This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Traditionally, a proteinÂ´s function was defined by a single reaction, such as the binding of a molecule or the catalysis of a certain reaction. This can be viewed as the molecular function of a protein (Eisenberg et al., 2000). Currently, the functional genomics view indicates that proteins function as nodes in an extensive network of interacting molecules and, as such, a proteinÂ´s function should be characterized in relation to the interactions formed with other proteins in the cell (Eisenberg et al., 2000).
Protein-protein interactions refer to any of a number of encounters between proteins, ranging from transient interactions to the formation of stable complexes (Kluger and Alagic, 2004). Such interaction often illuminate the molecular mechanisms that form the core of biological processes and can take place with various specifications and affinities. The interactions between proteins can also be influenced by a variety of other factors, such as the concentration and oligomeric state of the respective proteins, as well as the ionic strength, pH and type of counter ions of the solvent (Howell et al., 2006).
Protein-protein interactions cause various effects inside a cell: 1) the kinetic properties or stability of proteins can be altered, which can lead to differences in substrate affinity, catalytic activity or allosteric properties of the proteins ; 2) substrate channeling is often effected by protein-protein interactions; 3) the interaction can reveal a new binding site or 4) can inactivate a protein and 5) substrate specificity can also be altered by protein-protein interactions (Kluger and Alagic, 2004; Phyzicky and Fields, 1995).
Identification of protein-protein interactions
There are various experimental methods to identify protein-protein interactions, and they differ in the level of resolution, namely 1) atomic observation, through the use of X-ray structures, 2) direct interactions such as those identified by phage display and 3) determination of multiprotein complexes that only identify the proteins in a complex and not the binding site, e. g. through mass-spectrometry (MS) analysis. Lastly, activity bioassays can identity the results of an interaction but does not provide information on the proteins involved in the interaction itself. (Xanarios and Eisenberg, 2001).
Different methods can be used to determine direct interactions between proteins. These can be used either in vivo or in vitro, with various advantages and disadvantages. In vivo methods include Two-Hybrid based approaches, where the bait protein is typically fused to a DNA binding domain, whilst the prey (often proteins expressed from a cDNA library) forms a fusion protein with a DNA activation domain (Howell et al., 2006). Protein fragment complementation assay (PCA) or assisted protein reassembly can be used to detect protein-protein interactions in vivo and is based on the fact that several proteins such as Green Fluorescent Protein (GFP), ribonuclease and chymotrypsin inhibitor-2 can be reconstituted from their peptide fragments, if the correct dissection site is chosen (Ghosh et al., 2000). Upon protein-protein interaction, these two domains are brought into close enough proximity that a specific phenotypic effect can be observed. During PCA, cells that are concurrently expressing two different proteins that are fused to fragments of a reporter protein such as GFP, will fluoresce only if there is a physical interaction between the two proteins that can bring the fragments of the specific reporter protein into close enough proximity that refolding can take place (Remy and Michnick, 2004). Chemical cross linking can be used both in vivo and in vitro, and entails the coupling of a specific bait protein with those in near proximity through the use of a cross linking reagent. In vitro, co-immunoprecipitation studies, where prey proteins that adhere to a specific bait protein are co-precipitated by a bait-specific antibody, can be very useful if bait-specific antibodies are available. Affinity-tagged bait proteins are routinely used for the analysis of protein interactions in affinity purification of protein complexes (pull-down assays). Phage display is a high throughput method where prey proteins are fused to the viral coat proteins, leading to the identification of proteins with affinity to the bait protein through the process of bio panning. Protein chip arrays, where the bait proteins are coupled to a chip surface and exposed to a plethora of possible prey proteins, followed by MS analysis, is another high-throughput method for the detection of protein-protein interactions. Several biophysical techniques such as fluorescence resonance energy transfer (FRET) or surface plasmon resonance can also be used to investigate protein-protein interactions (Howell etal., 2006; Phisicky and Fields, 1995).
In 1985, G. P. Smith illustrated that fusion proteins can be expresse on the surface of E. coli filamentous phage, if the nucleotide sequence encoding the desired antibody fragment, peptide or protein is fused to the nucleotide sequence that encodes a phage coat protein (Phizicky and Fields, 1995; Smith, 1985; Willats, 2002). This process, called phage display, is used today as a straightforward functional genomics method from the identification of protein-ligand interactions (Mullen et al., 2006). This ranges from the identification of antibodies, (Bradbury and Marks, 2004), to interaction between peptides and various cellular proteins, (Szardenings, 2003; Uchiyama et al., 2005) to the identification of peptides with high binding affinity to inorganic compounds such as a diverse array of metals, (Kriplani and Kay, 2005). Phage display fas even been used to identify peptides that binds to Bacillus spores, an application which may be used for the detection of biological weapons, such as anthrax that is cased by B. anthracis (Turnbough, 2003).
Phage display is made possible by the fact that fusion proteins often have the same or similar biological effects as the original proteins from which they are derived (Uchiyama et al., 2005). It entails the fusion of foreign DNA sequences to one of the genes that encodes viral coat proteins, resulting in the expression of fusion peptides om the viral coat surface. There are basically two different types of libraries that are used in phage display, namely synthetic random libraries and natural peptide libraries. Synthetic random libraries are created using random peptides ranging from 5-20 amino acids, and it is possible to constrain the flexibility of these peptides by cyclisation (Uchiyama et al., 2005; Willats, 2002). The advantage of synthetic random libraries is the great diversity that can be generated (Mullen et al., 2005), as well as the fact that the library can be designed to include specific structural elements (Hoess, 2001). The process of bio panning with a synthetic random peptide library often leads to peptides with conserved consensus sequences, which can then be used as leads for synthetic peptide synthesis and further studies (Uchiyama et al., 2005). In contrast, natural peptide libraries are created from genome fragments of selected organisms, for example by fusing a cDNA library to one of the genes that encode coat proteins. This implies that the peptides that are displaced should occur naturally in the organism, which is why this type of library is often used for detection of in vivo protein-protein interactions. The disadvantages od this method is the theoretically only 1 in every 18 clones will be native peptides (only 1 in 3 will commence properly due to possible frame shifting, only 1 in 3 will finish correctly and only 1 clone in 2 will be the appropriate sense vs. antisense strand) (Mullen et al., 2006; Podi et al., 2001). In this way, phage antibody libraries were created from the variable regions (V genes) of unimmunized or immunized organisms (Bradbury and Marks, 2004). However, it must be noted that certain authors regard phage antibody libraries as artificial ligands (Konthur and Crameri, 2003). Additionally, specific protein domain can be displayed on the surface of phage particles, thus allowing subsequent interaction studies with a specific bait protein (Willats, 2002).
Originally, the filamentous phage was used for polyvalent display where either the major capsid protein (g8p) encoded by gene VIII or the minor adsorpsion protin (g3p) encoded by geme III were involved in the cloning and expression of the fusion proteins (Azzazy and Highsmith, 2002; Smith, 1985). However, since all the g3p or g8p proteins were then expressed as recombinant proteins, severe limitations were imposed on the size of the fused protein to be displayed in order to maintain the viability of the phage particles. This problem was overcome with the development of a monovalent phagemid system. Phagemids are plasmids that contain both an E. coli and phage origin of replication, gene III, multiple cloning sites for the insertion of foreign DNA as well as a suitable antibiotic resistance gene. A helper phage that contains the majority of the genes needed for the construction of phage particles and wild-type copies of the coat protein are co-infected with the phagemid into the E. coli host. Thus, fusion coat proteins encoded by the phagemid and wild-type coat proteins provided by the helper phage are packaged in the E. coli host into phage particles capable of re-infection (Azzazy and Highsmith, 2002; Baek et al., 2002; Fernández, 2004; Hoess, 2001; Mullen et al., 2006; Phizicky and Fields, 1995; Willats, 2002).
In spite of these advantages, the filamentous phages are severely restricted as display systems. The foreign DNA id fused to the N-terminal of gene III or gene VIII, making this system unsuitable for the expression of cDNA fragments that does not start with an initiation codon or that contain stop codons (Mullen et al., 2006). In addition, the non-lytic proliferation method of this type of phage imply that only peptides that can be exported through the bacterial inner membrane can be incorporated into the phage particle, since phage assembly takes place in the periplasm (Willats, 2002). It has also been shown that certain peptides and proteins are not effectively assembled on the virion capsid. The difference between the cytoplasmic and periplasmic chemical environments can also affect the stability and folding characteristics of the displayed protein (Castagnoli et al., 2001). These disadvantages led to the investigation of different systems that would not suffer from these limitation, such as lytic T7 phage.
Phage display cycle consists of 5 basic steps by which the large diversity in a library can be screened to obtain a manageable number of protein binding partners with affinity to the bait protein:
1. A diverse library such as a cDNA or a synthetic random peptide is created and cloned into phagemid or phage genomes to produce phage particles that expresses the recombinant peptide fused to a surface protein.
2. The phage particles are brought into contact with the immobilized protein target (bait) for which a protein ligand (prey) is sought.
3. The non-binding phage particles are washed off.
4. The phage particles that bound to the immobilized protein are elured, amplified by infection into host bacteria and screened again.
5. The phage particles are analysed to identify the binding proteins (Willats., 2002).
The bio panning steps (steps 2-4) are repeated between three to five times to generate a library that is greatly enriched in the number of phage with binding affinity to the immobilized bait protein. However, since a library contains phages with a diverse range of avidity to the target protein, care must be taken to ensure a balance between the avidity and selectivity of the enriched clones. For instance, too little washing may lead to the enrichment of phage clones with high binding avidity, but low selectivity, while stringent washing may lead to the loss of phage clones with high selectivity, but weak binding (Willats, 2002).
Phage display for the study of protein-protein interactions
One major advantage of the phage display system for the study of protein-protein interaction is that a very large number of protein ligands can be screened in a short time. As such, phage display libraries with several billion variants can be used to study antibody and receptor binding sites, or the interaction between proteins and ligands in a matter of weeks (Azzazy and Highsmith, 2002; Rodi et al., 2001). There is also a genetic and phenotypic linkage due to the fact that the genetic information that codes for the phenotypic effect is already cloned into the phage itself, which facilitates downstream reactions such as sequencing (Paschke, 2006). The peptide ligand identified during this process can also give an indication of the residues that are involved in the binding of the bait and prey proteins, since only ahort peptides are expressed (Phizicky and Fields, 1995; Rodi et al., 2001; Willats, 2002). However, false negatives can occur due to the use of a bacterial expression system and the fact that a fusion protein is generated. It is possible that an in vivo ligand of the bait protein is not identified due to misfolding or a decrease in the accessibility of the relevant residues of the displayed recombinant protein (Phyzicky and Fields, 1995).
Most proteins contain specific residues that are involved in binding to other proteins, over and above the active sites of the enzymes that have evolved to allow the binding to specific small molecules (substrates). As such, proteins are viable targets for the identification of peptide ligands via phage display, since the binding of the displayed peptides usually occur at biologically relevant pockets. either at the active site or at other domains that have evolved to allow molecular interactions (Kay and Hamilton, 2001; Szardenings, 2003). It is worth nothing that the concept of "convergent evolution" can play a role in the analysis of interacting peptides, where the sequences of synthetic random peptides that bind to a specific target may have homology to the in vivo protein partners of the bait protein. These in vico partners can then be identified using similarity searches of the specific proteome (Kay and Hamilton, 2001; Kay et al., 2000). As such, these isolated peptides can either inhibit the activity of the protein, aid in identifying the in vivo protein partners of a protein or elucidate the molecular basis (key residue "hot-spots") of particular interactions between different protein binding partners (Kay and Hamilton, 2001).