Heredity (2010) 105, 161–162 & 2010 Macmillan Publishers Limited All rights reserved 0018-067X/10 $32.00 www.nature.com/hdy NEWS AND COMMENTARY ...... Delineation of the S locus in ...... subulata ...... short homostyle did not reveal a simple explanation, but analysis of the long Homing in on heterostyly homostyle indicates that the dominant PM Gilmartin and J Li alleles responsible for development of a ...... short style (G) and large grains (P), but not that which determines Heredity (2010) 105, 161–162; doi:10.1038/hdy.2010.69; published online 9 June 2009 high anthers (A), have been lost. These data provide evidence of a coadapted linkage group in Turnera, often referred to as a ‘supergene’, as previously he majority of are hermaph- affect heteromorphic flower develop- proposed in Fagypyrum and Primula rodite and produce both male and ment, whereas previous studies in (see for example Richards, 1997 and T female gametes. In addition to the F. esculentum have focused on the Matsui et al., 2004). various contrivances by which plants identification of self-fertile mutants Although the mutant screen could facilitate outcrossing—using wind, in- (see for example Matsui et al., 2004 and have been extended to identify mutants sects and other animals—an elaborate the references therein). In Primula, self- showing the expected short-styled range of mechanisms has evolved to fertile and homostyle plants resulting phenotype that were rendered self- prevent self-fertilization. One such from recombination within the S locus fertile following mutagenesis, the large mechanism, known as floral heteromor- have defined different genetic functions deletions defined in those plants ana- phy, results in the development of at the Primula S locus responsible for lyzed suggest that it is unlikely that different forms of self-incompatible style length (G), anther position (A), such self-fertile short-styled plants flowers on different individual plants pollen size (P) and both pollen (IP) and would have been identified, as this (Darwin, 1877). Development of the style (IS) incompatibility phenotypes would have required deletion of genes distinct floral morphs with different demonstrating the presence of a co- responsible for pollen or stigma SI anther height and style lengths is adapted linkage group of genes rather functions (IP or IS) within the S locus orchestrated by the S locus (see than the presence of a single master without loss of the genes (G, P or A) Richards, 1997). In species such as regulator (Dowrick, 1956; Richards, responsible for heteromorphic architec- Turnera subulata (white ), Fagopyr- 1997). The existence of a similar co- ture (Dowrick, 1956; Richards, 1997). um esculentum (buckwheat) and Primula adapted linkage group has been pro- What is remarkable, however, is that in vulgaris (primrose), which produce two posed in F. esculentum (see Matsui et al., the three heteromorphic species cur- forms of flower with long and short 2004 and the references therein). rently under investigation, P. vulgaris, styles, this phenomenon is also known The Shore laboratory has pioneered F. esculentum and T. subulata, which fall as distyly and is coordinated by two the development of T. subulata as within three distinct orders, namely, alleles of the S locus (see for example a model for the study of heteromorphy, Ericales, Caryophyllales and Malpigh- Matsui et al., 2004; Li et al., 2007; and the scale and scope of this current iales, in different clades of the core Labonne et al., 2009). In contrast, most study is impressive. The data presented , independent evolution has self-incompatibility (SI) systems, in are an accumulation of nearly a decade resulted in similar predicted S locus which self- is prevented by of study starting with the generation of architectures. rejection of pollen following molecular genetically defined parental lines, an The identification of morph-specific self-recognition, occur in otherwise in- unusual homozygous (SS) short-styled genes and proteins in Primula (McCub- distinguishable homomorphic flowers , which provided the pollen for X- bin et al., 2006) and Turnera (Athanasiou (see for example Hiscock and McInnis, irradiation, and the long-styled (ss) and Shore, 1997) that are not asso- 2003). To date, much greater progress pollen recipient, through to the analysis ciated with the S locus provide an has been made toward a detailed mole- of 3982 progeny plants. All progeny opportunity to characterize the regu- cular understanding of homomorphic SI from this cross should have been short latory functions of the S locus control- in a range of species than SI associated styled with genotype Ss, but 10 long- ling heteromorphy in different species. with floral heteromorphy. However, the styled mutants were obtained, suggest- Indeed, a further key observation from recent study by Labonne et al.(thisissue) ing deletion of the dominant S allele, Labonne et al. (2010) is the finding provides a key step forward toward the together with a short-homostyle and a that expression of the two morph- identification of genes located at the long-homostyle plant. specific Turnera proteins known to be S locus that control floral heteromorphy In parallel to the mutagenesis studies, encoded by genes not associated with in distylous T. subulata. In this species, the Shore laboratory have painstakingly the S locus is absent in the long-styled long-styled plants are homozygous re- identified a number of S locus-linked and long-homostyle deletion mutants. cessive (ss) and short-styled plants are genes and DNA markers (Labonne et al., This observation defines a regu- heterozygous (Ss) with respect to the 2009), together with morph-specific latory role of the S locus in the modula- two S locus alleles. proteins (Athanasiou and Shore, 1997), tion of unlinked genes that encode In their paper, Labonne et al. (this issue) which were used to delimit the extent of morph-specific characteristics. present an X-ray deletion mutagenesis the S locus deletions within these 12 A number of unknowns remain to be screen of T. subulata. Although muta- mutants. As would be predicted by elucidated in Turnera and other hetero- genesis has been successfully used in analogy to the known architecture of morphic models, such as the relation- studies of homomorphic SI systems, this the Primula S locus (Dowrick, 1956; ship between genetic map distance and is the first published systematic screen Richards, 1997), the long-styled plants physical map distance in each species, in a heterostyled plant aimed at defin- seem to be derived from pollen in which and the extent of recombination sup- ing the S locus. This study focused the entire dominant S allele has been pression within the S locus. Some of the on the identification of mutants that deleted. Subsequent analysis of the S locus deletions presented in Labonne News and Commentary 162 et al. (2010) span up to 7 Mb, others are Professor PM Gilmartin and Dr J Li are at the Matsui K, Nishio T, Tetsuka T (2004). Genes smaller but result in the loss of only one School of Biological and Biomedical Sciences, outside the S supergene suppress S functions DNA marker and so the size cannot be University of Durham, Durham, DH1 3LE, UK. in buckwheat (Fagopyrum esculentum). Ann Bot 94: 805–809. estimated. However, a previous study e-mail: [email protected] McCubbin A (2008). Self-Incompatibility in Flowering suggested that the locus could be as Plants. ( Berlin: Springer.). Athanasiou A, Shore JS (1997). Morph-specific McCubbin AG, Lee C, Hetrick A (2006). compact as 32 kb (Labonne et al., 2009). proteins in pollen and styles of distylous Identification of genes showing differential Knowledge of the physical size of the Turnera (). Genetics 146: 669–679. expression between morphs in developing S locus in Primula and Fagopyrum also Darwin C (1877). The Different Forms of Flowers on flowers of Primula vulgaris. Sex Plant Reprod Plants of the Same Species. John Murray: remains elusive, but with rapid progress 19: 63–72. London. being made in the analysis of the S Richards AJ (1997). Plant Breeding Systems 2nd Dowrick VPJ (1956). Heterostyly and homostyly edn. Chapman and Hall: London. locus-linked genes, generation of genet- in Primula obconica. Heredity 10: 219–236. Yasui Y, Mori M, Matsumoto D, Ohnishi O, ic maps and screening of BAC libraries Hiscock SJ, McInnis SM (2003). The diversity of Campbell CG, Ota T (2008). Construction of a self-incompatibility systems in flowering (Li et al., 2007, 2010; Labonne et al., BAC library for buckwheat genome research— plants. Plant Biol 5: 23–32. 2008, 2009; McCubbin, 2008; Yasui et al., an application to positional cloning of agri- Labonne JDJ, Vaisman A, Shore JS (2008). Con- culturally valuable traits. Genes Genet Syst 83: 2008), it is only a matter of time before struction of a first genetic map of distylous 393–401. sequences corresponding to each S Turnera and a fine-scale map of the S-locus locus are available for comparison of region. Genome 51: 471–478. gene organization, function and genetic Labonne JJD, Goultiaeva A, Shore JS (2009). High- Editor’s suggested reading architecture. The current paper by resolution mapping of the S-locus in Turnera leads to the discovery of three genes tightly Brennan AC, Tabah DA, Harris SA, Hiscock SJ Labonne et al. (this issue) takes us a associated with the S-alleles. Mol Genet (2010). Sporophytic self-incompatibility in step closer to homing in on the genes Genomics 281: 673–685. Senecio squalidus (Asteraceae): S allele domi- that control floral heteromorphy; one of Li J, Dudas B, Webster MA, Cook HE, Davies BH, nance interactions and modifiers of cross- the most remarkable mechanisms for Gilmartin PM (2010). Hose in hose, an S locus- compatibility and selfing rates. Heredity linked mutant of Primula vulgaris, is caused by (e-pub ahead of print 7 April 2010; preventing self-pollination and promo- an unstable mutation at the Globosa locus. Proc doi:10.1038/hdy.2010.29). ting outbreeding in plants. Natl Acad Sci USA 107: 5664–5668. Labonne JDJ, Hilliker AJ, Shore JS (2007). Meiotic Li J, Webster MA, Furuya M, Gilmartin PM recombination in Turnera (Turneraceae): Conflict of interest (2007). Identification and characterization of extreme sexual difference in rates, but no pin and thrum alleles of two genes that evidence for recombination suppression The authors declare no conflict of co-segregate with the Primula S locus. Plant J associated with the distyly (S) locus. Heredity interest. 51: 18–31. 98: 411–418.

Heredity