(19)
(11)EP 3 546 567 A1

(12)EUROPEAN PATENT APPLICATION

(43)Date of publication:
02.10.2019 Bulletin 2019/40

(21)Application number: 18214552.4

(22)Date of filing:  05.05.2015
(27)Previously filed application:
 06.12.2014 US 201462088578 P
(51)Int. Cl.: 
C12N 5/0735  (2010.01)
C12N 5/077  (2010.01)
C12N 5/074  (2010.01)
C12N 5/0775  (2010.01)
(84)Designated Contracting States:
AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

(30)Priority: 06.12.2014 US 201462088578 P

(62)Application number of the earlier application in accordance with Art. 76 EPC:
15166347.3 / 3029137

(71)Applicant: GEMoaB Monoclonals GmbH
01307 Dresden (DE)

(72)Inventor:
  • Ehninger, Armin
    01326 Dresden (DE)

(74)Representative: Kailuweit & Uhlemann Patentanwälte Partnerschaft mbB 
Bamberger Straße 49
01187 Dresden
01187 Dresden (DE)

 
Remarks:
This application was filed on 20-12-2018 as a divisional application to the application mentioned under INID code 62.
 


(54)GENETIC MODIFIED PLURI- OR MULTIPOTENT STEM CELLS AND USES THEREOF


(57) The invention concerns pluri- or multipotent stem cells (SCs), e.g. human pluri- or multipotent stem cells (hSCs) engineered to express a multispecific antibody and which further expresses, on its surface, a human immune cell co-stimulatory ligand or an active fragment thereof.


Description

1. BACKGROUND



[0001] The use of recombinant bispecific antibodies (bsAbs) for retargeting effector T lymphocytes towards cancer cells is recently emerging as a promising immunotherapeutic tool for the treatment of hematologic malignancies. The high efficacy of this class of molecules in redirecting specifically CD8+ and CD4+ T cells to any chosen tumor associated antigen (TAA) on the surface of tumor cells by cross-linking their activating CD3 receptor has been reported in many preclinical and clinical studies. Nonetheless, due to their low molecular mass, bsAbs have shown a short life span in vivo and consequently have to be continuously administrated to patients over prolonged time spans of several weeks to achieve clinical responses. There is a need for improved therapies using bsAbs for treating disorders such as cancer.

2. SUMMARY



[0002] In one aspect, provided herein are mesenchymal stem cells (MSCs), such as human MSCs (hMSCs), engineered to express an antibody, e.g., multispccific antibody, such as a bispecific antibody targeting CD33 and CD3, and, optionally, a human T cell co-stimulatory ligand, such as the human T cell co-stimulatory 4-1BB ligand (CD137L) or an active fragment thereof. In specific aspects, also provided herein are methods for treating acute myeloid leukemia (AML) using such hMSCs, as well as related methods for inducing/enhancing T cell proliferation and related methods for activating T cells.

3. BRIEF DESCRIPTION OF THE FIGURES



[0003] 

Figure 1: BsAb CD33-CD3 specificity. (A) Schematic protein structure of the humanized recombinant bispecific anti-CD33-anti-CD3 Ab. (B) Direct cross-linking of T cell and CD33 antigen expressing AML blasts via the bsAb. (C) Efficacy of bsAb-mediated tumor cell killing evaluated via chromium release assay. 51Cr labeled antigen-positive or -negative cells were incubated with CD3+ T cells at effector-to-target (e:t) cell ratio of 5:1 in the presence of the bsAb at indicated concentrations. Data reported as the mean ± SD of triplicate wells.

Figure 2: Generation of bsAb CD33-CD3 releasing gene-modified hMSCs. (A) Intracellular EGFP signal correlating with the transgene expression in gene-modified MSCs was analyzed by flow cytometry. Positive cells are shown as percentages of all analyzed cells, after dead cells exclusion. (B) Quantitative analysis of the bsAb released by 48 h-cultured MSCs performed by ELISA. Data shown as the mean ± SD of two independent experiments.

Figure 3: Efficient T cell-mediated AML cell killing elicited by in situ hMSC-produced bsAb. (A) T cells and 51Cr labeled CD33high target cells were co-incubated at an e:t ratio of 5:1 together with 48 h-cultured wild type (wt), vector control or bsAb-releasing hMSCs (CD33-CD3) seeded at different concentrations. (B) Specific cell lysis of target AML cells was analyzed at early time points of 10 h or 20 h. Data as the mean ± SD of three individual donors.

Figure 4: Tumor cell killing is enhanced by the presence of a co-stimulatory signal. (A) Improvement of the bsAb-releasing cell system via transgenic expression of the T cell co-stimulatory ligand 4-1BBL on hMSCs surface. (B) Secretion of pro-inflammatory cytokines by T cells was determined after 24 h culture upon their cross-linkage with CD33+ target cells via hMSC-produced CD33-CD3 bsAb in the presence or absence of the 4-1BBL-expressing hMSCs. (C) Percentage of surviving eFluor670 proliferation dye-labeled CD33low tumor cells after 96 h of incubation with T cells at an e:t ratio of 1:1 in the presence or absence of gene-modified hMSCs seeded at limiting densities 48 h prior to the experiment. Summary of the results of four different donors. **p<0.01; ***p<0.001.

Figure 5: Redirection of AML patient-derived T cells against autologous AML blasts via bsAb-releasing hMSCs. (A) 1x105 AML patient-derived MNCs were cultured together with 48 h pre-seeded 1x104 hMSCs. After 96 h of co-cultivation the percentages of surviving HLA-DR+ AML blasts and CD3+ T cells were determined respectively as proportions of all CD45+ cells by flow cytometry analysis. (B) Total numbers of CD3-HLA-DR+CD45+ AML blasts detected after 96 h of co-cultivation with or without control/bsAb-releasing and 4-1BBL-expressing hMSCs are reported for three independent donors. (C) Absolute autologous T cell number was measured after 96 h and overall expansion of the cells in the presence of hMSCs was determined. Data are presented as means ± SD from three different donors. **p<0.01; ***p<0.001.


4. DETAILED DESCRIPTION



[0004] In one aspect, provided herein are human mesenchymal stem cells (hMSCs) engineered to express a multispecific antibody, such as a bispecific antibody targeting CD33 and CD3, and, optionally, a human T cell co-stimulatory ligand, such as the human T cell co-stimulatory 4-1BB ligand (CD137L) or an active fragment thereof. In specific aspects, also provided herein are methods for treating acute myeloid leukemia (AML) using such hMSCs, as well as related methods for inducing/enhancing T cell proliferation and related methods for activating T cells using such hMSCs.

[0005] In certain embodiments, the presently disclosed methods and uses enhance the T cell response against a cancer cell. Without being bound by theory, the improved T cell response against a cancer cell is accomplished by targeting of hMSCs to the cancer cell in a patient. In certain embodiments, at least 1x105, 2x105, 3x105, 4x105, 5x105, 6x105, 7x105, 8x105, 9x105, 1x106, 2x106, 3x106, 4x106, 3x106, 6x106, 7x106, 8x106, 9x106, or at least 1x107 MSCs as described herein are administered to a patient. In certain embodiments, at most 1x105, 2x105, 3x105, 4x105, 5x105, 6x105, 7x105, 8x105, 9x105, 1x106, 2x106, 3x106, 4x106, 5x106, 6x106, 7x106, 8x106, 9x106, or at least 1x107 MSCs as described herein are administered to a patient. In a more specific embodiment between 5x106 and 6x106 MSCs as described herein are administered to a patient.

[0006] In certain embodiments, at least 1mg, 2 mg, 3mg, 4mg, 5mg, 6mg, 7mg, 8mg, 9mg, or at least 10mg cell mass of an MSC as described herein can be transplanted into a patient. In certain embodiments, at most 1mg, 2 mg, 3mg, 4mg, 5mg, 6mg, 7mg, 8mg, 9mg, or at least 10mg cell mass of an MSC as described herein can be transplanted into a patient. In certain embodiments, about 1mg, 2 mg, 3mg, 4mg, 5mg, 5.5mg, 6mg, 7mg, 8mg, 9mg, or at least 10mg cell mass of an MSC as described herein can be transplanted into a patient.

[0007] MSCs can be used as in situ producer of anti-cancer therapeuticals. In certain embodiments, autologous gene-modified MSCs as described herein can be injected intravenously into patients. Without being bound by theory, thus promoting the migration to the tumor site and the local delivery of their payload. In other embodiments, MSCs could be trapped into an artificial scaffold matrix and transplanted subcutaneously in proximity of a tumor site. Again, without being bound by theory, thus the potential supportive role in tumor angiogenesis processes could be reduced, ensuring that transplanted MSCs cannot escape from their confined artificial environment. An additional benefit of this strategy can be that the delivery of the anti-cancer agent can be controlled and stopped after tumor clearance by removing the scaffold. Compte et al. [(2007) Cancer Gene Ther: 14[4]: 380-388.].

[0008] In certain embodiment, the use of MSCs to in situ production of anti-cancer therapeuticals (eg, bispecific antibodies) results in continuous delivery of the anti-cancer therapeutical in a patient. More specifically, such continuous production can be production for at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, or 14 days, at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 weeks, or at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, or 12 months, or at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 years. More specifically, such continuous production can be production for at most 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, or 14 days, at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 weeks, or at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, or 12 months, or at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 years.

Mesenchymal Stem Cells



[0009] Any mesenchymal stem cells known to the skilled artisan can be used with the methods and compositions provided herein. For example, methods for isolating primary hMSCs have been described, see, e.g., Penfornis and Pochampally, 2011, Methods Mol Biol., 698:11-21; and Qiao et al., 2008, Cell Biol. International, 32: 8-15. Mesenchymal stem cells, or MSCs, are multipotent stromal cells that can differentiate into a variety of cell types, including bone, cartilage, and fat cells. MSCs can be derived from bone marrow.

[0010] In a specific aspect, provided herein are mesenchymal stem cells (MSCs), such as human MSCs (hMSCs), recombinantly expressing an antibody (e.g., multispecific antibody, such as bispecific antibody, for example, anti-CD33-anti-CD3 bispecific antibody), and optionally a T cell co-stimulatory ligand, such as the co-stimulatory 4-1BB ligand (CD137L).

[0011] In a particular aspect, hMSCs provided herein engineered to express multispecific antibodies, e.g., bispecific antibodies (e.g., bispecific antibodies targeting CD33 and CD3) are derived from the single-cell-picked clone 1 (SCP-1) cell line (Böcker et al., 2008, J Cell Mol Med: 12: 1347-1359), derived from human mesenchymal stem cells and immortalized by lentiviral transduction of the gene coding for the human telomerase reverse transcriptase (hTERT).

[0012] In a certain aspect, hMSCs provided herein engineered to express multispecific antibodies, e.g., bispecific antibodies (e.g., bispecific antibodies targeting CD33 and CD3) are derived from primary hMSCs.

[0013] For the generation of permanent hMSCs expressing multispecific antibodies, e.g., bispecific antibodies (e.g., bispecific antibodies targeting CD33 and CD3), cDNA encoding a recombinant antibody (e.g., multispecific antibody, such as bispecific antibody, for example, anti-CD33-anti-CD3 bispecific antibody) construct can, for example, be cloned into a a viral vector (e.g., self-inactivating retroviral vector), such as a retroviral vector or lentiviral vector (e.g., p6NST50) to generate a transfer vector. In this viral vector, the expression of a recombinant antibody (e.g., multispecific antibody, such as bispecific antibody, for example, anti-CD33-anti-CD3 bispecific antibody) can be driven by a suitable promoter, such as a mammalian promoter, e.g., spleen focus forming virus (SFFV)-derived internal promoter. In certain embodiments, an enhanced green fluorescence protein (EGFP)-Zeocin fusion protein can be co-expressed by an internal ribosomal entry site (IRES). Lentiviral particles pseudotyped with an envelope protein, e.g., the Vesicular Stomatitis Virus envelope 120 (VSV-G), can be generated by transient transfection of a packaging cell, e.g., HEK293T cells, and virus supernatant are harvested and used to stably transduce hMSCs, such as SCP-1 cells.

[0014] For ectopic expression of a co-stimulary ligand, e.g., the co-stimulatory 4-1BB ligand (CD137L), on the surface of hMSCs, such as SCP-1 cells, a viral vector, such as a retroviral vector or a lentiviral vector, can be used which harbors an appropriate promoter, such as an internal minimal human elongation factor 1alpha promoter.

[0015] In a particular aspect, hMSCs can be co-transfected with two or more expression vectors (e.g., viral vectors) encoding polypeptides that associate (e.g., covalently or noncovalently) to form antigen binding domains (e.g., antigen-binding domains comprising a VH and a VL) of an antibody (e.g., multispecific antibody, such as bispecific antibody, for example, anti-CD33-anti-CD3 bispecific antibody) expressed by hMSCs.

[0016] In certain embodiments, at least 1pg, 5pg, 10pg, 50pg, 75pg, 100pg, 250pg of antibody are produced per cell over a period of 48 hours. In other embodiments at most 1pg, 5pg, 10pg, 50pg, 75pg, 100pg, 250pg of antibody are produced per cell over a period of 48 hours. In certain embodiments, 5-10, 5-50, 10-50, 25-75, or 8-70 pg of antibody are produced per cell over a period of 48 hours.

Antibodies



[0017] In specific aspects, provided herein are antibodies, such as bispecific antibodies, which are recombinantly expressed by hMSCs. In a particular embodiment, such bispecific antibodies target CD33, e.g., human CD33. In a specific embodiment, such bispecific antibodies target CD33, e.g., human CD33, and CD3, e.g., human CD3. In certain aspects, a bispecific antibody provided herein is a single chain bispecific antibody. In a particular embodiment, a bispecific antibody provided herein, such as a single chain bispecific antibody, is humanized (e.g., humanized bispecific antibody targeting CD33 and CD3).

[0018] As used herein, the terms "variable region" or "variable domain" refer to a portion of an antibody, generally, a portion of an antibody light or heavy chain, typically about the amino-terminal 110 to 120 amino acids in a mature heavy chain and about the amino-terminal 90 to 100 amino acids in a mature light chain. Variable regions comprise complementarity determining regions (CDRs) flanked by framework regions (FRs). Generally, the spatial orientation of CDRs and FRs are as follows, in an N-terminal to C-terminal direction: FR1-CDR1-FR2-CDR2-FR3-CDR3-FR4. Without wishing to be bound by any particular mechanism or theory, it is believed that the CDRs of the light and heavy chains are primarily responsible for the interaction of the antibody with antigen and for the specificity of the antibody for an epitope. In a specific embodiment, numbering of amino acid positions of antibodies described herein is according to the EU Index, as in Kabat et al. (1991) Sequences of Proteins of Immunological Interest, Fifth Edition, U.S. Department of Health and Human Services, NIH Publication No. 91-3242. In certain embodiments, the variable region is a human variable region. In certain embodiments, the variable region comprises murine (e.g., mouse or rat) CDRs and human framework regions (FRs). In particular embodiments, the variable region is a primate (e.g., human or non-human primate) variable region. In certain embodiments, the variable region comprises murine (e.g., mouse or rat) CDRs and primate (e.g., human or non-human primate) framework regions (FRs). As a non-limiting example, a variable region described herein is obtained from assembling two or more fragments of human sequences into a composite human sequence.

[0019] In certain aspects, the CDRs of an antibody can be determined according to (i) the Kabat numbering system (Kabat et al. (1971) Ann. NY Acad. Sci. 190:382-391 and, Kabat et al. (1991) Sequences of Proteins of Immunological Interest, Fifth Edition, U.S. Department of Health and Human Services, NIH Publication No. 91-3242); or (ii) the Chothia numbering scheme, which will be referred to herein as the "Chothia CDRs" (see, e.g., Chothia and Lesk, 1987, J. Mol. Biol., 196:901-917; Al-Lazikani et al., 1997, J. Mol. Biol., 273:927-948; Chothia et al., 1992, J. Mol. Biol., 227:799-817; Tramontano A et al., 1990, J. Mol. Biol. 215(1):175-82; and U.S. Patent No. 7,709,226); or (iii) the ImMunoGeneTics (IMGT) numbering system, for example, as described in Lefranc, M.-P., 1999, The Immunologist, 7:132-136 and Lefranc, M.-P. et al., 1999, Nucleic Acids Res., 27:209-212 ("IMGT CDRs"); or (iv) MacCallum et al., 1996, J. Mol. Biol., 262:732-745. See also, e.g., Martin, A., "Protein Sequence and Structure Analysis of Antibody Variable Domains," in Antibody Engineering, Kontermann and Dübel, eds., Chapter 31, pp. 422-439, Springer-Verlag, Berlin (2001).

[0020] In a specific aspect, provided herein are hMSCs expressing multispecific antibodies, e.g., bispecific antibodies, such as single chain bispecific antibodies, comprising a VL and a VH targeting CD33 (e.g., human CD33). In specific aspects, multispecific antibodies, such as, bispecific antibodies, e.g., single chain bispecific antibodies, described herein comprises one or more linkers (e.g., one, two, three, four, or five, or more linkers). In a certain embodiment, such a linker described herein comprises repetitions, for example, at least one, two, three, four or five repetitions, of glycine-serine, e.g., G4S linkers. In certain specific embodiments, a bispecific antibody for use with the methods and compositions dislcosed herein consists of or comprises the following structure: (humanized heavy chain variable region of anti-CD33)-(G4S)3-(humanized light chain variable region of anti-CD33)-AAARG-(humanizcd heavy chain variable region of anti-CD3)-(G4S)3-(humanized light chain variable region of anti-CD3). In certain even more specific embodiments, such a bispecific antibody further comprises at its N-terminus a signal peptide suitable for secretion of the bispecific antibody from the hMSC and/or an AAQPA sequence between the signal peptide and the humanized heavy chain variable region of anti-CD33; and/or a myc tag at the C-terminus; and/or a histidine tag at the C-terminus.

[0021] Non-limiting examples of antibodies comprising a VL and a VH targeting CD33 have been described, for example, U.S. Patent Application Publication No. US 2012/0251554 A1 (e.g., antibodies comprising CDRs of anti-CD33 antibody DRB1 or DRB2); Stamova et al., 2011, Molecular Immunol., 49: 474-482; Arndt et al., 2013, Leukemia, 27: 964-967; and Arndt et al., 2014, Leukemia, 28: 59-69, each of which is incorporated herein by reference in its entirety.

[0022] In a specific embodiment, provided herein are hMSCs expressing a multispecific antibody, e.g., bispecific antibody, such as a single chain bispecific antibody, which specifically binds to CD33 (e.g., human CD33) comprising a heavy chain variable region (VH) and a light chain variable region (VL), wherein (i) the VH comprises complementarity determining region CDR1 comprising the amino acid sequence DYVVH, CDR2 comprising the amino acid sequence YINPYNDGTKYNEKFKG, and CDR3 comprising the amino acid sequence DYRYEVYGMDY; and (ii) the VL comprises CDR1 comprising the amino acid sequence TASSSVNYIH, CDR2 comprising the amino acid sequence TSKVAS, and CDR3 comprising the amino acid sequence QQWRSYPLT.

[0023] In a specific embodiment, provided herein are hMSCs expressing a multispecific antibody, e.g., bispecific antibody, such as a single chain bispecific antibody, which specifically binds to CD33 (e.g., human CD33) comprising a heavy chain variable region (VH) and a light chain variable region (VL), wherein (i) the VH comprises CDR1, CDR2, and CDR3 comprising amino acid sequences as set forth in Table 1; and (ii) the VL comprises CDR1, CDR2, and CDR3 comprising amino acid sequences as set forth in Table 2.
Table 1: VH CDRs of representative anti-CD33 antibodies
VH CDR1VH CDR2VH CDR3
SYYIH VIYPGNDDISYNQKFXG EVRLRYFDV
DYNMH YIYPYNGGTGYNQKFKS GRPAMDY
DYNMY YIDPYKGGTIYNQKFKG QMITAYYFDY
DYVLH LINTYNGDVRYNQKFMG DYRYEYYAMDY
Table 2: VL CDRs of representative anti-CD33 antibodies
VL CDR1VL CDR2VL CDR3
KSSQSVFFSSSQKNYLA WASTRES HQYLSSRT
RASESVDNYGISFMN ASNQGS QQSKEVPWT
KASQDINKYIA TSTLQP LQYDNLLT
SANSSVSYIH TSKLAS QQWTSHPLT


[0024] In a specific aspect, provided herein are hMSCs expressing multispecific antibodies, e.g., bispecific antibodies, such as single chain bispecific antibodies, comprising (i) a VL and a VH targeting CD33 (e.g., human CD33) and (ii) a VL and a VH targeting CD3 (e.g., human CD3).

[0025] Non-limiting examples of antibodies comprising a VL and a VH targeting CD3 have been described, for example, U.S. Patent Nos. 7728114 and 5929212; and PCT International Patent Application Publication Nos. WO 2014/047231 and WO 2004/108158; each of which is incorporated herein by reference in its entirety.

[0026] In a specific embodiment, provided herein are hMSCs expressing a multispecific antibody, e.g., bispecific antibody, such as a single chain bispecific antibody, which specifically binds to CD33 (e.g., human CD33) and CD3 (e.g., human CD3), comprising a heavy chain variable region (VH) and a light chain variable region (VL) of a humanized antibody which specifically binds to human CD33, wherein (i) the VH comprises complementarity determining region CDR1 comprising the amino acid sequence DYVVH, CDR2 comprising the amino acid sequence YINPYNDGTKYNEKFKG, and CDR3 comprising the amino acid sequence DYRYEVYGMDY; and (ii) the VL comprises CDR1 comprising the amino acid sequence TASSSVNYTH, CDR2 comprising the amino acid sequence TSKVAS, and CDR3 comprising the amino acid sequence QQWRSYPLT. Non-limiting examples of fully humanized anti-CD33-anti-CD3 bispecific antibodies have been described, e.g., see Arndt et al., 2013, Leukemia, 27: 964-967.

[0027] In a specific embodiment, an anti-CD33-anti-CD3 bispecific antibody has the following configuration from the N-terminus to the C-terminus: heavy chain variable region targeting CD33 (VH/CD33)-linker1-light chain variable region targeting CD33 (VL/CD33)-linker2-heavy chain variable region targeting CD3 (VH/CD3)-linker3-light chain variable region targeting CD3 (VL/CD3).

[0028] In certain embodiments, the CDRs for the anti-CD3 portion of a multispecific or bispecific antibody are the CDRs of MT-301 (see, e.g., Arndt et al. 2014, Leukemia 28:59-69.

[0029] In specific embodiments, multispecific antibodies, such as, bispecific antibodies, e.g., single chain bispecific antibodies, described herein comprises at the N-terminus, a signal peptide, e.g., an Ig leader sequence as signal peptide, for protein secretion.

[0030] In certain embodiments, multispecific antibodies, such as, bispecific antibodies, e.g., single chain bispecific antibodies, described herein comprises at the C-terminus a tag, for example, myc-tag or and histidine (his)-tag (e..g, 6xhis-tag) for protein purification and/or detection.

[0031] In certain embodiments, antibodies (e.g., multispecific antibodies, such as bispecific antibodies) described herein can be of any type (e.g., IgG, IgE, IgM, IgD, IgA or IgY), any class, (e.g., IgG1, IgG2, IgG3, IgG4, IgA1 or IgA2), or any subclass (e.g., IgG2a or IgG2b, or a mixture thereof) of immunoglobulin molecule.

[0032] In specific embodiments, multispecific antibodies, such as, bispecific antibodies, e.g., single chain bispecific antibodies, described herein do not comprise an antibody constant region.

[0033] Methods for making multispecific (e.g, bispecific antibodies) have been described, see, for example, U.S. Patent Nos. 7951917, 7183076, 8227577, 5837242, 5989830, 5869620, 6132992, and 8586713.

Ligands of Co-Stimulatory Molecules



[0034] In a specific aspect, provided herein are hMSCs that recombinantly express an antibody (e.g., multispecific antibody, such as bispecific antibody, for example, anti-CD33-anti-CD3 bispecific antibody), and optionally a T cell co-stimulatory ligand, such as the co-stimulatory 4-1BB ligand (CD137L).

[0035] Non-limiting examples of co-stimulatory receptors having ligands which can be recombinantly expressed on the cell surface of hMSCs provided herein include, but are not limited to, those described in Table 3, e..g, CD28 ligand B7-1 and B7-2, Ox40 ligand OX40L, CD27 ligand CD70 (CD27L), LFA-1 ligand ICAM-1, -2, and -3, and SLAM ligand.
Table 3: Co-Stimulatory Receptors
Co-stimulatory MoleculeDescriptionGenbank Accession No.
4-1BB (Watts & DeBenedette, Current Opinion in Immunology, 1999. 11:286-293). U03397.1
CD28 CD28 is the primary co-stimulatory receptor for inducing high level IL-2 production and survival of naive CD4+ T cells.  
  Ligand: B7-1 and B7-2  
  (Watts & DeBenedette, Current Opinion in Immunology, 1999. 11:286-293).  
Ox40 (CD134) Promotes Th2 response; Sustains proliferation of Th1 and Th2 effectors; Enhances IL-2 and Th2 cytokine production; Up regulated on CD4+ T cells S76792.1
  Ligand: OX40L  
  (Watts & DeBenedette, Current Opinion in Immunology, 1999. 11:286-293).  
CD27 CD27 is a tumor necrosis factor receptor; Functions in T cell-B cell and T cell-T cell interations; Role in expansion of T cells after CD28 co-stimulation. NM_ 001242.4
  Ligand: CD70 (CD27L)  
  (Watts & DeBenedette, Current Opinion in Immunology, 1999. 11:286-293).  
CTLA-4 Cytotoxic T-lymphocyte-associated protein 4), also known as CD152 (cluster of differentiation 152), is a protein receptor that downregulates the immune system. CTLA4 is found on the surface of T cells. U90273.1
ICOS Ligand Inducible T-cell Costimulator (CD278) is a CD28-superfamily costimulatory molecule that is expressed on activated T cells. NM_ 012092.3
  (Watts & DeBenedette, Current Opinion in Immunology, 1999. 11:286-293).  
LFA-1 Lymphocyte function-associated antigen 1 ; Adhesion and cytoskeleton reorganization during T cell activation. NM_ 001114380.1
  Ligand: ICAM-1, -2, -3  
  Watts & DeBenedette, Current Opinion in Immunology, 1999. 11:286-293).  
SLAM Signaling lymphocytic activation molecule. SLAM is a CD2-related AY040554.1
  surface receptor expressed by activated T cells and B cells.  
  (Henning, Eur J Immunol. 2001 Sep;31(9):2741-50; Watts & DeBenedette, Current Opinion in Immunology, 1999. 11:286-293).  
  Ligand: SLAM  
PD-1 Program death 1 (PD-1) is an activation induced CD28 family member that contains an ITIM receptor. L27440.1
  (Kohlmeier & Benedict, Histol Histopathol, 2003, 18: 1195-1204).  

Uses and Methods



[0036] In specific aspects, provided herein arc methods for treating cancer, such as acute myeloid leukemia (AML) in a subject by administering a therapeutic effective amount of hMSCs expressing an antibody, e.g., multispecific antibody, such as bispecific antibody (for example, anti-CD33-anti-CD3 bispecific antibody), and optionally a T cell co-stimulatory ligand, such as the co-stimulatory 4-1BB ligand (CD137L). In certain embodiments, the cancer is a type of cancer associated with expression of CD33, such as AML, and one binding domain of the multispecific antibody is an anti-CD33 antibody. Also provided herein are related methods for inducing/enhancing T cell proliferation and related methods for activating T cells using such hMSCs.

[0037] In particular embodiments, provided herein are methods for managing, treating, preventing or protecting against AML in a subject (e.g., human subject) in need thereof, comprising administering to the subject a therapeutic effective amount of hMSCs expressing an antibody, e.g., multispecific antibody, such as bispecific antibody (for example, anti-CD33-anti-CD3 bispecific antibody), and optionally a T cell co-stimulatory ligand, such as the co-stimulatory 4-1BB ligand (CD137L).

[0038] In a particular embodiment, provided herein is a method for inducing or enhancing T cell proliferation comprising contacting T cells with an effective amount of hMSCs expressing an antibody, e.g., multispecific antibody, such as bispecific antibody (for example, anti-CD33-anti-CD3 bispecific antibody), and optionally a T cell co-stimulatory ligand, such as the co-stimulatory 4-1BB ligand (CD137L). In a specific embodiment, provided herein is a method for inducing or enhancing T cell proliferation in a subject in need thereof, comprising a therapeutic effective amount of hMSCs expressing an antibody, e.g., multispecific antibody, such as bispecific antibody (for example, anti-CD33-anti-CD3 bispecific antibody), and optionally a T cell co-stimulatory ligand, such as the co-stimulatory 4-1BB ligand (CD137L).

[0039] In particular embodiments, T cell proliferation is increased by at least 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100%, 150%, 200%, 250%, 300%, 400%, or 500% or more, with hMSCs than without hMSCs, as determined by methods known in the art, for example, those described herein (e.g., Examples Section).

[0040] In a particular embodiment, provided herein is a method for inducing or enhancing T cell activation, comprising contacting T cells with an effective amount of hMSCs expressing an antibody, e.g., multispecific antibody, such as bispecific antibody (for example, anti-CD33-anti-CD3 bispecific antibody), and optionally a T cell co-stimulatory ligand, such as the co-stimulatory 4-1BB ligand (CD137L). In a specific embodiment, provided herein is a method for inducing or enhancing T cell activation in a subject in need thereof, comprising a therapeutic effective amount of hMSCs expressing an antibody, e.g., multispecific antibody, such as bispecific antibody (for example, anti-CD33-anti-CD3 bispecific antibody), and optionally a T cell co-stimulatory ligand, such as the co-stimulatory 4-1BB ligand (CD137L).

[0041] In particular embodiments, T cell activation is increased by at least 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100%, 150%, 200%, 250%, 300%, 400%, or 500% or more, with hMSCs than without hMSCs, as determined by methods known in the art, for example, those described herein (e.g., Examples Section).

5. EXAMPLES



[0042] The examples in this section are offered by way of illustration, and not by way of limitation.

5.1 Example 1



[0043] The use of recombinant bispecific antibodies (bsAbs) for retargeting effector T lymphocytes towards cancer cells is recently emerging as a promising immunotherapeutic tool for the treatment of hematologic malignancies. The high efficacy of this class of molecules in redirecting specifically CD8+ and CD4+ T cells to any chosen tumor associated antigen (TAA) on the surface of tumor cells by cross-linking their activating CD3 receptor has been reported in many preclinical and clinical studies. Nonetheless, due to their low molecular mass, bsAbs have shown a short life span in vivo and consequently have to be continuously administrated to patients over prolonged time spans of several weeks to achieve clinical responses.

[0044] As a valid alternative to continuous exogenous infusions of short-lived Abs engineered bone marrow-derived human mesenchymal stem cells (hMSCs) can be used as cellular vehicles for the constant production of recombinant Abs. Without being bound by theory hMSCs can accumulate at the site of tumors including metastatic lesions, have limited immunogenicity.

[0045] BsAb CD33-CD3 specifity was determined. Figure 1A is a schematic protein structure of the humanized recombinant bispecific anti-CD33-anti-CD3 Ab. Figure 1B shows the direct cross-linking of T cell and CD33 antigen expressing AML blasts via the bsAb. Efficacy of bsAb-mediated tumor cell killing was evaluated via chromium release assay (Fig. 1C). 51Cr labeled antigen-positive or -negative cells were incubated with CD3+ T cells at effector-to-target (e:t) cell ratio of 5:1 in the presence of the bsAb at indicated concentrations.

[0046] BsAb CD33-CD3 releasing gene-modified hMSCs were generated. Intracellular EGFP signal correlating with transgene expression in gene-modified MSCs was analyzed by flow cytometry (Fig. 2A). Quantitative analysis of the bsAb released by 48 h-cultured MSCs was performed by ELISA and the results are shown in Figure 2B.

[0047] Efficient T cell-mediated AML cell killing was elicited by in situ hMSC-produced bsAb. T cells and 51Cr labeled CD33high target cells were co-incubated at an e:t ratio of 5:1 together with 48 h-cultured wild type (wt), vector control or bsAb-releasing hMSCs (CD33-CD3) seeded at different concentrations (Fig. 3A). Specific cell lysis of target AML cells was analyzed at early time points of 10 h or 20 h (Fig. 3B).

[0048] Tumor cell killing. Tumor cell killing is enhanced by the presence of a co-stimulatory signal. Improvement of the bsAb-releasing cell system was accomplished via transgenic expression of the T cell co-stimulatory ligand 4-1BBL on hMSCs surface (Fig. 4A). Secretion of pro-inflammatory cytokines by T cells was determined after 24 h culture upon their cross-linkage with CD33 target cells via hMSC-produced CD33-CD3 bsAb in the presence or absence of the 4-1BBL-expressing hMSCs (Fig. 4B). The percentage of surviving eFluor670 proliferation dye-labeled CD33low tumor cells after 96 h of incubation with T cells at an e:t ratio of 1:1 in the presence or absence of gene-modified hMSCs seeded at limiting densities 48 h prior to the experiment is shown in Figure 4C.

[0049] AML patient-derived T cells were redirected against autologous AML blasts via bsAb-releasing hMSCs. 1x105 AML patient-derived MNCs were cultured together with 48 h pre-seeded 1x104 hMSCs. After 96 h of co-cultivation the percentages of surviving HLA-DR+ AML blasts and CD3+ T cells were determined as proportions of all CD45+ cells by flow cytometry analysis (Fig. 5A). Total numbers of CD3-HLA-DR+CD45+ AML blasts detected after 96 h of co-cultivation with or without control/bsAb-releasing and 4-1BBL-expressing hMSCs are reported in Figure 5B for three independent donors. Absolute autologous T cell number was measured after 96 h and overall expansion of the cells in the presence of hMSCs was determined (Fig. 5C).

[0050] Conclusions. Demonstrated herewith was the feasibility of using genetically-modified MSCs as cellular production machines for the local release of a bispecific antibody designed to redirect autologous T-cells against CD33+ AML blasts. The bsAb secreted in high concentrations by hMSCs recruited T cells very efficiently against target cells, resulting in a rapid and specific anti-tumor response (Fig 3). Furthermore, the additional T cell stimulation provided via the 4-1BBL-expressing MSCs resulted in improvement of the cell system, leading to a more pronounced and long-lasting T cell-mediated response against AML cells expressing low levels of CD33 antigen even in the presence of limited bsAb-releasing MSCs number (Fig. 4C).

5.2 Example 2



[0051] Bispecific antibodies (bsAbs) engaging T cells against cancer cells are an emerging promising immunotherapeutic tool for the treatment of leukemia and solid tumors. Due to their low molecular mass, bsAbs have a short half-life and have to be continuously infused in patients over prolonged time spans of several weeks to achieve clinical responses. As an alternative to continuous intravenous infusion through extra-corporal medical devices the use of mesenchymal stem cells (MSCs) as autonomous cellular machines for the production and secretion of bsAbs.

[0052] A humanized anti-CD33-anti-CD3 bsAb which is capable to redirect human T-cells against CD33 expressing leukemic cells was expressed in the immortalized human MSC line SCP-1. See, Stamova S, et al. (2011) Leukemia: 25[6]: 1053-1056; Stamova S, et al. (2012) Anal Biochem: 423[2]: 261-268; Arndt C, et al. (2011) Blood : 118: 663-664; Arndt C, et al. (2013) Blood : 122[21]:930; Arndt C, et al. (2013) Leukemia: 27: 964-967. Gene-modified SCP-1 cells expressed the bsAb at sufficient amounts to redirect T cells efficiently against CD33 presenting target cells both in vitro as well as in an immunodeficient mouse model. The killing effect was independent of the antigen density on target cells. Moreover, T cells from patients suffering from acute myeloid leukemia (AML) in blast crisis eliminated autologous leukemic cells over time in the presence of the bsAb secreting MSCs. The immune response against AML cells could be further enhanced by providing T cells an additional co-stimulus via the CD137-CD137 ligand axis through CD137L expression on MSCs.

INTRODUCTION



[0053] T-cell engaging bispccific antibodies (bsAbs) have been described. See, c. g., Stamova S, et al. (2012), Antibodies: 1[2]: 172-198; Stamova S, et al. (2011), Leukemia: 25[6]: 1053-1056; Feldmann A, et al. (2012), J Immunol: 189[6]: 3249-3259; Bargou R, et al. (2008), Science: 321[1]: 974-977; Handgretinger R, et al. (2011), Leukemia: 25: 181-184; Schlereth B, et al. (2006) Cancer Immunol Immunother 55[5]: 503-514; Stork R, et al. (2008) J Biol Chem: 283[12]: 7804-7812. Human mesenchymal stem cells (MSCs) have been described [Hamada H, et al. (2005) Cancer Sci: 96[3]: 149-156; Compte M, et al. (2013) Biomatter: 3[1]: e-pub doi: 10.4161/biom.23897]. Without being bound by theory MSCs have a limited immunogenicity and are even poorly recognized by HLA incompatible hosts [Le Blanc K, et al. (2003) Exp Hematol: 31[10]: 890-896; Götherström C, et al. (2004) Am J Obstet Gynecol: 190[1]: 239- 245; Uccelli A, et al. (2008) Autoimmunity: 41[8]: 592-595] and have a tendency to accumulate at the site of tumors including metastatic lesions [Ren C, et al. (2008) Stem Cells: 26[9]: 2332-2338; Kim SM, et al. (2008) Cancer Res: 68[23]: 9614-9623; Uchibori R, et al. (2009) J Gene Med: 11[5]: 373-381]. Further, without being bound by theory, adoptive transfer of MSCs may provide an immunosuppressive environment helpful to reduce GvHD reactions [e. g. von Bonin M, et al. (2009) Bone Marrow Transplant: 43[3]: 245-251; Wehner R, et al. (2009) Haematologica: 94[8]: 1151-1156].

[0054] CD33 is predominantly expressed on myeloid-derived cells and found overexpressed on bone marrow cells from patients with AML as well as leukemic stem cells [Ehninger A, et al. (2014) Blood Cancer Journal: 4, e218: doi: 10.1038/bcj.2014.39]. Current conventional AML therapies do not achieve long-term remissions depending on age and subtype of the disease, therefore new adjuvant therapeutic strategies are urgently needed, especially for the elimination of minimal residual disease (MRD). It is demonstrated herein that gene-modified MSCs are able to (i) express the CD33-CD3 specific bsAb at high levels, and (ii) mediate an efficient lysis of AML blasts by human primary T-cells from both healthy donors and AML patients.

MATERIALS AND METHODS


Ethics statement



[0055] Human peripheral blood mononuclear cells (PBMCs) were isolated either from buffy coats supplied by the German Red Cross (Dresden, Germany) or from fresh blood of healthy donors or patients with their written consent. The study including the consent form was approved by the local ethics committee of the university hospital of the medical faculty of Carl-Gustav-Carus TU-Dresden (EK27022006).

[0056] NOD/SCID IL2Ry-/- (NSG) mice were provided by the animal facility of the Technical University of Dresden. All the procedures involving animals were performed according to the German animal protection law with permission from the responsible of the local authorities (Sächsische Landesdirektion).

Cell Lines



[0057] The human acute myeloid leukemia (AML) cell lines U937 (ACC 5) and MOLM-13 (ACC 554) were cultured in RPMI 1640 medium containing 10% of fetal calf serum (FCS, Biochrom AG), 100 µg/ml penicillin/streptomycin (Biochrom AG), whereas OCI-AML3 (ACC 582), HEK293T (ACC 635) and HEK293T genetically modified to ectopically express CD33 [Arndt C, et al. (2011) Blood : 118: 663-664] were cultured in DMEM medium (10% FCS, 100 µg/ml penicillin/streptomycin). The recently described single-cell-picked clone 1 (SCP-1) cell line [Böcker W, et al. (2008) J Cell Mol Med: 12[4]: 1347-1359], derived from human mesenchymal stem cells and immortalized by lentiviral transduction of the gene coding for the human telomerase reverse transcriptase (hTERT), were grown in RPMI 1640 medium (10% FCS, 100 µg/ml penicillin/streptomycin). All cell lines were maintained at 37°C and 5% CO2.

Generation of recombinant bsAb-releasing hMSCs



[0058] The development of the fully humanized anti-CD33-anti-CD3 bsAb was performed as recently described [Arndt C, et al. (2013) Leukemia: 27: 964-967]. For the generation of permanent hMSCs releasing the bsAb, the cDNA encoding the recombinant Ab construct was cloned into the lentiviral vector p6NST50 [Koristka S, et al. (2013) J Autoimmun: 42: 105-116] to generate the transfer vector p6NST50.bsAb.EGFP-Zeo. In this self-inactivating lentiviral vector, the expression of the CD33-CD3 bsAb is driven by a spleen focus forming virus (SFFV)-derived internal promoter. An enhanced green fluorescence protein (EGFP)-Zeocin fusion protein is co-expressed by an internal ribosomal entry site (IRES). Lentiviral particles pseudotyped with the Vesicular Stomatitis Virus envelope (VSV-G) were generated by transient transfection of HEK293T cells and virus supernatant was harvested as recently described [Cartellieri M, et al. (2014) PLoS One: 9[4]: e93745. doi: 10.1371/journal.pone.0093745] and used to stably transduce SCP-1 cells. For the ectopic expression of the co-stimulatory 4-1BB ligand (CD137L) on SCP-1 cell surface a lentiviral vector was used which harbors an internal minimal human elongation factor 1 alpha promoter instead of the SFFV promoter.

Flow-cytometry analysis



[0059] Transduced and parental wild type cells were stained with anti-CD45/VioBlue, anti-CD33/PE, anti-CD90/VioBlue, anti-CD105/PE and anti-CD73/APC (all Miltenyi Biotec, Bergisch-Gladbach, Germany) monoclonal antibodies (mAbs) to analyze the hMSC marker profile. To monitor for 4-1BBL transgene expression SCP-1 cells were stained with an anti-CD137L/PE (BD Bioscience, Heidelberg, Germany) mAb. Cells stained only with matched isotype control Ab (in grey) served as negative control. Samples were analyzed using a MACSQuant Analyzer® and MACSQuantify software (both Miltenyi Biotec).

Expression, purification and quantitative analysis of the recombinant bsAb



[0060] Culture supernatants of the stably transduced hMSCs were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) for the His-tagged anti-CD33-anti-CD3 bsAb purified by a single-step affinity chromatography on Ni-NTA columns (Qiagen, Hilden, Germany), and Western blotting using anti-penta-his mAb (Qiagen) and an alkaline phosphatase (AP)-coupled anti-mouse IgG antibody (Dianova, Hamburg, Germany) as previously described [e. g. Feldmann A, et al. (2011) Prostate: 71[9]: 998-1011]. The amount of anti-CD33-anti-CD3 bsAb secreted was quantified by enzyme-linked immunosorbent assay (ELISA) as follow: SCP-1 cells were seeded as triplets in 96-well plates at decreasing cell densities. After 48 h of culture the supernatant of each sample was collected and added at a defined dilution to the wells of F16 MaxiSorp ELISA plates (Nunc), beforehand coated with 1.5 µg/ml mouse anti-pentaHis capture antibody (Qiagen), capable of binding the C-terminal histidine (his)-tag of the CD33-CD3 bsAb. Standard samples were prepared as a 2-fold serial dilution from purified CD33-CD3 bsAb (GEMoaB, Dresden, Germany) starting from 100 ng/ml. For binding of the captured bsAb, a detection solution was prepared by diluting anti-Myc-HRP detection Ab (Miltenyi) 1:1000 in blocking buffer, and added to the samples for 2 h at RT. The substrate solution containing 100 µg/ml TMB, 10% DMSO, 0.05 M phosphate/citrate buffer and 30% H2O2 was added per well for 15 min to develop a color reaction and stopped by adding 1 M H2SO4 stop solution. The optical density was measured at 450 nm using the SUNRISE™ Microplate Reader (Tecan, Maennedorf, Switzerland) and used to calculate the concentration of hMSC-released bsAb. Indirect immunofluorescence analysis was performed to examine the binding properties of the hMSCs-rclcascd bsAb. A total of 5x10 CD33+ or CD33- cell lines and PBMCs were stained with 20 µg/ml of the purified bsAb. The bispecific construct was detected by flow cytometry using FITC-conjugated antibody against the myc-tag (Miltenyi) [Feldmann A, et al. (2012) J Immunol: 189[6]: 3249-3259; Feldmann A, et al. (2011) Prostate: 71[9]: 998-1011]. CD3+ T cells were discriminated using an anti-CD3/VioBlue mAb, whereas anti-CD56/APC (Miltenyi) and anti-16/PE (BD Biosciences) mAbs were used to identify CD3-CD56+CD16+ NK cells. Maternal anti-CD33 and anti-CD3 mAbs were used as control and detected with a PE-conjugated Goat F(ab)2 anti-Mouse IgG (Fcgamma) Ab (Immunotech, Marseille, France).

T cells isolation from human healthy donors



[0061] Human peripheral blood mononuclear cells (PBMCs) isolated from buffy coats of healthy volunteers after their informed consent were prepared by gradient centrifugation over polydextran based separating solution Biocoll (Biochrom). Pan T cells were freshly isolated from PBMCs by negative selection using the pan T cell isolation kit (Miltenyi Biotec), human CD4+ and CD8+ T cells by negative selection using the CD4+ and CD8+ T cell isolation kit (Miltenyi Biotec), respectively. Human T cells were cultured in RPMI 1640 medium containing 10% of fetal calf serum (FCS, Biochrom AG), 100 µg/ml penicillin/streptomycin (Biochrom AG), and 50 U/ml IL-2 (ImmunoTools, Friesoythe, Germany) overnight before starting the experiments the next day. T cells were washed twice with RPMI 1640 medium without any additives before added to the experimental cultures.

51Cr-release and flow-cytometry based Cytotoxicity Assay



[0062] The killing of CD33+ target tumor cells by T-cells redirected via hMSCs-released anti-CD33-anti-CD3 bsAb was examined either by standard 51Cr release assays or by flow cytometry-based assays as recently established [Cartellieri M, et al. (2014) PLoS One: 9[4]: e93745. doi: 10.1371/journal.pone.0093745; Feldmann A, et al. (2011) Prostate: 71[9]: 998-1011; Koristka S, et al. (2012) J Immunol: 188[3]: 1551-1558].

Determination of cytokine concentration



[0063] To determine amounts of secreted IFN-γ, IL-10, TNF-α, and IL-2 from co-cultures, cell-free supernatants were collected at the indicated time and analyzed for cytokine secretion using OptEIA ELISA Sets (BD Biosciences) according to the manufacturer's protocol. The absorption of the samples was measured after 30 min, and the obtained values were used to calculate the concentration of the cytokines in the samples, according to the values obtained for the standard series provided by the manufacturer.

T-cell activation and proliferation assays



[0064] The expression of the activation markers CD69 and CD25 were analyzed to determine the activation state of T cells. As target cells 1x104 transgenic CD33-expressing HEK293T cells were co-cultured with 5x104 untouched pan T cells in the presence or absence of 1x104 SCP-1 cells seeded 48 h before starting the experiment. After indicated time points, supernatant of each sample was collected and the cells of one triplet were pooled and stained with a mixture of anti-CD3/VioBlue, anti-CD69/FITC, anti-CD4/PerCP (all purchased from Miltenyi) and anti-CD25/PE (BD Biosciences). T cell proliferation assays were performed as previously described [Koristka S, et al. (2013) J Autoimmun: 42: 105-116; Koristka S, et al. (2012) J Immunol: 188[3]: 1551-1558].

Flow-cytometry killing assay with fresh AML samples



[0065] Mononuclear cell (MNC) samples from AML patients with hyperleukocytosis were obtained from leukapheresis products with informed consent and approval by the local institutional review board. MNCs were prepared by gradient centrifugation over polydextran based separating solution Biocoll (Biochrom). The redirection of autologous T cell towards AML blasts by the bsAbs CD33-CD3-releasing hMSCs was investigated by co-cultivation of 1x104 48 h pre-cultured genetically modified hMSCs with 1x105 AML patient-derived MNCs. The specific killing of myeloid cells was analyzed with MACSQuant® Analyzer (Miltenyi Biotec) at indicated time points. Moreover, MNCs of one triplet were pooled and stained with a mixture of anti-CD3/PECy7, anti-CD4/PerCP mAbs (Miltenyi) in order to analyze the relative percentage of T cells; and anti-CD123/PE, anti-HLA-DR/FITC and anti-CD45/VioBlue mAbs (Miltenyi Biotec) to discriminate the myeloid cell populations. Living cells were identified by being propidium iodide negative.

Mouse model



[0066] NOD/SCID IL2Ry-/- (NSG) mice were kept under standardized environmental conditions and received autoclaved food, water, and bedding. Prior to the injection, bsAb-releasing or vector control containing hMSCs (5x103) were cultured for 48 hours and subsequently incubated together with 1x105 MOLM-13 cells and freshly isolated human T cells at an e:t ratio of 5:1. Following additional 24 hours of co-culture the mixed cell population was administrated intravenously (i.v.) to two groups of eight- to ten-week-old NSG mice via the tail vein. One group (5 animals) received MOLM-13 cells, T cells and vector control hMSCs, whereas the other group (4 animals) was treated with MOLM-13 cells, T cells and bsAb-secreting hMSCs. Mice were daily monitored for posture, activity, fur texture and skin integrity. Animals were sacrificed according to local ethical committee guidelines when displayed pathological score and more than 15% body weight loss. The survival rate of bsAb-treated or untreated mice was determined.

Statistical analysis



[0067] Statistical analysis was performed using GraphPad Prism Software (La Jolla, CA, USA). One-way analysis of variance (ANOVA) with Bonferroni Multiple Comparison test was used for statistical significance when multiple experiments were compared. Survival data were analyzed by using a Kaplan-Meier survival analysis with a log rank method of statistics (***P<0.001, **P<0.01, *P<0.05).

RESULTS


Development and characterization of bispecific antibody-releasing hMSCs



[0068] The genetically modified hMSC lines used in this work were generated from the hTERT-immortalized single-cell derived hMSC line SCP-1 [Böcker W, et al. (2008) J Cell Mol Med: 12[4]: 1347-1359] using lentiviral gene transfer to stably express the bsAb CD33-CD3. The bsAb CD33-CD3 was constructed as single-chain bispecific tandem fragment variable (scBsTaFv). The VH and VL domains of each scFv were humanized by CDR grafting and connected via a linker comprised of three repeats of four glycine and one serine residues 3*(Gly4Ser). The N-terminus of the bsAb construct contains a signal peptide (SP) for the secretion of the bsAb into the cell culture medium, whereas its C-terminus tag harbors a myc- and his-tag used for immunochemical Ab detection and purification. The parental SCP-1 cell line was examined by flow cytometry at different passages to characterize the surface protein expression. SCP-1 cells expressed the typical MSCs marker proteins CD90, CD105 and CD73 whereas they did not display expression of CD33 antigen or hematopoietic antigens CD45 and CD34. For the identification and selection of successfully transduced cells with bsAb gene, the EGFP-zeocin fusion gene was co-expressed under the same promoter through an internal ribosomal entry site (IRES). A lentiviral vector containing only the EGFP-zeocin expression cassette was used for generation of a vector control cell line, whereas untransduccd SCP-1 cell line served as 'wild type' (wt) control. Following the SCP-1 cells transduction each hMSC line (wt, vector control and CD33-CD3) was analyzed for intracellular EGFP signal by flow cytometry. All cell lines were selected to nearly 100% purity. Next, it was tested if the genetically modified MSCs secrete the bsAb. Therefore culture medium was run over affinity chromatography columns to purify bsAb CD33-CD3 through the C-terminal his-tag. Indeed, a protein with the expected size of ∼60 kDa was detected in the elution fraction by COOMASSIE-BRILLIANT-BLUE®-staining. Western blotting confirmed the presence of a his-tagged protein at the corresponding molecular size. After successful bsAb production the amount of bsAb CD33-CD3 secreted by the modified MSCs was quantified by ELISA. The maximum bsAb concentration obtained from the bsAb-releasing hMSCs under these experimental conditions reached 4400 ng/ml at a starting density of 105 hMSC cells/well. Considering the sample volume of 200 µl it was calculated that a single MSC releases approximately 8.8 pg bsAb in 48 h. However, the highest calculated Ab amount released by a single MSC was obtained at the lowest seeding density (103 cells/well) reaching 73.7 pg/cell in 48 h. From the antibody release curve it was concluded that the amount of released bsAb starts to remain stable above 5x104 cells/well under the chosen experimental conditions. This observation is most likely due to a rapid nutrient consumption at high seeding densities of hMCSs during the 48 h cultivation period, which in turn down regulates protein synthesis including bsAb production in the cells. Next the binding specificity of the purified bsAb released by the gene-modified hMSCs was analyzed. In agreement with previous analysis, the binding of both parental monoclonal antibodies lead to a strong shift in MFI. The binding capabilities of the MSC-released anti-CD33-anti-CD3 bsAb is dependent on surface expression of the respective antigen, while no binding can be detected on antigen-negative e.g. HEK293T and SCP-1 cells. Furthermore, MSC-released bsAb shows a strong binding to the CD33 antigen which can be blocked in a concentration-dependent manner with the parental CD33-specific mab. In contrast to the anti-CD33 domain, the binding to the CD3 complex on T cells via the anti-CD3 domain is hardly detectable.

Specific CD33+ target cell killing by retargeted T cells via hMSCs-released bsAb CD33-CD3



[0069] The capability of the MSC-released bsAb to redirect T lymphocytes specifically in an antigen-dependent manner was evaluated with various AML cell lines expressing low (OCI-AML3), intermediate (U937) and high (MOLM-13) levels of CD33 antigen. HEK293T cells were included in these experiments as a CD33 negative control cell line. Using MOLM-13 cells as targets, it could be demonstrated that human T cells were redirected and mediated target cell killing in a concentration-dependent way in the prescence of the purified anti-CD33-anti-CD3 bsAb, whereas CD33 negative HEK293T cells were not attacked by human T cells in the prescence of the bsAb. Next, retargeting of T lymphocytes to AML cells via in situ bsAb-releasing modified hMSCs was evaluated. Therefore, 1x104 CD3+ T cells were incubated with 51Cr labeled CD33+ AML cell lines at an e:t cell ratio of 5:1 in the presence of 48 h-cultured hMSC lines. The cross-linkage of T cells and target cells via hMSC-released bsAb resulted in effective killing of target cells already at earlier time points independently of the CD33 antigen density on the surface of both analyzed cell lines. After prolonged incubation time of 20 h a large portion of target cells were lysed even at low MSC seeding densities. No relevant difference in the context of specific target cell lysis between the samples with different hMSC densities was detected, suggesting that T cells can be efficiently redirected against tumor cells even at very low bsAb concentrations released by the lowest cell density used in this experimental setting (5x102 cells/well). On the contrary no specific lysis was detected in the presence of the wt and vector control hMSC lines, confirming that the observed tumor cell killing was strictly dependent on the bsAb release. According to the findings herein, the killing efficacy of the hMSC-released bsAb CD33-CD3 is within the concentration range of previously published bsAb CD33-CD3 data [Stamova S, et al. (2011) Leukemia: 25[6]: 1053-1056.;Arndt C, et al. (2013) Leukemia: 27: 964-967; Arndt C, et al. (2014) Leukemia: 28[1]: 59-69]. Moreover, to further exclude any bsAb-dependent off-target effect on bsAb releasing MSCs, additional killing assays were performed. Therefore, 48 h-cultured 51Cr labeled gene-modified MSCs were incubated with peripheral blood mononuclear cells to better resemble a natural in vivo situation characterized by the presence of a more complex cell population. No off-target lysis of bsAb-expressing hMSCs was observed after 20 h of co-culture of hMSCs and PBMCs in the presence or absence of CD33+ target cells, whereas parallel experiments performed under the same conditions showed efficent killing of CD33+ target cells. The obtained results clearly indicate that MSC-secreted bsAb is able to redirect human T cells and induce a specific target cell lysis by triggering an efficient T cell activation. However, to ensure that during the development procedure of the bsAb the respective scFv portions did not contain fragments of the constant regions of the parental monoclonal antibodies, which could also recruit additional effector cell populations to contribute to the anti-tumor effect via bsAb-Fc-receptor interactions, additional experiments were performed. Therefore, isolated NK cells were incubated with CD33 expressing target cells in the prescence or absence of the purified bsAb. No enhanced target cell lysis resulted mediated by NK cells in the prescence of the bsAb in accordance to the fact that also no binding of the bsAb to NK cells could be detected by flow cytometry analysis.

Improvement of T cell-mediated tumor cell killing by the presence of a co-stimulatory ligand on the hMSCs surface



[0070] The bsAb-releasing hMSCs were futher modified by co-expression of the 4-1BBL molecule on the cell surface to prove that the additional co-stimulus led to an enhanced T cell-mediated response even against AML cells expressing low levels of CD33 antigen. Therefore the bsAb CD33-CD3 producing hMSC line was additionally genetically modified to express the 4-1BBL molecule. Surface expression of 4-1BBL in the newly generated sub-line could be verified by flow cytometry. Next, the immunotherapeutic effect exerted by the modified hMSC sub-lines was investigated by flow cytometry-based cytotoxicity assay. In order to define the lowest hMSCs density leading to a relevant target cell killing effect, hMSCs were seeded at very low concentrations and cultured for 48 h. Subsequently freshly isolated human T cells and EFLUOR®670-labeled CD33low OCI-AML3 target cells were co-incubated together with the modified hMSCs at an e:t ratio of 1:1, which better resembles natural in vivo conditions of AML patients in acute blast crisis having low T cell numbers. The numbers of living target cells were quantified after 24 and 48 h by flow cytometry. T cell mediated target cell lysis triggered by the bsAb released from both immunotherapeutic hMSC lines (CD33-CD3 and CD33-CD3 + 4-1BBL) was delayed and could not be observed before the 48 h time point. Co-stimulation by 4-1BBL presented on the surface of the modified hMSC line did not significantly enhance the killing abilities of bsAb redirected T cells under these experimental conditions.

[0071] As it is believed that 4-1BB-mediated co-stimulation is mainly involved in late phases of immune activation and, as such. a longer time may be required to observe a specific T cell response [Arndt C, et al. (2014) Leukemia: 28[1]: 59-69], the co-incubation time period under the same experimental conditions was prolonged. Thus, the percentages of living EFLUOR®670-positive tumor cells were further analyzed after 96 h. After a longer co-incubation time the additional T cell stimulation by hMSC presented 4-1BBL ameliorated the T cell response towards CD33low target cells compared to the hMSCs expressing only the bsAb, leading to a more pronounced specific tumor cell killing even at the lowest hMSCs density of 10 MSCs/well.

Cytokine secretion and expansion of redirected T cells upon their cross-linkage with 364 the CD33-CD3 bsAb and 4-1BBL



[0072] In order to verify the effects of the co-stimulatory signal via 4-1BBL/4-1BB interaction T cell activation, release of pro-inflammatory cytokines and T cell expansion were investigated. In these experiments transgenic CD33+ HEK293T cells, negative for T cell co-stimulatory ligands like B7.1, B7.2, Ox40, 4-1BB and ICOS ligand, were used as target cells instead of CD33+ native AML cell lines to exclude additional T cell stimulation exerted by co-stimulatory molecules. Both CD4+ and CD8+ T cells up-regulated activation markers CD69 and CD25 in the presence of both CD33-CD3 bsAb producing hMSC lines at comparable levels. Thus, cross-presentation of 4-1BBL seems not to enhance the up-regulation of these specific T cell activation markers. However, analyzing the amount of pro-inflammatory cytokines secreted into the supernatants, dramatic differences were detected. Co-stimulation via 4-1BBL expressed on the modified hMSCs increased significantly the levels of IFN-γ and TNF-α secretion up to 8-fold for TNF and more than 10-fold for IFN-γ. On the contrary, T cell cross-linkage with CD33+ HEK293T cells without the co-stimulatory signal provoked lower IFN-γ and TNF-α release, while nearly no cytokines were detectable in presence of the wt and vector control hMSCs. In accordance to the increased cytokine secretion, cross-linking with immunoligand expressing hMSCs induced stronger proliferation of bsAb activated T cells leading to higher expansion rates of T cells in the presence of the 4-1BBL presenting hMSCs. T cells expanded approximately two to three fold in the presence of hMSCs expressing the co-stimulatory ligand, whereas no increase in absolute T cells numbers could be observed either in the presence of the CD33-CD3-releasing hMSCs or the parental wt and vector control sub-lines respectively.

Retargeting of autologous T cells towards AML blasts



[0073] In the next step, CD33-CD3 bsAb-producing hMSCs were co-cultivated with AML patient-derived MNCs to investigate their capability to redirect autologous T cells towards AML blasts. Freshly isolated MNCs were phenotyped via immunostaining to measure the proportions of CD45+, CD123+ and HLA-DR+ AML blasts and CD45+, CD3+ T cells respectively. As in previous experimental settings, MSC sub-lines were seeded 48 h beforehand and then co-cultivated with patient-derived MNCs. The relative percentages and the absolute cell numbers of either leukemic cells or leukocyte populations were monitored by flow cytometry upon staining for specific surface markers after 24, 48 and 96h of co-culture to quantify tumor cell killing and T cell expansion respectively. For one representative donor, already 48 h after starting the co-culture, absolute leukemic blast numbers dramatically decreased, leading to a nearly complete eradication of tumor cells within 96 h. On the contrary in the presence of the control hMSCs the number of leukemic blasts remained nearly stable over time.

[0074] On the one hand, the AML blasts number strongly decreased in the presence of both bsAb-releasing hMSCs in comparable manner. On the other hand, the presence of the 4-1BB co-stimulatory signal significantly enhanced the proliferation of T cell population after 96 h of co-culture, leading to a doubling of total T cell number and to a pronounced T cell expansion, approximately 3 fold higher compared to the expansion attained in the other samples. Altogether, these results show that the continuous delivery of bsAb CD33-CD3 together with a constant stimulation of T cells by modified MSCs improve specific killing of autologous AML blasts and increase patient-derived T cells proliferation over time. Moreover, the data show that a co-stimulatory signal for bsAb-redirected and -activated T cells need not necessarily be mediated via interaction with the target cells but can also be provided by independent cells.

In vivo functionality of the bsAb-releasing hMSCs system



[0075] The efficacy of the bsAb-secreting gene-modified cell system was investigated in vivo by co-injection of 24 h-preincubated MOLM-13 cells, T cells and bsAb CD33-CD3 releasing- or vector control containing hMSC lines in NOD/SCID IL2Rγ-/- mice, respectively distinguished as treatment and control group. Daily monitoring of the animals revealed that the mice of the control group injected with vector control-modified hMSCs developed evident signs of leukemic cells engraftment between the second and the third week after transplantation displaying weight loss, reduced locomotion, hunched posture and ruffled fur. In contrast mice of the treatment group receiving the bsAb-releasing hMSCs appeared to be protected from onset of the disease during the experiment. At the time when all untreated control mice had to be killed, all treated mice were still alive and did not show signs of disease (significance of survival **p<0.01).

DISCUSSION



[0076] Without being bound by theory , mesenchymal stem cells can have the potential to migrate into inflamed tissues and tumors. Inflamed tissues and tumors secrete cytokines and chemokines such as stroma-derived factor one alpha (SDF-1α) for which MSCs express the respective appropriate receptors on their surface. In this study MSCs were gene-modified to secrete locally a bispecific antibody designed to recruit autologous T-cells against AML blasts. The rationale for developing a CD33-CD3 T-cell recruiting bispecific antibody is that the long-term success of conventional AML therapy including combination of chemotherapy and hematopoietic stem cell transplantation is still limited resulting in high incidents of relapse and refractory diseases. Moreover, most clinical experience has been gained with CD33 as target antigen, CD33 has been found expressed on leukemic stem cells, and only CD33+ leukemic stem cells gave rise to leukemia in mice. However, T-cell activation against CD33-expressing cells with the bsAb used in the present study does not impair hematopoietic engraftment in NOD scid gamma knock-out mice [Arndt C, et al. (2014) Leukemia: 28[1]: 59-69], but leads to efficient eradication of AML blasts both in vitro and in vivo [Arndt C, et al. (2013) Leukemia: 27: 964-967].

[0077] Although it has been reported that MSCs are able to suppress activation of T cells via different pathways [Glennie S, et al. (2005) Blood: 105[7]: 2821-2827; Nauta AJ, et al. (2007) Blood: 110[10]: 3499-3506], in the present experiments no suppression of tumor cell killing by bsAb-redirected T cells was observed in the presence of MSCs. On the other hand, bsAb-activated T cells seem to not attack modified MSCs, pointing out the high specificity of bsAb-mediated target cell killing, but non-immunogenic properties of human MSCs might also contribute to this observation [Ménard C, Tarte K. (2013) Stem Cell Res Ther: 4[3]: 64; Castro-Manrreza ME, et al. (2014) Stem Cells Dev: 23[11]: 1217-1232].

[0078] Nonetheless, additional modifications of the MSCs to enhance local anti-tumor inflammatory reactions might also be beneficial. One way to augment anti-tumor effects of T cells was examined in our study. The additional T cell stimulation provided through tumor necrosis factor receptor signaling by the 4-1BBL-expressing MSCs resulted in a more pronounced and long-lasting T cell-mediated response, enhancing T cell proliferation potential and their cytotoxic activity against AML blasts expressing different levels of CD33 antigen. The findings presented herein and the observation that targeting the 4-1BB pathway in cancer treatment ameliorates the elimination of tumor cells and prevents activation-induced cell death (AICD) are in line with recent published data [Hernandez-Chacon JA, et al. (2011) J Immunother: 34[3]: 236-250; Vinay DS, Kwon BS (2014) BMB Rep: 47[3]: 122-129]. The specific anti-tumor effect achieved by the use of CD33-CD3 bsAb acts in concert with the 4-1BB stimulus, leading to a significant improvement of the specific T cell response represents an additional benefit in respect to future in vivo use of gene-modified MSCs.

[0079] Together the data presented herein underline that continuous in situ delivery of bsAb CD33-CD3 by genetically modified hMSCs can represent a valid alternative to the exogenous administration of short-lived immunoagents for antigen-specific immunotherapy of AML patients. Moreover, the immunosuppressive potential of MSCs does not limit their use as in vivo bsAb factory.

[0080] All references (e.g., publications or patents or patent applications) cited herein are incorporated herein by reference in their entirety and for all purposes to the same extent as if each individual reference (e.g., publication or patent or patent application) was specifically and individually indicated to be incorporated by reference in its entirety for all purposes.

[0081] Other embodiments are within the following claims.


Claims

1. A pluri- or multipotent stem cell (SC) engineered to express a bispecific antibody comprising at least a target cell specific binding domain and at least an immune cell specific binding domain, wherein the target cell specific binding domain is preferably specific for a tumor associated antigen or an antigen of a pathogen, parasite or parasitoid.
 
2. The SC of claim 1, wherein the pluri- or multipotent stem cell (SC) is a human pluri- or multipotent stem cell (hSC) or a mesenchymal stem cell (MSC).
 
3. The SC of any one of claims 1 or 2, wherein said bispecific antibody is a single chain bispecific antibody.
 
4. The SC of any one of claims 1 to 3, wherein said bispecific antibody is a humanized bispecific antibody.
 
5. The SC of any one of claims 3 or 4, wherein the bispecific antibody is capable of binding to CD33 and CD3.
 
6. The SC of any one of claims 1 to 5, wherein said bispecific antibody comprises a variable light (VL) antibody domain and variable heavy (VH) antibody domain that specifically bind to CD33 (CD33 VL and CD33 VH), wherein said CD33 VH comprises a VH CDR1 comprising an amino acid sequence selected from the group consisting of SEQ ID NOs: 3 and 9-12, a VH CDR2 comprising an amino acid sequence selected from the group consisting of SEQ ID NOs: 4 and 13-16, and a VH CDR3 comprising an amino acid sequence selected from the group consisting of SEQ ID NOs: 5, 17-20; and wherein said CD33 VL comprises a VL CDR1 comprising an amino acid sequence selected from the group consisting of SEQ ID NOs: 6 and 21-24, a VL CDR2 comprising an amino acid sequence selected from the group consisting of SEQ ID NOs: 7 and 25-28, and a VL CDR3 comprising an amino acid sequence selected from the group consisting of SEQ ID NOs: 8 and 29-32.
 
7. The SC of claim 6, wherein said CD33 VH comprises a VH CDR1 comprising SEQ ID NO: 3, a VH CDR2 comprising SEQ ID NO: 4, and a VH CDR3 comprising SEQ ID NO: 5; and said CD33 VL comprises a VL CDR1 comprising SEQ ID NO: 6, a VL CDR2 comprising SEQ ID NO: 7, and a VL CDR3 comprising SEQ ID NO: 8.
 
8. The SC of any one of claims 1 to 8, further expressing a T cell co-stimulatory ligand or an active fragment thereof.
 
9. The SC of claim 9, wherein the T cell co-stimulatory ligand or active fragment thereof is selected from the group consisting of CD28 ligand B7-1 and B7-2, Ox40 ligand OX40L, 4-1BB ligand 4-1BBL (CD137L), CD27 ligand CD70 (CD27L), CD40 ligand CD154, ICOS ligand ICOSL, LFA-1 ligand ICAM-1, -2, and -3, 2B4 ligand and SLAM ligand.
 
10. The SC of claim 8, wherein the T cell co-stimulatory ligand or active fragment thereof is 4-1BB ligand 4-1BBL (CD137L).
 
11. A pharmaceutical composition comprising the SC of any one of claims 1 to 10.
 
12. The SC of any one of claims 1 to 11, for use in treating a cancer.
 
13. The SC according to claim 12, wherein the cancer is associated with expression of CD33.
 
14. The SC of any one of claims 12 or 13, wherein the cancer is acute myeloid leukemia (AML).
 




Drawing






































REFERENCES CITED IN THE DESCRIPTION



This list of references cited by the applicant is for the reader's convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.

Patent documents cited in the description




Non-patent literature cited in the description