Self-Rejection--a New Kinase Connection

See allHide authors and affiliations

Science  05 Mar 2004:
Vol. 303, Issue 5663, pp. 1474-1475
DOI: 10.1126/science.1095764

Self-incompatibility is a strategy adopted by many flowering plants to reject “self” pollen, thus promoting mating between unrelated individuals. When the pollen and stigma recognize each other as being related, pollen-tube growth is inhibited and fertilization is prevented. Extensive research has unraveled the molecular and cellular pathways of pollen rejection in plants that exhibit self-incompatibility. In Brassica, a member of the cabbage family, self-incompatibility depends on recognition between an S receptor kinase (SRK) expressed by stigma cells and its ligand (SP11; also called SCR) present on the surface of pollen grains (1). This initial recognition step activates an SRK-mediated signaling pathway in the pistil that results in rejection of self pollen (see the figure). The only component identified in this pathway so far is ARC1, an E3 ubiquitin ligase (2, 3). Enter Murase et al. (4) on page 1516 of this issue with their report of a new player—the M locus protein kinase (MLPK)—in the self-incompatibility signaling pathway of Brassica. Their discovery raises an intriguing new aspect of plant signaling in which a receptor kinase may work with a nonreceptor kinase to activate the signaling pathway.

The self-incompatibility response in Brassica.

In the absence of its allele-specific SP11/SCR ligand, SRK is inhibited by thioredoxin h (Th). After pollination with self-incompatible pollen, the allele-specific SP11/SCR ligand binds to SRK, resulting in recruitment of MLPK to the activated complex. Activation and phosphorylation of the SRK and MLPK domains lead to the activation of one or more intracellular signaling pathways. The signaling component ARC1, an E3 ubiquitin ligase, may be recruited to the kinase complex and activated. ARC1 then targets unknown substrates for ubiquitination (Ub) and degradation by the proteasome. The degradation of these substrates could cause pollen rejection and prevent fertilization (14).


The recessive modifier (m) locus in the Brassica rapa variety Yellow Sarson causes a complete breakdown of the self-incompatibility response in m/m plants (5). However, the m locus segregates independently of the self-incompatibility S locus (containing the tightly linked SP11/SCR and SRK genes) (5). Originally, the m locus was thought to encode an aquaporin-like protein, MIP-MOD, but this turned out not to be the case (6, 7). Using map-based cloning, Murase et al. identified MLPK as the m gene candidate and demonstrated that mutations in this gene are responsible for the modifier phenotype. They identified a mutation in the MLPK gene that results in a loss of kinase activity in vitro as well as a complete absence of the protein in m/m plants. Using an innovative transient expression assay in stigmatic papillae (the receptive pistil surface for pollen grains), the authors demonstrated that expression of the wild-type MLPK gene restored the rejection of self pollen in the m/m plants.

MLPK belongs to a group of protein kinases called the receptor-like cytoplasmic kinases (RLCKs). In the model plant Arabidopsis, these kinases cluster phylogenetically with receptor-like kinases (RLKs) but do not contain an extracellular domain or transmembrane domain (8). Very little is known about the RLCKs, and MLPK is the first member of this group to be linked to a receptor kinase signaling pathway.

Where does MLPK fit into the SRK signaling pathway? MLPK is enriched in the plasma membrane fraction derived from pistil cells. When transiently expressed in tobacco cells, it has a predicted myristoylation site required for plasma membrane localization. The complete breakdown of self-incompatibility in m/m Brassica plants indicates that MLPK is required for all SRK-mediated signaling pathways (assuming that there are multiple pathways) leading to pollen rejection. This raises the possibility that MLPK forms a signaling complex with SRK to mediate the rejection response. Typically, phosphorylation of these signaling complexes can lead to activation of the kinase partner and establishment of binding sites for downstream signaling components. In this model, ARC1 may operate downstream of the SRK-MLPK complex (see the figure). Suppression of ARC1 activity with the expression of an antisense cDNA causes a partial breakdown in the self-incompatibility response (2). The incomplete breakdown of this response implies that there may be a branch in the signaling pathway after the SRK-MLPK complex. However, it is also very likely that the incomplete breakdown phenotype is due to partial RNA suppression in the antisense transgenic plants or to some functional redundancy with other related E3 ubiquitin ligases. Interestingly, the introduction of related Arabidopsis lyrata SP11/SCR and SRK genes into A. thaliana elicits a self-pollen rejection response for a brief period after flowering (9). A. thaliana has an MLPK ortholog that is closely related to the Brassica MLPK (4), and has more distantly related ARC1 orthologs (10) that may participate in SRK-mediated signaling.

The notion of plant receptor kinases working with nonreceptor kinases in heteromeric complexes has emerged in other systems. The kinases in the heteromeric complexes can both be receptor kinases. For example, the BRI1 leucine-rich repeat receptor kinase detects the plant hormone brassinosteroid (11). Recently, an activation-tagging screen (12) and a yeast two-hybrid analysis (13) in Arabidopsis identified the receptor kinase BAK1 as a BRI1-interacting protein. Genetic and biochemical analyses suggest that the formation of BRI1-BAK1 heterodimers is required for brassinosteroid signaling (12, 13). Another example is provided by the NFR1 and NFR5 LysM receptor kinases identified in a screen for legume mutants with defects in rhizobial symbiosis and nodulation (14). To recognize Nod factors, NFR1 is proposed to form a heterodimer with NFR5, which is missing the activation loop in the kinase domain (14).

The Murase et al. work presents a new scenario in which receptor kinases cooperate with plasma membrane-localized RLCKs. Perhaps this will become a common theme in plant receptor kinase signaling and certainly would account for the large number of RLCKs and RLKs found in plant genomes (8). Future studies will need to address the precise biochemical and cellular relationships among SRK, MLPK, and ARC1 in the plant self-incompatibility response.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
View Abstract

Stay Connected to Science

Navigate This Article