Detecting interactions between membrane proteins in vivo using chimeras

Analysis of Sec61p and Ssh1p interactions in the ER membrane using the split-ubiquitin system

BACKGROUND

The split-ubiquitin system monitors interactions of transmembrane proteins in yeast. It is based on the formation of a quasi-native ubiquitin structure upon interaction of two proteins to which the N- and C-terminal halves of ubiquitin have been fused. In the system we use here ubiquitin formation leads to proteolytic cleavage liberating a transcription factor (PLV) from the C-ubiquitin (C) fusion protein which can then activate reporter genes. Generation of fusion proteins is, however, rife with problems, and particularly in transmembrane proteins often disturbs topology, structure and function.

RESULTS

We show that both the Sec61 protein which forms the principal protein translocation channel in the endoplasmic reticulum (ER) membrane, and its non-essential homologue, Ssh1p, when fused C-terminally to CPLV are inactive. In a heterozygous diploid Sec61-CPLV is present in protein translocation channels in the ER membrane without disturbing their function and displays a limited set of protein-protein interactions similar to those found for the wildtype protein using biochemical methods. Although its expression level is similar, Ssh1-CPLV interactions are less strong, and, in contrast to Sec61p, Ssh1p does not distinguish between Sbh1p and Sbh2p. We show that interactions can be monitored by reporter gene activity or directly by PLV cleavage, which is more sensitive, but leads to quantitatively different results.

CONCLUSIONS

We conclude that the split-ubiquitin system we used here has high fidelity, but low sensitivity and is of limited use for detection of new, transient interactions with protein translocation channels in the ER membrane.

Complexes with mixed primary and secondary cellulose synthases are functional in Arabidopsis plants

In higher plants, cellulose is synthesized by so-called rosette protein complexes with cellulose synthases (CESAs) as catalytic subunits of the complex. The CESAs are divided into two distinct families, three of which are thought to be specialized for the primary cell wall and three for the secondary cell wall. In this article, the potential of primary and secondary CESAs forming a functional rosette complex has been investigated. The membrane-based yeast two-hybrid and biomolecular fluorescence systems were used to assess the interactions between three primary (CESA1, CESA3, CESA6), and three secondary (CESA4, CESA7, CESA8) Arabidopsis (Arabidopsis thaliana) CESAs. The results showed that all primary CESAs can physically interact both in vitro and in planta with all secondary CESAs. Although CESAs are broadly capable of interacting in pairwise combinations, they are not all able to form functional complexes in planta. Analysis of transgenic lines showed that CESA7 can partially rescue defects in the primary cell wall biosynthesis in a weak cesa3 mutant. Green fluorescent protein-CESA protein fusions revealed that when CESA3 was replaced by CESA7 in the primary rosette, the velocity of the mixed complexes was slightly faster than the native primary complexes. CESA1 in turn can partly rescue defects in secondary cell wall biosynthesis in a cesa8ko mutant, resulting in an increase of cellulose content relative to cesa8ko. These results demonstrate that sufficient parallels exist between the primary and secondary complexes for cross-functionality and open the possibility that mixed complexes of primary and secondary CESAs may occur at particular times.

Apple sucrose transporter SUT1 and sorbitol transporter SOT6 interact with cytochrome b5 to regulate their affinity for substrate sugars

Sugar transporters are central machineries to mediate cross-membrane transport of sugars into the cells, and sugar availability may serve as a signal to regulate the sugar transporters. However, the mechanisms of sugar transport regulation by signal sugar availability remain unclear in plant and animal cells. Here, we report that a sucrose transporter, MdSUT1, and a sorbitol transporter, MdSOT6, both localized to plasma membrane, were identified from apple (Malus domestica) fruit. Using a combination of the split-ubiquitin yeast two-hybrid, immunocoprecipitation, and bimolecular fluorescence complementation assays, the two distinct sugar transporters were shown to interact physically with an apple endoplasmic reticulum-anchored cytochrome b5 MdCYB5 in vitro and in vivo. In the yeast systems, the two different interaction complexes function to up-regulate the affinity of the sugar transporters, allowing cells to adapt to sugar starvation. An Arabidopsis (Arabidopsis thaliana) homolog of MdCYB5, AtCYB5-A, also interacts with the two sugar transporters and functions similarly. The point mutations leucine-73 --> proline in MdSUT1 and leucine-117 --> proline in MdSOT6, disrupting the bimolecular interactions but without significantly affecting the transporter activities, abolish the stimulating effects of the sugar transporter-cytochrome b5 complex on the affinity of the sugar transporters. However, the yeast (Saccharomyces cerevisiae) cytochrome b5 ScCYB5, an additional interacting partner of the two plant sugar transporters, has no function in the regulation of the sugar transporters, indicating that the observed biological functions in the yeast systems are specific to plant cytochrome b5s. These findings suggest a novel mechanism by which the plant cells tailor sugar uptake to the surrounding sugar availability.

Yeast as a model for medical and medicinal research

In the past, studies using the yeast Saccharomyces cerevisiae enabled major breakthroughs in the understanding of basic cellular and molecular processes. Today, the use of yeast is undergoing a "rebirth" in both fundamental and applied research. Indeed, advances in yeast technology have paved the way for a variety of new genome-wide screening approaches. Experimental strategies using yeast aim to unravel disease-related molecular events and to discover novel medicinal compounds. In this article, the impact of yeast as an experimental tool for disease-related studies is summarized and the use of yeast in high-throughput screenings for pharmacological purposes is evaluated. The recently applied and promising approach of so-called humanized yeast systems is also discussed.

Identification of novel ErbB3-interacting factors using the split-ubiquitin membrane yeast two-hybrid system

Analysis of membrane protein interactions is difficult because of the hydrophobic nature of these proteins, which often renders conventional biochemical and genetic assays fruitless. This is a substantial problem because proteins that are integral or associated with membranes represent approximately one-third of all proteins in a typical eukaryotic cell. We have shown previously that the modified split-ubiquitin system can be used as a genetic assay for the in vivo detection of interactions between the two characterized yeast transmembrane proteins, Ost1p and Wbp1p. This so-called split-ubiquitin membrane yeast two-hybrid (YTH) system uses the split-ubiquitin approach in which reconstitution of two ubiquitin halves is mediated by a protein-protein interaction. Here we converted the split-ubiquitin membrane YTH system into a generally applicable in vivo screening approach to identify interacting partners of a particular mammalian transmembrane protein. We have demonstrated the effectiveness of this approach by using the mammalian ErbB3 receptor as bait and have identified three previously unknown ErbB3-interacting proteins. In addition, we have confirmed one of the newly found interactions between ErbB3 and the membrane-associated RGS4 protein by coimmunoprecipitating the two proteins from human cells. We expect the split-ubiquitin membrane YTH technology to be valuable for the identification of potential interacting partners of integral membrane proteins from many model organisms.

Interactions between co-expressed Arabidopsis sucrose transporters in the split-ubiquitin system

BACKGROUND

The Arabidopsis genome contains nine sucrose transporter paralogs falling into three clades: SUT1-like, SUT2 and SUT4. The carriers differ in their kinetic properties. Many transport proteins are known to exist as oligomers. The yeast-based split ubiquitin system can be used to analyze the ability of membrane proteins to interact.

RESULTS

Promoter-GUS fusions were used to analyze the cellular expression of the three transporter genes in transgenic Arabidopsis plants. All three fusion genes are co-expressed in companion cells. Protein-protein interactions between Arabidopsis sucrose transporters were tested using the split ubiquitin system. Three paralogous sucrose transporters are capable of interacting as either homo- or heteromers. The interactions are specific, since a potassium channel and a glucose transporter did not show interaction with sucrose transporters. Also the biosynthetic and metabolizing enzymes, sucrose phosphate phosphatase and sucrose synthase, which were found to be at least in part bound to the plasma membrane, did not specifically interact with sucrose transporters.

CONCLUSIONS

The split-ubiquitin system provides a powerful tool to detect potential interactions between plant membrane proteins by heterologous expression in yeast, and can be used to screen for interactions with membrane proteins as baits. Like other membrane proteins, the Arabidopsis sucrose transporters are able to form oligomers. The biochemical approaches are required to confirm the in planta interaction.

Intra- and intermolecular interactions in sucrose transporters at the plasma membrane detected by the split-ubiquitin system and functional assays

Interaction of two separately expressed halves of sucrose transporter SUT1 was detected by an optimized split-ubiquitin system. The halves reconstitute sucrose transport activity at the plasma membrane with affinities similar to the intact protein. The halves do not function independently, and an intact central loop is not required for membrane insertion, plasma membrane targeting, and transport. Under native conditions, the halves associate into higher molecular mass complexes. Furthermore, the N-terminal half of the low-affinity SUT2 interacts functionally with the C-terminal half of SUT1. Since the N terminus of SUT2 determines affinity for sucrose, the reconstituted chimera has lower affinity than SUT1. The split-ubiquitin system efficiently detects intramolecular interactions in membrane proteins, and can be used to dissect transporter structure.

Interaction of the endoplasmic reticulum α1,2-mannosidase Mns1p with Rer1p using the split-ubiquitin system

The α1,2-mannosidase Mns1p involved in the N-glycosidic pathway in Saccharomyces cerevisiae is a type II membrane protein of the endoplasmic reticulum. The localization of Mns1p depends on retrieval from the Golgi through a mechanism that involves Rer1p. A chimera consisting of the transmembrane domain of Mns1p fused to the catalytic domain of the Golgi α1,2-mannosyltransferase Kre2p was localized in the endoplasmic reticulum of Δpep4 cells and in the vacuoles of rer1/Δpep4 by indirect immunofluorescence. The split-ubiquitin system was used to determine if there is an interaction between Mns1p and Rer1p in vivo. Co-expression of NubG-Mns1p and Rer1p-Cub-protein A-lexA-VP16 in L40 yeast cells resulted in cleavage of the reporter molecule, protein A-lexA-VP16, detected by western blot analysis and by expression of β-galactosidase activity. Sec12p, another endoplasmic reticulum protein that depends on Rer1p for its localization, also interacted with Rer1p using the split-ubiquitin assay, whereas the endoplasmic reticulum protein Ost1p showed no interaction. A weak interaction was observed between Alg5p and Rer1p. These results demonstrate that the transmembrane domain of Mns1p is sufficient for Rer1p-dependent endoplasmic reticulum localization and that Mns1p and Rer1p interact. Furthermore, the split-ubiquitin system demonstrates that the C-terminal of Rer1p is in the cytosol.