Journal of Experimental Botany, Vol. 54, No. 380, pp. 157-168,
January 1, 2003
© 2003 Oxford University Press
Just how complex is the Brassica S-receptor complex?
Received 4 July 2002; Accepted 12 September 2002
Department of Biology and Biochemistry, University of Bath, Bath BA2 7AY, UK
1 To whom correspondence should be addressed. Fax: +44 (0)1225 386779. E-mail: bssjd{at}bath.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionA model for S-receptor...Variations on a theme?Molecular basis of dominanceSignal transduction and...The role of SLGPCPsSI and interspecific...References |
|---|
Of the plant self-incompatibility (SI) systems investigated to date, that possessed by members of the Brassicaceae is currently the best understood. Whilst the recent demonstrations of interactions between the male determinant (S-locus cysteine rich protein, SCR) and the female determinant (S-locus receptor kinase, SRK) indicate the minimal requirement for SI in Brassica, no consensus exists as to the nature of these molecules in vivo and the potential involvement of accessory molecules in establishing the active S-receptor complex. Variation between S haplotypes appears to be present in the molecular composition of the receptor complex, the regulation of downstream signalling and the requirement for accessory molecules. This review discusses what constitutes an active receptor complex and highlights potential differences between haplotypes. The role of accessory molecules, in particular SLG (S-locus glycoprotein) and low molecular weight pollen coat proteins (PCPs), in pollination are discussed, as is the link between SI and unilateral incompatibility (UI).
Key words: Brassica, interspecific incompatibility, ligand, pollination, receptor, recognition, SCR, self-incompatibility, signalling, SRK.
|
Introduction |
|---|
|
TopAbstractIntroductionA model for S-receptor...Variations on a theme?Molecular basis of dominanceSignal transduction and...The role of SLGPCPsSI and interspecific...References |
|---|
Many flowering plant species possess mechanisms that promote outbreeding. Amongst these, the self-incompatibility (SI) systems, where pollen between closely related individuals is recognized and rejected by the pistil, have been considered to be the most important, being highly efficient and arguably the most evolutionarily advanced. Indeed some estimates suggest that at least half of the 250 000 species of angiosperms possess SI, further highlighting its evolutionary importance (Brewbaker, 1959; Darlington and Mather, 1949). SI systems can be broadly divided into two groups, these being defined by the genetic control of incompatibility in pollen. Thus in gametophytic SI the incompatibility phenotype of the pollen is determined by its own haploid genome, whereas in sporophytic SI the genotype of the pollen producing plant (the diploid sporophyte) determines the pollen incompatibility phenotype. Typically, for both gametophytic and sporophytic systems SI is controlled by a single genetic locus (S-locus; sterility locus) having multiple allelic versions or specificities; when an allele is shared between pollen and pistil then an incompatible response will follow.
Investigations into the molecular basis of SI in a range of angiosperm families have revealed spectacular diversity in the mode of recognition and rejection of self-pollen. This is perhaps unsurprising, as great selective pressures will constantly act on plants to acquire and maintain outbreeding mechanisms (Charlesworth, 1995; Dickinson et al., 1998). So far, three mechanistically distinct systems have been characterized in some detail at the molecular level in widely divergent plant families. This diversity has been taken as compelling evidence for a polyphyletic origin of SI (Hiscock and Kues, 1999). Self-incompatible members of the Solanaceae including Nicotiana (McClure et al., 1989), Petunia (Ai et al., 1990), Lycopersicon (Tsai et al., 1992), and Solanum (Xu et al., 1990) species reject pollen via the action of a stigmatic S-locus-encoded ribonuclease (S-RNase). The stigmatic S-RNase is held to degrade RNA molecules present in incompatible pollen tubes, which, in turn, leads to an arrest of pollen tube development through interference with protein biosynthesis (McClure et al., 1990). Exactly how this specificity is achieved is at present unclear, as S-RNase has been shown to enter all pollen tubes regardless of their compatibility. The pollen determinant of SI in the Solanaceae has yet to be identified and, until it is, the exact mode of action can only be guessed at. One hypothesis has suggested that it may act as a general RNase inhibitor, inhibiting the action of all S-RNases except where the alleles are matched (Kao and McCubbin, 1996). In Papaver, SI appears to operate through a calcium-dependent signal transduction pathway sited in the pollen that also involves the phosphorylation of specific downstream proteins (Franklin-Tong, 2002). The stigmatic determinants have been identified as small secreted proteins (
15 kDa) that probably act as ligands to an as yet unidentified pollen receptor likely to be located in the pollen plasma membrane (Jordan et al., 2000b). Activation of this signalling cascade following an incompatible pollination ultimately leads to the arrest of pollen tube growth through what appears to be a process carrying many of the hallmarks of programmed cell death (PCD) (Jordan et al., 2000a). The SI system found in members of the Brassicaceae, however, is currently the best understood with both the primary male and female determinants having been characterized. Like the Papaver system just described, a ligand and receptor are involved, but in this case the ligand is carried by the pollen grain and the receptor resides in the papilla cells that cover the stigmatic surface. The stigmatic SI receptor, SRK (for S-locus receptor kinase), is a plasma membrane-spanning serine/threonine receptor kinase, the extracellular domain (or S domain) of which is highly polymorphic between S alleles, as would be expected for a molecule determining specificity (Stein et al., 1991). The S domain of SRK binds the pollen-borne ligand SCR (for S-locus cysteine-rich protein; also designated SP-11) in an allele-specific manner such that the interaction only occurs when both SCR and SRK are from the same haplotype (Kachroo et al., 2001; Takayama et al., 2001). SCR is a small (6–8 kDa), basic, cysteine-rich polypeptide (Schopfer et al., 1999) located on the surface of the pollen grain in the highly lipidic pollen coat (Kachroo et al., 2001; Shiba et al., 2001; Stephenson et al., 1997). SCR is likely to gain speedy access to SRK following pollination, for within minutes of arrival of the grain at the stigmatic surface the pollen coat flows to form a meniscus or ‘foot’ between the interacting partners. EM studies have revealed that the pollen coat rapidly infiltrates microchannels in the stigmatic cuticle establishing hydraulic and thus molecular continuity between pollen and stigma (Dickinson, 1995). Although the timing of the incompatible response is variable between incompatibility haplotypes, rejection of incompatible pollen usually occurs within 45 min. This suggests that interaction of the primary determinants is likely to have occurred well within this time-frame (Dickinson, 1995). In addition to SRK, a second highly polymorphic S-locus-encoded protein, SLG (for S-locus glycoprotein), is expressed in the stigmatic tissue of most Brassica haplotypes (Nasrallah et al., 1985). SLG shares considerable sequence identity with the S domain of SRK and is secreted into the papillar cell wall. By contrast to SRK, SLG is an abundant constituent of the papillar cell wall (abundance 10–100-fold that of SRK) and appears to play an accessory role in SI rather than being absolutely required for pollen rejection (Takasaki et al., 2000). The signalling process downstream of SCR-SRK is as yet poorly understood, although the kinase domain of SRK has been shown to interact in a phosphorylation-dependent manner with ARC1 (for arm repeat-containing protein) and that ARC1 is an essential component of the SI signalling pathway (Stone et al., 1999). The precise mechanism by which incompatible pollen is rejected is still a matter of some considerable conjecture, but a range of studies point to stigmatic regulation of water flow to the pollen grain as being central to the process.
Although the primary determinants of SI in Brassica have been identified and their absolute requirement confirmed, there are likely to be other molecules that have accessory roles in establishing a stable active receptor complex. Experimental evidence is accumulating that SLG is likely to play such a role in many cases and in this respect it is intriguing that PCP-A1 (for Pollen Coat Protein–Class A, 1), a pollen protein with similar characteristics to SCR, binds SLG and possibly SRK (Doughty et al., 1998, 2000).
In addition to the specific molecules that are considered central to functional SI, it must not be forgotten that the SI ‘system’ is embedded and has evolved within a complex molecular environment where other crucial aspects of the pollen–stigma interaction are regulated. For instance, factors governing pollen adhesion, hydration, germination, pollen tube penetration, and directional tube growth are clearly important for compatible pollinations, thus it would not be surprising if SI acted on a number of these stages to varying degrees to ensure efficient pollen rejection (Dickinson, 1995). Some indication that this may be the case has come from the observation that certain S haplotypes are ‘weaker’ than others. These weak or late-acting haplotypes generally support a greater degree of pollen hydration, often permitting the production of a pollen tube that may, in some cases, initiate penetration of the stigmatic cuticle. However, in all cases further pollen tube development is rapidly inhibited, suggesting that tubes that ‘escape’ the initial line of SI defence are halted by back-up mechanisms (Dickinson, 1995; Kachroo et al., 2002). Such inferences should be treated with caution, however, as it is quite possible that such early stages of pollen development could be controlled by the primary mechanism alone, perhaps, as discussed above, by the limitation of water flow from the stigma to the grain.
Over and above the molecular complexities of SI, consideration must also be given to the role of SI-factors in interspecific incompatibility. A growing body of genetic, molecular and physiological data from a broad range of flowering plant families supports a link between SI and interspecific incompatibility (Hiscock and Dickinson, 1993; McClure et al., 2000).
Pollen–stigma recognition in Brassica SI is seen as a paradigm for cell–cell signalling systems in plants; although the primary determinants have been identified and shown to interact, little else is known about the nature and regulation of the interaction. It is clear that the story is far from straightforward, for instance, recent data points to probable haplotypic variation in what constitutes the ‘active’ S-receptor complex. This article will focus on recent advances in an understanding of the molecular mechanism that controls SI in Brassica and will discuss this in the context of the pollen–stigma interaction and interspecific pollination relationships.
|
A model for S-receptor activation in Brassica |
|---|
|
TopAbstractIntroductionA model for S-receptor...Variations on a theme?Molecular basis of dominanceSignal transduction and...The role of SLGPCPsSI and interspecific...References |
|---|
Of the self-incompatibility systems studied, only in Brassica have both male and female components been isolated. In addition, a number of potential accessory molecules and components of downstream signalling pathways have been isolated (Table 1). The female determinant, SRK (S-locus receptor kinase), is a receptor-like kinase with an extracellular domain, a membrane spanning region, and a functional cytoplasmic serine/threonine kinase domain (Stein and Nasrallah, 1993). The extracellular domain protrudes from the membrane into the stigmatic cell wall (Letham et al., 1999) and is termed the S domain because, depending on haplotype, it has between 75% and 99% identity to SLG (S-locus glycoprotein), the first S-locus encoded stigma-specific protein to be isolated (Nasrallah et al., 1985). SRK was deduced to be essential to SI in Brassica because SI broke down in transgenic plants with either silenced SRK (Conner et al., 1997) or expressing dominant negative SRK (Stahl et al., 1998). Proof that SRK was necessary and sufficient for pollen rejection was shown by introducing SRK28 into S60 homozygotes enabling them to reject S28 pollen (Takasaki et al., 2000).
|
The male determinant of Brassica SI was isolated by sequencing large tracts of the S-locus and finding a small gene similar to PCP-A1, previously identified to interact with SLG (Doughty et al., 1998). Termed SCR (S-locus cysteine rich protein) (Schopfer et al., 1999) or SP11 (S-locus protein 11) (Suzuki et al., 1999) the male determinant is a small, basic protein of typically 50–66 amino acids. SCR is highly polymorphic between haplotypes and the only conserved feature in the mature protein are eight cysteines (though some alleles lack the C terminal cysteine (Watanabe et al., 2000)) and a glycine. Generally, all eight of the cysteines are involved in intramolecular disulphide bonds (Takayama et al., 2001). However, SCR homodimers have recently been observed in the presence of SDS and in the absence of reducing agents (Kachroo et al., 2001) indicating that the dimers may be disulphide linked through one or more cysteine residues. When S2 homozygotes were transformed with SCR6 the pollen of transformants was unable to develop on S6 stigmas (Schopfer et al., 1999). Similar results were obtained by introducing SCR8 and SCR9 into a S52S60 heterozygote (Shiba et al., 2001). In addition, pretreatment of stigmas with SCR either produced in E. coli (Kachroo et al., 2001; Shiba et al., 2002; Takayama et al., 2000b) or chemically synthesized (Takayama et al., 2001) is sufficient to induce pollen rejection. SCR is, therefore, the only molecule needed for male specificity. Both SCR6 (Kachroo et al., 2001) and SCR8 (Takayama et al., 2001) have been shown to interact with SRK in an allele-specific manner and it is likely that this direct physical interaction will occur for all haplotypes.
The current model of S-receptor activation is that following contact between the pollen grain and the stigma papaillar cell, SCR diffuses through the extracellular matrix where it is bound by SRK. The binding induces phosphorylation of SRK which initiates a signalling cascade leading to pollen rejection.
|
Variations on a theme? |
|---|
|
TopAbstractIntroductionA model for S-receptor...Variations on a theme?Molecular basis of dominanceSignal transduction and...The role of SLGPCPsSI and interspecific...References |
|---|
Within the strict framework of SRK and SCR determining the outcome of pollen–stigma interactions, variations between haplotypes are becoming apparent. Whilst SRK is the only stigmatic protein absolutely required for S specificity, it is probable that other proteins interact with SRK in stigmas forming a receptor complex. Evidence for this comes from a number of studies where the association of SRK with other molecules (most notably SLG) has been shown either directly by cross-linking and immunoprecipitation or inferred from mutant and transgenic plants. When stigma microsomal fractions of S3 homozygotes were solublized with Triton X-100 (a mild non-ionic detergent) and cross-linked with glutaraldehyde, two complexes were observed that cross-reacted with an SRK antibody. One of 161 kDa, probably consisting of SRK3 complexed with SLG3 or eSRK3, (a truncated version of SRK consisting only of the S domain which is found in the S3 haplotype), and one of 233 kDa, probably composed of two SRK3 molecules (Giranton et al., 2000). Velocity sedimentation of unpollinated S3 stigma extracts on a sucrose gradient indicated that SRK3 is mainly oligomeric under native conditions, in contrast to eSRK and SLG which were mainly monomeric. When the sucrose gradients were run with the addition of 0.5% SDS the SRK complexes were disrupted, indicating that the association was mediated by non-covalent interactions rather than by disulphide bonds between SRK and its interacting partners (Giranton et al., 2000). An interaction between SRK and SLG has also been inferred from the observation that depletion of SLG protein in a number of mutants leads to a lack of SRK protein accumulation despite normal levels of SRK transcript (Dixit et al., 2000). A likely explanation is that mature SRK protein in these haplotypes is unstable and SLG acts either directly or indirectly to stabilize SRK. Further evidence of SLG interacting with SRK in order to stabilize the mature protein was found when co-expression of SLG6 and SRK6 in transgenic tobacco plants prevented the formation of SRK aggregates (Dixit et al., 2000). However, no direct physical interaction between SLG and SRK was observed in unpollinated S8 stigmas, possibly because gels were run in the presence of SDS which disrupts most non-covalent interactions (Dixit et al., 2000). It is likely that the requirement of SLG to stabilize SRK is haplotype-specific and may even be exceptional, especially as some haplotypes do not require SLG for SRK to be functional (Suzuki et al., 2000).
Whilst a complex between SRK and SLG appears to be present in the unpollinated stigmas of at least some haplotypes, the role of such a complex in ligand binding has not been elucidated. By analogy with the CLAVATA complex in Arabidopsis thaliana, where CLV1 and CLV2 appear to form a receptor complex for the ligand CLV3 (DeYoung and Clark, 2001), it can be speculated that SLG and SRK form a complex to bind SCR. It has not yet been reported which of the two SRK3 complexes described by Giranton et al. (2000) bind SCR3 and a receptor complex was not reported in the binding of SCR6 to SRK6 (Kachroo et al., 2001). However, the binding of SCR to a putative receptor complex has been shown in S8 stigmas. Microsomal fractions were incubated with 125I SCR8 and cross-linked with bis[sulphosuccinimidyl] suberate (BS3) before separating on SDS-PAGE. Two proteins bound SCR8; SRK8 and a 65 kDa protein which is probably SLG8 (Takayama et al., 2001). Evidence of an SLG/SRK/SCR complex was indicated by immunoprecipitating solubilized microsomal fractions, which had previously been incubated with 125I SCR8 and cross-linked with BS3, with an SRK antibody raised against the kinase domain. The antibody precipitated SLG and SRK, both of which had bound SCR, when membranes were solubilized with Triton X-100, but only precipitated SRK when the membranes were solubilized with SDS (Takayama et al., 2001). It is interesting that BS3 did not cross-link the SCR8/SLG8/SRK8 complex. BS3 cross-links primary amines that are up to 11 Å away from each other and it is possible that the association between SLG8 and SRK8 is not this close. However, both SRK3 homodimers and SRK3/SLG3 heterodimers were cross-linked using glutaraldehyde (Giranton et al., 2000) which cross-links primary amines 5 Å apart. These differences may be explained either by variation between haplotypes or by assuming that SCR binding alters the complex, forcing SLG and SRK apart. It is noteworthy that not only do SRK8 and SLG8 appear to form a complex, but that both molecules bind SCR8. These data indicate that, in this haplotype, SLG may be an integral part of the S-receptor complex.
The presence of an SRK/SLG complex to bind the SCR may not be ubiquitous, indeed, in the S6 interaction SCR6 and eSRK6 produced in heterologous expression systems (E. coli and Nicotiana benthamiana, respectively) interacted in vitro in the absence of any other molecules (Kachroo et al., 2001). When a similar experiment was attempted using synthetic SCR8 and eSRK8 (produced in insect cells) no in vitro binding was detected (Takayama et al., 2001). The lack of binding can be explained by various practical considerations, such as incorrect folding of heterologously produced eSRK, or the necessity for a functional kinase domain in this haplotype in addition to the need for a receptor complex. The fact that SRK and SCR can interact in vitro in the absence of other molecules does not negate the involvement of other molecules in the stigma. Although no SRK/SLG complexes were found in S6 stigmas, high molecular weight bands (reported to be SLG oligomers) from microsomal membrane fractions of S6 homozygotes bound SCR6 (Kachroo et al., 2001). The presence of a complex to bind SCR appears to be haplotype-specific, but, interestingly, eSRK6 appears to have only one binding site for SCR6 (Kachroo et al., 2001), whereas two were reported from the microsomal membranes of the S8 haplotype (Takayama et al., 2001). Whilst it is tempting to speculate that SLG8 provides the second SCR binding site, until more extensive biochemical studies on SCR/S-receptor complex interactions are published no hard conclusions can be drawn.
On balance, the binding of SCR to SRK is probably independent of other molecules and a property of the extracellular domain alone. However, in planta SRK is probably complexed to other molecules, forming binding site(s) for SCR. This complex may be essential for effective signal transmission, needed to provide extra SCR binding sites, be a result of SLG acting as a shuttle for SCR or may be required to counter the inherently instability of some SRK alleles.
|
Molecular basis of dominance |
|---|
|
TopAbstractIntroductionA model for S-receptor...Variations on a theme?Molecular basis of dominanceSignal transduction and...The role of SLGPCPsSI and interspecific...References |
|---|
It has been known for some time that Brassica S haplotypes are subject to complex dominance relationships in both stigma and pollen (Ockendon, 1974). Co-dominance of S haplotypes is common in both pollen and stigma and dominance relationships are more prevalent in pollen than the stigma (Uyenoyama, 2000). The dominance levels of the pollen and stigma are independent (for example, S52 is dominant to S60 in pollen but co-dominant in the stigma (Takasaki et al., 1999)) and, recently, the molecular basis of these dominance relationships has started to be unravelled.
The division of S haplotypes into Class I and Class II was originally based on the observation that SRK and SLG alleles fall into two distinct families based on sequence homology (Nasrallah et al., 1991). It is thought that the two classes diverged some 40 million years ago (Uyenoyama, 1995). Class II S haplotypes tend to be pollen recessive to Class I haplotypes and, recently, Class II SCR sequences from both B. rapa and B. oleracea have been isolated in order to elucidate how pollen dominance is determined at a molecular level. Although SCR genes are highly divergent, Class II SCR sequences are markedly different from Class I SCRs, notably in the signal peptide, one of the few conserved features of SCR genes (Shiba et al., 2002). Class II SCRs also appear to be less polymorphic than Class I SCRs (63–94% identity compared to 19–76% identity), an observation that may help elucidate regions of the molecule important for specificity. Interestingly, alternative splicing at the intron–exon boundary appears to give two different mRNA species in at least some Class II haplotypes, a phenomenon not found in Class I SCRs (Shiba et al., 2002). Although it is not known if both mRNA species are translated, the presence of two subtly different SCR molecules in the pollen coat could be important in Class II pollen recognition. Class II SCRs also show an exclusively tapetal expression pattern (Shiba et al., 2002) in contrast to the tapetal and gametophytic expression of Class I SCRs (Takayama et al., 2000b). The molecular mechanism governing dominance relationships in pollen was revealed by looking at SCR transcript levels in Class II/Class I heterozygotes. In a S52 (Class I)/S60 (Class II) heterozygote no SCR60 mRNA was detected, in contrast to S60 homozygotes where high levels of SCR transcript were detected (Shiba et al., 2002). These results indicate that the dominance relationships seen between Class I and Class II SCRs are regulated at the RNA level, with dominant alleles suppressing the transcription of recessive alleles. A similar relationship appears to be present in self-incompatible Arabidopsis lyrata, where SCRa (which is only tapetally expressed compared to SCRb which shows tapetal and gametophtic expression) transcripts are suppressed in SaSb heterozygotes (Kusaba et al., 2002). The best characterized mechanisms for suppression of transcript levels are methylation of genomic DNA (Paszkowski and Whitham, 2001) and post-transcriptional gene silencing (Voinnet, 2001) and one would assume that one of these mechanisms is involved in suppressing SCR transcript levels. However, bisulphite sequencing and methylation-sensitive restriction digestion of genomic DNA from the anthers and leaves revealed no differential methylation of SCRa in heterozygotes compared with homozygotes. Further, 21–25 nt degradation products of SCRa (which are indicative of post-transcriptional gene silencing) were not found in the anthers of SaSb heterozygotes (Kusaba et al., 2002). This indicates that SCRa is neither imprinted nor post-transcriptionally silenced, however, it is possible that the regulation is only seen in the tapetal cells which may be swamped by background from other anther cells.
By contrast to pollen dominance, stigma dominance appears to be a property of SRK alone and was not correlated with differences in SRK transcript levels (Hatakeyama et al., 2001). In natural heterozygotes, dominant SRK28 mRNA levels were 1.6–3-fold higher than recessive SRK transcript levels. However, in transgenic lines low SRK28 mRNA levels (about 10-fold less than recessive SRK43) could totally mask the SRK43 phenotype (Hatakeyama et al., 2001). In addition, the A. lyrata Sa haplotype has a weakened SI response in SaSb stigmas compared to the stigmas of Sa homozygotes and this weakening did not correlate with reduced transcript levels (Kusaba et al., 2001). In these studies stigmatic SRK protein levels were not ascertained, so the possibility of post-transcriptional regulation accounting for the dominance effects can not be ruled out. However, assuming that protein from both SRK alleles is synthesized, other explanations for dominance include interaction between the two SRKs or competition for downstream signalling components.
The elucidation of the mechanisms of S allele dominance poses some interesting questions concerning the evolutionary benefits of dominance. Examination of the allelic series in B. oleracea reveals fewer pollen recessive alleles than dominant alleles. However, recessive alleles are maintained at higher frequency than dominant alleles in natural populations (Ockendon, 1974). Recessive pollen alleles have the evolutionary advantage of being able to spread by pollinating heterozygotes containing one copy of the recessive ‘self’ allele (Kusaba et al., 2002). This could explain why pollen dominance relationships appear to be mediated by a mechanism of regulation where transcription is actively inhibited. This indicates either that the recessive allele ‘allows’ itself to be turned off or that the dominant allele actively inhibits the recessive allele. By contrast, stigmatic dominance is much less frequent in Brassica. This could be a function of a less ‘active’ form of regulation, which appears to be mediated either by interaction between SRK molecules or by competition for downstream effector molecules, and not by a specific dominance regulation mechanism.
|
Signal transduction and downstream events |
|---|
|
TopAbstractIntroductionA model for S-receptor...Variations on a theme?Molecular basis of dominanceSignal transduction and...The role of SLGPCPsSI and interspecific...References |
|---|
Signal transduction in Brassica SI is predicted to follow that of animal receptor kinases, where binding of the ligand causes kinase domain phosphorylation, leading to the recruitment of downstream effector molecules which pass on the signal (Schlessinger, 2000). It is likely that SRK is phosphorylated after binding SCR and that this phosphorylation is tightly regulated, as the effect of non-specific activation of the signalling pathway (pollen rejection) would lead to sterile plants. Detectable levels of in vivo SRK phosphorylation were observed 60 min after the addition of pollen in the S3 haplotype (Cabrillac et al., 2001). However, considering that by 60 min compatible pollinations are well advanced, with grains of some haplotypes already producing tubes (Dickinson, 1995), the ‘self’ rejection cascade must initiate before this detectable phosphorylation event.
The kinase activity of receptor kinases is highly regulated to avoid spontaneous activation of signalling pathways. In animals, a number of mechanisms such as inhibitors and phosphatases, have been well characterized to down-regulate kinase activity (Cock et al., 2002). The SRK kinase domain (Goring and Rothstein, 1992) and the complete SRK molecule (Giranton et al., 2000) can autophosphorylate in the absence of ligand in vitro whereas in vivo SRK requires the addition of pollen coat to autophosphorylate (Cabrillac et al., 2001) indicating that SRK must be held in an inactive state in the stigma. Two potential regulators of SRK kinase activity are the thioredoxin-h-like proteins (THL1 and THL2), which were found to interact with the kinase domain of SRK in a yeast two-hybrid screen (Bower et al., 1996). The autophosphorylation of SRK3 was found to be inhibited by the addition of both commercial thioredoxins and THL1 expressed in E. coli (Cabrillac et al., 2001). This thioredoxin-mediated inhibition of SRK3 kinase activity can be alleviated by the addition of pollen coat proteins in a haplotype-specific manner (Cabrillac et al., 2001). The evidence suggests that THL1 and THL2 interact with SRK3 in a reversible manner to prevent constitutive activation of the SI signalling cascade, whereas binding of SCR would lead to a conformational change causing de-repression of kinase activity. However, in another study, synthetic SCR8 (but not SCR9) was able to increase the phosphorylation level of SRK8 extracted from microsomal membranes in the absence of thioredoxin or other regulators of SRK activity (Takayama et al., 2001). It is possible that synthetic SCR has a higher kinase stimulating activity than crude pollen coat proteins; the higher level of phosphorylation would be detectable above the background level of SRK autophosphorylation, whereas crude coat proteins may not contain enough SCR to elicite such a large response.
Other potential regulators of SRK kinase activity include a kinase-associated protein phosphatase which binds to the kinase domain of at least one allele (Braun et al., 1997) and a recently isolated nucleoside diphosphate kinase, capable of phosphorylating SRK (Matsushita et al., 2002). Clearly, the phosphorylation state of SRK must be tightly controlled in stigmas to avoid spontaneous pollen rejection and it is probable that a number of processes are involved in this regulation. As mentioned earlier the decision to reject pollen takes place rapidly and perhaps only subtle changes in the phosphorylation status of SRK are needed to trigger rejection.
The only SRK interacting partner isolated to date with a proven role in SI is ARC1 (arm (Armadillo) repeat-containing protein 1). ARC1 was found in a yeast-2-hybrid screen using the kinase domain of SRK as bait (Gu et al., 1998). ARC1 has stigma-specific expression and binds to the SRK kinase domain is a phosphorylation-dependent manner (Gu et al., 1998). A conclusive role for ARC1 in SI was shown when transgenic plants expressing antisense ARC1 showed a partial breakdown in SI (Stone et al., 1999). This partial breakdown in SI can either be assigned to incomplete silencing (mRNA levels were reduced by up to 90%, but not totally ablated) or the action of another component of SI (Stone et al., 1999). In addition to ARM repeats (involved in protein–protein interactions), ARC1 also has a U box (Azevedo et al., 2001). The U-box is a domain recently discovered in a yeast protein (UFD2) which functions to transfer ubiquitin moieties to proteins targeted for proteasome degradation (Koegl et al., 1999). It is possible that ARC1 targets specific proteins for degradation through the U box, these targets could either be inhibitors of SI or stimulators of pollen growth, degradation of either would lead to pollen rejection. However ubiquitination does have roles other than protein degradation, including transcriptional regulation (Conaway et al., 2002), intercellular targeting of proteins and recruitment to molecular complexes (Kachroo et al., 2002).
With the exception of ARC1 the molecules downstream of the SRK/SCR interaction that lead to pollen rejection are unknown, and this presents a challenge for future research. The presence of ‘weak’ S alleles has shown that pollen rejection can take place at different stages; the strongest rejection response is the failure of self-pollen to hydrate and the decision to deny water must be taken quickly to account for the speed of this rejection response (Dickinson, 1995). Weaker rejection responses take place after the initiation of pollen tube growth, it is therefore not unreasonable to assume that SI can act at several levels and that the molecular mechanisms leading to rejection may vary between haplotypes.
|
The role of SLG |
|---|
|
TopAbstractIntroductionA model for S-receptor...Variations on a theme?Molecular basis of dominanceSignal transduction and...The role of SLGPCPsSI and interspecific...References |
|---|
The putative function of SLG appears to be in limbo at the moment. It is not required for the SI reaction (Takasaki et al., 2000) and revelations that self-incompatible A. lyrata (Kusaba et al., 2001) and at least three S haploypes of B. oleracea (Okazaki et al., 1999; Suzuki et al., 2000) do not produce SLG indicate that it is not involved in the SI systems of some haplotypes. However SLG28 enhances the SI response in transgenic S60 plants expressing SRK28, although no rejection-enhancing effects were seen when SLG910 was expressed in B. napus Westar plants expressing SRK910 (Silva et al., 2001).
In addition to being a possible component of a receptor complex, a number of other potential functions for SLG can been speculated upon; including post-transcriptional maturation of SRK, pollen adhesion and acting as a shuttle for SCR. In two self-compatible mutant lines of Brassica there was a correlation between depletion of SLG protein and lack of SRK protein, despite normal levels of SRK transcript, indicating that the breakdown of SI in these lines may be due to a lack of stable SRK (Dixit et al., 2000). In addition SRK6 produced in transgenic tobacco plants was found to migrate as very high molecular weight aggregates under non-reducing conditions, but not when co-expressed with SLG6 (Dixit et al., 2000). These data indicate that, in some haplotypes, SRK is an unstable molecule that requires SLG for proper folding or processing.
A role for SLG in pollen adhesion has been suggested based on anti-SLG antibodies having subtle effects on pollen adhesion (Luu et al., 1999). However, the S60 haplotype of B. oleracea (which produces no SLG protein) had no significant loss of pollen adhesion when compared to S2–b (which produces abundant SLG) (Suzuki et al., 2000). Pollen adhesion is a complex process which may also involve the lipidic component of pollen coat and physical interaction of the pollen exine with the stigma (Zinkl and Preuss, 2000). Other proteins such as SLR1 (for S-locus related protein 1) which is closely related to SLG (Luu et al., 1999) have also been implicated in pollen adhesion. The function of SLG as a molecular shuttle, bringing SCR and possibly other pollen coat proteins through the papillar cell wall to the membrane anchored SRK, may be hinted at by observations that SCR6 binds SLG in an allele specific manner. Although interactions between SCR and SLG are much weaker than that of SRK and SCR (Kachroo et al., 2001) given the large quantities of SLG present in stigmas, this interaction may be physiologically relevant. If SLG were to act as a shuttle it would follow that the binding would be weaker than that of the receptor to the ligand in order to allow efficient transfer.
The apparent multitude of SLG functions may be explained by the presence of several forms of SLG in stigmas. A membrane anchored form of SLG has been observed in S2 and S15 stigmas, in addition to the usual soluble form (Cabrillac et al., 1999; Tantikanjana et al., 1993). Different forms of SLG, as differentiated by different isoelectric points, have been known for some time (Nishio and Hinata, 1977). In addition, disulphide-linked SLG homodimers have been reported in S29 (Doughty et al., 1998), S3 (Giranton et al., 2000) S8, Sf1 (Dixit et al., 2000), and SLG6 produced in transgenic tobacco (Dixit et al., 2000). Whilst homodimers appeared as the dominant form of SLG in S29 stigmas (Doughty et al., 1998) they are much rarer in S3, S8 and Sf1 stigmas (Dixit et al., 2000; Giranton et al., 2000).
Distinctions have been made between soluble and membrane-associated (but not anchored) SLG. Membrane-associated SLG8 and SLGf1 can form oligomers in planta, however, the soluble form exists only as monomers (Dixit et al., 2000). By contrast, both soluble and membrane-associated SLG29 form dimers in planta (Doughty et al., 1998). Soluble SLG8 has little affinity for SCR8 whereas the form associated with SRK8 (and thus membrane associated) does appear to bind SCR8 (Takayama et al., 2001). However, both forms of SLG6 weakly bind SCR6 (Kachroo et al., 2001). It is possible that the membrane-associated SLG represents protein in transit through the secretory pathway (Stein et al., 1996) which would explain the different properties observed.
|
PCPs |
|---|
|
TopAbstractIntroductionA model for S-receptor...Variations on a theme?Molecular basis of dominanceSignal transduction and...The role of SLGPCPsSI and interspecific...References |
|---|
In consideration of the potential molecular complexity relating to pollen–stigma recognition in Brassica one must also take account of the PCP-A family (for Pollen Coat Protein–Class A) of pollen coat proteins (Doughty et al., 2000). These proteins are small (6–7 kDa), highly basic (pIs 8.5–10) and characterized by a motif of eight cysteine residues, all of which are likely to be involved in the formation of intramolecular disulphide bridges (Doughty et al., 1998, 2000). Significantly, the arrangement of cysteine residues in this group of proteins reveals that they are related to the SCRs (Schopfer et al., 1999; Takayama et al., 2000b). One member of this family, PCP-A1, has a general affinity for SLGs and there is some evidence that it may also form associations with SRK (Doughty et al., 1998). PCP-A1 is not encoded by the S-locus and appears to be present in all S haplotypes examined to date (Doughty et al., 1998; BP Kemp, J Doughty, unpublished observations).
Clearly, as discussed in previous sections, if SLG is present in the S-receptor complex then PCP-A1 is highly likely to be associated with it. What function PCP-A1 may serve is as yet unclear, but it may conceivably function along with SLG to stabilize the complex in vivo. The possibility that PCP-A1 also forms associations with SRK is intriguing and raises the question as to whether it may act as a regulator of SRK activity, perhaps competing for a binding site with SCR. It is apparent from previous studies (see previous sections) that binding of SCR to SRK is not dependent on the presence of other pollen coat proteins, therefore the proposition that PCP-A1 may perform a stabilizing or regulatory role in the S-receptor complex is an attractive one. Further research is clearly urgently needed to determine the identity of proteins that associate with SRK in both incompatible and compatible pollinations.
Another member of the PCP-A family of proteins, SLR-BP1 (for SLR1 binding protein 1; also known as PCP-A2, Doughty et al., 2000), has high affinity for the stigmatic protein SLR1 (Hiscock and Dickinson, 1993; Hiscock et al., 1998; McClure et al., 2000; Takayama et al., 2000a). One study has indicated that SLR1 probably functions in pollen adhesion along with SLG (Luu et al., 1999); in this respect interaction with SLR-BP1 and PCP-A1 may be highly significant. Data such as this further complicates the building of models relating to the question, ‘what constitutes the S-receptor complex?’.
How can SLG function both at the level of pollen adhesion and activation of SI? It is generally held that SI systems are likely to have evolved from mechanisms that were already present in the plant, but fulfilling other roles. In this context the Brassica SI system could have evolved from families of stigmatic molecules (S-domain-like proteins) and pollen coat proteins (PCP-A class and SCR-like proteins) that together mediated compatible pollen–stigma interactions. It is not inconceivable that the control of pollen adhesion would have been central to such early evolutionary adaptions and may explain why a potential cross-over is seen between the two mechanisms today.
|
SI and interspecific incompatibility |
|---|
|
TopAbstractIntroductionA model for S-receptor...Variations on a theme?Molecular basis of dominanceSignal transduction and...The role of SLGPCPsSI and interspecific...References |
|---|
A link is now well established between interspecific incompatibility and SI (Hiscock and Dickinson, 1993; Hiscock et al., 1998; Lewis and Crowe, 1958; McClure et al., 2000; Murfett et al., 1996; Sampson, 1962). This is perhaps most clearly illustrated in reciprocal crosses between closely related SI and SC (self-compatible) species. Typically, such crosses display unilateral incompatibility (UI), that is the stigmas of the SI partner are able to reject pollen from the SC partner whereas the reciprocal cross is usually successful. UI has been extensively documented in the Brassicaceae (Hiscock and Dickinson, 1993; Sampson, 1962) and is widespread amongst the angiosperms (Lewis and Crowe, 1958).
Thus it is clear that the stigmas of SI species present a barrier to ‘foreign’ pollen that is absent from the stigmas of self-compatible species. This tight correlation between UI and the possession of SI indicates that the molecular mechanisms of interspecific and intraspecific pollen rejection may overlap and points to an involvement of the S-locus or closely related factors (Hiscock and Dickinson, 1993; Sampson, 1962).
Direct evidence for a role of S-locus products in interspecific incompatibility has come from plant transformation experiments in Nicotiana sp. where involvement of the stigmatic S-RNase has been demonstrated (McClure et al., 2000; McCubbin and Kao, 1996). Interestingly, both S-RNase-dependent and S-RNase-independent mechanisms of interspecific pollen rejection were identified, highlighting the complexity of interspecific pollination relationships.
Whether products of the S-locus are involved in Brassica is as yet far from clear, however, the means by which ‘incompatible’ interspecific pollen is rejected is nearly indistinguishable from that of ‘self’ pollen. For instance, both recognition/rejection systems are sited in the stigmatic papilla cells, with pollen being rejected at the stigma surface (Hiscock and Dickinson, 1993). Further, both systems are acquired as the stigmatic papilla cell matures, i.e. immature stigmas do not present a barrier to either ‘self’ pollen or interspecies pollen and both systems require continued protein synthesis, as demonstrated by cycloheximide treatment of stigmas (Hiscock and Dickinson, 1993; Hiscock et al., 1998; Sarker et al., 1988).
Although it is clear that a strong correlation exists between the occurrence of SI and UI amongst members of the Brassicaceae, suggestive of an involvement of the S-locus, there is currently no evidence to suggest that either SRK or SLG play an active role in this process. One study demonstrated that the stigmas of an SC line of Brassica oleracea, which lacked a functional SRK, were still capable of rejecting pollen from SC Arabidopsis thaliana, indicating that UI was still intact (Kandasamy et al., 1994). Further, SI Arabidopsis lyrata, which has the same SI system as that found in Brassica, appears to lack SLG but displays UI with A. thaliana (Fig. 1). Until more detailed molecular analyses are carried out coupled with extensive interspecific crossing studies, a role for the S-receptor complex cannot be dismissed. If stigmatic products of the S-locus are found not to be directly involved in UI in the Brassicaceae, then how can the correlation between the possession of SI and UI be explained? One explanation could be that signalling components of the SI system downstream of SRK (that ultimately bring about pollen rejection) are shared with a ‘receptor’ molecule (unlinked to the S-locus) that ‘detects’ interspecific pollen from SC species. As has been seen, the mechanism by which interspecific pollen is rejected in Brassica appears highly similar to that by which ‘self’ pollen is rejected, thus adding weight to this hypothesis. Although the downstream components are not genetically linked to S, they are mechanistically inseparable; therefore one might expect that if the primary determinants of SI were ever lost then selective pressure to maintain these downstream factors over evolutionary time would be severely diminished.
|
Although such a scenario seems logical, work recently published by the Nasrallah group has indicated that fully functional SI can be regained in A. thaliana simply through the introduction of SRK and SCR from A. lyrata (Nasrallah et al., 2002). This elegant experiment clearly demonstrates that A. thaliana, which lacks functional orthologues of these genes, has retained the downstream signalling and effector molecules required for pollen rejection. Maintenance of this pathway in A. thaliana, following mutation of SRK and SCR, suggests that it may perform essential functions unrelated to SI. In this respect it is possible that interspecific function (which required this pathway) was retained following mutation of SRK/SCR and has only been lost comparatively recently, leaving the bulk of the pathway intact. Further, in order to clarify the potential role of SRK in interspecific incompatibility it will be important to determine whether this A. thaliana line has acquired the ability to reject a range of interspecific pollen.
As unravelling the complexities of the S-receptor complex and the mechanism of pollen rejection in intraspecific pollination get closer, it is likely that such data will greatly facilitate future studies directed at elucidating the perhaps even more complex story of interspecific pollen rejection.
|
Acknowledgements |
|---|
The authors would like to thank Hugh Dickinson and Simon Hiscock for helpful discussions, Hannah Gomes for her work on interspecific incompatibility, and the UK BBSRC for financial support.
|
References |
|---|
|
TopAbstractIntroductionA model for S-receptor...Variations on a theme?Molecular basis of dominanceSignal transduction and...The role of SLGPCPsSI and interspecific...References |
|---|
Ai YJ, Singh A, Coleman CE, Ioerger TR, Kheyrpour A, Kao TH. 1990. Self-incompatibility in Petunia inflata—isolation and characterization of cDNAs encoding 3 S-allele-associated proteins. Sexual Plant Reproduction 3, 130–138.
Azevedo C, Santos-Rosa MJ, Shirasu K. 2001. The U-box protein family in plants. Trends in Plant Science 6, 354–358.[CrossRef][ISI][Medline]
Bower MS, Matias DD, Fernandes-Carvalho E, Mazzurco M, Gu T, Rothstein SJ, Goring DR. 1996. Two members of the thioredoxin-h family interact with the kinase domain of a Brassica S locus receptor kinase. The Plant Cell 8, 1641–50.[Abstract]
Braun DM, Stone JM, Walker JC. 1997. Interaction of the maize and Arabidopsis kinase interaction domains with a subset of receptor-like protein kinases: Implications for transmembrane signaling in plants. The Plant Journal 12, 83–95.[CrossRef][ISI][Medline]
Brewbaker JL. 1959. Biology of the angiosperm pollen grain. Indian Journal of Genetics and Plant Breeding 19, 121–133.
Cabrillac D, Cock JM, Dumas C, Gaude T. 2001. The S-locus receptor kinase is inhibited by thioredoxins and activated by pollen coat proteins. Nature 410, 220–223.[CrossRef][Medline]
Cabrillac D, Delorme V, Garin J, Ruffio-Chable V, Giranton JL, Dumas C, Gaude T, Cock JM. 1999. The S15 self-incompatibility haplotype in Brassica oleracea includes three S gene family members expressed in stigmas. The Plant Cell 11, 971–86.
Charlesworth D. 1995. Multi-allelic self-incompatibility polymorphisms in plants. Bioessays 17, 31–38.
Cock JM, Vanoosthuyse V, Gaude T. 2002. Receptor kinase signalling in plants and animals: distinct molecular systems with mechanistic similarities. Current Opinion in Cell Biology 14, 230–236.[CrossRef][ISI][Medline]
Conaway RC, Brower CS, Conaway JW. 2002. Gene expression—emerging roles of ubiquitin in transcription regulation. Science 296, 1254–1258.
Conner JA, Tantikanjana T, Stein JC, Kandasamy MK, Nasrallah JB, Nasrallah ME. 1997. Transgene-induced silencing of S-locus genes and related genes in Brassica. The Plant Journal 11, 809–823.[CrossRef][ISI]
Darlington CD, Mather K. 1949. The elements of genetics. London, UK: Allen and Unwin Ltd.
DeYoung BJ, Clark SE. 2001. Signaling through the CLAVATA1 receptor complex. Plant Molecular Biology 46, 505–513.[CrossRef][ISI][Medline]
Dickinson H. 1995. Dry stigmas, water and self-incompatibility in Brassica. Sexual Plant Reproduction 8, 1–10.[ISI]
Dickinson HG, Doughty J, Elleman CJ, Hiscock SJ. 1998. Evolution of angiosperm self-incompatibility mechanisms; data from Brassica oleracea. In: Bennett M, Owens SJ, eds. Reproductive biology in systematics, conservation and economic botany. Kew Symposium Series. London: Academic Press, 15–29.
Dixit R, Nasrallah ME, Nasrallah JB. 2000. Post-transcriptional maturation of the S receptor kinase of Brassica correlates with co-expression of the S-locus glycoprotein in the stigmas of two Brassica strains and in transgenic tobacco plants. Plant Physiology 124, 297–311.
Doughty J, Dixon S, Hiscock SJ, Willis AC, Parkin IA, Dickinson HG. 1998. PCP-A1, a defensin-like Brassica pollen coat protein that binds the S-locus glycoprotein, is the product of gametophytic gene expression. The Plant Cell 10, 1333–1347.
Doughty J, Wong HY, Dickinson HG. 2000. Cysteine-rich pollen coat proteins (PCPs) and their interactions with stigmatic S (incompatibility) and S-related proteins in Brassica: putative roles in SI and pollination. Annals of Botany 85, 161–169.
Franklin-Tong VE. 2002. The difficult question of sex: the mating game. Current Opinion in Plant Biology 5, 14–18.[CrossRef][ISI][Medline]
Giranton JL, Dumas C, Cock JM, Gaude T. 2000. The integral membrane S-locus receptor kinase of Brassica has serine/threonine kinase activity in a membranous environment and spontaneously forms oligomers in planta. Proceedings of the National Academy of Sciences, USA 97, 3759–3764.
Goring DR, Rothstein SJ. 1992. The S-locus receptor kinase gene in a self-incompatible Brassica napus line encodes a functional serine/threonine kinase. The Plant Cell 4, 1273–81.
Gu T, Mazzurco M, Sulaman W, Matias DD, Goring DR. 1998. Binding of an arm repeat protein to the kinase domain of the S-locus receptor kinase. Proceedings of the National Academy of Sciences, USA 95, 382–387.
Hatakeyama K, Takasaki T, Suzuki G, Nishio T, Watanabe M, Isogai A, Hinata K. 2001. The S receptor kinase gene determines dominance relationships in stigma expression of self-incompatibility in Brassica. The Plant Journal 26, 69–76.[CrossRef][ISI][Medline]
Hiscock SJ, Dickinson HG. 1993. Unilateral incompatibility within the Brassicaceae: further evidence for the involvement of the self-incompatibility (S)-locus. Theoretical and Applied Genetics 86, 744–753.
Hiscock SJ, Doughty J, Dickinson H. 1998. Unilateral interspecific incompatibility and the S (self-incompatibility) locus. In: Owens SJ, Bennett M, eds. Reproductive biology in systematics, conservation and economic botany. Kew Symposium Series. London: Academic Press, 31–46.
Hiscock SJ, Kues U. 1999. Cellular and molecular mechanisms of sexual incompatibility in plants and fungi. International Review of Cytology 193, 165–295.[ISI][Medline]
Jordan ND, Franklin FCH, Franklin-Tong VE. 2000a. Evidence for DNA fragmentation triggered in the self- incompatibility response in pollen of Papaver rhoeas. The Plant Journal 23, 471–479.[CrossRef][ISI][Medline]
Jordan ND, Ride JP, Rudd JJ, Davies EM, Franklin-Tong VE, Franklin FCH. 2000b. Inhibition of self-incompatible pollen in Papaver rhoeas involves a complex series of cellular events. Annals of Botany 85, 197–2002.
Kachroo A, Nasrallah ME, Nasrallah JB. 2002. Self-incompatibility in the Brassicaceae: receptor-ligand signalling and cell-to-cell communication. The Plant Cell 14, 227–238.
Kachroo A, Schopfer CR, Nasrallah ME, Nasrallah JB. 2001. Allele-specific receptor–ligand interactions in Brassica self-incompatibility. Science 293, 1824–1826.
Kandasamy MK, Nasrallah JB, Nasrallah ME. 1994. Pollen–pistil interactions and developmental regulation of pollen-tube growth in Arabidopsis. Development 120, 3405–3418.[Abstract]
Kao TH, McCubbin AG. 1996. How flowering plants discriminate between self and non-self-pollen to prevent inbreeding. Proceedings of the National Academy of Sciences, USA 93, 12059–12065.
Koegl M, Hoppe T, Schlenker S, Ulrich HD, Mayer TU, Jentsch S. 1999. A novel ubiquitination factor, E4, is involved in multiubiquitin chain assembly. Cell 96, 635–644.[CrossRef][ISI][Medline]
Kusaba M, Dwyer K, Hendershot J, Vrebalov J, Nasrallah JB, Nasrallah ME. 2001. Self-incompatibility in the genus Arabidopsis: characterization of the S locus in the outcrossing A. lyrata and its autogamous relative A. thaliana. The Plant Cell 13, 627–643.
Kusaba M, Tung CW, Nasrallah ME, Nasrallah JB. 2002. Monoallelic expression and dominance interactions in anthers of self-incompatible Arabidopsis lyrata. Plant Physiology 128, 17–20.
Letham DLD, Blissard GW, Nasrallah JB. 1999. Production and characterization of the Brassica oleracea self-incompatibility locus glycoprotein and receptor kinase in a baculovirus infected insect cell culture system. Sexual Plant Reproduction 12, 179–187.[CrossRef]
Lewis D, Crowe LK. 1958. Unilateral interspecific incompatibility in flowering plants. Heredity 12, 233–256.[Medline]
Luu DT, Marty-Mazars D, Trick M, Dumas C, Heizmann P. 1999. Pollen–stigma adhesion in Brassica spp. involves SLG and SLR1 glycoproteins. The Plant Cell 11, 251–62.
Matsushita Y, Suzuki T, Kubota R, Mori M, Shimosato H, Watanabe M, Kayano T, Nishio T, Nyunoya H. 2002. Isolation of a cDNA for a nucleoside diphosphate kinase capable of phosphorylating the kinase domain of the self- incompatibility factor SRK of Brassica campestris. Journal of Experimental Botany 53, 765–767.
McClure BA, Cruz-Garcia F, Beecher B, Sulaman W. 2000. Factors affecting inter- and intraspecific pollen rejection in Nicotiana. Annals of Botany 85, 113–123.
McClure BA, Gray JE, Anderson MA, Clarke AE. 1990. Self-incompatibility in Nicotiana alata involves degradation of pollen ribosomal-RNA. Nature 347, 757–760.[CrossRef]
McClure BA, Haring V, Ebert PR, Anderson MA, Simpson RJ, Sakiyama F, Clarke AE. 1989. Style self-incompatibility gene products of Nicotiana alata are ribonucleases. Nature 342, 955–957.[CrossRef][Medline]
McCubbin AG, Kao TT. 1996. Molecular mechanisms of self-incompatibility. Current Opinion in Biotechnology 7, 150–154.
Murfett J, Strabala TJ, Zurek DM, Mou BQ, Beecher B, McClure BA. 1996. S RNase and interspecific pollen rejection in the genus Nicotiana: multiple pollen-rejection pathways contribute to unilateral incompatibility between self-incompatible and self-compatible species. The Plant Cell 8, 943–958.[Abstract]
Nasrallah J, Kao T, Goldberg M, Nasrallah M. 1985. A cDNA clone encoding an S-locus-specific glycoprotein from Brassica oleracea. Nature 5, 373–384.
Nasrallah JB, Nishio T, Nasrallah ME. 1991. The self-incompatibility genes of Brassica: expression and use in genetic ablation of floral tissues. Annual Review of Plant Physiology and Plant Molecular Biology 42, 393–422.[CrossRef][ISI]