Patent Application
20210355441
A Sequence Listing accompanies this application and is submitted as an ASCII text file of the sequence listing named “960296_04147_ST25.txt” which is 18.5 KB in size and was created on Apr. 30, 2021. The sequence listing is electronically submitted via EFS-Web with the application and is incorporated herein by reference in its entirety.
De novo production of hematopoietic stem cells (HSCs) from in vitro expandable human cells, such as pluripotent stem cells (hPSCs), represents a promising approach for stem cell-based therapies and modeling of hematologic diseases. However, generation of HSCs and lymphoid cells from hPSCs remains a significant challenge. During development, blood cells and HSCs arise from hemogenic endothelium (HE) via a definitive hematopoiesis program that produces the entire spectrum of adult-type erythro-myeloid progenitors (EMP) and lymphoid cells. Importantly, these in vivo-produced cells have the capacity to provide long-term cell repopulation in an adult recipient following engraftment. Although previous studies have successfully generated hematopoietic progenitors (HPs) with a HSC phenotype and limited engraftment potential from pluripotent stem cells (PSCs), cells with the capacity for robust and consistent engraftment with recapitulation of the full spectrum of terminally differentiated hematopoietic cells, including lymphoid cells, has not been achieved. Thus, identifying key cellular and molecular programs required for proper lymphoid and HSC specification in vitro is essential to overcome the current roadblocks to advance the lymphoid cell and HSC manufacturing technology.
The present invention provides methods of enhancing HOXA gene expression and arterial specification of hemogenic endothelium with superior lymphomyeloid potential in differentiating human pluripotent stem cells (hPSCs). The methods comprise (a) introducing an inducible SOX17 transgene into a population of hPSCs; (b) culturing the hPSCs for at least two days under conditions to differentiate the hPSC into KDR+ mesoderm cells; and (c) inducing expression of the SOX17 transgene in the KDR+mesoderm cells and culturing for at least two days, such that DLL4+CXCR4+ arterial hemogenic endothelium (AHE) cells are obtained.
In some embodiments, the methods involve transducing the hPSCs with a vector comprising an inducible promoter operably linked to the SOX17 transgene.
In another aspect, the present invention provides cell populations produced by the methods disclosed herein.
In another aspect, the present invention provides isolated in vitro populations of DLL4+CXCR4+ AHE cells differentiated from an hPSC population comprising a SOX17 transgene.
In another aspect, the present invention provide a method of expansion of hematopoietic progenitors comprising a) generating hemogenic endothelium (HE) cells in presence of SOX17 upregulation; b) culturing the HE cells on OP9 or OP9-DLL4 cells in medium comprising FLT3L, TPO SCF, IL6, and IL3 for at least an additional 5 days; c) collecting the floating hematopoietic progenitor cells (HP); and d) culturing the HPs of step c in medium comprising FLT3L, TPO SCF, IL6, and IL3 for at least an additional 5 days to expand HPs with myeloid and lymphoid potential. In some aspects, the method further comprises (e) passaging the cells of step (d) for at least two weeks in medium comprising SCF, FLT3L and IL-7 for at least two weeks to produce CD4+CD8+ T cells.
In some embodiments, the cells are further differentiated to form an isolated in vitro T cell population that may be used for several downstream applications, such as the generation of exogenous chimeric antigen receptors (CARs).
The generation of functional hematopoietic stem cell (HSC)-like cells from pluripotent stem cells (PSCs) has been a long-sought goal in hematology research. Previous efforts to generate cells with myeloid and T cell hematopoietic potential from human pluripotent stem cells (hPSCs) have produced few, if any, cells capable of engrafting in irradiated mice.
Recent advances in understanding the major bottlenecks in the derivation of engraftable hematopoietic cells and definitive lympho-myeloid progenitors from PSCs have identified deficiencies in NOTCH and HOXA signaling as major contributing factors to the observed functional deficiencies. However, the hierarchy of molecular events that are critical to establishing these programs is still poorly understood. In the present application, the inventors identified the transcription factor SOX17 as a critical upstream activator of these pathways. As is described in the Examples, the inventors generated SOX17-knockout and SOX17-inducible human hPSCs, and using cell biology and molecular profiling, they show that SOX17 activates HOXA and arterial programs in hemogenic endothelium (HE) and establishes definitive lympho-myeloid hematopoiesis. The inventors further show that SOX17 produces these effects through activation of NOTCH and CDX2 signaling.
Based on these findings, the inventors have devised a novel strategy for producing superior HE that expresses HOXA cluster genes and has enhanced T lymphoid potential. Importantly, these methods enable enhanced T cell production, which could facilitate the development of off-the-shelf immunotherapies derived from hPSCs.
The present invention provides methods of enhancing arterial specification of hemogenic endothelium in differentiating human pluripotent stem cells (hPSCs). The methods involve (a) introducing an inducible SOX17 transgene into a population of hPSCs; (b) culturing the hPSCs for at least two days under conditions to differentiate the hPSC into mesoderm cells; and (c) inducing expression of the SOX17 transgene in the mesoderm cells on at least about day two of differentiation, such that DLL4+CXCR4+ arterial hemogenic endothelium (AHE) cells are obtained.
As is described in the Examples, the inventors discovered that they could promote the arterial program and HOXA gene expression in hemogenic endothelium derived from hPSCs in vitro by overexpressing SOX17 at a specific time (i.e., starting at day 2) and for a defined period (i.e., 2-4 days) during differentiation. SOX17 is a member of the Sry-related high mobility group domain (SOX) family of transcription factors, and is key developmental regulator of endothelial and hematopoietic lineages. In the methods of the present invention, differentiating hPSCs are forced to overexpress SOX17 during the mesoderm differentiation by introducing an inducible SOX17 transgene into the population of mesodermal cells to facilitate formation of AHE with robust lymphomyeloid potential. In addition, inventors discovered that an SOX17 expands in suspension cultures lymphomyeloid progenitors generated from hemogenic endothelium. This method provided increased amounts of progenitors which could differentiate into lymphoid cells.
The SOX17 transgene used with the present invention may comprise any nucleic acid sequence encoding the SOX17 protein. For Example, the SOX17 transgene may be obtained by amplifying the SOX17 gene sequence from the genomic locus in human cells or by amplifying SOX17 mRNA from hPSCs differentiated into endothelial and blood cells and converting it into cDNA. Alternatively, genomic DNA or cDNA clones can be obtained commercially (e.g., from Sino Biological, Origene, IDT, etc.). In some embodiments, the transgene comprises SEQ ID NO:58, a cDNA sequence encoding the human SOX17 protein.
In some embodiments, the SOX17 transgene further comprises a vector sequence that can be used to drive the expression of the SOX17 transgene within the cells. In these embodiments, the transgene is introduced by into the population of hPSCs by transducing the cells with said vector. As used herein, the term “vector” refers to a nucleic acid molecule capable of propagating another nucleic acid to which it is linked. The term includes the vector as a self-replicating nucleic acid structure as well as the vector incorporated into the genome of a host cell into which it has been introduced. Certain vectors are capable of directing the expression of nucleic acids to which they are operatively linked. Such vectors are referred to herein as “expression vectors”. Vectors suitable for use with the present invention comprise a nucleotide sequence encoding a SOX17 transgene and a heterogeneous sequence necessary for proper propagation of the vector and expression of the encoded polypeptide. The heterogeneous sequence (i.e., sequence from a different species than the transgene) can comprise a heterologous promoter or heterologous transcriptional regulatory region that allows for expression of the polypeptide. Suitable vectors for the expression of the SOX17 transgene include plasmids and viral vectors. In a preferred embodiment, the vector comprises heterologous sequence that allows the transient and/or inducible expression of the encoded SOX17 protein.
In some embodiments, the vector includes a transposase system, such as the PiggyBac transposon system (see Examples). The PiggyBac transposon is a TTAA-specific mobile genetic element that efficiently transposes between vectors and chromosomes via a “cut and paste” mechanism. PiggyBac transposase recognizes transposon-specific inverted terminal repeat sequences (ITRs) and moves the intervening contents to a TTAA insertion site in a chromosome or another vector. Thus, inserting a gene of interest between two ITRs in a transposon vector allows one to efficiently insert the gene into a target genome. Other suitable transposase systems for use with the present invention include, for example, Sleeping Beauty.
In other embodiments, the vector is a plasmid, a viral vector, a cosmids, or an artificial chromosome. Suitable plasmids include, for example, E. coli cloning vectors. Many suitable viral vectors are known in the art and include, but are not limited to, an adenovirus vector; an adeno-associated virus vector; a pox virus vector, such as a fowlpox virus vector; an alpha virus vector; a baculoviral vector; a herpes virus vector; a retrovirus vector, such as a lentivirus vector; a Modified Vaccinia virus Ankara vector; a Ross River virus vector; a Sindbis virus vector; a Semliki Forest virus vector; and a Venezuelan Equine Encephalitis virus vector. In one particular embodiment, the vector comprises SEQ ID NO:1.
In some embodiments, the vector is an expression vector that comprises a promoter that drives the expression of the SOX17 transgene, preferably transient or inducible expression of the SOX17 transgene. As used herein, the term “promoter” refers to a DNA regulatory region capable of binding RNA polymerase in a cell and initiating transcription of a coding sequence. Although promoters are most commonly found immediately upstream of a coding sequence, they may also be found downstream of or within the coding sequence. Promoters may be derived in their entirety from a native gene or may be composed of multiple elements, including elements derived from promoters found in nature or elements comprising synthetic DNA sequences. It is understood by those skilled in the art that different promoters may direct the expression of a gene in different tissues or cell types, at different stages of development, or in response to different environmental conditions. Preferably, the promoters used with the present invention are inducible promoters. An “inducible promoter” is a promoter that is activated (i.e., initiates transcription) only in the presence of a particular molecule. Inducible promoters allow tight control the expression of a transgene within cells. Many suitable inducible expression systems are known in the art and include, for example, Tet-On gene expression systems that allow one to induce the expression of a gene by administering tetracycline (Tc) or tetracycline-derivatives like doxycycline (DOX). Suitable Tet-On systems for use with the present invention include, without limitation, Tet-On Advanced and Tet-On 3G. Tet-On systems utilize several promoters, including both minimal promoters (e.g., CMV) flanked by a tetracycline response element (TRE) and engineered Tet-inducible promoters (e.g., TRE2 and TREtight). For instance, in the Examples, the SOX17 transgene is inserted (i.e., via conventional cloning methods) downstream of the doxycycline-inducible TREtight promoter within a vector. This vector was introduced into the hPSCs, allowing the inventors to induce expression of the SOX17 transgene at the desired stage of differentiation by adding doxycycline to the cell culture to activate expression from the TREtight promoter. Those of skill in the art are aware of many additional inducible gene expression systems, including both chemical-inducible and temperature-inducible systems. Other suitable inducible gene expression systems for use with the present invention include, without limitation, the glucocorticoid-responsive mouse mammary tumor virus promoter (MMTVprom), the tamoxifen-responsive hormone-binding domain of the estrogen receptor (ERTAM), the ecdysone-inducible promoter (EcP), heat shock inducible promoters (e.g., Hsp70 or Hsp90-derived promoters), and the T7 promoter/T7 RNA polymerase system (T7P). The SOX17 transgene may be introduced into the hPSCs using any suitable method, for example by transfection or transduction. In one embodiment, the transgene is introduced by transducing the hPSCs with a vector comprising the SOX17 transgene. In another embodiment, the hPSCs are transduced with an exogenous SOX17 mRNA. In yet another embodiment, the hPSCs are transduced with the SOX17 protein.
In the present methods, the hPSCs are cultured under conditions for at least two days to differentiate these cells into mesoderm cells. Notably, mesoderm cells can be identified by their KDR+ phenotype. Methods of differentiating hPSCs into progenitor mesoderm cells are known in the art. In one embodiment, mesoderm cells are obtained by culturing hPSCs in a chemically defined culture medium for about 2 days to about 4 days, whereby a cell population comprising mesoderm cells is obtained. For example, the hPSCs may be cultured in xenogen-free, serum-albumin free chemically defined medium comprising BMP4, activin A, LiCl, and FGF2, as described in Uenishi et al. (Stem cell reports (2014): 1073-1084) and U.S. Pat. No. 9,938,499, which is incorporated by reference in its entirety. For example, the chemically defined medium comprises about 10 ng/ml to about 50 ng/ml FGF2, about 50 ng/ml to about 250 mg/ml of BMP4 (e.g., 50 ng/ml to about 500 ng/ml BMP4), about 10 ng/ml to about 15 ng/ml Activin A, and about 1 to 2 mM LiCl under hypoxic conditions. In other embodiments, the cells are attached to a culture plate via extracellular matrix proteins. For example, in one embodiment, the cells are attached via collagena, fibronectin, matrigel™ or Tenascin C (TenC). In a preferred embodiment, the cells are cultured on plates coated with Collagen IV, as described in Uenishi et al. and U.S. Pat. No. 9,938.499.
The term “defined culture medium” is used herein to indicate that the identity and quantity of each medium ingredient is known. As used herein, the terms “chemically-defined culture conditions,” “fully defined, growth factor free culture conditions,” and “fully-defined conditions” indicate that the identity and quantity of each medium ingredient is known and the identity and quantity of supportive surface is known. As used herein the term “xenogen-free” refers to medium that does not contain any products obtained from a non-human animal source. As used herein, the term “serum albumin-free” indicates that the culture medium used contains no added serum albumin in any form, including without limitation bovine serum albumin (BSA) or any form of recombinant albumin. Standardizing culture conditions by using a chemically defined culture medium minimizes the potential for lot-to-lot or batch-to-batch variations in materials to which the cells are exposed during cell culture. Accordingly, the effects of various differentiation factors are more predictable when added to cells and tissues cultured under chemically defined conditions. As used herein, the term “serum-free” refers to cell culture materials that do not contain serum or serum replacement, or that contain essentially no serum or serum replacement. For example, an essentially serum-free medium can contain less than about 1%, 0.9%, 0.8%, 0.7%, 0.6%, 0.5%, 0.4%, 0.3%, 0.2%, or 0.1% serum. “Serum free” also refers to culture components free of serum obtained from animal blood and of animal-derived materials, which reduces or eliminates the potential for cross-species viral or prion transmission. Further, serum-containing medium is not chemically defined, producing a degree of variability in the culture conditions. Suitable defined media include, but are not limited to, E8 medium.
Expression of the SOX17 transgene may be induced using any suitable methods known in the art. For example, in embodiments in which the transgene is introduced as a vector comprising an inducible promoter operably linked to the SOX17 transgene, transgene expression can be induced by contacting the cells with the signaling molecule required for activation of the chosen promoter. In embodiments in which the transgene is introduced as SOX17 mRNA or protein, induction is accomplished by transducing with cells with the mRNA or protein at the appropriate stage of hPSC differentiation.
In a preferred embodiment, expression of the SOX17 transgene is induced at about day 2 of hPSC differentiation, i.e., in the mesoderm stage of development. In embodiments in which the transgene is provided as a vector, induction may involve treating the cells with the signaling molecule required to activate an inducible promoter (e.g., DOX). Alternatively, one can induce SOX17 expression by transfecting the hPSCs with mRNA encoding SOX17 or with the SOX17 protein at about day 2 of differentiation. The inventors discovered that inducing SOX17 expression for two to four days (i.e., from day 2 to day 4-6 of differentiation) generated the greatest number of CD43+/CD45+ cells (
In another embodiment, SOX17 was upregulated in CD43+ hematopoietic progenitors generated from hemogenic endothelium on day 9 of hematopoietic differentiation to further expand lymphomyeloid progenitors. This additional SOX17 upregulation resulted in an greater expansion of lymphomyeloid progenitors when compared to the method without such upregulation.
The human pluripotent stem cells (hPSCs) used with the present methods may be embryonic stem cells (ESCs) or induced pluripotent stem cells (iPSCs). Many stem cells lines are commercially available. For example, the inventors utilized a H9 hESC (WA09) line from WiCell.
The goal of the present invention is to produce a population of arterial type hemogenic endothelial cells (AHEs) with broad lymphoid potential that are able to give rise to engraftable hematopoietic cells. These superior AHE cells can be identified as having the DLL4+CXCR4+ phenotype. CXCR4 is a HSC homing receptor that is not present in hematopoietic progenitors produced by traditional hESC differentiation methods, and DLL4 (Delta Like Canonical Notch Ligand 4) is a NOTCH-signaling ligand expressed by HE in vivo. In the Examples, the inventors show that by inducing SOX17 expression in differentiating hPSCs they produce a greater number of AHE cells with these desirable characteristics. Specifically, they found that SOX17 expression increases the production of AHE cells by about 50%.
As is demonstrated in the Examples, inducing the expression of SOX17 causes upregulation of genes encoding HOXA family members and arterial markers, which facilitates arterial specification of HE from mesoderm cells. In the present methods, hPSCs are cultured under conditions to differentiate the hPSC into mesoderm cells for about 4 days, and 24 hours later (on day 5) the HE begin to specify into DLL4+CXCR4+/− arterial HE and non-arterial DLL4− CXCR4− HE. Enhanced arterial specification may be detected at day 4 and 5 as (1) an increase in the percentage of VEC+CD73−CD43− HE cells generated, (2) an increase in the percentage of cells expressing DLL4 and CXCR4, which express increased levels of arterial marker genes (EFNB2, NOTCH4, HEY1, CXCR4, DLL4) and produce higher numbers of blood cells with colony forming cell (CFC) potential and/or T cell potential, as compared to the cells generated when SOX17 expression is not induced.
The inventors discovered that SOX17 expression enhances arterial specification via activation of several signaling pathways, including the Notch signaling pathways. Additionally, the inventors demonstrated that SOX17 is essential for the expression of HOXA genes in AHE. Thus, in some embodiments, the AHE cells produced by the methods disclosed herein express one or more HOXA genes selected from the group consisting of HOXA5, HOXA7, HOXA9, HOXA10, and HOX11. HOX genes have been implicated in hematopoiesis across species. For example, HOXA9 is the key homeotic gene that defines HSC identity, supporting HSC renewal during embryogenesis and stress hematopoiesis. HOXA5, HOXA9, and HOXA10 are transcriptional targets of Notch signaling in T cell progenitors, suggesting a role for these protein in T-lymphopoiesis. While SOX17 expression was previously known to be correlated with HOXA expression, it was not known to be causative. Thus, the inventors' discovery that SOX17 activates HOXA expression represents a significant advance in understanding the mechanisms of hematopoietic specification.
Further, the inventors have determined that SOX17 activates HOXA expression by binding to the promoter of CDX2 (a master regulator of HOX gene expression) and upregulating its expression. Thus, in some embodiments, the AHE cells produced by the methods disclosed herein express CDX2.
The AHE cells produced by the methods of the present invention have definitive lympho-myeloid potential and are, thus, useful for the production of myeloid and lymphoid progenitor cells. Those of skill in the art may consult a standard cell differentiation protocol (e.g., Stem cell reports (2014): 1073-1084; Nature protocols (2011): 296-313; Stem cells and development (2011): 1639-1647; and Blood (2005): 617-626) to obtain cell populations of the desired hematopoietic cell type. For example, the AHE of the present invention may be further differentiated into T cells, beta-hemoglobin-producing red blood cells, megakaryocytic cells, and multipotential myeloid progenitors, including granulocyte, erythrocyte, megakaryocyte, macrophage (GEMM) and granulocyte-macrophage (GM) colony forming cells (CFCs), and mature myelomonocytic cells. In some embodiments of the present methods, the cells are further differentiated into lymphoid cell lines.
In the Examples, the inventors demonstrate that DLL4+CXCR4+ AHE cells produced by the present methods have increased T cell potential, i.e., they give rise to an increased number of T cells. Specifically, in DOX-treated conditions, DLL4+CXCR4+ AHE cells generate 54-fold more CD4+CD8+ T cells as compared to DLL4−CXCR4− HE cells. Scalable T cell production is essential to advance iPSC-based immunotherapies into the clinic. Thus, in preferred embodiments, the cells are further differentiated into T cells. T cells may be differentiated using known methods, including the method disclosed in the Examples (see the Materials and Methods section titled “T cell differentiation”). The T cells can be identified as CD4+CD8+. In some embodiments, the T cells are identified as CD7+CD5+, CD8+CD4+, or a combination thereof (CD7+CD5+ and CD8+/CD4+).
Suitably, in one embodiment, the day 4 HE or day 5 AHE cells were cultured on OP9-DLL4 in presence of TPO (about 20-70 ng/ml, e.g., 50 ng/ml), SCF (about 20-70 ng/ml, e.g., 50 ng/ml), IL-6 (about 10-40 ng/ml, e.g., 20 ng/ml), IL-3 (about 5-30 ng/ml, e.g. 10 ng/ml) and FLT3L (3-20 ng/ml, e.g., 10 ng/ml) for about 4-5 days to produce hematopoietic progenitors enriched in T cell potential which were differentiated into T cells by culturing in the presence of delta-like ligand 4, e.g., cells overexpressing DLL4 (DLL4-expressing cells, e.g., OP9-DLL4 cells) for a sufficient time to produce T cells (e.g., at least 2 weeks, preferably at least three weeks or more). The media further comprise about 5-20 ng/ml SCF, about 3-20 ng/ml FLT3L and about 3-20 ng/ml IL-7 to augment T cell production. For example, in one embodiment, the AHE cells are passaged weekly onto fresh OP9-DLL4 cells in media comprising about 10 ng/ml SCF, about 5 ng/ml FLT3L and about 5 ng/ml IL-7 for about 3 weeks. Cells can be analyzed by flow cytometry for T cell surface markers.
In another embodiment, the hematopoietic progenitors obtained from AHE cells are cultured in media comprising 5-50 ng/ml SCF, about 3-20 ng/ml FLT3L and about 3-20 ng/ml IL7 in the presence of DLL4-Fc (commercially available), DLL4-Fc ligands or on DLL4-OP9 cells for a sufficient time to differentiate into T cells.
It was also surprisingly found that expansion of hemogenic progenitors (HP) which are subject to a second increased expression of SOX17 at day 4+5 through day 4+5+5 led to an increased expansion of myeloid and T lymphoid progenitors, as demonstrated in
Thus, in another embodiment, CD43+ hematopoietic progenitors generated from day 4 hemogenic endothelium in coculture with OP9 or OP9-DLL4 in medium comprising 20-70 ug/ml TPO and SCF (e.g., 50 ug/ml), 10-30 ng/ml IL6 (e.g., 20 ng/ml), and 5-25 ng/ml IL3 and FLT3L (e.g., 10 ng/ml) for 5 days, were collected and cultured in low attachment plate with IF9S media with 50-150 ng/ml FLT3L (e.g. 100 ng/ml) FLT3L, TPO and SCF, and about 10-40 ng/ml (e.g., 20 ng/ml) IL6, 3-20 ng/ml, e.g. 10 ng/ml IL3, with 2 μM Doxycyclin to induce expression of SOX17. As demonstrated in
T cells produced by the methods disclosed herein may be used for the production of various therapeutics, including chimeric antigen receptors (CAR) T cells. CAR T cells are T cells that have been genetically engineered to produce an artificial T cell receptor (i.e., a CAR) that allows them to target a specific protein of choice. CARs comprise an antigen-specific recognition domain that binds to specific target antigen or cell and a transmembrane domain linking the extracellular domain to an intracellular signaling domain. CAR T cells are commonly designed to recognize cancer cells (e.g., via recognition of an antigen that is present on the tumor surface) for use in cancer immunotherapies. Thus, in one embodiment, the methods of the present invention further comprise using the AHE to generate CAR expressing T cells that can be used to kill tumor cells. Methods of designing and producing CART cells are known in the art.
The present invention also encompasses cell populations produced by the methods disclosed herein. In some embodiments, the AHE cells are sorted from the cell culture, e.g., based on expression of DLL4 and CXCR4. The resulting hemogenic cell population will contain a SOX17 transgene, and may be at least 90%, 95% or 99% pure.
In another aspect, the present invention provides hPSC populations that comprise a SOX17 transgene and are capable of differentiating into DLL4+CXCR4+ arterial hemogenic endothelium (AHE) cells. In some embodiments, the hPSC cells comprise a vector comprising the SOX17 transgene. In some embodiments, said vector comprises an inducible promoter operably linked to the SOX17 transgene (e.g., the vector of SEQ ID NO:1 or a vector comprising SEQ ID NO:58), allowing for induction of SOX17 expression within the cells. Suitable vectors are discussed in the previous section.
Additionally, the present invention provides isolated in vitro populations of DLL4+CXCR4+ arterial hemogenic endothelium (AHE) cells differentiated from an hPSC population comprising a SOX17 transgene. As used herein, the phrase “isolated in vitro population” refers to a population of cells that is grown outside of a living organism (e.g., in a test tube, flask, or culture dish) under defined conditions.
These DLL4+CXCR4+ AHE cells may be further differentiated into T cells, forming an isolated in vitro T cell population. T cell differentiation may be accomplished using known methods, as discussed above. In some embodiments, the T cell population comprises more than 90% CD4+CD8+ T cells, alternatively at least 95% CD4+CD8+ T cells.
In some embodiments, the AHE derived herein can produce hematopoietic progenitors (HP) with a high T cells potential. As used herein, the phrase “hematopoietic progenitors with high T cell potential” refers to a population of HE cells that is able to differentiate into T cells at least 10 fold more efficiently.
Further, in some embodiments, the present invention provides hematopoietic progenitors (HP) derived from the AHE that have high myeloid and T cell potential, as demonstrated in Example 2. These cells are differentiated by a method comprising at least two induction of SOX17 expression, leading to a more robust production of cells with myeloid and T cell potential as compared to without SOX17 expression.
As is discussed in the previous section, T cells are useful for the production of therapeutics, including, for example, chimeric antigen receptors (CAR) T cells. Thus, in some embodiments, the T cells are engineered to express an exogenous chimeric antigen receptor (CAR).
The following non-limiting examples are included for purposes of illustration only, and are not intended to limit the scope of the range of techniques and protocols in which the compositions and methods of the present invention may find utility, as will be appreciated by one of skill in the art and can be readily implemented. The present invention has been described in terms of one or more preferred embodiments, and it should be appreciated that many equivalents, alternatives, variations, and modifications, aside from those expressly stated, are possible and within the scope of the invention.
SOX17 has been implicated in arterial specification and the maintenance of hematopoietic stem cells (HSCs) in the murine embryo. However, knowledge about molecular pathways and stage-specific effects of SOX17 in humans remains limited. Here, using SOX17-knockout and SOX17-inducible human pluripotent stem cells (hPSCs), paired with molecular profiling studies, we reveal that SOX17 is a master regulator of HOXA and arterial programs in hemogenic endothelium (HE), and is required for the specification of HE with robust lympho-myeloid potential and DLL4+CXCR4+ phenotype resembling arterial HE at sites of HSC emergence. Along with activation of NOTCH signaling, SOX17 directly activates CDX2 expression leading to the upregulation of the HOXA cluster genes. Since deficiencies in NOTCH signaling and HOXA regulation were identified as major contributing factors to the impaired engraftment potential of hPSC-derived hematopoietic cells, identification of SOX17 as a key regulator linking arterial and HOXA programs in HE may help to program the HSC fate from hPSCs.
Sox17 has been found to be expressed in the arterial vasculature (Liao et al., 2009) and the hemogenic endothelium (HE) in aorta-gonad-mesonephros (AGM) region (Clarke et al., 2013; Corada et al., 2013), in which it is required for arterial specification (Corada et al., 2013) and essential for establishing the definitive, but not primitive, hematopoietic program (Clarke et al., 2013) within the murine embryo. Although Sox17 actively prevents endothelial-to-hematopoietic transition (EHT) by repressing Runx1 (Lizama et al., 2015), Sox17 remains critical for maintaining intra-aortic hematopoietic clusters (IAHC) and fetal liver HSCs (Kim et al., 2007; Nobuhisa et al., 2014; Saito et al., 2018). Transduction of human embryonic stem cell (hESC)-derived CD34+ HE/OP9 cocultures with a tamoxifen-inducible murine Sox17 transgene revealed that tamoxifen treatment expands CD34+CD43+CD45−/low cells co-expressing the endothelial marker VE-cadherin (VEC) (Nakajima-Takagi et al., 2013). Although these expanded cells possessed the capacity to form compact colonies in hematopoietic CFC medium with SCF, TPO and IL3, they were interpreted as HE cells. In mouse studies, the effects of Sox17 were attributed to the activation of the NOTCH signaling pathway by its direct binding to Dll4, Notch1 and Notch4 loci (Clarke et al., 2013; Corada et al., 2013). However, no activation of NOTCH pathway following Sox17 overexpression was observed during hESC differentiation (Nakajima-Takagi et al., 2013). While these studies established an important role of SOX17 in specification of definitive hematopoiesis and its diverse effects on EHT and HSCs, the molecular program induced by SOX17 at distinct stages of hematopoietic development, especially in humans, remains poorly understood.
To define the mechanisms of SOX17 action during specification and diversification of HE, we established SOX17-knockout and SOX17-inducible hESC lines and assessed their differentiation in a 2D chemically defined, feeder- and xeno-free human pluripotent stem cell (hPSC) differentiation system in which all stages of hematopoietic development are temporally, phenotypically, and functionally defined (Uenishi et al., 2014). In this study we specifically focused on the earliest stages of HE emergence and its arterial specification which have not been previously assessed. We reveal that SOX17 is required for the activation of HOXA expression and establishing arterial-type HE (AHE) with robust lympho-myeloid potential that can be identified by DLL4+CXCR4+ phenotype resembling AHE at sites of HSC emergence in vivo. Furthermore, the SOX17 effects are mediated by CDX2. These findings are important for understanding the molecular mechanisms controlling HE and definitive blood lineage development and designing strategies for specifying HSC fate from hPSCs.
We generated an inducible SOX17 cell line using the PiggyBac system (Park et al., 2018a). Human SOX17 CDS was cloned into PiggyBac transposon vector (Transposagen) downstream of TREtight promoter of pTRE-P2A-Venus-rpEF1a-Zeo plasmid, and cotransfected with pEF1α-M2rtTA-T2A-Puro and transposase plasmid into H9 hESCs using human stem cell nucleofector kit 2 (Lonza). Cells were selected in Zeocin (0.5 μg/ml, Thermofisher) and Puromycin (0.5 μg/ml, Sigma) for 10 days and resistant clones screened for Venus expression following DOX (Sigma) treatment. To generate SOX17−/− knockout H9 ESC line, two single guide RNAs were designed in CRISPR design tool (Synthego). The two sgRNA sequences are provided as SEQ ID NO:2 and SEQ ID NO:3. H9 ESCs were electroporated with the two sgRNAs and Cas9 protein (PNA Bio), and then plated at a low density on 6 well plate. After 7 days, individual colonies were picked and further expanded. After expansion, individual clones were screened by genomic PCR for the acquisition of 830 bp deletion in wildtype SOX17 allele using primers P1 and P2.
hESC Lines Maintenance and Hematopoietic Differentiation
hPSCs (H9 hESC (WA09) line from WiCell), iSOX17 H9 line and knockout SOX17 H9 line were maintained and passaged on Matrigel in mTeSR1 media (WiCell). The cell lines were differentiated on collagen IV (ColIV)-coated plate (Uenishi et al., 2014). Cell lines were plated at a density of 5,000 cells/cm2 onto 6 well plates with E8 media containing 10 μM Rock inhibitor (Y-27632, Cayman Chemicals). The following day, the media was changed to IF9S media with 50 ng/ml FGF2 (PeproTech), 50 ng/ml BMP4 (PeproTech), 15 ng/ml Activin A (PeproTech), and 2 mM LiCl (Sigma), and cultured in hypoxia (5% CO2, 5% O2). On day 2, the media was changed to IF9S media with 50 ng/ml FGF2, 50 ng/ml VEGF (PeproTech), and 2.5 μM TGF-β inhibitor (SB-431542, Cayman), and cultured in hypoxia (5% CO2, 5% O2). On days 4 and 6, the media was changed to IF9S media with 50 ng/ml FGF2, 50 ng/ml VEGF, 50 ng/ml TPO (PeproTech), 50 ng/ml IL-6 (PeproTech), 20 ng/ml SCF (PeproTech), and 10 ng/ml IL-3 (PeproTech), and cultured in normoxia (20% CO2, 5% O2). DOX (Sigma) was added to cultures on day 2 of differentiation at concentration of 2 μg/ml.
HB-CFCs were detected using a semisolid colony-forming serum-free medium (CF-SFM) containing 40% ES-Cult M3120 methylcellulose (2.5% solution in IMDM, Stem Cell Technologies), 25% StemSpan serum-free expansion medium (SFEM, Stem Cell Technologies), 25% human endothelial serum-free medium (ESFM, ThermoFisher), 10% BIT 9500 supplement (Stem Cell Technologies), GlutaMAX (1/100 dilution, ThermoFisher), Ex-Cyte (1/1000 dilution, Millipore), 100 μM MTG, 50 μg/ml ascorbic acid and 20 ng/ml FGF (Peprotech) (Vodyanik et al., 2010). Hematopoietic CFCs were detected using serum containing H4435 MethoCult with FGF, SCF, IL-3, IL-6 and EPO (Stem Cell Technologies) following plating 1000 CD43+ cells/dish in duplicates. CFCs numbers recalculated per 105 cells.
Immature/primordial HE cells were isolated from knockout SOX17 or DOX+ and DOX− iSOX17 differentiation cultures by CD31 MACS (Miltenyi Biotec) at D4. Isolated cells were plated on OP9 or OP9-DLL4 in α-MEM (Gibco) with 10% FBS (Hyclone) with TPO, SCF (50 ng/ml), IL-6 (20 ng/ml), IL-3 and FLT3L (10 ng/ml; all from Peprotech). The media was changed 24 hours later, and extra media was added another 2 days later. After 5 days in secondary culture, cells were collected and assessed for CFC and T cell potential.
H9, iSOX17 and SOX17−/− ESCs were collected on day 5 of differentiation, singularized by 1× TrypLE, and stained for VEC (CD144), CD73, CD43, DLL4, CXCR4 with dead cells excluded using Ghost Dye Violet 540 (Tonbo Biosciences). FMO controls for flow cytometric analysis are shown in
Expansion of Hematopoietic Progenitors.
On day 4, HE cells were isolated from DOX+ or DOX− cultures by CD31 MACS. Isolated cells were plated on OP9 or OP9-DLL4 in α-MEM with 10% FBS with 50 ug/ml TPO and SCF, 20 ng/ml IL6, and 10 ng/ml IL3 and FLT3L. The media was changed 24 hours later, and extra media was added another 2 days later. After 5 days in in secondary culture, floating cell were collected and plated lin ow attachment 24 well plate. Cell were cultured with IF9S media with 100 ng/ml FLT3L, TPO and SCF, 20 ng/ml IL6, 10 ng/ml IL3, and with or without 2 μM Doxycyclin, and extra media was added another 2 days later. After 5 days, cells were collected and assessed for CFC and T cell potential.
Floating hematopoietic cells were collected from day 9 differentiation cultures or day 5 secondary OP9 or OP9-DLL4 cocultures (D4 HE+5 or D5 HE +5), and were cultured on OP9-DLL4 in α-MEM with 20% FBS, 10 ng/ml SCF, 5 ng/ml FLT3L and IL-7 (PeproTech) on OP9-DLL4 for 3 weeks. Cells were passaged weekly onto fresh OP9-DLL4 cells. Cells were analyzed by flow cytometry for T cell surface markers after 21 days.
DAPT Treatment and CDX2 Knockdown in Differentiation Cultures Using siRNA
Notch signaling was blocked by DAPT (-secretase inhibitor, 10 μM, Cayman Chemical) added on day 3 of differentiation. For knockdown of CDX2 expression, DOX-treated iSOX17 cells were transfected with 100 nM CDX2 siRNA SMARTpool (Dharmacon) or Scramble negative control siRNA (Dharmacon) on day 3 of differentiation using Lipofectamine RNAiMAX (ThermoFisher). Next day, differentiation media was replaced with fresh media and cells were harvested at day 5 of differentiation.
Apoptosis was detected by flow cytometry using Annexin V (BD). For cell-cycle analysis, D5 cells were incubated in culture medium with BrdU (10 μM, BD Pharmingen) for 2 hours and stained with antibodies. For BrdU detection, the BrdU flow kit with 7 AAD was used and performed per the manufacturer's instructions. Fluorescent reagents used for analysis, cell viability, apoptosis, and proliferation are listed in Table 1.
Cell extracts were prepared by adding IP Lysis buffer (ThermoFisher) with protease inhibitor cocktail (Sigma). Cell lysates (10 μg) were separated by Mini-protean TGX gels (Bio-rad). The separated proteins were transferred to a PVDF membrane, and were stained with antibodies for SOX17 (R&D) and GAPDH (Santa Cruz). Immunoblots were visualized using the ECL PLUS detection kit (Amersham Pharmacia) and analyzed using ChemiDox XRS+ Image Lab Software Version 5.2.1 (Bio-Rad).
Real-Time qPCR
RNA was extracted using the RNeasy Plus Micro Kit (Qiagen). RNA was reverse-transcribed into cDNA using random hexamer primers (Qiagen) with SMART MMLV reverse transcriptase (TaKaRa). qPCR was conducted using TB Green Advantage qPCR Premix (TaKaRa). RPL13A was used as the reference gene to normalize the data. Primer sequences are listed in Table 2.
One hundred nanograms of total RNA was used to prepare sequencing libraries using the Ligation Mediated Sequencing (LM-Seq) protocol, according to the paper guidelines (Hou et al., 2015) and quantified with the Qubit fluorometer (ThermoFisher). Final cDNA libraries were quantitated with the Qubit Fluorometer (ThermoFisher), multiplexed, loaded at a final concentration of 2.5 nM, and sequenced as single reads on the Illumina HiSeq 3000 (Illumina).
Chromatin immunoprecipitation (ChIP) analysis of day 4 HE was performed, as described in the protocol included in the EZ-Magna ChIP A/G Chromatin Immunoprecipitation Kit (Millipore Sigma). Five nanograms of IP or control DNA was used to prepare sequencing libraries using the TruSeq ChIP Sample Preparation Kit (Illumina) as per the manufacturer instructions and quantified with the Qubit fluorometer (Life Technologies). All six TruSeq ChIP indexed samples were pooled per lane, loaded at a final concentration of 2.5 nM, and sequenced as single reads on the Illumina HiSeq 3000 (Illumina).
Day 4 HE cells were harvested and frozen in culture media containing FBS and 5% DMSO. Cryopreserved cells were sent to Active Motif to perform the ATAC-seq assay. The cells were then thawed in a 37° C. water bath, pelleted, washed with cold PBS, and tagmented as previously described (Buenrostro et al., 2013). Briefly, cell pellets were resuspended in lysis buffer, pelleted, and tagmented using the enzyme and buffer provided in the Nextera Library Prep Kit (Illumina). Tagmented DNA was then purified using the MinElute PCR purification kit (Qiagen), amplified with 10 cycles of PCR, and purified using Agencourt AMPure SPRI beads (Beckman Coulter). Resulting material was quantified using the KAPA Library Quantification Kit for Illumina platforms (KAPA Biosystems), and sequenced with PE42 sequencing on the NextSeq 500 sequencer (Illumina).
RNA-seq analyses were performed on three biological replicates in DOX− and DOX+ conditions. Sequencing fragments were aligned by STAR (version 2.5.2b) to human genome (hg38) with gene annotations from GENCODE (version 27). Transcript expression levels were quantified by RSEM (version 1.3.0) and differentially expression analysis was performed by DESeq2 (version 1.22.2). KEGG gene sets were defined by MSigDB (version 6.1).
ATAC-seq analyses were performed on two biological replicates in DOX− and DOX+ conditions. Sequencing fragments were pre-processed by the company Active Motif, Inc. Briefly, ATAC-seq reads were mapped to the human genome by BWA with default settings. Only reads that passed Illumina's purity filter, aligned with no more than 2 mismatches, and mapped uniquely to human genome were used in the subsequent analysis. Duplicate reads (“PCR duplicates”) were removed. To calculate signals, human genome was divided into 32 bp bins and the number of reads in each bin was counted. In order to smooth the data, reads were extended to 200 bp. To normalize signals across ATAC-seq datasets, the number of reads in each dataset was reduced by random sampling to the smallest number of reads present in the datasets.
ATAC-seq peaks were called by Active Motif, Inc using MACS2. For DOX− or DOX+ condition, we defined condition-specific peaks by selecting those existing in both ATAC-seq replicates of that condition and not overlapping with any peak from the two replicates of the other condition. From condition-specific peaks, we identified ‘promoter peaks’ by choosing those overlapped with protein-coding transcript's 5 kb upstream region and do not overlap with any intron or exon. DNA sequences for the 250 bp flanking regions to the center of promoter peaks were prepared for motif enrichment analysis. Motif enrichment was performed by MEME (version 5.0.4)'s CentriMo function with default settings based on motifs from HOCOMOCO human database (version 11).
ATAC-seq signals were calculated for gene's promoter region, which was defined as the 5 kb region upstream of its transcription start site (TSS). If a gene encoded for multiple transcripts, the most upstream TSS will be used as this gene's TSS. A gene's ATAC-seq promoter signal change upon DOX+ activation was computed by taking the difference of averaged signals from the two ATAC-seq replicates under DOX− or DOX+ condition. The top 5% genes that have the largest increase of ATAC-seq promoter signals were collected for GO term analysis by the Bioconductor package limma's function goana. P-values were adjusted by Benjamini & Hochberg method.
SOX17 and IgG control ChIP-seq fragments from were aligned by BWA (version 0.7.15) with a quality threshold at 5 for read trimming and all the other options in default settings. Normalized SOX17 ChIP-seq signals were calculated by MACS2 by using all the tags at the same loci. SOX17's fold enrichment over IgG control were calculated by MACS2 using all default options.
Experiments were analyzed using GraphPad Prism versions 8 (GraphPad Software Inc.) and Microsoft Excel (Microsoft Corporation). Tests for statistical significance are listed with each experiment and included two-sided Student's t-test for paired analyses and one-way ANOVA, and two-way ANOVA for experiments with multiple comparisons of or grouped variables, accompanied by Tukey and Sidak post hoc tests indicated as appropriate by the software. *p<0.05, **p<0.01, ***p<0.001, and ****p<0.0001. All error bars represent the mean±SD and duplicated or triplicated independent experiments.
SOX17 Knockout Impairs AHE Specification and Definitive Lympho-Myeloid Hematopoiesis from hPSCs
To assess the effect of SOX17 on hematopoietic development, we generated SOX17 knockout H9 human embryonic stem cell (hESC; SOX17−/−) lines using CRISPR/Cas9 (
To assess the hematopoietic potential of SOX17−/− and SOX17+/+ HE cells and their dependence on NOTCH signaling, we isolated D4 HE cells and co-cultured them on OP9 or OP9-DLL4 (
Overall, these results demonstrate the critical role of SOX17 in the specification of definitive lympho-myeloid hematopoiesis and DLL4+CXCR4+ AHE from hPSCs. Although both, DLL4 and CXCR4, are considered markers of AHE (Chong et al., 2011; Yamamizu et al., 2010), DLL4 expression has been found in arterial vessels of yolk sac and aorta (Duarte et al., 2004; Herman et al., 2018; Robert-Moreno et al., 2005), while CXCR4 expression was detected in aorta and vitelline/umbilical arteries (McGrath et al., 1999; Werner et al., 2020), i.e. vasculature harboring precursors capable of maturing into definitive HSCs (Dzierzak and Medvinsky, 1995; Gordon-Keylock et al., 2013), but not in yolk sac (McGrath et al., 1999; Venkatesh et al., 2008; Werner et al., 2020). Thus, SOX17 is the most essential factor for the formation of AHE with the CXCR4+ phenotype typical of HE with HSC potential in vivo.
To further characterize the role of SOX17 during hematoendothelial development, we engineered an H9 hESC line with a transgene cassette that expresses SOX17 upon treatment with doxycycline (DOX; iSOX17-hESCs;
As we previously demonstrated, D4 HE specifies 24 hours later (D5) into DLL4+CXCR4+/− arterial HE and non-arterial DLL4−CXCR4− HE (Park et al., 2018b; Uenishi et al., 2018). Assessment of HE phenotype on D5 (
Collectively, these studies suggest that SOX17 upregulation promotes definitive lympho-myeloid hematopoiesis from hPSCs through enhancement of AHE specification with CXCR4+DLL4+ phenotype, typical for HE at sites of HSC emergence.
To understand the molecular mechanisms of the effect of SOX17, we performed molecular profiling of D4 HE from DOX+ and DOX− cultures using RNA-seq and ATAC-seq. To analyze chromatin binding of SOX17 by ChIP-seq, we used DOX+ cultures because under DOX− conditions, SOX17 expression was absent in D4 HE, i.e., before AHE was formed (Uenishi et al., 2018) (
ATAC-seq analysis of D4 HE isolated from DOX-treated cultures identified 93,615 and 100,036 open chromatin regions in the two ATAC-seq replicates, respectively, of which 5,130 of which were specific to DOX+ conditions. Gene Ontology (GO) analysis of genes with increased ATAC-seq counts at promoters upon DOX treatment revealed enrichment in categories associated with development and morphogenesis, including blood vessel morphogenesis (Table 4), suggesting that SOX17 facilitates the establishment of gene regulatory networks essential for early morphogenesis, including vascular development. Motif-enrichment analysis of ATAC-seq peaks at promoters in DOX+ and DOX− conditions revealed enrichment in ETS-binding motifs for both conditions, consistent with the endothelial nature of the analyzed cells. However, in DOX+ conditions we observed a unique enrichment in retinoic receptor alpha (RARA) and estrogen receptor 2 (ESR2) motifs at open chromatin regions (
To identify direct targets of SOX17 in D4 HE, we analyzed overlapping SOX17 ChIP-seq and ATAC-seq peaks at promoters and intragenic regions of differentially expressed genes (DEGs) between DOX+ and DOX− conditions. The set of DEGs bound by SOX17 at open chromatin regions was enriched in the Hippo, Wnt, TGFβ, and Notch signaling pathways, and included genes important for NOTCH regulation, arterial specification and hematopoietic development such as NOTCH4, CDX2 (
To confirm the role of SOX17 in establishing the HOXA pattern in HE, we evaluated expression of arterial and HOXA genes in the three major subsets of HE on D5 (
QPCR analysis of phenotypically similar HE populations generated from SOX17−/− and wild type hESCs revealed substantial reduction of arterial genes in both DLL4+CXCR4− and DLL4+CXCR4+ AHE subsets as compared to wild type cells. However, significant reduction in HOXA5-HOX10 gene expression was observed only in DLL4+CXCR4+ AHE subset (
Taken together, our molecular profiling studies indicate that SOX17 acts as a key factor in activating the arterial program and HOXA expression in HE.
To determine whether SOX17 induction promotes arterial specification through activation of NOTCH signaling, we evaluated hematopoiesis following SOX17 upregulation in the presence of the NOTCH signaling inhibitor N—[N-(3,5-difluorophenacetyl)-L-alanyl]-S-phenylglycine t-butyl ester (DAPT) (
Despite significant upregulation of HOXA cluster genes in SOX17-expressing HE, our molecular profiling studies lacked of evidence for their direct regulation by SOX17. However, we found that SOX17 binds to and increases ATAC-seq counts of H3K27ac levels at the CDX2 promoter, along with the upregulation of CDX2 gene expression (
Overall, these observations indicate that the effect of SOX17 expression on establishing HOXA signature in HE is mediated through CDX2 signaling.
The critical role of SOX17 in HSC development has been well-recognized (Clarke et al., 2013; Kim et al., 2007; Lizama et al., 2015; Nobuhisa et al., 2014; Saito et al., 2018). Although SOX17 regulates multiple steps along the HSC developmental path, including HE specification, EHT, and HSC maintenance and expansion, the stage-specific molecular mechanisms of SOX17 are not well understood. Previously, overexpression of Sox17 was demonstrated to decrease cell numbers within the IAHC, while loss of Sox17 had the opposite effect (Lizama et al., 2015). However, the HSC potential of IAHC of manipulated AGM cells has not been characterized. The majority of cells within IAHC are differentiated hematopoietic cells, with only two of those cells possessing HSC potential (Kumaravelu et al., 2002; Solaimani Kartalaei et al., 2015). Thus, increase in IAHC cells following Sox17 downregulation could be associated with increase of lineage-committed progenitors which is accompanied by impaired HSC generation. This hypothesis is supported by observation that Sox17 loss leads to loss of CD45+VEC+ HSCs in AGM and fetal liver (Clarke et al., 2013; Kim et al., 2007). Demonstration of reduced multilineage CFC potential following knockdown of Sox17 in CD45lowCD117high cells from IAHC and expansion of undifferentiated hematopoietic cells following Sox17 overexpression (Nobuhisa et al., 2014) supports this hypothesis. In addition to regulating EHT and HSCs, SOX17 is essential for arterial specification and HE formation in the AGM (Clarke et al., 2013; Corada et al., 2013). Murine studies revealed that Sox17 effects are mediated through NOTCH signaling (Clarke et al., 2013; Lizama et al., 2015), while no effect of SOX17 on NOTCH signaling was observed in hESC cultures (Nakajima-Takagi et al., 2013).
A prior study revealed that Sox17 overexpression in hESC-derived during EHT expands VEC+CD34+CD43+CD45−/low cells with hematopoietic colony-forming potential (Nakajima-Takagi et al., 2013). In our study, we focused on defining the cellular and molecular pathways by which SOX17 regulates the earliest stages of HE specification and diversification from the mesoderm. Previously, we and others demonstrated that emerging HE cells lack arterial or venous characteristics (Ditadi et al., 2015; Uenishi et al., 2018) and express high levels of the mesodermal gene HAND1 (Uenishi et al., 2018). Therefore, we defined these cells as immature/primordial HE (Uenishi et al., 2018). When primordial HE cells get exposed to NOTCH signaling, they undergo arterial specification and formation of DLL4+CXCR4+/− AHE (Park et al., 2018b; Uenishi et al., 2018). Here, we show that SOX17 plays a critical role in specifying of AHE by upregulation of NOTCH4, DLL1 and DLL4 eventually leading to the formation of AHE with the DLL4+CXCR4+ phenotype typical of AHE at sites of HSC emergence, but not yolk sac AHE, which expresses DLL4, but not CXCR4 (McGrath et al., 1999; Venkatesh et al., 2008; Werner et al., 2020). Along with activating the arterial program, SOX17 is essential for expressing HOXA genes in AHE (
We found that SOX17 binds directly to the CDX2 promoter and increases chromatin accessibility and levels of H3K27ac activating histone modification leading to upregulated CDX2 expression. CDX2 knockdown with siRNA revokes SOX17-mediated effects on HOXA genes, thus demonstrating the critical role of a SOX17-CDX2 axis in establishing HOXA pattern in AHE. Genes of the CDX family (CDX1, CDX2 and CDX4) are well-known master regulators of HOX genes that mediate anterior-posterior patterning (Charite et al., 1998; Subramanian et al., 1995; van den Akker et al., 2002). In Wnt-activated epiblast stem cells, CDX2 binds to all four HOX cluster genes, including HOXA genes and is required for opening up the HOX cis-sequences (Neijts et al., 2017). Deficiency of cdx1 and cdx4 results in severe blood defects and altered expression of HOX genes in zebrafish (Davidson et al., 2003; Davidson and Zon, 2006). Similarly, impaired hematopoiesis from Cdx1, Cdx2 or Cdx4-deficient ESCs was observed in murine studies (Wang et al., 2008). Although ectopic expression of CDX4 enhanced definitive hematopoiesis from human and murine ESCs (Creamer et al., 2017; Wang et al., 2005), and hematopoietic engraftment in adult mice from murine ESCs (Wang et al., 2005), Cdx1 and Cdx4 double mutant mice were viable and did not show any hematopoietic defect (van Nes et al., 2006), which could be due to the observed functional redundancy of genes within the Cdx family (Davidson and Zon, 2006; Wang et al., 2008). It has been demonstrated that Cdx2 is the predominant Cdx gene expressed the AGM HE, while the expression of other Cdx genes is substantially lower (Gao et al., 2018). Cdx2 deficiency caused the most significant impairment in blood production from mouse ESCs (Wang et al., 2008) and Cdx1-Cdx2 compound conditional null mice failed to produce any blood at E11.5 (Foley et al., 2019), suggesting that among the Cdx family, Cdx2 is the most critical factor required for establishing hematopoiesis. Our finding of direct regulation of CDX2 expression by SOX17 provides an insight into the mechanisms responsible for establishing a CDX-HOXA pathway required for the formation of definitive AGM-like HE and lympho-myeloid hematopoiesis from hPSCs. Although previous studies with mouse ESCs found that overexpression of Cdx2 inhibits hematopoietic differentiation (McKinney-Freeman et al., 2008), such effect was not observed following upregulation of CDX2 by SOX17 in hESCs. This could be explained by differences in levels of upregulation or molecular programs activated by upregulation of CDX2 alone or in the context of SOX17 overexpression.
The de novo production of hematopoietic stem cells (HSCs) with robust multilineage reconstitution potential from human pluripotent stem cells (hPSCs) has long been sought after, but remains an elusive goal. Recent advances in understanding the molecular differences between HE and HSCs developed in vivo and their phenotypic counterparts produced from PSCs in vitro, have revealed that deficiencies in NOTCH and HOXA signaling are major factors responsible for aberrant functionality of PSC-derived hemogenic progenitors (Dou et al., 2016; Doulatov et al., 2013; McKinney-Freeman et al., 2012; Ng et al., 2016; Salvagiotto et al., 2008; Sugimura et al., 2017). It is well-established that NOTCH signaling is essential for arterial specification and development of HSCs (Burns et al., 2005; Kumano et al., 2003). Knock-out of the HOXA cluster in adult mice severely compromised HSC activity (Lebert-Ghali et al., 2016). In humans, HOXA5 and HOXA7 were shown to be critical for the expansion of engraftable fetal liver HSCs (Dou et al., 2016). Although overexpression of single or multiple medial HOXA genes in PSC-derived CD34+ cells was insufficient to confer HSC function (Dou et al., 2016; Ramos-Mejia et al., 2014), overexpression of HOXA5, HOXA9, and HOXA10 along with ERG, LCOR, RUNX1 and SPI1 hPSC-derived HE was capable of generating engraftable hematopoietic cells (Sugimura et al., 2017). In the present study, we provided a compelling evidence that SOX17 is a master regulator that integrates HOXA and arterial signature in HE through modulation of CDX2 signaling. This important finding may contribute to the strategic targeting of NOTCH and HOXA pathways to enhance lymphoid and engraftable hematopoietic cell production from hPSCs for the therapies of hematologic and oncologic diseases, including off-the-shelf immunotherapies.
SOX17 expands lymphomyeloid progenitors generated from day 4 hemogenic endothelium.
To evaluate the role of SOX17 in expansion of hematopoietic progenitors, we isolated day 4 HE from DOX+ and DOX− cultures (phase I DOX treatment) and cultured these cells on OP9 or OP9-DLL4. After 5 days, floating hematopoietic progenitors were collected and cultured with or without DOX for additional 5 days (phase II DOX treatment) and evaluated for CFC and T lymphoid potential (
Park, M. A., Kumar, A., Jung, H. S., Uenishi, G., Moskvin, O. V., Thomson, J. A., and Slukvin, II (2018b). Activation of the Arterial Program Drives Development of Definitive Hemogenic Endothelium with Lymphoid Potential. Cell Rep 23, 2467-2481.
Solaimani Kartalaei, P., Yamada-Inagawa, T., Vink, C. S., de Pater, E., van der Linden, R., Marks-Bluth, J., van der Sloot, A., van den Hout, M., Yokomizo, T., van Schaick-Solerno, M. L., et al. (2015). Whole-transcriptome analysis of endothelial to hematopoietic stem cell transition reveals a requirement for Gpr56 in HSC generation. J Exp Med 212, 93-106.
This application claims priority to U.S. Provisional Application No. 63/026,494 filed on May 18, 2020, the contents of which are incorporated by reference in their entireties.
This invention was made with government support under HL142665, OD011106, CA014520, and HL134655 awarded by the National Institutes of Health. The government has certain rights in the invention.
| Number | Date | Country | |
|---|---|---|---|
| 63026494 | May 2020 | US |
