|Home | About | Journals | Submit | Contact Us | Français|
Self-incompatibility (SI) has emerged as an evolutionary strategy to enhance the genetic variability of plant species. In Brassica, it is controlled by a single multiallelic locus, the S-locus, encoding a receptor kinase (SRK) expressed in the stigma papilla cells and its ligand, a small protein (SCR) located in the pollen coat. Pollen rejection is achieved only when the receptor recognizes SCR coming from the same S-allele. If a single papilla cell is simultaneously pollinated by a self- and a cross-pollen grain, it is capable of distinguishing between the two and responding accordingly, rejecting self while accepting cross pollen. This phenomenon reveals that SI response is strictly localized and does not involve the whole papilla cell. It also suggests that the distribution of SRK inside the cell may play an important role in regulating this dual response. We have recently demonstrated that SRK is mostly intracellular, only small amounts being present in distinct domains of the plasma membrane (PM), where interaction with SCR occurs. Following ligand recognition, the receptor-ligand complex is endocytosed and degraded. Based on this, we propose a model of the significance of SRK intracellular trafficking for the functioning and specificity of SI response.
Several recent studies have demonstrated the active trafficking of plant receptor kinases1–4 and its importance in the regulation of receptor functions. In the case of the self-incompatibility response in the Brassicaceae, among the several known SRK interactors is a sorting nexin, SNX1.5 SNX1 is involved in endosome functions in plants.6,7 This suggested that vesicular trafficking is a factor in regulating SI response and prompted us to investigate the intracellular distribution of SRK by fluorescent immunolocalization on Brassica oleracea (S3 haplotype) papilla cells. We found that the majority of SRK3 is concentrated in sorting endosomes, with less SRK3 in the Golgi apparatus, the trans-Golgi network, endoplasmic reticulum (ER) and very low amounts at the PM. Despite the low and patchy distribution of SRK3 at the PM, using an anti-SRK3 antibody as a mimetic ligand for SCR3, we were able to demonstrate that interaction between SRK and its ligand occurs at the PM.
Immunolocalization on electron microscopy sections confirmed the localization pattern and revealed presence of SRK3 in small vesicles in the vicinity of the PM, suggesting that the pattern observed is a result of a dynamic process, rather than a static distribution of SRK. As the anti-SRK3 antibody can recognize both the full-length receptor and the different glycosylated forms of its splice variant (known as eSRK), it was important to understand the nature of the antigen at the PM. We performed PM purification from 4,000 Brassica oleracea stigmas by aqueous two-phase partitioning8 and we verified the presence of full-length SRK3 by western blot detection (Fig. 1). The PM H+-ATPase9 was used as a control for the PM-enriched fraction, while probing against the ER protein BIP10 and α tubulin was done to verify the purity of the PM fraction. Signal for BiP was seen in the microsomal and also surprisingly in the soluble fraction. α tubulin, which is a component of the cytoskeleton and can also be associated with intracellular compartments, was observed in the soluble and faintly in the microsomal fraction. Importantly, no BiP or tubulin signal was present in the PM-enriched fraction, where H+-ATPase and SRK3, as well as the 62.8 kDa eSRK3 form could be found.
In a previous study, it has been demonstrated that eSRK alone is not able to bind the ligand due to the fact that it cannot form homodimers.11 Therefore, the full-length SRK is the factor that is responsible for ligand recognition at the PM, with eSRK probably acting as a coreceptor.12
The peculiar patchy distribution of SRK3 at the PM is understandable in the light of the observation that SRK3 can auto-activate and cross phosphorylate in high concentrations.13,14 In papilla cells, maintenance of low amounts of SRK may be required to avoid unspecific autoactivation. We propose that SRK at the PM is in an uninhibited, ready-to-be-activated state and may or may not be in complex with eSRK. The PM localized SRK is not randomly distributed but is grouped into receptor-rich regions, which we designate Self-Incompatibility domains (SI domains, Fig. 2). Such domains can operate as independent units and are in constant contact with the underlying endosomes, evidenced by the small SRK3-containing vesicles close to the PM. The endosomes function as a regulatory and exchange center and the high SRK amounts there are kept in an unactivated state by the inhibiting action of the thioredoxin THL1.14–16
If self-pollination occurs, the interaction between SRK and the ligand SCR will trigger the activation (phosphorylation) of the receptor complex. This will lead to a chain reaction spreading the activation to the neigboring SRK molecules but due to the specific PM SRK distribution, the effect will finally be contained within the borders of the affected SI domain. Thus, the majority of the PM stays unstimulated and ready to respond to independent pollination events. The activated receptorligand complexes, which probably include other components required for the SI response, are then internalized to the sorting endosomes and then degraded, allowing the “resetting” of the SI domain by the arrival of naive receptor molecules from the endosome.
At the moment, there is no solid evidence where the actual receptor signaling takes place, but our model favors the PM as the place where SRK triggers the SI response. Indirect evidence for the PM signaling is the internalization of the activated complexes to the endosome, where they fall under the influence of the negative regulator THL1. It cannot be excluded though, that SRK signaling may be a multi-step process, where the first step is the marking of the contact position at the PM, followed by activating additional signaling components at the endosome. Like this, the effectors of SI can be recruited anywhere in the cell and targeted to the marked PM region.
Previously published online: www.landesbioscience.com/journals/psb/article/9714