Chimeric polynucleotides and polypeptides enabling secretion of a polypeptide of interest in association with exosomes and use thereof for the production of immunogenic compositions
09546371 ยท 2017-01-17
Assignee
- Centre National De La Recherche Scientifique (Paris, FR)
- UNIVERSITE MONTPELLIER 2 SCIENCE ET TECHNIQUES (Montpellier, FR)
Inventors
- Robert Zaine El Abiddine Mamoun (St Andre de Sangonis, FR)
- Bernadette Nadine Trentin (St Andre de Sangonis, FR)
- Michel Vidal (Juvignac, FR)
Cpc classification
A61P31/00
HUMAN NECESSITIES
C07K2319/01
CHEMISTRY; METALLURGY
C12N2740/14022
CHEMISTRY; METALLURGY
C12N15/625
CHEMISTRY; METALLURGY
A61P35/00
HUMAN NECESSITIES
International classification
C07K19/00
CHEMISTRY; METALLURGY
Abstract
The present invention concerns a chimeric polypeptide comprising several polypeptide domains, which is capable of being secreted in association with membrane vesicles and in particular exosomes. The present invention also concerns a polypeptide constituted by one of said domains for the secretion of a peptide or polypeptide of interest in association with membrane vesicles and in particular exosomes. The invention also concerns the use of polypeptides of the invention and of polynucleotides coding for said polypeptides for the production of immunogenic compositions based on exosomes or based on DNA or for screening protein interactions. The present invention also concerns exploiting the properties of exosomes comprising a polypeptide of the invention and of immunogenic compositions of the invention for the prophylaxis and/or treatment of an infection by a pathogenic agent, a pathogenic organism, a tumour antigen or a cytoplasmic antigen, in particular to elicit or promote, in vivo, in a host, in particular in a human or non-human mammal or a bird, a humoral and/or cellular response against a virus, bacterium, parasite or tumour.
Claims
1. A membrane vesicle comprising a chimeric polypeptide, wherein said chimeric polypeptide comprises or consists of the following domains: (i) a peptide or polypeptide of interest; (ii) a transmembrane domain having the capacity to anchor in the lipid bilayer of a cell membrane; and (iii) a mutated cytoplasmic domain (CD) of the transmembrane protein (TM) of the Bovine Leukaemia Virus (BLV), wherein a sequence of the mutated cytoplasmic domain comprises or consists of the sequence SEQ ID NO: 12, wherein said mutated cytoplasmic domain has the capacity of addressing of said chimeric polypeptide to membrane vesicles and/or to the cell compartment(s) involved in the formation of said membrane vesicles, and wherein the sequence of said mutated cytoplasmic domain is a 45-80 amino acid sequence.
2. The membrane vesicle of claim 1, which is an exosome.
3. The membrane vesicle of claim 1, wherein the peptide or polypeptide of interest is exposed, partly or completely, on the outside of said membrane vesicle.
4. An immunogenic composition, which comprises one or several membrane vesicle(s) of claim 1.
5. The immunogenic composition of claim 4, which is a drug.
6. The immunogenic composition of claim 4 or 5, which is a drug for the prophylaxis and/or treatment of a bacterial, viral or parasitic infection, or of a tumour, by induction or stimulation of a humoral and/or cellular response against said tumour, virus, bacterium or parasite.
7. A recombinant exosome-producing cell, which comprises or has absorbed a membrane vesicle according to claim 1.
8. A method for in vitro production of membrane vesicles, which comprises bringing an exosome-producing cell into contact with one or several membrane vesicle(s) according to claim 1; a) growing said exosome-producing cell; and b) recovering the membrane vesicles produced by said exosome-producing cell.
9. The method of claim 8, which further comprises an intermediate step between steps a) and b), during which said cell is selected and/or stimulated to induce and/or increase the secretion of exosomes or to induce a specificity in the composition of the exosomes in certain cellular proteins.
10. A method for preparing a polyclonal serum directed against one or several antigenic peptide(s) or polypeptide(s) of interest expressed on the surface of membrane vesicles, comprising the following steps: a) administering to a non-human animal, the membrane vesicle of claim 1, or the immunogenic composition of claim 4, associated or not with an adjuvant, to induce the production of antibodies by said non-human animal; and b) recovering from said non-human animal the antibodies, which bind to said antigenic peptide(s) or polypeptide(s) of interest.
11. A method for preparing monoclonal antibodies directed against one or several antigenic peptide(s) or polypeptide(s) expressed on the surface of membrane vesicles, comprising the following steps: a) fusing, with myeloma cells, spleen cells previously obtained from a host to which the membrane vesicle of claim 1, or the immunogenic composition of claim 4 has been administered to produce hybridomas; b) growing said hybridomas under conditions allowing the production of antibodies; c) recovering the monoclonal antibodies, which are directed against said antigenic peptide(s) or polypeptide(s) of interest.
12. The membrane vesicle of claim 1, wherein the sequence of the mutated cytoplasmic domain is a 45-60 amino acid sequence.
13. The membrane vesicle of claim 1, wherein the sequence of the mutated cytoplasmic domain is devoid of the sequence KCLTSRLLKLLRQ (SEQ ID NO: 126).
14. The membrane vesicle of claim 1, wherein said peptide or polypeptide of interest (i) comprises or consists of one or several domain(s) of an extracellular protein; or one or several cytoplasmic domain(s) of a transmembrane protein; or one or several domain(s) of a cytosolic protein.
15. The membrane vesicle of claim 1, wherein said peptide or polypeptide of interest (i) is a viral peptide or polypeptide, or a bacterial peptide or polypeptide, or a tumor antigen.
16. The membrane vesicle of claim 1, wherein said peptide or polypeptide of interest (i) is the ectodomain of the haemagglutinin (HA) of an influenza virus, or a fragment comprising or consisting of one or several epitope(s) of said ectodomain; or the ectodomain of the HA protein of the H5N1 avian influenza virus, or a fragment comprising or consisting of one or several epitope(s) of said ectodomain; or the ectodomain of the Spike protein of a coronavirus of severe acute respiratory syndrome (SARS) or a fragment comprising or consisting of one or several epitope(s) of said ectodomain.
17. The membrane vesicle of claim 1, wherein said transmembrane domain (ii) is the transmembrane domain of the membrane protein of a virus, a bacterium or a tumor; or is a mutated derivative of said transmembrane domain, wherein said mutated derivative of transmembrane domain is defined by substitution, deletion and/or insertion of one or several residue(s) in the sequence of said transmembrane domain and by the capacity to anchor in the lipid bilayer of a cell membrane, and wherein the sequence of said mutated derivative of transmembrane domain consists of 10-50 amino acids and is at least 80% identical to the sequence of said transmembrane domain.
18. The membrane vesicle of claim 1, wherein said peptide or the polypeptide of interest (i) and said transmembrane domain (ii) are not fragments of the same protein.
19. The membrane vesicle of claim 1, wherein said chimeric polypeptide further comprises at least one linker linking two of said domains (i), (ii) and (iii).
20. The recombinant exosome-producing cell of claim 7, which is a cell of the immune system.
21. The recombinant exosome-producing cell of claim 7, which is a lymphocyte.
Description
DESCRIPTION OF THE FIGURES
(1) The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the office upon request and payment of the necessary fee.
(2)
(3)
(4)
(5)
(6) The bands corresponding to super-coiled plasmid DNA are framed. A: DNA obtained by the STET method and deposited on the column. E and E: Fractions retained on the column then eluted. FT: Non-retained fraction. M: Size marker.
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
(19)
(20)
(21)
(22)
(23)
(24)
(25)
(26)
(27)
(28)
(29)
(30)
(31)
(32)
(33)
EXAMPLES
Example 1
(34) Method and Apparatus
(35) I. Preparations of plasmidic DNA:
(36) A. Transformation of Bacteria:
(37) Competent DH5 bacteria (200 l) were transformed by thermal shock with 12.5 ng of each of the 13 DNAs studied coding for the wild type or mutant CD of the BLV virus, as well as by a plasmid devoid of the insert (pLPCX) acting as a negative control.
(38) The bacteria were then spread onto a LB/Agar medium containing 50 g/mL of ampicillin, at 37 C. for 16 h. They were then stored at 4 C.
(39) B. Pre-culture and Culture of Bacteria:
(40) A colony of each type of bacteria was then pre-cultured in 3 mL of LB medium containing 100 g/mL of ampicillin, at 37 C. with stirring, for approximately 8 h.
(41) Each pre culture served to inoculate, at 1/200, two flasks each containing 250 mL of LB/Amp (100 g/mL). Incubation was carried out at 37 C., with stirring, for 12 to 16 h.
(42) C. STET Maxipreparation:
(43) The cultures obtained were centrifuged (GR 412, Jouan) for 20 min at a speed of 3600 rpm and at a temperature of 4 C. The residues were taken up in 25 mL of STET buffer (Sucrose 8%, Triton X100 5%, Tris HCl pH 8, 50 mM, EDTA 50 mM) to which 500 l of lysozyme (10 mg/mL; Sigma) and 250 l of RNase (2 mg/mL; Sigma) were added. The tubes were then incubated for 10 min at 100 C. and centrifuged at 16000 rpm for 30 min. The supernatants obtained were incubated for 45 min at 65 C. in the presence of 1 mg/mL of proteinase K (Amresco). The supernatants were removed and the DNA was precipitated with 0.15 volume of sodium acetate (AcNa) 3 M, pH 6 and 0.6 volume of isopropanol. The solutions were centrifuged (Aventi 30, Beckman) at 4 C. for 15 min at 15000 rpm. The residues obtained were washed with 10 mL of ethanol before being centrifuged again using the same parameters. The nucleic acids obtained were dried and taken up in 2 mL of TE 1 then stored at 4 C.
(44) The presence of super-coiled DNA was verified for each Maxipreparation product by electrophoresis of 0.2 l and 1 l of the products by 0.8% agarose gel electrophoresis.
(45) D. Column Purification:
(46) STET Maxipreparations can be used to produce large quantities of plasmids, but the degree of purity could be improved. To this end, we purified each of the 14 plasmids by means of two passes over AX100 columns (Kit Nucleobond PC 100, Macherey Nagel). After precipitation with isopropanol, the purified plasmids obtained were taken up in 500 l of TE1 and stored at 4 C.
(47) The presence of super-coiled DNA was then verified by agarose gel electrophoresis (0.8%) for each eluate obtained as well as for the fraction not retained on the columns (FT=flow-through).
(48) The DNA concentration of the solutions obtained was evaluated by spectrophotometry, at a wavelength of 260 nm.
(49) E. Taking Aliquots and Precipitation with EtOH:
(50) Each DNA type was placed into tubes in aliquots of 50 or 100 g, then precipitation was carried out with ethanol (EtOH) and NaCl, in a laminar flow hood, in order to sterilize the plasmids. The residues obtained were taken up in an amount of 200 l of TE1 per 100 g of plasmids, i.e. a concentration of 500 ng/l. The presence of DNA in the aliquots was verified by 0.8% agarose gel electrophoresis.
(51) F. Spectrophotometric Assay of Aliquots Obtained:
(52) The aliquots were assayed by spectrophotometry at a wavelength of 260 nm. The samples were diluted to 1/50 and were assayed in a final volume of 500 l.
(53) G. Enzymatic Digestions:
(54) The identity of each plasmid was checked by digestion of 20 ng of each of them with restriction enzymes (New England Biolabs): the Hind III/Not I pair, Xba I, Aat II, Pac I, Sfo I. The plasmids were differentiated by the number of bands obtained and their molecular weight, after agarose gel electrophoresis (0.8%) of each digestion product.
(55) II. Analysis of the Expression of Chimeras
(56) A. Cell Culture:
(57) HEK293 cells (Human Embryonic Kidney cells) were cultivated in DMEM (Dulbecco's modified Eagle's medium), supplemented with 10% foetal calf serum (FCS) and 20 g/mL of gentamicin, at 37 C. under 5% CO.sub.2.
(58) B. Transfection:
(59) For the transfections, 510.sup.5 cells were plated onto 2 mL of medium per 9.6 cm.sup.2 well (6 well plates). After incubating overnight at 37 C. under 5% CO.sub.2, the medium was replaced with complete DMEM without antibiotics.
(60) The cells were then transfected by a polyplex formed by complexing, in a NaCl buffer (0.15 M), 6 l of Jet PEI (Qbiogen) and 3 g of each test nucleic acid; a plasmid expressing LacZ acted as the positive control for transfection. After 24 h of incubation at 37 C. under 5% CO.sub.2, the medium was eliminated and replaced with complete DMEM with 20 g/mL de gentamicin. Optimum plasmid expression was obtained 48 h after commencing the transfection.
(61) In certain cases, in order to analyze the importance of vesicular transport in the degradation and targeting of CD, we used vesicular transport inhibitors, namely bafilomycin and Ly 294002. 32 h post-transfection, each inhibitor was added to the culture medium in a concentration of 0.5 M for bafilomycin and 10 M for Ly 294002.
(62) C. Protein Extraction:
(63) 48 h post-transfection, the cells were lysed by a 0.5% THE-NP40 buffer to which 0.1 mM of PMSF had been added. After clarification by centrifuging (14000 rpm, 15 min, 4 C.; Eppendorf 5417R), the lysates were assayed by spectrophotometry (at 595 nm) using the Bradford technique.
(64) For each sample, 200 g de proteins was removed which was supplemented with lysis buffer for a final volume of 30 l to which 10 l of 4 buffer sample was added (CB 4: NaOH 200 mM, EDTA 20 mM, SDS 2%, bromocresol green 0.05%, glycerol 10%).
(65) D. Preparation of Exosomes:
(66) Before lysing the cells, the culture media were recovered and pre-centrifuged at 10000 rpm for 20 min at a temperature of 4 C. (Aventi 30, Beckman). The supernatants obtained were then ultracentrifuged (Optima LE-80K, Ti 50 rotor, Beckman) at 100000 g for 2 h at a temperature of 4 C. The residues obtained are taken up in 100 l of CB 1.
(67) We also analyzed the exosomes by sedimentation on a sucrose density gradient. The residues obtained after ultracentrifuging the culture media were taken up in 100 l of a 0.25 M sucrose solution.
(68) The vesicles were then deposited on a sucrose gradient prepared with 8 layers (of 1.2 mL) of different densities (molarity): 0.5/0.75/1/1.25/1.5/1.75/2/2.5. The tubes were then centrifuged (Optima LE-80K, SW 41 rotor, Beckman) at 39 000 rpm for 16 h, at a temperature of 4 C.
(69) After centrifuging, the gradient was recovered in 700 l fractions. The proteins were precipitated by adding the same volume of 30% TCA. The tubes were stored for 2 h at 4 C. then centrifuged (Eppendorf, 5417R) at a temperature of 4 C. for 20 min at 13000 rpm. The residues were taken up in 500 l of acetone then centrifuged again, as before.
(70) The residues thus obtained were taken up in 80 l of CB1 then analyzed by acrylamide-SDS gel migration and Western Blot.
(71) E. Western Blot and Immunoprecipitation:
(72) The protein samples obtained were analyzed by Western Blot: after migration and separation on acrylamide gel (12.5%), the proteins were transferred onto a hydrophobic PVDF membrane (Immobillon-P, Millipore).
(73) The presence of mutants of the TM protein was revealed by immunolabelling using the following antibodies: primary antibody: BLV anti-rabbit CD antiserum (dilution 1/200). secondary antibody: Anti-rabbit IgG labelled with peroxidase (dilution 1/5000; Jackson Immuno Research, JIR) The presence of transferrin receptors was revealed by immunolabelling using the following antibodies: primary antibody: anti-human TFr mouse IgG (dilution 1/200; Zymed) secondary antibody: anti-mouse IgG (dilution 1/5000; JIR)
(74) In order to selectively increase the concentration of the protein being studied, we also carried out immunoprecipitations on the protein lysates, prior to gel migration. After normalizing the protein quantities, lysate adsorption was carried out on sepharose 6B, in order to eliminate non-specific reactions and the immunoprecipitation was carried out with protein sepharose 6A in the presence of 2.5 g of the following antibodies: primary antibody: anti-mouse CD8 mouse IgG (19/178) (JIR) anti-mouse CD8 mouse rat IgG (53/672) (JIR) secondary antibody: Anti-mouse IgG (JIR) Anti-rat IgG (JIR)
The residues obtained were then taken up in 70 l de CB 1.5.
III. Localisation by Immunofluorescence
(75) A. Preparation of Slides.
(76) Coverslip type slides were sterilized in absolute EtOH, in a laminar flow hood, then placed in 1.9 cm.sup.2 wells (24 well plates) before being coated with poly-L-Lysine (25 g/mL, Sigma) for 1 h at 37 C. After washing with PBS, the slides were stored at 4 C., in 1 mL of PBS.
(77) B. Cell Culture and Transfection.
(78) The HEK293 cells were cultivated and transfected according to the method described in the section Analysis of chimeric expression. 24 h post-transfection, the cells were taken up in 35 mL of complete DMEM, then distributed into 1.9 cm.sup.2 wells (24 well plates) containing the slides which had been sterilized and coated in an amount of 1 mL of cellular dilution per well. The cells, thus undergoing culture, were incubated for 24 h at 37 C., under 5% CO.sub.2.
(79) C. Fixing and Permeabilization.
(80) The cells were then fixed for 30 min with a solution of formaldehyde (4%) then permeabilized with Triton X-100 (0.2% final). After three 10 min rinses with PBS, the slides were stored with 1 mL de PBS, at 4 C.
(81) D. Labels for Immunofluorescence.
(82) The various antibodies at our disposal meant that we could test several labels for localisation by immunofluorescence:
(83) CD8 Labels:
(84) No 1: mouse IgG/anti-mouse CD8 (19/178) (JIR)+anti-mouse IgG FITC (JIR) No 2: mouse Ascites/anti-mouse CD8 (19/178) (JIR)+anti-mouse IgG FITC (JIR) No 3: rat IgG/anti-mouse CD8 (53/672) (JIR)+anti-rat IgG FITC (JIR) No 4: rat IgG (53/6.7)/anti-mouse CD8 FITC (Pharmingen)
Labelling of Intra-Cell Compartments: No 5: mouse IgG/anti-human CD63 (Lamp3) (Zymed)+anti-mouse IgG Cy3 (Sigma) No 6: mouse IgG/anti-human Lamp1+anti-mouse IgG Cy3 (Sigma) No 7: mouse IgG/anti-human Tf2 (Zymed)+anti-mouse IgG Cy3 (Sigma) No 8: rabbit IgG/anti-caveolin (BD)+anti-rabbit IgG TRITC (JIR)
Results
(85) Knowledge of the exosome formation process is incomplete. Similarly, the functions associated with the cytoplasmic domain of Env are poorly characterized. The present study is based on the hypothesis that the BLV model is a good tool for studying the phenomena of exosome formation and viral particle formation and that the cytoplasmic domain of the TM protein (CD) of BLV is a potentially important tool for the development of vaccination by exosome display.
(86) In order to evaluate this potential and in order to characterize the functions responsible for targeting the TM protein of BLV, and more particularly the role of palmitoylated or non palmitoylated cysteine residues, we developed several vectors allowing expression, in eukaryotic cells, of chimeric transmembrane vectors comprising the ectodomain of the mouse CD8 protein and the CD domain (CD8-CD chimeric proteins). Wild type CD8-CD chimeras as well as mutated CD8-CD chimeras have been expressed.
(87) In a human cell, the mouse CD8 ectodomain is a neutral element that does not interfere with cell receptors. The chimeras used thus guarantee the absence of interactions between the ectodomain of the proteins and the cell structures.
(88) As a consequence, these chimeras mean that the cytoplasmic and transmembrane domains of the TM protein can be specifically studied, avoiding phenomena caused by the ectodomain of the TM protein and by the SU protein associated therewith.
(89) Three types of constructs were used (see
(90) These constructs allowed analysing, at the level of the targeting of the TM protein: the consequences of the mutation of cysteine residues, the importance of the integrity of the trans-membrane domain of the TM protein, and the importance of the cytoplasmic domain of the TM protein (CD).
(91) I. Preparation of Plasmid DNA:
(92) Initially, we produced large quantities of each of our 13 DNAs as well as a vector with no insert (pLPCX) acting as a negative control. After culture of bacteria transformed by our various plasmids, the DNA obtained was purified in succession by the STET method then on a column. The identity of the plasmids obtained was then checked by enzymatic digestion.
(93) A. DNA Preparations by the STET Method:
(94) In order to verify the integrity of the plasmid DNA and the yield of the STET preparations, we carried out a migration of 1 l and 0.20 l of each of them on agarose gel (0.8%; see
(95) B. Checking the Presence of DNA in the Column Purification Products:
(96) The DNA obtained by the STET method was purified on columns; the retained then eluted fractions as well as the non-retained fractions were then analyzed by gel electrophoresis in order to verify the integrity of the purified DNA obtained. The gel presented in
(97) C. Spectrometric Assays of Aliquots and Adjustment of DNA Concentrations After Enzymatic Digestion:
(98) The purified DNA was then assayed by spectrophotometry. The concentrations thus obtained varied, as a function of the aliquots, between 149 g/mL and 584 g/mL. After precipitation with EtOH, the DNA from the aliquots was taken up in TE1 and adjusted to the same concentration.
(99) In order to verify the concentrations of DNA, we linearized the plasmids by enzymatic digestion using the Hind III/Not I pair, then we analyzed them by gel electrophoresis (see
(100) D. Checking the Identity of the DNA After Enzymatic Digestion.
(101) 1) General Principle:
(102) In order to check the identity of the 14 DNA prepared, they were all checked by digestion with the aid of the restriction enzymes Hind III and Not I, Xba I, Aat II, Pac I as well as Sfo I, which should result in different combinations of gel profiles for each DNA. Agarose gel (0.8%) migration of the digestion products allowed the DNA to be differentiated as a function of the number and size of the bands obtained, as indicated in Table 1 (NB: bands smaller than a hundred base pairs in size were not visible on our electrophoresis gels):
(103) TABLE-US-00004 TABLE 1 Theoretical numbers of bands obtained after digestions N Plasmid Hind III/Not I Xba I Aat II Sfo I Pac I 1 pX2 wt TM 2 3 8 4 0 {close oversize brace} (i) 2 pX2 MCA 2 3 8 4 1 3 pX2 M2 2 3 9 4 {close oversize brace} (k) 1 {close oversize brace} (j) 4 pX2 M14 2 3 9 4 0 5 pX2 M15 2 {close oversize brace} (a) 3 {close oversize brace} (e) 8 5 0 6 pX2 M16 2 3 8 5 1 7 pX2 M17 2 3 9 5 {close oversize brace} (l) 1 8 pX2 M18 2 3 9 5 0 9 pX3 wt TM 2 2 8 4 0 {close oversize brace} (b) {close oversize brace} (f) 10 pX3 M15 2 2 9 4 1 11 pX4 wt TM 1 2 8 4 0 {close oversize brace} (c) {close oversize brace} (g) 12 pX4 M15 1 2 8 5 0 13 pX4 Stp 1 (d) 2 (h) 8 4 0
(104) Digestion of the constructs studied was used to confirm the identity of each plasmid. The large CD8-CD construct types were differentiated by number and size analysis of the bands obtained after digestion with the enzymes Hind III/Not I and Xba I. The various cysteine residue mutants were differentiated after digestion with the enzymes Aat II, Pac I and Sfo I. The number of bands thus obtained was larger than with the digestions with Hind III/Not I and Xba I, and differentiation was principally carried out based on the presence or absence of specific bands (in bold).
(105) Expected Size of Bands (bp):
(106) TABLE-US-00005 (a): 7206/49 (g): 6552/610 (b): 6868/311 (h): 6552/385 (c): 7162 (i): 4285/1055/801/498/294/186/83/53 (d): 6937 (j): 3423/1055/862/801/294/186/83/53 (e): 6822/367/66 (k): 4276/2538/310/131 (f): 6812/367 (l): 4276/1839/699/310/131
(107) 2) Example:
(108) For the differentiation of mutants pX2 AAC and pX2 CCA, the gels obtained are shown in
(109) Interpretation:
(110) Digestion with Hind III/Not I: single band obtained, at approximately 7206 bp, corresponds to the presence of constructs X2 or X4 in each of the two samples. Digestion with XbaI: obtaining two bands at approximately 6822 (or 6812) and 367 by corresponded to the presence of constructs X2 or X3 in each of the two samples.
(111) These two results cross-check and confirm that we were indeed in the presence of two constructs of type X2.
(112) We then sought to differentiate the X2 mutants among themselves (see
(113) Interpretation:
(114) Digestion with AatII: Sample 4: obtaining one specific band of approximately 3423 by (red), corresponds to the presence of one of the following X2 plasmids: pX2 CAC, pX2 AAC, pX2 CAA or pX2 AAA. Sample 5: obtaining one specific band of approximately 4285 by (blue) corresponds to the presence of one of the following X2 plasmids: pX2 CCC, pX2 ACC, pX2 CCA or pX2 ACA.
(115) Digestion with Sfo I: Sample 4: obtaining one specific band of approximately 2538 by (red) corresponds to the presence of one of the following X2 plasmids: pX2 CCC, pX2 ACC, pX2 CAC, pX2 AAC.
(116) This result cross-checks with that of digestion with Aat II, and so we could deduce therefrom we were in the presence of one of the following plasmids: pX2 AAC or pX2 CAC. Sample 5: obtaining 2 specific bands of approximately 1839 by (black) and 699 (green) by corresponds to the presence of one of the following X2 plasmids: pX2 CCA, pX2 ACA, pX2 CAA, pX2 AAA. This result cross-checks with that of digestion with Aat II, and so we could then deduce that we were in the presence of one of the following plasmids: pX2 CCA or pX2 ACA.
(117) Digestion with Pac I:
(118) For each sample, we obtained one low intensity band corresponding to non super-coiled DNAp as well as a high intensity band corresponding to our super-coiled DNAp (blue), demonstrating the absence of digestion with Pac I for these sample.
(119) Samples 4 and 5 correspond respectively to mutants pX2 AAC and pX2 CCA.
(120) This process was applied to the identification of all of the plasmids.
(121) 3) Conclusion:
(122) During this step for preparing DNA, we obtained several milligrams of each of the 14 purified plasmids and their identities were all confirmed after checking by enzymatic digestions and gel analyses. Further, all of the DNA have been adjusted to the same concentration.
(123) II. Analysis of the Expression of Chimeras:
(124) In order to evaluate DNA expression, the same quantity of each of the plasmids was transfected into HEK293 cells. The proteins expressed thereby were analyzed by gel migration followed by a transfer onto PVDF membrane. The membranes were then revealed with the aid of an anti-rabbit CD serum and of a secondary Anti-Rabbit antibody labelled with peroxidase.
(125) A. Analysis of the Presence of CD8-CD in Cell Lysates:
(126) 1) Analyses of Crude Lysates:
(127) After 48 h of transfection, the cells are lysed and the extracts obtained were assayed using the Bradford technique. We analyzed 200 g of crude proteins brutes, derived from these lysates, by gel migration and Western Blot revealed with an anti-CD antibody (see
(128) Various specific signals could then be observed in the samples: a double band at 50 kDa (blue) corresponding to the presence of CD8-CD. The two bands correspond to two levels of glycosylation of CD8. a signal at approximately 20 kDa (red) of indeterminate origin.
(129) The samples pX2 CCC, pX2 ACC, pX2 CAC, pX2 AAC, pX4 stp and pLPCX did not have these bands.
(130) 2) Analysis of Lysates After Immunoprecipitation by Anti-CD8 Antibodies:
(131) In order to reduce the detection threshold for CD8-CD in cell lysates, we used a large quantity of extract and concentrated the chimeras by immunoprecipitation with specific monoclonal antibodies of a conformational epitope of the ectodomain of CD8.
(132) After 48 h of transfection, the cells were lysed and the extracts obtained were assayed using the Bradford technique. After normalisation of the protein concentrations, the extracts were immunoprecipitated in the presence of Anti-CD8 antibody and protein A sepharose. The products obtained were analyzed by gel migration and Western Blot, and revealed with a anti-CD antibody (see
(133) In this case, the pX2 AAC sample had a detectable signal at 50 kDa, in contrast to the analysis carried out without immunoprecipitation. In contrast, no signal was detectable for the samples pX2 CCC, pX2 ACC, pX2 CAC, pX4 stp and pLPCX.
(134) 3) Summary of the Analysis of the Expression of CD8-CD in Cell Lysates:
(135) By correlating the data obtained by Western Blot on the cell lysates with the characteristics of each chimera (integrity of tmD BLV, presence of third cysteine residue and number of cysteine residues), we could draw up Table 2 below.
(136) It appears that, in addition to CD of BLV, two factors may act on the presence of CD8-CD chimeras. These factors are: the presence or not of tmD BLV, and the presence or not of the third cysteine residue (Cys 213).
(137) TABLE-US-00006 TABLE 2 Number of Signal Construct tmD DLV cysteines CD BLV pX2 CCC Complete 3 None pX2 ACC Complete 2 None pX2 CAC Complete 2 None pX2 AAC Complete 1 + pX2 CCA Complete 2 ++++ pX2 ACA Complete 1 ++++ pX2 CAA Complete 1 ++++ pX2 AAA Complete 0 ++ pX3 CCC - 15 AA 3 +++ pX3 CAC - 15 AA 2 +++ pX4 --C Absent 1 ++++ pX4 --A Absent 0 +++++ pX4 stp Absent 0 None pLpcX Absent 0 None
(138) B. Analysis of the Association of CD8-CD with Exosomes:
(139) The various DNA used only varied by a few nucleotides. It is thus probable that the quantities of CD8-CD synthesised were equivalent. However, during the analysis of the expression of CD in cell lysates, it was observed that certain chimeras were not detectable 48 h post-transfection. This absence of a signal could reflect the disappearance of some of our chimeric proteins. This disappearance could be due either to a rapid degradation of these chimeras, or to their secretion in the form of exosomes.
(140) We thus sought to detect the presence of CD8-CD within exosomes.
(141) 1) CD8-CD Content in Vesicles Isolated by Centrifuging:
(142) After 48 h of transfection, the culture media were recovered and centrifuged in order to isolate the exosomes. The residues obtained were then analyzed by gel migration and Western Blot and revealed with a anti-CD antibody (see
(143) Some samples exhibited a signal at 55 kDa corresponding to the presence of CD8-CD. This signal is not detectable for the samples pX2 CCC, pX2 ACC, pX2 CAC, pX2 AAC, pX4 stp and pLPCX.
(144) 2) CD8-CD Content After Sedimentation of Vesicles on Sucrose Density Gradient:
(145) In order to confirm that the signal obtained after ultracentrifuging the culture media was indeed due to the presence of chimeras in the exosomes and not to the presence of cellular debris containing CD8-CD, we analyzed the residues of exosomes obtained according to the method described above by sedimentation on a sucrose density gradient. The exosomes floated at a density of 1.13 g/mL to 1.20 g/mL as a function of cells used and the composition of the vesicles. After sedimentation by ultracentrifuging, the gradient was removed in 700 l fractions. The proteins were precipitated using TCA and analyzed by gel migration then Western Blots and revealed with a anti-CD antibody and an anti-transferrin receptor (RTf) antibody to detect the presence of cellular vesicles (endosomes or exosomes). This analysis was carried out for the mutants pX4 -C and pX4 --A. (see
(146) In pX4 --C and pX4 --A, a signal corresponding to CD8-CD was detected in fractions with a density of 1.13 g/mL to 1.25 g/mL. For these same densities, we were also able to detect a signal corresponding to RTf. These results reveal the presence of CD8-CD in the fractions containing the exosomes.
(147) 3) Summary of the Analysis of the Association of Exosomes with CD8-CD:
(148) The association of CD8-CD with the exosomes was detected for all of the mutants containing CD except for the pX2 with Cys 213. The pX4 --C and pX4 --A mutants appeared to have been very efficiently targeted in the exosomes. The pX3 and pX2 mutants without Cys 213 were also found in the exosomes, but in smaller proportions than for the pX4 --C and pX4--A mutants.
(149) It thus appears that, in addition to the presence of viral CD, two factors may act on the targeting of CD8-CD chimeras in exosomes.
(150) These factors are:
(151) the presence or not of the transmembrane domain of BLV. the presence or not of Cys 3.
(152) C. Influence of Inhibiting Vesicular Transport on the Expression of CD8-CD Chimeras:
(153) Since the absence of detection of the chimeras pX2 CCC, pX2 ACC, pX2 CAC and pX2 AAC was not due to increased exosomal secretion, we investigated whether it could be due to degradation by sorting of MVB proteins to the lysosomes.
(154) To this end, we used vesicular transport inhibitors, namely bafilomycin and Ly294002. These inhibitors can block the lysosomal degradation pathway, thereby encouraging the secretion of proteins by the exosomes. The inhibitors were added to the culture media 32 h post-transfection; 16 h later, the cells were lysed and the culture media were recovered then centrifuged to isolate the exosomes. The samples obtained were then analyzed by gel migration and revealed by Western Blot with a anti-CD antibody. We thus analyzed the expression of the chimeras pX2 CCC, pX2 CCA, pX3 CCC, pX4 -- C, pX4 -- A and pX4 stp (see
(155) It appears that, for these mutants, the Western Blot profiles for the proteins obtained in the presence of inhibitors were the same as during the analysis without inhibitor, both for the cell lysates and for the exosomes.
(156) The absence of the chimeras pX2 CCC, pX2 ACC, pX2 CAC and pX2 AAC is thus apparently due neither to an accelerated exosomal exit, nor to a degradation in the lysosomes. It is probable that it is the consequence of a very early degradation at the level of the system for controlling folding in the endoplasmic reticulum (RE) and TGN.
(157) III. Localisation by Immunofluorescence:
(158) In order to assess the membrane targeting of CD8-CD as well as its localisation in the cell compartments, we carried out observations using confocal immunofluorescence microscopy. 48 h post-transfection, the cells were fixed then labelled using various antibodies.
(159) A. Choice of Antibodies:
(160) Initially, each type of label (see Methods and Apparatus) was tested using variable antibody dilutions on fixed cells expressing pX4 -- C or pLPCX. After observation on a conventional immunofluorescence microscope (ZEISS Axiovert 200 M), we tested the various available antibodies (see Methods and Apparatus section) and determined their efficacy as well as their optimum dilutions. It was seen that several anti-intra-cell compartments antibodies had a zero or aspecific signal.
(161) For confocal imagery observation, we could thus use only two types of labels:
(162) CD8-CD Label: rat IgG (53/6.7) anti-mouse CD8 FITC (Pharmingen, dilution 1/50)
(163) Lamp3 Label: Anti-human CD63 (Lamp3) mouse IgG (Zymed, dilution 1/50)+anti-mouse IgG Cy3 (Sigma, dilution 1/500)
(164) B. Observations of the Distribution of CD8-CD:
(165) After fixing and labelling the cells using our various antibodies, we observed the distribution of labelling derived from CD8-CD as well as its co-localization with Lamp3 with the aid of a confocal microscope (ZEISS LSM 510).
(166) The cells expressing pLPCX (negative control) did not exhibit a FITC signal. This control thus allowed us to confirm that the FITC fluorescence visualised for the other mutants was indeed derived from the presence of CD8-CD.
(167) Surprisingly, despite an absence of detection by Western Blot, the pX2 constructs comprising Cys 3 were visible, albeit weakly, in immunofluorescence.
(168) Analysis of General Phenotypes:
(169) The HEK293 cells were transfected for 48 h then fixed. Observation was carried out with the aid of a ZEISS LSM 510 confocal microscope (X63 immersion objective). CD8: FITC label (green) revealed the presence of chimeras containing CD8. Lamp3: Cy3 label (red) revealed the presence of the Lamp3 protein characteristic of late endosomes. CD8-CD/Lamp3: Superimposition of FITC and Cy3 images. The yellow shade obtained proves the co-localization of CD8-CD with Lamp3.
(170) From the observations, we can distinguish 5 general phenotypes (see
(171) Analysis of Perinuclear Zones Having a Strong FITC Signal:
(172) For all of the samples except pX4 stp, a large homogeneous CD8 perinuclear signal was almost always found at the periphery of the nucleus. The intensity parameters required to visualize the major portion of the FITC fluorescence present in the cells caused saturation of this zone. Such a saturated signal is not very useful, and so we captured weaker intensity parameters by concentrating on an analysis of these zones, in particular in order to determine the pertinence of the co-localizations they have with Lamp3. CD8-CD/Lamp3 [a]: Co-localization of CD8-CD (green) with Lamp3 (red). The intensity parameters for acquisition, while weak, caused saturation of the perinuclear signal and the appearance of yellow shades at this level, demonstrating partial co-localization between these two labels. Because of the signal saturation, it was impossible to say whether the co-localization occurring was real or artificial. CD8-CD [d]: FITC signal from CD8-CD. The intensity parameters were substantially reduced in order to eliminate any signal saturation phenomenon. The FITC signal appeared diffuse, homogeneous and not punctuated. CD8-CD/Lamp3 [d]: Co-localization of CD8-CD (green) with Lamp3 (red). The intensity parameters were substantially reduced in order to eliminate any signal saturation phenomenon. This resulted in an absence of yellow shades, demonstrating the absence of co-localization between FITC and Lamp3.
(173) According to this data, the labelling derived from Lamp3 appeared to be located inside this perinuclear zone, but it was never co localized with the FITC label (see
(174) Furthermore, the labelling derived from CD8-CD of these perinuclear zones appeared diffuse, homogeneous and not punctuated, and appeared to demonstrate the presence of CD8-CD within a cellular structure. The localisation and appearance of this labelling, as well as the absence of genuine co-localization with Lamp3, suggest the presence of CD8-CD in the Golgi apparatus.
(175) This phenotype is found, to a greater or lesser extent, in all mutants except for pX4 stp. In pX4 --C and pX4 -A mutants, this phenotype is only visible in a minority of cells, in contrast to the mutants pX2 and pX3 in which it is always present.
(176) Thus, the specific observation of perinuclear zones strongly labelled by FITC allowed us to demonstrate the probable presence of CD8-CD in TGN.
(177) We used the data obtained to construct the table below (see Table 3).
(178) These observations confirm that, in addition to viral CD, two factors play a role in the stability and targeting of our chimeras.
(179) These factors are:
(180) the presence or not of tmD BLV; the presence or not of the N-terminal residues of BLV CD; the presence or not of Cys 3.
(181) TABLE-US-00007 TABLE 3 Localization of FITC labelling Phenotype Mutants FITC intensity Vesicular Membrane TGN A pX2 CCC + +++ ++ pX2 ACC + +++ ++ pX2 CAC + +++ ++ pX2 AAC + +++ ++ B pX2 CCA +++ ++++ ++ pX2 ACA +++ ++++ ++ pX2 CAA +++ ++++ ++ PX2 AAA +++ ++++ ++ C pX3 CCC ++++++ + +++ +++++ pX3 CAC ++++++ + +++ +++++ D pX4 --C ++++++ ++ ++++ + pX4 --A ++++++ ++ ++++ + E pX4 stp ++++ + +++
Conclusion:
(182) During our analyses, it has been shown that the chimera pX4 stp, composed solely of the ectodomain and the tmD of CD8, accumulate in HEK293 cells by being efficiently targeted to the plasma membrane.
(183) In the same manner, the chimeras pX4 --C and pX4 --A accumulate and are found in the plasma membrane, but in larger proportions than pX4 stp. These chimeras are secreted very effectively within the exosomes.
(184) The pX3 chimeras are definitely present in the Golgi apparatus, in contrast to the pX4 chimeras. Their targeting in the plasma membrane and in the exosomes appears less effective than in the pX4 chimeras, but remains considerable.
(185) The pX2 chimeras appear less stable and are found in TGN but not in the plasma membrane. The pX2 constructs comprising Cys 3 are much less stable in the cells and are not detected in the exosomes. Substitution of the terminal Cys in this type of construct appears to contribute to a gain in stability of the chimeric proteins. In fact, the pX2 constructs without Cys 3 are detectable in the cells and in the exosomes.
(186) In immunofluorescence, all of the chimeras studied have a vesicular label which is partially co-localized with a marker for late endosomes.
(187) This study has shown the importance of the presence or not of Cys 3 in the stability and in the targeting of CD8-CD chimeras. It appears that the absence of this cysteine residue favours stability and the membrane targeting of the chimeras studied as well as their presence in the exosomes. These phenomena could be independent of the hyperpalmitoylation associated with the absence of Cys 3. In fact, the construct pX2 AAC, which is non palmitoylated, is more stable than the three pX2 mutants having Cys 3 as well as Cys 1 and/or Cys 2, which can be palmitoylated. However, a different implication of palmitoylation in the stability and targeting of our chimeras cannot be excluded.
(188) This study has shown the fundamental importance of BLV tmD. In fact, the deletion of all or part of the tmD BLV appears to substantially increase the stability of the chimeras as well as their targeting in the plasma membrane. Thus, surprisingly, the presence of tmD BLV could contribute to the early degradation of the chimeras. This degradation could interfere in the system controlling folding in the cell compartments, namely the RE and TGN, since the lysosomal degradation pathway is not the cause. The absence of vesicular transport inhibitor effects on the stability and exosomal targeting of CD8-CD chimeras as well as their very partial co-localization with late endosomes, even for the least stable chimeras, supports this hypothesis.
(189) The principle attraction of our studies resides in the discovery of potentially effective tools for the development of a mode of vaccination based on the concept of exosome display. In fact, the pX4 --C and especially pX4 --A chimeras appear to have a molecular machinery that allows highly effective targeting of the peptide antigen (here CD8) to the exosomes. This machinery would thus be located in the cytoplasmic domain of the TM protein of BLV.
Example 2
Identification of Amino Acids Involved in Exosomal Targeting of CD Peptide
(190) In order to determine the exact nature of the machinery located in the cytoplasmic domain of the BLV TM protein before proceeding to the first immunization trials, we studied the effect of 18 types of mutations in the cytoplasmic domain of the BLV TM protein (see
(191) The substitution and deletion mutants were obtained by directed mutagenesis; the substituted residues were replaced by an alanine residue. Translation of the deletion mutants was halted by addition of a stop codon (TGA, TAG or TAA codon).
(192) The DNA sequences coding for the 18 mutants, as well as the wild type DNA CD sequence in which only the 13 N-terminal residues had been deleted (SEQ ID NO: 7; sequence termed wild type sequence hereinafter) were sub-cloned downstream of a sequence coding for the ectodomain of murine CD8. The 19 chimeric genes obtained were then cloned into a viral expression vector. The recombinant vectors obtained thereby were transfected into eukaryotic cells (HEK cells) in order to analyze targeting to the exosomes of the resulting chimeric proteins. 48 hours later, protein expression in the cells was examined by Western Blot. At the same time, the exosomes were purified by ultracentrifuging in order to evaluate sorting of the chimeric protein in the exosomes by FACS and Western Blot.
(193) The results shown below demonstrate the need for two peptide motifs individually recognized in the literature for their interactions with the proteins associated with the ESCRT machinery and with the intermembrane transfer machinery (in particular adaptins including AP3).
(194) This is the first time that experimental evidence has been provided that these two peptide motifs have a synergistic, determining role in exosomal targeting.
(195) These results provide novel and important information in terms of intercellular signalling using exosomes. Thus, we have a clearly defined tool to hand for targeting proteins with exosomes. From an industrial viewpoint, they will facilitate the production of a new generation of vaccine and provide a unique tool for screening therapeutic molecules or antibodies, for example.
(196) 1Obtaining Molecular Constructs
(197) APreparation of Inserts by PCR:
(198) The three substitution mutants S (Ser.fwdarw.Ala), D (Ac. Asp.fwdarw.Ala) and E (Ac.Glut.fwdarw.Ala) were obtained by directed mutagenesis using a double PCR employing the following mutation primers:
(199) TABLE-US-00008 PrimerStoA,sense: 5 CCCTAAACCCGATGCTGATTATCAGGCGTTGCTACCATCC3 PrimerStoA,anti-sense: 5 CGCGGATGGTAGCAACGCCTGATAATCAGCATCGGGTTTA3 PrimerDtoA,sense: 5 CCACCAAGCCGGCATACATCAACCT3 PrimerDtoA,anti-sense: 5 TCGAAGGTTGATGTATGCCGGCTTGGT3 PrimerEtoA,sense: 5 GCTACCATCCGCGCCAGCGATCTAC3 PrimerEtoA,anti-sense: 5 GTAGATCGCTGGCGCGGATGGTA3.
(200) The PCR was carried out using the Expand High Fidelity PCR system kit (Roche) which had an enzymatic mixture containing thermostable Taq DNA polymerase and a thermostable Tgo polymerase provided with a corrective activity (3-5-exonuclease) to limit errors during polymerisation and obtain fragments with blunt ends. Two mixtures of reagents were prepared at 4 C.:
(201) A (25 L):10 ng of DNA to be amplified, 1 L of 10 mM dNTP (200 M final each), 1.5 L of each of the two sense and anti-sense primers, at 10 M (300 nM final), sterile water (qsp 25 L); and
(202) B (25 L):5 L of Expand High Fidelity buffer, 10 with 15 mM MgCl.sub.2 (1.5 mM final), 0.75 L of High Fidelity enzyme mixture (2.6 U final), sterile water (qsp 25 L).
(203) A and B were mixed at 4 C. then the following amplification cycles were carried out: denaturing of the double strand DNA, 2 minutes at 94 C.; 4 cycles of denaturing (94 C., 10 seconds), hybridization (50 C., 15 seconds) and elongation (72 C., 20 seconds); 25 cycles: 10 seconds at 94 C., 15 seconds at 64 C. then 20 seconds at 72 C.; final elongation for 7 minutes at 72 C.
(204) The PCR product obtained was kept at 4 C.
(205) The DNA sequences coding for the 18 mutants, as well as the wild type DNA sequence were modified by PCR (directed mutagenesis) so that they were framed by particular restriction sites; the protocol indicated above was carried out using two primers having the restriction sites XbaI and NotI.
(206) BCloning of PCR Products in TOPO-bluntII and Controls
(207) Each of the 19 DNAs was ligated into a TOPO cloning vector (see
(208) a) Ligation into the Plasmid and Transformation in Top10
(209) Each of the various PCR products was integrated into a TOPO-BluntII plasmid that was already open and had blunt ends and carried the kanamycin resistance gene. The chemocompetent Top10 bacteria were transformed by these plasmids and cultured on LB/agar dishes (100 g/mL of kanamycin). The TOPO-BluntII plasmid with no insert acted as a negative control and another control (1 ng of PUC19 plasmid) was used as the positive control for transformation. After culture, only the bacteria transformed by a plasmid containing the insert or PUC19 (positive control) developed in the presence of the selection agent (antibiotic).
(210) b) Targeting of Good Clones and Sequencing:
(211) Depending on the cloning results, 2 to 10 colonies were amplified to carry out extraction of the plasmid DNA. For each construct and each clone, we obtained a plasmid DNA volume of 100 L at approximately 150 ng/L. The absorption at 260 nm/absorption at 280 nm ratio for each purification had a value in the range 1.8 to 2, providing evidence of the purity of the preparation (a value of less than 1.8 provides evidence of protein contamination).
(212) In order to confirm the presence of an insert in the plasmid, the plasmid DNA was digested by the EcoRI restriction enzyme which framed the sequences of interest. These digestion were visualized on 2% agarose gel where the presence of a fragment (of approximately 300 pb) constituted proof of recombinant DNA.
(213) Once good clones had been identified, 2 g to 4 g of plasmid DNA from each mutant was sequenced to check the integrity of each sequence of interest and thus of the open reading frame. The mutations and restriction sites added were checked at the same time.
(214) CObtaining Chimeric Genes in a pKSII Cloning Vector
(215) Each of the 19 sequences (1 wild type and 18 mutated) was placed downstream of a sequence coding for the mouse CD8 ectodomain, to provide 19 constructs.
(216) a) Preparation of a pKSII-CD8 Cloning Vector
(217) The pKSII-CD8 vector (see
(218) b) Preparation of Inserts
(219) The various inserts framed by the XbaI-NotI restriction sites were digested with the same restriction enzymes as the plasmid so that it could integrate them.
(220) c) Obtaining Chimeric Plasmids by Cloning
(221) The DNAs were inserted in the linearized pKSII-CD8 plasmid:
(222) The digested inserts were purified on 2.5% agarose gel. Their reinsertion in the pKSII-CD8 vector was carried out using the gel ligation technique. A fragment of virgin gel acted as the negative ligation control.
(223) Having used the same restriction enzymes for the vector and the inserts, they thus had cohesive, complementary XbaI and NotI sites: the plasmid and inserts should be able to establish bonds between them. One enzyme, ligase, catalyses the formation of a phosphodiester linkage between a 3-OH end and a 5-Phosphate end of two nucleic acids.
(224) Once the ligation step is complete, the DH5 bacteria were transformed by these plasmids carrying the ampicillin resistance gene. After culture of the various bacteria transformed at 37 C. on LB/agar (50 g/mL of ampicillin) and comparison with the negative controls, colony development was observed only on cells transformed by the ligation products in the presence of the CD insert. This suggests that the colonies obtained had indeed been transformed by a vector containing an insert between the XbaI and NotI sites.
(225) d) Screening:
(226) In order to confirm that chimeric plasmids had been obtained, various clones of each mutant were screened by digesting plasmid DNA with the two enzymes, XhoI/NotI, and after migration of the digestion products on 0.8% agarose gel. This double digestion excises all of the created chimeric gene, i.e. with the CD8 in phase with the CD.
(227) For each clone and each construct (mutated or wild type pKSII-CD8-CD), a first band at 2.9 kbp corresponding to the size of the linearized pKSII plasmid was observed. A second band was noted at approximately 950 bp; it corresponded to the gene coding for the mutated or non-mutated CD8-CD chimeras.
(228) DObtaining Chimeric Genes in the Retroviral Expression Vector pLPCX:
(229) While it is easy to manipulate because of its size and its restriction sites, the expression vector pKSII does not allow protein expression in eukaryotic cells. We therefore selected pLPCX (Clontech Laboratories Inc., see
(230) Each chimeric gene was excised from the pKSII plasmid between the XhoI-NotI sites for purification by extraction on 2% agarose gel using the Nucleospin Extract II kit (Macherey-Nagel).
(231) The expression vector pLPCX had also been digested by the XhoI-NotI enzyme pair, dephosphorylated then precipitated with isopropanol (this step was to eliminate the short fragment of DNA (less than 100 pb) situated between XhoI and NotI liberated during digestion).
(232) The ligation of chimeric genes with pLPCX was carried out using T4 DNA ligase (Biolabs) (insert/vector ratio of approximately 3/1 molecule to molecule). Chemo-competent Stbl2 bacteria (MAX Efficiency Stb12 Competent Cells, Invitrogen) were transformed by these ligation products. A positive control (1 ng of the pUC19 plasmid) and a negative control (pLPCX ligated without insert) were prepared at the same time. After bacterial culture at 30 C. on gelose medium containing 50 g/mL of ampicillin, only the bacteria transformed by the positive control or by the ligation products with inserts had developed.
(233) In order to be able to proceed to screening and have a sufficient quantity of DNA plasmid necessary for the transfections in eukaryotic cells, Midipreparations were carried out. The presence of inserts in the plasmid was verified on 2% agarose gel after digestion with XhoI then NotI. For each construct (mutated and wild type pLPCX-CD8-CD; see
(234) 2Expression and Analysis of Targeting to Exosomes:
(235) ATransfections in HEH 293T Cells:
(236) In order to confirm the transfection of HEK 293T eukaryotic cells and estimate the percentage of transduced cells, the cells were transfected by the plasmid containing the LacZ gene and incubated in a solution of X-Gal.
(237) Initially by eye and then by observation with an optical microscope (magnification X40), we observed that more than 50% of the cells had been stained blue, which proved that the great majority had been transduced by LacZ plasmid. The same conditions had been applied as for the transfection of our chimeric genes, and so it was probable that more than 50% of the cells had been transduced by the chimeric genes.
(238) At the same time, twenty simultaneous transfections were carried out in HEK 293T cells. They corresponded to each of the plasmids and to a negative control (pLPCX-CD8 without CD).
(239) BExpression of Chimeric Proteins in the Cell and Targeting to the Exosomes
(240) a) Western Blot Analysis:
(241) The cell and exosomal lysates derived from the transfections were analyzed by 10% SDS-PAGE gel migration, followed by a transfer onto PVDF membrane (polyvinylidene difluoride, Immobilon-P, Millipore). These membranes were then revealed with the aid of a primary anti-rabbit CD serum followed by a secondary anti-rabbit IgG antibody coupled to peroxidase. After revealing, these antibodies were eliminated and the transfer membranes were revealed in the same manner but using an anti-rabbit CD8 serum.
(242) The results are presented in
(243) Comparison of Levels of Expression of Chimeras in Cells and Exosomes:
(244) It was observed that the cell lysates exhibited only weak expression of chimeric proteins. In contrast, the presence of certain chimeric proteins in the exosomes was sometimes very strong.
(245) In addition to the levels of expression, the major difference which was noted between the cell proteins and exosome proteins was the presence of 2 or 3 bands for the cells and only one for the exosomes. This was because the cell contains non-glycosylated forms and forms which are glycosylated to a greater or lesser extent. Only the correctly glycosylated form is found in the exosomes.
(246) Western Blot Analysis of Exosomal Targeting of Positive (CD8-CD) and Negative (CD8 Alone) Controls:
(247) With the anti-CD serum, a band migrating to 31 kDa was present in the exosomal lysate of the wild type CD8-CD control, while it was absent in the lysate from cells transfected by the negative control. This band is characteristic of the expected chimeric protein.
(248) With the anti-CD8 serum, the negative CD8 control alone had a band near 27 kDa which corresponded to expression and to exosomal targeting of CD8 devoid of CD. As before, a band migrating to 31 kDa was present in the exosomal lysate of the wild type CD8-CD control. The difference in intensity between the 31 kDa and 27 kDa bands indicates that CD8 is much more targeted onto the exosomes when it is fused to CD.
(249) Western Blot Analysis of Variations in Targeting of Mutated CD8-CD:
(250) Only the results obtained with the anti-CD8 serum were mutually comparable. The results obtained with the anti CD serum were only used to confirm the preceding results. As a function of the mutation of the sequence coding for the CD, these results show a variation in expression and targeting of chimeric proteins to exosomes. This is particularly clear for mutations which strongly inhibit targeting of proteins to exosomes. The mutants concerns were the mutants KM8, KM13, D and KS8.
(251) From these observations it may be concluded that the motif PSAP (KM8 mutant) and the motif DY (at the last motif YxxL (mutants KM13 and D)) are important for exosomal targeting.
(252) b) Quantification of Chimeric Proteins on Exosomes by FACs:
(253) The presence of chimeric proteins was also investigated by a cytofluorometric analysis (FACscan) using a fluorescent monoclonal anti-CD8 antibody.
(254) After fixing exosomes on latex beads (IDC (Interfacial Dynamics Corporation) ultraclean aldehyde/sulphate Latex beads), the chimeric proteins present on the surface of the exosomes were labelled with the aid of a monoclonal mouse anti-CD8 antibody coupled to fluorescein (53-6.7 antibodies from Pharmingen) and analyzed by cytofluorimetry (FACScan).
(255) The results obtained are particularly clear for mutants KM8, KM13 and D, which reveal the importance of mutated amino acids in targeting chimeric proteins on exosomes (see
(256) Conclusion:
(257) During this study, we constructed chimeric genes for expressing the mutated or wild type CD pilot peptide fused with mouse CD8. These mutations of the amino acids or remarkable motifs of CD were intended to identify amino acids and consensus motifs that are important in exosome targeting.
(258) The various chimeric genes were integrated into a retroviral expression vector in order to transfect HEK 293T eukaryotic cells in order to obtain expression of these chimeric proteins. The bands observed in Western Blot suggest that, like the native CD8 protein, these proteins are glycosylated differently during their passage through the Golgi apparatus. Only correctly glycosylated proteins were found in the exosomes. These proteins have undergone adequate post translational modifications, a condition which is indispensable to the expression of conformational epitopes essential to the future production of vaccine immunity or to screening therapeutic molecules. However, these glycosylations, which are present to a greater or lesser extent, result in multiple diffuse bands which hindered us during comparative protein quantification. To overcome this problem, the lysates had to be treated with an endoglycosylase in order to observe a single band on the gel.
(259) According to these results, the motifs PSAP and DY (from the last YxxL) are indispensable to expression and targeting of chimeric proteins to exosomes. These results are novel and are interesting both on the fundamental front and on the industrial application front.
(260) It is probable that the PSAP motif is responsible for an interaction with the Tsg101 protein of the ESCRT complex. Regarding the motif DYxxL, it could be implicated in the interaction with the ALIX protein of the ESCRT complex. Thus, for the first time, experimental data suggest that the ESCRT complex is involved in the formation of exosomes.
Example 3
Targeting of Receptors with Transmembrane Domains to Exosomes
(261) Membrane receptors are the major targets for the development of therapeutic molecules. In general, high throughput screening of drugs is carried out, in particular with receptors with multiple membrane domains expressed on cells in culture. In addition to difficulties in obtaining strong expression of receptors on the cell surface, this technique causes difficulties for robotic automation. However, it is currently the only solution since the use of purified recombinant receptors is currently technically unenvisageable.
(262) In this context, exosomes carrying receptors, in particular receptors with multiple domains, would constitute a tool which was simple to use and well suited to screening because of their stability and ease of manipulation.
(263) The present study was aimed as producing exosomes carrying receptors with single or multiple transmembrane domains, in particular the receptor CxCR4 (receptor for the chemokine SDS-1 (CXCL-12) and HIV) and the CD4 receptor (HIV receptor).
(264) Three chimeric chimeric genes were synthesized. They comprised, at the 3 end, the CD-BLV peptide of sequence SEQ ID NO: 8, and at the 5 end, a DNA coding for the CxCR4 human receptor, for a version of the CxCR4 receptor truncated at the C-terminal part comprising 307 amino acids (CxCR4 (307)) or for a version of the CD4 receptor truncated at the C-terminal part comprising 403 amino acids (CD4 (403)).
(265) The CD4 and CxCR4 receptors respectively comprise one and seven transmembrane domains.
(266) The three chimeric genes were cloned into a retroviral expression vector pLPCX. These various plasmids were transfected into HEK293T human eukaryotic cells in order to observe the expression of the various chimeric proteins in these cells as well as their sorting towards the exosomes.
(267) The cloning and sub-cloning strategy used was similar to that described for Example 2:
(268) The DNAs coding for the receptors CxCR4, CxCR4 (307) and CD4 (403) as well as that coding for the pilot peptide CD were amplified by PCR using primers comprising the sequences for the restriction sites which were to be integrated at each end of the amplified fragments (the fragments CxCR4, CxCR4 (307) and CD4 (403) will be flanked at the 5 end by the EcoRI site and at the 3 end by the XbaI site, and CD/BLV will be flanked at the 5 end by the XbaI site and at the 3 end by the NotI site).
(269) The inserts produced were cloned into the Topo amplification vectors (see
(270) The CD/BLV insert was then excised from the Topo vector by enzymatic digestion using the XbaI/NotI couple then sub-cloned into the pKS2 amplification vector (see
(271) The various constructs thus obtained were excised from the recombinant pKS2 plasmids by digestion with the EcoRI/NotI enzyme pair then sub-cloned into the retroviral expression vector pLPCX (see
(272) The various pLPCX plasmids were transfected into HEK293T human eukaryotic cells in order to express the chimeric proteins CxCR4/CD, CxCR4(307)/CD and CD4 (403)/CD.
(273) An extract of the total cellular proteins (100 g) and the proteins of a suspension of exosomes produced by each of the batches of transfected cells then underwent SDS-PAGE (10%) migration in the presence or absence of -mercaptoethanol. The proteins of interest were revealed by Western Blot using a anti-rabbit CD primary serum and anti-rabbit IgG secondary antibody coupled to peroxidase. The antibodies were revealed in a dark room using an ECL solution. Finally, the protein fingerprint of each sample was revealed by staining with Coomassie blue.
(274) The results are presented in
(275) Note that the chimeras CxCR4/CD, CxCR4 (307)/CD and CD4 (403)/CD were expressed in the cellular protein extracts (see
(276) Several bands characterize the chimera CxCR4/CD: 36, 42, 62 and 87 kDa. The expression of the wild type CxCR4 receptor is characterized by several isoforms, in particular in the HEK293T cells; the 34, 40, 47, 62, 73 and 80 kDa bands may be identified (Sloane J A, et al.). The larger sizes of the bands of the CxCR4/CD chimera in the HEK293T were due to the presence of pilot peptide (CD domain) in the CxCR4/CD chimera. Similarly, the 30, 42, 60 and 83 kDa bands revealed the presence of the CxCR4 (307)/CD chimera. The variation in the size of the bands representing the various isoforms of this chimera is explained by the fact that the CxCR4 receptor is truncated. The CD4 (403)/CD chimera is characterized by a clearly visible 53 kDa band.
(277) Further, these chimeras are all sorted to the exosomes, as shown by the presence of 36, 42, 62 and 87 kDa (for CxCR4/CD) bands; 30, 38, 48, and 83 kDa bands (for CxCR4 (307)/CD) and the 53 kDa band (for CD4 (403)/CD) (see
(278) Revealing the various protein fingerprints with Coomassie blue shows that the total quantities of proteins of exosomal origin (
(279) Conclusion
(280) Transfection of HEK293T cells with various pLPCX plasmids has been used to express the chimeras CxCR4/CD-BLV, CxCR4(403) /CD-BLV and CD4 (403)/CD-BLV. Western Blot analysis has demonstrated their presence in the cell lysates.
(281) As expected, several isoforms of the chimeras CxCR4/CD-BLV and CxCR4 (307)/CD-BLV are expressed in HEK293T cells transfected with the plasmids pLPCX CxCR4/CD-BLV and pLPCX CxCR4 (307)/CD-BLV. The chimeras CxCR4/CD-BLV and CxCR4 (307)/CD-BLV are effectively sorted towards the exosomes and it appears that these fusion proteins are equally shared between the cells and the exosomes.
(282) Further, the presence of the chimera CD4 (403)/CD-BLV is observed in HEK293T cells transfected with the plasmid pLPCX CD4 (403)/CD-BLV. It appears that the pilot peptide CD-BLV favours major sorting of the chimera CD4 (403)/CD-BLV towards the exosomes.
(283) The results described above show that the pilot peptide CD-BLV is equally capable of sorting proteins with single or with multiple transmembrane domains towards the exosomes.
(284) It is known that it is very difficult to work on these receptors in solution as they are not integrated into a plasma membrane, and they do not retain their native structure. Current studies carried out on these proteins are often made using stable cell lines expressing the receptors of interest. However, this is a constraint in terms of the time and the cost of acquiring, cultivating and maintaining those lines which may die at any time if they are maltreated. For this reason, the fact of using exosomes carrying receptors with multiple transmembrane domains represents an interesting solution as the exosomes have all of the advantages of a cell for these studies with none of the disadvantages since they are not alive.
(285) Exosomes with recombinant proteins integrated into their membrane, and in particular proteins comprising multiple transmembrane domains, could be used in vaccinology and as a screening tool.
REFERENCES
(286) Alberts B, Bray D, Lewis J, Raff M, Roberts K, Watson J D (1995). Molecular Biology of the Cell (3rd Ed.) Garland, N.Y. Cann A J, Churcher M J, Boyd M, O'Brien W, Zhao J O, Zack J, Chen I S (1992). The region of the envelope gene of human immunodeficiency virus type 1 responsible for determination of cell tropism. J Virol. 1992; 66(1):305-9. Chaput N, Taeb J, Schartz N, Andre F, Angevin E, Zitvogel L. (2004). Exosomes-based immunotherapy. Cancer Immunol Immunother; 53:234-239. Colino, J. and Snapper C. M. (2006). Journal of Immunology, 177:37576 De Gassart A, Geminard C, Hoekstra D, Vidal M. (2004). Exosome secretion: the art of reutilizing nonrecycled proteins. Traffic.; 5:896-903. Delamarre L, Rosenberg A R, Pique C, Pham D, Callebaut I, Dokhelar M C. (1996). The HTLV-I envelope glycoproteins: structure and functions. J Acquir Immune Defic Syndr Hum Retrovirol; 13 Suppl 1:S85-91. Delcayre, A., et al. 2005, Blood Cells Mol Dis 35:158; Thery C. et al, (2002); Nature Immunology 3:1156. Delcayre A, Le Pecq J B. (2006) Exosomes as novel therapeutic nanodevices. Curr Op in Mol Ther.; 8:1464-1471. Gould S J, Booth A, Hildret J E. (2003) The Trojan exosome hypothesis. Proc Natl Acad Sci USA.; 100:10592-10597. Herbein G, Coaquette A, Perez-Bercoff D, Pancino G. (2002) Macrophage activation and HIV infection: can the trojan horse turn into a fortress? Curr Mol Med.; 2:723-738. Levine A J. (1992). Viruses (Scientific American Library, No. 37). W. H. Freeman/Scientific American Library Pornillos O, Garrus J, Sundquist W I. (2002). Mechanism of enveloped RNA virus budding. Trends Cell Biol. 2; 12:569-579. Raposo G, Moore M, Innes D, Leijenderkker R, Leigh-Brown A, Benaroch P, Geuze H. (2002). Human macrophages accumulate HIV-1 particles in MHC II compartments. Traffic.; 3:718-729. Sloane J A, et al. (2005). Marked structural and functional heterogeneity in CxCR4: separation of HIV-1 and SDF-1alpha responses. Immunology and Cell Biology; 83: 129-143. Straub O C, Levy D. (1999). Bovine immunodeficiency virus and analogies with human immunodeficiency virus. Leukemia.;13 Suppl 1:S106-9. Zitvogel L, Regnault A, Lozier A, Wolfers J, Tenza D, Raposo G, Amigorena S. (1998). Dendritic cell-derived exosomes elicit potent antitumour immune responses in vivo. Nat. Med.; 4:594-600.