Bacterial ghosts as carrier and targeting vehicles

Abstract

The invention concerns the use of bacterial ghosts as carrier and targeting vehicles for active substances.

Claims

1. A method for preparation of a bacterial ghost containing an active substance encapsulated therein, comprising: (i) preparing a bacterial ghost comprising a receptor, wherein the receptor is a heterologously expressed polypeptide integrated into a cytoplasmic membrane of the bacterial ghost comprising the receptor, and (ii) encapsulating the active substance in the bacterial ghost comprising the receptor prepared in step (i), wherein the active substance is present in the bacterial ghost in the immobilized form by interactions with the receptor.

2. The method of claim 1, wherein the active substance is immobilized on the receptor by non-covalent interactions.

3. The method of claim 1, wherein the active substance is immobilized on the receptor by non-covalent interactions that are not antibody-antigen binding.

4. The method of claim 1, wherein the active substance is not an antibody.

5. The method of claim 1, wherein the active substance is immobilized by non-covalent interactions between avidin or streptavidin and biotin or biotin analogs on the receptor located on an inner side of the cytoplasmic membrane of the bacterial ghost.

6. The method of claim 1, wherein the heterologously expressed polypeptide is a fusion polypeptide.

7. The method of claim 1, wherein the heterologously expressed polypeptide is a fusion polypeptide containing streptavidin or avidin.

8. The method of claim 1, wherein the active substance is a fusion polypeptide.

9. The method of claim 1, wherein the active substance is derivatized with receptor binding groups.

10. The method of claim 1, wherein the active substance is a biotinylated active substance.

11. The method of claim 1, wherein the active substance is biotinylated alkaline phosphatase.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIG. 1 shows a schematic representation of the streptavidin-anchoring plasmid pAV1 which contains a fusion gene E′-FXa-StrpA under the control of the inducible lac promoter (lacPO), the origin of replication ColE1 and the ampicillin resistance gene bla.

(2) FIG. 2 shows the reaction kinetics of alkaline phosphatase bound in streptavidin ghosts.

EXAMPLES

(3) 1. Materials and Methods

(4) 1.1 Construction of Streptavidin-Encoding Plasmids

(5) The plasmid pBGG9 (British Biotechnology Limited) was cleaved with the restriction enzymes NdeI and HindIII. A 495 by DNA fragment containing the complete streptavidin gene (Argarana et al., Nucleic acids Res. 14 (1986), 1871-1882) was isolated by agarose gel electrophoresis and subsequent electroelution. The NdeI restriction site was filled in by Klenow polymerase and the fragment was inserted between the HincII and HindIII restriction sites of M13K11RX (Waye et al., Nucleic acids Res. 13 (1985), 8561-8571) The resulting phagemid M13FN contains 160 codons of the streptavidin gene fused to the 3′-end of a short sequence which codes for the recognition sequence ile-glu-gly-arg of the protease factor Xa (FXa). This FXa-StrpA cassette was isolated by restriction cleavage with BamHI and the resulting 509 by DNA fragment was subcloned into the BamHI-linearized plasmid pSK (Stratagene, Cleveland, Ohio) to create the plasmid pFN6. The same 509 by BamHI fragment was also inserted into the BamHI-cleaved membrane targeting vector pKSEL5 to obtain the plasmid pAV1 containing the streptavidin gene fused to the 5′-terminal membrane anchor sequence E′ (FIG. 1).

(6) 1.2 Preparation of Streptavidin Ghosts

(7) E. coli NM522 cells (Stratagene) were simultaneously transformed with the lysis plasmid pML1 (Szostak et al., J. Biotechnol. 44 (1996), 161-170) and the streptavidin-encoding plasmid pAV1. The transformants were cultured at 28° C. in LB medium (10 g/l tryptone, 5 g/l yeast extract, 5 g/l NaCl) containing ampicillin (200 μg/ml) and kanamycin (50 μg/ml). One liter of medium was inoculated with an overnight culture which was derived from a single transformant colony and used as a preculture for a fermenter (type MRD 60TE, Meredos GmbH, Bovenden, Germany). The bacteria were cultured in the fermenter in a volume of 10 l with aeration and agitation until an optical density at 600 nm of 0.4 was reached. IPTG was then added to a final concentration of 3 mM in order to induce the expression of streptavidin. After 30 min 0.2 M MgSO.sub.4 was added and 20 min thereafter the expression of the lysis protein E was induced by increasing the temperature from 28° C. to 42° C. After 1 h the cells were harvested by centrifugation at 4000 g. Resuspension of the pellets in distilled water (final volume 5 l) led to their immediate lysis. The ghosts were washed twice in a large volume of Tris-buffered saline (TBS) and subsequently lyophilized.

(8) 1.3 Light and Electron Microscopy

(9) Examinations by light microscopy were carried out using an Olympus AX70 True Research System Microscope. Transmission electron micrographs were taken with a Siemens Elmiscop 101 electron microscope. Scanning electron micrographs were taken with a Hitachi S-800 field emission scanning electron microscope. The fixation of cells and preparation for electron microscopy were carried out as described by Witte et al. (J. Bacteriol. 172 (1990), 4109-4114).

(10) For the detection of streptavidin the ghosts were incubated while shaking for 20 min at 37° C. with gold-labelled albumin-biotin (10 nm, Sigma Immunochemicals) diluted in Tris buffer (10 mM Tris, 150 mM NaCl, pH 7.4), washed, and fixed for electron microscopy.

(11) 1.4 SDS-polyacrylamide Gel Electrophoresis and Western Blot

(12) Ghosts or protein samples were boiled for 5 min in gel loading buffer (2% SDS, 5% 2-mercaptoethanol, 10% glycerol and 0.003% bromophenol blue in 0.063 M Tris-HCl buffer, pH 6.8) and separated on a 10% SDS-polyacrylamide gel by the method of Laemmli (Nature 227 (1970), 680 to 685). Western blots were carried out as described by Towbin et al (Biotechnology 24 (1992), 145-149). Blots were blocked in TBS containing 1% bovine serum and incubated with anti-streptavidin antiserum from rabbits (Sigma Immunochemicals).

(13) 1.5 Binding of Biotinylated Alkaline Phosphatase and Fluorescent-Labelled Biotin

(14) Biotinylated alkaline phosphatase (Pierce) was diluted 1:1000 in Tris buffer. 2 mg of lyophilized ghosts was suspended in 500 μl diluted alkaline phosphatase solution and incubated for 30 min at 37° C. while shaking. The samples were centrifuged at 10000 g and washed three times in 20 ml Tris buffer and a fourth time in diethanolamine buffer (10 mM diethanolamine, 0.5 mM MgCl.sup.2, pH 9.5) and subsequently divided into six aliquots. Substrate (2.5 mM p-nitrophenyl phosphate in diethanolamine buffer) was added and the reactions were stopped after 0.5, 1, 2, 4, 8 or 16 min by adding an equal volume of 7 M NaOH. The samples were centrifuged at 10000 g and the supernatants were measured at 410 nm.

(15) A molar absorption coefficient ε=18.5×10.sup.3xM.sup.−1×cm.sup.−1 was determined and used to calculate the amount of p-nitrophenol formed according to the Lambert-Beer equation. The number of molecules of alkaline phosphatase bound per ghost was calculated assuming that one unit of alkaline phosphatase activity corresponds to the release of 1 μmol nitrophenol per min at pH 9.5 and 37° C. One unit of alkaline phosphatase corresponds to 0.7 μg, its molecular weight is 140000 and 1 mg of ghosts contains 6.7×10.sup.8 individual envelopes.

(16) For the binding of fluorescent-labelled biotin (FITC-biotin), ghosts were washed repeatedly in PBS until the relative fluorescence intensity in the supernatants was less than 0.5 at 530 nm (excitation at 490 nm). 1 mg lyophilized ghosts was incubated while shaking for 30 min in 2 ml of a solution containing 0.4056 μg FITC-biotin/100 ml TBS. The samples were centrifuged at 10000 g and the fluorescence intensities were measured in the supernatants.

(17) 1.6 Biotinylation of Polylysine

(18) Poly-L-lysine hydrobromide, molecular weight 18000 (Sigma), was biotinylated using the following protocol: 6 mg polylysine was taken up in 1 ml phosphate-buffered saline (PBS). 50 μl of a solution of 640 μg biotin-N-hydroxysuccinimide ester (Boehringer Mannheim) in 200 μl DMSO was added and the pH was adjusted to 10 using 0.5 M NaOH. The reaction mixture was stirred overnight at room temperature and subsequently dialysed for 48 h against water. A HABA test (Sigma) yielded a binding ratio of 2 mol biotin per mol polylysine.

(19) 1.7 Fluorescent Labelling of DNA

(20) A randomly selected plasmid (pUC18) was used to generate fluorescent-labelled DNA. Labelling was carried out using the polymerase chain reaction and labelled nucleotides (Cy3-dCTP, Pierce). The reaction mixtures contained 200 μM dATP, 200 μM dGTP, 200 μM dTTP, 200 μM dCTP (75% thereof is Cy3-dCTP), 1 μM of each oligonucleotide primer, 0.2 ng/μl linearized plasmid DNA and 0.02 U/μl Taq DNA polymerase in polymerase buffer. The reaction protocol was as follows: predenaturation for 4 min at 95° C.; 35 cycles: 1 min 95° C./1 min 60° C./3 min 72° C.; 5 min final extension at 72° C. The samples were phenolized, precipitated with ethanol, resuspended in 10 mM Tris-HCl (pH 8.0) and stored at −20° C.

(21) 1.8 Binding of Fluorescent-Labelled Dextran and DNA/Polylysine.

(22) 1 mg lyophilized streptavidin-ghosts was suspended in 1 ml Tris buffer. 50 μl of an aqueous solution (1 mg/ml) of biotinylated fluorescent-labelled dextran (Molecular Probes Europe BV) was added and the mixture was incubated for 1 h at 37° C. while shaking. The ghosts were washed three times in 1.5 ml Tris buffer and analysed by light microscopy. Diluted solutions of DNA (0.1 μg/μl in HBS [150 mM NaCl, 20 mM HERPES, pH 7.3]) and poly-L-lysine (1 μg/μl in HBS) were prepared in order to form complexes of fluorescent-labelled DNA and biotinylated polylysine. The solutions were combined at a weight ratio of DNA to polylysine of 10:1 and mixed rapidly. Streptavidin-ghosts were suspended therein, incubated for 1 h at 37° C. while shaking, washed and analysed by light microscopy.

(23) 2. Results

(24) 2.1 Membrane Anchoring of Streptavidin

(25) If the bacterial ghosts are used as a vehicle to transport active substances, the active substance should be fixed within the bacterial envelope. Recombinant ghosts which contain streptavidin anchored in their envelope are able to bind biotinylated compounds with high affinity. The plasmid pAV1 was constructed for this as described in section 1.1. It contains a hybrid gene consisting of the 54 5′-terminal codons of gene E of the bacteriophage PhiX174 (E′) followed by an in frame-fusion of the FXa-StrpA cassette. This plasmid is shown schematically in FIG. 1.

(26) 2.2 Production of Streptavidin-Ghosts

(27) Several E-specific lysis plasmids with different gene E expression control sequences, origins of replication and selection marker genes are available (Szostak et al., J. Biotechnol. 44 (1996), 161-170). The plasmid pML1 used here contains the gene E under the transcriptional control of the λP.sub.R-c1857 system. Onset of lysis can be observed in the E. coli strain NM522/pML1 by a decrease of optical density at 600 nm approximately 10 min after increasing the culturing temperature from 28° C. to 42° C.

(28) For the production of ghosts by an alternative E-lysis protocol, 0.2 M MgSO.sub.4 was added to the culture medium 20 min prior to inducing gene E expression. In this procedure protein E is incorporated into the bacterial cell wall complex, but cell lysis is inhibited by the high salt concentration in the surrounding medium. Gene E expression is allowed to proceed for 1 h in cultures treated with MgSO.sub.4 and the cells are subsequently harvested by centrifugation. Resuspension of these cells in water or low ionic strength buffers results in immediate, explosive lysis which creates substantially larger lysis holes than normal E-lysis.

(29) 2.3 Microscopic Visualisation of Ghosts Produced by Alternative Lysis

(30) Ghosts can be distinguished from living cells by light microscopic examination in which they appear distinctly more transparent than intact bacteria. Examination of ghosts by light microscopy which have been produced by alternative E-lysis showed cells exhibiting polar caps that had been blasted off or cracks in the middle opening them up into two halves. The ghosts appeared slightly elongated.

(31) 2.4 Detection of Streptavidin with Gold-Conjugated Biotin

(32) In order to detect the localization of streptavidin in the ghosts, streptavidin-ghosts were incubated with gold-labelled albumin-biotin particles, washed and examined by electron microscopy. Ultrathin sections revealed gold particles distributed exclusively along the inner membrane of the ghosts.

(33) 2.5 Determination of Streptavidin Anchorage in Ghosts

(34) Streptavidin-ghosts were analysed by SDS-polyacrylamide gel electrophoresis together with defined amounts of pure streptavidin as a control in order to determine their streptavidin content. The gels were transferred onto nitrocellulose membranes and treated with anti-streptavidin antiserum. Densitometric analysis of the streptavidin-specific bands on Western blots revealed a streptavidin content of approximately 5% of the total cell weight.

(35) 2.6 Functional Binding of Biotinylated Alkaline Phosphatase and FITC-Biotin, and Quantification of the Binding Sites

(36) An enzymatic assay was developed to determine the biotin-binding capacity of streptavidin-ghosts. Streptavidin-ghosts and streptavidin-negative ghosts (ghosts without streptavidin anchored to their membrane) which had both been prepared by the alternative lysis protocol, and streptavidin-ghosts produced by standard lysis, were incubated with biotinylated alkaline phosphatase. After extensive washing, the amount of retained enzyme was measured using p-nitrophenyl nitrophosphate as a substrate. Whereas almost no reaction was observed in streptavidin-negative ghost samples, alternatively lysed streptavidin-ghosts exhibited a bright yellow colouration. The reaction was stopped after controlled intervals and the absorbance of the sample supernatants was measured at 410 nm.

(37) The number of molecules of alkaline phosphatase bound per ghost was determined as approximately 200 by the calculation method described in section 1.5. Interestingly, streptavidin-ghosts produced by normal lysis were negative in the enzymatic test. Consequently, the larger holes created by the alternative lysis protocol are necessary for large molecules of active substances like alkaline phosphatase to allow an efficient diffusion into the interior of the ghosts.

(38) FIG. 2 shows the kinetics of the reaction of biotinylated alkaline phosphatase which was bound to streptavidin-ghosts produced by alternative lysis. Streptavidin-negative ghosts produced by alternative lysis were used as a control.

(39) A similar test was carried out using fluorescent-labelled biotin. Ghost and streptavidin-ghost samples were incubated with FITC-biotin, centrifuged and the residual fluorescence of unbound label was measured in the supernatants. The number of these much smaller molecules (molecular weight 832) that were bound was 2060.+−0.400 per ghost.

(40) 2.7 Binding of Fluorescent-Labelled Biotinylated Dextran and Fluorescent-Labelled DNA

(41) Fluorescent-labelled biotinylated dextran and fluorescent-labelled DNA were used as a model to demonstrate the fixation of compounds that could be used for the targeting of active substances in streptavidin-ghosts. For this streptavidin-ghosts were incubated with a mixture of biotinylated poly-L-lysine and fluorescent-labelled DNA or with fluorescent-labelled biotinylated dextran and analysed by fluorescence light microscopy. In both cases the fluorescent label was detected on the ghosts. Negative controls (ghosts without streptavidin) were not stained.