HAEMATOPOIETIC STEM/PROGENITOR CELLS
The invention relates to an isolated cell having, after culturing a pluripotent stem cell (PSC) in a medium comprising a WNT agonist and an ACTIVIN antagonist, increased HOXA gene expression relative to a PSC not cultured in a medium comprising a WNT agonist and an ACTIVIN antagonist, wherein the cell with increased HOXA gene expression is capable of generating a definitive haematopoietic stem/precursor cell. The invention also relates to use of the cell for generating a definitive haematopoietic stem/precursor cell, and a method for differentiating a PSC into a definitive haematopoietic stem/precursor cell and a definitive haematopoietic stem/progenitor cell differentiated from a PSC by the method. The invention further relates to a therapeutic composition comprising a cell of the invention, and to therapeutic methods and uses of a cell of the invention. The invention also relates to a reporter cell comprising distinguishable SOX 7 and RUNXIC reporters, for use in tracking the differentiation of a PSC into a definitive haematopoietic stem/progenitor cell.
The invention relates to differentiating a pluripotent stem cell (PSC) into a definitive haematopoietic stem/progenitor cell, and to therapeutic uses of the haematopoietic stem/progenitor cell.
BACKGROUNDHematopoietic stem cell (HSC) transplantation permits the reconstitution of the blood cell compartment, for example in patients receiving myeloablative therapy. Because many patients do not have an optimal matched donor, the provision of HSCs from differentiated PSCs would be beneficial. Despite considerable efforts, in vitro haematopoietic differentiation remains skewed towards embryonic, yolk-sac like blood that lacks re-population activity, i.e. primitive haematopoiesis.
It follows that production of definitive haematopoietic stem/progenitor cells from PSCs, in particular human PSCs (hPSCs), remains a significant challenge, and identification of differentiation conditions that replicate early definitive mammalian blood cell development is needed.
It is to be understood that if any prior art publication is referred to herein, such reference does not constitute an admission that the publication forms a part of the common general knowledge in the art in Australia or any other country.
SUMMARYRUNX1 and SOX17 encode transcription factors critical to definitive haematopoiesis. SOX17 expression is a marker of haemogenic endothelium, whereas one isoform of RUNX1, RUNX1C is a marker of haematopoietic stem/progenitor cells. In view of this, the inventors produced PSCs comprising a SOX17 reporter and/or a RUNX1C reporter and used these PSCs to investigate definitive haematopoiesis.
Investigation of homozygous and heterozygous RUNX1C reporter PSCs together with repopulation-competent PSCs from cord blood revealed the surprising finding that HOXA genes were not expressed in the RUNX1C reporter PSCs.
The inventors then set about increasing expression of HOXA genes in RUNX1C/SOX17 dual reporter PSCs using a combination of WNT agonist and ACTIVIN antagonist, and established a robust in vitro differentiation system that encompasses the developmental period from mesoderm formation to AGM-like definitive haematopoiesis, providing a platform for definitive haematopoiesis and the generation of repopulating HSCs.
Based on the inventors' findings, a first aspect provides an isolated cell having, after culturing a pluripotent stem cell (PSC) in a medium comprising a WNT agonist and an ACTIVIN antagonist, increased HOXA gene expression relative to a PSC not cultured in a medium comprising a WNT agonist and an ACTIVIN antagonist, wherein the cell with increased HOXA gene expression is capable of generating a definitive haematopoietic stem/precursor cell.
A second aspect provides an isolated cell, comprising a SOX17 reporter and a RUNX1C reporter, wherein the SOX17 reporter and the RUNX1C reporter are distinguishable.
A third aspect provides use of the cell of the first aspect for generating a definitive haematopoietic stem/precursor cell.
A fourth aspect provides use of the cell of the second aspect for tracking differentiation of a PSC into a definitive haematopoietic stem/precursor cell.
A fifth aspect provides a method for differentiating a PSC into a definitive haematopoietic stem/precursor cell, the method comprising increasing HOXA gene expression by culturing the PSC in medium comprising a WNT agonist and an ACTIVIN antagonist, wherein increased HOXA gene expression is relative to a PSC not cultured in medium comprising a WNT agonist and an ACTIVIN antagonist.
A sixth aspect provides a definitive haematopoietic stem/progenitor cell differentiated from a PSC by the method of the fifth aspect.
A seventh aspect provides a therapeutic composition comprising the cell of the first or sixth aspect.
An eighth aspect provides an infusion bag comprising the cell of the first or sixth aspect or the therapeutic composition of the seventh aspect.
A ninth aspect provides a method for treating a condition, disease or disorder requiring HSC transplantation, the method comprising administering to a subject the cell of the first or sixth aspect or the therapeutic composition of the seventh aspect.
Alternatively or additionally to the ninth aspect, use of the cell of the first or sixth aspect in the manufacture of a medicament for treating a condition, disease or disorder requiring HSC transplantation is provided.
Alternatively or additionally, the cell of the first or sixth aspect or the therapeutic composition of the seventh aspect for use in a method for treating a condition, disease or disorder requiring HSC transplantation, the method comprising administering to a subject the cell or therapeutic composition is provided.
(a) Schematic representation of the RUNX1 locus, the targeting vector and the targeted allele with GFP sequences inserted into exon 1(e1). The distal (D) and proximal (P) promoters are shown. Exons (e) are numbered with non-coding exons boxed in white, coding exons in black and exons encoding the runt homology domain in red. e7A and e7B are alternately spliced exons that give rise to RUNX1A and RUNX1B transcripts. RES indicates a cassette encoding either G418 (Neo) or hygromycin resistance flanked by loxP sites (black triangles). The positions of primers used to identify homologously integrated alleles are shown (p1-p7), with the sizes of PCR products indicated. p3 identifies the hygromycin resistance gene whilst p7 is specific for the G418 resistance gene (b) A representative example of the PCR screen to detect homologous integration of the left and right homology arms with the hygromycin cassette containing vector. ‘+’ refers to a clone where the targeting construct was homologously integrated into the RUNX1 locus, and ‘−’ indicates a randomly integrated clone. The combinations of p1/p2 and p3/p4 identify correctly integrated left and right homology arms of the vector. p1/p5 identifies sequences of a wild type allele and p3/p6 identifies vector sequences present even in random integrations. Sequences for each primer are indicated in Table 1. (c) Homologous integration into the RUNX1 locus was further verified by cloning and sequencing the integration sites. An example using primers p1/p2 and p3/p4 is shown. Junction sequences from the genomic locus are shown in lower case. (d) RUNX1cGFP/w EBs harvested on day 13, 14 or 17 were flow-sorted on the basis of GFP and CD34 expression. cDNA synthesized from the sorted samples was analysed by RT-PCR using primers that detected both RUNX1B and RUNX1C (RUNX1B/C) or RUNX1C alone. Note that RUNX1B/C primers revealed expression in GFP−CD34+ cells, consistent with expression of the RUNX1B isoform in a subset of CD34+ endothelial cells. Gene expression is shown relative to GAPDH. UN, unfractionated cells. (e) Schematic outlining the strategy for generation of the RUNX1CGFP/GFP cell line by re-targeting the wild type RUNX1C allele with the hygromycin targeting vector. (f) Schematic outlining the PCR based screening strategy to identify doubly targeted cell lines, prior to excision of the targeting cassette. (g) PCR analysis using primers shown in (e) to identify GFPHygro, GFP and wild type (w) alleles. Clone 180 depicts an example where the GFPHygro targeting vector has replaced the previous GFP allele, leaving one wild type allele, clone 22 represents depicts one previously targeted GFP allele and one wild type allele and clone 253 represents a clone in which GFPHygro has correctly integrated into the wild type allele resulting in both RUNX1C alleles being targeted. (h) PCR analysis of sorted fractions of RUNX1CGFP/w and RUNX1CGFP/GFP differentiated cells confirming the absence of RUNX1C transcripts from the RUNX1CGFP/GFP line without disruption of the RUNX1B transcripts. Cells were sorted at day 14 based on GFP (G) and CD34 expression. UN, unfractionated cells. (i) Schematic representation of the SOX17 locus, the targeting vector and the targeted allele with mCHERRY sequences inserted into exon 1(e1). The loxP flanked Neo selectable marker was excised with CRE recombinase. (j) Intracellular flow cytometry profiles of SOX17mCHERRY/w hESC differentiated for 5d to endoderm, showing specific staining of mCHERRY+ cells with FITC-conjugated SOX17 antibody.
(a-d) Expression of CD34, CD43 and CD41 precedes GFP expression in RUNX1CGFP/w EBs. Representative flow cytometry analyses of differentiating RUNX1CGFP/w EBs indicating that expression of CD34 at day 5 (a) followed by CD41 and CD43 at day 7 (b) antedated GFP expression. Correlation of GFP expression with haematopoietic cell surface markers at day 11 (c) and day 16 (d) revealed that all GFP cells initially co-expressed CD34, CD31, and most expressed CD41. At day 16, GFP expression remained confined to haematopoietic cells expressing the indicated cell surface markers. Red boxed areas represent CD34+CD31+CD41−CD43−CD33−CD45− endothelial cells and the blue boxes mark GYPA+ erythroid cells, both of which were GFP−. A high proportion of d16 GFP+ cells co-expressed VECAD and CD45, a combination of surface markers found in mouse haematopoietic stem/progenitor cells. (e-h) Flow cytometry analysis of differentiation timecourse comparing RUNX1CGFP/w heterozygous cells (clone 67.4) with RUNX1CGFP/GFP nullizygous clone A5 at (e) day 7, (f) day 11, (g) day 14 and (h) day 18. The percentage of GFP+ cells and CD34 and CD45 expression patterns were similar for all clones. (i) Strategy for flow cytometry sorting of d14 differentiating RUNX1CGFP/w and RUNX1CGFP/GFP cultures based on GFP and CD34 expression.
(a) Differentially expressed probe sets between the indicated fractions and GFP−CD34− cells for samples including both d14 RUNX1CGFP/w and RUNX1CGFP/GFP EBs. (b) Venn diagram showing overlap between probe sets up-regulated in each cell fraction in (a). (c) Hierarchical cluster analysis of the fractions sorted on the basis of CD34 and GFP expression, as well as undifferentiated hESCs isolated from the heterozygote RUNX1CGFP/w 67.4 line and the homozygous null RUNX1CGFP/GFP A5 line, based on the top 25% of probes with the greatest variance. (d) Scatter plot representations of the logarithmic signal intensities of the pairwise comparisons between the CD34 and GFP fractions as well as undifferentiated hESCs from the heterozygous and homozygous null lines. Numbers in the top left and bottom right corners indicate the number of probes that are outside of the 2-fold cut-off (indicated by red lines). The correlation co-efficient for each differentiated population is similar to the correlation co-efficient for the undifferentiated cells, indicating the similarity of the differentiated progeny. (e) Table indicating the Pearson correlation co-efficient for each of the crosswise comparisons. Note the high correlation co-efficients (r2≥0.96) between the equivalent fractions from heterogygous and homozygous RUNX1C cell lines and between the RUNX1C expressing populations (G+34− and G+34lo; r2≥20.94). (f) Heatmap showing HOX gene expression in sorted d14 RUNX1CGFP/w EBs and three samples of CB 34+ cells. (g) Heatmap depicting expression of HOX cluster genes during differentiation of RUNX1CGFP/w cells for 8 days. Note that the probes for HOXA7 and HOXB9 function poorly in this array, based on comparisons by RT-PCR and RNA-seq,
(a) Relative expression by RT-PCR of HOXA and CDX genes at d7 of differentiation in cultures supplemented with SB431542 (SB), CHIR99021 (CHIR) and the combination of these agents (SB/CHIR). Data is the mean±SEM for 3-4 experiments. *, p<0.05; **, p<0.01; ***, p<0.001 compared to control using 2 way ANOVA with Holm-Sidak's multiple comparisons test. Data for the d4 timepoint is shown in
(a) Control and SB/CHIR differentiation protocols. (b, c) Multidimensional scaling plots of RNA-Seq data from d10 control (Con) and SB/CHIR (SB/CH) cultures sorted into endothelium, blood and CD34 adherent cell fractions demonstrating the clustering of sorted fractions by immunophenotype (b) and by culture condition (c). (d) Log2 scale bar plot showing the expression by RNA-Seq of HOXA genes in SOX17+34+ and SOX17−34+ endothelial cells in control and SB/CHIR cultures (n=3, mean±SEM). Expression in SB/CHIR cultures was higher for asterisked (*) genes (p<0.05). See also HOXA expression in other sort fractions in
(a-e) Bar plots showing fold change (log2) in HOX gene expression in SB/CHIR versus control cultures for each sorted fraction. Red fill indicates the difference was statistically significant (p<0.05). Bars are arranged in order of increasing p value from top to bottom. (f, g) Bar plot showing the expression by RNA-Seq of HOXA genes in (f) CD43+ blood cells and (g) SOX17+ and SOX17− CD34− adherent cells in control and SB/CHIR cultures (n=3, mean±SEM). Expression in SB/CHIR cultures was higher for asterisked (*) genes (p<0.05). (h) Expression of genes by RNA-Seq analysis discriminating arterial and venous gene signatures in umbilical artery and vein endothelium as reported by Aranguren et al. Blood 122, 3982-3992 (2013). Genes shown in red type were selectively expressed in CD34+ sorted fractions as shown in
(a) Heatmap showing transcription factors and cell surface markers differentially expressed between d10 SOX17+ and SOX17− endothelial cells in SB/CHIR cultures. Expression in sorted haematopoietic cells (43+) is shown to demonstrate that genes increased in SOX17− endothelium include many haematopoietic regulators and surface markers. (b) Sorting strategy to determine the clonal frequency of haemogenic capacity of SB/CHIR endothelium, showing sequential gating on SOX17, CD34 and CD43, and CD73. SOX17 expression was stratified as bright (++), dull (+) and negative (−). Most CD73lo/− cells were also DLL4− and CXCR4−. (c) RNA-Seq and DNA methylation (ME) array, and chromatin accessibility assessed by ATAC-Seq, comparing the HOXA cluster in control and SB/CHIR CD34 endothelial fractions. In the DNA ME array, the blue and red lines represent the fraction of methylated probe at each locus in control and SB/CHIR cultures, respectively. The shading represents 95% confidence intervals for the DNA ME value. Scale represents counts per million for RNA-Seq and ATAC-Seq. DNA ME is scored for mean probe methylation from 0 (unmethylated) to 1 (methylated).
(a) Multidimensional scaling plots of DNA methylation array data from d10 control (C) and SB/CHIR (S/C) cultures sorted into SOX17− and SOX17− endothelium, and SOX17−34− adherent cell fractions demonstrating the clustering of sorted fractions by phenotype and by culture condition. The closer clustering of the control and SB/CHIR samples for the SOX17+CD34+ endothelium suggests more similar levels of methylation in these samples. (b) Heatmap of the 1002 methylation probes that encompassed the HOXA-D loci showing global hypomethylation in all samples. Hierarchical clustering indicates additional grouping of samples by treatment type.
(c) Multidimensional scaling plots of ATAC-Seq data from d10 control (C) and SB/CHIR (S/C) cultures sorted into SOX17+ and SOX17− endothelium, demonstrating the clustering of sorted fractions by phenotype and by culture condition. (d-f) RNA-Seq and DNA methylation (ME) array, and chromatin accessibility assessed by ATAC-Seq, comparing the (d) HOXB, (e) HOXC, and (f) HOXD clusters in control and SB/CHIR CD34+ endothelial fractions. In the DNA ME array, the blue and red lines represent the fraction of methylated probe at each locus in control and SB/CHIR cultures cells respectively.
(a, b) Strategy used to sort RUNX1CGFP/w cells differentiated for 7 days in spinner cultures under Control (a) and SB/CHIR (b) conditions into RUNX1C+34+, SOX17+34+, RUNX1C−43+ and SOX17−34+ fractions prior to co culture with OP9 DL4 stromal cells. (c) Bar plot comparing the frequency of cells in each fraction indicating that RUNX1C−43+ (R−43+) cells were less frequent in SB/CHIR cultures, consistent with suppression of primitive hematopoieisis, and that SOX17+34+ (S+34+) cells increased in response to SB/CHIR. Data represent mean±SEM from 4 experiments. *, p<0.05 Student's t test. (d) Day 7 SOX17+CD34+ and RUNX1C+CD34lo populations from control and SB/CHIR cultures after 5d on OP9 DL4 showing emerging haematopoietic cells. Scale bar, 50 μM (e) Image of SOX17−CD34+ sorted cells from SB/CHIR cultures after 9 days on OP9 DL4 showing the generation of SOX17+ cells and emergence of haematopoietic cells. Scale bar, 50 μM. (f, g) Representative flow cytometry profiles of control (f) and SB/CHIR (g) d7 sorted fractions after 21 days co-culture on OP9 DL4. Live cells were gated for CD45, then for the NK cell marker CD56 (left panels). T cell progenitors (CD45+CD56−CD7+CD5±) are shown in red and NK cells (CD45+CD56hiCD7±) are shown in blue. Whilst T cells and NK cells were generated from all sorted fractions, the yield appeared greater from the SB/CHIR treated cultures. (h, i) Bar plot showing the percentages of cells falling into the CD7 and CD5 quadrants shown in (e, f) for the CD56 T cell progenitors (h) and the CD56hi NK cells (i). (j) Generation of CD4+CD8+ T cells following 27 days co-culture on OP9 DL4 of SOX17−CD34+ cells from SB/CHIR treated cultures.
(a) Day 11 EBs from control and SB/CHIR cultures showing developing SOX17+ vascular structures within and growing out from the EB, contrasting with the lack of identifiable vessels and the early production of blood cells in the control EBs. Boxed areas in left panels are magnified on the right. Scale Bar, 100 μM. (b) Tiled images of d15 EBs from control and SB/CHIR cultures showing multiple SOX17+ vascular structures emanating from the SB/CHIR EBs. Arrows indicate two of the vascular structures. Scale bar, 500 μM. (c-f) Images of a d20 SB/CHIR EB in which some SOX17+ vascular segments enclose RUNX1C+ haematopoietic cells. Panels (d,e,f) show higher power views of boxed areas (d′,e′,f′). Scale bars, (c) 300 μM, (d-f) 100 μM. (g) Image of boxed area (g′) showing RUNX1C+ cells enclosed by SOX17+ endothelium (h) Image of boxed area (h′) showing RUNX1C+ and RUNX1C cells within a SOX17+ vascular structure. Scale bar, (g,h) 50 μM. (i) increased viability of haematopoietic cells in d23-d29 cultures of SB/CHIR cells (n=6, mean±SEM; *, p<0.05 Student's t test). (j) Brightfield and RUNX1C+ overlaid images of control and SB/CHIR cultures at d18 showing dying cells (arrows) in the control cultures. Scale bar, 50 μM. (k) Flow cytometry plots showing increased frequency of haematopoietic genes expressing the indicated stem cell markers at d23 in SB/CHIR cultures. (1) Bar plots showing the increased percentage of haematopoietic cells expressing CD34, GPI80, FCER1A and ACE in d23-29 SB/CHIR cultures (n=4, mean±SEM; *, p<0.05 Student's t test).
(a-f) Multifocal erythroid burst colonies (a,b), mixed erythroid myeloid colony (c), myeloid colony (d), and erythroid colonies apposed to stromal cysts in MATRIGEL™ supplemented methycellulose (e, f). Panel (a) was a colony from a d23 culture, the remaining panels came from d31 colonies. Scale bar, 100 μM. (g) Proportion of myeloid and erythroid/mixed colony types in methylcellulose in SB/CHIR treated cultures assayed at the day of differentiation indicated. Data represent results from 7 independent experiments with two of the experiments assayed at both earlier and later time points. (h) Bar plot showing the frequency of CFCs in control and SB/CHIR (S/C) cultures. Data is the mean±SEM for 5 experiments (d23-33), p<0.01 for S/C versus control using Student's t test. (i) Log scale bar graph showing the relative gene expression of β globin genes in SB/CHIR colonies compared with primitive erythroid colonies (PRIM). Data is the mean±SEM for 6 SB/CHIR and 14 primitive erythroid samples. *, p<0.005 using Student's t test. (j) Log scale bar graph showing the ratio of γ/ε and γ/β globin expression in the same data set as (i). ***, p<0.001 using Student's t test. (k) Log scale bar graph of the relative gene expression of HOXA genes in control (CON), SB/CHIR cultures (S/C), erythroid colonies (COL) and d10 differentiated cord blood CD34+ cells (CB) Data is the mean±SEM for 3 experiments. *, p<0.05 versus CON using Student's t test (l) Log scale bar graph of the ratio of γ/ζ, γ/ε and γ/β globin chain expression in the same experiments as (k). *, p<0.01 for γ/ε □□□□ S/C versus CON and CB versus S/C, for α/ζ in COL versus S/C and for γβ in CB versus CON, using Student's t test. (m) Log scale bar graph of the relative gene expression of BCL11A and KLF1 in the same experiments as (k) *, p<0.05 versus CON using Student's t test. (n) Summary of the experiments in panels (k-m), showing that HOXA expression induced by SB/CHIR translates into persistent clonogenic cells in 3-4 week cultures. These late colonies have increased BCL11A and KLF1 expression and down-regulated embryonic EPSILON and ZETA globin genes, similar to an early foetal liver stage of development.
(a) Confocal images of a d18 SB/CHIR culture showing a vessel containing a cell co-expressing SOX17 and RUNX1C. (b) Gating strategy showing the isolation of SOX+RUNX−, SOX+RUNX+ and SOX−RUNX+ populations that were further stratified by CD34 and KIT expression. (c) Bar plot of the percentage of cells in each indicated fraction across a developmental time course from d14 to d26 of differentiation, showing reciprocal changes in proportions of SOX+RUNX− and SOX−RUNX+ populations. (d) Clonogenic frequency at each time point of populations sorted on SOX17 (S), RUNX1C (R), CD34 (34), and KIT (K). The left panel shows the total number of myeloid and erythroid colonies and the right panel shows only the erythroid-containing colony frequency. The inset shows a multifocal mixed erythroid/myeloid colony from a d26 S+R+34+K+ culture. Scale bar, 100 μM. These data represent a series of 4 independent experiments.
Sorting strategy for generating (a) SOX17mCHERRY/wRUNX1CGFP/w d18 endothelial (S+R−34+90+) (d18 S+R− En), stem/progenitor (S+R+34+90+) (d18 S+R+S/P) and progenitor (S−R+34+90lo) (d18 S−R+ Pr1) populations, (b) SOX17mCHERRY/w RUN1CGFP/w d22 progenitor (S+R+34+K+) (d22 S+R+ Pr2) and (SR+34+K+) (d22 S−R+ Pr3) populations, (c) 5 week AGM endothelium (34+90+43−)(En), cells transiting from haemogenic endothelium to haematopoietic stem/progenitors (34+90+43+) (S/P) and committed progenitor cell (34+90−43+) (Pr1) populations and (d) 17 week foetal liver (FL) sample (CD43+CD45+) enriched for haematopoietic stem/progenitor cells (34+38lo/−90+) (S/P). For panels (a-d), the percentage of live cells sorted for each fraction is shown in red type. The RUNX1C+ fractions in panels (a, b) also expressed CD43 or CD45. (e) Expression of HOX genes by RNA-Seq in FL, AGM, and d18 and d22 SOX/RUNX fractions sorted as shown in panels (a-d). (f) Expression of the list of 32 genes differentially expressed between cord blood and d14 RUNX1C+34+ cells (see
RNA-seq profiling comparing human foetal liver (FL), human AGM and SOX/RUNX (S/R) populations sorted at d18 and d22. ‘S/P’ refers to populations with a stem/progenitor phenotype, ‘En’ marks endothelium and ‘Pr1-3’ refer to committed progenitor populations with an increasingly mature phenotype. See text for further details. Heatmaps show (a) cell surface markers, (b) transcription factors, (c) NOTCH, (d) BMP/TGFβ (e) WNT, (f) retinoic acid signaling pathway genes and (g) globin genes. Red shaded gene groups are expressed in FL S/P cells, and blue shaded genes are selectively expressed in the AGM S/P fraction. Scale is in log2 RPKM. (h) Bar graph showing expression of HOXA genes in FL, AGM and d18 S+R+ S/P populations. These data represent the results of a single flow sorting experiment. (i) Summary of the developmental stages, culture format, growth factors and the haematopoietic differentiation pathway during the transition from hPSC to AGM-like stem/progenitor cells. Thumbnail images illustrate the developmental stages.
Repopulating HSCs have an origin distinct from the yolk sac, and are first located in the aorta-gonad-mesonephros (AGM) and placenta. Cell clusters budding from the ventral aortic endothelium in the AGM harbour the earliest HSCs and express the transcription factor RUNX1, without which clusters do not form and foetal liver haematopoiesis fails.
Three isoforms of RUNX1 (RUNX1A-C) are transcribed from two promoters in humans and mice. In mouse embryos, Runx1b marks haemogenic endothelium, whilst Runx1c is expressed in haematopoietic progenitor cells and is the dominant isoform in foetal liver haematopoietic cells. Runx1b is required in AGM endothelium for definitive haematopoiesis and Runx1b heterozygous mice display defects in HSC and progenitor generation. Conversely, loss of both Runx1c alleles only modestly impacts haematopoiesis and a phenotype for haploinsufficient Runx1c has not been reported.
SOX17 is a second critical transcription factor required for early definitive haematopoiesis in the mouse. In haematovascular development, SOX17 is largely expressed in the arterial system of the embryo. It marks haemogenic endothelium in the AGM and is required for the generation of nascent HSCs from the AGM and for their maintenance in the foetal liver. As with RUNX1C, a haploinsufficient phenotype was not reported for SOX17.
Based on these key roles in haemogenic endothelium and in identifying haematopoietic progenitors, the inventors generated reporter cell lines for SOX17 and/or RUNX1C. In initial studies, RUNX1C marked a subset of CD34+ cells highly enriched for haematopoietic progenitors that homed to the bone marrow, but did not engraft. Exploring molecular differences between PSC-derived and repopulation-competent cord blood (CB) CD34+ cells, the inventors revealed that the RUNX1C+CD34+ cells failed to express HOXA genes, indicating incorrect mesoderm patterning.
Consequently, the inventors have shown that simultaneous modulation of WNT and ACTIVIN signaling yields haematopoietic stem/progenitor cells with HOXA codes that more closely resemble those of cord blood. These cultures generate a network of aorta-like SOX17+ vessels, from which RUNX1C+ blood cells emerge-reminiscent of definitive haemogenesis within the AGM. Transcriptional profiles of nascent haematopoietic progenitors and corresponding cells sorted from human AGM displayed striking similarity in expression of cell surface receptors, signaling molecules and transcription factors. The inventors demonstrate that HOXA codes are an indicator of mesoderm patterning, thereby enabling generation of definitive haematopoietic lineages from PSCs.
Using HOXA gene expression of human CB progenitor cells as a guide, the inventors devised a differentiation protocol that generates AGM-like structures from hPSCs (
The inventors were struck by the low levels of HOXA expression in hPSC-derived vascular and haematopoietic progenitors derived under control culture conditions. Given that HOXA5, HOXA7, HOXA9 and HOXA10 are expressed in HSCs, the inventors hypothesized that acquisition of a HOXA ‘signature’ during hPSC differentiation would underpin the capacity to eventually generate repopulating cells.
Because HOX genes are first expressed in the primitive streak, the inventors trialed factors that influenced mesodermal patterning. These experiments led to the finding that 2 days of culture with the combination of SB and CHIR up-regulated CDX and HOXA gene expression to a greater extent than either agent alone. The inventors found that 2 days was important, as no or minimal effect was observed at 24 hours. The inventors also observed that the synergistic effect of SB/CHIR was unexpected owing to the fundamental unpredictability in the art.
Chromatin accessibility was increased following SB/CHIR and, in the context of hypomethylation of the HOX clusters, indicated that a permissive environment was generated for transcription factor-mediated HOXA gene expression.
A striking consequence of SB/CHIR patterning was the emergence of SOX17+ vessels from day 10 (d10) of differentiation. The expression of aorta-associated genes and the appearance of haematopoietic cells from these vessels resembled AGM development. The restriction of haematopoiesis to discrete regions shows that blood cells only arose from a subset of the SOX17+34+ endothelium, consistent with reports that haemogenic endothelium displays a distinct phenotype. Indeed, these studies identified that the d10 SOX17+ haemogenic precursors were restricted to the small CD73-negative fraction and that the frequency was greatly increased in cells with lower SOX17 expression.
In previous studies, deletion of Hoxa9 or the whole Hoxa cluster in mesoderm led to significant haematopoietic defects. Additionally, knockdown of HOXA5 or HOXA7 in human foetal liver depleted cells with an HSC-like phenotype and engraftment ability, confirming the importance of HOXA genes in human definitive haematopoiesis. Despite this, studies aiming to generate HSCs through overexpression of HOXA genes have not been successful. A cocktail of genes, of which HOXA9 was one component, was insufficient to generate long term engraftable cells from hPSC-derived CD34+ cells in one study. In a second study, enforced HOXA9 expression in hESCs enhanced haematopoiesis but did not generate HSCs. In the HOXA5 or HOXA7 knockdown study above, it was concluded that even overexpression of multiple HOXA genes did not confer HSC properties on hPSC derived haematopoietic progenitors. These results show that the ability of HOXA genes to orchestrate a functional HSC phenotype requires the correct cellular context.
In contrast to control cultures, SB/CHIR cultures generated late progenitors with globin chain synthesis indicative of a switch to a foetal developmental stage. Erythroid differentiated hPSCs usually express EPSILON and GAMMA globin, although prolonged culture in human plasma or on stromal layers has yielded predominantly GAMMA or GAMMA and BETA globin expressing cells. A switch from EPSILON to GAMMA globin exclusively in SB/CHIR cultures, associated with the increased expression of KLF1 and its target gene BCL11A was identified. In addition, ZETA globin down-regulation was observed later, in the progeny of colony forming cells from SB/CHIR cultures, indicating that distinct factors regulate silencing at the ALPHA globin locus (
Molecular analysis of human AGM has been limited to date, and the present study represents the first example of transcriptional profiling of sorted fractions from this tissue, with comparisons to human foetal liver and AGM-like populations derived from hPSCs. It has been shown elsewhere that AGM-derived HSCs exclusively derived from the ventral aorta, possessed a CD34+CDH5+CD45+KIT+THY1+ ENG+RUNX1+CD38−/loCD45RA− phenotype. It is noteworthy that the AGM S/P and d18 S+R+ S/P fractions match this phenotype (
One laboratory transcriptionally profiled endothelial and haematopoietic populations from day 10.5-11 AGM. The laboratory identified transcription of the ‘heptad’ transcription factors (Tal1, Lyl1, Erg, Fli1, Runx1, Gata2 and Lmo2), in addition to Cdh5, Tek, Esam, Kdr and Eng in haemogenic endothelium and in the HSC compartment. They observed that many additional transcription factors influencing HSC development were selectively expressed in their HSC compartment. The majority of the genes they described were enriched in both the AGM and d18 S+R+ En and S/P fractions (
These transcriptional data of human primary AGM and hPSC-derived samples show a high degree of concordance with existing human and mouse data, supporting the inventors' conclusion that the hPSC-derived samples represent definitive, human AGM haematopoietic cells, and that there is substantial conservation of definitive haematopoietic development between the species.
In summary, the in vitro differentiation of human PSCs that are ‘switched’ to definitive haematopoiesis through exposure to SB/CHIR, timed to overlap with the peak expression of primitive streak genes is described herein. Analysis of these cultures revealed the early and sustained expression of HOXA genes, followed by the development of SOX17+ vascular structures from which haematopoietic cells then emerged, resembling the developing mammalian AGM. This is the first report of a culture system that encompasses human haematopoietic development from mesoderm specification to an AGM-like stage and provides robust generation of transplantable human haematopoietic stem cells from pluripotent stem cells.
Unless defined otherwise in this specification, technical and scientific terms used herein have the same meaning as commonly understood by the person skilled in the art to which this invention belongs and by reference to published texts.
It is to be noted that the term “a” or “an” refers to one or more, for example, “a molecule,” is understood to represent one or more molecules. As such, the terms “a” or “an”, “one or more,” and “at least one” may be used interchangeably herein.
In the claims which follow and in the description of the invention, except where the context requires otherwise due to express language or necessary implication, the word “comprise” or variations such as “comprises” or “comprising” is used in an inclusive sense, i.e. to specify the presence of the stated features but not to preclude the presence or addition of further features in various embodiments of the invention.
The term “about” as used herein contemplates a range of values for a given number of ±25% the magnitude of that number. In other embodiments, the term “about” contemplates a range of values for a given number of ±20%, ±15%, ±10%, or ±5% the magnitude of that number. For example, in one embodiment, “about 3 μM” indicates a value of 2.7 to 3.3 μM (i.e. 3 μM±10%), and the like.
Similarly, while differentiation processes include ordered, sequential events, the timing of the events may be varied by at least 25%. For example, while a particular step may be disclosed in one embodiment as lasting one day, the event may last for more or less than one day. For example, “one day” may include a period of about 18 to about 30 hours. In other embodiments, periods of time may vary by ±20%, ±15%, ±10%, or ±5% of that period of time. Periods of time indicated that are multiple day periods may be multiples of “one day,” such as, for example, two days may span a period of about 36 to about 60 hours, and the like. In another embodiment, time variation may be lessened, for example, where day 2 is 48±3 hours from day 0; day 4 is 96±3 hours from day 0, and day 5 is 120 hours±3 hours from day 0.
As used herein, a “WNT agonist” is a substance that mimics or increases WNT signaling. A WNT agonist is not to be restricted to a substance acting directly on WNT as the substance may act elsewhere in the WNT signaling pathway.
Non-limiting examples of WNT agonists according to the first or fifth aspects of the invention include small molecules CHIR99021 (CAS 252917-06-9), a 2-amino-4,6-disubstituted pyrimidine, e.g. BML 284 (CAS 853220-52-7), SKL 2001 (CAS 909089-13-0), WAY 262611 (CAS 1123231-07-1), WAY 316606 (CAS 915759-45-4), SB 216763 (CAS 280744-09-4), IQ 1 (CAS 331001-62-8), QS 11 (CAS 944328-88-5), deoxycholic acid (CAS 83-44-3), BIO (CAS 667463-62-9), kenpaullone (CAS 142273-20-9), or a (hetero)arylpyrimidine. A WNT agonist may also be an agonist antibody or functional fragment thereof or an antibody-like polypeptide.
In one embodiment of the first or fifth aspects of the invention, the WNT agonist is CHIR99021 ((CHIR) CAS 252917-06-9), and the PSC is cultured in a medium comprising about 3 μM. In other embodiments, the medium may comprise about 2 μM, about 2.1 μM, about 2.2 μM, about 2.3 μM, about 2.4 μM, about 2.5 μM, about 2.6 μM, about 2.7 μM, about 2.8 μM, about 2.9 μM, about 3 μM, about 3.1 μM, about 3.2 μM, about 3.3 μM, about 3.4 μM, about 3.5 μM, about 3.6 μM, about 3.7 μM, about 3.8 μM, about 3.9 μM, about 4 μM, about 4.1 μM, about 4.2 μM, about 4.3 μM, about 4.4 μM, about 4.5 μM, about 4.6 μM, about 4.7 μM, about 4.8 μM, about 4.9 μM, or about 5 μM CHIR.
Lithium chloride (LiCl) activates WNT signaling and is included in differentiation medium to improve mesoderm induction during the first 48 hours of differentiation, which is understood in the art. The concentration of LiCl in the differentiation medium may be about 0.5 mM, about 0.6 mM, about 0.7 mM, about 0.8 mM, about 0.9 mM, about 1 mM, about 1.1 mM, about 1.2 mM, about 1.3 mM, about 1.4 mM, about 1.5 mM, about 1.6 mM, about 1.7 mM, about 1.8 mM, about 1.9 mM, about 2 mM, about 2.1 mM, about 2.2 mM, about 2.3 mM, about 2.4 mM, or about 2.5 mM. Preferably, as understood in the art, concentration of LiCl in the medium may be about 2 mM.
As used herein, an “ACTIVIN antagonist” is a substance that inhibits or decreases ACTIVIN signaling. An ACTIVIN antagonist is not to be restricted to a substance acting directly on ACTIVIN as the substance may act elsewhere in the ACTIVIN signaling pathway.
Non-limiting examples of ACTIVIN antagonists according to the first or fifth aspects of the invention include small molecules SB 431542 (CAS 301836-41-9), SB 505124 (CAS 694433-59-5), LDN 193189 (CAS 1062368-24-4), LDN 193719 (CAS 1062368-49-3), Dorsomorphin (CAS 866405-64-3), A 83-01 (CAS 909910-43-6), DMH 1 (CAS 1206711-16-1), RepSox (CAS 446859-33-2), or LY 364947 (CAS 396129-53-6). An ACTIVIN antagonist may also be an antagonist antibody or functional fragment thereof or an antibody-like polypeptide.
In one embodiment of the first or fifth aspects of the invention, the ACTIVIN antagonist is SB 431542 ((SB) CAS 301836-41-9), and the PSC is cultured in a medium comprising about 3-4 μM SB. In other embodiments, the medium may comprise about 2 μM, about 2.1 μM, about 2.2 μM, about 2.3 μM, about 2.4 μM, about 2.5 μM, about 2.6 μM, about 2.7 μM, about 2.8 μM, about 2.9 μM, about 3 μM, about 3.1 μM, about 3.2 μM, about 3.3 μM, about 3.4 μM, about 3.5 μM, about 3.6 μM, about 3.7 μM, about 3.8 μM, about 3.9 μM, about 4 μM, about 4.1 μM, about 4.2 μM, about 4.3 μM, about 4.4 μM, about 4.5 μM, about 4.6 μM, about 4.7 μM, about 4.8 μM, about 4.9 μM or about 5 μM SB.
In some embodiments of the first or fifth aspect, the medium may further comprise a retinoid, e.g. retinol or all trans retinoic acid, or a retinoic acid analogue, e.g. AM580 or EC23. The retinoid or retinoic acid analogue may be used in culturing at concentrations between 1×10−9 M and 1×10−4 M, e.g. 1×10−9 M, 0.5×10−8 M, 1×10−8 M, 0.5×10−7 M, 1×10−7 M, 0.5×10−6 M, 1×10−6 M, 0.5×10−5 M, 1×10−5 M, 0.5×10−4 M, or 1×10−4 M for a period ranging from one day to one week during differentiation. The retinoid or retinoic acid analogue may be added at or about d1, d2, d3, d4, d5, d6, or d7, d8, d9, d10, d11, or at later time points of differentiation. The retinoid or retinoic acid analogue may be added at or about d4 of differentiation, at or about d7 of differentiation or at or about d12 of differentiation. The retinoid or retinoic acid analogue may be added at more than one differentiation time point.
Determining an effective concentration of a WNT agonist other than SB and/or determining an effective concentration of an ACTIVIN antagonist
As used herein, “pluripotent stem cell” or “PSC” refers to a cell that has the ability to reproduce itself indefinitely, and to differentiate into any other cell type. There are two main types of pluripotent stem cell: embryonic stem cells (ESCs) and induced pluripotent stem cells (iPSCs).
As used herein, “embryonic stem cell” or “ESC” refers to a cell isolated from a five to seven day-old embryo donated with consent by subjects who have completed in vitro fertilisation therapy, and have surplus embryos. The use of ESCs has been hindered to some extent by ethical concerns about the extraction of cells from human embryos.
Suitable ESCs for use in the first or fifth aspects of the invention include, but are not limited to, H1 and H9 human ESCs.
As used herein, “induced pluripotent stem cell” or “iPSC” refers to an ESC-like cell derived from adult cells. iPSCs have very similar characteristics to ESCs, but avoid the ethical concerns associated with ESCs, since iPSCs are not derived from embryos. Instead, iPSCs are typically derived from fully differentiated adult cells that have been “reprogrammed” back into a pluripotent state.
Suitable human iPSCs for use in the first or fifth aspects of the invention include, but are not limited to, iPSC 19-9-7T, MIRJT6i-mND1-4 and MIRJT7i-mND2-0 derived from fibroblasts and iPSC BM119-9 derived from bone marrow mononuclear cells. Other suitable iPSCs may be obtained from Cellular Dynamics International of Madison, Wis., USA.
As used herein, “definitive haematopoietic stem/precursor cell” refers to a cell or population of cells responsible for producing all mature blood cells throughout the lifespan of an organism. Clinically, they are important for transplantation in blood-related diseases, particularly in subjects undergoing myeloablative therapy. Definitive haematopoiesis is distinguished from primitive haematopoiesis that occurs transiently during early development and arises from the yolk sac.
A definitive haematopoietic stem/precursor cell of the first aspect of the invention or produced by the method of the fifth aspect of the invention may be further characterized by gene and protein expression as detailed in the examples and figures.
As used herein, “embryoid body” and “EB” refers to a three-dimensional aggregate of PSCs. Advantageously, EBs in may be cultured in suspension, thus making EB cultures scalable for clinical applications. Additionally, the three-dimensional structure of EBs, including the establishment of complex cell adhesions and paracrine signaling within the EB microenvironment, enables differentiation and morphogenesis which yields microtissues that are similar to native tissue structures, for example the AGM as disclosed herein.
In one embodiment of the first or fifth aspects of the invention, the PSCs of the present disclosure are cultured as an EB.
Methods for culturing EBs are known in the art. A specific example of a spin EB method that may be used in the present invention is described in Ng et al. Nature Protocols 3, 768-776 (2008).
As used herein, “increased HOXA gene expression” refers to a PSC expressing one or more genes from the HOXA cluster at a greater level as a result one or more specific culture conditions relative to expression of one or more genes from the HOXA cluster by a reference PSC cultured in the absence of the specific culture conditions. The specific culture conditions encompassed by the present invention are culture of the PSC in the presence of a WNT agonist and an ACTIVIN antagonist.
Gene expression may be measured by a number of means known in the art. For example, gene expression may be measured quantitatively by real time RT-PCR, or gene expression may be measured qualitatively by microarray.
Increased HOXA gene expression may be binary, either “on” or “off”. Alternatively, increased HOXA gene expression may be about: a 1% increase, about a 2% increase, about a 3% increase, about a 4% increase, about a 5% increase, about a 6% increase, about a 7% increase, about a 8% increase, about a 9% increase, about a 10% increase, about a 20% increase, about a 30% increase, about a 40% increase, about a 50% increase, about a 60% increase, about a 70% increase, about a 800 increase, about a 90% increase, about a 100, or greater increase. Alternatively, increased HOXA gene expression may be about a 2-fold, about a 3-fold, about a 4-fold, about a 5-fold, about a 6-fold, about: a 7-fold, about an 8-fold, about a 9-fold, about a 10-fold, or more increase.
In some embodiments of the first or fifth aspects of the invention, the HOXA gene is one or more of HOXA1, HOXA2, HOXA3, HOXA4, HOXA5, HOXA6, HOXA7, HOXA8, HOXA9, HOXA10, HOXA11, and HOXA13. In some embodiments of the first or fifth aspects of the invention, the HOXA gene is one or more of HOXA3, HOXA5 and HOXA9, HOXA10, HOXA11, and HOXA13. In some embodiments, the HOXA gene is one or more of HOXA3, HOXA7, HOXA9 and HOXA10.
As used herein, a “reporter” refers to an expression product, i.e. a nucleic acid or a protein, expressed from a specific locus that is measurable and indicative of transcription from the locus. The reporter may be endogenous or heterologous. Preferably, the reporter is a heterologous protein, examples of which include enzymes, immunoaffinity tags, and fluorescent proteins. Preferably, the reporter is a fluorescent protein. Many examples of fluorescent proteins are known in the art. Examples include GFP, EGFP, YFP, tdTomato, mCherry, mBanana, mTagBFP2 and so on.
According to the present disclosure, a PSC may comprise two or more reporters targeted to different loci, which therefore report expression of the different loci. To be meaningful where multiple reporters are potentially expressed in the same cell, the reporters are selected such that they are distinguishable. With respect to fluorescent proteins, this generally means non-overlapping or minimally-overlapping emission spectra.
Where multiple reporters are potentially expressed in the same cell, the reporters may be expressed at separate times or simultaneously. In one embodiment of the second aspect, the SOX17 reporter and the RUNX1C reporter are expressed at separate times or simultaneously.
In one embodiment of the second aspect, the SOX17 reporter comprises a first fluorescent protein expressed by the cell and the RUNX1C reporter comprises a second fluorescent protein expressed by the cell.
As used herein, “differentiating” and “differentiation” refers to a process of a cell changing from one cell type to another, in particular a less specialized type of cell becoming a more specialized type of cell.
As used herein, “tracking differentiation” refers to the ability to observe differentiation, for example, to observe the endothelial (SOX17) to haematopoietic (RUNX1C) transition of during PSC differentiation to definitive haematopoietic stem/precursor cells.
As used herein, “medium” or its plural “media” refers to a liquid or gel designed to support the growth of cells. In some embodiments, the cell culture medium comprises APEL medium.
As used herein, “APEL” medium refers to the Albumin Polyvinylalcohol Essential Lipids medium described in Ng et al. Nature Protocols 3, 768-776 (2008). APEL medium is available commercially. In an embodiment of the first or fifth aspects of the invention, the medium is APEL medium.
All proteins described herein are known to the person skilled in the art, and most if not all proteins described herein are available commercially.
Although the presently disclosed media may include specific components (e.g. morphogens, small molecules, and hematopoietic cytokines), it is contemplated that other components with the same, equivalent, or similar properties may be used in addition to or in place of those disclosed, as are known in the art.
In some embodiments of the first or fifth aspects of the invention, the medium comprises 20-40 ng/mL BMP4, 20-30 ng/mL VEGF, 40 ng/mL SCF, 10-20 ng/mL ACTIVIN A, and 10 ng/mL FGF2.
In some embodiments of the first or fifth aspects of the invention, the concentration of BMP4 in the medium is about 5 ng/mL, about 10 ng/mL, about 15 ng/mL, about 20 ng/mL, about 25 ng/mL, about 30 ng/mL, about 35 ng/mL, about 40 ng/mL, about 45 ng/mL, about 50 ng/mL, about 55 ng/mL, or about 60 ng/mL.
In some embodiments of the first or fifth aspects of the invention, the concentration of VEGF in the medium is about 10 ng/mL, about 15 ng/mL, about 20 ng/mL, about 25 ng/mL, about 30 ng/mL, about 35 ng/mL, about 40 ng/mL, about 45 ng/mL, about 50 ng/mL, about 55 ng/mL, about 60 ng/mL, about 65 ng/mL, or about 70 ng/mL.
In some embodiments of the first or fifth aspects of the invention, the concentration of SCF in the medium is about 10 ng/mL, about 15 ng/mL, about 20 ng/mL, about 25 ng/mL, about 30 ng/mL, about 35 ng/mL, about 40 ng/mL, about 45 ng/mL, about 50 ng/mL, about 55 ng/mL, about 60 ng/mL, about 70 ng/mL, about 80 ng/mL, about 90 ng/mL, about 100 ng/mL, about 110 ng/mL, or about 120 ng/mL.
In some embodiments of the first or fifth aspects of the invention, the concentration of ACTIVIN A in the medium is about 1 ng/mL, about 2 ng/mL, about 3 ng/mL, about 4 ng/mL, about 5 ng/mL, about 6 ng/mL, about 7 ng/mL, about 8 ng/mL, about 9 ng/mL, about 10 ng/mL, about 11 ng/mL, about 12 ng/mL, about 13 ng/mL, about 14 ng/mL, about 15 ng/mL, about 16 ng/mL, about 17 ng/mL, about 18 ng/mL, about 19 ng/mL, about 20 ng/mL, about 21 ng/mL, about 22 ng/mL, about 23 ng/mL, about 24 ng/mL, about 25 ng/mL, or about 30 ng/mL.
In some embodiments of the first or fifth aspects of the invention, the concentration of FGF2, also known as basic fibroblast growth factor, in the medium is about 1 ng/mL, about 2 ng/mL, about 3 ng/mL, about 4 ng/mL, about 5 ng/mL, about 6 ng/mL, about 7 ng/mL, about 8 ng/mL, about 9 ng/mL, about 10 ng/mL, about 11 ng/mL, about 12 ng/mL, about 13 ng/mL, about 14 ng/mL, about 15 ng/mL, about 16 ng/mL, about 17 ng/mL, about 18 ng/mL, about 19 ng/mL, or about 20 ng/mL.
In some embodiments of the first or fifth aspects of the invention, the concentration of IGF2 in the medium is about 20 ng/mL, about 25 ng/mL, about 30 ng/mL, about 35 ng/mL, about 40 ng/mL, about 45 ng/mL, or about 50 ng/mL. Preferably, the medium comprises IGF2 at 30 ng/mL.
In some embodiments of the first or fifth aspects of the invention, the concentration of IL-6 in the medium is about 10 ng/mL, about 15 ng/mL, about 20 ng/mL, about 25 ng/mL, about 30 ng/mL, about 35 ng/mL, about 40 ng/mL, about 45 ng/mL, or about 50 ng/mL. Preferably, the concentration of IL-6 in the medium is 25 ng/mL to 30 ng/mL.
In some embodiments of the first or fifth aspects of the invention, the concentration of IL-3 in the medium is about 10 ng/mL, about 15 ng/mL, about 20 ng/mL, about 25 ng/mL, about 30 ng/mL, about 35 ng/mL, about 40 ng/mL, about 45 ng/mL, or about 50 ng/mL. Preferably, the concentration of IL-3 in the medium is 25 ng/mL to 30 ng/mL.
In some embodiments of the first or fifth aspects of the invention, the concentration of TPO in the medium is about 10 ng/mL, about 15 ng/mL, about 20 ng/mL, about 25 ng/mL, about 30 ng/mL, about 35 ng/mL, about 40 ng/mL, about 45 ng/mL, or about 50 ng/mL. Preferably, the concentration of TPO in the medium is 25 ng/mL to 30 ng/mL.
In some embodiments of the first or fifth aspects of the invention, the concentration of FLT3L in the medium is about 1 ng/mL, about 2 ng/mL, about 3 ng/mL, about 4 ng/mL, about 5 ng/mL, about 6 ng/mL, about 7 ng/mL, about 8 ng/mL, about 9 ng/mL, about 10 ng/mL, about 15 ng/mL, about 20 ng/mL, about 25 ng/mL, about 30 ng/mL, about 35 ng/mL, about 40 ng/mL, about 45 ng/mL, about 50 ng/mL, about 55 ng/mL, about 60 ng/mL, or about 70 ng/mL.
In some embodiments of the first or fifth aspects of the invention, the concentration of EPO in the medium is about 1 U/mL, about 2 U/mL, about 3 U/mL, about 4 U/mL, about 5 U/mL, about 6 U/mL, about 7 U/mL, about 8 U/mL, about 9 U/mL, or about 10 U/mL.
Preferably, the medium comprises 3 U/mL, 4 U/mL or 5 U/mL EPO.
In some embodiments of the first or fifth aspects of the invention, the concentration of GM-CSF in the medium is about 10 ng/mL, about 15 ng/mL, about 20 ng/mL, about 25 ng/mL, about 30 ng/mL, about 35 ng/mL, about 40 ng/mL, about 45 ng/mL, or about 50 ng/mL. Preferably, the concentration of GM-CSF in the medium is 25 ng/mL.
Media disclosed herein may be also made in concentrated, including dried, forms that are diluted prior to use, such as 2×, 10×, 100×, or 1000× concentrations.
As used herein, “culturing” refers to the process by which cells are grown under controlled, in vitro conditions.
As used herein, “extracellular matrix” refers to the non-cellular component of all tissues and organs that provides essential physical scaffolding for the cellular components and initiates crucial biochemical and biomechanical cues that are required for tissue morphogenesis, differentiation and homeostasis. It follows that an “extracellular matrix protein” is a protein present in and/or derived from the extracellular matrix. Examples of extracellular matrix proteins include collagens, elastin, fibronectin, laminin, and proteoglycans. Extracellular matrix proteins, including mixtures, are available commercially, e.g. MATRIGEL™.
As used herein, a “condition, disease or disorder requiring HSC transplantation” may be a malignant condition, disease or disorder, for example acute myeloid leukemia (AML), chronic myeloid leukemia (CML), acute lymphoblastic leukemia (ALL), Hodgkin lymphoma (relapsed, refractory), Non-Hodgkin (relapsed or refractory) lymphoma, neuroblastoma, Ewing sarcoma, multiple myeloma, a myelodysplastic syndrome, a glioma, or other solid tumour, or may be a non-malignant condition, disease or disorder, for example thalassemia, sickle cell anemia, aplastic anemia, Fanconi anemia, an immune deficiency syndrome, or an inborn error of metabolism.
A haematopoietic stem/progenitor cell of the first or sixth aspect or produced according to fifth aspect of the invention may be used for treating a condition, disease or disorder requiring HSC transplantation following myeloablative therapy in a subject. Alternatively, haematopoietic stem/progenitor cell of the first or sixth aspect or produced according to fifth aspect of the invention may be used for treating a condition, disease or disorder requiring HSC transplantation that does not require myeloablative therapy in a subject.
As used herein, “myeloablative therapy” refers to treatment, generally radiation or chemotherapy, that kills cells in the bone marrow, including cancer cells.
It will be appreciated by the person skilled in the art that the exact manner of administering to a subject a therapeutically effective amount of a haematopoietic stem/progenitor cell of or produced according to the invention for treating a condition, disease or disorder will be at the discretion of the medical practitioner. The mode of administration, including dosage, combination with other agents, timing and frequency of administration, and the like, may be affected by the diagnosis of a subject's likely responsiveness to treatment with the haematopoietic stem/progenitor cell of or produced according to the invention, as well as the subject's condition and history.
As used herein, the term “therapeutic composition” refers to a composition comprising a haematopoietic stem/progenitor cell of or produced according to the invention that has been formulated for administration to a subject. Preferably, the therapeutic composition is sterile. In one embodiment, the therapeutic composition is pyrogen-free.
The haematopoietic stem/progenitor cell of or produced according to the invention will be formulated, dosed, and administered in a fashion consistent with good medical practice. Factors for consideration in this context include the particular type of condition, disease or disorder being treated, the particular subject being treated, the clinical condition of the subject, the site of administration, the method of administration, the scheduling of administration, possible side-effects and other factors known to medical practitioners. The therapeutically effective amount of the haematopoietic stem/progenitor cell of or produced according to the invention to be administered will be governed by such considerations.
The haematopoietic stem/progenitor cell of the first or sixth aspect or produced according to the fifth aspect of the invention may be administered to a subject by any suitable method including intravenous (IV), intra-arterial, intramuscular, intraperitoneal, intracerobrospinal, subcutaneous (SC), intra-articular, intrasynovial, intrathecal, intracoronary, transendocardial, surgical implantation, topical and inhalation (e.g. intrapulmonary) routes. Most preferably, the haematopoietic stem/progenitor cell of or produced according to the invention is administered IV.
The term “therapeutically effective amount” refers to an amount of the haematopoietic stem/progenitor cell of or produced according to the invention effective to treat a condition, disease or disorder in a subject.
The terms “treat”, “treating” or “treatment” refer to both therapeutic treatment and prophylactic or preventative measures, wherein the aim is to prevent or ameliorate a condition, disease or disorder in a subject or slow down (lessen) progression of a condition, disease or disorder in a subject. Subjects in need of treatment include those already with the condition, disease or disorder as well as those in which the condition, disease or disorder is to be prevented.
The terms “preventing”, “prevention”, “preventative” or “prophylactic” refers to keeping from occurring, or to hinder, defend from, or protect from the occurrence of a condition, a disease or disorder, including an abnormality or symptom. A subject in need of prevention may be prone to develop the condition, disease or disorder.
The term “ameliorate” or “amelioration” refers to a decrease, reduction or elimination of a condition, a disease or disorder, including an abnormality or symptom. A subject in need of treatment may already have the condition, disease or disorder, or may be prone to have the condition, disease or disorder, or may be in whom the condition, disease or disorder is to be prevented.
As used herein, the term “subject” refers to a mammal. The mammal may be a primate, particularly a human, or may be a domestic, zoo, or companion animal. Although it is particularly contemplated that the method and its resulting definitive haematopoietic stem/progenitor cell or population of definitive haematopoietic stem/progenitor cells disclosed herein are suitable for medical treatment of humans, they are also applicable to veterinary treatment, including treatment of domestic animals such as horses, cattle and sheep, companion animals such as dogs and cats, or zoo animals such as felids, canids, bovids and ungulates.
EXAMPLESMethods
Generation and Validation of Targeted RUNX1CGFP/w, RUNX1CGFP/GFP, SOX17mCHERRY/w and SOX17mCHERRY/w RUNX1CGFP/w hESC Lines
The RUNX1C targeting vectors comprised a 5.9 kb 5′ homology arm, sequences encoding GFP, a loxP flanked antibiotic resistance cassette with a PGK promoter driving either neomycin phosphotransferase (PGKneo) or hygromycin resistance (PGKhygro) genes and a 3 kb 3′ homology arm (
A list of primers used in this study, their sequences and locations within the RUNX1 gene or the targeting vector is provided in Table 1. Primers p1-p10 were used for screening of targeted clones and the remaining primers were used for RT-PCR to distinguish the B (Human Fujita a and Human Fujita c) and C (Human Fujita b and Human Fujita c) isoforms or identify the combination of the B and C isoforms (denoted common RUNX1) (fAML1a and rAML1a). See also
To generate the RUNX1CGFP/GFP knockout hESC line, the RUNX1CGFP/w line (clone 67.4) was electroporated with the RUNX1C PGKhygro construct described above, resulting in the identification of 2/308 hygromycin resistant clones that were targeted at the 3′ end following PCR screening using p3/p4 primers (
The SOX17mCHERRY/w H9 line has been described previously. Briefly, The SOX17-mCHERRY targeting vector comprised an 8.3 kb 5′ homology arm that encompassed genomic sequences located immediately upstream of the SOX17 translational start site, sequences encoding mCHERRY, a loxP-flanked PGK-Neo antibiotic resistance cassette and a 3.6 kb 3′ SOX17 homology arm. The human ESC line H9 (purchased from WiCell) was electroporated with the linearized vector and correctly targeted clones identified using a PCR based screening strategy as described above. The antibiotic resistance cassette was excised using CRE recombinase. The SOX17mCHERRY/w human ESC line used in this study was validated by demonstrating the correlation between SOX17 and protein and mCHERRY expression on endodermally differentiated cells (
The dual reporter SOX17mCHERRY/w RUNX1CGFP/w line was generated in H9 by retargeting GFP to the RUNX1C locus in the SOX17mCHERRY/w human ESC line, as outlined above.
For all lines, genomic integrity was confirmed using the Illumina HumanCytoSNP-12 v2.1 array.
Gene Expression Analysis
Total RNA was prepared from ES cells and dissociated EBs using the Qiagen RNeasy kit or the Bioline Isolate II RNA Mini Kit according to the manufacturers' instructions. cDNA was reverse transcribed using random hexamer priming and Superscript III (Invitrogen) or Tetro (Bioline) cDNA synthesis kits as specified by the manufacturer. Human RUNX1 isoform specific primers (Table 1) were based on those designed for mouse Runx1 previously. For other genes, TaqMan gene expression probes and reagents (Applied Biosystems) were used for quantitative real-time PCR analysis and GAPDH was used as the reference gene to normalize data.
Culture and Differentiation of hESCs
The HES3 and H9 human embryonic stem cell lines used in these studies were obtained from ES Cell International (now owned by Biotime Inc) and WiCell. The human ESC work was approved by the Monash Medical Centre and Royal Children's Hospital Human Research Ethics Committees. Culture and enzymatic passaging of hESC lines was performed as reported in Ng et al. Nature Protocols 3, 768-776 (2008). Differentiation of hESC lines was performed using the spin EB method in APEL medium as described in Ng et al. Nature Protocols 3, 768-776 (2008) supplemented for the first 4 days with 20-40 ng/mL recombinant human (rh) bone morphogenetic protein 4 (BMP4, R&D Systems), 20-30 ng/mL rh vascular endothelial growth factor (VEGF, PeproTech) and 20-40 ng/mL rh stem cell factor (SCF, PeproTech), 10-20 ng/mL rh ACTIVIN A (R&D Systems) and 10 ng/mL rh FGF2 (PeproTech) (
Differentiation of hESCs in Spinner Flasks
To obtain larger yields of haematopoietic cells for d7 T cell assays, after the first 4 days of spin EB culture, the EBs were transferred to 125 ml spinner flasks (Corning Life Sciences) at a density of 60 EBs per 60 ml and cultured for a further 3 days in APEL medium/0.1% methylcellulose supplemented with 50 ng/mL rhVEGF, 3 U/mL rh EPO, 50 ng/mL rh SCF, 10 ng/mL rh FGF2, 30 ng/mL rh TPO, 30 ng/mL rh FLT3L and 30 ng/mL rh IGF2. For analysis, the EBs were harvested and dissociated into a single cell suspension using TrypLE Select and passaged through a 23-gauge needle prior to cell sorting as described in Ng et al. Nature Protocols 3, 768-776 (2008).
Flow Cytometry and Cell Sorting
Antibodies directed against the following cell surface antigens (fluorochrome, manufacturer, catalogue number, clone [where known]) were used for staining dissociated EBs for flow cytometric analysis: CD34 (phycoerythrin[pe]-cy7, Biolegend #343515, clone 581) CD45 (brilliant violet, Biolegend #304032, clone H130), VE-CADHERIN (allophycocyanin [apc], Biolegend #348508, clone bv9), CD33 (apc, BD Pharmingen #340474), TIE2 (BD Pharmingen #557039, clone 33), platelet-derived growth factor receptor alpha (PDGFRa)(BD Pharmingen #556001, clone aR1), CD41a (apc, BD Pharmingen #559777, clone HIP8), CD43 (apc, Biolegend #343206, clone 10G7), Glycophorin A (apc, BD Parmingen #551336, clone GA-R2[HIR2]), CD7 (apc, Biolegend #343107, clone 6B7), CD5 (pe-cy7, Biolegend #300622, clone UCHT2), CD4 (apc, BD Pharmingen #555349, clone RPA-T4), CD8 (pe-cy7, Biolegend #344712, clone SK1), CD56 (pe, BD Pharmingen #555516, clone B159), CD90 (apc, BD Pharmingen #559869, clone 5E10), KIT (apc, Biolegend #313206, clone10D2) and DLL4 (apc, Biolegend #346508, MHD4-46). Unconjugated primary antibodies (TIE2 #557039, clone 33 and PDGFRa #556001, clone aR1, both BD Pharmingen) were detected with anti-mouse secondary antibodies conjugated to apc (BD Pharmingen #550826). Flow cytometric analysis was performed using FACSCALIBUR™ and BD FORTESSA™ analysers and cell sorting was done using FACS DIVA™, FACS ARIA™ and BD INFLUX™ cell sorters (BD Biosciences). Flow cytometry plots shown in
T Cell Assays
Flow sorted fractions of d7-8 differentiated SOX17mCHERRY/w RUNX1CGFP/w cells were co-cultured with OP9 DL4 cells in αMEM (Invitrogen) supplemented with 20% FCS, 10 ng/nl rh FLT3L, 10 ng/mL rh IL7 (PeproTech) and 25 ng/mL rh SCF. Twenty thousand sorted cells were cultured per well of OP9 DL4 on 12-well plates (Costar). Cells were passaged weekly onto fresh OP9 DL4 layers by pipetting the cultures and passing cell through a 40 μM cell strainer. Cultures were analysed by flow cytometry after ˜21 and ˜28 days of culture on OP9 DL4. Brightfield and fluorescent images in
Colony Forming Assays
Hemangioblast colonies (Bl-CFCs) were identified by culturing 5×103 dissociated cells from d2-4 EBs in serum-free MethoCult (StemCell Technologies) or in a formulation denoted MC-APEL (1% methylcellulose in APEL medium) supplemented with 30 ng/mL VEGF, 50 ng/mL SCF and 3 U/mL EPO, with 20 ng/mL BMP4 and 10 ng/mL FGF2 also added to MC-APEL cultures. For the identification of hematopoietic CFCs emerging at later times, 2-20×103 differentiated cells were cultured in serum free MethoCult or MC-APEL supplemented with 5 U/mL rh EPO, 50 ng/mL rh VEGF, 30 ng/mL rh IL-3, 30 ng/mL rh IL-6, 25 ng/mL rh GM-CSF, 25 ng/mL rh TPO, 50 ng/mL rh FLT3L and 100 ng/mL rh SCF. Plates were scored for colony formation between 7 and 14 days. Cytocentrifuge preparations of harvested colonies were stained with May-Grunwald-Giemsa (Sigma) according to the manufacturer's instructions. Brightfield and fluorescence images of stained cytocentrifuge preparations shown in
Transcriptional Profiling by Illumina Microarray
In the first series of experiments, differentiated cultures of RUNX1CGFP/w cells were harvested daily for 8 days. In a second experiment, differentiated cultures of RUNX1CGFP/w and RUNX1CGFP/GFP cells were dissociated at d14 and flow sorted based on their expression of CD34 and GFP using a BD INFLUX™ cell sorter. Finally, three cord blood cell samples sorted for CD34 expression were profiled in parallel with the d14 differentiated RUNX1CGFP/w and RUNX1CGFP/GFP samples. In all experiments, total RNA was extracted from the collected cell fractions (High Pure RNA Isolation Kit, Roche Applied Science) and RNA quality was assessed using a NanoDrop 2000 analyzer (Thermo Scientific). Total RNA from each fraction was amplified, labeled, and hybridized to Human HT12 v3 and v4 BeadChips according to Illumina standard protocols (Illumina) at the Australian Genome Research Facility. Initial data analysis was performed using GenomeStudio Version 2010.2 (Illumina), using average normalization across all the samples. Further data analysis was performed in R/BioConductor using algorithms within the lumi package (function: bgAdjust.affy and quantile normalization). Subsequent data analysis was performed using MultiExperiment Viewer. Principle component analysis was performed to determine the overall relationship between samples. Differentially expressed genes were identified using SAM (Significance Analysis for Microarrays) with a zero false discovery rate. Hierarchical clustering of genes was performed using Pearson correlation with average linkage clustering. Differentially expressed genes were subjected to pathway over-representation and functional clustering analysis using the DAVID public database (Database for Annotation, Visualization and Integrated Discovery). The microarray data reported here are deposited with Array Express, accession number E-MEXP-3356, and Gene Expression Omnibus, accession number GSE64876.
The experiment describing HOX gene expressing in differentiated cultures of d4 MIXL1GFP/w cells used data derived from a previous sorting experiment. The Array Express accession number for this data is E-MEXP-2495.
Transcriptional Profiling Using RNA-Seq
Differentiated SOX17mCHERRY/wRUIVXICGFP/w cells at d10/11 cultured with or without additional SB/CHIR from d2-4 were flow sorted as shown in
Assay for Transposase-Accessible Chromatin with High Throughput Sequencing (ATAC-Seq)
Differentiated SOX17mCHERRY/wRUNX1CGFP/w cells at d10 cultured with or without additional SB/CHIR from d2-4 were flow sorted as shown in
DNA Methylation Analysis Using Illumina Infinium 450 Mircoarray
Differentiated SOX17mCHERRY/wRUNX1CGFP/w cells at d10 cultured with or without additional SB/CHIR from d2-4 were flow sorted as shown in
Mice
NOD/SCID mice were purchased from ARC (Perth, Australia) and housed in specific pathogen free (SPF) conditions at Monash University Animal Services (MAS, Melbourne, Australia). NOD.Cg-Prkdcdscid Il2rgtm1wj1/SzJ mice (NOD/SCID/IL2Rγ−/−, NSG) (stock number 005557, The Jackson Laboratory) were bred at MAS in SPF conditions. The colony was regularly checked for γ-c deficiency by PCR according to the Jackson Laboratory protocol. The Monash University institutional animal ethics committee approved all animal protocols, and experiments were carried out under its guidelines for the care and use of laboratory animals.
Human Tissue Collection, Processing and Sequencing
i) AGM Tissue Procurement and Processing
This study was performed in accordance with institutional guidelines and was approved by the local research ethics committee (University of Tubingen IRB #406/2011B02, #710/2012BO2 and #690/2013BO2). Human first trimester tissue (5 weeks of gestation) was obtained from an electively aborted fetus following informed consent and de-identification. After procurement, the tissues was immediately washed in sterile Dulbecco's phosphate buffered saline (DPBS, Invitrogen, Cat#14190-169), placed in medium (DMEM/F12, Gibco, Cat#31330-095) that was supplemented with 5% foetal bovine serum (ThermoFisher, Cat#16000-044), 1% Penicillin-Streptomycin (Gibco, Cat#15140-122) and 2.5 μg/mL Amphotericin B (Sigma Aldrich, Cat# A9528), and processed for gene expression analyses within 48 hours. The embryo was dissected to isolate the AGM region. The tissue was digested in 2.5 U dispase (Gibco), 90 mg collagenase A (Worthington), and 0.075 mg DNase I (Sigma) per ml in PBS containing 10% FBS, for 90 minutes at 37° C. Cells were then disaggregated by pipetting, and filtered through a 70 μm cell strainer.
ii) Foetal Liver Tissue Collection and Processing
Foetal liver was de-identified, discarded material obtained from elective termination of a second trimester pregnancies following informed consent. Specimen age for this study (17 weeks) is denoted as developmental age, two weeks less than gestational age, and was determined by ultrasound or estimated by the date of the last menstrual period. Tissues were harvested into PBS 5% FBS (Hyclone), Ciprofloxacin HCl (10 ng/mL, Sigma), amphotericin B (2.50 μg/mL, Invitrogen), 1% penicillin/streptomycin, transported on ice and processed the same day. Tissues were mechanically dissociated using scalpels and syringes. Mononuclear cells were enriched on a Ficoll layer according to manufacturer's protocol (GE Healthcare Biosciences AB) and strained through a 70 μm mesh.
iii) Antibody Staining and Flow Cytometric Sorting
Cells were stained with the following antibodies: human-CD45-APC-H7 cl.2D1 (Biolegend 368516; 1:100), CD34-APC cl. 581 (BD 555824; 1:20), CD90-BV421 cl. 5E10 (BD 562556; 1:100), CD38-PE-Cy7 cl. HIT2 (BD 560677; 1:100), CD43-FITC cl.1G10 (BD 555475; 1:20), GPI-80-PE cl. 3H9 (MBL D087-5; 1:50). The indicated populations (
iii) RNA Extraction and Library Preparation and RNA-Sequencing.
Total RNA from 500-2000 sorted cells was extracted using the RNeasy Mini kit (Qiagen) and library was constructed using Ovation Rna-seq system System v2 (Nugen), followed by KAPA LTP Library Preparation Kit. Libraries were sequenced using HIseq-2000 (Illumina) to obtain single end 50 bp long reads. Demultiplexing of the reads based on the barcoding was performed using in house Unix shell script. Sequencing was performed at the University of California, Los Angeles, core sequencing facility.
Cord Blood Cells
Samples of human umbilical cord blood from healthy subjects were obtained from the Mercy Hospital for Women, Studley Road, Heidelberg, Victoria 3084, under auspices of the hospital Human Research Ethics Committee. Mononuclear cells or flow sorted CD34 cells were isolated and cryopreservation for subsequent use in transcriptional profiling or transplantation assays.
Analysis of Bone Marrow Homing of CFSE-Labeled Differentiated Cells
i) CFSE Labeling of Differentiated hPSCs
Differentiated RUNX1CGFP/w and RUNX1CGFP/GFP cells were flow sorted at d14 based on CD34 and GFP expression, or at earlier time points on the basis of CD34 and CD41 expression. Viable cells were labeled with 5-(and 6)-carboxyfluorescein diacetate succinimidyl ester (CFSE; Molecular Probes). Briefly, cell populations were resuspended in PBS 0.5% heat-inactivated (HI) foetal calf serum (FCS; Hyclone, Logan, Utah, USA) at a density of 1×106 cells/mL and pre-incubated at 37° C. for 3 minutes. CFSE was diluted to 5 mM in dimethyl sulfoxide (DMSO) and then to 5 μM in PBS. CFSE was added to the cells to give a final concentration of 0.5 μM, and the dye solution/cell mixture was incubated at 37° C. for a further 10 minutes. Staining was stopped by adding 10 times the dye solution/cell volume of ice-cold PBS containing 20% FCS.
ii) Irradiation of Immunodeficient Mice
The ability of cells to reconstitute haematopoiesis was analyzed in NSG mice receiving a near-lethal dose of irradiation (2.75 Gy), delivered from two opposing 137Cs sources (Gammacell 40; Atomic Energy of Canada, Ottawa, ON, Canada) at a dose rate of 1.4 Gy/minute.
iii) Cell Homing
Sorted, CFSE-labeled cells were transplanted in 0.2 ml of PBS by injection into the lateral tail vein of near-lethally irradiated NSG mice, with 3 recipients per sorted cell fraction, for each of 4 experiments. As a result, the data accumulated are the result of multiple sorts at each time point. On average, 1×105 sorted, CFSE+ cells were co-injected with 2×106 irradiated (15 Gy), unlabelled human cord-blood mononuclear cells. The homing ability was analysed 15 hours post-transplant. For the analysis of homing, bone marrow samples were prepared from individual mice as described below. The analysis of the proportion of CFSE+ donor cells in each fraction was performed using the number of live white blood cells (WBC) as the denominator for the total number of cells analysed by flow cytometry. No statistical method was used to predetermine sample size and the experiments were not randomized. The investigators were not blinded to allocation during experiments and outcome assessment.
iv) Isolation of Hematopoietic Cells
NSG mice were humanely killed by cervical dislocation. Bone marrow was routinely collected from femurs, tibiae and iliac crests.
These bones were thoroughly ground in a mortar and pestle in phosphate-buffered saline (PBS) supplemented with 2% heat inactivated FCS. Bone fragments were washed, and the supernatant cell suspension was filtered through a 40 μM filter (Becton Dickinson, N.J., USA). The marrow was centrifuged (400 g, 5 minutes) and resuspended in fresh buffer. The cell supernatant was re-filtered through a 40 μM filter and diluted to 1×107 cells/mL PBS supplemented with 2% heat inactivated FCS (buffer). Donor cells labeled with CFSE that homed to the bone marrow were identified by flow cytometry.
Statistical Analysis
Experiments were analysed using GraphPad Prism versions 5 and 6 (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 ANOVA for experiments with multiple comparisons of one or more grouped variables, accompanied by the post-tests (Dunnett, Tukey, Holmes-Sidak) indicated as appropriate by the software. No statistical method was used to predetermine sample size.
Results
RUNX1C is Expressed in Human hESC-Derived Haematopoietic Cells
To identify haematopoietic stem/progenitor cells during hPSC differentiation, reporter lines targeting GFP to the RUNX1C locus (
Sequential expression of the haematopoietic cell surface markers CD34, CD43 and CD41, antedated RUNX1C expression (
RUNX1C Expression Identifies CD34+ Cells with Clonogenic and Bone Marrow Homing Ability
Given the hematopoietic specificity of the RUNX1C isoform, homozygous RUNX1CGFP/GFP cells to assess whether RUNX1C was required for human haematopoiesis (
Homing to bone marrow of d14 differentiated cells; an ability needed for long-term engraftment was evaluated. Significant homing was observed in GFP+CD34lo cells sorted from the heterozygous, but not the null line (
hPSC-Derived and Cord Blood Progenitor Cells Display Discordant HOX Gene Expression
To explore the reasons for engraftment failure of hPSC-derived hematopoietic cells, transcriptional profiles of cells sorted from d14 RUNX1CGFP/w and RUNX1CGFP/GFP EBs with repopulation-competent CB CD34+ cells were compared. RUNX1CGFP/w and RUNX1CGFP/GFP EBs were transcriptionally nearly identical, judged by hierarchical clustering of sorted fractions, scatter plot analysis, Pearson correlation coefficients (
Conversely, HOXA5, HOXA9 and HOXA10 are expressed in the primitive streak and also in HSCs. Furthermore, deletion of Hoxa9 or the whole Hoxa cluster in mesoderm is associated with severe haematopoietic defects. Lastly, failure to express HOXA genes was a major difference between hPSC-derived haematopoietic cells and CB CD34+ cells (
HOXB, but not HOXA, genes were first expressed at d2-3 of hPSC differentiation, coincident with the expression of MIXLI1 and BRACHYURY (T) primitive streak genes, and the CDX genes that regulate HOX expression (
HOXA Expression in Mesoderm is Induced by a Combination of WNT Agonist and ACTIVIN Antagonist
The effects of adding the WNT agonist, CHIR99021 (CHIR), and the ACTIVIN receptor-like kinase (ALK) inhibitor, SB431542 (SB), from differentiation days 2 to 4, the period during which HOX gene expression is initiated (
RNA-Seq Analysis Confirms Enhanced HOXA Gene Transcription as a Major Consequence of SB/CHIR Treatment
It was hypothesized that the SB/CHIR protocol (
The transcriptional profiles of d10 control and SB/CHIR differentiated SOX17mCHERRY/wRUNX1CGFP/w cells were compared, sorting cultures on CD34, CD43 and SOX17− expression into haematopoietic (43+), endothelial (SOX17+34+ and SOX17−34+) and CD34− adherent cells (SOX17+34− and SOX17−34−) (
HOXA cluster genes were up-regulated in all populations sorted from SB/CHIR cultures (
Endothelia from SB/CHIR Treated Cultures Express Genes Indicative of Aortic Fate
To define the endothelium developing under control and SB/CHIR conditions, these populations were examined for the expression of genes defining umbilical cord arterial and venous endothelia (
Importantly, genes were not identified whose expression indicated that SB/CHIR directed endothelium towards an aorta-like fate, from which definitive haematopoeitic cells could subsequently arise. The angiotensin II type 2 receptor (AGTR2), expressed in foetal rat aorta, was transcribed in SB/CHIR patterned SOX17+ endothelium, whilst the type 1 receptor (AGTR1) was expressed in control endothelia (
The transcriptional programs of the SOX17+34+ and SOX17−34+ SB/CHIR derived endothelial populations were then compared. It was observed that HOXA and NOTCH pathway genes were more highly expressed in SOX17+34+ cells, whilst haematopoietic transcription factors and surface markers were up-regulated in SOX17−34+ populations (
Adherent CD34− cells from control and SB/CHIR cultures expressed a host of growth factor genes including WNT3, WNT5A, BMP4, JAG1, KITLG and VEGF. Additional genes such as CXCL12, DLL1, FGF2, and ALDH1A2 were selectively up-regulated in SB/CHIR patterned CD34 cells, highlighting the potential supportive role that these cells might play (
A Pulse of SB/CHIR Increases Chromatin Accessibility of HOX Cluster Genes
The mechanisms underpinning the sustained increase in HOXA expression induced by SB/CHIR were explored. Evaluation of CpG methylation in d10 SOX17+ and SOX17− endothelia and SOX17−34− cells by microarray showed that the HOX clusters were hypomethylated in all populations in both treatment groups, indicating that changes in methylation did not play a major role in HOXA gene induction by SB/CHIR (
Correlating the RNA-Seq and ATAC-Seq experiments revealed that the increased chromatin accessibility was accompanied by selective gene expression (
Overall, these analyses indicate that SB/CHIR changes the transcriptional profile of multiple cell populations within treated cultures, opening chromatin in the HOXA cluster and generating SOX17+ endothelial cells that bear markers of human aortic cells.
T Cells are Generated from SB/CHIR and Control Cultures
SB/CHIR treatment was correlated with T lymphoid differentiation capacity of d7 endothelial (SOX17−34+ and SOX17+34+) and haematopoietic (RUNX1C−43+ and RUNX1C+34lo) cells using previously published protocols (
Haemogenic SOX17+ Vascular Structures Arise Exclusively from SB/CHIR Treated EBs
When d7 EBs were transferred onto MATRIGEL™-coated plates, it was that observed SOX17+ vascular structures from d10, exclusively in SB/CHIR cultures (
SB/CHIR Cultures Retain Cells Expressing Haematopoietic Stem/Progenitor Cell Surface Markers and Progenitors for Erythroid Colonies Expressing Foetal Globin Genes
Differences between control and SB/CHIR cultures were not restricted to the SOX17+ endothelium. The percentage of viable haematopoietic cells, compared to controls, was increased in SB/CHIR cultures. These cultures also included increased numbers of CD34+ cells, some of which co-expressed foetal liver HSC markers such as ACE and GPI80 (VNN2) or the FCER1A receptor, up-regulated in CB CD34+ cells (
Clonogenic assays from d20-33 revealed that CFCs persisted in SB/CHIR cultures, with an initial myeloid preponderance followed later by predominantly erythroid or mixed colonies, frequently forming multifocal bursts (
Globin gene expression between colonies derived from d23-26 SB/CHIR cultures and primitive haemangioblast colonies generated from control cultures at d4 were compared, and marked reduction in EPSILON globin and increase in BETA globin expression in the former (
HOXA9 and HOXA10 expression levels in the haematopoietic and stromal components of d23-26 SB/CHIR cultures and the d39-44 derived haematopoietic colonies were similar to in vitro differentiated CB CD34+ cells, whilst minimal HOXA expression was seen cells from control cultures (
Nascent Haematopoietic Progenitors Transiently Co-Express SOX17 and RUNX1C.
It was hypothesized that haematopoietic progenitors emerging from the SOX17+ vascular structures would progress from a SOX17+RUNX1C− phenotype, via a SOX17+RUNX1C+ intermediate, to a more mature SOX17−RUNX1C+ phenotype. In support of this, CHERRY+GFP+ haematopoietic cells were observed by microscopy, a result confirmed by flow cytomtery (
hPSC-Derived Stem/Progenitor Cells Share Expression of Key Haematopoietic Genes with Human AGM.
The transcriptional profiles of SOX17 and RUNX1C populations from SB/CHIR cultures with populations that included developing human HSCs sorted from a human 5 week AGM (equivalent to mouse E11.5) and a human 17 week foetal liver (FL) were compared. The AGM sample was fractionated on CD34, CD90 and CD43 into 34+90+43− endothelium (AGM En), a 34+90+43+ population of cells transiting from haemogenic endothelium to haematopoietic stem/progenitors (AGM S/P), and 34+90 43+ committed progenitor cells (AGM Pr1). The CD43+CD45+ FL sample was sorted further on CD34, CD38 and CD90 to enrich for haematopoietic stem/progenitor cells (34+38lo/−90+)(FL S/P). Day 18 hPSC samples included similar populations to the AGM, with SOX17+RUNX1C−34+90+ endothelial (d18 S+R− En), SOX17+RUNX1C+34+90+ haematopoietic stem/progenitor (d18 S+R+ S/P), and SOX17−RUNX1C+34+90lo committed progenitor (d18 S+R+ Pr1) fractions. In addition, d22 samples were included, enriched for committed progenitor cells, SOX17+RUNX1C+34+KIT+ (d22 S+R+ Pr2) and SOX17−RUNX1C+34+KIT+ (d22 SR+ Pr3) (
If the differentiation cultures indeed generated AGM-like haematopoiesis, it was anticipated that the day 18 S+R− En and S+R+ S/P cells would share expression of key genes with the AGM En and S/P fractions, whilst the more mature day 18 S−R+ Pr1, day 22 S+R+ Pr2 and S−R+ Pr3 samples might show greater similarity to the AGM Pr1 committed progenitors.
It was observed that most hematopoietic and vascular cell surface markers expressed on FL S/P cells were shared with AGM S/P and d18 S+R+ S/P cells (
Very similar patterns of gene expression in corresponding AGM and d18 hPSC En and S/P samples were observed for the NOTCH, BMP/TGFβ and WNT signaling genes (
Finally, the list of genes differentially expressed between CB and the HOXA− RUNX1C+34+ cells (
Expression of Anterior and Posterior HOXA Genes Requires a Combination of Retinoic Acid Signaling and SB/CHIR
hPSCs were differentiated in medium supplemented with CHIR99021, BMP4, VEGF, SCF, ACTIVIN A and FGF2 (denoted CH BVSAF in
In the absence of retinoic acid signaling, SB/CHIR, but not CH BVSAF, induced low levels of anterior (
Retinoic acid signaling from d2-4 induced predominantly anterior HOXA genes irrespective of SB CHIR treatment (
Combining EC23 with SB CHIR induced a balance of both anterior and posterior HOXA genes (
Both 10−5 and 10−7 M concentrations of EC23 were effective (
A comparison of
Treating Acute Myeloid Leukaemia
Haematopoietic stem/progenitor cells produced according to the present disclosure will be administered to a subject diagnosed with acute myeloid leukaemia (AML) at a dose of 3×106 cells/kg who has undergone myeloablative therapy according to standard protocols prior to allogeneic HSC transplantation. Administration of the haematopoietic stem/progenitor cells will result in an equivalent or improved response in the subject.
Treating Acute Lymphocytic Leukaemia
Haematopoietic stem/progenitor cells produced according to the present disclosure will be administered to a subject diagnosed with acute lymphocytic leukaemia (ALL) at a dose of 3×106 cells/kg who has undergone myeloablative therapy according to standard protocols prior toallogeneic HSC transplantation. Administration of the haematopoietic stem/progenitor cells will result in an equivalent or improved response in the subject.
Claims
1.-8. (canceled)
9. A method for differentiating a PSC into a definitive haematopoietic stem/progenitor cell, the method comprising increasing HOXA gene expression by culturing the PSC in a medium comprising a WNT agonist and an ACTIVIN antagonist, wherein increased HOXA gene expression is relative to a PSC not cultured in a medium comprising a WNT agonist and an ACTIVIN antagonist.
10. The method of claim 9, wherein the HOXA gene is one or more of HOXA3, HOXA5, HOXA7, HOXA9, HOXA10, HOXA11, and HOXA13.
11. (canceled)
12. The method of claim 9, wherein culturing the PSC in a medium comprising a WNT agonist and an ACTIVIN antagonist comprises culturing for about 2 days.
13. The method of claim 12, wherein culturing the PSC in a medium comprising a WNT agonist and an ACTIVIN antagonist comprises adding the WNT agonist and the ACTIVIN antagonist to the medium at about day 3.
14. The method of claim 9, wherein the WNT agonist is CHIR99021, optionally wherein the medium comprises 3 μM CHIR99021.
15. (canceled)
16. The method of claim 9, wherein the ACTIVIN antagonist is SB431542, optionally wherein the medium comprises 3-4 μM SB431542.
17. (canceled)
18. The method of claim 9, wherein the medium further comprises bone morphogenic protein 4 (BMP4), vascular endothelial growth factor (VEGF), stem cell factor (SCF), ACTIVIN A and fibroblast growth factor 2 (FGF2), and culturing is from about day 0 to about day 4, optionally wherein the medium comprises about 20-ng/mL BMP4, about 20-30 ng/mL VEGF, about 40 ng/mL SCF, about 10-20 ng/nL ACTIVIN A, and about 10 ng/mL FGF2.
19. (canceled)
20. The method of claim 9, wherein the medium further comprises a retinoid or retinoic acid analogue from day 2, day 3, day 4 or day 5.
21. The method of claim 9, wherein the PSC is cultured as an embryoid body.
22. The method of claim 9, further comprising replacing the medium with medium comprising BMP4, VEGF, SCF, FGF2 and insulin-like growth factor 2 (IGF2) at about day 5, optionally wherein the medium comprises about 20 ng/mL BMP4, about 50 ng/mL VEGF, about 50 ng/mL SCF, about 10 ng/mL FGF2, and about 30 ng/mL IGF2.
23. (canceled)
24. The method of claim 22, comprising culturing the PSC for about 2 days.
25. The method of claim 9, further comprising transferring the PSC to a culture surface comprising an extracellular matrix protein at about day 7 and culturing the PSC adherently in medium comprising VEGF, SCF, FGF2, IGF2, interleukin 6 (IL-6), thrombopoietin (TPO), FLT3 receptor ligand (FLT3L), and optionally interleukin 3 (IL-3) and/or erythropoietin (EPO), optionally wherein the medium comprises about 50 ng/mL VEGF, about 50 ng/mL SCF, about 25 ng/mL IL-6, about 30 ng/mL TPO, about ng/mL FLT3L, about 3 U/mL EPO, about 10 ng/mL FGF2, about 50 ng/mL SCF and about ng/mL IGF2, and optionally about 25-30 ng/mL IL-3 and/or about 3 U/mL EPO.
26. (canceled)
27. The method of claim 9, wherein the medium is APEL medium.
28. The method of claim 9, wherein the definitive haematopoietic stem/progenitor cell is capable of producing a SOX17+ blood vessel cell.
29. The method of claim 9, wherein the definitive haematopoietic stem/progenitor cell is capable of producing a RUNX1C+ blood cell.
30. The method of claim 9, wherein the PSC is an embryonic stem cell (ESC) or an induced PSC (iPSC).
31. The method of claim 9, wherein the PSC is human.
32. A definitive haematopoietic stem/progenitor cell differentiated from a PSC by the method of claim 9.
33.-34. (canceled)
35. A method for treating a condition, disease or disorder requiring HSC transplantation, the method comprising administering to a subject a definitive haematopoietic stem/progenitor cell differentiated from a PSC by the method of claim 9.
36. (canceled)
37. The method of claim 35, wherein the condition, disease or disorder is acute myeloid leukemia (AML), chronic myeloid leukemia (CML), acute lymphoblastic leukemia (ALL), Hodgkin lymphoma (relapsed, refractory), Non-Hodgkin (relapsed or refractory) lymphoma, neuroblastoma, Ewing sarcoma, multiple myeloma, a myelodysplastic syndrome, a glioma, other solid tumour, thalassemia, sickle cell anemia, aplastic anemia, Fanconi anemia, an immune deficiency syndrome, or an inborn error of metabolism.
Type: Application
Filed: May 12, 2017
Publication Date: May 16, 2019
Inventors: Elizabeth NG (Parkville, VIC), Edouard STANLEY (Ascot Vale, VIC), Andrew ELEFANTY (Surrey Hills, VIC)
Application Number: 16/300,490
