Segregation distortion caused by weak hybrid necrosisin recombinant inbred lines of common wheat

Shigeo Takumi • Yoichi Motomura •

Julio Cesar Masaru Iehisa • Fuminori Kobayashi

Received: 4 July 2013 / Accepted: 18 October 2013 / Published online: 22 October 2013

� Springer Science+Business Media Dordrecht 2013

Abstract Segregation distortion of molecular markers is

closely related to hybrid incompatibility in progeny from

intraspecific crosses. Recent reports in higher plants have

demonstrated that hybrid sterility results in segregation

distortion at the causal gene regions in progeny of intra-

specific crosses. Ne1 and Ne2 complementary loci are

known to control hybrid necrosis in intraspecific crosses of

common wheat cultivars. Here, we examine the effect of a

weak necrosis allele Ne1w on the segregation ratio of

molecular markers in recombinant inbred lines (RILs) of

common wheat. Some RILs showed accelerated cell death

in the leaves at the heading stage due to the epistatic

interaction between two quantitative trait loci (QTL) on

chromosomes 5B and 2B. Chromosomal localization of

these QTL corresponding to Ne1w and Ne2 showed dis-

torted segregation ratios of assigned markers having

oppositely biased direction. Although the Ne1w and Ne2

interaction had no obvious effect on seed fertility, Ne1w

reduced completion of grain development under the Ne2-

homozygous background. This reduction might be one of

causes that induces segregation distortion in the 5B and 2B

chromosomal regions of RILs. The present study demon-

strated that weak hybrid necrosis has limited phenotypic

effects; it causes segregation distortion in progeny from

intraspecific crosses.

Keywords Epistatic interaction � Hybrid necrosis �QTL analysis � Reproductive isolation � Segregation

distortion � Triticum aestivum L.

Introduction

Distorted segregation, or significantly biased segregation

ratio, is frequently observed in crossed progenies. Under-

lying genetic causes for segregation distortion can be

hybrid incompatibility and non-functional gamete forma-

tion. Hybrid incompatibility between two diverging lin-

eages acts as a post-zygotic reproductive barrier, and it

plays a significant role in intraspecific differentiation and

plant speciation (Rieseberg and Willis 2007). The evolu-

tionary process for generating hybrid incompatibility is

simply explained by the Dobzhansky–Muller (DM) model,

which proposes two genetic models involving interlocus

epistasis and allelic interaction at a single locus (Bomblies

and Weigel 2007; Presgraves 2010). In higher plants, some

DM loci resulting in hybrid incompatibility have been

elucidated, and several causative genes of hybrid necrosis,

a well-known reproductive isolation mechanism in plant

species (Bomblies and Weigel 2007), have been identified

as disease resistance-related genes in Arabidopsis and let-

tuce (Bomblies et al. 2007; Jeuken et al. 2009; Alcazar

et al. 2009, 2010). Autoimmune response is generally

considered to trigger hybrid necrosis through epistatic

interactions among disease resistance-related genes in

hybrids (Bomblies and Weigel 2007). Recent studies in

Arabidopsis and rice have demonstrated that segregation

distortion is caused by hybrid sterility through epistatic

interaction of hybrid incompatibility genes or alleles (Bi-

kard et al. 2009; Mizuta et al. 2010; Yang et al. 2012).

Reciprocal gene loss of duplicated genes participates in the

S. Takumi (&) � Y. Motomura � J. C. M. Iehisa

Laboratory of Plant Genetics, Graduate School of Agricultural

Science, Kobe University, Nada-ku, Kobe 657-8501, Japan

e-mail: takumi@kobe-u.ac.jp

F. Kobayashi

Plant Genome Research Unit, National Institute of

Agrobiological Sciences, 2-1-2 Kannondai, Tsukuba, Ibaraki

305-8602, Japan

123

Genetica (2013) 141:463–470

DOI 10.1007/s10709-013-9745-2

induction of hybrid sterility (Bikard et al. 2009; Mizuta

et al. 2010). Many examples of pollen sterility in hybrid

plants are attributed to either allelic interaction at a single

DM locus or epistasis between two DM loci, and it has

been proven that hybrid sterility is a reproductive barrier in

rice between japonica and indica varieties and between

rice and its relatives (Long et al. 2008; Yamagata et al.

2010; Kubo et al. 2011). Thus, the hybrid sterility can

generally be expected to result in segregation distortion at

causal gene regions in progeny of intra- and interspecific

crosses of higher plants.

In intraspecific crosses of common wheat cultivars, the

loci Ne1 and Ne2 are known to control hybrid necrosis

(Tsunewaki 1960, 1970). These complementary genes are

located on chromosome arms 5BL and 2BS, respectively,

and the induction of necrotic cell death by Ne1–Ne2 is

called type I hybrid necrosis (Tsunewaki 1960, 1992; Chu

et al. 2006). Leaves of necrotic hybrid wheat have high

superoxide content and enhanced antioxidant enzyme

activity (Khanna-Chopra et al. 1998; Dalal and Khanna-

Chopra 2001; Sugie et al. 2007). Recent wheat breeding

programs have generated significantly biased frequencies

of Ne1 and Ne2 alleles among wheat populations. Increased

frequency of the Ne2 allele was observed recently in

European and American varieties, while increased fre-

quency of Ne1 was observed in the middle of Asia and

Africa (Pukhalskiy et al. 2000). This distribution of Ne1

and Ne2 alleles was proposed to result from the selection

for beneficial traits such as rust fungus resistance (Bom-

blies and Weigel 2007). Thus, the Ne1–Ne2 system func-

tions as a reproductive barrier contributing intraspecific

differentiation in common wheat.

Three and four alleles were classified based on the

strength of necrotic symptoms in the Ne1 and Ne2 loci,

respectively (Hermsen 1963a; Pukhalskiy et al. 2009).

Complementary interaction between the strong alleles,

Ne1s and Ne2s, induces severe necrotic cell death in F1

hybrid plants (Hermsen 1963a; Sugie et al. 2007), and the

necrotic symptoms are suppressed at high temperature

(Dhaliwal et al. 1986). F1 hybrid plants with two weak

alleles, Ne1w and Ne2w, showed necrotic symptoms in the

two youngest leaves at the end of the heading stage and

generally exhibited normal one-thousand kernel weight

values (Hermsen 1963a). The Ne1s–Ne2s interaction pro-

duced severe necrosis and was generally lethal in F1 plants

under normal growth condition. If selfed seeds (F2 gener-

ation) are produced from the severely necrotic F1 plants

under artificial conditions, necrotic and normal plants show

a 9:7 segregation ratio (Hermsen 1963a). Therefore, a

severely necrotic phenotype caused by the Ne1s–Ne2s

interaction is expected to strongly affect segregation ratios

at the loci around the causal gene-containing chromosomal

regions in crossed progeny. Moreover, Hermsen (1963a)

observed segregation of lethal alleles for a portion of F3

progeny resulting from crosses of the weak allele-con-

taining cultivars of common wheat, confirming the effect of

weak alleles of the hybrid necrosis genes on the segrega-

tion ratio. In our previous study, recombinant inbred lines

(RILs) of M808 and CS were established by the single seed

descent method in common wheat (Kobayashi et al. 2010).

Our objective in the present study was to elucidate the

effect of a weak necrosis allele on the segregation ratio of

molecular markers in the RILs of common wheat.

Materials and methods

Plant material

Two common wheat cultivars, Mironovskaya 808 (M808)

and Chinese Spring (CS), were used as parental cultivars

for the mapping population. M808, reported to be one of

the hardiest winter wheat cultivars (Veisz and Sutka 1990),

was bred in the Mironovska Institute, Ukraine, and has a

genotype of ne1Ne2s (Pukhalskiy et al. 2000). CS, a spring-

type wheat cultivar, has the Ne1wne2 genotype (Hermsen

1963a). The mapping population of 210 RILs was estab-

lished at the F7 generation by the single-seed descent

method from an F2 population derived from a M808 and

CS cross (Kobayashi et al. 2010). The parental cultivars

and RILs were grown individually in pots arranged ran-

domly in a Kobe University experimental field in the

2009–2010 season.

A common wheat variety S-615 has no dominant alleles

of Ne1 and Ne2. The near-isogenic lines (NILs) carrying

dominant alleles Ne1s and Ne2s are designated as Ne1-S615

and Ne2-S615, respectively (Tsunewaki and Koba 1979).

The Ne1s and Ne2s alleles originate from common wheat

cultivars Prelude and Kharkov, respectively (Tsunewaki

1960), and these dominant alleles were introduced into

S-615 genetic background through 10 backcrosses. The

NILs and their crossed progeny were grown individually in

pots arranged randomly in a greenhouse of Kobe Univer-

sity in the 2011–2012 season.

Phenotype evaluation

The degree of necrotic symptom development was desig-

nated as the position of the uppermost leaf showing

necrotic and/or senescence-related cell death on the leaf

blade at the heading stage. In plants with the strongest

symptoms, designated as degree 1, the flag leaf is withered

and shows necrotic cell death. The degree was determined

for the three earliest tillers of two plants, and the mean and

standard deviation was calculated for each RIL. In addi-

tion, the degree of necrotic symptom, selfed seed fertility

464 Genetica (2013) 141:463–470

123

and grain weight were measured for selected RILs, M808,

CS, S-615, Ne1-S615, Ne2-S615 and their F1 progeny.

Selfed seed fertility was measured using at least five spikes,

and grain weight was taken as the mean of at least 50

seeds; all parameters were determined on two plants from

each group.

QTL mapping

A linkage map of M808 and CS was previously constructed

using 210 RILs (Kobayashi et al. 2010). Genetic mapping

was performed using MAPMAKER/EXP version 3.0b

software (Lander et al. 1987). The threshold for log-like-

lihood (LOD) scores was set at 3.0, and genetic distances

were calculated with the Kosambi function (Kosambi

1944). Quantitative trait loci (QTL) analysis was carried

out by composite interval mapping using Windows QTL

Cartographer (ver. 2.5) software (http://statgen.ncsu.edu/

qtlcart/WQTLCart.htm; Wang et al. 2011) with the forward

and backward method with 1,000 repetitions. The per-

centage of phenotypic variation explained by a QTL for a

trait and any additive effect were also estimated. Epistatic

interactions between QTL were evaluated using the mul-

tiple interval mapping (MIM) method using Bayesian

Information Criteria (BIC-M3).

Analysis of segregation ratios of the marker genotypes

In the M808/CS RILs, segregation of each SSR marker

mapped on the linkage map was expected to be in a 1:1

ratio. At the 410 SSR loci, Chi square test probabilities for

segregation distortion were calculated against the predicted

1:1 segregation ratio.

Results

QTL analysis of necrotic cell death

Plants exhibiting feeble growth at the heading stage were

found among some of the 210 RILs of M808/CS; the feeble

plants generated more withered leaves more than normal

plants. Thus, RILs with feeble growth showed necrotic cell

death earlier than normal RILs. To evaluate the feebleness

or vigorousness in each RIL, the position of withered

leaves, which is where necrotic cell death occurs, at the

heading stage was scored. Mean values for the parental

cultivars, M808 and CS, were 6, indicating that senes-

cence-related cell death was visible on the sixth leaf and

younger leaves at the heading stage (Fig. 1). Leaves of

most RILs looked normally vigorous, whereas 16 RILs

with scores between 1 and 4 showed feeble growth and had

leaves that became withered earlier than those of the nor-

mal growth lines.

The QTL for the position of withered leaves was

detected in the M808/CS linkage map using composite

interval mapping. The linkage map showed 410 SSR loci,

and the total map length was 2,814.5 cM with an average

spacing of 6.9 cM between markers (Kobayashi et al.

2010). Only two QTL taken together showed significant

LOD scores (P \ 0.05) (Fig. 2): the QTL located on the

short arm of chromosome 2B had an LOD score of 3.63

and contributed 6.5 % of the phenotypic variance and the

QTL located on the long arm of chromosome 5B had an

LOD score of 5.08 and explained 9.0 % of the variation.

The additive effect of the 2B QTL (0.35) was opposite that

of the 5B QTL (-0.42). These results indicate that the

M808 allele of the 2B QTL and the CS allele of the 5B

QTL induce necrotic cell death. In addition, a significant

(P \ 0.001) additive-by-additive epistatic interaction

among the QTL by the MIM method was detected between

QTL identified by the MIM method, and the total of the

proportion of the phenotypic variance explained by the

QTL was 40.9 % (Table 1). The epistatic interaction

between the 2B and 5B QTL was related to the induction of

necrotic cell death in leaves at the heading stage in the RIL

population.

Segregation distortion in RILs of M808 and CS

To study segregation distortion of the SSR markers, the

segregation ratio of each mapped marker was measured

and tested for statistical deviation from the ratio of 1:1.

Significantly distorted segregation (P \ 0.05) was found in

Fig. 1 Frequency of RILs for the position of the uppermost withered

leaf at the heading stage in the M808/CS population

Genetica (2013) 141:463–470 465

123

a total of 41 out of the 410 loci in the M808/CS map

(Table 2). Most of the markers showing segregation dis-

tortion were located on the B genome chromosomes. Par-

ticularly, significant segregation distortion was observed in

three chromosomal regions from Xgwm429 to Xbarc18 on

the short arm of chromosome 2B, from Xbarc109 to

Xhbd149 on the long arm of chromosome 5B, and from

Xhbg268 to Xbarc24 and from Xgwm219 to Xhbe169 on

the long arm of chromosome 6B (Fig. 3). In the two

chromosomal regions on 2BS and 6BL, frequencies of the

Table 1 Epistatic interaction of QTL for the expression of necrosis

in the RILs

QTL (pair) Type Position

(cM)

Effect Effect

(%)

Empirical

P value

2B A 87.23 0.75 12.1 \0.001

5B A 58.45 -0.76 15.0 \0.001

2B 9 5B AA 0.79 13.8 \0.001

A, additive effect; AA, additive by additive effect

Table 2 Distribution of loci showing distorted segregation (P \ 0.05 in v2 test)

Three genomes A genome B genome D genome

Group CS M808 Chr. CS M808 Chr. CS M808 Chr. CS M808

1 (72) 3 1A (22) 1 1B (31) 2 1D (19)

2 (77) 9 3 2A (36) 1 2B (23) 9 2D (18) 2

3 (47) 2 3A (23) 1 3B (12) 3D (12) 1

4 (43) 2 1 4A (17) 1 1 4B (12) 1 4D (14)

5 (59) 8 5A (18) 1 5B (19) 7 5D (22)

6 (60) 12 6A (19) 1 6B (15) 11 6D (26)

7 (52) 1 7A (16) 7B (19) 1 7D (17)

410 26 15 3 4 22 9 1 2

The numbers of the distorted segregation loci on each chromosome are represented as biased to the CS-type allele (CS) and M808-type allele

(M808)

Total number of the analyzed SSR markers in each chromosome or homologous group is indicated in the parenthesis

Fig. 2 Linkage maps and QTL-

likelihood curves of LOD scores

showing the localization of

alleles on chromosomes 2B and

5B associated with withered

leaves. Genetic distances are

represented in centimorgans on

the left of each chromosome.

A LOD threshold score of 3.0 is

indicated by a dashed line

466 Genetica (2013) 141:463–470

123

homozygous M808-allele were significantly lower than

expected, while the homozygous CS allele was more fre-

quent. In contrast, the frequency of the homozygous CS

allele was significantly lower, and that of the homozygous

M808 allele was highly observed in the 5BL chromosomal

region.

Effects of the identified QTL on phenotypes

To examine genetic relationships between the 2BS and 5BL

QTL on the position of withered leaves and the presence of

type I hybrid necrosis genes Ne1 and Ne2, eight RILs were

selected from the 210 RIL of M808/CS based on SSR

marker genotypes in the 2BS and 5BL QTL regions

(Table 3). Two RILs, #129 and #138, had the M808 allele in

the 2B QTL and the CS allele in the 5B QTL and exhibited

withered flag leaves at the heading stage, indicating the

occurrence of accelerated cell death in the leaves, while the

other six RIL showed a normal growth phenotype (Table 3).

Most of the F1 plants of the four RILs with CS alleles in

the 2B QTL showed normal phenotypes with respect to the

position of the uppermost withered leaf, high selfed seed

fertility, and high grain weight (Table 4). Only when two

RILs with CS alleles at both the 2B and 5B QTL were

crossed with Ne2-S615 individuals, the F1 plants exhibited

accelerated cell death in the leaves, but high seed fertility

and normal grain weight were observed. On the other hand,

only for two RILs with M808 alleles in both the 2B and 5B

QTL crossed with Ne1-S615, both accelerated cell death

along with reduced seed fertility and collapsed seeds was

observed in the F1 plants.

To confirm the genetic relationships between Ne1 and

Ne2, the parental cultivars were crossed with the Ne1 and

Ne2 NILs, and the resulting F1 phenotypes were com-

pared. F1 plants produced by the cross between Ne1-S615

and Ne2-S615 showed severe hybrid necrosis, and no

selfed progeny were obtained from the F1 plants

(Table 5). F1 plants produced by the cross of Ne1-S615

and M808 also showed accelerated cell death in the

leaves first, followed by reduced fertility and collapsed

seeds. Among other F1 plants, no obvious abnormalities

were observed.

Fig. 3 Frequency of CS-type alleles for each SSR marker on linkage

maps of chromosomes 2B, 5B and 6B. Open diamonds indicate SSR

markers with significantly distorted segregation, and closed ones

indicate SSR markers with normal segregation frequencies of the CS

alleles

Table 3 Postulated genotypes at two QTL in eight selected RILs

RIL # Position of a

withered leaf

2B QTL (Ne2) 5B QTL (Ne1)

Xgwm374 Xwmc474 Xbarc361-2 Xbarc230 Xwmc540 Xbarc74 Xgwm213

10 7 M808 M808 M808 M808 M808 M808 M808

31 8 M808 M808 M808 M808 M808 M808 M808

27 7 CS CS CS CS CS CS CS

41 7 CS CS CS CS CS CS CS

80 7 CS CS CS CS M808 M808 M808

142 7 CS CS CS CS M808 M808 M808

129 1 M808 M808 M808 M808 CS CS CS

138 1 M808 M808 M808 M808 CS CS CS

Alleles with positive effects are underlined

Genetica (2013) 141:463–470 467

123

Discussion

The feeble growth of plants among the 210 RILs of M808/

CS was due to epistatic interaction between the 2BS and

5BL QTL determining the position of the uppermost

withered leaf (Table 1). The locations of 2BS and 5BL

chromosome QTL in the M808/CS map (Fig. 2) corre-

sponded to those of Ne2 and Ne1, respectively, in the SSR

map previously reported by Chu et al. (2006). Phenotypic

characteristics in common among RILs showing feeble

growth were the same as the previously reported weak

phenotypes of type I hybrid necrosis (Hermsen 1963a). The

M808-type allele in the 2B QTL and CS-type allele in the

5B QTL positively affected the acceleration of cell death in

leaves. These results clearly indicate that M808 and CS

possess Ne2 and Ne1, respectively. The F1 plants produced

in the cross between Ne1-S615 and M808 showed necrotic

cell death at the seedling stage while the F1 plants produced

in the cross between Ne2-S615 and M808 were normal

(Table 5), which supports the previously reported M808

genotype for type I necrosis of ne1ne1Ne2sNe2s (Pukhal-

skiy et al. 2000). On the other hand, the F1 plants produced

in the cross between CS and Ne2-S615 showed normal

growth, implying that CS contains the weak allele, Ne1w,

as previously reported (Hermsen 1963a). Therefore, the CS

genotype could be postulated to be Ne1wNe1wne2ne2.

Normal plant growth and high seed fertility in M808/CS F1

plants (Ne1wne1Ne2sne2) could be explained to be due to

the weak allele of Ne1 from CS.

The M808/CS RILs with genotype Ne1wNe1wNe2sNe2s

showed severely accelerated cell death in the leaves at the

heading stage and reduced grain weight but high seed

fertility (Tables 3, 4). The F1 plants between these RILs

and Ne1-S615 with the genotype of Ne1sNe1wNe2sne2

were lethal hybrids, while F1 plants between these RILs

and Ne2-S615, Ne1wne1Ne2sNe2s, showed the same phe-

notype as the RILs (Table 4). Although RILs with the

genotype Ne1wNe1wne2ne2 showed normal growth and

fertility, accelerated cell death was observed in the leaves

of F1 plants produced in the cross between these RILs and

Ne2-S615, with the genotype Ne1wne1Ne2sne2, and the

severity of cell death was less than that in the F1 plants with

the genotype, Ne1wne1Ne2sNe2s. These results indicate

that phenotypes of Ne1w-homozygous and -heterozygous

plants are identical, at least under the Ne2s-homozygous

background. Epistatic interaction between Ne1s and Ne2s

induced severe hybrid necrosis in F1 plants, and the F1

plants with genotype Ne1sne1Ne2sne2 were sometimes

semi-lethal with the production of collapsed seeds as

observed in RIL#10/Ne1-S615 and RIL#31/Ne1-S615

(Table 4). The Ne1-S615/Ne2-S615 F1 plants, with geno-

type Ne1sne1Ne2sne2, were lethal hybrids, with lethality

determined by the genetic background. On the other hand,

the interaction between Ne1w and Ne2s affected only

necrotic cell death in leaves at the heading stage. There

were no obvious effects of the interaction between Ne1w

and Ne2s on seed fertility and grain weight. Hermsen

(1963a) also reported almost normal one-thousand grain

weight in F1 plants with weak hybrid necrosis due to Ne1w.

Thus, phenotypic effects of Ne1w were quite limited

compared to those of Ne1s due to the interaction with Ne2s.

Regions with continuous segregation distortion were on

chromosomes 2B, 5B and 6B in the M808/CS RILs

(Table 2; Fig. 3). The distorted regions of chromosomes

2B and 5B corresponded to Ne2 and Ne1, respectively. The

biased directions of segregation distortion were opposite

Table 4 Effect of epistatic interaction of two QTL on three traits in

selected RILs and F1 plants

Selected RILs and

their F1 plants

Position of a

withered leaf

Selfed seed

fertility (%)

Grain

weight (mg)

RIL#10 (self) 5 93.32 36.96

RIL#10/S-615 5 96.0 36.17

RIL#10/Ne1-S615 1 52.5 3.27

RIL#10/Ne2-S615 5 96.25 34.60

RIL#31 (self) 5.5 97.9 37.44

RIL#31/S-615 5.5 97 40.7

RIL#31/Ne1-S615 1 18.64 3.77

RIL#31/Ne2-S615 6 97 41.37

RIL#27 (self) 5 95.34 39.16

RIL#27/S-615 5 96.18 44.78

RIL#27/Ne1-S615 5 96.10 48.04

RIL#27/Ne2-S615 3 96.24 42.33

RIL#41 (self) 6 95.20 40.51

RIL#41/S-615 5 98.50 39.53

RIL#41/Ne1-S615 5 98.44 38.10

RIL#41/Ne2-S615 2 97.12 39.55

RIL#80 (self) 6 98.34 39.88

RIL#80/S-615 5 91.50 43.26

RIL#80/Ne1-S615 5 90.44 48.23

RIL#80/Ne2-S615 5.5 93.40 45.57

RIL#142 (self) 6 96.42 48.49

RIL#142/S-615 5 98.50 45.42

RIL#142/Ne1-S615 5.5 98.98 45.48

RIL#142/Ne2-S615 5 98.38 43.97

RIL#129 (self) 1 95.78 36.23

RIL#129/S-615 5.5 98.40 44.39

RIL#129/Ne1-S615 1 n.d. n.d.

RIL#129/Ne2-S615 2 97.92 41.31

RIL#138 (self) 1 91.22 26.91

RIL#138/S-615 5 97.34 45.30

RIL#138/Ne1-S615 1 n.d. n.d.

RIL#138/Ne2-S615 1 94.06 24.92

n.d. not determined

468 Genetica (2013) 141:463–470

123

for the two chromosomal regions; high frequencies of the

CS-type alleles were observed in the 2B markers from

Xgwm429 to Xbarc18 and high frequencies of the M808-

type alleles were observed from Xbarc109 to Xhbd149 on

5B. These observations strongly suggest that an epistatic

interaction between Ne1w and Ne2s results in segregation

distortion on the 2B and 5B chromosomal regions. Sig-

nificantly higher frequencies of the CS-type alleles on the

2B chromosome can be explained by the contribution of

Ne2s from M808, and the significantly lower frequencies of

the CS-type alleles on the 5B chromosome can be

explained by the contribution of Ne1w from CS. The

interaction between Ne1s and Ne2s resulted not only in

severe hybrid necrosis but also in remarkably reduced

fertility, as well as the segregation distortion around the

Ne1s and Ne2s chromosomal regions in the progeny. Grain

weight was reduced while the fertility of selfed seeds was

normal in two RILs with the M808 allele in the 2B QTL

and the CS allele in the 5B QTL (Table 4). The F1 plants

produced in the crosses of each of these two RILs and Ne2-

S615 showed high seed fertility, although accelerated cell

death was observed. When the same RILs were crossed

with Ne1-S615, the F1 plants failed to develop normal

spikes due to severely accelerated cell death and no selfed

progeny could be obtained from the F1 plants. Although the

interaction between Ne1w and Ne2s had no obvious effects

on seed fertility, plants with Ne1w showed reduced grain

completion under the Ne2s-homozygous background. Thus,

the Ne1wne1Ne2sNe2s and Ne1wNe1wNe2sNe2s plants

induced hybrid necrosis symptom and reduced grain

weight, which might be one of causes of segregation dis-

tortion in the 2B and 5B chromosomal regions in M808/CS

progeny. Previously, abnormal segregation was reported

even in the progeny of crosses between plants with Ne1w

and Ne2w alleles (Hermsen 1963b). Not only strong alleles

but also weak ones for hybrid necrosis appear to cause

distorted segregation of the necrosis genes and their linked

alleles.

Genetic incompatibility resulting from interactions

among DM model genes represents a potential source of

postzygotic reproductive barriers (Bomblies and Weigel

2007). Recent studies in Arabidopsis and rice showed that

two-way interactions in distorted marker segregation were

significantly detected between the causal DM gene regions,

implying that hybrid sterility induces segregation distortion

at the causal gene regions in the progeny of intraspecific

and intersubspecific crosses (Bikard et al. 2009; Mizuta

et al. 2010). The interaction of the DM loci for hybrid

sterility occurs among the paralogs of an essential gene in

Arabidopsis and rice (Bikard et al. 2009; Mizuta et al.

2010). The present study demonstrated that although the

phenotypic effects are limited for weak hybrid necrosis, the

effects are evident in the segregation distortion in progeny

from the intraspecific cross as well as in hybrid sterility.

Many common wheat cultivars and landraces contain either

Ne1 or Ne2 (Hermsen 1963b, c; Pukhalskiy et al. 2000,

2009). The prevalence of Ne2 has particularly increased in

modern wheat cultivars due to selection for beneficial

linked traits, including rust fungus resistance (Pukhalskiy

et al. 2000; Bomblies and Weigel 2007). For the intro-

gression of agriculturally important genes from Ne1-con-

taining landraces to modern cultivars, segregation

distortion at the Ne1 chromosomal region would be a

barrier if the target genes were located near Ne1. Devel-

opment of tightly linked molecular markers for Ne1 and

Ne2 should be required for the separation of Ne1 and Ne2

from the target genes.

In a large region of chromosome 6B, segregation dis-

tortion was observed in the M808/CS RILs, and the fre-

quencies of CS-type alleles were significantly higher. In

contrast to Ne1–Ne2 hybrid necrosis, however, the QTL on

chromosome 6B solely causes segregation distortion

without any genetic interaction with other loci. Hybrid

sterility caused by a single genetic locus has been reported

in japonica–indica hybrids of rice. A killer-protector sys-

tem at the rice S5 locus, which is the location of three

tightly linked genes, regulates fertility in intersubspecific

hybrid rice, specifically affecting marker segregation ratios

in the S5 region (Yang et al. 2012). Another hybrid male

sterility gene, S24, acts as a pollen killer gene and strongly

affects male sterility and segregation distortion in the

progeny of the intersubspecific rice hybrids (Kubo et al.

2011). The allelic interactions at a single locus also result

Table 5 Effect of the epistatic interaction of two QTL on three traits

in the NILs of Ne1-S615 and Ne2-S615 and F1 plants

Cultivars, NILs

and their F1 plants

Position of a

withered leaf

Selfed seed

fertility (%)

Grain weight

(mg)

M808 6 95.65 41.88

CS 5 98.14 40.91

S-615 5 95.0 42.63

Ne1-S615 5 93.35 43.49

Ne2-S615 5 95.42 45.71

CS/S-615 5 99.17 45.12

M808/S-615 5 99.50 47.91

M808/CS 5 93.76 47.58

Ne1-S615/S-615 5 94.53 42.34

Ne1-S615/M808 1 9.37 22.50

Ne1-S615/CS 5 91.80 48.18

Ne2-S615/S-615 5 94.43 42.75

Ne2-S615/M808 5 98.48 47.39

Ne2-S615/CS 4 96.63 43.96

Ne1-S615/Ne2-S615 1 n.d. n.d.

n.d. not determined

Genetica (2013) 141:463–470 469

123

in segregation distortion at the affected chromosomal

region. A pollen killer gene, Ki, of CS was previously

reported on the long arm of chromosome 6B (Sears and

Loegering 1961). Therefore, the segregation distortion

observed on chromosome 6B might be due to the pollen

killer gene. The wheat SSR markers showing distorted

segregation were widely distributed in the 6B chromo-

somal region, indicating that Ki strongly affects segrega-

tion distortion in the progeny of intraspecific wheat crosses.

However, there is no mapping study of Ki to the best of our

knowledge. Further genetic studies are required to clarify

the relationship between the pollen killer gene and segre-

gation distortion on chromosome 6B.

Acknowledgments The authors thank emeritus professor Dr. Koi-

chiro Tsunewaki for helpful discussion and supplying seeds of cul-

tivars Ne1-S615 and Ne2-S615. This work was supported by Grants-

in-Aid for Scientific Research (B) No. 21380005 and No. 25292008 to

ST from the Ministry of Education, Culture, Sports, Science and

Technology (MEXT) of Japan.

References

Alcazar R, Garcıa AV, Parker JE, Reymond M (2009) Incremental

steps toward incompatibility revealed by Arabidopsis epistatic

interactions modulating salicylic acid pathway activation. Proc

Natl Acad Sci USA 106:334–339

Alcazar R, Garcıa AV, Kronholm I, de Meaux J, Koornneef M, Parker

JE, Reymond M (2010) Natural variation at Strubbelig Receptor

Kinase 3 derives immune-triggered incompatibilities between

Arabidopsis thaliana accessions. Nat Genet 42:1135–1139

Bikard D, Patel D, Le Mette C, Giorgi V, Camilleri C, Bennett MJ,

Loudet O (2009) Divergent evolution of duplicate genes leads to

genetic incompatibilities within A. thaliana. Science 323:623–626

Bomblies K, Weigel D (2007) Hybrid necrosis: autoimmunity as a

potential gene-flow barrier in plant species. Nat Rev Genet

8:382–393

Bomblies K, Lempe J, Epple P, Warthmann N, Lanz C, Dangl JL,

Weigel D (2007) Autoimmune response as a mechanism for

Dobzhansky–Muller-type incompatibility syndrome in plants.

PLoS Biol 5:e236

Chu CG, Faris JD, Friesen TL, Xu SS (2006) Molecular mapping of

hybrid necrosis genes Ne1 and Ne2 in hexaploid wheat using

microsatellite markers. Theor Appl Genet 112:1374–1381

Dalal M, Khanna-Chopra R (2001) Differential response of antiox-

idant enzymes in leaves of necrotic wheat hybrids and their

parents. Physiol Plant 111:297–304

Dhaliwal HS, Sharma SK, Randhawa AS (1986) How to overcome

hybrid necrosis in wheat? Wheat Inf Serv 61:27–28

Hermsen JGT (1963a) The genetic basis of hybrid necrosis in wheat.

Genetica 33:245–287

Hermsen JGT (1963b) Hybrid necrosis as a problem for the wheat

breeder. Euphytica 12:1–16

Hermsen JGT (1963c) Sources and distribution of the complementary

genes for hybrid necrosis in wheat. Euphytica 12:147–160

Jeuken MJW, Zhang NW, McHale LK, Pelgrom K, den Boer E,

Lindhout P, Michelmore RW, Visser RGF, Niks RE (2009) Rin4

causes hybrid necrosis and race-specific resistance in an

interspecific lettuce hybrid. Plant Cell 21:3368–3378

Khanna-Chopra R, Dalal M, Kumar PG, Laloraya M (1998) A genetic

system involving superoxide causes F1 necrosis in wheat

(Triticum aestivum L.). Biochem Biophys Res Commun

248:712–715

Kobayashi F, Takumi S, Handa H (2010) Identification of quantitative

trait loci for ABA responsiveness at seedling stage associated

with ABA-regulated gene expression in common wheat. Theor

Appl Genet 121:629–641

Kosambi DD (1944) The estimation of map distance from recombi-

nation values. Ann Eugen 12:172–175

Kubo T, Yoshimura A, Kurata N (2011) Hybrid male sterility in rice

is due to epistatic interactions with a pollen killer locus. Genetics

189:1083–1092

Lander ES, Green P, Abrahamson J (1987) MAPMAKER: an

interactive computer package for constructing primary genetic

linkage maps of experimental and natural populations. Genomics

1:174–181

Long Y, Zhao L, Niu B, Su J, Wu H, Chen Y, Zhang Q, Guo J,

Zhuang C, Mei M, Xia J, Wang L, Wu H, Liu YG (2008) Hybrid

male sterility in rice controlled by interaction between divergent

alleles of two adjacent genes. Proc Natl Acad Sci USA

105:18871–18876

Mizuta Y, Harushima Y, Kurata N (2010) Rice pollen hybrid

incompatibility caused by reciprocal gene loss of duplicated

genes. Proc Natl Acad Sci USA 107:20417–20422

Presgraves DC (2010) The molecular evolutionary basis of species

formation. Nat Rev Genet 11:175–180

Pukhalskiy VA, Martynov SP, Dobrotvorskaya TV (2000) Analysis of

geographical and breeding-related distribution of hybrid necrosis

genes in bread wheat (Triticum aestivum L.). Euphytica

114:233–240

Pukhalskiy VA, Udachin RA, Bilinskaya EN (2009) Hybrid necrosis

genes in aboriginal wheats of Middle Asia in the light of the

problem of the primary centers of biodiversity of the Triticum L.

genus. Euphytica 165:533–543

Rieseberg LH, Willis JH (2007) Plant speciation. Science

317:910–914

Sears ER, Loegering WQ (1961) A pollen-killing gene in wheat.

Genetics 46:897

Sugie A, Murai K, Takumi S (2007) Alteration of respiration capacity

and transcript accumulation levels of alternative oxidase genes in

necrosis lines of common wheat. Genes Genet Syst 82:231–239

Tsunewaki K (1960) Monosomic and conventional analyses in

common wheat. III. Lethality. Jpn J Genet 35:71–75

Tsunewaki K (1970) Necrosis and chlorosis genes in common wheat

and its ancestral species. Seiken Ziho 22:67–75

Tsunewaki K (1992) Aneuploid analyses of hybrid necrosis and hybrid

chlorosis in tetraploid wheats using the D genome chromosome

substitution lines of durum wheat. Genome 35:594–601

Tsunewaki K, Koba T (1979) Production and genetic characterization

of the co-isogenic lines of a common wheat Triticum aestivum

cv. S-615 for ten major genes. Euphytica 28:579–592

Veisz O, Sutka J (1990) Frost resistance studies with wheat in natural

and artificial conditions. In: Panayotov I, Pavlova S (eds)

Proceedings of the international symposium on cereal adaptation

to low temperature stress. Albena, Bulgaria, pp 12–17

Wang S, Basten CJ, Zeng ZB (2011) Windows QTL cartographer 2.5.

Department of Statics, North Carolina State University, Raleigh,

NC. http://statgen.ncsu.edu/qtlcart/WQTLCart.htm

Yamagata Y, Yamamot E, Aya K, Win KT, Doi K, Ito T, Kanamori

H, Wu J, Matsumoto T, Matsuoka M, Ashikari M, Yoshimura A

(2010) Mitochondrial gene in the nuclear genome induces

reproductive barrier in rice. Proc Natl Acad Sci USA

107:1494–1499

Yang J, Zhao X, Cheng K, Du H, Ouyang Y, Chen J, Qiu S, Huang J,

Jiang Y, Jiang L, Wang J, Zu C, Li X, Xhang Q (2012) A killer-

protector system regulates both hybrid sterility and segregation

distortion. Science 337:1336–1340

470 Genetica (2013) 141:463–470

123