Stable proteins

Abstract

The invention provides a fusion protein comprising, from N-terminus to C-terminus: a) a first portion of a Family B G-protein coupled receptor (GPCR) that comprises transmembrane helix (TM)-1, TM2 and TM3 of the GPCR; b) a stable protein domain; and c) a second portion of the GPCR comprising TM4, TM5, TM6 and TM7 of the GPCR. The invention also provides a method of crystallising a GPCR comprising providing the fusion protein of the invention and crystallising it to obtain crystals.

Description

The invention relates to proteins that are not readily crystallisable and particularly to GPCRs that are not readily stabilised and therefore not readily crystallised. The invention also relates to methods for crystallising such proteins and various uses of them. The proteins are useful for drug discovery and development studies.

GPCRs constitute a very large family of proteins that control many physiological processes and are the targets of many effective drugs. Reference is made particularly to Overington et al (2006) Nature Rev. Drug Discovery 5, 993-996 which indicates that over a quarter of present drugs have a GPCR as a target. They are of considerable pharmacological importance. A list of GPCRs is given in Foord et al (2005) Pharmacol Rev. 57, 279-288, which is incorporated herein by reference.

GPCRs are generally unstable when isolated, and despite considerable efforts, it has only been possible to crystallise a few GPCRs including bovine rhodopsin, which naturally is exceptionally stable and the beta 2 adrenergic receptor which was crystallised as a fusion protein or in complex with an antibody fragment.

GPCRs are thought to exist in multiple distinct conformations which are associated with different pharmacological classes of ligand such as agonists and antagonists, and to cycle between these conformations in order to function (Kenakin T(1997) Ann NY Acad Sci 812, 116-125). Switching between conformations also contributes to the difficulty inobtaining crystal structures of receptors.

Based on sequence homology and molecular architecture, GPCRs can be classified into three families (A, B and C), although they all share the characteristic seven transmembrane (TM) domain. Family A, the largest group, consists of receptors that are homologous to Rhodopsin. Family B, also referred to as Secretin receptor family, are 30 receptors that are regulated by large peptides hormones such as the glucagon hormone family; the members of this family are characterised by a relatively large extracellular N-terminus which contains several cysteines that form a network of disulphide bridges and is part of the ligand binding pocket. Family C consists of receptors that are homologous to metabotropic glutamate receptors; these receptors are characterised by a very long extra-cellular N-terminus as well as a long carboxy-tail, and the N-terminus forms the ligand binding pocket that has been shown to form a disulphide linked dimer, resembling a Venus fly trap in its shape.

Over the last few years the structure of a number of Family A GPCRs has been solved, and these milestones have been achieved by developing a number of key techniques. One such technology is the insertion of T4 lysozyme (T4L) in the intracellular cytoplasmic loop (ICL) 3, which is thought to create a large hydrophilic area that allows crystal contacts to form [2] [3]. The application of this technology in combination with Lipidic Cubic Phase crystallography has allowed high resolution structure determination of Beta2, A2a, CXCR4 and D3 receptors [2]. Thus, significant information has been gleaned from these structures about the orientation and organisation of the TM bundle of Family A receptors. However, little information is available for the members of Family B and C receptors and given the high sequence divergence, it is likely that significant differences exist in the architecture and organisation of the TM domains between families [1].

In Family A receptors, T4L was inserted in the ICL3 because it is thought that the distance between helices 5 and 6 is similar to the distance between the N- and C-termini of T4L. It is thus possible to accommodate the fusion protein in this position, whereas the distances between other helices are not believed to be conducive to the insertion of a fusion partner. Indeed, T4L has been fused to a number of different Family A receptors in ICL3 and functional protein has been expressed in each case with the added benefit of reducing the flexibility of the receptor, thus increasing overall stability.

We tested the effect of inserting T4L in the internal loops of Family B receptors and particularly in ICL3. Our data indicate that Family B receptors cannot tolerate T4L fusion in ICL3, however, surprisingly and unexpectedly, in view of the architecture of Family A receptors, adding T4L to ICL2 improves the biochemical properties of Family B receptors. ICL2 connects the portion of the GPCR comprising transmembrane helix (TM)-1, TM2 and TM3 to the portion of the GPCR comprising TM4, TM5, TM6 and TM7. The present data suggests that unlike Family A receptors, the distance between helices 3 and 4 in Family B receptors is more similar to the distance between the N- and C-termini of T4L, than is the distance between helices 5 and 6. Thus, insertion of a stable protein domain between these two portions of a GPCR is believed to represent a new technique for facilitating crystallisation of GPCRs that could not have been previously predicted.

Accordingly, a first aspect of the invention provides a fusion protein comprising, from N-terminus to C-terminus:

• • a) a first portion of a Family B G-protein coupled receptor (GPCR) that comprises TM1, TM2 and TM3 of the GPCR;
  • b) a stable protein domain; and
  • c) a second portion of the GPCR comprising TM4, TM5, TM6 and TM7 of the GPCR.

By “GPCR” we mean a G protein coupled receptor or polypeptide that has the signalling activity of a GPCR and retains an intact 7TM region. Standard nomenclature in the art designates the transmembrane helices of a GPCR from N-terminus to C-terminus as TM1, TM2, TM3, TM4, TM5, TM6 and TM7. The transmembrane helices are joined by stretches of amino acids extracellularly between TM2 and TM3, between TM4 and TM5, and between TM6 and TM7, referred to as extracellular loop (ECL)s 1, 2 and 3, respectively. The transmembrane helices are joined by stretches of amino acids intracellularly between TM1 and TM2, between TM3 and TM4, and between TM5 and TM6, referred to as intracellular loop (ICL)s 1, 2 and 3, respectively. Thus, the first and second GPCR portions as defined above are naturally joined by the 1CL2 region, i.e. ICL2 connects the first portion N-terminal to 1CL2 comprising TM1, TM2 and TM3 to the second portion C-terminal to ICL2 comprising TM4, TM5, TM6 and TM7.

The GPCR is preferably derived from full length wild type sequences including natural polymorphisms or mutant GPCR molecules that have been altered, for example so as to improve one or more properties of the GPCR eg stability.

The GPCR may be derived from wildtype and mutant GPCRs wherein mutant GPCRs may be stabilised GPCRs biased towards a particular conformation such as agonist or antagonist. For example, a stable protein domain may then be inserted between TM3 and TM4 of a conformationally stabilised GPCR.

We have previously developed a methodology for the stabilisation of a GPCR in a biologically relevant conformation (see WO 2008/114020) describing the production of stabilised GPCRs known as StaRs™ that enables the purification of recombinant G protein coupled receptors that maintain their conformation, stability and function when purified from the cell membrane. In addition, this platform technology also provides the means to engineer receptors biased either towards agonist conformation or the antagonist conformation (see also Magnani et at, 2008; Serrano-Vega et al, 2008; Shibata et al, 2009), i.e. they have increased stability in a particular conformation. Such stabilised receptors may be used in the present invention and have a number of advantages, for example stability, elevated yields of purified protein, reduced denaturation and reduced non-specific binding. Where a stable mutant GPCR is used in the present invention it is preferably selected and prepared using any of the methods as described in PCT applications WO 2008/114020, WO 2009/114020 and WO 2009/081136. Preferably the first and second GPCR portions are from a GPCR which has increased stability in a particular conformation relative to a parent GPCR (i.e. increased conformation stability). By increased conformational stability we include the meaning that a particular conformation of a mutant GPCR has, compared to the same conformation of the parent GPCR, increased stability (eg an extended lifetime) when exposed to a denaturant or denaturing conditions. Examples of denaturants/denaturing conditions include heat, detergent, a chaotropic agent and an extreme of pH. As is well known in the art, such denaturants or denaturing conditions can affect secondary and tertiary structures of a protein but not the primary sequence.

Suitable GPCRs for use in the practice of the invention include any Family B GPCR such as any of a glucagon-like peptide 1 receptor (GLP1R), glucagon-like peptide 2 receptor (GLP2R), calcitonin receptor (CT), amylin/CGRP receptor (AMY₁α), amylin receptor (AMY₂α), amylin/CGRP receptor (AMY₃α), CGRP/adrenomedullin receptor (CGRP₁α), adrenomedullin/CGRP receptor (AM₁α), adrenomedullin/CGRP receptor (AM₂α receptor), corticotropin releasing factor receptor (CRF₁), urocortins receptor (CRF₂), growth hormone releasing hormone receptor (GHRH), gastric inhibitory polypeptide receptor (GIP), glucagon receptor, secretin receptor, TIP-39 receptor (PTH2), parathyroid hormone receptor (PTH1), VIP/PACAP receptor (VPAC₁), PACAP receptor (PAC₂) and VIP/PACAP receptor (VPAC₂). In a particularly preferred embodiment, the GPCR is GLP1R. Other suitable GPCRs are well known in the art and include those listed in Overington et al supra. In addition, the International Union of Pharmacology produce a list of GPCRs that includes Family B GPCRs (Foord et al (2005) Pharmacol. Rev. 57, 279-288, and this list is periodically updated at http:/www.iuphar-db.org/GPCR/ReceptorFamiliesForward; Family B GPCRs are listed in Table 2 as Class 2 GPCRs).

The amino acid sequences (and the nucleotide sequences of the cDNAs which encode them) of many GPCRs are readily available, for example by reference to GenBank. In particular, Foord et al supra gives the human gene symbols and human, mouse and rat gene IDs from Entrez Gene (http://www.ncbi.nlm.nih.gov/entrez). It should be noted, also, that because the sequence of the human genome is substantially complete, the amino acid sequences of human GPCRs can be deduced therefrom.

Although the GPCR may be derived from any source, it is particularly preferred if it is from a eukaryotic source. It is particularly preferred if it is derived from a vertebrate source such as a mammal. It is particularly preferred if the GPCR is derived from rat, mouse, rabbit or dog or non-human primate or man. For the avoidance of doubt, we include within the meaning of “derived from” that a cDNA or gene was originally obtained using genetic material from the source, but that the protein may be expressed in any host cell subsequently. Thus, it will be plain that a eukaryotic GPCR (such as an avian or mammalian GPCR) may be expressed in a prokaryotic host cell, such as E. coli, but be considered to be avian- or mammalian-derived, as the case may be.

In some instances, the GPCR may be composed of more than one different subunit. For example, the calcitonin gene-related peptide receptor requires the binding of a single transmembrane helix protein (RAMP1) to acquire its physiological ligand binding characteristics. Effector, accessory, auxiliary or GPCR-interacting proteins which combine with the GPCR to form or modulate a functional complex are well known in the art and include, for example, receptor kinases, G-proteins and arrestins (Bockaert et al (2004) Curr Opinion Drug Discov and Dev 7, 649-657). In some instances, the GPCR may be bound to a GPCR ligand. By “ligand” we include any molecule which binds to the GPCR. Many ligands are known, for example from WO 2008/114020 and Neubig et al (2003) Pharmacol. Rev. 55, 597-606, both of which are incorporated herein by reference. Thus, the fusion protein may comprise a portion of the GPCR comprising TM1, TM2 and TM3 connected to a portion of the GPCR comprising TM4, TM5, TM6 and TM7, wherein the GPCR is bound to a GPCR binding partner. In this way, it is possible to gain structural insights into GPCR interactions by being able to crystallise complexes of GPCR with other molecules. It is preferred if the molecules are not ones that bind to ICL2 of the GPCR.

For any given GPCR, the TM helices can be determined by the skilled person using standard techniques in the art. For example, computer programs are available which model transmembrane regions of GPCRs based on hydrophobicity (Kyle & Dolittle (1982) J. Mol. Biol. 157, 105-132). Likewise transmembrane prediction algorithm servers are widely available on the World Wide Web (eg Expasy), many of which rely on hydropathy analysis. TMHMM is a membrane protein topology prediction method that may be used, based on a hidden Markov model (TMHMM Server v. 2.0; http://www.cbs.dtu.dklservices/TMHMM/). Where the transmembrane regions are already known for a given GPCR, for example by structural analysis or hydropathy analysis, the analogous regions in a further GPCR can also be identified by multiple or pairwise sequence alignment. For example, the alignment may be carried out using the Clustal W program (Thompson et al., 1994). The parameters used may be as follows: Fast pairwise alignment parameters: K-tuple(word) size; 1, window size; 5, gap penalty; 3, number of top diagonals; 5. Scoring method: x percent. Multiple alignment parameters: gap open penalty; 10, gap extension penalty; 0.05. Scoring matrix: BLOSUM.

FIG. 6 lists the amino acid sequences of Family B GPCRs and highlights the position of TM3, ICL2 and TM4. For example, for human GLP1R, TM3 ends with Phe 257, ICL2 corresponds to Ser 258 to Ser 261, and TM4 begins with Glu 262. Conveniently therefore, when the GPCR is a Family B receptor, the positions of TM3, ICL2 and TM4 can be identified by locating the amino acid residues that correspond to the amino acids that define the boundaries of TM3, ICL2 and TM4 in FIG. 6, when the sequences are aligned, for instance using CLUSTAL W.

It will be appreciated, however, that the boundaries are not absolute and they may well depend on the model provided for GLP1R that has been used to define them. In FIG. 1, for example, TM3 of human GLP1R ends with Leu 254, ICL2 corresponds to Leu 255 to Trp 264, and TM4 begins with Ile 265. Also, the loop regions may be defined as amino acid structures that join alpha helices or amino acid structures that are predicted to be outside the membrane, and depending on which definition is used, the boundaries will change.

In one embodiment, the stable protein domain is inserted into ICL2. Thus, the invention provides a GPCR into which a stable protein domain has been inserted into ICL2. By ‘inserted into ICL2’ we include both the addition of the amino acid sequence that defines the stable protein domain into the amino acid sequence of ICL2 without the deletion of any amino acids of ICL2, and also the replacement of one or more or all amino acids of ICL2 with the amino acid sequence encoding the stable protein domain. It will be appreciated that in this embodiment, the first and/or second portion of the GPCR may comprise at least part of ICL2, in addition to the requisite transmembrane helices. The first portion of the GPCR may comprise TM1, TM2 and TM3 and an N-terminal part of ICL2. The second portion of the GPCR may comprise TM4, TM5, TM6 and TM7 and a C-terminal part of ICL2.

It is appreciated that the stable protein domain may be inserted into ICL2 and flanked by one or two spacer moieties at its N- and/or C-terminus. In this way, the stable protein domain is not directly linked to ICL2 but is indirectly linked. The spacer moieties may be used to help reduce tension on the helices.

Preferably, the stable protein domain is inserted into ICL2 by replacing one or more consecutive amino acids (eg 2, 3, 4 or 5 or more amino acids) in the amino acid sequence of ICL2 with the amino acid sequence of the stable protein domain. In one embodiment, the one or more amino acids that are replaced is/are at least one or two amino acids from the C-terminus of TM3 and/or the N-terminus of TM4. In other words, the fusion protein may have at least one or two amino acids of ICL2 on at least one of the sides of the stable protein domain.

As described in Example 1, we have inserted T4 lysozyme at various positions in ICL2 of GLP1R, and insertions between Phe 257 and Ser 261 resulted in productive fusion GLP1R receptors. Thus, it is particularly preferred that the stable protein domain is inserted into the ICL2 region of the GPCR at a position between amino acid residues that correspond to amino acids Phe 257 and Ser 261 according to the numbering of human GLP1R as set out in FIG. 6.

Accordingly, the amino acid of the stable protein domain may be inserted into the ICL2 region of the GPCR after an amino acid corresponding to amino acid Phe 257 and before an amino acid corresponding to amino acid Ser 261 or Phe 260 or Vat 259, according to the numbering of human GLP1R as set out in FIG. 6. For example, the amino acid sequence of the stable protein domain may replace the amino acid corresponding to Ser 258, or it may replace amino acids corresponding to Ser 258 and Val 259, or it may replace amino acids corresponding to Ser 258, Val 259 and Phe 260, according to the numbering of human GLP1R as set out in FIG. 6. Inserting the stable protein domain at such positions corresponds to the GLP1R-T4 lysozyme fusion constructs 1c, 2c and 3c shown in FIG. 1.

Similarly, the amino acid of the stable protein domain may be inserted into the ICL2 region of the GPCR after an amino acid corresponding to amino acid Ser 258 and before an amino acid corresponding to amino acid Ser 261 or Phe 260 or Val 259, according to the numbering of human GLP1R as set out in FIG. 6. For example, the amino acid sequence of the stable protein domain may replace the amino acid corresponding to Val 259, or it may replace amino acids corresponding to Val 259 and Phe 260, according to the numbering of human GLP1R as set out in FIG. 6. Inserting the stable protein domain at such positions corresponds to the GLP1R-T4 lysozyme fusion constructs 1d, 2d and 3d shown in FIG. 1.

By “corresponding amino acid residue” we include the meaning of the amino acid residue in another GPCR that aligns to the given amino acid residue in human GLP1R when the human GLP1R receptor and the other GPCR are compared using MacVector and CLUSTALW.

Although it is preferred that the stable protein domain is inserted into the ICL2 region of the GPCR at a position between amino acid residues that correspond to amino acids Phe 257 and Ser 261 according to the numbering of human GLPIR as set out in FIG. 6, it is appreciated that it may be inserted outside of this region.

It is appreciated that the function of the stable protein domain is to increase the hydrophilic surface for crystal contacts and to reduce the inherent flexibility of GPCRs in order to, for example, improve the crystallisation properties of the GPCR. Accordingly, by ‘stable protein domain’ we include the meaning of any soluble, folded polypeptide that provides a hydrophilic surface for crystal lattice contacts. Further, the protein domain is stable such that in its folded form it is resistant to denaturation (eg is stable to heat, detergents and chaotropic agents etc.). Tests for protein stability are well known in the art and include those described in W02008/114020.

Typically, the stable protein domain is one that folds autonomously from the GPCR portions of the fusion protein in the cell.

Conveniently, the stable protein domain is one that is readily crystallisable. Thus, the stable protein domain may be a protein whose crystal structure has been solved, for example one whose coordinates has been deposited in the Protein Data Bank (http://www.pdb.org/).

Particularly preferred characteristics of the stable protein domain are:

• • 1. the domain is soluble, well folded and can be expressed easily in one or more expression systems;
  • 2. the N- and C-termini of the domain are close together in space; typically in the range of 5-17 A eg 6-16 Å, 7-15 Å, 7-10 Å, 10-13 Å or 12-15 Å;
  • 3. the domain is resistant to thermal and chemical denaturation as well as to proteolytic degration;
  • 4. the domain is highly crystallisable in a variety of space groups and crystal packing arrangements.

It is preferred that the domain does not contain cysteine residues so as to prevent disulphide bond formation either within the domain or with the GPCR portion of the fusion protein. It will be understood that since the domain is soluble it should not be hydrophobic or have the propensity to aggregate in a disordered fashion.

In one embodiment, the length of the stable protein domain is between 50 and 1000 amino acids, preferably between 50 and 300 amino acids or 100 and 300 amino acids, or between 150 and 250 amino acids.

Once a suitable polypeptide has been found for the stable protein domain, it may be necessary to modify the polypeptide by deleting or adding amino acid residues from or to the N-terminus, the C-terminus or both termini of the polypeptide such that the closest alpha carbon atoms in the backbone at the termini of the polypeptide are spaced by a distance of in the range of 5-17 Å eg 6-16 Å, 7-15 Å, 7-10 Å, 10-13 Å or 12-15 Å.

It is preferred if insertion of the stable protein domain does not affect a biological activity of the GPCR, such as a binding activity or a signalling pathway modulation activity. Ideally, the fusion protein should retain at least 60% or 70% or 80% or 90% of its biological activity, and most ideally 100% of its biological activity relative to the level of the same activity in the absence of the stable protein domain. Methods for assessing GPCR binding and GPCR signalling are well known in the art and are described for example in WO 2008/114020 and WO 2009/101383, both of which are incorporated herein by reference. Thus, where the biological activity is a binding activity, binding to any GPCR binding partner may be assessed using routine binding assays known in the art; where the biological activity is a signalling pathway modulating activity, the activity can be assessed by any suitable assay for the particular signalling pathway (eg reporter gene assays).

It is appreciated that retaining ligand binding ability is more important for crystallisation purposes than is retaining signalling activity, and that it may be desirable to only assess ligand binding ability prior to crystallisation. Thus in a particularly preferred embodiment, the stable protein domain does not affect a binding activity of a GPCR.

For the avoidance of doubt, the stable protein domain is not ICL2 or part thereof of the particular GPCR.

In a preferred embodiment, the stable protein domain is lysozyme. Lysozyme is known to be readily crystallisable and the structures of various wild type and variant lysozymes have been deposited in the Protein Data Bank (www.rcsb.org). Suitable examples include 135L, 193L, 194L, 1AKI, 1GBS, 1IEE, 1LZ1, 1P7S, 1REX, 1VDQ, 2ANV, 2ANX, 2D4K, 2FBB, 2IHL, 2NWD, 2XBR, 2XBS, 2Z2F, 2ZYP, 3A8Z, 3K2R, 3N9A, 3N9C, 3N9E and 3OD9

Although lysozyme derived from any source may be used, it is particularly preferred if the lysozyme is derived from T4 phage. Two amino acid sequences of T4 phage lysozyme are provided in FIG. 7, and either of the sequences may be used in the context of the invention. The lysozyme may be derived from full length wild type sequences including natural polymorphisms or it may be a mutant lysozyme that has been altered, for example to improve one or more properties. Thus, it is understood that variants of the amino acid sequences provided in FIG. 7 may be used, such as amino acid sequences with at least 60%, 65%, 70%, 75%, 80%, 85% or 90% sequence identity with either of the sequences set out in FIG. 7, and more preferably at least 95% or 99% sequence identity with either of the sequences set out in FIG. 7.

Sequence identity may be measured by the use of algorithms such as BLAST or PSI-BLAST (Altschul et at, NAR (1997), 25, 3389-3402) or methods based on Hidden Markov Models (Eddy S et al, J Comput Biol (1995) Spring 2 (1) 9-23). Typically, the percent sequence identity between two polypeptides may be determined using any suitable computer program, for example the GAP program of the University of Wisconsin Genetic Computing Group and it will be appreciated that percent identity is calculated in relation to polypeptides whose sequence has been aligned optimally. The alignment may alternatively be carried out using the Clustal W program (Thompson et al., 1994) as mentioned above.

While lysozyme is a preferred example of a stable protein domain, the general principles may be used to employ any number of polypeptides that have the characteristics discussed above. Thus, suitable candidates include those containing the amino acid sequence of proteins that are readily crystallisable, for example as found by interrogating the protein data bank or other crystallization databases known in the art. Other examples include those mentioned in Engel et al (2002) BBA 1564: 38-46 (incorporated herein by reference), such as cytochrome_b562, flavodoxin, β-lactamase and 70 kDa heat shock ATPase domain.

The fusion protein may be modified so that it can be more easily detected, for example by biotinylating it or by incorporating any detectable label known in the art such as radiolabels, fluorescent labels or enzymatic labels. In a particularly preferred embodiment, the label is a fluorescent label such as EGFP. Similarly, the fusion protein may be modified to facilitate purification, for example by incorporating any affinity moiety known in the art such as a GST tag, 6× His tag, MBP or other epitope tag. Such modifications may be at the N-terminus or C-terminus of the GPCR or in an external loop.

As demonstrated in Example 1, it is believed that the fusion proteins of the invention have improved biochemical properties compared to the biochemical properties of the GPCR without the insertion of the stable protein domain. Such improved properties make the fusion protein more amenable to crystallisation. Thus, the fusion protein is expected to have a larger hydrophilic surface for crystal contacts. Similarly, the fusion protein is expected to be more soluble, eg displaying less aggregation in detergent solution, than the GPCR without insertion of the stable protein domain. Methods for assessing GPCR solubility are well known in the art and include size exclusion chromatography such as fluorescent size exclusion chromatography used to assess solubility in DDM as described in Example 1. The fusion protein may also be more stable (eg to any of heat, detergent or chaotropic agents) than the GPCR without the insertion of the stable protein domain. Methods for assessing GPCR stability are known in the art, including those described in WO 2008/114020.

Conveniently, the fusion protein is produced by standard molecular biology and recombinant DNA techniques. For example, DNA fragments encoding the first and second GPCR portions and the stable protein domain may be made using standard cloning techniques and PCR as is well known in the art. The fragments can then be ligated together in-frame in accordance with conventional practice, e.g., by employing blunt-ended or stagger-ended termini for ligation, restriction enzyme digestion to provide for appropriate termini, filling-in of cohesive ends as appropriate, alkaline phosphatase treatment to avoid undesirable joining, and enzymatic ligation.

Equally, the fusion constructs can be made using ligase independent cloning strategies such as InFusion or Gateway. The construct may also be made synthetically through de novo gene synthesis.

It is appreciated that one or both of the GPCR and stable protein domain may be mutated so as to improve any of solubility, stability, expression and crystallisability.

Molecular biological methods for cloning and engineering genes and cDNAs, for mutating DNA, and for expressing polypeptides from polynucleotides in host cells are well known in the art, as exemplified in “Molecular cloning, a laboratory manual”, third edition, Sambrook, J. & Russell, D. W. (eds), Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y., incorporated herein by reference.

Suitable expression systems include constitutive or inducible expression systems in bacteria or yeasts, virus expression systems such as baculovirus, semliki forest virus and lentiviruses, or transient transfection in insect or mammalian cells. Suitable host cells include E. coli, Lactococcus lactis, Saccharomyces cerevisiae, Schizosaccharomyces pombe, Pichia pastoris, Spodoptera frugiperda and Trichoplusiani cells. Suitable animal host cells include HEK 293, COS, S2, CHO, NSO, DT40 and so on. It is known that some GPCRs require specific lipids (eg cholesterol) to function. In that case, it is desirable to select a host cell which contains the lipid. Additionally or alternatively the lipid may be added during isolation and purification of the fusion protein. Purification may be carried out by standard techniques such as affinity chromatography.

A second aspect of the invention provides a polynucleotide encoding the fusion protein according to the first aspect of the invention. The polynucleotide may be RNA (eg mRNA) or DNA, although typically it is DNA.

It will be appreciated that the polynucleotide may be incorporated into a vector, and so the invention also provides a vector comprising a polynucleotide according to the second aspect of the invention.

Suitable vectors are ones which propagate in and/or allow expression of the fusion protein in prokaryotic (eg bacterial) or eukaryotic (eg mammalian) cells. For example, the vector may be a plasmid, a cosmid, a phage or a bacterial artificial chromosome (BAC). The polynucleotide sequence of the vector will depend upon the nature of the intended host cell, the manner of the introduction of the polynucleotide of the second aspect of the invention into the host cell, and whether episomal maintenance or integration is desired. Conveniently, the vector comprises at least one selectable marker such as antibiotic resistance (eg kanamycin or neomycin).

Vectors are useful to replicate the polynucleotide of the second aspect of the invention, and are also useful to transfect cells with the polynucleotide, and may also promote expression of the fusion protein.

Typical prokaryotic vector plasmids are: pUC18, pUC19, pBR322 and pBR329 available from Biorad Laboratories (Richmond, Calif., USA); pTrc99A, pKK223-3, pKK233-3, pDR540 and pRIT5 available from Pharmacia (Piscataway, N.J., USA); pBS vectors, Phagescript vectors, Bluescript vectors, pNH8A, pNH16A, pNH18A, pNH46A available from Stratagene Cloning Systems (La Jolla, Calif. 92037, USA).

A typical mammalian cell vector plasmid is pSVL available from Pharmacia (Piscataway, N.J., USA). This vector uses the SV40 late promoter to drive expression of cloned genes, the highest level of expression being found in T-antigen-producing cells, such as COS-1 cells. Another example is pcDNA3.1 (neo) (Invitrogen) for use in COS-1 or COS-7 cells. An example of an inducible mammalian expression vector is pMSG, also available from Pharmacia (Piscataway, N.J., USA). This vector uses the glucocorticoid-inducible promoter of the mouse mammary tumour virus long terminal repeat to drive expression of the cloned gene.

Useful yeast plasmid vectors are pRS403-406 and pRS413-416 and are generally available from Stratagene Cloning Systems (La Jolla, Calif. 92037, USA). Plasmids pRS403, pRS404, pRS405 and pRS406 are Yeast Integrating plasmids (Yips) and incorporate the yeast selectable markers HIS3, TRP1, LEU2 and URA3. Plasmids pRS413-416 are Yeast Centromere plasmids (YCps).

In a preferred embodiment, the vector comprising the polynucleotide of the second aspect of the invention is pcDNA3.1 (http://products.invitrogen.com/ivgn/product/V79020).

Any suitable method known in the art may be used to construct vectors containing the polynucleotide of the second aspect of the invention including the ligation techniques described above.

A third aspect of the invention provides a cell comprising a polynucleotide according to the second aspect of the invention, or a vector comprising said polynucleotide. Such cells may be used to replicate the polynucleotide of the second aspect of the invention, or may be used to express the fusion protein of the first aspect of the invention.

The cell can be either prokaryotic or eukaryotic.

It is appreciated that construction and amplification of the polynucleotide of the second aspect of the invention is conveniently performed in bacterial cells. Expression of the polynucleotide may be carried out in cells such as mammalian cells or bacterial cells.

Bacterial cells are preferred prokaryotic host cells and typically are a strain of E. coil such as, for example, the E. coil strains DH5 available from Bethesda Research Laboratories Inc., Bethesda, Md., USA, and RR1 available from the American Type Culture Collection (ATCC) of Rockville, Md., USA (No ATCC 31343). Preferred eukaryotic host cells include yeast, insect and mammalian cells or cell lines, preferably vertebrate cells or cell lines such as those from a mouse, rat, monkey or human. Particularly preferred cells are human embryonic kidney cells such as HEK293T cells.

Cells used for expressing the fusion protein may be either stably or non-stably transfected.

A fourth aspect of the invention provides a method of crystallising a fusion protein according to the first aspect of the invention, the method comprising providing a fusion protein according to the first aspect of the invention and crystallising it to obtain crystals.

In an embodiment, the fusion protein is provided by culturing a host cell according to the third aspect of the invention to express the fusion protein and isolating the protein.

Any suitable crystallisation method may be used to crystallise the fusion protein, such as any of those reviewed in “Crystallisation of Biological Macromolecules” (Alexander McPherson; ISBN: 0-87969-617-6), which is incorporated herein by reference.

In a preferred embodiment, the crystallisation is carried out using lipidic cubic phase crystallography (see US 2011/0031438 incorporated herein by reference).

A fifth aspect of the invention provides a crystal comprising a fusion protein of the first aspect of the invention.

The fusion proteins disclosed herein are useful for crystallisation studies and are useful in drug discovery programmes. They may be used in biophysical measurements of receptor/ligand kinetic and thermodynamic parameters eg by surface plasmon resonance or fluorescence based techniques. They may be used in ligand binding screens, and may be coupled to solid surfaces for use in high throughput screens or as biosensor chips. Biosensor chips containing the fusion proteins may be used to detect molecules, especially biomolecules.

The invention will now be described with the aid of the following figures and examples.

FIGS. 1A-1B. Design of T4L fusion constructs with GLP1R. T4L was inserted after the indicated residues in ICL2 (left) and ICL3 (right). Construct 1 a means that T4L was inserted between L255 and 8261. The model of TM domains and loops are based on reference [4].

FIG. 2. Total EGFP signal of GLP1R-T4L fusion constructs compared to the wild-type (WT) and the mock transfected (U) samples. Each measurement was done on 50 ug of total cellular material in duplicate. Error bars represent standard deviation from mean.

FIG. 3. Typical fSEC elution profile of wild-type GLP1R.

FIGS. 4A-4C. fSEC elution profiles DOM solubilised GLP1 R-T4L fusion constructs in the ICL2 overlaid the wild-type profile. In each case, the wild-type profile is shown in red the fusion constructs in blue.

FIGS. 5A-5E. fSEC elution profiles DOM solubilised GLP1R-T4L fusion constructs in the ICL3 overlaid the wild-type profile. In each case, the wild-type profile is shown in red the fusion constructs in blue.

FIG. 6. Amino acid sequences of Family B GPCRs showing position of TM3, ICL2 and TM4 (SEQ ID Nos: 1-22). The portion of ICL2 that was replaced with T4L in GLP1R fusion construct 1c is highlighted in other Family B receptors from mouse, rat and human.

FIG. 7. Amino acid sequences of T4 phage lysozyme: (A) Sequence inserted into ICL3 of Family A receptors [2], [3] (SEQ ID No: 23); (B) Sequence inserted into ICL2 of Family B receptors (SEQ ID No: 24) (see Examples). Differences are indicated in boxes.

EXAMPLE 1

Insertion of T4L into ICL2 Improves Biochemical Properties of GLP1R

Summary

We have tested the effect of inserting T4L in the internal loops of Family B receptors. Our data indicate that Family B receptors cannot tolerate T4L fusion in ICL3, however, adding T4L to ICL2 improves the biochemical properties of the receptor.

Results

The loop regions of GLP1R were determined according to the model of GLP1R and the DNA encoding T4L was inserted in different locations within ICL2 and ICL3 (FIG. 1). The GLP1R construct was C-terminally tagged with the EGFP in order to monitor total expression as well as monodispersity using fluorescent-detection size exclusion chromatography.

Following sequence confirmation, these constructs were expressed in HEK293T transiently. As an initial analysis, the EGFP signal in whole cells was measured to assess the total levels of expression. Interestingly, the constructs in ICL3 failed to produce any EGFP signal, indicating that T4L fusion in this region of GLP1R is incompatible with the overall architecture of this receptor. However, fusions in the ICL2 resulted in the robust expression of GLP1R (FIG. 2).

In order to analyse the biochemical properties of GLP1R-T4L fusions, cells expressing these constructs were solubilised in dodecyl maltoside (DDM) and applied to fluorescence-detection size-exclusion chromatography (fSEC). fSEC has been used widely to provide data regarding the monodispersity and the aggregation status of the proteins, particularly in pre-crystallisation screens [5]. In general, the more favourable conditions will result in more monodispersity and reduced aggregation. The fSEC elution profile of DDM-solubilised wild-type GLP1R shows the presence of the main monodispersed peak with an aggregation shoulder as well as free EGFP species that are the result of proteolytic degredation (FIG. 3).

Consistent with the EGFP signal data showed in FIG. 2, the elution profiles of ICL2 T4L fusion constructs 1a, 1b, 2a, 2b, 3a and 3b indicates that cells failed to express these fusions (FIG. 4). This is most likely due to the proximity of the N-terminus of T4L to TM III, leading to the disruption of the overall structure. In contrast, the elution profiles of constructs 1c, 1d, 2c, 2d, 3c and 3d revealed that productive fusion receptors were expressed and more significantly it appears that T4L fusion results in the reduction of the aggregation peak and concomitant improvement of the monodispersed peak, which together indicate that T4L insertion in this region of the receptor has beneficial effect on the biochemical properties of solubilised receptor (FIG. 4). This effect is most pronounced in the constructs 1c and 2c.

The same analysis was carried out for the ICL3 fusion constructs and in agreement with the EGFP signal data showed in FIG. 2, none of the T4L fusions in the ICL3 resulted in the expression of any productive fusion receptor (FIG. 5).

Taken together, these data indicate that T4L fusion in the third cytoplasmic loop of GLP1R is not tolerated, however, insertion of T4L in certain positions in the second cytoplasmic loop not only is tolerated, it also improves the biochemical properties of the solubilised receptor. Given the high sequence homology amongst members of the Family B GPCRs, we suggest that these observations can be extended to other members of this Family. The portion of the ICL2 that was replaced in the best construct (1c) is highlighted in other Family B members as shown in FIG. 6.

Methods and material

T4 lysozyme was inserted in the second and third cytoplasmic loops of human GLP1R using standard molecular biology techniques. These constructs were transiently expressed from a modified pcDNA3.1 in HEK293T cells, generating receptors fused to EEGFP at their C-termini. Transfections were carried out using GeneJuice (Merck Biosciences) according to the manufacturer's guideline. Typically, 6 ug of DNA was used to transfect 3×10⁶ adherent cells in 10 cm plates. Cells were harvested about 40 hours post transfection and re-suspended in 50 mM HEPES pH 7.5/150 mM NaCl/0.5 mM EDTA complemented with Complete EDTA-free protease inhibitor cocktail (Roche). Typically 650 ug of each sample was solubilised with 1% DDM in total volume of 200 uL for 1 hour at 4° C. followed by centrifugation at 50000 rpm for 30 minutes. 50 uL of the supernantant was loaded onto BioSEep-SEC-S3000 column (Phenomenex), pre-equilibrated with SEC buffer (50 mM HEPES pH 7.5/150 mM NaCl/0.5 mM EDTA/0.03% DDM) and run at the flow rate of 1 mL/minute for 15 minutes. The eluent was passed through an on-line fluorometer with the following settings: excitation 490 nm, emission 513 nm and gain of 13.

REFERENCES

[1] Kristiansen, K Molecular mechanisms of ligand binding, signaling, and regulation within the superFamily of G-protein-coupled receptors: molecular modeling and mutagenesis approaches to receptor structure and function. Pharma & Therap 103, 21-80 (2004).

[2] Bill R M, Henderson P J, Iwata S, Kunji E R, Michel H, Neutze R, Newstead S, Poolman B, Tate C G and Vogel H. Overcoming barriers to membrane protein structure determination. Nat Biotechnol. 29(4), 335-340 (2011).

[3] Kobilka B K, Kobilka T S, Daniel K, Regan J W, Caron M G and Lefkowitz R J. Chimeric alpha 2-,beta 2-adrenergic receptors: delineation of domains involved in effector coupling and ligand binding specificity. Science 240(4857) 1310-6 (1988).

[4] Frimurer T M and Bywater R P. Structure of the integral membrane domain of the GLP1 receptor. Proteins 35(4), 375-86 (1999).

[5] Kawate T and Gouaux E. Fluorescence-detection size-exclusion chromatography for precrystallization screening of integral membrane proteins. Structure 14(4), 673-81 (2004).

Claims

1. A fusion protein comprising, from N-terminus to C-terminus: a. a first portion of a Family B G-protein coupled receptor (GPCR) that comprises transmembrane helix (TM)-1, TM2 and TM3 of the GPCR, wherein the TM1 and the TM2 are joined by intracellular loop 1 (ICL1), and the TM2 and TM3 are joined by extracellular loop 1 (ECL1);b. a stable protein domain; andc. a second portion of the GPCR comprising TM4, TM5, TM6 and TM7 of the GPCR wherein the TM4 and the TM5 are joined by extracellular loop 2 (ECL2), the TM5 and the TM6 are joined by intracellular loop 3 (ICL3), and the TM6 and the TM7 are joined by extracellular loop 3 (ECL3);wherein the stable protein domain comprises a soluble, well-folded polypeptide that provides N- and C-termini, the distance between which approximates the distance between helices 3 and 4 in the Family B GPCR, and that provides a hydrophilic surface for crystal lattice contacts, thereby facilitating crystallisation,wherein the stable protein domain is inserted into the intracellular loop 2 (ICL2) region of the GPCR which loop joins the TM3 in the first portion of the GPCR and the TM4 in the second portion of the GPCR, andwherein the fusion protein displays reduced aggregation in the presence of a detergent solution, as compared to a GPCR without insertion of the stable protein domain into the ICL2 region of the GPCR.

2. The fusion protein according to claim 1, wherein the stable protein domain is inserted into the ICL2 region of the GPCR at a position between amino acid residues that correspond to amino acids Phe 257 and Ser 261 of human GLP1R, wherein Phe 257 corresponds to the fifth amino acid in SEQ ID NO: 30 and Ser 261 corresponds to the ninth amino acid in SEQ ID NO: 30.

3. The fusion protein according to claim 2, wherein the stable protein domain is inserted into the ICL2 region of the GPCR after an amino acid corresponding to amino acid Phe 257 of human GLP1R and before an amino acid corresponding to amino acid Ser 261 or Phe 260 or Val 259, wherein Phe 257 corresponds to the fifth amino acid in SEQ ID NO: 30, Ser 261 corresponds to the ninth amino acid in SEQ ID NO: 30, Phe 260 corresponds to the eighth amino acid in SEQ ID NO: 30, and Val 259 corresponds to the seventh amino acid in SEQ ID NO: 30.

4. The fusion protein according to claim 2, wherein the stable protein domain is inserted into the ICL2 region of the GPCR after an amino acid corresponding to amino acid Ser 258 of human GLP1R and before an amino acid corresponding to amino acid Ser 261 or Phe 260 or Val 259 of human GLP1R, wherein Ser 258 corresponds to the sixth amino acid in SEQ ID NO: 30, Ser 261 corresponds to the ninth amino acid in SEQ ID NO: 30, Phe 260 corresponds to the eighth amino acid in SEQ ID NO: 30, and Val 259 corresponds to the seventh amino acid in SEQ ID NO: 30.

5. The fusion protein according to claim 1, wherein the stable protein domain comprises a lysozyme and wherein the amino acid sequence of the lysozyme is at least 90% identical to SEQ ID NO: 24.

6. The fusion protein according to claim 1, further comprising a detectable moiety.

7. The fusion protein according to claim 6, wherein the detectable moiety is EGFP.

8. A crystal comprising the fusion protein of claim 1.

9. The fusion protein according to claim 1, which is in a solubilised form or which is substantially free of other proteins or which is immobilised to a solid support.

10. The fusion protein according to claim 1, wherein the stable protein domain reduces the inherent flexibility of the GPCR fusion protein.

11. The fusion protein according to claim 1, wherein the stable protein domain: a. has an N-terminus that is within 5-17 Å of its C-terminus;b. is resistant to thermal and chemical denaturation, as well as proteolytic denaturation; andc. is highly crystallisable in a variety of space groups and crystal packing arrangements.

12. The fusion protein according to claim 1, wherein the stable protein domain has an N-terminus that is within 6-16 Å, or within 7-15 Å, or within 7-10 Å, or within 10-13 Å, or within 12-15 Å of its C-terminus.

13. The fusion protein according to claim 1, wherein the fusion protein is characterised in that it is crystallisable.

14. A fusion protein according to claim 1, wherein the stable protein domain comprises a polypeptide selected from a cytochromeb562, a flavodoxin, a β-lactamase and a 70 kDa heat shock ATPase domain.

15. The fusion protein according to claim 1, wherein the Family B GPCR is a glucagon-like peptide 1 receptor (GLP1R), glucagon-like peptide 2 receptor (GLP2R), or glucagon receptor.

16. The fusion protein according to claim 1, wherein the Family B GPCR is a calcitonin receptor (CT).

17. The fusion protein according to claim 1, wherein the Family B GPCR is an amylin/CGRP receptor (AMY1α), amylin receptor (AMY2α), amylin/CGRP receptor (AMY3α), CGRP/adrenomedullin receptor (CGRP1α), adrenomedullin/CGRP receptor (AM1α), or adrenomedullin/CGRP receptor (AM2α receptor).

18. The fusion protein according to claim 1, wherein the Family B GPCR is a corticotropin releasing factor receptor (CRF1), or urocortins receptor (CRF2).

19. The fusion protein according to claim 1, wherein the Family B GPCR is a growth hormone releasing hormone receptor (GHRH).

20. The fusion protein according to claim 1, wherein the Family B GPCR is a gastric inhibitory polypeptide receptor (GIP).

21. The fusion protein according to claim 1, wherein the Family B GPCR is a secretin receptor.

22. The fusion protein according to claim 1, wherein the Family B GPCR is a TIP-39 receptor (PTH2), or parathyroid hormone receptor (PTH1).

23. The fusion protein according to claim 1, wherein the Family B GPCR is a VIP/PACAP receptor (VPAC1), PACAP receptor (PAC2), or VIP/PACAP receptor (VPAC2).

24. The fusion protein according to claim 5, wherein the lysozyme is T4 lysozyme.

25. The fusion protein according to claim 6, wherein the detectable moiety is a fluorescent label, a radiolabel, or an enzymatic label.