Resistance gene analogs are conserved and clustered in soybean. Proc Nat Acad Sci USA 93: 11746-11750

Abstract

Sequences of cloned resistance genes from a wide range of plant taxa reveal significant similarities in sequence homology and structural motifs. This is observed among genes conferring resistance to viral, bacterial, and fungal pathogens. In this study, oligonucleotide primers designed for conserved sequences from coding regions of disease resistance genes N (tobacco), RPS2 (Arabidopsis) and L6 (flax) were used to amplify related sequences from soybean [Glycine max (L.) Merr.]. Sequencing of amplification products indicated that at least nine classes of resistance gene analogs (RGAs) were detected. Genetic mapping of members of these classes located them to eight different linkage groups. Several RGA loci mapped near known resistance genes. A bacterial artificial chromosome library of soybean DNA was screened using primers and probes specific for eight RGA classes and clones were identified containing sequences unique to seven classes. Individual bacterial artificial chromosomes contained 2-10 members of single RGA classes. Clustering and sequence similarity of members of RGA classes suggests a common process in their evolution. Our data indicate that it may be possible to use sequence homologies from conserved motifs of cloned resistance genes to identify candidate resistance loci from widely diverse plant taxa.

Full text

Proc.
Natl.
Acad.
Sci.
USA
Vol.
93,
pp.
11746-11750,
October
1996
Genetics
Resistance
gene
analogs
are
conserved
and
clustered
in
soybean
(disease/mapping/multi-gene
families/evolution)
VLADIMIR
KANAZIN*,
LAURA
FREDRICK
MAREK*,
AND
RANDY
C.
SHOEMAKER*tt
*Department
of
Agronomy,
and
tField
Crops
Research
Unit,
Agricultural
Research
Service,
U.S.
Department
of
Agriculture,
Iowa
State
University,
Ames,
IA
50011
Communicated
by
Steven
D.
Tanksley,
Cornell
University,
Ithaca,
NY,
August
19,
1996
(received
for
review
May
3,
1996)
ABSTRACT
Sequences
of
cloned
resistance
genes
from
a
wide
range
of
plant
taxa
reveal
significant
similarities
in
sequence
homology
and
structural
motifs.
This
is
observed
among
genes
conferring
resistance
to
viral,
bacterial,
and
fungal
pathogens.
In
this
study,
oligonucleotide
primers
de-
signed
for
conserved
sequences
from
coding
regions
of
disease
resistance
genes
N
(tobacco),
RPS2
(Arabidopsis)
and
L6
(flax)
were
used
to
amplify
related
sequences
from
soybean
[Glycine
max
(L.)
Merr.].
Sequencing
of
amplification
products
indi-
cated
that
at
least
nine
classes
of
resistance
gene
analogs
(RGAs)
were
detected.
Genetic
mapping
of
members
of
these
classes
located
them
to
eight
different
linkage
groups.
Several
RGA
loci
mapped
near
known
resistance
genes.
A
bacterial
artificial
chromosome
library
of
soybean
DNA
was
screened
using
primers
and
probes
specific
for
eight
RGA
classes
and
clones
were
identified
containing
sequences
unique
to
seven
classes.
Individual
bacterial
artificial
chromosomes
con-
tained
2-10
members
of
single
RGA
classes.
Clustering
and
sequence
similarity
of
members
of
RGA
classes
suggests
a
common
process
in
their
evolution.
Our
data
indicate
that
it
may
be
possible
to
use
sequence
homologies
from
conserved
motifs
of
cloned
resistance
genes
to
identify
candidate
resis-
tance
loci
from
widely
diverse
plant
taxa.
The
sequences
of
cloned
plant
disease
resistance
genes
show
that
the
majority,
whether
conferring
resistance
to
viral,
bacterial,
or
fungal
pathogens,
contain
similar
sequences
and
structural
motifs.
The
Arabidopsis
genes,
RPS2
(1,
2)
and
RPM1
(3),
conferring
resistance
to
the
bacterial
blight
patho-
gen
Pseudomonas
syringae,
the
tobacco
gene
N
(4,
5),
confer-
ring
resistance
to
tobacco
mosaic
virus,
the
rice
gene
Xa2l
(6),
conferring
resistance
to
Xanthomonas
oryzae,
the
flax
gene
L6
(7),
conferring
resistance
to
a
rust
fungus,
and
the
tomato
gene
Cf-9
(8),
conferring
resistance
to
the
fungal
pathogen
Clas-
doporium
fulvum,
all
contain
leucine-rich
repeats
that
encode
protein
motifs
often
associated
with
protein-protein
interac-
tions
or
ligand
binding
(9).
Motifs
for
a
conserved
nucleotide
binding
site
are
also
found
in
the
RPS2,
RPM],
N,
and
L6
coding
region.
It
has
been
proposed
that
the
similarities
among
resistance
genes
may
make
it
possible
to
take
advantage
of
sequence
homologies
to
identify
other
resistance
genes
(10).
Southern
hybridization
with
a
clone
from
the
Pto
locus
of
tomato
to
genomic
DNA
of
six
dicotyledonous
and
five
monocotyledonous
species
detected
homologs
in
all
species
(11)
and
suggested
that
gene
families
of
these
homologs
may
exist,
similar
to
that
observed
in
tomato.
Similarly,
conserved
motifs
from
receptor
proteins
have
been
used
to
identify
multigene
families
of
odorant
receptors
in
rat
(12)
and
con-
served
motifs
from
regulatory
proteins
have
been
used
to
identify
other
homeobox
proteins
in
Xenopus
(13).
Although
not
all
resistance
genes
have
been
demonstrated
to
reside
in
clusters,
tight
linkage
associations
of
many
resis-
tance
genes
have
been
well
established.
Genetic
linkage
of
resistance
genes
has
been
reported
in
maize
for
the
Rpl
cluster
(14-16),
in
barley
for
the
Mla
cluster
(17-20),
in
lettuce
for
a
wide
range
of
Dm
loci
(21,
22),
in
oat
for
the
Pc
cluster
(23),
and
in
flax
for
the
L
gene
cluster
(24-26).
Clustering
of
resistance
genes
suggests
that
a
common
genetic
mechanism
has
been
involved
in
their
evolution
(16).
In
this
paper
we
demonstrate
that
the
nucleotide
conserva-
tion
observed
within
disease
resistance
genes
cloned
from
widely
diverse
taxa
can
be
used
to
advantage
to
isolate
sequences
with
strikingly
similar
motifs
from
a
species
from
which
no
disease
resistance
genes
have
yet
been
cloned.
We
demonstrate
that
resistance
gene
analogs
(RGAs)
exhibit
microclustering
in
the
genome,
that
clusters
of
RGAs
contain
only
members
of
the
same
family,
and
that
mapping
of
RGA
sequences
can
place
genetic
markers
in
close
proximity
to
known
resistance
genes.
MATERIALS
AND
METHODS
Nucleic
Acid
Manipulations.
Soybean
genomic
DNA
was
prepared
as
described
in
Dellaporta
et
al.
(27).
RNA
was
prepared
according
to
Sambrook
et
al.
(28).
Electrophoresis,
blotting,
and
hybridization
was
done
using
standard
techniques
(28).
Restriction
enzyme
digestions
were
conducted
using
conditions
recommended
by
the
manufacturers.
PCR
Amplification,
Cloning,
Sequence
Analysis.
Regions
of
amino
acid
identity
in
the
N
gene
from
tobacco,
the
RPS2
gene
from
Arabidopsis,
and
the
L6
gene
from
flax
were
used
to
design
degenerate
primers.
Primer
LM638
was
designed
from
the
conserved
P-loop
sequence.
Primer
LM637
was
designed
from
a
second
region
of
amino
acid
identity
which
in
the
RPS2
protein
is
proposed
to
reside
in
a
transmembrane
region.
Primer
sequences
were
as
follows:
LM638,
5'-GGIGGIGTIG-
GIAAIACIAC-3',
and
LM637,
5'
-A(A/G)IGCTA(A/
G)IGGIA(A/G)ICC-3'.
PCR
was
performed
in
a
total
volume
of
100
,ul,
with
a
3-min
initial
denaturation
step
at
94°C
followed
by
35
cycles
as
follows:
94°C
for
1
min,
45°C
for
30
sec,
and
72°C
for
30
sec.
PCR
products
were
cloned
into
the
pGEM-T
vector
(Promega).
Clones
were
sequenced
using
the
Applied
Biosystems
model
377
PRISM
automated
sequencer,
or
manually
using
the
Sequenase
DNA
sequencing
kit
(United
States
Biochemical).
Multiple
sequences
were
obtained
for
each
RGA
class.
DNA
sequence
analysis
was
carried
out
with
the
DNASIS
(Hitachi)
and
GCG
(University
of
Wisconsin
Genetics
Com-
puter
Group,
Madison)
sequence
analysis
packages.
Align-
ment
of
sequences
was
done
using
the
PILEUP
function
of
the
Genetics
Computer
Group.
Phylogenetic
analysis
of
amino
Abbreviations:
RGA,
resistance
gene
analog;
BAC,
bacterial
artificial
chromosome.
Data
deposition:
The
sequences
reported
in
this
paper
have
been
deposited
in
the
GenBank
data
base
(accessions
nos.
U55803-
U55812).
*To
whom
reprint
requests
should
be
addressed.
11746
The
publication
costs
of
this
article
were
defrayed
in
part
by
page
charge
payment.
This
article
must
therefore
be
hereby
marked
"advertisement"
in
accordance
with
18
U.S.C.
§1734
solely
to
indicate
this
fact.

Proc.
Natl.
Acad.
Sci.
USA
93
(1996)
11747
acid
sequences
was
performed
using
MEGA
version
1.0
(29).
The
tobacco
N
sequence
was
used
as
an
outgroup.
Genetic
Mapping
of
RGAs.
RGA
sequences
were
mapped
in
a
G.
max
x
G.
soja
population
containing
56
individuals
(30)
using
MAPMAKER
(31).
Additionally,
one
soybean
bacterial
artificial
chromosome
(BAC)
was
mapped
in
a
G.
max
x
G.
max
population
containing
196
individuals
and
in
which
the
disease
resistance
genes
Rps2
and
Rmd,
and
the
Rhizobium
responsive
gene
Rj2,
were
mapped
(32).
Data
sets
from
linkage
group
J
from
both
populations
were
combined
and
an
inte-
grated
map
was
constructed
using
the
computer
program
JOINMAP
(33).
The
Kosambi
mapping
function
was
selected
and
a
minimum
logarithim
of
odds
(lod)
score
of
3
required
for
a
two-point
linkage
to
be
included
in
analyses.
The
order
of
"anchored
loci"
defined
by
JOINMAP
output
agreed
with
the
published
order.
Therefore,
specification
of
fixed
sequences
was
not
necessary.
BAC
Library
Construction
and
Screening.
The
soybean
BAC
library
was
constructed
as
described
(34)
from
megabase-
sized
DNA
isolated
from
the
soybean
cultivar
Williams
82.
The
4-5
genome
equivalent
library
contains
-40,000
individually
picked
clones
with
an
average
insert
size
of
150
kb
and
it
is
stored
in
384-well
microtiter
dishes.
Three-dimensional
BAC
pools
for
PCR
screening
were
set
up
using
entire
plates
and
individual
rows
or
columns
from
groups
of
20
plates.
Plasmid
DNA
was
purified
from
each
pool
by
standard
alkaline
lysis
midiprep
technique
(28).
Class-specific
primers
were
designed
for
each
RGA
class
(Table
1).
PCR
amplification
was
done
as
described
above
using
an
annealing
temperature
of
55°C.
Microtiter
dishes
were
replicated
onto
nylon
membranes
(Zeta-
Probe
GT;
Bio-Rad)
and
selected
membranes
were
used
to
confirm
results
of
PCR-based
screening.
RESULTS
Cloning
and
Sequence
Analysis.
PCR
amplification
of
soy-
bean
genomic
DNA
using
degenerate
RGA
primers
resulted
in
a
product
that
appeared
to
be
a
large,
single
band
on
a
1%
agarose
gel.
However,
digestion
of
this
product
with
several
restriction
enzymes
recognizing
4-bp
sites
resulted
in
many
fragments,
whose
sum
was
much
greater
than
the
molecular
weight
of
the
original
PCR
product.
The
presence
of
a
heterogeneous
PCR
product
suggested
the
involvement
of
a
multigene
family.
These
PCR
products
were
cloned
and
-450
clones
were
analyzed.
The
clones
were
grouped
into
nine
classes
which
did
not
cross-hybridize
under
stringent
hybrid-
ization
conditions
[0.1
x
standard
saline
citrate
(SSC)/0.1%
SDS/60°C
wash].
Clones
representing
each
class
were
hybrid-
Table
1.
Sequences
of
class-specific
RGA
primer
pairs
RGA
class
ized
to
Southern
blots
of
soybean
genomic
DNA
digested
with
various
restriction
enzymes
to
identify
polymorphisms
useful
for
genetic
mapping
of
RGAs
and
to
estimate
copy
number.
Four
classes
were
used
as
probes
on
Northern
blots
containing
total
RNA
from
different
soybean
organs
(leaves,
stems,
and
roots).
Lower
levels
of
RGA
message
were
detected
in
stems
and
roots
compared
with
that
observed
from
leaves
(Fig.
1).
Differential
expression
or
accumulation
was
not
observed
in
these
tissues
as
a
result
of
wounding
or
Phytophthora
inocula-
tion
(data
not
shown).
One
to
five
clones
from
each
class
were
sequenced,
and
the
deduced
amino
acid
sequences
of
repre-
sentative
clones
from
each
class
are
shown
in
Fig.
2.
Among
sequenced
clones
only
class
4
sequences
showed
heterogeneity
(two
groups).
Alignment
of
the
amino
acid
sequences
estab-
lished
that
the
cloned
RGAs
contain
two
additional
conserved
nucleotide
binding
protein
domains
also
present
in
N,
RPS2,
and
L6.
A
search
of
GenBank
using
the
BLAST
algorithm
revealed
that
the
RGA
sequences
were most
similar
to
the
L6,
N,
and
RPS2
gene
products.
Remote
similarity,
limited
to
the
conserved
motifs
was
found
to
other
P-loop
containing
pro-
teins
(myosin
heavy
chain
homolog,
Arabidopsis;
ATPase,
Plasmodium).
Pairwise
comparisons
between
different
classes
and
be-
tween
each
class
and
the
homologous
N
and/or
L6
sequences
revealed
that
amino
acid
identities
ranged
from
30
to
66%;
similarities
ranged
from
50
to
75%.
Clone
RGA9
is
likely
to
be
a
pseudogene
because
it
contains
multiple
stop
codons
and
frame-shift
mutations.
However
it
did
show
strong
similarity
with
RGA4
(77%).
Genetic
mapping
placed
RGA9
and
RGA4
at
one
position
on
linkage
group
H.
Class
4
contained
two
subclasses,
a
and
b,
which
showed
88%
amino
acid
identity.
These
subclasses
differed
at
the
nucleotide
level
but
were
not
distinguishable
by
Southern
blot
analysis.
Classes
3
and
7
were
66%
identical
at
the
amino
acid
level;
classes
5
and
8
were
51%
identical.
CA
0
0
-o
1.5
Sequences,
5'
-*
3'
RGA1
AGTTTATAATT(C/T)(C/G)ATTGCT-
ACTACGATTCAAGACGTCCT
RGA2
AGTTTATAATT(C/T)(C/G)ATTGCT-
CACACGGTTTAAAATTCTCA
RGA3
AGTTTATAATT(C/T)(C/G)ATTGCT-
CTCTCGATTCAAAATATCAT
RGA4
TGTTACTGCTTTGTTTGGTA-
TACATCATGTGTTACCTCT
RGA5
TGCTAGAAAAGTCTATGAAG-
TCAATCATTTCTTTGCACAA
RGA6
AGCCAAAGCCATCTACAGT-
AACTACATTTCTTGCAAGT
RGA7
AGTTTATAATT(C/T)(C/G)ATTGCT-
CCGAAGCATAAGTTGCTG
RGA8
AGCGAGAGTTGTATTTAAG-
AGCCACTTTTGACAACTGC
1.5
kb
FIG.
1.
Northern
blot
analysis
of
total
soybean
RNA
from
leaves,
stems,
and
roots.
(Upper)
Ethidium
bromide-stained
RNA
gel.
(Bot-
tom)
Results
of
hybridization
of
a
Northern
blot
of
this
gel
with
an
RGA3
probe.
Hybridizations
using
RGA1,
RGA2,
and
RGA4
probes
yielded
similar
results.
Genetics:
Kanazin
et
al.

Proc.
Natl.
Acad.
Sci.
USA
93
(1996)
RGA4a
RGA4b
RGA3
RGA7
RGA1
RGA2
RGA5
RGA8
RGA6
N
L6
RGA4a
RGA4b
RGA3
RGA7
RGA1
RGA2
RGA5
RGA8
RGA6
N
L6
ol
.20
.40
.60
.80
.100
GGVGKTTLVTALFGKISP...
..QYDARCFIDDLNKKCGN
..
FGAISAQKQLLSLALHQG
......................
NMBIHNLSHGTMLIRTRL
GGVGKTTLVTALFGKISP
......
QYDARCFIDDLNKYCGD
..
FGATSAQKQLLCQALNQG.
.
...........
NMEIHNLSHGTMLVRTRL
GGVGKTTLAVAVYNSIAG
......
HFEASCFLENVRETSN.K.KGLQHLQSILLSKTVGE
.......................
KKIRLTNWREGIPIIKHRL
GGVGKTTLAAAVYNSIAD
......
HFFALCFLENVRETSK.K.HGIQHLQSNLLSETVGE
.H..
.IGVKQGISIIQHRL
GGVGKTTLALAVYNLIAL
......
HFDESCFLQNVREESN.K.HGLKHLQSIILSRLLGE
.......................
RDINLTSWQEGASMIQHRL
GGVGKTTIAREVYNLIAD
......
QFEWLCFLDNVRENSI.K.HGLVHLQKTLLSKTIGE
.......................
SSIRLGSVHEGIPIIKHRF
GGVGKTTIARKVYEAIKG
......
DFDVSCFLENIREVS
..
KTNGLVHIQK.ELSNLGVSCFLEKCKTNGLVPIVEEVFRDQLRIVDFDNLHDGRMIIANSL
GGVGKTTLARVVFKKIRN
......
RFDISCFLENVREISQ.NCDGMLSLQGKLLSHM
........................K
RMLKIQNLDEGKSIIGGIL
GGVGKTTSAKAIYSQIHR
......
RFMDKSFIESIRSVCETDSKGHVHLQEQLLSDVLNT
........................
KVRVHSIGMGTTIIEKRL
GGVGKTTIARAIFDTLLGRMDSSYQFDGACFLKDIKE
....
NRGMHSLQNALLSELLRE
.
.......................
KANYNNEEDGKHQMASRL
GGIGKTTTAKAVYNRISSC
......
FDCCCFIDNIRET.
.QEDGVVVLQKKLVSEILRI
................
DSGSVGFNNDSGGRKTIKERV
.120
.140
CHLKTLIVLDNVDQVEQLENLAL.HREYLGEGSRTIIISTNMHILRNYGVD.
.160
.180
.200
.GVYNVQLLNIWKALQLLCKFAFKSDD.IVKGYEEVTHDVLKYVNGLPLA
.KVYNVQLLKDKALQLLCKKAFKSDD.IEKGYEEVTYDVLKYVNGLPLA
KQKKVLLILDDVDEHKHLQAI.IGSPDWFGCGSRVIITTRNEHLLALHNV...ITYKVRELNERnALQLLTQKAFELEKEVDSSYNDILNRALIYASGLPLA
QQQKILLILDDVDKREQLQAL.AGRPDLFGLGSRVIITTRDKQLLACHGVE..RTYEVNELNEEHALELLSWKAFKLEK.VDPFYKDVLNRAATYASGLPLA
QRZKVLLILDDVDKRQQLKAII.VGRPDWFGPGSRVIITTRDKHILKYHEVE..RTYEVKVLNQSAALQLLKYNAFKREKN.DPSYEDVLNRVVTYASGLPLA
LLRKVLLVVDDVDDPDQLQAI.VGGTDWFGSASSVIITTRDKHLLTCHGVT..STYEVDGLNKEKALKLLSGTAFKIDK.VDPCYMRILNRVVTYASGLPLA
SNKKVLLVLDDVSELSQLENLA.GKQEWFGPGSRVIITTRDKHLLKTHGVH..LTCKARALAQNEALQLICLKAIKRDQPKKG.YLNLCKEMIECARGLPLA
FNNNVLLVLDDVNDIRQLENFSVNDQKWLGPGSRIIIITRDHEVLRSHGTV..ESYKIDLLNSGESLQLFSQKAFKRDQPLEH.ILQLSKVAVQQAGGLPLA
SGKRVLIVLDDVNEIGQLENL.CGNCEWFGQGSVIIITTRDVGLLNLFKVD..YVYXMEENDENESLELFCLHAFGEPNPRED.FNELARNVVAYCGGLPLA
RSRRIVLDDIDNKDHYLEYLAGDIDWFGNGSRIIITTRDKHLIEKNDI....
IYEVTALPDHESIQLFKQHAFGKEVPNEN.FERLSLEVVNYAKGLPLA
SRFKILVVLDDVDEKFKFEDMLGSPKDFISQ.SRFIITSRSMRVLGTLNENQCKLYEVGSMSKPRSLELFSKA.FKKNTP.PSYYETLANDVVDTTA
P
FIG.
2.
Alignment
of
the
deduced
amino
acid
sequences
of
RGAs
from
soybean.
Arrows
indicate
location
of
PCR
primers
used
to
amplify
RGA
sequences.
Underlined
regions
correspond
to
additional
conserved
domains.
Dotted
regions
indicate
gaps
in
sequence
introduced
to
maximize
alignment.
The
tobacco
N
and
the
flax
L6
amino
acid
sequences
are
included
for
comparison.
Amino
acid
sequences
were
aligned
and
neighbor-joining
analysis
resulted
in
the
production
of
a
phylogenetic
tree
with
the
nine
RGA
sequences
divided
into
several
subclades
(Fig.
3).
Mapping
of
RGA
Loci.
Genetic
mapping
of
members
of
the
different
classes
placed
them
on
8
of
the
26
linkage
groups
comprising
the
soybean
genetic
map
(Fig.
4).
The
RGA
loci
mapped
both
singly
and
in
clusters
and
were
located
on
several
of
the
linkage
groups
to
which
known
disease
resistance
genes
have
been
mapped.
A
class
6
RGA
mapped
to
linkage
group
Ni
near
Rps1
(35).
A
large
cluster
of
RGAs
representing
five
of
the
different
classes
mapped
to
linkage
group
J
and
encompassed
a
resistance
gene
cluster
including
Rmd,
Rps2,
and
Rj2
(32)
and
near
a
quantitative
trait
locus
for
soybean
cyst
nematode
resistance
(N.
Young,
personal
communication).
We
found
no
map
positions
near
other
known
resistance
genes:
Rhg4,
linkage
group
A2
(36);
Rps3
(35),
Rsv
(37),
linkage
group
F;
and
Rps4
(35),
linkage
group
G.
To
determine
more
precisely
the
map
positions
of
RGAs
relative
to
mapped
resistance
genes
on
linkage
group
J
the
0.158
RGA3
0.103
0.155
RGA7
|
0.171
RGAI
_
_
~~~~0.20
RGA2
0.086
0.262RGRGA
RGA2
0.333
L6
0.288
0.056
RGA4a
0.064
RGA4b
0.302
RGA6
0.205
N
FIG.
3.
Phylogenetic
tree
based
on
alignment
of
amino
acid
sequences
of
tobacco
N,
flax
L6,
and
nine
soybean
RGA
classes.
Numbers
above
lines
indicate
branch
length
(proportion
of
amino
acid
differences
distinguishing
classes).
composite
map
shown
in
Fig.
5
was
constructed
using
the
computer
program
JOINMAP
(33)
and
markers
common
to
both
populations.
These
data
indicate
that
at
least
two
RGAs
map
within
this
multigene
cluster.
BAC
Library
Screening.
Our
initial
screen
of
the
BAC
library
using
the
original
degenerate
primers
was
unsuccessful
because
the
primers
also
amplified
products
from
DH1OB,
the
Escherichia
coli
host
used
to
grow
and
maintain
the
library
(data
not
shown).
To
overcome
this
obstacle
we
designed
class-specific
primers
for
eight
of
the
RGA
classes
(Table
1)
and
identified
50
BACs
representing
seven
classes.
Copy
number
of
RGA
sequences
within
each
BAC
was
estimated
by
digesting
the
BACs
with
restriction
enzymes
that
did
not
have
recognition
sites
within
the
RGA
probe
sequences
and
by
hybridizing
with
each
class-specific
probe.
Fig.
6
demonstrates
results
for
six
BACs
belonging
to
classes
1,
2,
and
3.
Class
1
BACs
each
contained
3-10
copies
of
the
RGA1
sequence.
This
agreed
with
the
8-10
copies
expected
based
on
genomic
Southern
hybridization
patterns
(data
not
shown).
Genomic
Southern
blot
analyses
predicted
two
to
four
copies
of
class
2
RGAs
and
thus
far
the
class
2
BACs
isolated
appear
to
have
two
to
three
copies
of
the
class
2
RGA
sequence.
We
identified
eight
BACs
containing
class
3
RGAs.
Each
of
these
BACs
appeared
to
have
two
to
five
copies
of
RGAs.
Restriction
digests
indicated
that
some
of
the
BACs
have
common
hy-
C2
D1
H
J
L
M
Nl
P
I
I
I
I
I
I
I
I
-RGAI
RGA1
RGA6
*RGA6
RGA3
.RGA2
I~AS
RGA1
-RGA1
-RGA1
-RGA7
-RGA2
FIG.
4.
Distribution
of
RGA
markers
on
the
soybean
genetic
map.
I
11748
Genetics:
Kanazin
et
al.

Proc.
Natl.
Acad.
Sci.
USA
93
(1996)
11749
50.5
12.3
2.9
1.8
2.0
1
2
3
4
5
6
1
2
3
4
5 6
1
2
3
4
5
6
A450-1
RGA6
4.3
-
2.0-
0.5-
411
A
kb
RGA3
RGA2
RGA5
RGA1
K375
RGA
1
NA724
A233
'.RBa
\RGA*
Al199_2
L.J
RGA1
FIG.
5.
A
portion
of
linkage
group
J
showing
the
integration
of
RGA
positions
and
the
map
positions
of
Rps2,
Rmd,
and
Rj2.
The
map
was
generated
through
the
use
of
JOINMAP
(33).
Anchored
markers
used
to
create
this
composite
map
were
K375, A724,
and
A233.
Numbers
to
the
left
of
the
map
indicate
estimated
genetic
distances
between
selected
loci
in
centimorgans.
The
position
of
the
RGA1
identified
with
an
asterisk
(*)
has
been
confirmed
by
independant
mapping
(see
Materials
and
Methods).
bridizing
fragments
and
appear
to
overlap,
forming
a
contig
(see,
for
example
Fig.
6C,
lanes
4-6),
although
some
BACs
may
represent
duplicated
chromosomal
segments.
This
cluster
appears
to
contain
eight
RGA
copies.
Based
on
the
size
of
individual
BACs
the
cluster
spans
"200-300
kb,
indicating
that
the
distance
between
individual
RGA
copies
averages
20
kb.
The
class
3
RGA
mapped
to
only
one
position
on
the
soybean
map
(Fig.
4,
linkage
group
J;
and
Fig.
5).
Based
on
genomic
Southern
blot
analyses,
we
expected
five
to
eight
copies
of
class
3
RGAs;
therefore,
these
BACs
may
contain
the
entire
RGA
class
3
cluster.
DISCUSSION
Nearly
50
restriction
fragment
length
polymorphism
were
mapped
using
13
restriction
enzymes
and
9
RGA
class
se-
quences
as
probes.
Many
polymorphisms
were
likely
due
to
sequence
variations
among
class
members.
However,
given
the
microclustering
of
RGA
class
members,
and
the
small
popu-
lation
size
of
56
individuals
in
which
these
were
mapped,
it
was
not
possible
to
distinguish
very
tightly
linked
loci
from
a
single
_
..
<....
:.....
...*...:
*......
*.
:.
.-..;
n
D
._h
I.....
:C_
FIG.
6.
Southern
hybridization
of
identical
membranes
containing
six
HaeIII
digested
BACs
probed
with
RGA1
(A),
RGA2
(B),
and
RGA3
(C).
Numbered
lanes
indicate individual
BACs.
Molecular
weights
are
shown on
the
left.
Note
that
each
BAC
contains
only
one
RGA
family.
locus.
Therefore,
these
polymorphisms
resolved
to
16
positions
on
8
different
linkage
groups.
Although
very
few
resistance
genes
have
been
mapped
in
soybean,
we
mapped
RGA
sequences
close
to
several
known
genes.
One
reported
gene
cluster
in
soybean,
consisting
of
the
Rps2
locus
conferring
resistance
to
the
fungal
pathogen
Phy-
tophthora
sojae
Kaufmann
&
Gerdemann,
the
Rmd
locus
conferring
resistance
to
the
powdery
mildew
pathogen
Mi-
crosphaera
diffusa
Cooke
&
Peck,
and
the
Rj2
locus
controlling
a
nodulation
response
to
Bradyrhizobia
japonicum
(Kirchner)
Jordan,
maps
within
a
3.8-centimorgan
region
of
linkage
group
J
(32).
A
QTL
associated
with
resistance
to
soybean
cyst
nematode
resistance
is
also
placed
near
this
region
(N.
Young,
personal
communication).
Surprisingly,
a
large
group
of
RGAs
representing
five
of
the
nine
classes
mapped
in
the
region
of
this
cluster.
The
consensus
map
generated
by
JOINMAP
(33)
placed
two
RGAs
within
this
3.8-centimorgan
cluster.
This
was
confirmed
by
independant
mapping
of
an
RGA1
BAC
within
the
population
segregating
for
all
three
genes.
In
this
popu-
lation
we
mapped
the
BAC
within
the
1.8-centimorgan
region
between
Rj2
and
Rmd
(Fig.
5).
This
finding
demonstrates
that
mapping
of
RGA
sequences
can
be
beneficial
in
landing
markers
tightly
linked
to
known
resistance
genes
and
possibly
in
identifying
candidate
resistance
loci.
The
fact
that
members
of
tight
clusters
of
resistance
genes
can
confer
resistance
to
different
pathogens
is
not
surprising
considering
that
members
of
the
same
gene
family
often
maintain
only
partial
redundancy;
they
retain
a
shared
set
of
preserved
functions
but
acquire
unique
specificities
(12,
38).
For
example,
this
would
allow
tightly
linked
members
of
the
same
family
to
retain
structural
motifs
necessary
to
function
in
similar
pathways
(e.g.,
disease
resistance),
while
each
could
respond
to
unique
signals.
Analysis
of
microclusters
of
RGAs
may
have
important
implications
in
identifying
functional
genes
for
any
number
of
signal-responsive
traits.
In
tomato
the
Pto
gene
belongs
to
a
complex
locus
consisting
of
a
tightly
linked
cluster
of
five
to
seven
genes.
Pto
confers
resistance
to
P.
syringae
(11),
while
the
tightly
linked
homolog,
Fen,
confers
sensitivity
to
an
organo-
phosphate
insecticide
(39,
40).
Similar
examples
of
related
genes
that
have
acquired
unique
roles
can
be
found
among
gene
families
involved
in
the
regulation
of
floral
identity,
in
the
reception
of
specific
light
spectra,
and
in
cell
differentiation
(see
ref.
38).
The
clustering
of
signal-responsive
genes
suggests
that
common
genetic
processes-e.g.,
unequal
crossing-over
and
gene
conversion-have
acted
upon
them
during
their
evolution;
although
no
conserved
mechanisms
by
which
these
results
are
obtained
has
been
established
(16).
Genetics:
Kanazin
et
aL

Proc.
Natl.
Acad.
Sci.
USA
93
(1996)
It
is
unlikely
that
each
member
of
an
evolving
gene
family
will
remain
functional.
Without
positive
or
negative
selection
acting
to
retain
function
of
gene
copies
within
a
gene
family
the
accumulation
of
deleterious
mutation
would
quickly
result
in
the
silencing
of
redundant
genes
(38).
However,
within
an
environment
in
which
a
population
is
continually
challenged
by
a
broad
range
of
stressful
conditions
(e.g.,
pathogens),
it
could
be
possible
for
a
rich
repetoire
of
functionally
similar
genes
responsive
to
unique
signals
to
develop
(see
refs.
12
and
41).
Our
findings
demonstrate
that
conserved
sequences
from
resistance
genes
cloned
from
a
diverse
range
of
plant
taxa
can
be
used
to
identify
evolutionarily
related
genes
from
soybean.
These
related
sequences
are
distributed
throughout
the
ge-
nome,
exist
in
microclusters
of
gene
classes,
and
are
associated
with
known
resistance
genes.
The
identification
of
candidate
resistance
genes
by
restriction
fragment
length
polymorphism
mapping
using
RGA
sequences
may,
however,
have
limita-
tions.
Indeed,
it
is
likely
that
the
silenced
pseudogenes,
by
virtue
of
accumulation
of
mutation,
will
be
the
source
of
polymorphisms
between
genotypes.
If
this
is
true,
we
can
predict
that
it
will
be
these
sequences
that
are
mapped.
Therefore,
genetic
mapping
of
RGA
sequences
may
be
more
important
in
landing
markers
close
to
resistance
genes
for
subsequent
map-based
cloning,
than
for
direct
identification
of
resistance
genes.
We
are
grateful
to
Dr.
C.
Nickell
(Agricultural
Research
Service,
U.S.
Department
of
Agriculture)
for
making
available
germplasm
used
in
this
study.
We
are
also
grateful
for
the
financial
support
from
the
Iowa
Soybean
Promotion
Board.
This
paper
is
a
joint
contribution
of
Midwest
Area,
Agricultural
Research
Service,
U.S.
Department
of
Agriculture,
Field
Crops
Research
Unit
and
Journal
Paper
No.
J-16991 of
the
Iowa
Agricultural
and
Home
Economics
Experiment
Station,
Ames;
Project
3236.
1.
Bent,
A.,
Kunkel,
B.,
Dahlbeck,
D.,
Brown,
K.,
Schmidt,
R.,
Giraudat,
J.,
Leung,
J.
&
Staskawicz,
B.
(1994)
Science
265,
1856-1860.
2.
Mindrinos,
M.,
Katagiri,
F.,
Yu,
G.-L.
&
Ausubel,
F.
(1994)
Cell
78,
1089-1099.
3.
Grant,
M.,
Godiard,
L.,
Straube,
E.,
Ashfield,
T.,
Lewald,
J.,
Sattler,
A.,
Innes,
R.
&
Dangl,
J.
(1995)
Science
269,
843-846.
4.
Dinesh-Kumar,
S.,
Whitham,
S.,
Choi,
D.,
Hehl,
R.,
Corr,
C.
&
Baker,
B.
(1995)
Proc.
Nati.
Acad.
Sci.
USA
92,
4175-4180.
5.
Whitham,
S.,
Dinesh-Kumar,
S.
P.,
Choi,
D.,
Hehl,
R.,
Corr,
C.
&
Baker,
B.
(1994)
Cell
78,
1101-1115.
6.
Song,
W.-Y.,
Wang
G.-L.,
Chen,
L.-L.,
Kim,
H.-S.,
Pi,
L.-Y.,
Holsten,
T.,
Gardner,
J.,
Wang,
B.,
Zhai,
W.-X.,
Zhu,
L.-H.,
Fauquet,
C.
&
Ronald,
P.
(1995)
Science
270,
1804-1806.
7.
Lawrence,
G.,
Finnegan,
E.
J.,
Ayliffe,
M.
&
Ellis,
J.
(1995)
Plant
Cell
7,
1195-1206.
8.
Jones,
D.
A.,
Thomas,
C.
M.,
Hammond-Kosack,
K.
E.,
Balint-
Kurti,
P.
J.
&
Jones,
J.
D.
G.
(1994)
Science
266,
789-793.
9.
Kobe,
B.
&
Deisenhofer,
J.
(1994)
Trends
Biochem.
Sci.
19,
415-421.
10.
Staskawicz,
B.,
Ausubel,
F.,
Baker,
B.,
Ellis,
J.
&
Jones,
J.
(1995)
Science
268,
661-667.
11.
Martin,
G.,
Brommonschenkel,
S.,
Chunwongse,
J.,
Frary,
A.,
Ganal,
M.,
Spivey,
R.,
Wu,
T.,
Earle,
E.
&
Tanksley,
S.
(1993)
Science
262,
1432-1436.
12.
Buck,
L.
&
Axel,
R.
(1991)
Cell
65,
175-187.
13.
King,
M.
W.
&
Moore,
M.
J.
(1994)
Nucleic
Acids
Res.
22,
3990-3996.
14.
Hulbert,
S.
H.
&
Bennetzen,
J.
L.
(1991)
Moi.
Gen.
Genet.
226,
377-382.
15.
Saxena,
K.
M.
S.
&
Hooker,
A.
L.
(1968)
Genetics
61,
1300-1305.
16.
Sudapak,
M.
A.,
Bennetzen,
J.
L.
&
Hulbert,
S.
H.
(1993)
Ge-
netics
132,
119-125.
17.
Mahadevappa,
M.,
DeScenzo,
R.
&
Wise,
R.
(1994)
Genome
37,
460-468.
18.
DeScenzo,
R.
A.,
Wise,
R.
P.
&
Mahadevappa,
M.
(1994)
Mol.
Plant-Microbe
Interact.
7,
657-666.
19.
Jorgensen,
J.
H.
(1992)
Euphytica
63,
141-152.
20.
Gorg,
R.,
Hollricher,
K.
&
Schulze-Lefert,
P.
(1993)
Plant
J.
3,
857-866.
21.
Bonnier,
F.
J.
M.,
Reinink,
K.
&
Groenwald,
R.
(1994)
Phyto-
pathology
84,
462-468.
22.
Maisonneuve,
B.,
Bellec,
Y.,
Anderson,
P.
&
Michelmore,
R.
W.
(1994)
Theor.
Appl.
Genet.
89,
96-104.
23.
Rayapati,
P.
J.,
Lee,
M.,
Gregory,
J.
W.
&
Wise,
R.
P.
(1994)
Theor.
Appl.
Genet.
89,
831-837.
24.
Shepherd,
K.
&
Mayo,
G.
(1972)
Science
175,
375-380.
25.
Ellis,
J.
G.,
Lawrence,
G.
J.,
Finnegan,
E.
J.
&
Anderson,
P.
A.
(1995)
Proc.
Natl.
Acad.
Sci.
USA
92,
4185-4188.
26.
Islam,
M.
R.
&
Sheperd, K.
W.
(1991)
Hereditas
114,
125-129.
27.
Dellaporta,
S.
L.,
Wood,
J.
&
Hicks,
J.
B.
(1983)
Plant
Mol.
Biol.
Rep.
1,
19-21.
28.
Sambrook,
J.,
Fritsch,
E.
&
Maniatis,
T.
(1989)
Molecular
Clon-
ing:
A
Laboratory
Manual
(Cold
Spring
Harbor
Lab.
Press,
Plainview,
NY),
2nd
Ed.
29.
Kumar,
S.,
Tamura,
K.
&
Nei,
M.
(1993)
MEGA:
Molecular
Evolutionary
Genetics
Analysis
(Pennsylvania
State
Univ.,
Uni-
versity
Park),
Version
1.0.
30.
Keim,
P.,
Diers,
B.,
Olson,
T.
&
Shoemaker,
R.
C.
(1990)
Genetics
126,
735-742.
31.
Lander,
E.,
Green,
P.,
Abrahamson,
J.,
Barlow,
A.,
Daly,
M.,
Lincoln,
S.
&
Newburg,
L.
(1987)
Genomics
1,
174-181.
32.
Polzin,
K.
M.,
Lohnes,
D.
G.,
Nickell,
C.
D.
&
Shoemaker,
R.
C.
(1994)
J.
Hered.
85,
300-303.
33.
Stam,
P.
(1993)
Plant
J.
3,
739-744.
34.
Marek,
L.
F.
&
Shoemaker,
R.
C.
(1996)
Soybean
Genet.
Newsl.
23,
in
press.
35.
Diers,
B.
W.,
Mansur,
L.,
Imsande,
J.
&
Shoemaker,
R.
C.
(1992)
Crop
Sci.
32,
377-383.
36.
Weisemann,
J.
M.,
Matthews,
B.
E.
&
Devine,
T.
E.
(1992)
Theor.
Appl.
Genet.
85,
136-138.
37.
Yu,
Y.
G.,
Saghai-Maroof,
M.
A.,
Buss,
G.
R.,
Maughan,
P.
J.
&
Tolin,
S.
A.
(1994)
Phytopathology
84,
60-64.
38.
Pickett,
F.
B.
&
Meeks-Wagner,
D.
R.
(1995)
Plant
Cell
7,
1347-1356.
39.
Martin,
G.,
Frary,
A.,
Wu,
T.,
Brommonschenkel,
S.,
Chun-
wongse,
J.,
Earle,
E.
&
Tanksley,
S.
(1994)
Plant
Cell
6,
1543-
1552.
40.
Loh,
Y.-T.
&
Martin,
G.
B.
(1995)
Proc.
Natl.
Acad.
Sci.
USA
92,
4181-4184.
41.
Dangl,
J.
L.
(1995)
Cell
80,
363-366.
11750
Genetics:
Kanazin
et
aL