BIOCHEMICAL AND FUNCTIONAL STUDIES OF S-RNASE-BASED SELF-INCOMPATIBILITY IN PETUNIA INFLATA

Open Access
Author:
Hua, Zhihua
Graduate Program:
Plant Biology
Degree:
Doctor of Philosophy
Document Type:
Dissertation
Date of Defense:
February 20, 2008
Committee Members:
  • Teh Hui Kao, Committee Chair
  • Hong Ma, Committee Member
  • Richard Cyr, Committee Member
  • David Braun, Committee Member
Keywords:
  • PiSLF
  • S-RNase
  • Self-Incompatibility
  • Petunia inflata
Abstract:
Self-incompatibility (SI) allows the female reproductive tissue, pistil, to distinguish between self- and non-self pollen during sexual reproduction in flowering plants. In simplest cases, this self/non-self recognition is controlled by a highly polymorphic locus, named the S-locus. If the S-haplotype of pollen is also carried by the pistil, the pollen is recognized by the pistil as self-pollen and rejected. If the S-haplotype of pollen is different from both S-haplotypes carried by the pistil, the pollen is recognized as non-self pollen and its tube is allowed to grow through the pistil to effect fertilization. Thus, SI allows flowering plants to avoid inbreeding and generate genetic diversity in the offspring. My thesis research focuses on the type of SI that has so far been found in the Solanaceae and two other families, and I have used Petunia inflata, a wild species of petunia, as a model. The Kao Lab has identified the S-RNase gene and the PiSLF (P. inflata S-locus F-box) gene as the genes that control pistil and pollen SI function, respectively. The overall goal of my thesis research is to study how S-RNase and PiSLF interact inside a pollen tube to result in specific growth inhibition of self-pollen tubes. There are several clues to the biological functions of S-RNase and PiSLF. First, the RNase activity of S-RNase is essential for its function in growth inhibition of self-pollen tubes. Second, PiSLF is a member of a large family of proteins, named F-box proteins, and a typical F-box protein is a component of an E3 ligase complex, named SCF, which consists of Skp1, Cullin 1, F-box protein and Rbx 1. Third, a typical SCF complex, along with two other proteins, E1 and E2, catalyzes the attachment of poly-ubiquitin chains to a subset of proteins through specific recognition by the F-box protein. The poly-ubiquitin chains allow the target proteins to be recognized by the 26S proteasome and degraded. Based on these biochemical clues, I have designed experimental approaches utilizing a variety of techniques to study the biochemical mechanism of S-RNase-based SI. In Chapter 2, I describe the use of various protein-protein interaction assays to show that PiSLF is not a typical F-box protein, as the PiSLF-containing complex consists of only two other proteins, Cullin 1 and PiSBP1, with PiSBP1 possibly playing the dual role of Skp1 and Rbx1. This finding suggests that PiSLF is likely involved in ubiquitin-mediated protein degradation. Further in vitro binding experiments showed that a PiSLF interacted with its non-self S-RNases (produced by different S-haplotypes) much more strongly than with its self S-RNase, and an S-RNase interacted with its non-self PiSLFs much more strongly than with its self PiSLF. This preferential binding with non-self S-RNases would allow the PiSLF-containing complex to specifically target non-self S-RNases for ubiquitination and degradation, but allow the self S-RNase to exert its RNase activity to degrade pollen RNAs. This finding thus provides a biochemical explanation for why a pistil only rejects its self-pollen tubes during SI interactions. I then developed a cell-free ubiquitination and degradation system using extracts of in vitro germinated pollen tubes, and showed that S-RNases were ubiquitinated and degraded via the ubiquitin-26S proteasome protein degradation pathway in vitro, albeit not in an S-haplotype-specific manner. Extensive sequencing of the S-locus region in Antirrhinum and several species in the Rosaceae family, all of which possess S-RNase-based SI, has revealed the existence of additional F-box genes at the S-locus. In Chapter 3, I describe the identification of genes encoding four PiSLF-like proteins that share many properties with PiSLF, and present both in vitro and in vivo comparative studies of PiSLF-like proteins and PiSLF. The results showed that none of the PiSLF-like proteins interacted with S-RNases to any significant degree, or functioned in SI, suggesting that PiSLF has a unique function in SI. Sequence comparison between PiSLF and these PiSLF-like proteins has revealed three domains that are specific to PiSLF. I used various chimeric proteins between PiSLF1 and PiSLF2, and between PiSLF2 and one of the PiSLF-like proteins, to show that one of the domains is responsible for the strong interaction with non-self S-RNases, and the other two domains together specifically suppress the interactions between PiSLF and its self S-RNase. This finding provides the biochemical basis for why a PiSLF preferentially interacts with its non-self S-RNases as described in Chapter 2. To further test the involvement of ubiquitin-26S proteasome-mediated protein degradation in SI, I studied whether any of the 20 lysine residues in S3-RNase of P. inflata might be targets for ubiquitination. In Chapter 4, I report the finding that six lysine residues near the C-terminus, when changed to arginines, significantly reduced ubiquitination and degradation of the mutant S3-RNase, GST:S3-RNase (K141-164R), in pollen tube extracts. I further showed that GST:S3-RNase (K141-164R) had similar RNase activity as GST:S3-RNase, suggesting that their degradation was not likely caused by an ER-associated protein degradation pathway that removes mis-folded proteins. Finally, I showed that PiSBP1 (P. inflata S-RNase Binding Protein 1), the RING-HC subunit of the PiSLF (P. inflata SLF)-containing E3-like complex identified in Chapter 2, could target S-RNase for ubiquitination in vitro. All these results suggest that ubiquitin-26S proteasome-dependent degradation of S-RNase is likely an integral part of the S-RNase-based SI mechanism. Two biochemical models, a degradation model and a sequestration (compartmentalization) model, have recently been proposed to explain the S-RNase-based SI mechanism. In the first part of Chapter 5, I provide a critical evaluation of these two models in view of the results I have obtained in my thesis research, and discuss why the model invoking specific degradation of non-self S-RNases via the ubiquitin-26S proteasome pathway can better explain some key aspects of SI. I have developed several projects and collaborated with three other graduate students in the Kao Lab to further test the validity of this model. In the second part of Chapter 5, I discuss the preliminary results from these projects, as well as future directions.