Copyright 2002 by the Genetics Society of America
A Genetic Map of Gibberella zeae (Fusarium graminearum)
J. E. Jurgenson,* R. L. Bowden,† K. A. Zeller,† J. F. Leslie,†,1 N. J. Alexander‡ and R. D. Plattner‡
*Department of Biology, University of Northern Iowa, Cedar Falls, Iowa 50614, †Department of Plant Pathology, Kansas State University,
Manhattan, Kansas 66506-5502 and ‡Mycotoxin Research Unit, USDA/ARS National Center for Agricultural Utilization Research,
Peoria, Illinois 61604
Manuscript received November 1, 2001
Accepted for publication December 26, 2001
ABSTRACT
We constructed a genetic linkage map of Gibberella zeae (Fusarium graminearum) by crossing complementary nitrate-nonutilizing (nit) mutants of G. zeae strains R-5470 (from Japan) and Z-3639 (from Kansas).
We selected 99 nitrate-utilizing (recombinant) progeny and analyzed them for amplified fragment length
polymorphisms (AFLPs). We used 34 pairs of two-base selective AFLP primers and identified 1048 polymorphic markers that mapped to 468 unique loci on nine linkage groups. The total map length is ⵑ1300 cM
with an average interval of 2.8 map units between loci. Three of the nine linkage groups contain regions
in which there are high levels of segregation distortion. Selection for the nitrate-utilizing recombinant
progeny can explain two of the three skewed regions. Two linkage groups have recombination patterns
that are consistent with the presence of intercalary inversions. Loci governing trichothecene toxin amount
and type (deoxynivalenol or nivalenol) map on linkage groups IV and I, respectively. The locus governing
the type of trichothecene produced (nivalenol or deoxynivalenol) cosegregated with the TRI5 gene (which
encodes trichodiene synthase) and probably maps in the trichothecene gene cluster. This linkage map
will be useful in population genetic studies, in map-based cloning, for QTL (quantitative trait loci) analysis,
for ordering genomic libraries, and for genomic comparisons of related species.
IBBERELLA zeae (anamorph Fusarium graminearum) is the most important causal agent of Fusarium head blight (scab) of wheat and barley in the
United States (McMullen et al. 1997) and China (Chen
et al. 2000). In the 1990s, scab caused an estimated $3
billion losses to wheat and barley farmers in the United
States alone (Windels 2000). Scab reduces wheat baking quality (Seitz et al. 1986) and harvested grain often
is contaminated with mycotoxins such as nivalenol
(NIV), deoxynivalenol (DON), and zearalenone (Marasas et al. 1984; Tanaka et al. 1988).
G. zeae is homothallic (Nelson et al. 1983; Yun et al.
2000) and may produce abundant perithecia in the
field, but can be outcrossed under laboratory conditions
(Bowden and Leslie 1999). Despite being homothallic,
the amount and distribution of genetic heterogeneity in
field populations of this fungus suggest that outcrossing
occurs at a significant rate in the field (Bowden and
Leslie 1992; Walker et al. 2001). Recently, O’Donnell
et al. (2000) used DNA sequences of elongation factor
(EF-1␣), phosphate permease genes (PHO), -tubulin (TUB), UTP-ammonia ligase (URA), trichothecene
3-O-acetyltransferase (TRI101), and a putative reductase
(RED) to resolve a set of G. zeae strains into at least seven
distinct phylogenetic lineages. Differences between strains
in distinct lineages include qualitative differences in
G
1
Corresponding author: Department of Plant Pathology, 4002 Throckmorton Plant Sciences Center, Kansas State University, Manhattan,
KS 66506-5502. E-mail: jfl@plantpath.ksu.edu
Genetics 160: 1451–1460 (April 2002)
toxin production (DON, 3-acetyldeoxynivalenol, 15-acetyldeoxynivalenol, NIV, 4-acetylnivalenol, and zearalenone)
as well as in DNA sequence-based markers. The degree
of genetic isolation and pathogenic specialization among
lineages remains unresolved (Carter et al. 2000).
Bowden and Leslie (1999) established that, under
laboratory conditions, members of at least three of the
phylogenetic lineages described by O’Donnell et al.
(2000) can interbreed and produce viable, recombinant
progeny. The fertility of these interlineage crosses can
be relatively high and suggests that these strains are
members of geographically separated and genetically
distinct populations rather than of distinct species.
O’Donnell et al. (2000) described one putative naturally occurring hybrid strain (collected in Nepal by A. E.
Desjardins) between lineages 2 and 6. If isolates from
these genetically divergent populations interbreed, then
there is the potential for the production of new genotypes that carry novel combinations of genes for pathogenicity, host range, or toxin production (Brasier 2000).
A cross between isolates from two such distantly related
populations also should be rich in polymorphic markers
that could be used to generate a detailed genetic map.
Amplified fragment length polymorphism (AFLP)
analysis is a PCR-based DNA analysis technique that can
detect variations in restriction fragment length polymorphisms (RFLP) on a genome-wide basis (Vos et al. 1995).
Like restriction fragment length polymorphism analysis,
AFLPs can detect size differences in restriction fragments caused by DNA insertions, deletions, or changes
1452
J. E. Jurgenson et al.
in target restriction site sequences. As compared to
RFLP analysis, however, the labor required to detect
genetic polymorphisms with AFLPs is considerably reduced. AFLP analysis yields dominant band/no-band
type markers that can be used to study genetic diversity
in fungal populations (e.g., Gonzalez et al. 1998; Purwantara et al. 2000; Zeller et al. 2000) and to define
and distinguish species of Fusarium (Marasas et al.
2001). AFLPs have been used to develop recombinationbased genetic maps in mapping populations of higher
plants such as barley, soybeans, and maize (Vuylsteke
et al. 1999; Yin et al. 1999; Hua et al. 2000) and to
supplement mapping efforts in fungi, e.g., in Phytophthora (van der Lee et al. 2001). When using 2-bp extensions in specific AFLP reactions, as is common in fungi,
there are 256 potential primer-pair combinations that
can each be used to generate a unique DNA fingerprint
pattern for each pair of restriction enzymes used in the
initial digestion of the DNA. The genomic distribution
of markers generated by AFLPs is limited only by the
distribution of the restriction sites used to generate
them.
Our objective in this study was to establish a recombination-based genetic linkage map of G. zeae by crossing
phenotypically and genetically divergent strains from
different continents. Only one other detailed genetic
map is presently available for any Gibberella species
(Xu and Leslie 1996). Generation of a genetic linkage
map for G. zeae will permit the correlation of physical
sequences with segregating phenotypes, the localization
of genes for toxin production, the identification of genetically independent markers that can be used in characterization of field populations, and the identification
of genomic sequences that might be of particular importance in the evolution of this species.
MATERIALS AND METHODS
Mapping cross: One of the parents of this cross was derived
from a DON-producing strain, Z-3639, originally isolated from
wheat in Kansas (Bowden and Leslie 1992) and belonging
to lineage VII as described by O’Donnell et al. (2000). The
other parent was derived from a NIV-producing strain, R-5470,
originally isolated from barley in Japan and obtained from
Paul E. Nelson (Department of Plant Pathology, Pennsylvania
State University, University Park, PA), which belongs to lineage
VI as described by O’Donnell et al. (2000). These strains were
selected because of the difference in toxin production and
to give a large number of polymorphic markers, due to the
presumptive isolation of the populations from which they were
derived.
Crosses were performed essentially as described previously
(Bowden and Leslie 1999). Mycelial plugs of a nit1 mutant
(Z-11570) of R-5470 and a nit3 mutant (Z-11572) of Z-3639
were simultaneously inoculated onto carrot agar medium
(Klittich and Leslie 1988). nit mutants were generated as
previously described (Bowden and Leslie 1992) and characterized by using standard Fusarium protocols (Correll et
al. 1987; Klittich and Leslie 1988). The parents [Z-11570,
Fungal Genetics Stock Center (FGSC) 8632; Z-11572, FGSC
8633] of and the progeny (FGSC 8634–8732) from the mapping cross are available from the Fungal Genetics Stock Center
(Department of Microbiology, University of Kansas Medical
Center, Kansas City, KS; http://www.fgsc.net). We analyzed 99
single ascospore-derived progeny from this cross. The frequency of nitrate-utilizing ascospores was ⬍1%, because most
of the perithecia were homothallic selfs of Z-11572.
We isolated ascospores from mature perithecia by inverting
the carrot agar cross plates and collecting ascospores on the
plate lid. These ascospores were suspended in 5 ml of sterile
water and then dilution plated onto a minimal agar medium
(Correll et al. 1987) amended with tergitol and sorbose
(Bowden and Leslie 1999). Plates were incubated for 5–7
days at 24⬚. Recombinant progeny, identified as nitrate utilizers, were collected and transferred to minimal medium
slants. Each of the recombinant progeny was subcultured from
a single macroconidium. Macroconidia were separated with
a Cailloux stage-mounted micromanipulator (Stoelting, Chicago). Cultures were maintained on minimal medium and
stored as spore/hyphal fragment suspensions in 15% glycerol
at ⫺70⬚ at Kansas State University.
Analysis of DNA polymorphisms in the mapping population:
We inoculated 50 ml of liquid complete medium (Correll
et al. 1987) in 125-ml Erlenmeyer flasks with ⵑ5 ⫻ 105 macroconidia suspended in ⵑ1 ml of a 2.5% aqueous (v/v) solution
of Tween 60 (Sigma, St. Louis). Cultures were incubated for
2–3 days at room temperature (22⬚–25⬚) on a rotary shaker
(150 rpm). Tissue from each culture was collected by filtration
through a nongauze milk filter (Ken Ag Milk Filter, Ashland,
OH), washed with 100 ml sterile water, and blotted dry with
paper towels. The tissue was frozen at ⫺20⬚ until DNA was
extracted.
DNA extraction: DNA was isolated with a cetyltrimethyl ammonium bromide procedure (Kerényi et al. 1999) modified
from that of Murray and Thompson (1980). We estimated
final DNA concentrations (in TE buffer) by comparison of
DNA fluorescence of diluted aliquots of each DNA sample
against that of HindIII-digested bacteriophage DNA with an
IS-1000 version 2.0 digital imaging system (Alpha Innotech,
San Leandro, CA). Samples and sample dilutions were run in
1% agarose gels containing TAE (40 mm Tris-acetate, 1 mm
EDTA pH 8.0) and 0.5 g/ml ethidium bromide. DNA yields
ranged from 100 to 1000 g of DNA per culture. The concentration of each DNA sample was adjusted to 20 g/ml for use
in AFLP analysis.
AFLPs: AFLPs were generated with the protocol of Vos et
al. (1995) as modified by Zeller et al. (2000). AFLP primers
were synthesized by Integrated DNA Technologies (Coralville,
IA). The EcoRI primers in the final specific amplification reactions were 5⬘ end-labeled with [␥-33P]ATP (NEN Life Sciences,
Boston). Dried gels were exposed to X-ray film (Classic Blue
Sensitive, Molecular Technologies, St. Louis) for 2–5 days
at room temperature to identify DNA bands. We identified
polymorphic bands by eye and scored them manually. We
estimated molecular weights of AFLP fragments by comparisons with the Low Mass Ladder (Life Technologies, Bethesda,
MD) DNA standard that also was 5⬘ end-labeled with 33P. Most
polymorphisms were characterized as presence/absence of
bands although a few occurred in which the polymorphism
appeared as an apparent difference in molecular weight. Polymorphic bands were named using the nomenclature E_M_
0000_, where E_ denotes the EcoRI primer with the two additional selective nucleotides, M_ denotes the MseI primer with
the two additional selective nucleotides, the four-digit number
is an estimate of the size of the band in base pairs, and the
final blank is either “J” or “K” and denotes the parent that
was the source of the “band present” allele or was the source
of the larger band of a size-difference polymorphism. For
G. zeae Genetic Map
example, EAAMGT0234J is an AFLP polymorphism whose
presence allele is a DNA fragment 234 bp in length that originated from the Japanese parent (Z-11570) and was generated
by amplification with the primer pair EAA/MGT.
AFLPs were scored based on two DNA preparations that
began with independent cultures of each of the progeny. We
scored one primer pair from both DNA preparations for all
of the progeny. When results could not be scored clearly from
the first DNA preparation, the questionable progeny and the
parents from both the first and second DNA preparations
were run side-by-side on a second gel to resolve discrepancies
and to check for reproducibility. Thus the AFLP patterns from
the parents were checked with all primer pairs from two DNA
preparations, while AFLP patterns from the progeny were all
checked with one primer pair and were irregularly tested with
the remaining primer pairs. We did not score AFLP polymorphisms based on bands that were ⬍90 bp in length, as they
were not always consistent between the two DNA preparations.
Fertility and pigment: The parental strains differ with respect to pigment production (Kansas strain makes a bright red
pigment) and sexual reproduction (Kansas strain produces
numerous mature homothallic perithecia on carrot agar, but
the Japanese strain does not). These characters (PIG1 and
PER1) segregated in the progeny of the cross. PER1 was scored
by examination of 4-week-old cultures on carrot agar. PIG1
was scored by examination of 2-week-old cultures on complete
medium (Correll et al. 1987). Progeny were tested at least
twice to confirm phenotype designations.
Toxin assay: For toxin analysis, progeny and parental strains
were grown on sterilized cracked corn at 25⬚ for 4 weeks as
previously described (Leslie et al. 1992). The standard AOAC
method for the extraction of DON was used (Scott 1995), but
the determinative step was modified. To detect the presence of
both DON and NIV, gas chromatography/mass spectrometry
(GC/MS) of the trimethyl silane (TMS) derivative was used
instead of thin layer chromatography or electron capture GC
of the heptafluorobutyrate derivative as the detection step.
Briefly, the method was the following: Toxins were obtained
from cultures by extraction with 4 ml of acetonitrile:water
(84:16) per gram of culture material. DON and NIV were
detected as TMS derivatives by GC/MS as follows: A 100-l
aliquot of the extract was evaporated to dryness under nitrogen at 60⬚ and derivatized with 100 l trisil-TBT (trimethylsilylimadzole:bis-trimethylacetamide:trimethylsilylchloride 3:3:2)
reagent (Pierce, Rockford, IL) at 60⬚ for 1 hr. A total of 900
l of iso-octane was added to the solution and 1 l was injected
into the GC/MS at 70⬚ for analysis. Samples were analyzed on
a 15-m, 0.25-mm, 0.25-m Rtx-5MS column (Restec, Belefonte,
PA). The column was programmed at 30⬚/min to 180⬚, then
by 1⬚/min to 200⬚, and finally by 30⬚/min to 270⬚ and held
at 270⬚ for 5 min. The TMS derivatives of DON and NIV
eluted during the 180⬚–200⬚ gradient at 14.3 and 20.5 min,
respectively.
TRI5 analysis: The TRI5 gene is part of the trichothecene
gene cluster and encodes the enzyme trichodiene synthase
(Hohn and Beremand 1989; Brown et al. 2001). Forty micrograms of genomic DNA from each parental and progeny strain
was digested with 4 units of MseI (New England BioLabs,
Beverly, MA) for 4 hr. Fragments were separated on 1% agarose gels and blotted onto Nytran SuPer Charge membrane
(Schleicher and Schuell, Keene, NH) as previously described
(Sambrook et al. 1989). To detect polymorphisms in the TRI5
region, blots were probed with a 32P-labeled (Prime-a-Gene,
Promega, Madison, WI) fragment generated by PCR amplification of an ⵑ590 bp fragment containing the TRI5 region
from a larger, sequenced piece of the TRI gene cluster of F.
graminearum (GenBank accession no. AF359361). The primers
1453
used were the following: 5⬘-GGCATGGTTGTATACAGC-3⬘
and 5⬘-CAGAGTGATCTCATGGCAGG-3⬘. Amplification of
the fragment was performed with 30 cycles at 94⬚ for 30 sec,
52⬚ for 30 sec, and 72⬚ for 60 sec in a MJ PTC-100HB Thermocycler (MJ Research, Watertown, MA). Hybridization was performed in buffer containing formamide at 42⬚ and washes
were done as recommended by the membrane’s manufacturer. Hybridization was detected by exposing blots to Kodak
AR X-ray film at room temperature overnight.
Marker analysis: Genetic mapping of all characters was
performed using Map Manager QTX11 (http://mapmgr.
roswellpark.org/mmQTX) on a Macintosh G4 Power PC computer (Manly and Olson 1999). Data from gels were compiled as text files and imported into this program. We then
used Map Manager to distribute the data into linkage groups.
Program settings used for analysis were Kosambi mapping
function, search and linkage criteria set to a probability of
type I error for false linkage of P ⫽ 0.0001. The authors of
the program suggest that linkage relationships created with
this setting represent physical chromosomes. The mapping
program treated the data as a backcross with codominant
markers, with the paternal parent unique, as was necessary
for analysis of this haploid genome. No user-defined map
distances were known, so this function was not used. Following
the initial linkage group analysis, we inspected the aligned
phenotype data visually to minimize linkage distance based
on the assumption that single-locus double recombinants were
highly unlikely. These apparent double crossover events may
be due to gene conversion or to errors in data scoring and
data entry. We also converted unknown data or unscored
markers to their probable phenotypes on the basis of the
character states of scored flanking loci because Map Maker
V2.0 for the Macintosh program (Lander et al. 1987), which
we used to draw the map figures, treats unscored characters
as a third allele and inserts a crossover on each side of these
markers, which misrepresents the genetic distances between
the markers on the graphical output (Figure 1).
RESULTS
We utilized polymorphic bands generated by PCR
amplification with 34 different AFLP primer pairs. The
number of polymorphisms detected per primer pair
ranged from 23 to ⬎50, with an average of 32 per primer
pair. Approximately 0.8% of the scores from the first
DNA preparation resulted in unscorable or ambiguous
results and were scored on the basis of a second, independent DNA preparation. We found no duplicate or
completely complementary progeny in the progeny set.
Map Manager distributed the markers into nine linkage groups (Figure 1). Chromosome-sized linkage groups
vary in total genetic length from 281 cM for linkage
group I to 52 cM for linkage group IX (Table 1). We
estimate the length of the entire genome as ⵑ1300 cM
with an average distance of 2.8 cM between loci.
Of the 1070 markers analyzed, 22 were not associated
with any of the nine large linkage groups. Of these 22
markers, 15 comprise a small linkage group of 10 loci.
Six markers comprise 1 of the loci in this linkage group.
The remaining 7 of these 22 markers show no linkage
to each other or to any linkage group. The segregation
pattern of all 22 markers is distorted, i.e., not 1:1. The
Japanese allele is dominant for 18 of these markers and
Figure 1.—G. zeae linkage map. Loci are named by primer pair as described in the text. Loci that are represented by more
than one AFLP marker are indicated in the form xyy, where x is the number of the linkage group and yy is a letter or pair of
letters assigned in order along the linkage group. These names are followed in parentheses by the number of polymorphic AFLP
markers that map to this location.
G. zeae Genetic Map
1455
TABLE 1
Distribution of markers, loci, and crossovers across linkage groups
Observed crossover events per linkage group
Linkage group
None
No.
cM
No.
of loci
No. of
markers
Japanese
KS
1
2
3
4
5
6
7–10
Mean
I
II
III
IV
V
VI
VII
VIII
IX
281
263
219
182
102
63
69
55
52
94
90
88
69
31
26
26
22
22
171
203
214
158
71
71
69
41
53
7
33
19
0
9
57
28
20
27
18
0
6
24
7
5
33
36
48
9
2
14
22
69
19
20
29
7
15
24
16
16
7
13
9
10
9
13
3
20
23
4
5
6
3
5
19
25
12
3
2
0
1
1
3
8
1
7
9
1
0
0
0
0
6
6
2
1
0
0
2
0
0
4
6
3
1
0
0
0
0
0
2.65
2.43
2.27
1.95
1.09
0.61
0.73
0.63
0.63
Total
1286
468
1048
200
177
151
119
75
66
26
17
14
1.44
the Kansas allele for 4. These markers are probably not
associated with the mitochondria, since we would expect
all of the mitochondrial loci to have originated from
the Kansas strain that served as the female parent. These
bands could originate from multiple sequences that fortuitously are the same size; i.e., the sequences represented by any given band are not homologous or from
a multiple-copy DNA sequence, e.g., a transposable element, that is dispersed throughout the genome.
A large number of AFLP markers map to common
genetic loci. On each linkage group multiple loci are
represented by 2 or more polymorphic markers. We
named these loci based on their linkage group number
followed by one or more alphabetical characters. This
name is followed in Figure 1 by a number in parentheses
that indicates the number of AFLP markers that map to
that location. As many as 22 markers may map to one
location (4K, Figure 1) with 64 loci represented by 4 or
more AFLP markers. A total of 485 unique loci are
defined by the 1070 AFLP polymorphisms; 468 of these
loci map to unique positions on one of the nine linkage
groups.
Haplotype analysis: No loci on linkage groups VII,
VIII, and IX have segregation ratios that are statistically
different from 1:1 (Figure 2A). Linkage group I also is
generally unbiased, although a few loci have a slight
bias toward the Kansas genome. Five of the nine linkage
groups (Figure 2B) exhibit a segregation ratio of paternal-to-maternal alleles significantly different from the
1:1 expected ratio for progeny of a haploid genetic cross
(G test, P ⫽ 0.05; Weir 1990). Linkage group VI is
severely skewed to the Japanese genome along its entire
length. Linkage group V is skewed toward the Japanese
genome at one end and toward the Kansas genome
at the other. Skewing also occurs toward the Japanese
genome for the distal one-third of each end of linkage
group II. Segregation is biased toward the Japanese genome on one end of linkage group III and toward the
Kansas genome at one end of linkage group IV.
Haplotype analysis of linkage groups I, II, and IV
(Figure 3) illustrates representative patterns of recombination that we observed. On linkage group I, crossovers
appear to be distributed randomly and there is no segregation distortion (Figure 3A). For linkage group II, 33
of the progeny (Table 1) had no detected crossing over
in this linkage group and contained genetic material
from only the Japanese parent. Both terminal portions
of this linkage group are significantly skewed to the
Japanese parent alleles and are fixed at locus 2C for
Japanese alleles. Only in the central portion of linkage
group II (Figure 3B) is segregation not distorted. Of the
66 recombinant progeny, only 7 have an odd number of
crossovers. The remaining 59 have an even number
of crossovers, with those with two (24) and four (25)
crossovers predominating (Table 1).
Linkage group IV’s recombination pattern is similar
to that for linkage group II, but the direction of skewing
is reversed (Figure 3C). In this case the 24 nonrecombinant haplotypes are exclusively of the Kansas type. One
end of this linkage group is biased to the Kansas genome, and, at locus 4AD, all of the progeny have Kansas
alleles. The central portion of this chromosome also
has an apparent excess of even-numbered crossovers.
TRI5 analysis: Hybridization of a PCR probe homologous to the TRI5 gene identified a MseI RFLP polymorphism. The TRI5 probe hybridized to a 2.2-kb MseI fragment of R-5470 and to a 1.7-kb fragment of Z-3639.
This polymorphism segregated ⵑ1:1 in the progeny and
maps to linkage group I.
Toxin production: Toxin production by the parental
and progeny strains varied greatly under our culture
conditions. The Kansas parent produces ⵑ50 ppm DON
1456
J. E. Jurgenson et al.
Figure 2.—Marker segregation by linkage
group. (A) Linkage groups (I, VII, VIII, and IX)
with no significant segregation distortion. (B)
Linkage groups (II, III, IV, V, and VI) with significant segregation distortion. Values between the
two dotted lines are not statistically different from
a 1:1 segregation ratio (G test, P ⫽ 0.05).
while the Japanese parent produces much lower levels
of NIV (ⵑ1 ppm). Fifty-four of the progeny produced
levels of toxin that were high enough to characterize the
toxin by GC/MS. Of these high producers, 28 produced
DON and 26 produced NIV. No progeny produced both
toxin types at high levels. All high-level NIV producers
produce trace amounts of DON, but high-level DON
producers make no detectable NIV. Both toxin type
(DON/NIV) and toxin level (TOX1) segregated in the
cross as single Mendelian characters. TOX1 maps to one
end of linkage group IV and the locus controlling toxin
type cosegregated with the MseI polymorphism associated with the TRI5 gene on linkage group I (Figure 1).
Fertility and pigment: The loci for perithecia produc-
tion (PER1) and red pigment production (PIG1) both
mapped near the locus controlling high levels of toxin
production on linkage group IV (Figure 1).
DISCUSSION
This map of G. zeae is the second genetic map for a
Gibberella species and for any species with a Fusarium
anamorphic state. The G. zeae map includes many loci
that are represented by more than one AFLP polymorphism. Some explanations for this clustering include
map saturation (the average distribution of loci is 2.8
cM/locus); nonrandom distribution of AT-rich nucleotide regions in the genome, which would contain a
G. zeae Genetic Map
Figure 3.—Haplotype plots of linkage groups I (A), II (B),
and IV (C). Each vertical column represents an individual
progeny haplotype; some haplotypes represent more than one
of the progeny. Haplotypes are sorted by the number of crossovers and the relative position of the first crossover. Solid,
Kansas parental genome; open, Japanese parental genome.
higher number of EcoRI and MseI restriction sites; recombination suppression, perhaps caused by heterozygous inversions, deletions, or insertions that interfere
with recombination; or nonrandom distribution of restriction sites due to methylation, as has been observed
in soybeans (Young et al. 1999).
Segregation distortion in the progeny: As G. zeae is
homothallic we crossed strains carrying complementary
auxotrophic nit mutants in the parents and selected
nitrate-utilizing recombinant random ascospore prog-
1457
eny. This procedure was used by Bowden and Leslie
(1999) to demonstrate the potential for outcrossing and
hybridization between G. zeae strains. Selection of wildtype recombinants was the most practical method available to obtain sufficient ascospores from the rare hybrid
perithecia for the analysis. Because the selection we
applied selected for only one class of recombinant progeny, we expected and observed segregation distortion
in our cross due to the selection for the wild-type nit1
and nit3 alleles. All of the progeny in the 99-member
mapping population have at least two recombinant
chromosomes (98 have 3 or more) and no more than
80% of the markers in any one of the progeny are from
a single parent.
Distribution of crossovers: The distribution of crossovers across the linkage groups of the progeny is not
random (Table 1). In particular, there is an excess of
progeny with no crossovers within a linkage group and
a reduction in the number of progeny with linkage
groups in which recombination has occurred as compared to an expected Poisson distribution (2 test, P ⫽
0.05). For each of linkage groups VI, VII, VIII, and IX,
at least half of the progeny have no detectable crossover,
which could make estimates of linkage distances less
accurate than expected, given the number of progeny
analyzed.
There also is no evidence for chromosome loss as no
progeny have a linkage group on which all of the alleles
are of the no-band type. Thus, unlike F. moniliforme (Xu
and Leslie 1996) and F. solani (Miao et al. 1991), we
found no evidence for dispensable “B” chromosomes
in G. zeae.
We think that recombination might be generally suppressed in our cross. Recombination suppression has
been reported on chromosome 1 in the pseudohomothallic species Neurospora tetrasperma (Gallegos et al.
2000), but no mechanism to explain these observations
has been proposed. Grell (1962) proposed that recombination might be essential for a chromosome to find
its homolog and pass properly through meiosis. This
phenomenon has been studied in Drosophila (Hawley
et al. 1993) and more recently in yeast (Haber 1998).
Chromosomes without partners go through a distributive disjunction process where pairing apparently occurs
on the basis of size. In filamentous fungi, “B” chromosomes are thought to segregate in this manner (Miao
et al. 1991; Xu and Leslie 1996). In our cross, the number
of progeny carrying an intact parental linkage group is
high (Table 1), and we presume that distributive disjunction must be functioning for the corresponding
chromosomes to pass through meiosis. We do not know
the physical sizes for any of the chromosomes in G.
zeae and thus cannot determine whether the size of the
chromosome is playing a role in the segregation patterns
we observed or if aberrant segregation is reducing the
number of viable progeny produced. We also cannot
determine if the relatively high number of intact paren-
1458
J. E. Jurgenson et al.
tal linkage groups observed in the progeny is the result
of a lack of pairing and/or synapsis that could lead to
crossing over or if nonviable progeny result if crossing
over occurs other than in a few specific patterns. We
are currently testing some of these alternative explanations for our results.
Linkage groups with unusual properties: Linkage
group II has several unusual characteristics. The only
nonrecombinant haplotype for linkage group II is from
the Japanese parent, which is found in 33 of the 99
progeny (Table 1). The remaining 66 progeny also contain a small region near each end of this linkage group
that is predominantly Japanese genome in origin.
Among the progeny in which detectable recombination
has occurred on linkage group II, the central part of
the linkage group has a near 1:1 segregation ratio (Figure 2). Of the 66 recombinant progeny, 59 have an even
number of crossovers on linkage group II, with the two
and four crossover classes being approximately equally
frequent (Table 1). This pattern could result if there is
a large heterozygous inversion that includes most of the
central portion of the linkage group. Chromosomes
with odd numbers of crossovers within the inverted region would be duplicated for one region distal to the
inversion and deficient for the other. If the deficient
region carries essential genes, then the duplication/
deficiency progeny will be dead. The 7 progeny with an
odd number of crossovers on linkage group II all appear
to have an even number of crossovers within the putative
inverted region (a span of 220 cM from locus 2F to 2AJ)
and an odd number, usually a single, in one of the two
distal regions (Figure 3). The Japanese parent carries
the nit3 wild-type allele that was selected for in all of
the progeny. We expect nit3 to be in the small region
(locus 2C) for which all of the progeny have the Japanese
genome.
Linkage group IV recombination patterns also are
somewhat unusual. The 24 nonrecombinant progeny
for this linkage group are all of the Kansas type. It is
possible that the crossover type pattern between 4C and
4P (56 cM) could be due to a heterozygous inversion.
This linkage group also is fixed near one end at locus
4AD for the Kansas genome. This region is the only one
in the map that is 100% Kansas genome, and, therefore,
we predict that the nit1 gene maps on this linkage group
in or very near locus 4AD.
Trichothecene gene analyses: We mapped TRI5, and
presumably the rest of the trichothecene gene cluster,
near the middle of linkage group I in G. zeae. TRI5
encodes trichodiene synthase, the first step in the trichothecene toxin biosynthetic pathway (Hohn and van
Middlesworth 1986; Hohn and Beremand 1989;
Brown et al. 2001). The locus controlling toxin type
(DON or NIV) cosegregated with TRI5. This is the first
genetic proof that toxin type is controlled by a single
locus that is linked to the trichothecene gene cluster.
We did not give a new name to the toxin-type locus
because the responsible gene(s) in the trichothecene
gene cluster might already be named (Brown et al.
2001; Lee et al. 2001).
A locus controlling toxin amount was located on linkage group IV. This gene was not previously described
and was designated TOX1. Since this gene has a large
effect on toxin biosynthesis, it deserves further study.
We mapped several AFLP markers within 15 cM of TOX1
that could be useful for cloning the gene. Loci controlling red pigment production and perithecium formation also map near TOX1, but these distances and the
gene order might be affected by the putative inversion
on linkage group IV.
Map utilization: In future studies our map can be
utilized to locate genes of interest in several ways. If
mutations arise or are induced in one of the strains,
these mutations can be mapped by performing a new
cross between Z-3639 and R-5470 or their mutagenized
derivatives. Linkage to the AFLP markers on the map
should be readily determined by analysis with as few as
seven primer pairs. For example, the seven primer pairs
that use the EAA primer generate a skeleton map that
defines 70% of the total linkage map. Economically
important traits could include virulence, toxin production, competitive ability, and fungicide sensitivity. The
density of the map is high enough that it should be
possible to analyze these traits as quantitative trait loci
(QTL). If the gene of interest has been cloned or if it
is an expressed sequence tag (EST) sequence, either
hybridization with a PCR-amplified probe or the AFLP
mapping technique of Cato et al. (2001), which uses a
single 4-bp recognition-site restriction enzyme digest of
the ESTs, can be used to localize the gene within the
present mapping population. If the map is used to order
genomic libraries, then the process of sequencing the
G. zeae genome could be greatly simplified.
Our map provides markers with known linkage relationships that can be used for population studies. Such
studies are required to understand the current structure
and future changes in the pathogen’s populations and
their correlation with disease epidemics. AFLPs provide
a well-defined, relatively large set of markers that can
be used to monitor populations on a genome-wide basis.
These studies should more accurately reflect the population being studied since the bias created when linked
markers are treated as unlinked can be removed (Brändle et al. 1997).
The seven lineages described by O’Donnell et al.
(2000) appear genetically divergent based on the sequences of six genes. The fertility between various lineages needs to be quantified to help estimate the risk
of generating novel parasitic phenotypes when lineages
are commingled and to determine their degree of genetic isolation. We estimate that ⬍1% of the perithecia
were heterozygous outcrosses when Z-3639 and R5470
were mated. Up to 35% heterozygous perithecia were
produced in crosses of Z-3639 with two other Kansas
G. zeae Genetic Map
strains that belong to lineage VII (Bowden and Leslie
1999), so there may be lower fertility in crosses between
lineages. The map could be used to identify naturally
occurring hybrids between lineages and to estimate the
amount of genetic material within a hybrid strain that
originated from the different lineages.
On the basis of our AFLP study, the two parental
strains of the mapping population differ at ⵑ50% of
the observed bands. In the G. fujikuroi species complex
(Leslie et al. 2001; Marasas et al. 2001), we would
conclude that these strains were in the same species,
but perhaps not the same subspecies. [In the G. fujikuroi
species complex, strains in the same subspecies always
share 65% or more AFLP identity and those in different
species share ⬍40% AFLP identity. The only case in
which an intermediate value (55%) was observed (K.
Zeller and J. Leslie, unpublished data) is also a case
where some members of the two groups can occasionally
cross and produce perithecia with viable ascospores.]
However, many more G. zeae populations and related
species need to be examined before concluding that
the criteria applicable to G. fujikuroi also are applicable
to G. zeae.
The finding of at least two putative chromosome rearrangements in our cross also is suggestive of significant
genetic differentiation. Heterozygous inversions are postzygotic fertility blocks that reduce fertility and progeny
variability. These differences are certainly important in
Drosophila speciation (Anderson et al. 1991), but probably not in Neurospora speciation (Perkins 1997).
Within the G. fujikuroi species complex, ⵑ90% of the
RFLP markers tested remained on the same chromosome across six biological species (Xu et al. 1995). These
results indicate that translocations and transpositions
are not widespread, but provide no information on inversions, as an inversion would not alter the chromosome to which a marker hybridizes. To the degree that
chromosome rearrangements alter gene position, they
might alter expression due to specific position effects.
We have no evidence for this type of effect in Fusarium,
although rearrangements that affected toxin gene clusters or pathogenicity gene clusters might show such
effects. Limitations to recombination also could be an
effective means to lock particular allele combinations
into gene complexes that might have selective value and
could not easily be broken apart, e.g., the spore killer
complex in Neurospora (Raju 1994).
To test some of these hypotheses, follow-up studies
will be needed. Cytological studies have identified no
more than four chromosomes (Howson et al. 1963),
but studies with pulse-field chromosome separation are
likely to be more successful, given the relatively small
size of Fusarium chromosomes. Additional maps are
needed to confirm linkage group configurations and to
map other economically important traits, e.g., pathogenicity, host range, or toxin production, and to determine if the segregation distortion we observed is dis-
1459
torting our perception of the genomic organization of
G. zeae. For example:
i. Are the putative chromosome rearrangements we
observed peculiar to one of the strains in the present
cross?
ii. Is the suppression of crossing over a general property
of G. zeae outcrosses? Peculiar to interlineage crosses?
Specific for one (or both) of the strains used in the
present cross?
We thank Amy Beyer, Ann Clouse, and Amy Hanson for technical
assistance. J. E. Jurgenson was supported by a Professional Development Leave grant from the University of Northern Iowa. This work
was supported in part by U.S. Wheat and Barley Scab Initiative project
59-0790-9-029 and by the Kansas Agricultural Experiment Station. This
is contribution no. 02-67-J from the Kansas Agricultural Experiment
Station, Manhattan, KS.
LITERATURE CITED
Anderson, W., J. Arnold, D. Baldwin, A. Beckenbach, C. Brown et
al., 1991 Four decades of inversion polymorphism in Drosophila
pseudoobscura. Proc. Natl. Acad. Sci. USA 88: 10367–10371.
Bowden, R. L., and J. F. Leslie, 1992 Nitrate-nonutilizing mutants
of Gibberella zeae (Fusarium graminearum) and their use in determining vegetative compatibility. Exp. Mycol. 16: 308–315.
Bowden, R. L., and J. F. Leslie, 1999 Sexual recombination in
Gibberella zeae. Phytopathology 89: 182–188.
Brändle, U. E., U. A. Hammerli, J. M. McDermott and M. S. Wolfe,
1997 Interpreting population genetic data with the help of genetic linkage maps, pp. 157–171 in The Gene-for-Gene Relationship
in Plant-Parasite Interactions, edited by I. R. Crute, E. B. Holub
and J. J. Burdon. CAB International Press, Wallingford, UK.
Brasier, C., 2000 The rise of hybrid fungi. Nature 405: 134–135.
Brown, D. W., S. P. McCormick, N. J. Alexander, R. H. Proctor and
A. E. Desjardins, 2001 A genetic and biochemical approach
to study trichothecene diversity in Fusarium sporotrichioides and
Fusarium graminearum. Fungal Genet. Biol. 32: 121–133.
Carter, J. P., H. N. Rezanoor, A. E. Desjardins and P. Nicholson,
2000 Variation in Fusarium graminearum isolates from Nepal associated with their host of origin. Plant Pathol. 49: 452–460.
Cato, S. A., R. C. Gardner, J. Kent and T. E. Richardson, 2001
A rapid PCR-based method for genetically mapping ESTs. Theor.
Appl. Genet. 102: 296–306.
Chen, L.-F., G.-H. Bai and A. E. Desjardins, 2000 Recent advances
in wheat head scab research in China. National Agricultural Library (http://www.nal.usda.gov/pgdic/WHS/whsindex.html).
Correll, J. C., C. J. R. Klittich and J. F. Leslie, 1987 Nitrate
nonutilizing mutants of Fusarium oxysporum and their use in vegetative compatibility tests. Phytopathology 77: 1640–1646.
Gallegos, A., D. J. Jacobson, N. B. Raju, M. P. Skupski and D. O.
Natvig, 2000 Suppressed recombination and a pairing anomaly on the mating-type chromosome of Neurospora tetrasperma.
Genetics 154: 623–633.
Gonzalez, M., M. E. Z. Rodriguez, J. L. Jacabo, F. Hernandez,
J. Acosta et al., 1998 Characterization of Mexican isolates of
Colletotrichum lindemuthianum by using differential cultivars and
molecular markers. Phytopathology 88: 292–299.
Grell, R. F., 1962 A new model for secondary nondisjunction: the
role of distributive pairing. Genetics 47: 1737–1754.
Haber, J. E., 1998 Searching for a partner. Science 279: 823–824.
Hawley, R. S., K. S. McKim and T. Arbel, 1993 Meiotic segregation
in Drosophila melanogaster females: molecules, mechanisms, and
myths. Annu. Rev. Genet. 27: 281–317.
Hohn, T. M., and M. N. Beremand, 1989 Isolation and nucleotide
sequence of a sequiterpene cyclase gene from the trichotheceneproducing fungus Fusarium sporotrichioides. Gene 79: 131–138.
Hohn, T. M., and F. van Middlesworth, 1986 Purification and
characterization of the sesquiterpene cyclase trichodience syn-
1460
J. E. Jurgenson et al.
thase from Fusarium sporotrichioides. Arch. Biochem. Biophys. 251:
756–761.
Howson, W. T., R. C. McGinnis and W. L. Gordon, 1963 Cytological studies on the perfect stages of some species of Fusarium. Can.
J. Genet. Cytol. 5: 60–64.
Hua, J., L. L. Domier, S. Xuejen and F. L. Kolb, 2000 Combined
AFLP and RFLP mapping in two hexaploid oat recombinant
inbred populations. Genome 43: 94–101.
Kerényi, Z., K. Zeller, L. Hornok and J. F. Leslie, 1999 Molecular
standardization of mating type terminology in the Gibberella fujikuroi species complex. Appl. Environ. Microbiol. 65: 4071–4076.
Klittich, C. J. R., and J. F. Leslie, 1988 Nitrate reduction mutants of
Fusarium moniliforme (Gibberella fujikuroi). Genetics 118: 417–423.
Lander, E. S., P. Green, J. Abrahamson, A. Barlow, M. J. Daly
et al., 1987 Mapmaker: an interactive computer package for
constructing primary genetic linkage maps of experimental and
natural populations. Genomics 1: 174–181.
Lee, T., D. W. Oh, H.-S. Kim, J. Lee, Y.-H. Kim et al., 2001 Identification of deoxynivalenol- and nivalenol-producing chemotypes of
Gibberella zeae by using PCR. Appl. Environ. Microbiol. 67: 2966–
2972.
Leslie, J. F., F. J. Doe, R. D. Plattner, D. D. Shackelford and J. Jonz,
1992 Fumonisin B1 production and vegetative compatibility of
strains from Gibberella fujikuroi mating population “A” (Fusarium
moniliforme). Mycopathologia 117: 37–45.
Leslie, J. F., K. A. Zeller and B. A. Summerell, 2001 Icebergs and
speciation in species of Fusarium. Physiol. Mol. Plant Pathol. 59:
107–117.
Manly, K. F., and J. M. Olson, 1999 Overview of QTL mapping
software and introduction to Map Manager QT. Mamm. Genome
10: 327–334.
Marasas, W. F. O., P. E. Nelson and T. A. Toussoun, 1984 Toxigenic
Fusarium Species: Identity and Mycotoxicology. The Pennsylvania State
University Press, University Park, PA.
Marasas, W. F. O., J. P. Rheeder, S. C. Lamprecht, K. A. Zeller
and J. F. Leslie, 2001 Fusarium andiyazi sp.nov., a new species
from sorghum. Mycologia 93: 1203–1210.
McMullen, M. P., R. Jones and D. Gallenberg, 1997 Scab of wheat
and barley: a re-emerging disease of devastating impact. Plant
Dis. 81: 1340–1348.
Miao, V. P., S. F. Covert and H. D. Van Etten, 1991 A fungal gene
for antibiotic resistance on a dispensable (“B”) chromosome.
Science 254: 1773–1776.
Murray, M. G., and W. F. Thompson, 1980 Rapid isolation of high
molecular weight plant DNA. Nucleic Acids Res. 8: 4321–4325.
Nelson, P. E., T. A. Tousson and W. F. O. Marasas, 1983 Fusarium
Species: An Illustrated Manual for Identification. The Pennsylvania
State University Press, University Park, PA.
O’Donnell, K., H. C. Kistler, B. K. Tacke and H. H. Casper, 2000
Gene genealogies reveal global phylogeographic structure and
reproductive isolation among lineages of Fusarium graminearum,
the fungus causing wheat scab. Proc. Natl. Acad. Sci. USA 97:
7905–7910.
Perkins, D. D., 1997 Chromosomal rearrangements in Neurospora
and other filamentous fungi. Adv. Genet. 36: 239–398.
Purwantara, A., J. M. Barrins, A. J. Cozijnsen, P. K. Ades and B. J.
Howlett, 2000 Genetic diversity of the Leptosphaeria maculans
species complex from Australia, Europe and North America using
amplified fragment length polymorphism analysis. Mycol. Res.
104: 772–781.
Raju, N. B., 1994 Ascomycete spore killers: chromosomal elements
that distort genetic ratios among the products of meiosis. Mycologia 86: 461–473.
Sambrook, J., E. F. Fritsch and T. Maniatis, 1989 Molecular Cloning: A Laboratory Manual, Ed. 2. Cold Spring Harbor Laboratory
Press, Cold Spring Harbor, NY.
Scott, P. M. (Editor), 1995 AOAC official method 986.18. Deoxynivalenol in wheat. Gas chromatographic method, pp. 49–36–49.37
in Official Methods of the Association of Analytical Chemists, Vol. II,
Ed. 16. AOAC International, Gaithersburg, MD.
Seitz, L. M., W. D. Eustace, H. E. Mohr, M. D. Shogren and W. T.
Yamazaki, 1986 Cleaning, milling, and baking tests with hard
red winter wheat containing deoxynivalenol. Cereal Chem. 63:
146–150.
Tanaka, T., A. Hasegawa, S. Yamamota, U.-S. Lee, Y. Sugiura et
al., 1988 World-wide contamination of cereals by Fusarium mycotoxins nivalenol, deoxynivalenol, and zearalenone. 1. Survey
of 19 countries. J. Agric. Food Chem. 36: 979–983.
van der Lee, T., A. Robold, A. Testa, J. W. van ‘t Klooster and
F. Govers, 2001 Mapping of avirulence genes in Phytophthora
infestans with amplified fragment length polymorphisms selected
by bulk segregant analysis. Genetics 157: 949–956.
Vos, P., R. Hogers, M. Bleeker, M. Reijans, T. van de Lee et al.,
1995 AFLP: a new technique for DNA fingerprinting. Nucleic
Acids Res. 23: 4407–4414.
Vuylsteke, M., R. Mank, R. Antonise, E. Bastiaans, M. L. Senior
et al., 1999 Two high-density AFLP linkage maps of Zea mays L.:
analysis of distribution of AFLP markers. Theor. Appl. Genet.
99: 921–935.
Walker, S. L., S. Leath, W. M. Hagler, Jr. and J. P. Murphy, 2001
Variation among isolates of Fusarium graminearum associated with
Fusarium head blight in North Carolina. Plant Dis. 85: 404–410.
Weir, B. S., 1990 Genetic Data Analysis. Sinauer Associates, Sunderland, MA.
Windels, C. E., 2000 Economic and social impacts of Fusarium head
blight: changing farms and rural communities in the Northern
Great Plains. Phytopathology 90: 17–21.
Xu, J.-R., and J. F. Leslie, 1996 A genetic map of Gibberella fujikuroi
mating population A (Fusarium moniliforme). Genetics 143: 175–
189.
Xu, J.-R., K. Yan, M. B. Dickman and J. F. Leslie, 1995 Electrophoretic karyotypes distinguish the biological species of Gibberella
fujikuroi (Fusarium section Liseola). Mol. Plant-Microbe Interact.
8: 74–84.
Yin, X., P. Stam, C. J. Dourleijn and M. J. Kropff, 1999 AFLP
mapping of quantitative trait loci for yield-determining physiological characters in spring barley. Theor. Appl. Genet. 99: 244–253.
Young, W. P., J. M. Schupp and P. Keim, 1999 DNA methylation
and AFLP marker distribution in the soybean genome. Theor.
Appl. Genet. 99: 785–790.
Yun, S. H., T. Arie, I. Kaneko, O. C. Yoder and B. G. Turgeon,
2000 Molecular organization of mating type loci in heterothallic, homothallic, and asexual Gibberella/Fusarium species. Fungal
Genet. Biol. 31: 7–20.
Zeller, K. A., J. E. Jurgenson, E. M. El-Assiuty and J. F. Leslie,
2000 Isozyme and amplified fragment length polymorphisms
from Cephalosporium maydis in Egypt. Phytoparasitica 28: 121–130.
Communicating editor: R. H. Davis