Modified effector cell (or chimeric receptor) for treating disialoganglioside G.SUB.D2.-expressing neoplasia

Abstract

A modified effector cell includes a non-reversibly produced vector-encoded anti-G.sub.D2-BB- chimeric receptor for use in disialoganglioside G.sub.D2-expressing neoplasia, which is inserted in the cell, to obtain an effector cell that stably produces the anti-G.sub.D2-BB- chimeric receptor, the chimeric receptor having two distinct mutually fused portions, i.e. an intra-cytoplasmic portion and an extra-cytoplasmic portion.

Claims

1. A modified effector cell for treating disialoganglioside GD.sub.2-expressing neoplasia, comprising: a cytoplasm with a nucleus therein, the cytoplasm being enclosed in a membrane; and a chimeric receptor, wherein the chimeric receptor comprises an extracytoplasmic portion, a transmembrane portion, and an intracytoplasmic portion, wherein the extracytoplasmic portion comprises, in order: a signal peptide containing an intron sequence, a first sequence encoding for a variable region of a light chain of the anti-GD.sub.2 immunoglobulin M (IgM) antibody, a linker that allows folding of a GD.sub.2 antigen recognition region of an anti-GD.sub.2 IgM antibody, and a second sequence encoding for a variable region of a heavy chain of the anti-GD.sub.2 IgM antibody, and wherein the transmembrane portion and the intracytoplasmic portion comprise, in order: a hinge and transmembrane domain of a human lymphocyte CD8 molecule, an intracellular portion of a 4-1BB co-stimulatory molecule, and an intracellular portion of a human lymphocyte CD3- molecule.

2. The modified effector cell as claimed in claim 1, wherein the chimeric receptor is encoded by the sequence set forth in SEQ ID NO: 1.

3. The modified effector cell as claimed in claim 1, wherein the linker that allows folding of the GD.sub.2 antigen recognition region consists of 18 amino acids.

4. The modified effector cell as claimed in claim 1, wherein the anti-GD.sub.2 IgM antibody is from clone 126.

5. A polyclonal mixture of effector cells for treating disialoganglioside GD.sub.2-expressing neoplasia, comprising: a plurality of modified effector cells comprising: a cytoplasm with a nucleus therein, the cytoplasm being enclosed in a membrane; and a chimeric receptor, wherein the chimeric receptor comprises an extracytoplasmic portion, a transmembrane portion, and an intracytoplasmic portion, wherein the extracytoplasmic portion comprises, in order: a signal peptide containing an intron sequence, a first sequence encoding for a variable region of a light chain of the anti-GD.sub.2 immunoglobulin M (IgM) antibody, a linker that allows folding of a GD.sub.2 antigen recognition region of an anti-GD.sub.2 IgM antibody, and a second sequence encoding for a variable region of a heavy chain of the anti-GD.sub.2 IgM antibody, wherein the transmembrane portion and the intracytoplasmic portion comprise, in order: a hinge and transmembrane domain of a human lymphocyte CD8 molecule, an intracellular portion of a 4-1BB co-stimulatory molecule, and an intracellular portion of a human lymphocyte CD3- molecule, and wherein at least 20% of the modified effector cells comprises a CD3+/CD8+/CD56+ phenotype.

6. The polyclonal mixture of effector cells as claimed in claim 5, wherein the chimeric receptor is encoded by the sequence set forth in SEQ ID NO: 1.

7. The polyclonal mixture of effector cells as claimed in claim 5, wherein the linker that allows folding of the GD.sub.2 antigen recognition region consists of 18 amino acids.

8. The polyclonal mixture of effector cells as claimed in claim 5, wherein the anti-GD.sub.2 IgM antibody is from clone 126.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Further features and advantages of the invention will result more clearly from the description of a modified effector cell (also known as chimeric receptor) for treatment of anti-G.sub.D2-BB- tumors, which is stably provided to effector cells for treatment of abnormal G.sub.D2-expressing cells, which is described by way of illustration and without limitation in the following description and with reference to the annexed figures, in which:

(2) FIGS. 1A-1D illustrate different disclosures in the prior art and in the field of the present invention.

(3) FIG. 2A is a detailed representation of the various functional domains of the construct encoding for the anti-G.sub.D2-BB- chimeric receptor (and its anti-G.sub.D2-TM truncated version) where:

(4) The construct portion encoding for the extracytoplasmic region of the CR is composed of:

(5) a signal peptide containing an intron sequence;

(6) the sequence encoding for the variable region of the light chain of the anti-G.sub.D2 IgM antibody (clone 126);

(7) a linker of 18 amino acids that allow proper folding of the G.sub.D2 antigen recognition region;

(8) the sequence encoding for the variable region of the heavy chain of the anti-G.sub.D2 IgM antibody (clone 126);

(9) The construct portion encoding for the transmembrane region and the intracellular region of the desired anti-G.sub.D2-BB- CR molecule is composed of:

(10) the transmembrane region of the TRC receptor of cytotoxic T cells (CD8+);

(11) the intracellular portion of the 4-1 BB co-stimulatory molecule (CD137);

(12) the intracellular portion of the human lymphocyte CD3 molecule.

(13) FIG. 2B is a DNA sequence encoding for the molecule known as anti-G.sub.D2-BB- chimeric receptor in its full-length form, composed of 1612 base pairs. The scFv portion is in bold text, the STM is in italics. The truncated form of the CR, known as anti-G.sub.D2-TM CR, which lacks the intracytoplasmic activation domain, indicated in underlined text, is composed of a DNA sequence having a full length of 1087 base pairs.

(14) FIG. 3 is a schematic representation of the invention, showing that the previously described sequence encodes for a CR characterized by an extra-cytoplasmic region, a transmembrane region and an intra-cytoplasmic region. The extra-cytoplasmic region is required for G.sub.D2 antigen recognition on the surface of target tumor cells; the intra-cytoplasmic region has the role of activating and maintaining the cytolytic stimulus of the effector cells upon bonding with the G.sub.D2 antigen; the transmembrane region binds these two portions and allows proper localization thereof on the cell membrane.

(15) FIG. 4 is an analysis of the efficiency of transduction of the ECs with the three different constructs, by assessment of the amount of fluorescence emitted by the green fluorescent protein marker (GEP) encoded by the vector we used. The constructs encode for the (full-length) anti-G.sub.D2-BB- CR; the (truncated) anti-G.sub.D2-TM CR respectively; in addition a construct is provided which does not express the CR, known as an empty vector, that only expresses the GFP protein, and is used as a control (CN).

(16) FIG. 5 is a schematic representation of the technique of separation of the anti-G.sub.D2-BB- CR-expressing modified effector cells. The CR is recognized by anti-idiotypic antibodies, obtained by immunization of BALB/c mice using the Gene Gun method. After immunization, the mice developed anti-idiotypic antibodies that recognize the scFv (extracytoplasmic) region of the anti-G.sub.D2-BB- CR. A second antibody (IgG) of different animal origin (rat), conjugated with immunomagnetic beads, can recognize and bind with the anti-idiotypic antibody. This system has been used to purify, by magnetic separation, the population of efficiently transduced effector cells, which express the anti-G.sub.D2-BB- CR, from those that have not been transduced.

(17) FIG. 6 is a representation of the efficiency with which the effector cells are infected by the various vectors before and after cell sorting/separation. The detection of the GFP protein marker indirectly indicates vector transduction efficiency (gray and white bars), whereas direct detection of the efficiency of CR transduction on the surface of effector cells is provided using the anti-idiotypic antibody, followed by an antibody conjugated with a fluorochrome (striped and black bars).

(18) FIG. 7 shows data concerning the phenotypic characterization of the effector cells performed both after separation of such effector cells on the gradient of peripheral blood (day 0, gray bars) and after the transduction process, with the vectors encoding for CR (day 15, black bars).

(19) FIG. 8 shows a cytofluorimetry analysis characterization of the level of membrane expression of the G.sub.D2 antigen by various tumor cell lines. Particularly, neuroblastoma cell lines (SH-SY-5Y, SKnBE) typically express disialoganglioside G.sub.D2 on the membrane, whereas the cervix carcinoma cell line (Hela) does not express G.sub.D2.

(20) FIG. 9 is a detail of the disialoganglioside G.sub.D2 expression on the surface of the SH-SY5Y neuroblastoma tumor line, analyzed by immunofluorescence using a fluorescence microscope (10 magnification). The following antibodies were used for such assessment: primary anti-G.sub.D2 antibody (BD); secondary goat anti-mouse antibody, anti-mouse (all Ig classes) conjugated with rhodamine.

(21) FIG. 10 is an analysis of the cytotoxic ability of effector cells with the anti-G.sub.D2-BB- chimeric receptor or the receptor in the anti-G.sub.D2-TM truncated form, against tumor cell lines, as measured by the radioactive chromium.sup.51 release array, after 4 hours co-culture.

(22) FIG. 11 is an assessment of the reduction of G.sub.D2 positive tumor cells caused by the long-term (6-day) co-culture condition with modified effector cells.

(23) FIG. 12 is an assessment of the expression of the receptor molecule CD25 (Interleukin 2, IL-2 receptor) on the surface of effector cells active by tumor cell recognition.

DETAILED DESCRIPTION OF A PREFERRED EMBODIMENT

(24) According to the invention gene therapy and adoptive immunotherapy have been combined to provide a therapeutic strategy that involves gene modification of previously ex-vivo expanded effectors by providing them with surface molecules, known as receptors, which can recognize tumor antigens and activate a specific cytolytic activity against them.

(25) Gene Therapy with Chemical Receptor-Expressing Effector Cells.

(26) This approach consists in gene modification of effector cells using gene sequences, known as vectors that can induce the expression of surface molecules, known as chimeric receptors (CR). The CRs are transmembrane structures composed of an intra-cytoplasmic component and an extra-cytoplasmic component which acts as an anchor and can recognize and specifically bind the tumor antigen of interest. The extra-cytoplasmic region, which is responsible for recognition of the tumor Ag, consists of a variable fragment (scFv) of murine Ag-specific immunoglobulins (Ig), whereas the intra-cytoplasmic region consists of parts of molecules of human origin, which are responsible for immune response activation. The junction of these two basic functions provides a new molecule, defined as a chimeric molecule, that has the advantage of high specificity and efficiency and by-passes tumor escape mechanisms. Once the binding of the chimeric receptor and the specific antigen has taken place, the CR transmits the activation signal into the effector cell, thereby triggering the immune response in direct, highly specific fashion.

(27) Creation of the Anti-G.sub.D2-BB- CR.

(28) The construct (DNA chain) encoding for the anti-G.sub.D2-BB- CR is basically composed of two portions: an extra-cytoplasmic component that can specifically recognize the antigen sequence of interest (disialoganglioside G.sub.D2) and an intra-cytoplasmic component that can effectively transmit the effector cell activation signal.

(29) Then, the sequences of the two portions of the molecule known as anti-G.sub.D2-BB- CR were assembled by the technique known in the art as PCR Splicing by Overlap Extension (SOE-PCR). The cDNA encoding for the scFv region of the anti-G.sub.D2 monoclonal IgM, derived from hybridoma 126, was obtained by amplification of this region of the pcDNA3-G.sub.D2-hSIP molecule.

(30) On the other hand, the intra-cytoplasmic region is composed of various parts, including: the transmembrane portion of the TRC receptor of cytotoxic T lymphocytes (CD8a); the intracellular portion of the 4-1 BB co-stimulatory molecule and the intracellular activating portion of the lymphocyte OO3- molecules, which form together the Signal Transduction Machinery (STM).

(31) The two components of such chimeric receptor were assembled by using the Splicing by Overlap extension (SOE-PCR) technique. Such process provided the construct encoding for the anti-G.sub.D2-BB- CR. A truncated form of this CR was also generated. Such truncated form lacks the intra-cytoplasmic region comprising the 4-1 BB e CD3- molecules and only has the transmembrane portion of the CD8 molecule of the TRC of the cytotoxic T lymphocytes (CD8).

(32) The truncated form of the CR, known as anti-G.sub.D2TM CR, was generated by amplifying the DNA of the construct encoding for the anti-G.sub.D2-BB- CR, but using a different reverse primer (5-GCCTTAAGGCTTAGCAGTAAAGGGTGATAACCAGTGACAG-3) that contains a restriction site for the EcoRI enzyme in its 5 region. The construct so formed is shown in FIG. 2A, whereas the full-length DNA sequence encoding for the CR is described in FIG. 2B.

(33) Generation of the Viral Vector to Create the Anti-G.sub.D2 CR-Expressing Effector Cell.

(34) A population of carrier cells that can stably produce a pool of retroviral particles capable of infecting the effector cell population of interest, was created through two steps.

(35) The first step based on the obtainment of a cell line producing the retrovirus in transient mode and the second step aimed at obtaining the generation of a Producer Cell Line (PCL) capable of stably producing a retroviral progeny. For the transient step, embryonic renal fibroblasts (293 T cells), held at about 70% confluence, were transfected with a solution of 5 g (in a 25 cm.sup.2 flask) total plasmid DNA and with the help of polycations.

(36) Later, the retroviral supernatant obtained by transient transfection of 293T was collected and used to infect the PLC deriving from a human fibrosarcoma line; 24 h after infection the cells were analyzed by cytofluorimetry to check positivity of green fluorescent protein (GFP), an infection efficiency marker.

(37) The viral supernatant collected from the PCLs, was used to infect the effector cells.

(38) Transformation of Effector Cells.

(39) The effector cells (hereinafter shortly referred to as ECs) were separated from the rest of the peripheral blood cell components, by density gradient centrifugation (Ficoll) and were later pre-stimulated for 48 hours with Interleukin 2 (IL-2) and phytoematoagglutinin (PHA-M), as retroviruses only exhibit a high infecting power in the cells during active replication.

(40) The ECs, adhered to a plastic support using retronecting, were submitted to three infection hits, each 24 h, with the viral supernatant obtained from the PCLs and later showed a strong CR surface expression ability (48% average infection efficiency).

(41) This process was also carried out to produce ECs expressing the CR in its truncated form; therefore, three types of modified ECs were created:

(42) ECs transduced with the empty vector, known as EC-CN (not recognizing G.sub.D2 but expressing the GFP);

(43) ECs expressing the anti-G.sub.D2TM CR (recognizing G.sub.D2 but not reacting against cells having such antigen);

(44) ECs expressing the anti-G.sub.D2-BB- CR, (capable of both recognizing G.sub.D2-positive cells, and having a cytocidal action thereagainst).

(45) At the end of the transduction step, the ECs were left in their standard medium for a few days before cytofluorimetric analysis to check for expression of the GFP protein marker (FIG. 4) and later of the CR on the cell surface.

(46) The CR expression was determined by sequential use of two antibodies. The first is an anti-idiotypic antibody of murine origin that can recognize the extra-cytoplasmic region of the CR (particularly the variable fraction of the anti-G.sub.D2 antibody).

(47) Then, rat polyclonal fluorochrome-marked antibodies where used, capable of being associated with the anti-CR IgG antibodies and allowing cytofluorimetry analysis.

(48) Such anti-idiotype was obtained by immunization of BALB/C mice that received gold particles having a part of the CR-encoding DNA previously attached thereto, at skin level, by gene gun delivery.

(49) The gold particles promote development of a local immune reaction mediated by skin dendritic cells, which phagocytize these particles and process the DNA attached thereto, to lend them to the B-lymphocytes that activate a specific antibody response. The immunized mice produced relatively large amounts of anti-idiotypic polyclonal antibodies in the serum, that could recognize the scFv portion of the CR expressed on the ECs.

(50) Purification of the Membrane Chimeric Receptor-Expressing Lymphocyte Population by an Immunomagnetic Method.

(51) In order to implement purity of the chemical receptor-expressing EC population, and hence to exceed the 50% threshold, averagely obtained by infection thereof by retroviruses, a method was used of separating the transformed cells from the remaining cell population.

(52) The chimeric cell population is separated from the total population using MACS LD columns for immunomagnetic depletion sorting.

(53) The operation includes marking of CR-expressing ECs with a primary antibody, which is later recognized by a secondary antibody bonded to a ball, which is retained in the column by a magnetic charge.

(54) The bond thus obtained is given by the CR on the ECs with a (mouse anti-human) anti-idiotypic antibody, and a final antibody conjugated with the immunomagnetic balls (rat anti-mouse IgG1 MicroBeads) which retain the modified cells in the column inserted in the magnetic support, whereas unmodified cells move beyond the length of the column and may be thus separated (FIG. 5).

(55) By simply removing the column from the magnetic support, the CR-expressing ECs fall off, are collected and maintained in cell culture. This procedure can enrich the chimeric lymphocyte population from 50% to 70-80% (i.e. by about 20%) (FIG. 6).

(56) Analysis of Disialoganglioside G.sub.D2 Expression on Neuroblastoma Tumor Cells.

(57) Two neuroblastoma tumor cell lines are available at the laboratory (FIG. 8):

(58) SH-SY-5Y: clone isolated in 1970 from a 4 year-old girl (metastasized mass at bone level), 3.sup.rd subclone obtained from the SK-N-SH neuroblastoma cell line.

(59) SK-n-BE: clone isolated in 1972; a 2 year old boy with disseminated neuroblastoma (bone marrow biopsy) even after repeated chemotherapy cycles. The SK-N-BE (1) and (2) lines are obtained from the same patient but 5 months apart.

(60) The human cervix carcinoma cell line (Hela), with no surface-expressed target antigen, was considered as a control target line.

(61) Detection was carried out for the target G.sub.D2 antigen by cytofluorometry. Namely, tumor cells were first contacted with an antibody capable of recognizing the membrane-expressed G.sub.D2 antigen (Purified mouse anti-human disialoganglioside G.sub.D2 monoclonal antibodyBD), followed by the addition of an antibody defined as secondary antibody, capable of fluorescence emission after bonding with the primary antibody (APC Goat Anti-Mouse Ig polyclonalmultiple adsorption, BD).

(62) This analysis showed that all analyzed NB lines express relatively large amounts of G.sub.D2 in their membrane, respectively 99% SH-SY5Y and 33% SK-n-BE, whereas the Hela cell line confirmed the assumed absence of G.sub.D2 on the cell surface.

(63) Therefore, the tumor lines being considered can be suggested as optimal targets for the kind of therapy involved in this invention.

(64) Study of the Action of the Transformed Lymphocyte on Tumor Cell Lines and/or Primary Tumor Cells.

(65) The last invention implementation step is characterized by the study of the action of genetically modified effector cells on tumor cell lines and/or primary tumor cells, through the development of a number of assays for assessment of cytotoxicity, specificity and proliferative abilities.

(66) According to the invention, various laboratory studies have been conducted:

(67) study of the action of modified ECs against tumor cell lines in a short-term (4 h) cytotoxicity assay;

(68) study of the action of modified ECs against tumor cell lines in a long-term (6-day) cytotoxicity assay;

(69) assessment of actual activation of modified ECs after short-term (24 h) contact with tumor cell lines.

(70) 1. Study of the Action of Modified ECs Against Tumor Cell Lines in a Short-Term (4 h) Cytotoxicity Assay.

(71) This study is aimed at assessing the cytolytic action of the ECs against G.sub.D2-positive tumor cell lines, by an in-vitro cytotoxicity assay with Chromium.sup.51 (Cr.sup.51) conducted 4 hours after co-culture.

(72) The target cells were marked with an appropriate amount of Cr.sup.51, and the effector cells were later added in the desired ratios.

(73) The cell supernatant collected at the end of the 4 hours' co-culture contained free Cr.sup.51 amounts directly related to the amount of the tumor cells lysed by the ECs.

(74) The cell supernatants were analyzed using MicroBeta Trilux, and the percent release of specific Cr.sup.51 was calculated using the appropriate formula.

(75) The data obtained by the Cr.sup.51 assay shows a different cytotoxic activity of ECs with the anti-G.sub.D2-BB- CR as compared with those modified with the empty vector (CN), with respect of the SH-SY-5Y line, particularly after 4 h co-culture, with a Target cell: Effector cell (T:E) ratio of 1 to 20: the anti-G.sub.D2-BB- CR ECs showed a significant cytotoxic effect, i.e. 42.7%, against 11-7% obtained with the CN ECs (p=0.05) (FIG. 10).

(76) Likewise, the cytotoxic reaction against the SKnBE line by the anti-G.sub.D2-BB- CR ECs as compared with those modified with the empty vector (CN), or with the truncated anti-G.sub.D2-TM CR construct, proved to be highly specific and significant. The anti-G.sub.D2-BB- CR ECs exhibit a high short-time cytotoxic effect (8825%) which is statistically significant when compared with the cytotoxic effect obtained with the other two types of effector cells; particularly having a significance p<0.01 when compared with the anti-G.sub.D2-TM CR ECs; and a significance p=0.05 against CN ECs).

(77) The lack of a cytotoxic effect of ECs against the G.sub.D2-negative tumor cell (Hela) proves the high specificity of the cytolytic action of the ECs modified for expression of the anti-G.sub.D2-BB- CR.

(78) 2. To confirm such data, cell populations were characterized by cytofluorometry, at the end of 6-day co-cultures with a low E:T ratio=5:1. A significant decrease (from 92% to 239%, p<0.004) of G.sub.D2 expression was noted on the surface of NB cells from the SH-SY-5Y line, in the presence of anti-G.sub.D2-BB- CR ECs; whereas with the ECs with the anti-G.sub.D2-TM CR this percent remained almost unchanged (from 92% to 63.820%) (FIG. 1 1).

(79) A significant reduction in G.sub.D2 expression (from 37% to 9.024.4%) with anti-G.sub.D2-BB- CR ECs was also noted against the G.sub.D2-positive SKnBE tumor line; conversely, with the CN control cells, this percent had no significant decrease (from 37% to 29.4817%).

(80) 3. Study of the Activation of the Transformed Lymphocyte Against Tumor Cell Lines.

(81) The activating response of effector cells is typically also characterized by up-regulation of the IL-2 receptor (CD25) on the cell surface.

(82) This signal promotes cell proliferation, as well as cytokine production and secretion of cytolytic granules, with a consequent cytotoxic effect on target cells.

(83) In order to compare the different abilities of the two CRs to induce antitumor responses, we analyzed these parameters in response to the interaction between the ECs and the G.sub.D2-positive neuroblastoma lines.

(84) Short-term EC activation was assessed after a 24 h co-culture by cytofluorometry. The ECs were differentiated from the target cell for their expression of the GFP protein.

(85) After a 24 hours' co-culture with low E:T ratios (here E:T=3:1) we noted a significant up-regulation of the CD25 molecule on the surface of ECs with the anti-G.sub.D2-BB- CR as compared with both those modified with the anti-G.sub.D2-TM CR, and with the CN ECs (FIG. 12).

(86) Particularly, after a co-culture with SH-SY5Y cells, the ECs modified with the anti-G.sub.D2-BB- CR showed a high CD25 expression (551.3%) as compared with that detected with the anti-G.sub.D2-TM CR (441.1%) and with the CN control cells (19.40.9%).

(87) These differences are also statistically significant (p<0.01).

(88) The same considerations apply concerning EC activation triggered by the G.sub.D2-positive SKnBE tumor line, with no specific activation being found by the modified ECs against the G.sub.D2-negative Hela tumor line, to confirm that the activation of these CRs allow specific selection of G.sub.D2-positive tumor cells, with optimized cytolytic response by the effector cells modified with the anti-G.sub.D2-BB- CR.

(89) The invention has been found to fulfill the intended objects.

(90) The invention is susceptible to changes and variants within the inventive concept.

(91) Furthermore, all the details may be replaced by other technically equivalent elements, as needed, without departure from the scope as defined by the following claims.