Refolding of membrane proteins

In a method for production of membrane receptors folded into their native structure, first, receptors solubilized in a first detergent are provided. To induce folding of receptors into their native form, the first detergent is exchanged for a second detergent. Both for the first and for the second detergent, examples are shown.

Skip to: Description  ·  Claims  · Patent History  ·  Patent History
Description
CROSS-REFERENCE TO RELATED APPLICATIONS

This application is a continuation-in-part of co-pending U.S. application Ser. No. 10/069,433 filed on Feb. 19, 2002, which is a continuation of International application PCT/EP00/07763 filed on Aug. 10, 2002, and claims priority of German patent application DE 199 39 246.3 filed on Aug. 19, 1999, all of which are incorporated herein by reference in their entireties.

FIELD OF THE INVENTION

The present invention relates to a method for the production of proteins folded into their native or active structure, said protein being from the group of membrane receptors, excluding G-protein-coupled receptors, preferably ion channels, comprising:

    • providing a protein from the group of membrane receptors solubilized in a first detergent, and
    • exchanging said first detergent for a second detergent, to induce the folding of said protein into its native or active form.

DESCRIPTION OF THE RELATED ART

For membrane proteins, such a method is known from the article “Refolding of Escherichia coli produced membrane protein inclusion bodies immobilized by nickel chelating chromatography”, Rogl et al. in FEBS Letters 432 (1998) 21-26.

A method for refolding of receptor protein is known from the article “Expression of an Olfactory Receptor in Escherichia coli: Purification, Reconstitution, and Ligand Binding” by Kiefer et al. in Biochemistry 35 (1996) 16077-16084.

Both publications are based on the problem that membrane proteins, to which receptors also belong, can be produced in large quantities with the help of expression vectors in bacteria, but that the protein produced, however, is not active. The protein is, namely, not integrated into the membrane, but, first, is present in a denatured state and has to be refolded into the native or active structure. The aggregates of “inactive” protein are designated in English literature as inclusion bodies.

For the membrane proteins Toc75 and LHCP, Rogl et al. describe a method in which N-Lauroylsarcosine is used as first detergent and Triton X-100® is used as second detergent. By exchanging the chaotrope for the mild detergent, refolding of the aggregated protein was induced.

According to Kiefer et al., a G-protein-coupled olfactory receptor was transformed into the active structure during the binding onto a nickel column by detergent exchange from N-Lauroylsarcosine to digitonin.

In both cases, it could be shown that the aggregated protein first existing in the form of inclusion bodies was, first, solubilized in a denaturing detergent and, then, by the detergent exchange described, transformed into its active structure, which was verified by corresponding binding measurements.

There is a great scientific and commercial interest in membrane proteins, in particular in receptors in native or active form, since membrane proteins are components of all biological membranes and impart to the specificity of different cellular membranes, they are particularly responsible for the exchange of substances and signals.

The specific recognition of a chemical compound by the corresponding receptor has e.g. the consequence that the target cell changes its physiological state. That is why receptors are the most important target molecules for drugs, approximately ¾ of all commercially available pharmaceuticals act on receptors, most of which, again, act on so-called G-protein-coupled receptors, which have in the human genome several hundreds of representatives.

For the development of specific antibodies, of drugs etc. it is, in view of the above, most desirable to have membrane proteins, in particular receptors in active or native structure available in large quantities. Since these proteins occur, in tissue, only in very small concentrations, it is necessary to use a system for recombinant over-expression of membrane proteins and receptors. For this purpose, on the one hand, in eukaryotic cells (cells of mammals or insects), functional protein can be produced, however, the systems are expensive, and the expression rates are low, which is also disadvantageous. Functional protein can be obtained via bacterial expression as well, the expression rate, however, is even lower than in eukaryotic expression.

In view of the above, the two publications mentioned at the outset describe methods, in which the protein is expressed in the inner part of the cell, where it, however, aggregates, and hence is not functionally available. The advantage of this method is that very large quantities of protein can be produced, Kiefer et al. report that up to 10% of cell protein and, thus, 100-10,000 times more protein than with other expression systems can be produced. The inclusion bodies produced in that way, which Rogl et al. have also reported about, must then first of all, be solubilized and, via the exchange of detergents already described at the outset, be transformed into their native or active structure.

Of course, commercial interest is not only directed to membrane proteins and receptors in their naturally existing sequence, rather, also partial sequences, homologous sequences, mutated sequences or derived sequences of membrane proteins and receptors are an object of this invention, as they allow, depending on functionality, not only insights into the structure of membrane proteins and receptors, but also a rational drug design.

In this context it should be mentioned that the DNA sequence of many receptors is known, such sequences are contained in the EMBL database. As these DNA sequences, in most cases, do not contain introns, the coding sequence can be produced by PCR from genomic DNA or by RT-PCR from mRNA. This DNA can then be cloned into a corresponding expression vector.

However, the structure of the translation product is unknown, so that providing proteins which are an object of the invention in sufficient quantity allows crystallization experiments etc. to further elucidate the structure.

It should also be mentioned that receptors expressed in eukaryotic and in bacteria can be distinguished by glycosylation. G-protein-coupled receptors, namely, possess on the N-terminus one or more glycosylation sites, which are modified in the endoplasmic reticulum or later in the Golgi apparatus with an oligosaccharide. Bacteria, in contrast, do not modify these sequences.

By treating a portion of the protein with N-glycosidase F or N-glycosidase A, the saccharide can be cleaved off, so that on an SDS gel a different extend of migration of the protein can be distinguished before and after this treatment, if the protein was expressed in eukaryotic cells. For bacterially expressed protein, no differences in the extend of migration can be distinguished.

Although the methods for the production of membrane protein or receptor protein that are described in the publications mentioned above lead to active structures, the methods described, according to the knowledge of the inventors of the present application here submitted, are insofar not satisfying, as the yield is low and the method is poorly reproducible.

SUMMARY OF THE INVENTION

In view of the above, it is an object of the present invention to improve the method mentioned above to reach—with good reproducibility—a high yield of the protein in active or native structure.

According to the invention, this object is achieved by selecting the second detergent from the group:

    • Alkyl-N,N-dimethylglycine (alkyl=C8-C16)
    • Alkylglycosides (alkyl=C5-C12, also branched-chained or cyclic alkyl rests, glycoside=all mono- and disaccharides), including Dodecylmaltoside (DDM)
    • Saccharide fatty acid ester (e.g. sucrosemonododecanoate)
    • Alkylthioglycosides (alkyl=C5-C12, also branched-chained or cyclic alkyl rests, glycoside=all mono- and disaccharides with S- instead of O-glycosidic bond)
    • Bile acids (cholate, deoxycholate) and derivatives (e.g. CHAPS, CHAPSO)
    • Glucamides (MEGA-8 to -10, HEGA)
    • Lecithins and lysolecithins (e.g. DHPC, C12-lysolecithin)
    • Alkyl-Phosphorylcholine (Alkyl=C10-C16).

The object underlying the invention is in that way completely achieved.

The inventors of the submitted application have, namely, recognized that the low yield and the lacking reproducibility in the known methods is to be assigned to the second detergent. Surprisingly, the yields were, namely, distinctively higher and the results were more reproducible, if the second detergent was selected from the group mentioned above. Further, it has to be taken care that the second detergent, in its final concentration, is above the critical micellar concentration. This cmc-value hence reflects, in principle, the solubility of a detergent in water. Above the cmc-value, the concentration of solved detergent-monomers is constant.

The cmc-values of some detergents are described in the publication “Detergents: An Overview” J. M. Neugebauer in Methods in Enzymology 182 (1990), pages 239-253.

In this connection, it is particularly preferred if alkyl-phosphorylcholine with a chain length of C10-C16 is used as said second detergent. The inventors have recognized that this detergent provides for a high yield of refolded protein.

According to own measurements of the inventors, the cmc-values for alkyl-phosphorylcholine with an alkyl rest of C12, C13, C14 or C16 are 500, 150, 50 or 5 μM.

In view of the fact that only few detergents are able at all to keep membrane proteins or even receptors stably in solution, it is the more surprising that alkyl-phosphorylcholine, the use of which for membrane receptors has not yet been described in literature, is even able to induce a refolding into the native structure. In the case of an adenosine receptor, the inventors were able to prove that the refolded receptor has native binding properties if alkyl-phosphorylcholine is used as second detergent. Also for other receptors, the refolding could be shown with one of the detergents from the group mentioned above.

In this procedure, it is preferred if the protein is produced in form of inclusion bodies in a cell line transformed with an expression vector, which vector carries a gene coding for said protein, the protein being preferably part of a fusion protein and, before or after said exchange of detergents, is cleaved off from said fusion protein.

The expression of the DNA sequence coding for the protein according to the invention as fusion protein has, in comparison with the direct expression without carrier protein, the advantage that the carrier protein protects the protein which is desired, but, however, unknown to the expression system, against degradation by proteases and may result in a higher expression level. In particular by using glutathione-S-transferase (GST) as carrier protein, the solubility of proteins being expressed in large quantities is increased in the host cell and isolation is facilitated. The carrier protein can, further, be used for purifying fusion proteins, if suitable antibodies are provided. The same applies for purification methods with affinity chromatography.

In this method, it is further preferred if the inclusion bodies are purified and solubilized by adding the first detergent, wherein the first detergent is selected from the group:

N-Lauroylsarcosine, Alkyl(C16)phosphorylcholine, Dodecylsulfate, other charged detergents or urea or guanidiniumchloride in combination with charged or uncharged detergents.

It is important in this connection that the conditions to bring the protein in solution are denaturing, so that they do not allow a formulation of the native structure.

In this method, it is altogether preferred if the second detergent is present in a folding buffer with mixed lipid/detergent micelles, wherein the folding buffer contains preferably the second detergent and phospholipid from a natural source, preferably a lipid extract of tissue, in which the protein occurs naturally.

In this method, it is advantageous that, in comparison with the use of pure detergent micelles, the yield of native protein can even be improved. The lipid extract of the tissue, in which the receptor occurs naturally, can also be simulated by using lipids with a similar composition or by mixing same.

With reference to the exchange of detergents it is preferred, if same is done by a dialysis or ultrafiltration method or by chromatographic methods or by diluting said solubilized protein in a buffer which contains said second detergent.

The methods described insofar concerning the exchange of detergents are exchangeable amongst each other and offer, each for itself, specific advantages with reference to the handling, the duration of the method and the reachable yield.

After said exchange of detergents, at least one conserved disulfide bridge must be formed in the protein, preferably by adding a mixture of oxidized and reduced glutathione.

The folded protein can further be incorporated in proteoliposomes, which are artificially produced vesicles and represent a functional unit. With the help of these deliberately produced proteoliposomes, certain processes on the membrane proteins/receptors can be selectively investigated.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 Elution profile of purified and solubilized voltage gated ion channel from Salmonella typhimurium (Vic) (A), and PAGE visualizing refolded active/native tetrameric Vic (B).

FIG. 2 Elution profile of purified and solubilized MJ ion channel from Methanococcus jannaschii (MJ) (A, C) and PAGE visualizing refolded active/native tetrameric MJ (B).

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT EXAMPLE 1 Production of an Expression Vector with cDNA for Receptor Protein

DNA sequences for several receptor proteins and also membrane proteins are in the EMBL database, in most cases, they do not have introns. With the help of primers, the required DNA can be produced via PCR from genomic DNA or via RT-PCR from mRNA.

This DNA is then cloned into an expression vector, which was constructed for the expression of a fusion protein. The carrier protein can be e.g. glutathione-S-transferase (GST), as described in the article by Kiefer et al. mentioned above, wherein a fusion protein was produced from the receptor OR5 and GST. The expression vector is transformed into a cell line which expresses the fusion protein. The protein is, in this procedure, not incorporated into the membrane, but exists at least partly aggregated in form of inclusion bodies in cytoplasm and is, thus, not correctly folded.

EXAMPLE 2 Isolation of Expressed Protein

The cDNA of one of the following receptors is, in-frame, inserted into the vector pGEX2a-c-His: A0 adenosine receptor from the shark Squalus acanthias, human beta-2-adrenergic receptor, human neuropeptide YY1 receptor, human neuropeptide YY2 receptor, human melanocortine-1 receptor, human oxytocin receptor, tetrameric voltage gated ion channel from Salmonella typhimurium (Vic), tetrameric MJ ion channel from Methanococcus jannaschii (MJ). This vector contains downstream of the Tac-promotor the sequence encoding glutathione-S-transferase and a subsequent thrombin cleavage site, followed by a polylinker sequence and, finally, six histidine codons and a stop codon.

The vectors are transformed into E. coli, e.g. strains BL21 or BLR. The protein expression is induced by adding IPTG, and the cells are harvested after further three hours. After lysozyme treatment and homogenization by sonication, the membranes and inclusion bodies are separated from the soluble proteins by centrifugation.

EXAMPLE 3 Solubization of the Protein

3.1 Voltage Gated Ion Channel (Vic)

The inclusion bodies are solubilized by sonification and by the addition of solubization buffer [25 mM Tris/HCl pH 8.5, 250 mM NaCl, 1 mM DTT, 1% Alkyl(C16) phosphorylcholine (C16-FOS-Cholin) (first detergent), 0.01 mg/ml FOLCH lipid]. Thrombine is added to this solution and incubation is performed over night at room temperature to cleave of the GST part of the construct. After that, insoluble cell debris is separated from the solubilized ion channel protein by centrifugation at 4° C.

Ni-NTA-Agarose (Qiagen) equilibrated in 50 mM HEPES/NaOH pH

7.5, 250 mM NaCl, 1% C16-FOS-Cholin (first detergent) is added to the supernatant. Incubation is performed for one hour at 4° C., wherein the receptor binds to the nickel matrix. Thereafter, the nickel matrix is filled into a chromatography column, washed two times with washing buffer [50 mM HEPES/NaOH pH 7.5, 250 mM NaCl, 0.1 mg/ml FOLCH fraction 1 brain lipid extract, 1 mM GSH and 1 mM GSSG, 1% C16-FOS-Cholin (first wash), 0.1% C16-FOS-Cholin (second wash)].

Vic protein is eluted with elution buffer (50 mM HEPES/NaOH pH 7.5, 500 mM NaCl, 300 mM imidazole pH 7.0, 0.02% C16-FOS-Cholin). Fractions of 5 ml each are collected. A representative elution profile is shown in FIG. 1A demonstrating purity of the obtained Vic protein.

Fractions containing Vic protein were pooled and concentrated to 10 mg/ml using Amicon Ultra 30 kDa ultrafiltration units.

3.2 MJ Ion Channel (MJ)

The inclusion bodies are solubilized by sonification and by adding solubization buffer [25 mM Tris pH 8.4, 250 mM NaCl, 1% N-Lauroylsarcosine (first detergent), 10 mM β-mercaptoethanol]. Thrombin is added to this solution and incubation is performed for 1-2 hours at 20° C. to allow cleavage of the receptor from GST. After that, insoluble cell debris is separated from the solubilized ion channel protein by centrifugation at 4° C.

The supernatant is added to Ni-NTA-Agarose (Qiagen) equilibrated in solubization buffer and incubated over night at 20° C., wherein the receptor binds to the nickel matrix. Thereafter, the nickel material is filled into a chromatography column and washed two times with washing buffer [20 mM Tris pH 8.4 (first wash), 25 mM Tris pH 7.4 (second wash), 250 mM NaCl, 1% N-Lauroylsarcosine (first detergent)].

EXAMPLE 4 Refolding of the Protein Into its Native and Active Form

4.1 Voltage Gated Ion Channel (Vic)

Vic protein is refolded via reconstitution in proteoliposomes (PLS). Therefore 2 mg of monomeric Vic protein obtained as described under 3.1 are mixed with 20 mg lipid mixture consisting of PEPG [3:1 mixture of PE (Phosphatidyl-ethanolamine, natural extract from sheep brain (Cephalin™) with PC (1-Palmitoyl-2-oleoylphosphatidylcholine] or DMPC (di-Myristoyl-phosphatidyl-choline, 14:0) suspended in TBS buffer (25 mM Tris/HCl pH 7.5, 125 mM NaCl) containing Dodecylmaltoside (DDM, second detergent) at a final concentration of 0.04%. The total volume of the mixture is about 12 ml. Incubation is performed for two hours at room temperature under gentle agitation, to achieve a thorough and homogenous mixture of the components.

In order to remove the second detergent 1 ml Calbiosorb per 10 mg DDM is added and incubation is performed on a rotator at 18° C. over night (during this process the PLS are formed). The Calbiosorb is then removed by filtration. A centrifugation step at 4° C. is performed and the pelleted PLS are resolubilized in TBS buffer for further use.

Functionality assays analyzing the conductance of the ion channel demonstrate the rectification and the conductance (165 pS) of the obtained Vic channel. These studies in turn prove that the receptor is present in its native and active structure. Furthermore, CD spectroscopy and gel filtration studies allow the suggestion that the Vic ion channel has the correct tetrameric structure and the correct a-helix portion.

4.2 MJ Ion Channel

Subsequent to the procedure as described under point 3.2 the Ni-NTA-Agarose column is washed with detergent exchange buffer [25 mM Tris pH 7.5, 250 mM NaCl, 1% DDM (second detergent, first wash) and 0.05% DDM (second detergent, second wash)] so that folding of the ion channel in its native/active form is induced.

Elution is performed by eluting the Ni-NTA-Agarose column with elution buffer [25 mM Tris pH 7.4, 250 mM NaCl, 300 mM imidazole and 0.05% DDM (second detergent)]. Fractions of 1 ml each are collected. MJ protein containing fractions are pooled, subjected to a Superdex 200 HR column and an elution profile is established as shown in FIG. 2A. Aliquots containing pure refolded tetrameric ion channel protein are pooled and concentrated to 10 mg protein/ml.

For the reconstitution into PLS 13 mg of monomeric purified MJ ion channel protein are added to 10 mg of a lipid mixture consisting of PEPG [a 3:1 mixture of PE (phosphatidylethanolamine—natural extract from sheep brain (Cephalin™)), and PC (1-Palmitoyl-2-oleoylphosphatidylcholine)] suspended in reconstitution buffer (25 mM Tris/HCl pH 8.4, 250 mM NaCl) containing 0.15% Dodecylmaltoside (DDM, second detergent) and 0.15% Di-Myristoyl-phosphatidylcholine. The total volume of the mixture is about 33 ml. Incubation is performed for a sufficient time at room temperature under gentle agitation, to achieve a thorough and homogenous mixture of the components.

In order to remove the second detergent 1 ml Calbiosorb per 10 mg DDM is added and incubation is performed on a rotator at 18° C. over night (during this process the PLS are formed). The Calbiosorb is then removed by filtration. A centrifugation step at 4° C. is performed and the pelleted PLS are resolubilized in reconstitution buffer.

The inventors have performed functionality assays with so obtained MJ ion channel protein reconstituted in PLS. In doing so, they could prove conductivity of the tetrameric MJ ion channel by means of electrophysiological examinations. Furthermore, CD spectroscopy and gel filtration studies allow the suggestion that the MJ ion channel has the correct tetrameric structure. Hereby it is demonstrated that the receptor is present in native and active structure.

EXAMPLE 5 Re-Solubization of the Protein from the PLS

5.1 Voltage Gated Ion Channel (Vic)

200 μl PLS obtained as described under 4.1 are added to 20 μl 10% DDM. The solution is rotated for one hour at 18° C. until the solution has become clear. The solution is centrifuged, the supernatant is subjected to Ni-NTA-Agarose and incubation is performed over night at 20° C. Thereafter, the nickel material is applied to a column, washed with TBS buffer pH 7.4 containing 0.05% DDM.

Elution takes place with 300 mM imidazole in TBS buffer pH 7.4 containing 0.05% DDM. Gel filtration is performed in the presence of 0.02% DDM. A polyacrylamide gel electrophoresis (PAGE) is performed as shown in FIG. 1B: left lane=marker, middle lane unfolded Vic protein, right lane folded Vic protein (tetramer{circumflex over (=)}active/native ion channel, and monomer).

5.2 MJ Ion Channel

200 μl PLS containing reconstituted MJ ion channel protein obtained as described under 4.2 are added to 20 μl 10% DDM. The solution is incubated for 30 min at 18° C. and subsequently subjected to a centrifugation step.

The supernatant is incubated with Ni-NTA-Agarose and incubated for further 20 to 30 min at 18° C. The nickel material is applied to a column and washed with washing buffer (25 mM Tris-HCl pH 8.4, 250 mM NaCl, 0.03% DDM). Elution is performed with elution buffer (25 mM Tris/HCl pH 8.4, 250 mM NaCl, 300 mM imidazole, 0.03% DDM). Fractions of 600 μl each are collected. The resolubilized proteins are separated by PAGE as shown in FIG. 2B: left lane=marker, three lanes on the right demonstrating native/active tetrameric MJ ion channel protein.

Claims

1. A method for production of proteins folded into their native or active structure, said proteins being from the group of membrane receptors, excluding G-protein-coupled receptors, comprising:

providing a protein from the group of membrane receptors solubilized in a first detergent, and
exchanging said first detergent for a second detergent, to induce folding of said protein into its native or active form, wherein said second detergent is selected from the group consisting of: Alkyl-N,N-dimethylglycine (alkyl=C8-C16); Alkylglycosides (alkyl=C5-C12, also branched-chained or cyclic alkyl rests, glycoside=all mono- and disaccharides), including Dodecylmaltoside (DDM); Saccharide fatty acid ester (e.g. sucrosemonododecanoate); Alkylthioglycosides (alkyl=C5-C12, also branched-chained or cyclic alkyl rests, glycoside=all mono- and disaccharides with S- instead of O-glycosidic bond); Bile acids (cholate, deoxycholate) and derivatives (e.g. CHAPS, CHAPSO); Glucamides (MEGA-8 to -10, HEGA); Lecithins and lysolecithins (e.g. DHPC, C12-lysolecithin), and Alkyl-Phosphorylcholine (Alkyl=C10-C16).

2. The method of claim 1, wherein said membrane receptor is an ion channel.

3. The method of claim 1, wherein said second detergent is provided in a folding buffer with mixed lipid/detergent micelles.

4. The method of claim 3, wherein said folding buffer contains said second detergent and phospholipid from a natural source.

5. The method of claim 4, wherein said phospholipid is a lipid extract of tissue in which said protein occurs naturally.

6. The method of claim 1, characterized in that said exchange of detergents is done by a dialysis- or ultrafiltration method.

7. The method of claim 1, characterized in that said exchange of detergents is carried out via a chromatographic method.

8. The method of claim 1, wherein said exchange of detergents is carried out by diluting said solubilized protein in a buffer which contains said second detergent.

9. The method of claim 1, wherein after said exchange of detergents at least one disulfide bridge is formed in said protein.

10. The method of claim 9, wherein the disulfide bridge is formed by adding a mixture of oxidized and reduced glutathione.

11. The method of claim 1, wherein said folded protein is incorporated in proteoliposomes.

12. The method of claim 1, wherein said protein is produced in form of inclusion bodies in a cell line transformed with an expression vector which carries a gene coding for said protein.

13. The method of claim 12, further comprising solubilizing said inclusion bodies by adding said first detergent.

14. The method of claim 1, wherein said protein is part of a fusion protein and is cleaved off from said fusion protein.

15. The method of claim 1, wherein said first detergent is selected from: N-Lauroylsarcosine, Alkyl(C16)phosphorylcholine, Dodecylsulfate, other charged detergents or urea or guanidiniumchloride in combination with charged or uncharged detergents.

16. The method of claim 1, wherein said second detergent has a concentration that is above its critical micellar concentration.

17. The method of claim 1, wherein said second detergent is alkyl-phosphorylcholine with a chain length of C10-C16.

Patent History
Publication number: 20060058509
Type: Application
Filed: Aug 31, 2005
Publication Date: Mar 16, 2006
Inventors: Hans Kiefer (Tubingen), Klaus Maier (Stuttgart)
Application Number: 11/217,124
Classifications
Current U.S. Class: 530/350.000
International Classification: C07K 14/47 (20060101);