken MP, and Budge SP (1993) Use of fungal antagonists for
biocontrol of damping-off and Sclerotinia diseases. Pesticide Science 37: 309-313

Witsenboer H, Kesseli RV, F ortin MG, Stanghellini M, and Michelmore RW (1995)
Sources and genetic-structure of a cluster of genes for resistance to 3 pathogens in
lettuce. Theor Appl Gen 91: 178-188

Young ND (1996) QTL mapping and quantitative disease resistance in plants. Annual
Review of Phytopathology 34: 479-501

Zeng ZB (1993) Theoretical basis for separation-of multiple linked gene effects in
mapping quantitative trait loci. Proc Natl Acad Sci U S A 90: 10972-10976

Zeng ZB (1994) Precision mapping of quantitative trait loci. Genetics 136: 1457-1468

Zeng ZB, Kao CH, and Basten CJ (1999) Estimating the genetic architecture of
quantitative traits. Genet Res 74: 279 - 289

88

Summary

To introduce possible disease resistance resources from wild beet to sugarbeet,
susceptible sugarbeet (C869) was crossed to Aphanomyces resistant wild beet (P1540625).
A genetic map was built in the segregating F2 population. The map contained 163

AF LP/SAMPL markers on 9 linkage groups for a total length of 507.1 cM. A novel soil-
less zoospore inoculation protocol was developed in this study, which greatly reduced
time, space, labor and plant materials input. This disease resistance evaluation system
distinguished resistance and susceptibility reasonably well. The lab test results were
proved to be significantly (P=0.015) correlated with field evaluation results. Through
QTL analysis, two Aphanomyces resistance related QTLs were identified on two linkage
groups (LG2 and LG9). Both QTLs (acrI iand acr2) are statistically significant at 5%
level. When combined, the QTLs explained 63.2% variance of rAUDPC. In addition to
Aphanomyces resistance, several economically important traits were also investigated.
Two linkage groups (LG7 and LG9) were highly related to monogerm/multigerm trait.
Four linkage groups (LG3A, LG6, LG7 and LG9) controlled germination time. Genetic
male sterility was mainly manipulated by genes on LG2 and LG3B. Highly correlated

DNA markers were identified for each trait, and could be used for selection.

Future work to be done is listed as follows: 1) Generate more DNA markers and
build a marker-condensed map. 2) Use anchored DNA markers to build the connection
between linkage groups and chromosomes. Thus, information collected in other
populations can be utilized. 3) Develop more DNA markers in the interesting regions
(such as acrI and acr2 regions), and narrow down the potential QTL location. The final

purpose will be using DNA markers to facilitate selection in a marker assisted breeding

89

program. 4) Continue tracking the segregating F3 and higher population, and collect more
disease resistance data. 5) Test and validate the Aphanomyces resistance related QTLs

(acrl and acr2) in other populations.

90

Appendices

 

91

APPENDIX I

AFLP technique

DNA isolation and quantification

Plant DNA was isolated with automatic DNA isolation instrument (AutoGen850 PI-502,
AutoGen Inc. Framingharn, MA). Procedures recommended by manufacturer (Plant
tissue DNA protocol) were followed with slight modifications, i. e. PVP

(pol yvinylpyrrolidone) was added to the extraction solution, with a final concentration of
1% (w/v). PVP served to bind polysaccharides (Doyle and Doyle 1990), which were
abundant in sugarbeet. In brief, young leaves from plants were pulverized in liquid
nitrogen using mortar and pestle. An appropriate amount of leaf powder (about 0.2g) was
mixed with 500 1.11 Tris/EDTA/NaCl/CTAB (AutoGen Inc. Framingham, MA), with
addition of PVP, and incubated at 65 °C for 30 min prior to DNA isolation. Isolated
DNA was dissolved in 100 pl low TE buffer (10 mM Tris, 0.1 mM EDTA, pH 8.0).

DNA was quantified with a Hoefer DyNAQuant200 Fluorometer.

Adapter preparation

To anneal synthesized EcoRI or Pstl adapter oligos, solution was prepared as follows
(Table A.1):

Table A1. EcoRl or Pstl adapter preparation for AFLP

 

 

§o_mponents Stock Concentration Volume (1.1L)
EcoRI / Pstl forward 100 nmol/mL 10 11L
EcoRl / Pstl reverse 100 nmol/mL 10 1.1L
Millipore water 80 1.1L

Total 100 1.1L

 

For Msel adapter oligos, solution was prepared as follows (Table A2):

Table A2. Msel adapter preparation for AFLP

 

 

anrponents Stock Concentration Volume (11L)
Msel forward 100 nmol/mL 50 11L

Msel reverse 100 nrnol/mL 50 11L

marl 100 pL

 

92

Annealing was run on a PCR machine (MJ Research, Inc. Waltham, MA), and the
program used was: 95 °C 5 min, followed by 100 cycles of 90 °C 2 see (with 1 sec
extension and —0.7 °C per cycle). The annealed adapter was stored at -20 °C for several
weeks.

DNA restriction

Both EcoRI and Pstl were used for AF LP genotyping. These enzymes recognize 6 bp
restriction endonuclease recognition sequences. The restriction site for EcoRI is
GAATTC, and for Pstl it is CTGCAG. Seventy-five ng of DNA from each sample was
digested. The components for one digestion reaction (RL I) are shown in Table A.3.

Table A.3. Digestion reaction (RL l) for AFLP and SAMPL

 

 

Components Stock Volume for one reaction
Concentration“

DNA 0.5 pl (75ng)

10x RL buffer“ 3.0 pl

Restriction enzyme 10 U/pl 0.5 pl Pstl (5U) or 0.5 pl EcoRI (5U)

Msel 4 U/pl 1.0 pl (4U)

100x BSA 10 pg/pl 0.3 pl

Millipore water 24.7 pl

Total 30 pl

 

Note: *210x RL (restriction and ligation) buffer for EcoRI is: 0.1 M Tris-acetate, 0.1 M
Mg(CH3COO)2, 0.5 M KCH3COO, pH 7.5. 10x RL for Pstl is: 0.5 M Tris-HCI, 0.1 M
MgCl2, 0.5 M NaCI and 5mM DTT (dithiothreitol) pH 7.5.
#: concentration of stock solution.

DNA was digested at 37 °C for 2 hr.

Ligation

Following digestion, ligation components, referred as RL II (Table A.4), were added and
the solution was continuously incubated at 37 °C for 3 hours to over night, and heated to
70 °C for 15 min to inactivate enzymes.

93

Table A4. Digestion reaction (RL II) for AFLP and SAMPL

 

 

Components Stock Concentration Volume
EcoRI /Pstl adapter <= 10 nmol/rrrl 1.0 pl
Msel adapter <= 50 nmol/ml 1.0 pl
ATP 10 mM 1.0 pl
T4 ligase 1 U/pl 1.0 pl
10x RI. buffer 10x 1.0 pl
RL 1 mixture 30.0 pl
Millipore water 5.0 pl
Total 40.0 pl

 

Note: *210x RL (restriction and ligation) buffer for EcoRI is: 0.1 M Tris-acetate, 0.1 M
Mg(CH3COO)2. 0.5 M KCH3COO, pH 7.5. 10x RL for Pstl is: 0.5 M Tris-HCI, 0.1 M
MgCl2, 0.5 M NaCl and 5mM DTT (dithiothreitol) pH 7.5.

#: concentration of stock solution.

 

Preamplification

Digested and adapter-linked DNA was pre-amplified in the buffer shown in Table A.5.

Table A5. Preamplification reaction for AFLP

 

 

 

Components Stock Concentration Volume
DNA from RL II 1.0 pl
10X PCR buffer 10X 1.0 pl
EcoRI+l /Pstl+1 10 nmol/ml 0.5 pl
Msel+l 10 nmol/ml 0.5 pl
dNTP 1.25 mM 1.5 pl
Taq 5 U/pl 0.1 pl
Millipore water 5.4 pl
Total 10.0 pl

 

Note: EcoRI+1l Pstl+1 stands for preamplification primer with 1 extra selective nucleotide.
10X PCR buffer contains 15 mM MgCl2.

The preamplification reaction for Pst+TC and SAMPL markers was slightly
different from that of EcoRI AF LP, where no selective nucleotide was used in
preamplification. All PCR amplification was completed on FTC-200 machine (MJ
Research, Inc. Waltham, MA).

The PCR program used for AF LP preamplification is shown in Table A6 :

94

Table A.6. PCR program for AFLP pre-amplification

1 72 °C 1 min. 30 sec.

2 94 °C 1 rrrin. 30 sec.

3 94 °C 30 sec.

4 56 °C 30 sec.

5 72 °C 1 rrrin

6 Goto 3, 23 times

7 72 °C for 10 min. I
8 4 °C forever
9 end

 

After PCR, the reaction (10 pl) was diluted by adding 40 pl low TE (10 mM Tris,
0.1 mM EDTA, pH 7.5), mixed well and used for amplification.

In practice, a master mixture was prepared. An aliquot of the mixture was dispensed
into each tube. After adding DNA, tubes were gently vortexed and centrifuged briefly,
and subjected to PCR reaction. No oil was used to cover solution, and the heated lid of
the thennocycler was used to reduce condensation in the tube.

Amplification

Reaction components for AF LP amplification are listed in the following (Table A.7):

Table A.7. Amplification reaction for AFLP

 

 

 

Components Stock Concentration Volume
DNA from preanrplification 1.0 pl
10x Taq buffer 10x 1.0 pl
Labeled primer 1 nmol/ml 0.5 p1
Mse+3 10 nmol/ml 1.0 pl
DNTP 1.25 mM 1.5 pl
Taq 5 U/pl 0.1 pl
Millipore water 4.9 pl
Total 10.0 pl

 

95

A “touchdown” PCR program was used for AF LP amplification (Table A8).

Table A.8. PCR program for AFLP amplification

1. 94 °C 1 min. 30 sec.
2 94 °C 30 sec.
3. 65 °C 30 sec.
4 72°C 1min.

-0.7 °C per cycle
5. Goto 2, 12 times
6. 94 °C 30 sec.
7. 56°C 30 sec.
8. 72°C 1min.
9. Goto 6, 20 times
10. 72 °C 5 rrrin.
11. 4 °C forever
12. end

After amplification, 5 pl loading dye (Li-Cor, Lincoln, NE) was added to each
sample. The mixture was loaded into the gel as soon as possible, or stored in dark, since
the fluorescence faded quickly under light. Just before loading, samples were denatured
at 95 °C for 3 min, then placed on ice for 15 min.

Gel preparation

The genotyping system used here was Li-Cor NEN® 4200 (Lincoln, NE), which was able
to scan two labeling dyes, IRD700 and IRD800 concurrently.

Plates were set up following manufacturer’s instructions. In brief, a set of 25 cm
sequencing plates were carefully cleaned, and air-dried. Regions below comb area of the
front plate were coated with saline solution (100 pl binding saline from Li-Cor, mixed
with 100 pl 10% acetic acid). The 0.2 mm spacers and paper shark-tooth combs (64
teeth, 0.18 mm) were used. The 30 ml gel solution used was prepared as recommended by
manufacturer.

Electrophoresis conditions

Gel solution was mixed well by stirring, then filtered through a glass microfibre filter
(25mm Catalogue# 1820025. Whatman®, Whatrnan International Ltd. England),

96

degassed under vacuum (optional), and 200 pl 10% freshly made APS (Ammonium
persulfate, (NH4)2S2Og ) and 20 pl TEMED (N ,N,N’,N’-Tetramethylethylene-diamine)
was added. Solution was swirled gently, and poured into the fixed plates. After 1 hour
polymerization, the casting comb was removed, and the well region was rinsed
thoroughly with millipore water. 0.7X TBE running buffer was added.

e-SEQ software was used to control the LiCor instrument. The “electrophoresis
conditions” of e-SEQ program were set as follows: 1200 V, 30 W, 25 mA, 45° C, scan
speed 3, run time: 4 hours. Scanner was focused and the gel was pre-run for 30 minutes.
After pre-run, plate set was taken out of instrument; running buffer above gel was
decanted; small pieces of gel in the well region were removed. Then, well region was
dried with KimWipe paper, and shark-tooth paper comb was carefully inserted until the
front 0.2 mm of teeth was in the gel. After plate set was put back on Li-Cor Sequencer
and refilled with running buffer, wells were flushed with a 20 cc syringe to remove air
bubbles and urea precipitate. Finally, one pl denatured amplification solution was loaded
to each well, and gel was run for 4 hours. Data was collected automatically.

Molecular weight marker
Two sequencing reactions (one labeled with IRD700, the other labeled with IRD800)

from sequence-known samples were used as molecular weight marker. About 0.8 to 1.0
pl of each denatured marker reaction was loaded for each gel.

97

Appendix II Primers used In sugarbeet AFLP

 

 

Primer Sequence (5' TO 3') Length(bps)
Adapters:

ECORFWD CTC GTA GAC TGC GTA CC 17
ECORREV AAT TGG TAC GCA GTC TAC 18
MseFWD GAC GAT GAG TCC TGA G 16
MseREV TAC TCA GGA CTC AT 14
PstFWD ACG CAG TCT ACG AGT GCA G 19
PstREV CTC GTA GAC TGC GTA CC 17
Preamplification (+1) primers:

ECOR+A GAC TGC GTA CCA ATT CA 17
ECOR+C GAC TGC GTA CCA ATT CC 17
Mse+A GAT GAG TCC TGA GTA AA 17
Mse+C GAT GAG TCC TGA GTA AC 17
Pst+A CAG TCT ACG AGT GCA GA 17
Pst+C CAG TCT ACG AGT GCA GC 17
Amplification primers (+3) for Mse:

Mse+CAG GAT GAG TCC TGA GTA ACA G 19
Mse+CCA GAT GAG TCC TGA GTA ACC A 19
Mse+CTT GAT GAG TCC TGA GTA ACT T 19
Mse+CGA GAT GAG TCC TGA GTA ACG A 19
Mse+CCT GAT GAG TCC TGA GTA ACC T 19
Mse+CAT GAT GAG TCC TGA GTA ACA T 19
Mse+CGG GAT GAG TCC TGA GTA ACG G 19
Mse+ACG GAT GAG TCC TGA GTA AAC G 19
Mse+AGC GAT GAG TCC TGA GTA AAG C 19
Mse+ACC GAT GAG TCC TGA GTA AAC C 19
Mse+ACA GAT GAG TCC TGA GTA AAC A 19
Fluorescence labeled primers:

ECO+ACA GAC TGC GTA CCA ATT CAC A 19
ECO+AGC GAC TGC GTA CCA ATT CAG C 19
ECO+CAT GAC TGC GTA CCA ATT CCA T 19
ECO+CTC GAC TGC GTA CCA ATT CCT C 19
Pst+AC CAG TCT ACG AGT GCA GAC 18
Pst+CA CAG TCT ACG AGT GCA GCA 18
Pst+CAT CAG TCT ACG AGT GCA GCA T 19
Pst+AG CAG TCT ACG AGT GCA GAG 18
Pst+TC CAG TCT ACG AGT GCA GTC 18
SAMPLl CACACACACACACACTATAT 20
SAMPL2 GTGTGTGTGTGTGTGATAT 19

 

Note: FWD: abbreviation for forward primer. REV: for reverse primer.

98

CV 23ng E 8885 86.9883... 9 8888 E momgfi co.“ 3:52 25 fieoEew< .18. 2:me

2 .N 2:me 92: 05 E 63886 86.8 65 E coo—.88 a 82865 ”<

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

89:58. 6 832:. 8.0 be 838. 26 .8863. E 5:83..

 

 

 

 

 

 

E z; 5 ma mm m a m e a e mm Mm mm a fl 9 we a w a
2 1 11“ r1 - ID . 1 q 4 r 4 -< 1 il- - 1 B
a. w 9
L .F
k .w
um EN 1m
Q04
4 9:on ems—EA «mm L mm 8on owe—:5 Q04 B
a n m m w m N 2 AN WN we ma 12 m m m a
P p b p v 1 3 p P 41 d 1 1 b1 1*P1 1 0 11 h b
fit a 2. .o
.F F
m
mm. — m . r
.m 62 .m
8 8.2» 88.5 no.— .4 _ sea 8.33 83 .3.

 

 

99

C .N Rammed 88 05 E concomoe .568 05 E .8858 a 828%.: ”<
ANV 28883 E 8885 86.98.88? 8 8882 E momgfi com 33.0.8 3:0 38928 .N.< Semi

 

Nag 1 IR l Nmm l lfrtrxr
1 F

“LO
(:1

seawmeflfiao emvammfiaee

b p p u _ p . -
<<< a < {<38 < D a 1 111 a. «it... 4
I

 

 

 

 

 

 

 

 

 

 

 

 

100

 

 

 

 

 

 

.N
.m .
a; 8m
m as.» 8833 98 .3 a as.» 88.5 o3
m...s.._.._etmm...~. 688811.881. a
4 4 ID .q < 4 4‘ 4 4 ‘ d. 8 <1 41 .
.F
.m
E .m .3
6 868 ”was: no; .4 m 868 888: 83

 

 

 

 

 

   

1111111111111111