The invention relates to a method for expression of membrane proteins and to a method for searching interacting agents.
Membrane proteins are essential components of biological membranes. For a long time it is known, that these proteins are responsible for an immense number of biological activities. For example the different transport mechanisms found in nature are due to the so-called transport proteins of the membrane. An especially important point is signal transduction, which occurs via the membrane of cells and/or organelles, for example. In general, membrane proteins mediate the most important signal transduction phenomena. Therefore, the investigation of membrane proteins is of special interest since long time ago. This is especially true because the biologically phenomena regarding transport and signal transduction relate to medical and especially to pharmaceutical aspects, for example regarding the delivery of drugs to the point of use. Membrane proteins are very important targets for the development of pharmaceutically active agents. As one example, the membrane anchored adenylyl cyclase plays a key role in the hormone mediated signal transduction wherein second messengers like cyclic adenosine monophosphate (cAMP) are involved. In this signal cascade an hormone-binding receptor activates the so-called G-protein which itself activates the adenylyl cyclase, which is anchored in the plasma membrane. The activated adenylyl cyclase catalyses the formation of cAMP from adenosine triphosphate in the presence of Mg2+. cAMP is a prominent second messenger in the cell, which is responsible for the regulation of very many biological processes.
There are several pathological situations in which this adenylyl cyclase signal cascade is disturbed. One example is the disease cholera, which is mediated by the bacterium Vibrio cholera. This bacterium produces the so-called cholera toxin, which modulates the above-mentioned G-protein in such a manner that the following component in the signal cascade, the adenylyl cyclase, is constantly activated. This leads to an enormous level of cAMP, which results in the pathological symptoms of the disease cholera. This example shows that there is a special interest in analysing and consequently influencing the activity of membrane proteins, in particular members of the adenylyl cylcase family.
One main problem when investigating membrane proteins is the real impossibility to express such proteins in cellular systems by genetic engineering methods in such a way that the proteins are structural and functional comparable with the native membrane proteins. Consequently the common expression systems are not suited for an investigation of membrane proteins, because the proteins, e.g. enzymes, have lost their native characteristics, e.g. their activity, nothing to say about the structural deficiencies of the recombinant membrane proteins obtained by hitherto methods. Even if membrane proteins may be expressed in low yields by bacterial expression systems they are not inserted into the cell membrane but mostly located to inclusion bodies and intracellular structures, due to improper folding. Correct insertion into cell membranes is crucial for membrane protein function, since many membrane proteins get their native structure only in the environment of the correct phospholipid-bilayer, other associated membrane proteins or glycolipids, only to mention a few examples. Furthermore with the commonly used molecular biological methods it is not possible to obtain sufficient yields of recombinant membrane proteins, which are necessary for crystallographic structural analysis, for example. In this field yields in the range of mg are necessary.
Therefore, the present invention has the object to provide a method for expression of membrane proteins, which are structural and functional comparable with the native proteins. Furthermore the inventive method should be suited for comparatively high yields of recombinant protein, for example with respect to crystallography. This object is solved by a method as described in claim 1. The dependent claims 2 to 9 relate to preferred embodiments of the inventive method. Claims 10 to 13 concern fused nucleotide sequences and fused proteins, respectively. Claim 14 describes a kit and in claims 15 to 21 methods for searching active agents are elucidated. Claim 22 relates to an active agent. The wording of all claims is hereby made to the content of the specification.
The inventive method for expression of membrane proteins is characterized by the fusion of the nucleotide sequence or a portion of this sequence with a nucleotide sequence coding for an adenylyl cyclase of Mycobacterium tuberculosis or a portion of this sequence or a sequence, which is at least about 70% identical so such sequences. In a second step the fused nucleotide sequence is introduced into an organism and/or a living cell, wherein the introduced sequence is expressed by common procedures known by an expert in the art.
According to the present invention the fusion of the sequence for the adenylyl cylclase with any membrane protein leads to the expression of structural and functional correct recombinant proteins, in particular in high-yield-expression systems as Escherichia coli. By the inventive method the recombinant membrane proteins are targeted to the membrane displaying their characteristics like enzymatic activity comparable with the native protein. The term “protein” comprises proteins as well as peptides.
The invention comprises the use of the diverse open reading frames of Mycobacterium tuberculosis, which relate to adenylyl cyclase activity in the inventive manner. The diverse sequences are already described in the complete genomic analysis of Mycobacterium tuberculosis (Cole et al., 1998, Nature, 318, 618-624). Especially preferred are the genes Rv2435c and Rv1625c. In an especially preferred embodiment of the invention the sequence Rv1625c (SEQ ID NO 1) is used.
According to the invention the complete sequence as outlined in SEQ ID NO 1 or in SEQ ID NO 2 is used in order to form a fusion product with the sequence coding for any membrane protein of interest. In an especially preferred embodiment of the invention only a portion or several portions of said sequence are used. Especially preferred are portions of the sequence, which code for one ore more parts of the N-terminal domain of the adenylyl cyclase. The N-terminal domain of the adenylyl cyclase comprises six transmembrane spans, which are especially suited in order to target the membrane protein of interest to the membrane in the expression system. According to the invention sequences are used which code for one or more of the transmembrane spans or parts thereof. Furthermore it is preferred to use repeats of portions of the sequences, especially of portions of the sequence coding for the N-terminal domain of the adenylyl cylase.
According to the inventive method it is possible to express any membrane protein wherein the recombinant protein maintains essentially the structural and functional characteristics of the native membrane protein. The membrane protein could be of prokaryotic or eukaryotic origin. In a preferred embodiment of the invention the membrane proteins are enzymes. In an especially preferred embodiment of the invention the membrane proteins are adenylyl cyclases, especially mammalian adenylyl cyclases. Adenylyl cyclases play a crucial role in signal transduction as outlined above. Therefore, there is a special interest in the investigation of these components of signal pathways, especially with respect to the development of pharmaceutically active drugs.
The fusion of the sequence coding for the membrane protein of interest and the nucleotide sequence of Mycobacterium tuberculosis or portions of this sequence is performed by known molecular biological methods. The transfer of the sequence is performed by common methods known to an expert in the art. Advantageously, the introduction of the fused nucleotide sequence is performed by the use of an appropriate vector. In general, all common vectors are suited, for example DNA vectors, viral and retroviral vectors. Furthermore the invention comprises transfer methods like liposome mediated gene transfer or electroporation, for example.
In a preferred embodiment of the inventive method the organism, in which the fused nucleotide sequence is introduced and afterwards expressed, is a prokaryotic organism, especially a bacterium. In an especially preferred embodiment the organism is Escherichia coli. Escherichia coli is a well-known bacterium widely used in laboratory practice. Bacteria like Escherichia coli are easy to cultivate in large amounts. It is therefore especially suited in the inventive manner, for example in order to obtain high yields of recombinant protein. Nevertheless Escherichia coil is only one example of suited organisms.
In another preferred embodiment of the invention the organism, in which the fused nucleotide sequence is introduced and afterwards expressed, is a eukaryotic organism or a eukaryotic cell. In an especially preferred embodiment the eukaryotic cell is a mammalian cell. Advantageously cell cultures are used in the inventive manner. Nevertheless all cells and organisms of eukaryotic origin as well as other organisms like fungi are suited. Especially preferred is the use of Trypanosoma leishmanii.
One main advantage of the inventive method is, that the recombinant membrane protein can be effectively expressed in prokaryotic as well as in eukaryotic systems. This is independent of the origin of the membrane protein of interest. This is an especially surprising feature of the invention, because up to now, it was very difficult to express e.g. a eukaryotic protein in a prokaryotic system like Escherichia coli with sufficient yields in protein activity. As outlined above membrane proteins are particularly difficult to handle. It was not possible to express for example a prokaryotic membrane protein in an active form in another prokaryotic organism, nothing to say of the expression of an active form of a eukaryotic membrane protein in a prokaryotic organism. Now it is possible by the use of the inventive method to provide a structural and functional active membrane protein of prokaryotic origin in a eukaryotic organism and vice versa (so-called heterologous expression).
Furthermore the invention comprises fused nucleotide sequences. Theses sequences comprise a) at least a portion of a nucleotide sequence coding for an adenylyl cyclase of Mycobacterium tuberculosis and/or at least a portion of sequence, which is at least about 70% identical to such sequences and b) at least a portion of a nucleotide sequence coding for any membrane protein. Preferably the sequence coding for the adenylyl cyclase of Mycobacterium tuberculosis is Rv1625c and/or Rv2435c. Nevertheless other sequences as outlined above are also possible. In a preferred embodiment the membrane protein is an enzyme. Preferably the fused sequence comprises further components. In an especially preferred embodiment the inventive fused sequences are integrated in an appropriate vector, for example a DNA vector, which is suited for an introduction of the sequence into an appropriate host organism.
Additionally the invention comprises fused proteins. Theses proteins comprise a) at least a portion of a an adenylyl cyclase of Mycobacterium tuberculosis and/or at least a portion of an amino acid sequence at least about 70% identical to an adenylyl cylclase of Mycobacterium tuberculosis and b) at least a portion of a any membrane protein. In a preferred embodiment the membrane protein is an enzyme. The fused protein could be a protein expressed in an organism. In a preferred embodiment the fused protein is purified at least partly. Preferably the fused protein is suited for investigations of crystallographic structure.
Another aspect of the invention is a kit for expression of membrane proteins. This kit comprises at least one or more portions of the nucleotide sequence Rv1625c (SEQ ID NO 1) and/or Rv2435c (SEQ ID NO 2) of Mycobacterium tuberculosis. Regarding further details of said nucleotide sequence reference is made to the above description. Preferably said sequence is integrated in an appropriate vector. In another embodiment of the kit it comprises further components for performing an expression of membrane proteins. These components are e.g. means for the transfer of the nucleotide sequence, appropriate host cells, buffers and/or media.
Furthermore the present invention comprises a method for searching active agents, which interact directly or indirectly with adenylyl cyclases. According to this inventive method an adenylyl cyclase of Mycobacterium tuberculosis is expressed in an organism and/or a living cell. Preferably the expression of said enzyme is obtained by introduction of an appropriate nucleotide sequence into a host organism. Further details of this procedure can be taken form the examples. In a second step of the inventive method a potential active agent is brought into contact with the adenylyl cyclase and finally the influence of the active agent on the adenylyl cylclase is analysed.
As outlined above adenylyl cyclases play a crucial role in signal transduction. Therefore there is a special interest in investigating these membrane-anchored enzymes and to find substances, which are able to influence these important proteins. Up to now it was not possible to express these enzymes in a functional manner in recombinant cellular systems. Surprisingly by the use of the inventive method, wherein the adenylyl cylclase of Mycobacterium tuberculosis is expressed, a system with recombinant enzyme is provided, wherein the enzymes maintains essentially its native characteristics. This system is especially suited firstly to investigate adenylyl cyclases in general and secondly to find substances, which influence these enzymes. Such influencing substances, so-called active agents, are e.g. useful for the development of pharmaceutically compositions for the treatment of diseases, which are connected, with a disturbance of the signal cascade involving adenylyl cyclases. One example of such diseases is possibly tuberculosis.
The different nucleotide sequences of Mycobacterium tuberculosis coding for adenylyl cyclases are suited for this aspect of the invention. In an especially preferred embodiment of this aspect of the invention the nucleotide sequence coding for the adenylyl cylcase is the gene Rv1625c and/or Rv2435c or portions thereof. In general, all common methods for introducing nucleotide sequences into host cells are possible. In an especially preferred embodiment the nucleotide sequence is introduced into the host organism by the help of an appropriate vector, e.g. a DNA vector or retroviral vector. The host organism can be a prokaryotic organism like Escherichia coli for example, or a eukaryotic cell or organism. Advantageously the host organism is a mammalian organism or a mammalian cell. Preferably the host cells are cultivable cells, as often used in laboratory practice.
In the inventive method for searching active agents generally any substance is testable with respect to its influence on adenylyl cyclases. For example whole libraries of natural and/or artificial substances can be analysed in a screening based on the inventive method. In an especially preferred embodiment the potential active agents are selected by the way of computer aided modelling. In a preferred embodiment this method is automated in a way known to experts in the art in order to enable a high through-put-screening.
The influence of the potential active agent on the adenylyl cylase is analysable by different methods. One possibility is to analyse a direct interaction of the membrane anchored enzyme and the potential agent, for example by detection of complex formation. In another especially preferred embodiment the influence is analysed by an altered enzyme activity of the adenylyl cyclase. In this connection the enzyme activity is preferably measured before and after adding the potential agent. Details of the measurement of enzyme activity are described in the examples.
Finally the invention comprises active agents, which interact directly or indirectly with adenylyl cyclases. These active agents are obtained by a method as described above.
The described features of the invention and further features result in greater detail from the following description of examples in combination with the figures and the sub claims. The different features could be realized alone or in combination with each other.
In the figures it is shown
1. Introduction
In 1998, the complete genome sequence of Mycobacterium tuberculosis H37Rv was reported (Cole et al., 1998). Thus, genes of interest are now accessible by PCR using specific primers and genomic DNA as a template. In the M. tuberculosis genome 15 open reading frames have been identified which probably code for functional class III adenylyl cyclases (AC) (Cole et al., 1998; McCue et al., 2000). These cyclase isozymes belong to evolutionary segregated branches. Nine are predicted to be similar to ACs found in streptomyces, and four appear to be similar to ACs identified in bacteria such as myxobacterium Stigmatella aurantiaca (McCue et al., 2000). Two genes, Rv1625c and Rv2435c, code for proteins which are grouped with the mammalian ACs on a branch separate from other prokaryotic cyclases. The predicted protein of Rv1625c has a molecular mass of 47 kDa. It consists of a large N-terminal membrane domain which is made up of six transmembrane spans, and of a single C-terminal catalytic domain (Tang and Hurley, 1998). As such the predicted protein topology corresponds exactly to one half of a mammalian membrane-bound AC which is a pseudoheterodimer composed of two highly similar domains linked by a peptide chain designated as C1b (Tang and Hurley, 1998; see model in
2. Materials and Methods
2.1. Recombinant cDNAs
Genomic DNA from M. tuberculosis was a gift of Dr. Boettger, Medical School, Hannover, Germany. The mycoAC gene Rv1625c was amplified by PCR using specific primers and genomic DNA as a template. A BamHI restriction site was added at the 5′-end and a SacI site at the 3′-end. The DNA was inserted into either the multiple cloning site of pQE30 (Qiagen) with the N-terminal addition of a MRGSH6GS-peptide sequence or pQE60 which added a C-terminal His-tag. In conjunction with specific primers this clone served as a template to generate by PCR three DNA constructs comprising the catalytic C-terminal starting at D204 (mycoAC204-443), E213 (mycoAC213-443) or A221 (mycoAC221-443). After partial restriction of the products with BamHI and SacI the fragments were ligated into pQE30.
Two catalytic mycoAC204-443 segments were linked (C- to N-terminal) via the insertion of a DNA sequence coding for the peptide linker TRAAGGPPAAGGLE using conventional molecular biology techniques. Thus a pQE30 plasmid was generated which coded for the protein MRGSH6GS-mycoAC204-443-TRAAGGPPAAGGLE-mycoAC204-443, abbreviated (mycoAC204-443)2.
Single amino acid mutations in the catalytic domain of the mycoAC (D256A, K296A, D300A, D365A and R376A) were introduced by site-directed mutagenesis using respective PCR primers and the DNA coding for mycoAC204443 as a template. Nearby located restriction sites were used as appropriate. The fidelity of all constructs was verified by double-stranded sequencing. The sequences of all primers are available on request.
2.2. Expression and Purification of Bacterially Expressed Proteins
The constructs in the pQE30 expression plasmid were transformed into E. coli BL21 (DE3)[pREP4]. For the holoenzyme, the cya− strain DHP1[pREP4] was used. Cultures were grown in Lennox L broth at 25° C. containing 100 mg/liter ampicillin and 25 mg/liter kanamycin and induced with 30 μM isopropyl-β-D-thio-galactopyranoside at an A600 of 0.4. Bacteria were harvested after 3 h, washed once with 50 mM Tris-HCl, 1 mM EDTA, pH 8 and stored at −80° C. For purification frozen cells were suspended in 20 ml of cell lysis buffer (50 mM Tris-HCl, 50 mM NaCl, 10 mM β-mercaptoethanol, pH 8) and sonicated (3×10 s). Cell debris was removed by centrifugation (48,000×g, 30 min). To the supernatants 200 μl of nickel-NTA slurry (Qiagen) was added. After gentle rocking for 30 min at 0° C., the resin was poured into a column and washed (2 ml/wash; wash buffer A: 50 mM Tris-HCl, pH 8, 10 mM β-mercaptoethanol, 2 mM MgCl2, 400 mM NaCl, 5 mM imidazole; wash buffer B: wash buffer A with 15 mM imidazole; wash buffer C: wash buffer A with 10 mM NaCl and 15 mM imidazole). Proteins were eluted with 0.5 to 1 ml of buffer C containing 150 mM imidazole. Because the mycoAC was inhibited by more than 1 mM imidazole, the eluates were dialyzed overnight against 50 mM Tris-HCl, pH 7.5, 2 mM β-mercaptoethanol, 10 mM NaCl and 20% glycerol. The purified proteins could be stored in dialysis buffer at 4° C.
2.3. Expression of mycoAC in HEK293 Cells
For expression of the mycoAC in HEK293 cells the coding region in pQE60 (Qiagen) containing an RGSH6 sequence at the carboxy-terminal was excised from pQE60 with NcoI and HindIII, filled in and cloned into the EcoRV site of pIRES1neo (Clontech). HEK293 cells were grown at 37□C in minimum essential medium (Gibco-BRL # 31095-029) containing 10% fetal calf serum and 20 μg/ml gentamicin sulfate. For transfection, the calcium phosphate method was used with 10 μg plasmid DNA/10 cm dish (Kingston et al., 1992). Cells were selected with 800 μg/ml G418 starting 3 days after transfection. After 10 days, the surviving cells were propagated in the presence of 400 μg/ml G418.
For membrane preparation, cells were washed twice with phosphate-buffered saline (PBS; 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4, pH 7.4), suspended and lysed in 250 μl buffer (20% glycerol, 20 mM Tris-HCl, pH 7.5, 1% thioglycerol, 1 mM EDTA and 1 mM benzamidine; two freeze-thaw cycles). The suspension was diluted with 4 volumes of buffer (without glycerol) and sonicated (3×10 s, microtip setting 4). After removal of nuclei and debris (6,000×g) membranes of HEK293 cells were pelleted at 50,000×g for 30 min.
2.4. Adenylyl Cyclase Assay
Adenylyl cyclase activity was measured for 10 min in a final volume of 100 μl (Salomon et al., 1974). The reactions contained 22% glycerol, 50 mM Tris-HCl, pH 7.5, either 2 mM MnCl2 or 20 mM MgCl2, the indicated concentrations of [α-32P]ATP (25 kBq) and 2 mM [2,8-3H]-CAMP (150 Bq). 3 mM creatine phosphate and 1 unit of creatine kinase were used as an ATP-regenerating system when assaying crude extracts. To determine kinetic constants, the concentration of ATP was varied from 10 to 550 μM with a fixed concentration of 2 mM Mn2+ or 20 mM Mg2+. Assays were preincubated at 30° C. for 5 min (37° C. with HEK293 membranes) and the reaction was started by the addition of enzyme. Usually less than 10% of the ATP was consumed at the lowest substrate concentrations.
2.5. Western Blot Analysis
Protein was mixed with sample buffer and subjected to SDS-PAGE (15%). Proteins were blotted onto PVDF membranes and sequentially probed with a commercial anti-RGS-His4 antibody (Qiagen) and with a 1:5000 dilution of a peroxidase-conjugated goat-anti-mouse IgG secondary antibody (Dianova). Peroxidase detection was carried out with the ECL-Plus kit (Amersham-Pharmacia). An affinity-purified, highly specific antibody against mycoAC204-443 raised in rabbits was employed to detect AC expression in homogenates from E. coli and HEK293 cells.
2.6. Cross-Linking with Glutaraldehyde
The cross-linking of purified mycoAC204-443 was carried out in 30 μl at 22° C. in 50 mM Na2HPO4, 10 mM NaCl, 2 mM MgCl2, ±850 μM ATP, 20 mM glutaraldehyde and 20% glycerol, pH 7.4. After 60 min the reaction was quenched by with 5 μl of 1 M Tris-HCl, pH 7.5. After addition of 5 μl sample buffer, proteins were resolved by 15% SDS-PAGE and detected in a Western blot using the anti-RGS-His4 antibody.
3. Results
The gene Rv1625c from Mycobacterium tuberculosis codes for a membrane-anchored adenylyl cyclase corresponding to exactly one half of a mammalian adenylyl cyclase. An engineered, soluble form of Rv1625c was expressed in E. coli. It formed a homodimeric cyclase with two catalytic centers. Amino acid mutations predicted to affect catalysis resulted in inactive monomers. A single catalytic center with wild-type activity could be reconstituted from mutated monomers in stringent analogy to the mammalian heterodimeric cyclase structure. The proposed existence of supramolecular adenylyl cyclase complexes was established by reconstitution from peptide-linked, mutation-inactivated homodimers resulting in pseudo-trimeric and -tetrameric complexes. The mycobacterial holoenzyme was successfully expressed in E. coli and mammalian HEK293 cells, i.e. its membrane-targeting sequence was compatible with the bacterial and the eukaryotic machinery for processing and membrane insertion. The membrane-anchored mycobacterial cyclase expressed in E. coli was purified to homogeneity as a first step toward the complete structural elucidation of this important protein. As the closest progenitor of the mammalian adenylyl cyclase family to date the mycobacterial cyclase probably was spread by horizontal gene transfer.
3.1. Sequence Analysis of the Rv1625 Adenylyl Cyclase
The M. tuberculosis gene Rv1625c (Genbank accession number AF017731) codes for an AC (mycoAC) with six putative transmembrane-helices as a membrane anchor and a single catalytic domain (M and C;
3.2. Characterization of a Soluble Mycobacterial AC Homodimer
Three cytosolic segments of different lengths to generate a soluble, enzymatically active catalytic domain from the mycoAC (Tang and Gilman, 1995; Whisnant et al., 1996; Yan et al., 1996) were constructed. The N-terminally His-tagged constructs which started at D204, i.e. at the calculated exit of the last transmembrane span, at E213 and at A221 (marked in
The specific activity of mycoAC204-443 was dependent on the protein concentration. From 50 to 300 nM it increased five-fold suggesting the formation of a homodimer (
The kinetic properties of mycoAC204-443 and (mycoAC204-443)2 were analyzed using Mn2+ or Mg2+ (
3.3. Generation of an Active Heterodimeric from a Homodimeric AC
The crystal structure of the mammalian AC heterodimer predicts which amino acids of the mycoAC ought to be involved in catalysis (Liu et al., 1997; Sunahara et al., 1998; Tesmer et al., 1997; 1999; Yan et al., 1997; Zhang et al. 1997). In
In the soluble mycoAC204-443 D256, K296, D300, D365 and R376 to alanines were individually mutated. The constructs were expressed in E. coli and purified to homogeneity (
In contrast AC reconstitution to almost wildtype levels was encountered when mixing mycoAC204-443R376A with either mycoAC204-443D256A or mycoAC204-443D300A because these amino acids contribute to the catalysis from different protein domains very much like in a C1/C2 complex. A protein dose-response curve of mycoAC204-443R376A showed that it had no detectable AC activity up to 1 μM (
3.4. Evidence for Active Pseudo-Trimeric and -Tetrameric AC Species
The membrane-bound mammalian ACs are pseudoheterodimers in which the C1 and C2 domains are derived from an intramolecular association. Crystallographically, a tetrameric C2 structure has been reported (Zhang et al., 1997). The relevance of this tetramer remained unresolved because it has not been possible to experimentally examine the possibility of an intermolecular association, i.e. the formation of catalytic centers by an association of C1 and C2 domains originating from different AC proteins. With the available mycoAC mutant proteins such a possibility could be tested. Mutated homodimers linked by a tetradekapeptide chain were generated, i.e. mycoAC204-443R376A-TRAAGGPPAAGGLE-mycoAC204-443R376A (R376A-L-R376A), D300A-L-D300A and R376A-L-D300A. The constructs were expressed in E. coli and affinity purified to homogeneity (
Then it was investigated whether the R376A-L-R376A could be competed by the mycoAC204-443D300A monomer or even the D300A-L-D300A dimer with reconstitution of catalytic sites. 36 nM R376A-L-R376A which is inactive, was partially reactivated in a dose-dependent fashion by increasing amounts of mycoAC204-443D300A (
3.5. Expression of the Membrane-Bound mycoAC in E. coli and HEK293 Cells
M. tuberculosis carries the information for a mammalian type AC which requires membrane targeting. It was e attempted to express the mycoAC holoenzyme, i.e. including its huge membrane anchor, in E. coli as well as in mammalian HEK293 cells. So far no bacterial AC has been expressed in a mammalian cell and the bacterial expression and successful membrane targeting of a mammalian membrane-bound AC holoenzyme has not been reported. For expression in E. coli, the mycoAC ORF was cloned into pQE30 adding an N-terminal His6-tag. After transformation and induction a bacterial membrane preparation yielded a high AC activity (19 nmol cAMP/mg·min−1) indicating expression and successful dimerization in the bacterial membrane. This was demonstrated by a Western blot because a substantial amount of the protein remained dimerized during SDS-PAGE (
The mycobacterial protein could also be functionally expressed in stably transfected mammalian HEK293 cells. HEK293 membranes had an activity of 65 nmol cAMP/mg·min−1 compared with 24 pmol cAMP/mg·min−1 in the vector control. The Km for Mn-ATP was 65 μM. The robust expression of mycoAC in HEK293 cells was also detected by a Western blot using an antibody against the cytosolic portion of the protein (
3.6. Expression of Chimeric Adenylyl Cyclases in e. coli and HEK 293 Cells
To demonstrate functional expression of membrane proteins in different expression systems, chimeric adenylyl cyclases comprising parts of mycoAC and rabbit AC type V were constructed. A schematic representation of the chimeras is illustrated in
Briefly, in construct #88 the N-terminal membranous targeting sequence from mycoAC (MycoN) was used to transport the chimera to membranes. The first membrane anchor (M1) of rabbit type V AC (VM1) was fused C-terminally to MycoN. The resulting construct was investigated for activity. It was observed that construct #88 shows 5-fold increased activity compared to background (vector control), indicating that mammalian (rabbit type V AC) membrane anchor can be expressed in bacterial (E. coli) expression systems, when fused to Rv1625c. The membrane fractions of the resulting chimera were positive in Western-Blotting (50 kDa band). These results clearly demonstrate that mammalian membrane proteins can be expressed functionally (5-fold increased activity) and correct targeted (to membranes) in bacteria (E. coli).
In construct #113 the transmembrane regions TM 3-6 of Rv1625c were exchanged by the corresponding regions TM 3-6 of rabbit type V AC. The resulting chimera showed 4-fold increased adenylyl activity compared to background. The membrane fractions were positive in Western-Blotting (50 kDa band). These results demonstrate, that also functional parts of membrane proteins can be expressed in E. coli.
In construct #114, only the sixth transmembrane region of rabbit type V AC first membrane anchor was fused and the corresponding membrane anchor of mycoAC was deleted. Also in this chimera, Western-Blotting was positive and the activity of the chimera was increased 10-fold compared to background. These results demonstrate that also single functional membrane regions (TMs) of membrane proteins can be expressed in E. coli.
The following two contruct #102 and #103 were expressed, as mentioned-above, in mammalian (HEK 293; human embryonic kidney) cells.
In construct #103, the catalytic domains C1a and C2 of mammalian type V AC were replaced by the corresponding catalytic domains of Rv1625c (CMyco). The chimera was active (10-fold increased activity compared to background) and again positive in Western-Blotting.
In construct #102 the N-terminus and M2-membrane anchor (C-terminal loop consisting of 6 transmembrane regions) of rabbit type V AC (VM2) were fused with Cmyco as read-out. The activity of this chimera was increased 10-fold compared to background and in Western-Blots the corresponding band was detected. From these results it is clear, that with CMyco as read-out a native expression of VM2 is possible, although according to the publication of Cooper et al. (Current Biology, 11, 185-190, 2001) VIIIM2 (of type VII AC) is not transported out of the endoplasmatic reticulum. According to said publication, VIIIM2 builds dimeres/heteromultimeres. The presenting results demonstrate functionality of VM2. Furthermore, the results observed with the constructs #102 and #103 demonstrate that the catalytic domains of mycoAC can be used as read-out for mammalian AC membrane anchors in bacteria (E. coli) as also in mammalian (HEK 293) cells.
Summarizing the results observed by the chimeric adenylyl cyclases it can be noted that (mammalian) membrane enzymes can be fused to mycoAC for functionally expression in different systems. Enzymes which can be expressed by this method are for instance GPCR (G-protein-coupled receptors), the seven transmembrane regions or parts of these regions, other receptors or parts of them (e.g. membrane associated ligand binding regions), ion channels or parts of them (e.g. membrane associated ligand binding regions), transporters and porines or parts of them (e.g. membrane associated ligand binding regions) and electron transfer proteins or parts of them (e.g. membrane associated ligand binding regions).
4. Discussion
In the mycobacterial genome, 15 putative ACs are detected (Cole et al., 1998; McCue et al., 2000). This unique situation implies that the resulting signal transduction modalities using cAMP as a second messenger are of great importance to the tubercle bacillus. Although it has been reported that in macrophages with ingested mycobacteria cAMP levels are elevated and the phagosome-lysosome fusion is impaired, at this point it is not known in which function any of the mycobacterial cyclases may be required for pathogenesis (Lowrie, 1978). The protein product of Rv1625c was investigated because of its surprising topological identity and sequence similarity to mammalian membrane-bound ACs. The sequence-derived insights are now covered by the biochemical properties of the expressed protein. Dimerization was a prerequisite and sufficient for formation of the catalytic center. The molecular details were in agreement with the crystallographic description of a mammalian heterodimeric catalytic fold as evident from the studies with single amino acid mutants and from the reconstitution experiments. Because the mycoAC most likely is a symmetric homodimer, two identical and equivalent catalytic sites are generated. It is assumed that the measured activity represented that of a basal state. This leaves the question of its regulation in the bacillus to be resolved. One possibility is that the mycoAC uses intermediary proteins like the mammalian ACs. The presence in mycobacteria of proteins resembling the α, β, and γ-subunits of mammalian G-proteins has been suggested (Shankar et al., 1997), yet an evaluation of the mycobacterial genome has not turned-up proteins with evident structural similarities. Therefore, the molecular nature of these proteins remains to be defined. A plausible regulatory input to the membrane-anchored mycoAC may come from lipids of the complex mycobacterial cell envelope because they may affect the monomer/dimer ratio and thus AC activity in the pathogen.
In the mammalian heterodimer, forskolin binds to the catalytic core at the opposite end of the same ventral cleft that contains the active site. Thr410 and Ser942 (rat type II AC numbering) are implicated in forskolin binding (Tesmer et al., 1997). The correspondingly located amino acids in the mycoAC are Asp300 and Asn372 (
Several experimental observations have posed the question whether the pseudoheterodimeric mammalian AC might exist as a true dimer. A tetrameric x-ray structure was observed with the cytosolic C2 segment (Zhang et al., 1997). Further, target-size (Rodbell et al., 1981) and hydrodynamic analysis (Haga et al., 1977; Yeager et al., 1985) showed possible monomer/dimer transitions of the mammalian pseudoheterodimers and immunoprecipitation of recombinant type I AC yielded a functional oligomer (Tang et al., 1995). With the soluble constructs from mammalian ACs it was impossible to address the question of a potential true dimerization of two pseudoheterodimers. A first biochemical test was carried out using linked mutation-inactivated homodimers which were complementary for each other. Biochemically it was demonstrated the formation of catalytic centers within pseudo-trimeric and -tetrameric structures. The findings are consistent with the possibility of tetramers or higher order structures in mammalian ACs. Such processes could add additional levels of AC regulation in eukaryotic cells.
As a highly surprising finding the mycoAC gene could be expressed without further ado equally well in E. coli and in mammalian HEK293 cells (
Finally, bacterial membrane proteins provide attractive systems for structural characterization of homologues of integral membrane proteins present in eukaryotes. This has been convincingly demonstrated with examples such as the bacterial rhodopsin (Deisenhofer et al., 1985), the potassium channel from streptomyces (Chen et al., 1999) and the mechanosensitive channel from Mycobacterium tuberculosis (Chang et al., 1998). The expression in E. coli and purification of the mycobacterial homologue of a mammalian AC including its membrane anchor may be seen as the first step toward a structural elucidation of this important protein.
Number | Date | Country | Kind |
---|---|---|---|
01111655.5 | May 2001 | EP | regional |
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/EP02/05288 | 5/14/2002 | WO |