diversifying sunﬂower germplasm by integration and mapping

INVESTIGATION

Diversifying Sunflower Germplasm by Integrationand Mapping of a Novel Male Fertility

Restoration GeneZhao Liu,* Dexing Wang,† Jiuhuan Feng,* Gerald J. Seiler,* Xiwen Cai,* and Chao-Chien Jan‡,1

*Department of Plant Sciences, North Dakota State University, Fargo, North Dakota 58108, †Liaoning Academy of AgriculturalSciences, Shenyang, 110161, China, and ‡U.S. Department of Agriculture-Agricultural Research Service, Northern Crop Science

Laboratory, Fargo, North Dakota 58102

ABSTRACT The combination of a single cytoplasmic male-sterile (CMS) PET-1 and the corresponding fertility restoration (Rf) gene Rf1 isused for commercial hybrid sunflower (Helianthus annuus L., 2n = 34) seed production worldwide. A new CMS line 514A was recentlydeveloped with H. tuberosus cytoplasm. However, 33 maintainers and restorers for CMS PET-1 and 20 additional tester lines failed torestore the fertility of CMS 514A. Here, we report the discovery, characterization, and molecular mapping of a novel Rf gene for CMS514A derived from an amphiploid (Amp H. angustifolius/P 21, 2n = 68). Progeny analysis of the male-fertile (MF) plants (2n = 35)suggested that this gene, designated Rf6, was located on a single alien chromosome. Genomic in situ hybridization (GISH) indicatedthat Rf6 was on a chromosome with a small segment translocation on the long arm in the MF progenies (2n = 34). Rf6 was mapped tolinkage group (LG) 3 of the sunflower SSR map. Eight markers were identified to be linked to this gene, covering a distance of 10.8 cM. Twomarkers, ORS13 and ORS1114, were only 1.6 cM away from the gene. Severe segregation distortions were observed for both the fertility traitand the linked marker loci, suggesting the possibility of a low frequency of recombination or gamete selection in this region. This studydiscovered a new CMS/Rf gene system derived fromwild species and provided significant insight into the genetic basis of this system. This willdiversify the germplasm for sunflower breeding and facilitate understanding of the interaction between the cytoplasm and nuclear genes.

THE combination of cytoplasmic male-sterility (CMS) andcorresponding fertility restoration (Rf) genes has been

widely utilized for large-scale hybrid seed production ofmany crops, including cultivated sunflower (Helianthus annuusL., 2n = 34) (Serieys 1996; Horn et al. 2003). For over 40years, the hybrid sunflower seed industry has largely reliedon a single CMS, CMS PET-1, discovered from wild H. pe-tiolaris subsp. petiolaris Nutt. and its corresponding fertilityrestoration gene Rf1 (Leclercq 1969; Dominguez-Gimenezand Fick 1975; Miller and Fick 1997; Horn et al. 2003;Jan and Vick 2007). Alternative CMS/Rf gene systems couldexpand the diversity of the sunflower crop and reduce therisks inherent with using a single CMS/Rf system. Also, iden-tification and characterization of additional CMS/Rf gene

systems will enrich knowledge of the interactions betweencytoplasm and nuclear genes.

Seventy-two sunflower CMS sources have been identified(Serieys 2005), but only about a half of them have knowncorresponding Rf genes. Generally, one to four dominant Rfgenes are needed for fertility restoration (Serieys 1996).However, only seven Rf genes have been mapped, i.e., Rf1,Msc1, Rf3-RHA 340, Rf3-RHA 280, and Rf5 for CMS PET-1,Rf4 for a new alloplasmic CMS GIG2, and Rf-PEF1 for CMSPEF1 (Gentzbittel et al. 1995; Jan et al. 1998; Horn et al.2003; Abratti et al. 2008; Feng and Jan 2008; Schnabel et al.2008; Yue et al. 2010; Liu et al. 2012; Qi et al. 2012). Rf1was mapped to linkage group (LG) 6 on the RFLP map ofGentzbittel et al. (1995), and to LG 2 by Jan et al. (1998).This gene was also mapped to LG 13 of the SSR map, as wellas a recently mapped Rf5 gene, which is from a restorer lineRf ANN-1742 (Yu et al. 2003; Horn et al. 2003; Kustereret al. 2005; Yue et al. 2010; Qi et al. 2012). Msc1 was map-ped to LG 12 of the RFLP map of Gentzbittel et al. (1999).Both Rf3-RHA 340 and Rf3-RHA 280 were mapped to LG 7 ofthe SSR map (Abratti et al. 2008; Liu et al. 2012), and Rf4

Copyright © 2013 by the Genetics Society of Americadoi: 10.1534/genetics.112.146092Manuscript received September 21, 2012; accepted for publication December 22, 2012Supporting information is available online at http://www.genetics.org/lookup/suppl/doi:10.1534/genetics.112.146092/-/DC1.1Corresponding author: USDA–ARS–NCSL, 1605 Albrecht Blvd. N, Fargo, ND 58102-2765. E-mail: [email protected]

Genetics, Vol. 193, 727–737 March 2013 727

http://www.genetics.org/lookup/suppl/doi:10.1534/genetics.112.146092/-/DC1


mailto:[email protected]

was reported to be on LG 3 of the SSR map (Feng and Jan2008). The Rf-PEF1 gene was mapped to an AFLP linkagegroup that differed from LG 13 of the SSR map (Schnabelet al. 2008). Several Rf genes have been cloned from Arabi-dopsis, radish, rice, maize, and Petunia (Cui et al. 1996; Liuet al. 2001; Bentolila et al. 2002; Brown et al. 2003; Wanget al. 2006; Gillman et al. 2007). Extensive effort has beenmade to clone Rf1 in sunflower using a map-based cloningstrategy (Horn et al. 2003).

Amphiploids are derived from interspecific or interge-neric crosses by chromosome doubling of the F1 hybrids withcolchicine or by the spontaneous merging of two or moredifferentiated genomes. They have been used as an impor-tant “bridge” for transferring disease-resistance genes, abi-otic stress-resistance/-tolerance genes, and other genes fromwild relatives of wheat (Dvorák et al. 1988; Jiang et al.1994; Colmer et al. 1995; Martín et al. 1999; Solimanet al. 2001), rye (Wojciechowska and Pudelska 2005; Islamet al. 2007; Kang et al. 2011; Malik et al. 2011), and triticale(Kwiatek et al. 2012). Amphiploids are also useful for study-ing the evolution and genetic diversity within a genus, suchas Arabidopsis (Nasrallah et al. 2000), maize (Poggio et al.2005; González et al. 2006), wheat (Kumar et al. 2010), andBrassica (Allender and King 2010; Bansal et al. 2012).

Several interspecific sunflower amphiploids have beenproduced via embryo rescue and colchicine treatment withsuccessful gene transfer reported (Jan and Chandler 1989;Jan and Fernandez-Martinez 2002; Pérez-Vich et al. 2002;Feng and Jan 2008). Broomrape-resistance genes for race Fin Spain were transferred from several wild Helianthus spe-cies into cultivated sunflower using interspecific amphi-ploids (Jan and Fernandez-Martinez 2002; Pérez-Vichet al. 2002). Amphiploids have shown resistance to Sclero-tinia, a major fungal disease of sunflower (Jan et al. 2006;Feng et al. 2007). Also, Feng and Jan (2008) detected a newtype of CMS, CMS GIG2, in backcrossed progenies of anamphiploid (Amp) of H. giganteus 1934/HA 89. They iden-tified the Rf4 gene for this CMS in H. maximiliani 1631utilizing an Amp NMS HA 89/H. maximiliani 1631.

Recently, a CMS line, CMS 514A, derived from the crossbetween H. tuberosus and an inbred line 7718B, was devel-oped at the Liaoning Academy of Agricultural Sciences, Liaon-ing, China, but no Rf gene has been identified. Thirty-threemaintainer and restorer lines from five countries, as well as20 tester lines from the U.S. Department of Agriculture–

Agricultural Research Service (USDA–ARS) Northern CropScience Laboratory that are commonly used for Rf gene de-tection, failed to restore fertility in CMS 514A. This sug-gested the uniqueness of this CMS compared to other CMSsystems, including CMS PET-1, CMS CMG1, CMS CMG2,and CMS CMG3 (Wang et al. 2007). The objectives of thisstudy were to: (1) identify the Rf gene for CMS 514A fromfive interspecific amphiploids, and a hexaploid H. californi-cus (PI 664602); (2) introgress the Rf gene into a cultivatedsunflower background using traditional crossing and back-crossing method and study the inheritance of the Rf gene;(3) conduct mitotic cytogenetic studies and genomic in situhybridization (GISH) to characterize the alien chromosomeor segments in the progenies; and (4) map the Rf gene usingSSR and expressed sequence tag (EST)–SSR markers.

Materials and Methods

Plant materials

Five interspecific amphiploids (Amp H. atrorubens/HA 89,AmpH. mollis/P 21, AmpH. cusickii/P 21, AmpH. grosseserratus/P 21, and Amp H. angustifolius/P 21, 2n = 68), and the F1progeny of hexaploid H. californicus/HA 89 (2n = 68), werecrossed with CMS 514A in 2003. The male-fertile (MF) F1plants from these crosses were backcrossed with HA 89 andHA 821 to transfer the Rf gene into a cultivated background.HA 89 and HA 821 are oilseed maintainer lines publiclyreleased by USDA. HA 821 was used to increase the diversityof the background instead of using HA 89 only.

Mitotic chromosome counting and GISH

Root tips collected from seedlings were placed in distilledwater at 2� for 18 hr and fixed in ethanol:acetic acid (V:V) =3:1. Chromosome numbers in root tip cells were determinedfor the individual plants in each generation using the stan-dard Feulgen staining method. The MF plants with 2n = 35and 34 derived from the cross of CMS 514A · Amp H.angustifolius/P 21 (2n= 68) were used for cytogenetic anal-ysis. Chromosome spreads were made following the methodof Liu et al. (2007) with minor modifications. The root tipswere digested at 37� for 2.5 hr in an enzyme mixture con-sisting of 2% cellulase (Sigma, St. Louis, MO) and 24%pectinase (Sigma) in 10 mM sodium citrate buffer (4 mMcitric acid and 6 mM sodium citrate). The treated root tips

Table 1 The F1 progeny fertility restoration of five amphiploids (Amp) and the F1 of H. californicus/HA 89 crossedwith CMS 514A

Cross No. of plant No. of fertile plant

CMS 514A//H. californicus/HA 89 4 0CMS 514A//Amp H. atrorubens/HA 89 6 0CMS 514A//Amp H. mollis/P 21 11 0CMS 514A//Amp H. cusickii/P 21 6 0CMS 514A//Amp H. grosseserratus/P 21 0 0CMS 514A//Amp H. angustifolius/P 21 13 13Total 50

728 Z. Liu et al.

were squashed in 45% acetic acid. Cover slips were removedafter being frozen over liquid nitrogen for 5 min.

Genomic DNA of H. angustifolius (plant code G04/795)was used as a probe after being sheared in boiling water for10 min and labeled with digoxigenin–11-dUTP using thenick translation method according to the manufacturer’sinstructions (Roche Applied Science, Nutley, NJ). GenomicDNA of HA 89 was used as a blocking DNA after shearing inboiling water for 20 min and placed on ice for 5 min, withthe ratio of blocking DNA to probe DNA of 30:1. Labeledprobes were detected with anti-dig-rhodamine or anti-dig-fluorescein (Roche). Chromosomes were counterstainedwith 4’,6-diamidino-2-phenylindole (DAPI, Sigma) in Vecta-shield (Vector Laboratories, Burlingame, CT). Slides wereanalyzed using a fluorescence Axioplan2 imaging microscope(Zeiss, Germany). Images were captured by a charge-coupleddevice (CCD) camera (Zeiss AxioCam HRM) and processedusing Axiovision 3.1 software and Adobe Photoshop 6.0.

Mapping population

Ten to 12 progenies derived from three plants, which weregenetically similar to the F1 hybrids [G08/598, G08/613,and G08/621, pedigree CMS 514A/6/(CMS 514A//AmpH. angustifolius/P 21/3/2*HA 89/4/HA 821/5/HA 89 andSelf) SIB], were used as test populations. An F2 populationderived from G08/613 was used for the mapping of the Rfgene. The population was planted in the greenhouse in 2009

totaling 262 individuals. The MF F2 individuals were self-pollinated to obtain F3 seeds. Also, plants with poor pollenfertility were crossed with HA 89 to obtain adequate seed forprogeny testing.

Pollen fertility analysis and F2 phenotype confirmation

The pollen fertility of the F2 progenies was determined vi-sually and by pollen stainability for each MF plant. Pollenstaining followed Alexander’s method (Alexander 1969)and was analyzed as previously reported by Liu et al.(2012). The F3 and testcross progenies of the F2 populationwere visually scored to confirm the fertility of each F2 plant,using 20–50 progenies from each F2 individual grown in thefield in Fargo, North Dakota, in 2009.

DNA extraction and PCR analysis

Genomic DNAwas extracted according to the protocol of theQiagen DNAeasy 96 plant kit (Qiagen, Valencia, CA). Thebulked segregant analysis (BSA) method was used forpolymorphism screening (Michelmore et al. 1991), usingequal quantities of DNA from 10 plants for each bulk. Fourbulks were used, including the homozygous fertile (bulk F)and sterile (bulk S) bulks of F2 plants with 2n = 34, a fertilebulk with 2n= 35 (bulk 2n= 35) and a sterile bulk with 2n=34 (bulk 2n = 34) in the BC4F2 progeny. The PCR amplifi-cation and genotyping followed Liu et al. (2012).

Molecular marker screening

In total, 370 pairs of SSR primers mapped to the 17sunflower linkage groups from the Compositae database(http://compositdb.ucdavis.edu) were used for polymor-phism screening among the four bulks. An additional 65SSR markers and 28 EST–SSR markers from the candidateLG 3 of 23 maps in the Sunflower CMap Database (http://sunflower.uga.edu/cgi-bin/cmap/map_search) were used toscreen for polymorphisms among the parents and the F1plants. Polymorphic markers were used for genotyping themapping population after confirmation.

Statistical analysis and linkage map construction

The deviation analyses of the fertility trait and marker lociwere compared with the expected Mendelian ratios in the F2generation using the chi-square test. The MAPMAKER/Expv. 3.0b program (Whitehead Institute, Cambridge, MA)(Lander et al. 1987) was used for linkage analysis of the phe-notypes and molecular genotypes following Liu et al. (2012).

Results

Identification of the Rf gene in AmpH. angustifolius/P 21

The chromosome number of the five amphiploids and the F1hybrids of H. californicus/HA 89 (2n = 68) were stable andmaintained during sib-crossing. All of the F1 plants from thecrosses of CMS 514A with the F1s of H. californicus/HA 89and four of the five amphiploids were male sterile (MS)

Figure 1 Scheme for introgression of the fertility restoration gene Rf6 forCMS 514A from an interspecific amphiploid (Amp) of H. angustifolius/P21 (2n = 68) into cultivated sunflower.

Genetic Analysis of Rf6 in Sunflower 729

http://compositdb.ucdavis.edu

http://sunflower.uga.edu/cgi-bin/cmap/map_search

http://sunflower.uga.edu/cgi-bin/cmap/map_search

(Table 1). However, the F1 plants derived from the crosseswith Amp H. angustifolius/P 21 were all MF. Chromosomecounts revealed that 12 of the 13 F1s derived from this crosshad 51 chromosomes, and the remaining one had 47. SinceP 21 does not restore the male fertility of CMS 514A (Wanget al. 2007), it suggests that the Rf gene came from H.angustifolius of the Amp H. angustifolius/P 21 and was des-ignated Rf6.

Integration of Rf6 into cultivated sunflower

The F1 plants (2n= 51) derived from the cross of CMS 514A ·Amp H. angustifolius/P 21 were crossed with HA 89. All F1plants produced seeds, with an average seed set of 2.2%(593 seeds/26500 florets). The MF:MS ratio of 40 BC1F1plants was 11:29, with the pollen stainability averaging19.8% (range 0.7–79.1%), and a chromosome number of2n = 37-46 for the MF plants (supporting information, Ta-ble S1). Four plants that produced very small amount ofpollen (S+) were scored as MS. The MF progenies werebackcrossed with HA 89 or crossed with HA 821 to furtherreduce their chromosome number to 2n = 34 (Figure 1).Three MF BC1F1 plants, with pollen stainability of 52.3,79.1, and 19.3%, respectively, were crossed with HA 89.The chromosome numbers of BC2F1 progenies were reducedto 34–39, and 6 of 26 plants were MF with an averagepollen stainability of 47.3% (range 12–87%). After crossingwith HA 89, HA 821, or CMS BC2F1, 8 MF plants (2n = 35)were obtained among 36 progenies, with an average pollenstainability of 47.6% (range 9–96%).

After self-pollination and backcrossing with HA 89, sevenof the eight MF plants with 2n = 35 resulted in 1 MF prog-eny with 2n = 36 (pollen stainability of 27.4%), 10 MF with2n = 35 (pollen stainability of 56.5%, range 18.1–91.6%),and 2 MF with 2n = 34 (pollen stainabilities of 24.4 and56.9%) from a total of 62 progenies (Table S2). Significantvariation in pollen stainability among plants was observed.However, Rf6 restored the male fertility to .90% in somecases, suggesting the potential use of this material for sun-flower breeding. The frequency of MF and MS plants with2n = 36, 35, and 34 was 1.6 and 1.6%, 16.1 and 1.6%, and

3.2 and 75.8%, respectively. Since the majority of the MFplants (76.9%) had 2n= 35, it indicated that Rf6 was on thealien H. angustifolius chromosome in these 7 MF plants.Moreover, 79.0% of these 62 progenies had 2n = 34, withonly 2 MF plants (4.0%), which suggested a low transmis-sion rate of this alien chromosome in the progenies anda low frequency of recombination between the region con-taining Rf6 and the cultivated sunflower chromosome.

One MF plant, G07/517 (2n= 34), with pollen stainabilityof 56.9%, was crossed to an MS individual, G07/553 (2n =34). The seed set of this cross was 100%, suggesting that allthe female gametes were fertile. Of the 31 F1 hybrids derivedfrom this cross, six plants (19.4%) were MF, and three wereself-compatible with 100% seed set. These three plants, G07/610, G07/612, and G07/623, had improved pollen stainabilityof 74.7, 96.0, and 90.2%, respectively (Table S2), suggestingthat MF plants with 2n = 34 and acceptable male and femalefertility were found.

Genetic analysis of the alien chromosome carrying Rf6

Testcross progenies of CMS 514A pollinated with two MFplants, G07/596 and G07/598 (2n = 35, derived from theself-pollination of G07/513), resulted in 55 MS (2n = 34)and four MF plants (2n = 35). The MF plants were all 2n =35, whereas the MS plants were all 2n = 34 (Table 2),which suggested that Rf6 was located on the alien chromo-some in G07/596 and G07/598. The low number of theplants with 2n = 35 also indicated that this alien chromo-some did not segregate randomly into the daughter cells inmeiosis. The pollen stainability of G07/596 and G07/598was 60.4 and 81.9%, respectively, while those for the four2n = 35 MF progeny plants were improved further, averag-ing 95.0% (range 92.8–98.6%), including G08/672 andG08/683 (derived from CMS 514A · G07/596), and G08/638 and G08/651 (derived from CMS 514A · G07/598).

Comparison of the chromosome constitutions of the MS2n = 34 and MF 2n = 35 plants revealed a large chromo-some in the 2n = 35 plants (Figure 2A), which was notpresent in the 2n = 34 plants (Figure 2B). One MF plant,G09/2614 (2n = 35), derived from the self-pollination of

Table 2 Alien chromosome transmission in testcross and self-pollinated progenies of the male-fertile (MF)2n = 35 plants

Cross Total plants 2n = 34 MF:MS 2n = 35 MF:MS 2n = 35 (%)a 2n = 36 MF:MS 2n = 36 (%)b

CMS 514A · (G07/596and G07/598) (MF, 2n = 35)

59 0:55 4:0 6.8

G09/2614 (MF, 2n = 35) ·HA 89

59 0:39 20:0 33.9

G08/638 and G08/672(MF, 2n = 35) selfed-total

116 1:75 25:10 30.2 4:1 4.3

G08/638 (MF, 2n = 35)selfed (1)

57 1:36 8:9 29.8 2:1 5.3

G08/672 (MF, 2n = 35)selfed (2)

59 0:39 17:1 30.5 2:0 3.4

a Percentage = number of 2n = 35 plants/total plants · 100.b Percentage = number of 2n = 36 plants/total plants · 100.

730 Z. Liu et al.

http://www.genetics.org/lookup/suppl/doi:10.1534/genetics.112.146092/-/DC1/genetics.112.146092-1.pdf





G08/672 was emasculated and pollinated with HA 89 tostudy the transmission of the alien chromosome in the prog-eny when it was used as the female parent. Among the 59progenies of this cross, 20 were MF with 2n = 35, and theremaining 39 were MS with 2n = 34 (Table 2 and TableS3). A higher transmission rate of the alien chromosomecarrying Rf6 was observed when the MF plants (2n = 35)were used as the female parent (33.9%) vs. as the maleparent (6.8%). However, it was still lower than the expectedpercentage of 50% if this alien chromosome segregated ran-domly during meiosis. The above analysis indicated that theselection pressure on male gametes was stronger than thaton female gametes.

G08/638 and G08/672 (2n = 35) were self-pollinatedand produced 57 and 59 F2 plants, respectively. They wereexamined for chromosome number and male fertility (Table2). Among the 116 plants, 35 had 2n = 35, 5 had 2n = 36,and the remaining 76 had 2n = 34. Again, the distortedsegregation of the alien chromosomes was observed in thisgeneration. Pollen fertility examination of the 57 F2 plantsderived from G08/638 indicated that all except 1 with 2n =34 were MS and that not all plants with 2n = 35 and 36were MF. In the second group of plants derived from G08/672, 39 plants were 2n= 34 and MS, 2 plants were 2n= 36and MF, and 17 plants were 2n = 35 and MF, except for 1plant with 2n = 35, which was MS. The above genetic anal-yses indicated that the recombination rate between the alienchromosome and cultivated sunflower was very low; how-ever, the expression of Rf6 was complicated in other cases,such as for the F2 plants derived from G08/638. It was likelythat Rf6 had been recombined into the genome of the culti-vated sunflower. Thus, in the progenies there were both MFplant with 2n = 34 and MS plants with 2n = 35 or 36; i.e.,the large chromosomes in the gametes did not carry Rf6gene in the latter cases.

Genetic analysis of the Rf6 gene

Three MF plants, G07/610, G07/612, and G07/623 (2n =34), were checked for male fertility in their selfed progeniesand testcrossed with CMS 514A to study the inheritance andrestoration of Rf6. Of the 29 plants derived from the self-pollinated G07/610, 19 were MF (pollen stainability aver-aged 64.5%, but varied widely from 6.6 to 96.1%), and the

remaining 10 were MS. Similarly, the MF:MS ratio of the 29plants derived from G07/612 was 20:9. The pollen stainabil-ity also varied widely (range 27.3–99.4%), with a higheraverage of 78.9%. The ratio of MF:MS among the 30 plantsderived from G07/623 was 20:10, with an even higheraverage pollen stainability of 91.5% (range 51.1–98.3%).The overall ratio of the MF to MS (59:29) were closer to 2:1(x2 =0.01, P = 0.94) than to 3:1 (x2 =2.97, P = 0.085). Themale fertility and pollen stainability of each individual isshown in Table S4.

The testcrosses of G07/610, G07/612, and G07/623 toCMS 514A showed different ratios of MF to MS plants.Among the 12 progenies derived from each cross combina-tion, the ratio of MF to MS was 2:10, 4:8, and 8:4,respectively. The overall ratio of MF to MS in the threetestcrosses was 14:22, fitting the expected ratio of 1:1 (x2 =1.78,P = 0.18), indicating that these three male parents wereheterozygous at the Rf6 gene locus. Pollen stainability variedfrom 88.2 to 99.5%, with an average of 96.8%. Three plantswith 100% seed set, G08/598, G08/613, and G08/621,were selected for further study (Figure 1). Their pollen stain-ability was 96.9, 99.5, and 95.7%, respectively.

Mitotic GISH and cytogenetic analyses

The alien chromosome or segments from the H. angustifoliusgenome can be differentiated from the cultivated sunflowerchromosomes by GISH. Male fertility data and chromosomecounts for the testcross progenies of CMS 514A with G07/596 and G07/598, and the self-pollinated progeny of G08/672, as well as the progeny derived from G09/2614 · HA 89(Table 2), suggested that Rf6 was located on the alien chro-mosome in the MF plants with 2n = 35 chromosomes. GISHresults for both G08/638 and G08/672 (2n = 35) showedan obvious signal on one large chromosome compared toother cultivated sunflower chromosomes (Figure 3A forG08/638). The ratio of the short arm to the long arm of thisalien chromosome was 0.6351 based on five cell observa-tions. GISH results together with genetic analysis of the MFplants (2n = 35) suggested that this chromosome containsthe Rf6 gene for CMS 514A.

The F2 progenies derived from G08/598, G08/613, andG08/621 were tested for male fertility and the alien chro-mosome segments (Figure 3 and Table S5). A total of 27

Figure 2 Chromosome spreads of male-fertile(MF) (2n = 35) and male-sterile (MS) (2n = 34)plants. (A) Chromosome spread of an MF plantwith 2n = 35, G09/2617, derived from self-pollina-tion of G08/672. (B) Chromosome spread of an MSplant with 2n = 34, G09/2567, derived from self-pollination of G08/638. The arrow in A shows thelarger chromosome compared to other chromo-somes, which is assumed to be the alien chromo-some from H. angustifolius carrying Rf6. Bars,5 mm.





http://www.genetics.org/lookup/suppl/doi:10.1534/genetics.112.146092/-/DC1/genetics.112.146092-5.xls

plants were analyzed by GISH. Interestingly, three plantsheterozygous for Rf6 derived from G08/613 had only onesmall translocation, and two plants homozygous for Rf6 con-tained two small translocations (Figure 3, B and C), whereasseven plants heterozygous for Rf6 from both G08/598 andG08/621 had two translocations, i.e., one with a whole shortarm translocated and another with the same as G08/613progenies (Figure 3D). For the two homozygous MF plantsderived from G08/598, two small translocations togetherwith one whole short arm translocation were detected inone plant, G08/2670 (Figure 3E), and only the two smallsegment translocations were detected in G08/2680. In ad-dition, no alien chromosome segment was detected in threeMS plants (G08/2656, G08/2675, and G08/2676), and onlythe whole short arm translocation was detected in two otherMS plants, G08/2657 and G08/2658, derived from G08/621 (Figure 3F). Therefore, the GISH results for the F2 indi-viduals indicated that Rf6 was located on the small H. angus-tifolius chromosomal segment involved in the translocation.The translocation point for the small segment was located atabout the one-fourth distance from the end of the long armof the chromosome.

Moreover, three translocations were detected using GISHanalysis in two F2 plants with morphological abnormalities.One plant, G08/2651, derived from G08/621, had twowhole arm translocations and one small segment transloca-tion (Figure 3G) and was physically abnormal with a tiny

capitulum and a short plant (about 40 cm). The second plant,G08/2681, derived from G08/598, was MF and had one wholearm translocation and two small segment translocations, butwilted during flowering. This wilting trait was also observed intwo other MF F2 individuals (G08/2652 and G08/2672), andboth had one whole arm and one small segment translocations.Therefore, genetic unbalanced gametes might be producedwhen Rf6 linked with undesirable genes or the alien chro-mosome or segments negatively interact with the cultivatedsunflower background, causing morphological abnormalities.

Fertility segregation in the mapping population

Based on the fertility segregation, abnormal traits, and GISHanalysis, a population derived from G08/613 was used tomap Rf6. This population included 262 F2 individuals, with166 MF and 89 MS plants, producing 221 usable F3 or test-cross progeny families. Progeny test was not performed for34 MF plants due to shortage of seed and 7 of 9 wiltingplants died before flowering. Phenotypic analysis identified19 homozygous MF, 113 heterozygous MF, and 89 MSplants in the F2 population. Chi-square test indicated theratio of MS:MF or homozygous MF:heterozygous MF:homo-zygous MS phenotypes significantly deviated from theexpected Mendelian ratio 1:3 or 1:2:1 (x2 .13, P ,0.0005) (Table 3).

In this F2 mapping population, a large variation in fertil-ity was observed (15–100%). About 82% of the plants had

Figure 3 Genomic in situ hybridization (GISH) analyses of the alien H. angustifolius chromosome or segments in different progenies. The genomic DNAof H. angustifolius was labeled with digoxigenin–11-dUTP and detected by anti-dig-rhodamine (red), the chromosomes were counterstained by DAPI(blue). (A) a heterozygous MF plant, G08/638 (2n = 35), with an alien chromosome. (B) A heterozygous MF plant with 2n = 34, G08/2660, derived fromself-pollination of G08/613, with one small translocation. (C) A homozygous MF plant with 2n = 34, G08/2663, derived from self-pollination of G08/613, with two small translocations. (D) A heterozygous MF plant with 2n = 34, G08/2649, derived from self-pollination of G08/621, with one wholeshort arm and one small segment translocations. (E) A homozygous MF plant with 2n = 34, G08/2670, derived from self-pollination of G08/598, withone whole short arm and two small translocations. (F) An MS plant with 2n = 34, G08/2657, derived from self-pollination of G08/621, with one wholeshort arm translocation. (G) An abnormal plant, G08/2651, derived from self-pollination of G08/621, with two whole short arm and one small segmenttranslocations. The arrows show the alien chromosome or segments from H. angustifolius. Bars, 5 mm.

732 Z. Liu et al.

fertility .50%, and 73% .80%. The average pollen stain-ability for homozygous MF F2 plants was 93.9%. Lower pol-len stainability was observed for the heterozygous MF F2plants with 81.5%. For the MF F2s without progeny test,the pollen stainability was 87.2%. The overall average pol-len stainability in this population was 83.5%.

Chromosomal location of Rf6

A total of 463 molecular markers were used for mapping.Only nine markers (1.9%) showed no products or weakbands. The 370 pairs of SSR primers from 17 sunflowerlinkage groups were used to screen polymorphisms usingBSA, which averaged 21 primers per linkage group. Fourpolymorphic markers were identified for the two fertile(bulk F and bulk 2n= 35) and sterile bulks (bulk S and bulk2n= 34), including ORS433, ORS488, ORS822, and ORS1114.Two polymorphic markers (ORS432 and ORS1021) wereidentified for bulk 2n = 35 and bulk 2n = 34, but not forbulk F and bulk S. The polymorphic markers were validatedusing the 10 individuals constituting each bulk. Validation ofthe markers is shown in Figure 4, using the ORS433 primerpair. The ORS432, ORS488, and ORS1114 markers weremapped to LG 3 of Tang et al. (2003) and RHA 280 ·RHA 801_RIL (Sunflower CMap: http://www.sunflower.uga.edu/cgi-bin/cmap/viewer?data_source=pbio_cmap;ref_map_accs=RHA280xRHA801ril3), whereas ORS433,ORS822, and ORS1021 were multi-loci markers mappedto several LGs, including LG 3. The alien chromosome car-rying Rf6 in the MF 2n= 35 plants was expected to be on LG3 according to the results from the BSA. The four polymor-phic primers between the fertile and sterile bulks were usedto genotype the mapping population. The results showeda close linkage among the markers and Rf6, further confirm-ing the chromosomal location of Rf6 on LG 3.

In addition, 65 SSR and 28 EST-SSR markers on LG 3from 23 maps in the Sunflower CMap Database were used toscreen the polymorphism among the parents and the F1 plants.

Thirteen polymorphic markers were identified, including sevenSSR markers (ORS13, ORS134, ORS525, ORS683, ORS777,ORS1130, and ORS1144), and six EST-SSR markers (HT088,HT499, HT734, HT779, HT845, and HT1029). However, onlyORS13, ORS525, HT088, and HT734 were linked to Rf6 afterconfirmation using the individuals of the four bulks. These fourmarkers were also used for genotyping the mapping popula-tion. Segregations of Rf6 and markers using the whole F2 pop-ulation are shown in Table 3.

Molecular mapping of Rf6

The chi-square test showed that eight markers were severelydistorted from the expected Mendelian ratios. Using theeight markers mentioned above, Rf6 was located on a mapconstructed using 221 F2 individuals, covering a distance of10.8 cM (Figure 5). Rf6 was located between the co-domi-nant HT088 and two co-segregated markers ORS13 andORS1114. The closest markers were ORS13 and ORS1114,at a distance of 1.6 cM. Noticeably, the primer pair ORS433produced two dominant markers, i.e., ORS433-a (about 190bp) and ORS433-b (about 175 bp), with only ORS433-a linked to Rf6, whereas ORS433-b was not (Figure 4 andFigure 5). ORS433-b did not show the same segregationdistortion in the mapping population as ORS433-a (Table3). Therefore, the segregation distortion occurred only inthe mapped chromosomal region harboring Rf6. Comparedto the reference maps of Tang et al. (2003) (Figure 5A), andRHA 280 x RHA 801_RIL (in press) (Figure 5C) in the Sun-flower CMap Database, the eight linked marker loci for Rf6were grouped at the end of LG 3, although the order of themarkers were reversed (Figure 5B). The markers ORS1021and ORS432 were linked to the alien chromosome in the MFplants with 2n= 35, but not linked to Rf6 (region I on Figure5, A and C). Their map positions suggested that the breakpoint of the translocation carrying Rf6 might be located be-tween the markers ORS432 and ORS13 (Figure 5A) or be-tween ORS432 and ORS488 (Figure 5C).

Table 3 Segregation of the Rf6 locus and marker loci in the F2 population derived from the cross CMS 514A/6/(CMS 514A//AMP H.angustifolius/P 21/3/2*HA 89/4/HA 821/5/HA 89 and Self) SIB

Traits or markers No. of F2 plants

Observed no.

Ratio expected x2 P-valueA H B C

Rf6a 255 89 166 1:3 13.33 2.6 · 1024

Rf6b 221 89 113 19 1:2:1 44.46 2.2 · 10210

ORS822 220 85 135 1:3 21.82 3.0 · 10–6

HT088 220 85 118 17 1:3 43.20 4.9 · 10211

ORS433-a 220 82 138 1:3 17.67 2.6 · 10–5

ORS13 220 83 118 19 1:2:1 38.40 4.6 · 1029

ORS1114 220 83 118 19 1:2:1 38.40 4.6 · 1029

HT734 220 82 119 19 1:2:1 37.55 7.0 · 1029

ORS488 220 82 119 19 1:2:1 37.55 7.0 · 1029

ORS525 221 86 135 1:3 22.82 1.8 · 1026

ORS433-bc 216 58 158 1:3 0.40 0.53

Symbols: A, homozygous MS (rfrf); H, heterozygous MF (Rfrf); B, homozygous MF (RfRf); C, RfRf or Rfrf.a Phenotyping data in the F2 generation.b Phenotyping data of the F2 individuals after progeny test.c ORS433-b is not linked to the Rf6 gene, thus not included in Figure 5B.


http://www.sunflower.uga.edu/cgi-bin/cmap/viewer?data_source=pbio_cmap;ref_map_accs=RHA280xRHA801ril3



Discussion

CMS 514A/Rf6, a new CMS/Rf system

Fifty-nine different germplasm sources were used to identifythe Rf gene for CMS 514A (Wang et al. 2007; this study).The Rf6 gene was determined to have originated from H.angustifolius. These results suggested that Rf6 is probablydifferent from other reported Rf genes. The average pollenfertility of Rf1 is 98% for hybrid 894 (Seiler 2000) and 94%for Rf3 from RHA 280 (Liu et al. 2012), while the averagepollen fertility of Rf6 was �80%, with a large variation in theF2 mapping population. Rf6 was mapped to LG 3 of thesunflower SSR public map (Tang et al. 2003), with eightlinked markers in this study.

Seven Rf genes were previously mapped on the sunflowergenetic maps, with only Rf4 reported on LG 3 (Feng and Jan2008; reviewed by Liu et al. 2012). The closest markerlinked to Rf4 is ORS1114 at a distance of 0.9 cM. Rf6 wasalso mapped to LG 3, with the closest markers ORS13 andORS1114 at a distance of 1.6 cM. Considering the differentorigins, these two Rf genes are probably not the same. TheH. angustifolius amphiploid was the only one of the fivetested that restored male fertility for CMS 514A, while threeamphiploids (H. atrorubens, H. grosseserratus, and H. angus-tifolius) restored male fertility for CMS GIG2. The H. mollisamphiploid failed to restore the male fertility for either CMS(Sunflower CMap). Therefore, these two CMSs are likely notthe same. Currently, the allelic relationship analysis betweenRf4 and Rf6 is under investigation to provide more informa-tion for their practical use in sunflower breeding. The com-bination of the markers closely linked to Rf6 will be usefulfor marker-assisted selection.

We introgressed Rf6 into cultivated sunflower using a tra-ditional crossing and backcrossing scheme. The GISH anal-ysis with different fertile or sterile F2 plants indicated thatonly the translocation located terminally on a chromosome,estimated to be about one-fourth of the long arm, is relatedto fertility restoration of CMS 514A. Molecular marker anal-ysis suggested that the translocation break point on the alienchromosome for Rf6 in the MF plants (2n= 35) may be located

between markers ORS432 and ORS13 or ORS488 on thereference maps (region II on Figure 5, A and C). In addition,the primer pair ORS433 produced two markers, with onlyORS433-a closely linked to Rf6 in the F2 population. Consid-ering ORS433 is a multilocus marker in the public sunflowermaps and ORS433-b was not mapped onto LG 3 in this study,we are not sure whether the ORS433-a marker is the sameORS433 marker located on the region I of LG 3 or is a newmarker in this study. Therefore, additional markers are neededto characterize this region to more precisely determine thetranslocation break point.

Rf6 and the linked marker loci showedsegregation distortion

Segregation distortion is the deviation of the frequency ofgenotypes from the expected Mendelian ratio within a seg-regating population. Segregation distortion has been ob-served in fungi, plants, insects, and mammals (Lyttle 1991;Liu et al. 2010). In plants, segregation distortion has beenencountered in maize (Mangelsdor and Jones 1926), rice(McCouch et al. 1988), wheat (Zhang and Dvorák 1990;Faris et al. 1998; Kumar et al. 2007), barley (Graner et al.1991; Li et al. 2010), tobacco (Cameron and Moav 1957),tomato (Paterson et al. 1988), alfalfa (Echt et al. 1994), andcoffee (Ky et al. 2000). Segregation distortion, also called“meiotic drive,” may be caused by genetic elements, includ-ing gametic selection (pollen tube competition, lethal pollen,and preferential fertilization), zygotic selection, interspecificsterility genes (S), and chromosome translocation (Lyttle 1991;Kumar et al. 2007; Gutiérrez et al. 2010; Liu et al. 2010). It hasbeen suggested that meiotic drive elements are highly impor-tant for the evolution of recombination and sexual reproduction(Hurst and Werren 2001; Jaenike 2001; Li et al. 2010).

Segregation distortion has been reported for sunflowerpopulations derived from the interspecific crosses involvingwild species in the mapping of a downy mildew resistancegene on LG 1, PlARG, which originated from H. argophyllusTorrey and Gray (Dußle et al. 2004; Wieckhorst et al. 2010).Significant segregation distortion of the codominant markersclosely linked to the PlARG gene was observed in the F2

Figure 4 Representative results of the markers amplified by the ORS433 primer pair among the four bulks and the individuals constituting each bulk,respectively, on a nondenaturing polyacrylamide gel. (1) Bulk S; (2) bulk F; (3) bulk 2n = 34; (4) bulk 2n = 35. The bulks 3 and 4 consist of the MS plants(2n = 34) and MF plants (2n = 35) from the selfed progenies of a BC4F1 MF plant (2n = 35), respectively; M indicates a 100-bp plus ladder Gelpilot(Qiagen); the arrows indicates the dominant markers. Marker ORS433-a (�190 bp) is linked to Rf6, whereas marker ORS433-b (about 175 bp) is not.

734 Z. Liu et al.

population derived from the cross of CMS HA 342 ·ARG1575-2, but not in the ones derived from the HA 342 ·ARG1575-2 and NDBLOSsel · KWS04, indicating the influ-ence of the CMS cytoplasm and chromosome segment(s) fromthe wild species on the fertility and segregation ratios in thepopulation. In our study, a severe deviation was detected forRf6 and the linked marker loci in a mapping population, whichprobably indicated suppressed recombination or gamete selec-tion in this region. GISH and molecular marker results suggestedthat Rf6 was on the chromosome with the small segment trans-location. Moreover, only one amphiploid was discovered tocontain the Rf gene for CMS 514A after testing 59 differentgermplasm sources, including the maintainer of CMS 514A,which restores fertility to other CMS types (Wang et al.2007; this study). Taken together that CMS 514A has anH. tuberosus cytoplasm and the Rf6 gene was from the wildspecies H. angustifolius, the segregation distortion may becaused by several factors, such as gametic selection, inter-specific S gene, and chromosome translocation (Lyttle 1991;Kumar et al. 2007; Gutiérrez et al. 2010; Liu et al. 2010).

Due to the limited number of marker loci polymorphicbetween the parents, the linkage group covered a geneticdistance of only 10.8 cM. Chromosomal inversion could beone of the reasons for the suppressed recombination. Detailedcomparison of the marker orientations among the mapsconstructed here and the reference maps would help to explainthis question. Therefore, more markers such as single nucle-otide polymorphism (SNP) markers, SSR, or other types ofmarkers are necessary to fine map Rf6, as well as the segre-gation distortion regions. Additionally, a low transmission rateof the alien chromosome or its segments into the cultivatedsunflower was detected during backcrossing. Abnormal growth,such as reduced vigor, wilting before or near flowering stage,and sterile sections on the flowering capitulum of MF plants,was also noted in some backcrossing progenies.

In conclusion, this study identified an Rf gene, Rf6, thatrestores the male fertility of a recently identified CMS source,CMS 514A, originated from a wild species, H. angustifolius,via an interspecific amphiploid H. angustifolius/P 21 (2n= 68).This gene was introgressed into the cultivated sunflowerbackground after several crosses and backcrosses. The alienchromosome or segments were characterized using GISH andmolecular marker analyses. Rf6 was located on LG 3 of thesunflower public SSR map, with eight linked markers ina mapping population. Progenies with different translocationswere developed during the crossing process. These couldfacilitate the development of a unique fertility restorer forCMS 514A and could be useful in studying the interactionsbetween cytoplasm and nuclear genes.

Acknowledgments

The authors thank Lisa A. Brown for technical assistance inthis study and Ridhima Katyal, Jordan Hogness, Alexis Ganser,Yuni Chen, and Marjorie A. Olson for their help in conductingthis study. We appreciate Drs. Chengsong Zhu (Kansas StateUniversity), Wentao Li (University of California-Davis), ZahirulTalukder, and Yunming Long (North Dakota State University)for valuable discussion during data analysis. We also thankDrs. Larry G. Campbell, Lili Qi, Prem P. Jauhar, Steven S. Xu,and Brady A. Vick for critical review of the manuscript.

Literature Cited

Abratti, G., M. E. Bazzalo, and A. León, 2008 Mapping a novelfertility restoration gene in sunflower, pp. 617–621 in 17th In-ternational Sunflower Conference, edited by I. Velasco. Córdoba,Spain.

Alexander, M. P., 1969 Differential staining of aborted and non-aborted pollen. Stain Technol. 44: 117–122.

Figure 5 The position of the fertility restoration gene Rf6 on LG 3 of the sunflower map. (A) A partial map of LG 3 of Tang et al. (2003). (B) mappingresult of Rf6 on LG 3, using 221 F2 plants. (C) a partial map of LG 3 of RHA 280 · RHA 801_RIL (Sunflower CMap). The distances are given incentimorgans (cM). The corresponding markers are noted by lines between the maps. Region I indicates the region is not linked to Rf6, and region IIindicates the region where the possible break point of the translocation with Rf6 is located.


Allender, C. J., and G. J. King, 2010 Origins of the amphiploidspecies Brassica napus L. investigated by chloroplast and nuclearmolecular markers. BMC Plant Biol. 10: 54.

Bansal, P., S. Banga, and S. S. Banga, 2012 Heterosis as investi-gated in terms of polyploidy and genetic diversity using de-signed Brassica juncea amphiploid and its progenitor diploidspecies. PLoS ONE 7: e29607.

Bentolila, S., A. A. Alfonso, and M. R. Hanson, 2002 A pentatri-copeptide repeat-containing gene restores fertility to male-sterile plants. Proc. Natl. Acad. Sci. USA 99: 10887–10892.

Brown, G. G., N. Formanova, H. Jin, R. Wargachuk, C. Dendy et al.,2003 The radish Rfo restorer gene of Ogura cytoplasmic malesterility encodes a protein with multiple pentatricopeptide re-peats. Plant J. 35: 262–272.

Cameron, D. R., and R. M. Moav, 1957 Inheritance in Nicotianatabacum XXVII: pollen killer, an alien genetic locus inducingabortion of microspores not carrying it. Genetics 42: 326–335.

Colmer, T. D., E. Epstein, and J. Dvorák, 1995 Differential soluteregulation in leaf blades of various ages in salt-sensitive wheatand a salt-tolerant wheat · Lophopyrum elongatum (Host) A.Love amphiploid. Plant Physiol. 108: 1715–1724.

Cui, X., R. P. Wise, and P. S. Schnable, 1996 The rf2 nuclearrestorer gene of male-sterile T-cytoplasm maize. Science 272:1334–1336.

Dominguez-Gimenez, J., and G. N. Fick, 1975 Fertility restorationof male-sterile cytoplasm in wild sunflowers. Crop Sci. 15: 724–726.

Dußle, C. M., V. Hahn, S. J. Knapp, and E. Bauer, 2004 PlArg fromHelianthus argophyllus is unlinked to other known downy mil-dew resistance genes in sunflower. Theor. Appl. Genet. 109:1083–1086.

Dvorák, J., M. Edge, and K. Ross, 1988 On the evolution of theadaptation of Lophopyrum elongatum to growth in saline envi-ronments. Proc. Natl. Acad. Sci. USA 85: 3805–3809.

Echt, C. S., K. K. Kidwell, S. J. Knapp, T. C. Osborn, and T. J. McCoy,1994 Linkage mapping in diploid alfalfa (Medicago sativa).Genome 37: 61–71.

Faris, J. D., B. Laddomada, and B. S. Gill, 1998 Molecular map-ping of segregation distortion loci in Aegilops tauschii. Genetics149: 319–327.

Feng, J., and C. C. Jan, 2008 Introgression and molecular taggingof Rf4, a new male fertility restoration gene from wild sun-flower Helianthus maximiliani L. Theor. Appl. Genet. 117:241–249.

Feng, J., G. J. Seiler, T. J. Gulya, C. Li, and C. C. Jan, 2007 Sclerotiniastem and head rot resistant germplasm development utilizing in-terspecific amphiploids. 29th Sunflower ResearchWorkshop, January10–11, 2007, Fargo, ND. Available at: www.sunflowernsa.com/research/research-workshop/documents/Feng_etal_Amphiploids_2007.pdf

Gentzbittel, L., F. Vear, Y.-X. Zhang, A. Bervillé, and P. Nicolas,1995 Development of a consensus linkage RFLP map of culti-vated sunflower (Helianthus annuus L.). Theor. Appl. Genet. 90:1079–1086.

Gentzbittel, L., E. Mestries, S. Mouzeyar, F. Mazeyrat, S. Badaouiet al., 1999 A composite map of expressed sequences and phe-notypic traits of the sunflower (Helianthus annuus L.) genome.Theor. Appl. Genet. 99: 218–234.

Gillman, J. D., S. Bentolila, and M. R. Hanson, 2007 The petuniarestorer of fertility protein is part of a large mitochondrial com-plex that interacts with transcripts of the CMS-associated locus.Plant J. 49: 217–227.

González, G., C. Comas, V. Confalonieri, C. A. Naranjo, and L.Poggio, 2006 Genomic affinities between maize and Zea per-ennis using classical and molecular cytogenetic methods (GISH-FISH). Chromosome Res. 14: 629–635.

Graner, A., A. Jahoor, J. Schondelmaier, H. Siedler, K. Pillen et al.,1991 Construction of a RFLP map of barley. Theor. Appl.Genet. 83: 250–256.

Gutiérrez, A. G., S. J. Carabalí, O. X. Giraldo, C. P. Martínez, F.Correa et al., 2010 Identification of a rice stripe necrosis virusresistance locus and yield component QTLs using Oryza sativa ·O. glaberrima introgression lines. BMC Plant Biol. 10: 6.

Horn, R., B. Kusterer, E. Lazarescu, M. Prüfe, and W. Friedt,2003 Molecular mapping of the Rf1 gene restoring pollen fer-tility in PET1-based F1 hybrids in sunflower (Helianthus annuusL.). Theor. Appl. Genet. 106: 599–606.

Hurst, G. D., and J. H. Werren, 2001 The role of selfish geneticelements in eukaryotic evolution. Nat. Rev. Genet. 2: 597–606.

Islam, S., A. I. Malik, A. K. Islam, and T. D. Colmer, 2007 Salttolerance in a Hordeum marinum-Triticum aestivum amphiploid,and its parents. J. Exp. Bot. 58: 1219–1229.

Jaenike, J., 2001 Sex chromosome meiotic drive. Annu. Rev. Ecol.Syst. 32: 25–49.

Jan, C. C., and J. M. Chandler, 1989 Sunflower interspecific hy-brids and amphiploids of Helianthus annuus · H. bolanderi. CropSci. 29: 643–646.

Jan, C. C., and J. M. Fernandez-Martinez, 2002 Interspecific hy-bridization, gene transfer, and the development of resistance tothe broomrape race F in Spain. Helia 25: 123–126.

Jan, C. C., and B. A. Vick, 2007 Inheritance and allelic relation-ships of fertility restoration genes for seven new sources ofmale-sterile cytoplasm in sunflower. Plant Breed. 126: 213–217.

Jan, C. C., B. A. Vick, J. F. Miller, A. L. Kahler, and E. T. Butler III.1998 Construction of an RFLP linkage map for cultivated sun-flower. Theor. Appl. Genet. 96: 15–22.

Jan, C. C., J. Feng, G. J. Seiler, and T. J. Gulya, 2006 Amphiploidsof perennial Helianthus species · cultivated sunflower possessvaluable genes for resistance to Sclerotinia stem and head rot.28th Sunflower Research Workshop, January 11–12, Fargo, ND.Available at: www.sunflowernsa.com /research/research-workshop/documents /Jan_Amphiploids_06.pdf

Jiang, J., K. L. Morris, and B. S. Gill, 1994 Introgression of Elymustrachycaulus chromatin into common wheat. Chromosome Res.2: 3–13.

Kang, H., Y. Wang, G. Fedak, W. Cao, H. Zhang et al.,2011 Introgression of chromosome 3Ns from Psathyrostachyshuashanica into wheat specifying resistance to stripe rust. PLoSONE 6: e21802.

Kumar, S., B. S. Gill, and J. D. Faris, 2007 Identification andcharacterization of segregation distortion loci along chromo-some 5B in tetraploid wheat. Mol. Genet. Genomics 278: 187–196.

Kumar, S., B. Friebe, and B. S. Gill, 2010 Fate of Aegilops spel-toides-derived, repetitive DNA sequences in diploid Aegilops spe-cies, wheat-Aegilops amphiploids and derived chromosomeaddition lines. Cytogenet. Genome Res. 129: 47–54.

Kusterer, B., R. Horn, and W. Friedt, 2005 Molecular mapping ofthe fertility restoration locus Rf1 in sunflower and developmentof diagnostic markers for the restorer gene. Euphytica 143: 35–42.

Kwiatek, M., L. Błaszczyk, H. Wiśniewska, and B. Apolinarska,2012 Aegilops-Secale amphiploids: chromosome categoriza-tion, pollen viability and identification of fungal disease resis-tance genes. J. Appl. Genet. 53: 37–40.

Ky, C. L., P. Barre, M. Lorieux, P. Trouslot, S. Akaffou et al.,2000 Interspecific genetic linkage map, segregation distortionand genetic conversion in coffee (Coffea sp.). Theor. Appl.Genet. 101: 669–676.

Lander, E. S., P. Green, J. Abrahamson, A. Barlow, M. J. Daly et al.,1987 MAPMAKER: an interactive computer package for con-structing primary genetic linkage maps of experimental andnatural populations. Genomics 1: 174–181.

736 Z. Liu et al.

Leclercq, P., 1969 Une stérilité male cytoplasmic chez le tourne-sol. Ann. Amélior Plant 19: 99–106.

Li, H., A. Kilian, M. Zhou, P. Wenzl, E. Huttner et al.,2010 Construction of a high-density composite map and com-parative mapping of segregation distortion regions in barley.Mol. Genet. Genomics 284: 319–331.

Liu, F., X. Cui, H. T. Horner, H. Weiner, and P. S. Schnable,2001 Mitochondrial aldehyde dehydrogenase activity is re-quired for male fertility in maize. Plant Cell 13: 1063–1078.

Liu, X., L. Guo, J. You, X. Liu, Y. He et al., 2010 Progress ofsegregation distortion in genetic mapping of plants. Res. J.Agronomy 4: 78–83.

Liu, Z., D. Y. Li, and X. Y. Zhang, 2007 Genetic relationshipsamong five basic genomes St, E, A, B and D in Triticeae revealedby genomic Southern and in situ hybridization. J. Integr. PlantBiol. 49: 1080–1086.

Liu, Z., S. Mulpuri, J. Feng, B. A. Vick, and C. C. Jan,2012 Molecular mapping of the Rf3 fertility restoration geneto facilitate its utilization in breeding confection sunflower. Mol.Breed. 29: 275–284.

Lyttle, T. W., 1991 Segregation distorters. Annu. Rev. Genet. 25:511–557.

Malik, A. I., A. K. Islam, and T. D. Colmer, 2011 Transfer of thebarrier to radial oxygen loss in roots of Hordeum marinum towheat (Triticum aestivum): evaluation of four H. marinum-wheat amphiploids. New Phytol. 190: 499–508.

Mangelsdor, P. C., and D. F. Jones, 1926 The expression of Men-delian factors in the gametophyte of maize. Genetics 11: 423–455.

Martín, A., A. Cabrera, E. Esteban, P. Hernández, M. C. Ramírezet al., 1999 A fertile amphiploid between diploid wheat (Tri-ticum tauschii) and crested wheat grass (Agropyron cristatum).Genome 42: 519–524.

McCouch, S. R., G. Kochert, Z. Yu, Z. Wang, G. S. Khush et al.,1988 Molecular mapping of rice chromosomes. Theor. Appl.Genet. 76: 815–829.

Michelmore, R. W., I. Paran, and R. V. Kesseli, 1991 Identificationof markers linked to disease resistance genes by bulked segre-gant analysis: a rapid method to detect markers in specific ge-nomic regions by using segregating populations. Proc. Natl.Acad. Sci. USA 88: 9828–9832.

Miller, J. F., and G. N. Fick, 1997 Genetics of sunflower, pp. 441–495 in Sunflower Technology and Production, Agronomy Mono-graph 35, edited by A. A. Schneiter. ASA, CSSA, and SSSA,Madison, WI.

Nasrallah, M. E., K. Yogeeswaran, S. Snyder, and J. B. Nasrallah,2000 Arabidopsis species hybrids in the study of species differ-ences and evolution of amphiploidy in plants. Plant Physiol.124: 1605–1614.

Pérez-Vich, B., B. Akhtouch, J. Munoz-Ruz, J. M. Fernandez-Martinez,and C. C. Jan, 2002 Inheritance of resistance to a highly virulentrace F of Orobanche cumana Wallr. in a sunflower line derivedfrom interspecific amphiploids. Helia 25: 137–143.

Paterson, A. H., E. S. Lander, J. D. Hewitt, S. Peterson, S. E. Lincolnet al., 1988 Resolution of quantitative traits into Mendelianfactors by using a complete linkage map of restriction fragmentlength polymorphisms. Nature 335: 721–726.

Poggio, L., G. Gonzalez, V. Confalonieri, C. Comas, and C. A. Naranjo,2005 The genome organization and diversification of maizeand its allied species revisited: evidences from classical andFISH-GISH cytogenetic analysis. Cytogenet. Genome Res. 109:259–267.

Qi, L. L., G. J. Seiler, B. A. Vick, and T. J. Gulya, 2012 Geneticsand mapping of the R11 gene conferring resistance to recentlyemerged rust races, tightly linked to male fertility restoration, insunflower (Helianthus annuus L.). Theor. Appl. Genet. 125:921–932.

Schnabel, U., U. Engelmann, and R. Horn, 2008 Development ofmarkers for the use of the PEF1 cytoplasm in sunflower hybridbreeding. Plant Breed. 127: 587–591.

Seiler, G. J., 2000 Registration of ten interspecific germplasmsderived from wild perennial sunflower. Crop Sci. 40: 587–588.

Serieys, H., 1996 Identification, study and utilization in breedingprograms of new CMS sources. FAO Progress Report. Helia 19:144–158.

Serieys, H., 2005 Identification, study and utilization in breedingprograms of new CMS sources, pp. 47–53 in Proceedings of the2005 Sunflower Subnetwork Progress Report. FAO Subnetwork,Rome.

Soliman, M. H., D. Rubiales, and A. Cabrera, 2001 A fertile am-phiploid between durum wheat (Triticum turgidum) and the ·Agroticum amphiploid (Agropyron cristatum · T. tauschii). He-reditas 135: 183–186.

Tang, S., V. K. Kishore, and S. J. Knapp, 2003 PCR-multiplexes fora genome-wide framework of simple sequence repeat markerloci in cultivated sunflower. Theor. Appl. Genet. 107: 6–19.

Wang, D.-X., L.-J. Cui, and C. C. Jan, 2007 Establishment of newcytoplasmic male sterility by introduction of cytoplasm fromwild species in sunflower. Chinese J. Oil Sci. 29: 416–419.

Wang, Z., Y. Zou, X. Li, Q. Zhang, L. Chen et al., 2006 Cytoplasmicmale sterility of rice with boro II cytoplasm is caused by a cyto-toxic peptide and is restored by two related PPR motif genes viadistinct modes of mRNA silencing. Plant Cell 18: 676–687.

Wieckhorst, S., E. Bachlava, C. M. Dußle, S. Tang, W. Gao et al.,2010 Fine mapping of the sunflower resistance locus PlARGintroduced from the wild species Helianthus argophyllus. Theor.Appl. Genet. 121: 1633–1644.

Wojciechowska, B., and H. Pudelska, 2005 Production and char-acterization of amphiploids of Aegilops kotschyi and Ae. biuncia-lis with Secale cereale, and of backcross hybrids of Ae. biuncialis ·S. cereale amphiploids with 2· and 4· S. cereale. J. Appl. Genet.46: 157–161.

Yu, J. K., S. Tang, M. B. Slabaugh, A. Heesacker, G. Cole et al.,2003 Towards a saturated molecular genetic linkage map forcultivated sunflower. Crop Sci. 43: 367–387.

Yue, B., B. A. Vick, X. Cai, and J. Hu, 2010 Genetic mapping forthe Rf1 (fertility restoration) gene in sunflower (Helianthus an-nuus L.) by SSR and TRAP markers. Plant Breed. 129: 24–28.

Zhang, H. B., and J. Dvorák, 1990 Characterization and distribu-tion of an interspersed repeated nucleotide sequence from Lo-phopyrum elongatum and mapping of a segregation-distortionfactor with it. Genome 33: 927–936.

Communicating editor: F. F. Pardo Manuel de Villena


GENETICSSupporting Information


Diversifying Sunflower Germplasm by Integrationand Mapping of a Novel Male Fertility

Restoration GeneZhao Liu, Dexing Wang, Jiuhuan Feng, Gerald J. Seiler, Xiwen Cai, and Chao-Chien Jan

Copyright © 2013 by the Genetics Society of AmericaDOI: 10.1534/genetics.112.146092

Z. Liu et al. 2 SI

Table S1 Chromosome number, male fertility, and pollen stainability of the progenies of CMS 514A//Amp H. angustifolius/P 21/3/HA 89 Plant code Seed source 2n Male fertility Pollen stainability (%)

G04/54 G04/1-‐13 x HA 89 41 S

G04/55 G04/1-‐13 x HA 89 43 F 0.7

G04/56 G04/1-‐13 x HA 89 -‐ S

G04/57 G04/1-‐13 x HA 89 39 F 5.3

G04/58 G04/1-‐13 x HA 89 42 S

G04/59 G04/1-‐13 x HA 89 39 F -‐

G04/60 G04/1-‐13 x HA 89 41 S

G04/61 G04/1-‐13 x HA 89 40 S

G04/62 G04/1-‐13 x HA 89 40 S

G04/63 G04/1-‐13 x HA 89 39 S

G04/64 G04/1-‐13 x HA 89 41 S

G04/65 G04/1-‐13 x HA 89 38 S

G04/66 G04/1-‐13 x HA 89 38 S

G04/67 G04/1-‐13 x HA 89 39 S

G04/68 G04/1-‐13 x HA 89 43 F 5.1

G04/69 G04/1-‐13 x HA 89 43 S+

G04/70 G04/1-‐13 x HA 89 41 F 17.9

G04/71 G04/1-‐13 x HA 89 42 S

G04/72 G04/1-‐13 x HA 89 -‐ S

G04/73 G04/1-‐13 x HA 89 42 S

G04/74 G04/1-‐13 x HA 89 39 S

G04/75 G04/1-‐13 x HA 89 42 F 52.3

G04/76 G04/1-‐13 x HA 89 43 F 1.1

G04/77 G04/1-‐13 x HA 89 46 F 15.3

G04/78 G04/1-‐13 x HA 89 -‐ S+

G04/79 G04/1-‐13 x HA 89 39 S

G04/80 G04/1-‐13 x HA 89 37 F 79.1

G04/81 G04/1-‐13 x HA 89 41 S

G04/83 G04/1-‐13 x HA 89 40 F 19.3

G04/84 G04/1-‐13 x HA 89 42 S

G04/85 G04/1-‐13 x HA 89 38 S

G04/87 G04/1-‐13 x HA 89 40 S

G04/88 G04/1-‐13 x HA 89 38 S

G04/89 G04/1-‐13 x HA 89 43 S

G04/90 G04/1-‐13 x HA 89 38 S

G04/92 G04/1-‐13 x HA 89 39 S+

Z. Liu et al. 3 SI

G04/93 G04/1-‐13 x HA 89 43 S

G04/94 G04/1-‐13 x HA 89 44 F 1.9

G04/95 G04/1-‐13 x HA 89 39 S

G04/96 G04/1-‐13 x HA 89 41 S+

Z. Liu et al. 4 SI

Table S2 Chromosome number, male fertility, and pollen stainability of the progenies from seven MF plants (2n=35) derived from CMS 514A//Amp H. angustifolius/P 21/3/2*HA 89, and an MS plant G07/553 (2n=34) crossed with an MF plant G07/517 (2n=34) Plant code Seed source Pedigree 2n Male

fertility

Pollen

stainability (%)

G07/501 G06/44 CMS 514A//Amp H. angustifolius/P 21/3/2* HA 89/4/HA 821/5/HA 89 and Self 34 S









G07/510 G06/44 CMS 514A//Amp H. angustifolius/P 21/3/2* HA 89/4/HA 821/5/HA 89 and Self 35 F 91.6













G07/523 G06/52 CMS 514A//Amp H. angustifolius/P 21/3/2* HA 89/4/HA 821/5/HA 89 and Self 35 F


Z. Liu et al. 5 SI





G07/529 G06/55 CMS 514A//Amp H. angustifolius/P 21/3/2* HA 89/4/S x F, SIB/5/HA 89 and Self 35 F 18.1

G07/530 G06/55 CMS 514A//Amp H. angustifolius/P 21/3/2* HA 89/4/S x F, SIB/5/HA 89 and Self 36 S



G07/533 G06/55 CMS 514A//Amp H. angustifolius/P 21/3/2* HA 89/4/S x F, SIB/5/HA 89 and Self 35 F











G07/544 G06/75 CMS 514A//Amp H. angustifolius/P 21/3/2* HA 89/4/HA 821/5/HA 89 and Self 35 F-‐ 30.6









Z. Liu et al. 6 SI











G07/602 G07/553 x G07/517 CMS 514A//Amp H. angustifolius/P 21/3/2* HA 89/4/HA 821/5/HA 89 and Self/6/S x F, SIB

S


S


S


S


S


S


S


S


F 74.7


S


F 96


S


S


S


S


S


S


S

Z. Liu et al. 7 SI


S


S


S


F 90.2


F 33


S


S


S


S


F 81.3


S


S

G07/633 G07/553 x G07/517 CMS 514A//Amp H. angustifolius/P 21/3/2* HA 89/4/HA 821/5/HA 89 and Self/6/S x F, SIB F 45.6

Z. Liu et al. 8 SI

Table S3 Root tip chromosome number and male fertility of the progenies derived from the cross of G09/2614 (2n=35) x HA 89 Plant code Temp ID. 2n Note Male fertility

G10/1001 G09/2614 x HA 89-‐1 34

S

G10/1002 G09/2614 x HA 89-‐2 35 one large chromosome F

G10/1003 G09/2614 x HA 89-‐3 34

S



G10/1006 G09/2614 x HA 89-‐6 34

S

G10/1007 G09/2614 x HA 89-‐7 34

S

G10/1008 G09/2614 x HA 89-‐8 34

S

G10/1009 G09/2614 x HA 89-‐9 34

S

G10/1010 G09/2614 x HA 89-‐10 34

S

G10/1011 G09/2614 x HA 89-‐11 34

S

G10/1012 G09/2614 x HA 89-‐12 34

S


G10/1014 G09/2614 x HA 89-‐14 34

S


G10/1016 G09/2614 x HA 89-‐16 34

S




G10/1020 G09/2614 x HA 89-‐20 34

S


G10/1022 G09/2614 x HA 89-‐22 34

S

G10/1023 G09/2614 x HA 89-‐23 34

S

G10/1024 G09/2614 x HA 89-‐24 34

S


G10/1026 G09/2614 x HA 89-‐26 34

S

G10/1027 G09/2614 x HA 89-‐27 34

S


G10/1029 G09/2614 x HA 89-‐29 34

S

G10/1030 G09/2614 x HA 89-‐30 34

S

G10/1031 G09/2614 x HA 89-‐31 34

S

G10/1032 G09/2614 x HA 89-‐32 34

S

G10/1033 G09/2614 x HA 89-‐33 34

S

G10/1034 G09/2614 x HA 89-‐34 34

S


G10/1036 G09/2614 x HA 89-‐36 34

S

Z. Liu et al. 9 SI

G10/1037 G09/2614 x HA 89-‐37 34

S



G10/1040 G09/2614 x HA 89-‐40 34

S

G10/1042 G09/2614 x HA 89-‐42 34

S

G10/1043 G09/2614 x HA 89-‐43 34

S

G10/1044 G09/2614 x HA 89-‐44 34

S


G10/1046 G09/2614 x HA 89-‐46 34

S

G10/1047 G09/2614 x HA 89-‐47 34

S


G10/1049 G09/2614 x HA 89-‐49 34

S

G10/1050 G09/2614 x HA 89-‐50 34

S


G10/1052 G09/2614 x HA 89-‐52 34

S

G10/1053 G09/2614 x HA 89-‐53 34

S

G10/1054 G09/2614 x HA 89-‐54 34

S


G10/1056 G09/2614 x HA 89-‐56 34

S



G10/1059 G09/2614 x HA 89-‐59 34

S

G10/1060 G09/2614 x HA 89-‐60 34 S

Z. Liu et al. 10 SI

Table S4 Male fertility and pollen stainability of the self-‐pollinated progenies of three MF 2n=34 plants, G07/610, G07/612 and G07/623 Plant code Seed source Male fertility Pollen stainability (%)

G08/501 G07/610 selfed F 6.6

G08/502 G07/610 selfed S

G08/503 G07/610 selfed S

G08/504 G07/610 selfed S

G08/505 G07/610 selfed S

G08/506 G07/610 selfed F (50% Sectional S) 19.2

G08/507 G07/610 selfed S

G08/508 G07/610 selfed S

G08/509 G07/610 selfed F 90.4

G08/510 G07/610 selfed F 79.8

G08/511 G07/610 selfed F 82.1

G08/512 G07/610 selfed F 88

G08/513 G07/610 selfed S

G08/514 G07/610 selfed F 91.7

G08/516 G07/610 selfed F 64.8

G08/517 G07/610 selfed S

G08/518 G07/610 selfed F 84.9

G08/519 G07/610 selfed F

G08/520 G07/610 selfed F 33.2

G08/521 G07/610 selfed F 92.5

G08/522 G07/610 selfed S

G08/523 G07/610 selfed F 83.1

G08/524 G07/610 selfed F 96.1


G08/526 G07/610 selfed F

G08/527 G07/610 selfed S

G08/528 G07/610 selfed F 11.1

G08/529 G07/610 selfed F 86.7


G08/531 G07/612 selfed S

G08/532 G07/612 selfed S

G08/533 G07/612 selfed F 92.5

G08/534 G07/612 selfed F 91.6

G08/535 G07/612 selfed F 99.3

G08/536 G07/612 selfed F 34.5

G08/537 G07/612 selfed S

Z. Liu et al. 11 SI

G08/538 G07/612 selfed F 29.6

G08/539 G07/612 selfed F 27.3

G08/540 G07/612 selfed S

G08/541 G07/612 selfed F 91.2

G08/542 G07/612 selfed F 93

G08/543 G07/612 selfed S

G08/544 G07/612 selfed F 85.9

G08/545 G07/612 selfed F 89.7

G08/546 G07/612 selfed F 93

G08/547 G07/612 selfed F 96.3

G08/548 G07/612 selfed S

G08/549 G07/612 selfed S

G08/550 G07/612 selfed F 88.6

G08/551 G07/612 selfed F 96.8

G08/553 G07/612 selfed F 96.9

G08/554 G07/612 selfed S

G08/555 G07/612 selfed F 87

G08/556 G07/612 selfed F

G08/557 G07/612 selfed F 28.1

G08/558 G07/612 selfed S

G08/559 G07/612 selfed F 82.2

G08/560 G07/612 selfed F 95.1

G08/561 G07/623 selfed S

G08/562 G07/623 selfed F 95.2

G08/563 G07/623 selfed F 94.3

G08/564 G07/623 selfed F 97.4

G08/565 G07/623 selfed F 97.3

G08/566 G07/623 selfed F 98.1

G08/567 G07/623 selfed S

G08/568 G07/623 selfed S

G08/569 G07/623 selfed S

G08/570 G07/623 selfed S

G08/571 G07/623 selfed F 96.8

G08/572 G07/623 selfed F 98.1

G08/573 G07/623 selfed F 82.7

G08/574 G07/623 selfed F-‐

G08/575 G07/623 selfed F 93.4

G08/576 G07/623 selfed F 92.5

G08/577 G07/623 selfed F 91.5

Z. Liu et al. 12 SI

G08/578 G07/623 selfed S

G08/579 G07/623 selfed S

G08/580 G07/623 selfed F 51.1

G08/581 G07/623 selfed F 81.4

G08/582 G07/623 selfed S

G08/583 G07/623 selfed S

G08/584 G07/623 selfed S

G08/585 G07/623 selfed F 98.3

G08/586 G07/623 selfed F 83.8

G08/587 G07/623 selfed F 96.7

G08/588 G07/623 selfed F 98.2

G08/589 G07/623 selfed F 94.3

G08/590 G07/623 selfed F 96.7

Z. Liu et al. 13 SI

Table S5 Male fertility, pollen stainability, F3 progeny test, GISH and molecular marker analyses of the F2 plants Table S5 is available for download at http://www.genetics.org/lookup/suppl/doi:10.1534/genetics.112.146092/-‐/DC1.

diversifying sunﬂower germplasm by integration and mapping

Documents