The recent identification of several proteins playing key roles in S-RNase-based gametophytic self-incompatibility has led both to a greater understanding of the molecular biology of this response, as well as to questions regarding the precise mechanism by which compatible pollen tubes are recognized and accepted. A proposed variant SCFSLF (where SCF is SSK1/cullin/F-box and SLF is S-locus F-box) ubiquitin ligase complex is thought to play a central role in recognizing and inhibiting non-self S-RNases, but the exact role of ubiquitination remains unclear. How the possible sequestration of non-self S-RNases in a pollen vacuolar compartment can be reconciled with the need for protein interaction between S-RNase and the SCFSLF complex needs to be determined. Current work to answer these questions focuses on more precisely defining quantitative protein interactions and subcellular localization of proteins involved in S-RNase-based gametophytic self-incompatibility.
S-RNase-based GSI (gametophytic self-incompatibility)
Self-incompatibility is the ability of an otherwise fertile hermaphroditic plant to recognize and reject its own (‘self’) pollen, thus preventing fertilization, while accepting cross or ‘non-self’ pollen from a different individual in the same species. This behaviour was first recognized and described by Darwin, in his book The Effects of Cross and Self Fertilisation in the Vegetable Kingdom . Later, Mather , Linskens  and de Nettancourt , among others, recognized that self-incompatibility recognition, and pollen rejection, was controlled by a single genetic locus with multiple alleles, termed the S-locus. In subsequent years, different self-incompatibility systems have been studied and characterized [4–13]. These include S-RNase-based GSI (the subject of the present paper), GSI in Papaver, SSI (sporophytic self-incompatibility), self-incompatibility in heterostylous plants and self-incompatibility in grasses. Of these systems, the ones that have been best characterized at the molecular level are S-RNase-based GSI in Solanaceae, Plantaginaceae and Rosaceae [5–9], the GSI system in Papaver rhoeas  and SSI in Brassicaceae [11–13]. Although all three of these systems have the same end effect (i.e. the rejection of self-pollen), the genetics, timing and specific genes and proteins involved differ among the three systems. In the two GSI systems, recognition specificity is based on the haploid genotype of individual pollen grains. In such systems, so-called ‘half-compatible’ pollinations are possible with self-pollen being rejected, and non-self pollen accepted within the same pistil . In SSI, pollen recognition specificity is determined by the diploid genotype of the pollen parent, because recognition proteins synthesized by the diploid parent are deposited in pollen cell walls during pollen maturation. Advances in our understanding of the molecular basis of GSI in Papaver rhoeas and of SSI are discussed by other authors in this issue of Biochemical Society Transactions. The remainder of the present paper therefore concentrates on our current understanding of the molecular biology of GSI in the S-RNase-based system, with the emphasis on our understanding of this response in Petunia hybrida.
Genes and proteins involved in S-RNase-based GSI
Over the last several years, a number of different genes have been identified and shown to play critical or presumed roles in GSI. Two of these genes, S-RNase and SLF/SFB (S-locus F-box), have been identified as pistil-S and pollen-S (i.e. the genes controlling recognition specificity in pistil and pollen respectively). Three other genes, HT-B, 120K and SSK1 (SLF-interacting Skp1-like1), have been shown to be essential for pollen rejection or acceptance. A sixth gene, SBP1 (S-RNase-binding protein 1), is known to interact with S-RNase and SLF and (along with SLF and SSK1) may be a component of a proposed SCFSLF (where SCF is SSK1/cullin/F-box) ubiquitin ligase complex.
S-RNase is the pistil component of GSI
The ability to selectively inhibit the growth of self pollen is determined in the style by an S-locus-encoded ribonuclease known as the S-RNase. All S-RNase proteins characterized to date contain five conserved and two hypervariable domains. Conserved domains C3 and C4 contain the active site of the ribonuclease, and ribonuclease activity is required for the ability to reject pollen . The two hypervariable domains have been demonstrated to control allele specificity , and the protein crystal structure has been determined for the SF11S-RNase of Nicotiana alata . Immunolocalization experiments have demonstrated that both incompatible and compatible S-RNases are imported into pollen tubes [17,18]. The extensive knowledge of S-RNase structure, expression and function has enabled researchers to design different genetic and biochemical experiments (e.g. protein interaction assays) to dissect further the molecular basis of GSI.
SLF/SFB is the pollen-recognition component of GSI
The pollen-recognition component of GSI, SLF (SFB in Rosaceae) was initially cloned via chromosome walking and sequencing to identify pollen-expressed genes linked to the S-locus [19,20]. Confirmation that SLF was pollen-S came from experiments that phenocopied the well-documented phenomenon of competitive interaction, where expression of two different pollen-S alleles in pollen results in the breakdown of GSI [21,22]. SLF has an N-terminal F-box domain characteristic of the recognition components of SCF E3 ubiquitin ligase complexes .
SBP1 interacts with S-RNase and SLF and is an E3 ubiquitin ligase
To attempt to identify pollen-expressed proteins that interact with the S-RNase, Sims and Ordanic  screened a yeast two-hybrid library from mature pollen of P. hybrida with a bait construct for the N-terminal half of the P. hybrida S1-RNase. This screen identified a gene, named PhSBP1 (P. hybrida SBP1) that bound to N-terminal, but not C-terminal, regions of the S-RNase. Sequence characterization of SBP1 (as PhSBP1 will be referred to henceforth) indicated that it contained a C-terminal RING-HC (or C3HC4) protein domain. Such domains have been shown to be characteristic of E3 ubiquitin ligases, the components of the ubiquitin–proteasome system that interact with specific substrates targeted for ubiquitination and protein turnover . Subsequently, SBP1 was cloned from Solanum tuberosum, Petunia inflata and Nicotiana alata [26–28].
HT-B and 120K proteins are required for pollen rejection
Two style-expressed proteins, HT-B and 120K, are also imported into growing pollen tubes, and transgenic experiments have shown that both are required for the ability to reject incompatible pollen [29,30] Recently, Puerta et al.  cloned HT-B from P. hybrida. Down-regulation of this gene by RNAi (RNA interference) caused a partial breakdown of incompatibility, with only highly suppressed plants showing this partial breakdown.
SSK1 interacts with SLF and cullin and may be an essential component of a SCFSLF complex
Huang et al.  identified a Skp1-like protein, AhSSK1 (Antirrhinum hispanicum SSK1), in yeast two-hybrid screens with AhSLF1 (A. hispanicum SLF1) as bait. AhSSK1 interacted with another scaffold component of SCF complexes, CUL1. Zhao et al.  have cloned the orthologue of AhSSK1 from P. hybrida (PhSSK1), and confirmed that it interacts with PhSLF-S1L and PhSLF-S3L. Furthermore, down-regulation of PhSSK1 in transgenic P. hybrida, although not affecting the ability to reject incompatible pollen, severely reduced the fertility of compatible pollen.
Models for pollen recognition and rejection in GSI
An understanding of the biochemical properties of the proteins described above, along with the genetic characteristics of GSI have allowed the development of models for pollen rejection (incompatible pollination) and pollen acceptance (compatible pollination) for S-RNase-based GSI. A generally accepted model for pollen rejection proposes a cytotoxic mechanism resulting in incompatibility. According to this model, in an incompatible pollination, S-RNase imported into pollen tubes acts to degrade pollen-tube RNA, leading to inhibition of protein synthesis and decreased growth. What is also generally accepted is that compatible pollination results from the inhibition of S-RNase activity inside compatible pollen tubes [5–9]. At present, however, two different models (which are not necessarily mutually exclusive) have been proposed to explain the mechanism of S-RNase inhibition in self-compatible pollinations. One model proposes the involvement of ubiquitination of S-RNase by an SCFSLF E3 ubiquitin ligase complex, the other model proposes that compatible S-RNases remain sequestered in a pollen vacuole during compatible pollinations.
The ubiquitination model
Much of the evidence for this model comes from protein interaction assays, along with the known characteristics of the interacting proteins. Pollen-S (SLF) is an F-box protein, and F-box proteins are the recognition components of SCF ubiquitin ligase complexes . SBP1, which interacts with S-RNase, SLF, PhUBC (P. hybrida ubiquitin-conjugating enzyme, an E2 conjugation protein), and Cullin1 [24,27,34] is a RING-HC protein, which are also E3 ubiquitin ligases , and SBP1 has E3 ubiquitin ligase activity in vitro . AhSSK1  and PhSSK1  are SKP1-like proteins that interact with SLF and Cullin. Pollen extracts have been shown to ubiquitinate S-RNase proteins, albeit in an allele-independent manner . Together, these results have led to the proposal that a SCFSLF-like complex acts to recognize and ubiquitinate S-RNases, leading to the inhibition of S-RNase activity in compatible pollen tubes. This complex is proposed to differ from a canonical SCF complex, because neither SKP1 orthologues [7,27,33] nor RBX1 (RING box 1)  interact with SLF or cullin. Instead either (or both) SBP1 and SSK1 have been proposed to replace RBX1 and/or SKP1 in this complex [5–7,33]. According to the simplest version of this model, recognition of non-self S-RNases by the SCFSLF complex would lead to polyubiquitination and degradation of S-RNase by the 26S proteasome complex [5–7]. To date, however, there is little evidence for degradation of S-RNase in compatible pollinations . An alternative possibility is that mono-ubiquitination of S-RNase could result in directing this protein to an endosomal compartment [33,35], although, again, there is little evidence for this. One prediction of the SCFSLF ubiquitin ligase complex model is that down-regulation of SLF, SBP1 or SSK1 should render all pollen tubes incompatible, regardless of genotype. To date, results of experiments to down-regulate SLF or SBP1 have not been published, but experiments where SSK1 was down-regulated convert compatible pollinations into incompatible pollinations. Conversely, mutants in the Rosaceae that lack either the 3′ end of SFB or appear to have deleted this gene altogether, are self-compatible . This result has led to the suggestion that details of pollen rejection may differ in Solanaceae and Rosaceae . Figure 1 summarizes the known protein interactions and components of the proposed SCFSLF E3 ubiquitin ligase complex.
Proposed SCFSLF ubiquitin ligase complex
The sequestration model
Goldraij et al. , working with Nicotiana, hybridized anti-S-RNase antibodies along with anti-callose, anti-aleurain (a marker for vacuolar lumen) and anti-vPPase (a marker for vacuolar membrane) to fixed paraffin-embedded sections, then visualized fluorescence using confocal microscopy. These authors concluded that S-RNase was initially sequestered in a vacuolar compartment in pollen in both compatible and early-stage (16 h) incompatible pollen tubes, but that this compartment broke down at later stages (36 h) of incompatible pollinations, releasing S-RNase into the pollen-tube cytoplasm. The 120K protein appeared to localize to the periphery of this presumed vacuolar compartment, whereas HT-B, which could be detected in incompatible pollen tubes, was not detected in compatible pollen tubes. This model suggests possible roles for the requirement of HT-B and 120K expression for rejection of incompatible pollen tubes, but is difficult to reconcile with the need for SLF and S-RNase to interact (since, presumably the two proteins would be on opposite sides of the vacuolar membrane).
Current experiments and future prospects
The above models suggest certain predictions for both protein interaction strength and subcellular localization that we are currently attempting to test. For example, the ubiquitination model predicts that protein interactions between SCFSLF complex proteins should be stronger with non-self S-RNase compared with self-S-RNase. This model also predicts that SLF, S-RNase and other SCFSLF complex components should co-localize in the same subcellular compartment. For the sequestration model, it will be important to confirm, using approaches giving the highest degree of resolution and tissue preservation, the precise subcellular location of different proteins at different times following compatible and incompatible pollinations.
Quantitative protein interaction assays
As a first step to determine the relative strength of interaction of non-self S-RNase–SCFSLF interactions compared with self-interactions, we have used a quantitative yeast two-hybrid assay to measure the relative degree of lacZ reporter gene expression between different subdomains of S1- and S3-RNase alleles with SLF-S1 from P. hybrida. These assays used a fluorescent substrate for β-galactosidase activity that is substantially more sensitive than previous substrates . When subdomains of the S1-RNase were used in two-hybrid assays with the SLF-S1 pollen-S allele from P. hybrida, the level of β-galactosidase activity detected was significantly less than the level seen when the equivalent domains of the S3-RNase were used in interaction assays with SLF-S1 (Figure 2). These data are consistent with the predictions of the ubiquitination model (and possibly with the sequestration model as well). These experiments are currently being refined and extended using a variety of bait and prey constructs, by expressing potential competing protein constructs from the pTFT1 vector , and by using BiFC (bimolecular fluorescence complementation)  to assay protein interactions in planta.
Quantitative yeast two-hybrid assays
The isolation of the pollen-S protein, SLF/SFB, along with putative SCFSLF complex proteins SBP1 and SSK1, and the further defining of the roles of HT-B and 120K proteins has given us a better understanding of self-incompatibility interactions. These advances have also led to alternative, possibly competing, models for the detailed mechanism of how S-RNases are recognized and either inhibited or released in an apparent cytotoxic response. In the future, it will be critical to determine more precisely the exact subcellular location of the different proteins involved in S-RNase-based GSI, at different times following compatible or incompatible pollinations. The use of high-resolution immunoelectron microscopy as well as in planta protein interaction assays should help to resolve these questions.
Cell–Cell Communication in Plant Reproduction: A Biochemical Society Focused Meeting held at University of Bath, Bath, U.K., 14–16 September 2009. Organized and Edited by James Doughty (Bath, U.K.) and Rod Scott (Bath, U.K.).
This work was supported by the Plant Molecular Biology Center of Northern Illinois University.