Biochemical investigations into the regulation and specificity of SERK co-receptors

Research output: Thesisinternal PhD, WU


The transduction of extracellular signals to intracellular responses is of vital importance for all living organisms. The perception of signalling molecules outside the cell is often accomplished via ligand binding proteins spanning the plasma membrane, i.e. receptor proteins. In plants, the majority of plasma membrane receptors are part of the family of leucine-rich-repeat receptor-like-kinases (LRR-RLKs) (Shiu and Bleecker 2001). These receptors contain an extracellular LRR domain, coupled to an intracellular kinase domain via a single pass transmembrane helix. The plant RLKs and their animal counterparts, the receptor-tyrosine-kinases (RTKs), are thought to form dimers or oligomers to make activation of their cytoplasmic kinase domains possible. Sometimes, additional membrane spanning proteins are part of these higher order complexes. If so, we can distinguish between the main ligand binding receptor, which contains both the domains to perceive signal and to confer this across the membrane to intracellular reactions and co-receptors (or auxiliary receptors), which on their own cannot perceive the signal or transduce the signal into the cell. The somatic embryogenesis receptor kinases ( SERKs ) are a family of co-receptors found in plants and show multitasking functions in several, seemingly unrelated, pathways (Albrecht et al. 2005, Albrecht et al. 2008, Chinchilla et al. 2009, Roux et al. 2011). Except for SERK5, that has no reported function at this time, the SERK genes in Arabidopsis thaliana show partial redundancy, with each SERK functioning in its own specific subset of signalling pathways.

In this thesis I have taken a closer look at the SERK family of receptor kinases to understand their role as co-receptors in plant signalling cascades and the regulation of their activity. Studies on animal signalling indicate that co-receptors can have many different functions in signalling complexes and form an additional layer of regulation in signalling cascades, which I have discussed in more detail in the introductory Chapter 1. Also in this chapter, I have reviewed the current crystal structures available on plant leucine-rich-repeat receptor-like-kinases (Santiago et al. 2013, Sun et al. 2013). These structures provide us with the first clues into the molecular basis of plant RLK function. Interestingly, the current crystal structures of SERK extracellular domains indicate that these co-receptors have interactions not only with the main ligand binding receptor, but also with the different ligands. The SERK residues that are implicated in ligand interaction are conserved in the SERK protein family, so this new evidence does not seem to be able to explain the functional divergence observed between the SERK family members. Current structural, biochemical and genetic evidence define the A. thaliana SERK proteins as true co-receptors, part of active signalling complexes via interaction with different main ligand binding receptors (Albrecht et al. 2012, Gou et al. 2012, Santiago et al. 2013, Sun et al. 2013).

The wide functional diversity seen between different members of the SERK protein family must be inherent to differences in protein sequence. In some ways, this is rather counter-intuitive, because the SERK proteins are very homologous and show only minor changes. In Chapter 2, I pinpoint which residues/motifs or domains are important for the different functions of the SERKs. A sequence comparison with SERK sequences from different plant species showed that SERK genes are ancient, present already with multiple copies in genomes of non-vascular plants (e.g. the moss Physcomitrella patens). The SERKs present in these evolutionary older genomes closely resemble the A. thaliana SERK1 and 2 sequences; apparently the SERKs have not changed much throughout 2 M years of evolution. This is a strong indication that SERKs are indeed essential for important plant signalling pathways. The A. thaliana SERK3 and 4 genes, however, have diverged more in protein sequence and belong to a SERK-clade only present in dicotyledons. This divergence is probably linked to different functionality of these proteins. Chimeric SERK proteins, in which the extracellular domain of one SERK is coupled to the cytoplasmic domain of another, show that SERK specificity is rather complex and cannot be attributed to any one residue, motif or even domain.

To further investigate the functional plasticity of SERKs, we aimed to obtain additional high-resolution structural data on these co-receptors. In recent years, the first crystal structures of plant LRR-RLK domains have been reported. These include the kinase domain of SERK3/ BAK1, the extracellular domain of SERK1 in complex with that of the main receptor BRI1 (Brassinosteroid Insensitive 1) and of SERK3 with the main receptor FLS2 (Flagellin Sensing 2). These structures have given important clues into the function of SERK co-receptors. High-resolution structural data on other SERKs could provide insights into the functional plasticity and specificity observed in the SERK protein family. In Chapter 3 I describe our attempts to solve the crystal structures of the kinase domains of different A. thaliana SERK proteins. This was hampered by low protein production and difficulties in purification. We managed to obtain two crystals of the SERK1 kinase domain, which diffracted to around 2.5 Å resolution. However, structure elucidation was not possible.

Besides structural information, biochemical investigations into receptor activation mechanisms are essential and in Chapter 4I used a combination of spectroscopic data and activity assays to understand more of the regulation of activity of the SERK kinase domain. Co-receptors can modulate signalling activity, but only if their activity is also tightly regulated to avoid aberrant activation of the entire receptor complex. In this thesis, and in other reports, it has been shown that kinase domains of receptors and co-receptors are intrinsically active when produced in vitro. This means that mechanisms need to be in place in vivo to keep kinase domains in their inactive state. In Chapter 4we investigated one such way of keeping proteins inactive, namely intrinsic disorder. In this manner, proteins are kept disordered, or unfolded/unstructured, until their activity is needed. Folding, or structuring, can subsequently be induced when the protein is at its proper location or in the proper complex conformation. Our results show that the SERK1 kinase domain has the ability to refold itself to an active kinase domain in vitro, thus in the absence of any cellular components. About 80% of the kinase activity is regained after only 30 minutes of refolding. We conclude that the SERK1 kinase is not only intrinsically active, but folds efficiently to its active state. Therefore it seems unlikely that intrinsic disorder is a manner in which the phosphorylation activity of the kinase domain is regulated.

Protein function is closely coupled to its location in the cell. SERKs, as RLKs, are located at the plasma membrane (PM), which seems to be the major site for signalling. The PM is not homogenous and many proteins do not display a free diffusion through this lipid bilayer. In Chapter 5 the distribution of the SERK3/ BAK1 co-receptor and BRI1 main receptor on the plasma membrane is shown, using Variable Angle Epifluorecence Microscopy (VAEM), a technique that makes use of the difference in refractive index between the sample and cover slide in order to create an evanescence wave of excitation light. This evanescence wave only penetrates the sample to a certain depth (depending on the angle of the laser) and thus excites only those fluorophores in close proximity to the cover slide. The technique makes it possible to visualise fluorescent proteins with high resolution and low background signal in the plane of the plasma membrane or just below. In this chapter we show that this technique can be used to visualise the PM of A. thalianalive epidermal root cells, and several intracellular membrane compartments in the cytoplasm just below the PM. In addition, VAEM was used to show that SERK3/ BAK1 and BRI1 are not homogenously distributed across the PM, but in fact are present in nanoclusters. These nanoclusters are not influenced by addition or depletion of ligand, and BRI1 is still present in clusters in the absence of SERK1 and SERK3/ BAK1. Overexpression of BRI1, or diminishing the number of PM localised BRI1 receptors by tyrphostin A23 treatment, influenced the number of receptors per cluster, but not the cluster density. We propose that the nanocluster distribution in the PM is a general feature of plant PM receptors.


In summary, this thesis provides new insights in the role of SERKs as co-receptors, their origin, regulation and localisation. A combination of different techniques, biochemical, molecular, bioinformatics and structural, was used to understand how SERKs function as co-receptors in plant signalling complexes.

Original languageEnglish
QualificationDoctor of Philosophy
Awarding Institution
  • Wageningen University
  • de Vries, Sacco, Promotor
  • Boer, D.R., Co-promotor, External person
  • van Mierlo, Carlo, Co-promotor
Award date4 Jun 2014
Place of PublicationWageningen
Print ISBNs9789461739346
Publication statusPublished - 4 Jun 2014


  • kinases
  • biochemical receptors
  • chemical structure


Dive into the research topics of 'Biochemical investigations into the regulation and specificity of SERK co-receptors'. Together they form a unique fingerprint.

Cite this