Acute myeloid leukemia transforms the bone marrow niche into a leukemia-permissive microenvironment through exosome secretion.

Little is known about how leukemia cells alter the bone marrow (BM) niche to facilitate their own growth and evade chemotherapy. Here, we provide evidence that acute myeloid leukemia (AML) blasts remodel the BM niche into a leukemia growth-permissive and normal hematopoiesis-suppressive microenvironment through exosome secretion. Either engrafted AML cells or AML-derived exosomes increased mesenchymal stromal progenitors and blocked osteolineage development and bone formation in vivo. Preconditioning with AML-derived exosomes 'primed' the animals for accelerated AML growth. Conversely, disruption of exosome secretion in AML cells through targeting Rab27a, an important regulator involved in exosome release, significantly delayed leukemia development. In BM stromal cells, AML-derived exosomes induced the expression of DKK1, a suppressor of normal hematopoiesis and osteogenesis, thereby contributing to osteoblast loss. Conversely, treatment with a DKK1 inhibitor delayed AML progression and prolonged survival in AML-engrafted mice. In addition, AML-derived exosomes induced a broad downregulation of hematopoietic stem cell-supporting factors (for example, CXCL12, KITL and IGF1) in BM stromal cells and reduced their ability to support normal hematopoiesis. Altogether, this study uncovers novel features of AML pathogenesis and unveils how AML cells create a self-strengthening leukemic niche that promotes leukemic cell proliferation and survival, while suppressing normal hematopoiesis through exosome secretion.

Preparing clinical grade Ag-specific T cells for adoptive immunotherapy trials.

The production of clinical-grade T cells for adoptive immunotherapy has evolved from the ex vivo numerical expansion of tumor-infiltrating lymphocytes to sophisticated bioengineering processes often requiring cell selection, genetic modification and other extensive tissue culture manipulations, to produce desired cells with improved therapeutic potential. Advancements in understanding the biology of lymphocyte signaling, activation, homing and sustained in vivo proliferative potential have redefined the strategies used to produce T cells suitable for clinical investigation. When combined with new technical methods in cell processing and culturing, the therapeutic potential of T cells manufactured in academic centers has improved dramatically. Paralleling these technical achievements in cell manufacturing is the development of broadly applied regulatory standards that define the requirements for the clinical implementation of cell products with ever-increasing complexity. In concert with academic facilities operating in compliance with current good manufacturing practice, the prescribing physician can now infuse T cells with a highly selected or endowed phenotype that has been uniformly manufactured according to standard operating procedures and that meets federal guidelines for quality of investigational cell products. In this review we address salient issues related to the technical, immunologic, practical and regulatory aspects of manufacturing these advanced T-cell products for clinical use.

Manufacturing of gene-modified cytotoxic T lymphocytes for autologous cellular therapy for lymphoma.

BACKGROUND: The production of therapeutic T-cell populations for adoptive immunotherapy of cancer requires extensive ex vivo cell processing, including the isolation or creation of Ag-specific T cells and their subsequent propagation to clinically relevant numbers. These procedures must be performed according to the principles of current good manufacturing practices (cGMP) for phase I clinical trials to ensure the identity, purity potency and safety of the cellular product. In this report we describe our approach to manufacturing and characterizing bulk populations of gene-modified autologous T cells for use in treating follicular lymphoma. METHODS: PBMC from healthy donors, obtained after informed consent, were stimulated in vitro with Ab to CD3epsilon (OKT3) and recombinant human IL-2 and then electroporated with plasmid DNA containing a human CD19-specific chimeric Ag receptor (CAR) gene and HSV-1 thymidine kinase (TK) gene. Stably transfected cells were selected in cytocidal concentrations of hygromycin B over multiple 14-day stimulation culture cycles and then cryopreserved. Vials of cryopreserved/selected T cells were used to initiate T-cell expansion cultures to produce cell products for clinical infusion. These cultures were characterized for phenotype, function and suitability for use in adoptive immunotherapy studies. RESULTS: Our results demonstrate that bulk populations of gene-modified T cells derived from peripheral blood of healthy donors express CD19+ chimeric Ag receptor at low levels and can specifically lyse CD19+ target cells in vitro. These cells display a differentiated T-effector phenotype, are sensitive to ganciclovir-mediated killing and display a non-transformed phenotype. TCR Vbeta usage indicated that all populations tested were polyclonal. Ex vivo cell expansion from cryopreserved cell banks is sufficient to produce doses of between 5 x 10(9) and 1 x 10(10) cells/run. One of three transductions resulted in a population of cells that was not suitable for infusion but was identified during release testing. No populations displayed any evidence of bacterial, fungal or mycoplasma contamination. DISCUSSION: We have established a manufacturing strategy that is being used to produce T cells for a phase I clinical trial for follicular lymphoma. Genetically modified T cells have been characterized by cell-surface marker phenotype, functional activity against CD19+ targets and requisite safety testing. These pre-clinical data confirm the feasibility of this approach to manufacturing T-cell products.

Sustained retroviral gene marking and expression in lymphoid and myeloid cells derived from transduced hematopoietic progenitor cells.

The expression of antiviral genes in human hematopoietic stem or progenitor cells has been proposed as a strategy for gene therapy of AIDS. To be successful, this strategy requires safe and efficient transfer of the therapeutic gene into hematopoietic cells and gene expression has to be maintained in HIV susceptible cells following differentiation. We have used retroviral vectors to transfer the gene for a transdominant inhibitor of HIV replication (RevM10) into CD34+ stem/progenitor cells isolated from human umbilical cord blood (UCB). Following transduction, cells were allowed to differentiate either in vitro in clonogenic assays and long-term stromal cell cultures or in human thymus implanted in immunodeficient scid/scid mice in vivo (SCID-hu). Following differentiation and expansion, multiple lineages of cells were shown to carry the transgene. A higher percentage of gene-marked progenitor cells (10-30% in most cases) were detected in methylcellulose colony assays and in long-term stromal cell cultures (1-5%). In contrast, gene-marked T cells derived from transduced CD34+ cells in a SCID-hu model were detected at an even lower frequency (0.01-1%). RevM10 RNA expression was detected in CD34+ cells immediately after transduction and was maintained after in vitro differentiation of those cells into CD14+ myeloid cells. In T cells, the RevM10-specific RNA was detectable by RT-PCR and also by semiquantitative RNase protection. These findings demonstrate that LTR-driven gene expression is sustained in relevant cells derived from retrovirus-transduced hematopoietic progenitor cells after extensive differentiation in vitro and in vivo and suggest that stringent in vivo, rather than in vitro assays, may be a better preclinical system to improve gene marking and expression in hematopoietic cells.

Hematopoietic potential of cryopreserved and ex vivo manipulated umbilical cord blood progenitor cells evaluated in vitro and in vivo.

The hematopoietic potential of cryopreserved and ex vivo manipulated umbilical cord blood (UCB) samples was evaluated in vitro and in vivo. Phenotypic analysis shows that approximately 1% of cord blood mononuclear cells express high levels of CD34 antigen on their surface (CD34hi), but none of a panel of lineage antigens (Lin-), suggesting that they are hematopoietic progenitor cells that have not yet committed to a specific lineage. Approximately 1% of CD34hi/Lin- cells are primitive hematopoietic progenitors that produce B lymphoid and multiple myeloid progeny for up to 7 weeks in stromal cell cultures. Twenty-one percent (+/- 13%) of CD34hi/Lin- cells also express low levels of the Thy-1 antigen and are threefold to fourfold enriched over CD34hi/Lin- cells in primitive hematopoietic potential as measured by long-term culture and phenotypic analysis. One-week liquid cultures of CD34-enriched UCB progenitor cells in the presence of interleukin (IL)-3, IL-6, and stem cell factor (SCF) results in a two-fold to threefold expansion of progenitors capable of reinitiating long-term stromal cell cultures. Only the CD34hi/Thy-1+/Lin- cell population was capable of maintaining progenitors with secondary transfer potential in long-term stromal cell cultures and is thus postulated to contain all of the primitive hematopoietic stem cells in UCB. The in vivo transplantation potential of UCB was also measured. Ex vivo manipulated UCB progenitor cells were used to engraft irradiated human thymus fragments implanted in severe combined immunodeficiency (SCID) mice. Thymic engraftment with >5% donor-derived cells and a normal CD4/CD8 distribution was observed in 19 of 23 tissues tested. UCB cells from in vitro expansion cultures engrafted with efficiencies comparable to nonexpanded cells. Similar results were obtained for UCB engraftment of human bone fragments implanted in SCID mice. In all cases, engraftment was achieved in competition with endogenous competitor stem cells and across major histocompatibility barriers. Taken together, this data demonstrates that human UCB is a rich source of multipotent hematopoietic progenitors that can be cryopreserved, enriched by physical methods, and expanded in a limited fashion without measurable loss of long-term culture or in vivo engrafting potential as measured in these assays.

An analysis of sequence variation in the beta chain framework and complementarity determining regions of an allo-reactive T cell receptor.

Current models of T cell receptor (TCR) structure are generally based on the homology observed between the TCR and the immunoglobulins. Furthermore, these models have predicted the locations of framework and complementarity determining regions within the alpha- and beta-chain variable regions. In order to test the validity of these models, we have generated a series of mutations within the V beta domain of an allo-reactive TCR and determined their effect on antigen recognition.

Single cell analysis of calcium mobilization in anti-immunoglobulin-stimulated B lymphocytes.

A rapid increase in the concentration of intracellular free calcium ([Ca++i]) in B cells after mIg crosslinking has been documented previously by fluorimetric analysis of cell populations loaded with the fluorescent Ca++ indicator Quin 2. Although providing a valuable indication of Ca++ mobilization in the population as a whole, it has not been possible to determine whether only a subpopulation of the cells or the entire population exhibits this response. In this report, we describe the marriage of flow cytometry and Quin 2 technology, which permits discrimination of Ca++ mobilization by subpopulations of cells, as well as in whole populations. We have determined that the entire mIg+ population exhibits a synchronous increase in [Ca++i] rapidly after stimulation. Furthermore, all members of the population appear to undergo an approximately equal response.

mIgM:mIgD ratios on B cells: mean mIgD expression exceeds mIgM by 10-fold on most splenic B cells.

The relative frequency of mIgM and mIgD molecules on B cell surfaces is important in determining, in large part, the isotype involvement in antigen binding and signal transduction. Although it is generally assumed that on most mature B cells, mIgM and mIgD occur in roughly equal quantities, no formal analysis of this question has been reported. In this report, we describe such an analysis based on the quantitation of anti-Fab or anti-kappa specific immunofluorescence of splenic B cells before or after capping with rabbit anti-IgD or anti-IgM antibodies or both. The results indicate that, whereas mean expression of IgD exceeds IgM on splenocytes by threefold, members of the major B cell subpopulation (60 to 70% of cells) express 10-fold more IgD than IgM.