Reprogramming committed murine blood cells to induced hematopoietic stem cells with defined factors

Abstract

Hematopoietic stem cells (HSCs) sustain blood formation throughout life and are the functional units of bone marrow transplantation. We show that transient expression of six transcription factors Run1t1, Hlf, Lmo2, Prdm5, Pbx1, and Zfp37 imparts multilineage transplantation potential onto otherwise committed lymphoid and myeloid progenitors and myeloid effector cells. Inclusion of Mycn and Meis1 and use of polycistronic viruses increase reprogramming efficacy. The reprogrammed cells, designated induced-HSCs (iHSCs), possess clonal multilineage differentiation potential, reconstitute stem/progenitor compartments, and are serially transplantable. Single-cell analysis revealed that iHSCs derived under optimal conditions exhibit a gene expression profile that is highly similar to endogenous HSCs. These findings demonstrate that expression of a set of defined factors is sufficient to activate the gene networks governing HSC functional identity in committed blood cells. Our results raise the prospect that blood cell reprogramming may be a strategy for derivation of transplantable stem cells for clinical application.

Figure 1. Identification of factors capable of imparting alternative lineage potential in vitro

(A) Heat map showing relative expression (green;high, to purple;low) of 36 regulatory genes identified as HSC-specific in the indicated cell types (see also Table S1). (B) Schematic representation of lentivirus transgene expression cassette (top), and flow cytometry plots showing reporter cassette (ZsGr) expression in Pro/Pre B-cells +/− doxycycline induction (48 hours post). (C) Schematic representation of in vitro screening strategy for cell fate conversion. (D) Representative images of wells showing colonies arising in methylcellulose from Pro/Pre B cells transduced with ZsGr or 36-factor cocktail. (E) Colony number and type arising in methylcellulose from Pro/Pre B cells transduced with ZsGr or 36-factor cocktail. Four independent experiments are shown and each condition performed in triplicate. See Also Table S1.

Figure 2. Identification of factors capable of imparting multi-lineage engraftment potential onto committed progenitors in vivo

(A) Schematic of experimental strategy to identify factors capable of imparting multi-lineage engraftment potential on committed progenitors in vivo. (B) Representative flow cytometry plots showing donor (CD45.2) reconstitution of mice transplanted with control (ZsGr) or 36-factor transduced Pro/Pre B cells or CMPs 16-weeks post-transplant. (C) Donor reconstitution of mice transplanted with ZsGr or 36-factor transduced Pro/Pre B cells or CMPs at indicated time points post-transplantation. Only mice with >1% donor chimerism (dotted line) were considered reconstituted. Recipients transplanted; Pro/PreB;ZsGr n=15, Pro/PreB;36-factor n=15, CMP;ZsGr n=8, and CMP;36-factor n=8. (D) Reconstitution of indicated peripheral blood cell lineages of individual recipients showing >1% donor chimerism presented as % of donor. (E) PCR analysis of immunoglobulin rearrangement showing heavy (J_H), and light chain (J_Lλ, J_Lκ) in bone marrow (BM) cells including B-cells (B220+), stem/progenitor (LSK) cells, myeloid progenitors (Myl Pro), and peripheral blood (PB) cells including B-cells (B220+), recipient myeloid cells (Mac1+ Rec), and donor myeloid cells (Mac1+ Donor) originating from Pro/Pre B cell;36-factor experiment. Loading control; genomic PCR for CD45. (F) PCR-based strategy to identify virally integrated factors and discriminate from endogenous genes. (G) Summary of data showing presence (yellow) or absence (black) of each of the indicated factors in donor B-, T-, and myeloid cells in each of the reconstituted mice shown in (C). See also Figure S1–S2.

Figure 3. Transient ectopic expression of six transcription factors in committed progenitors is sufficient to alter lineage potential in vitro and impart long-term engraftment potential on committed progenitors in vivo

(A) Representative images of wells showing colonies arising in methylcellulose from Pro/Pre B cells transduced with ZsGr or 6-TF cocktail. (B) Colony number and indicated colony type arising in methylcellulose from Pro/Pre B cells transduced with ZsGr or 6-TF cocktail. 3 independent experiments are shown with each condition performed in triplicate. (C) Colony number and type arising in methylcellulose from Pro/Pre B cells transduced with ZsGr, 6-TF cocktail, or 6-TF minus the indicated factor. Each condition performed in triplicate. (D) Donor reconstitution of mice transplanted with ZsGr or 6-TF transduced Pro/Pre B cells or CMPs at indicated time points post-transplantation. Only mice with >1% donor chimerism (dotted line) were considered reconstituted. Recipients transplanted; Pro/PreB;ZsGr n=10, Pro/PreB;6-TF n=12, CMP;ZsGr n=9, and CMP;6-TF n=9. (E) Representative flow cytometry plots showing donor reconstitution and lineage composition of mice transplanted with control (ZsGr) or 6-TF transduced Pro/Pre B cells or CMPs 16-weeks post-transplant. Lineage contribution to Mac1+ myeloid cells, B220+ B-cells, and CD3/4/8+ T-cells is shown. (F) Reconstitution of indicated peripheral blood cell lineages of individual recipients showing >1% donor chimerism presented as % of donor. (G) PCR analysis of immunoglobulin heavy (J_H) chain rearrangement in recipient myeloid cells (Mac1+ Rec), and donor myeloid cells (Mac1+ Donor) originating from Pro/Pre B cell;6-TF experiment. Loading control; genomic PCR for CD45. See also Figure S2.

Figure 4. Inclusion of Meis1 and Mycn and use of polycistronic viruses improves in vivo reprogramming efficiency

(A) Schematic representation of RHL (Runxt1t1, Hlf, Lmo2) and PZP (Pbx1, Zfp37, Prdm5) polycistronic, and Meis1 and Mycn single factor viral constructs. (B) Donor reconstitution of mice transplanted with ZsGr, 8-TF (8 single factor viruses), or 8-TF^Poly (RHL, PZP polycistronic viruses plus Meis1 and Mycn viruses), transduced Pro/Pre B cells at indicated time points post-transplantation. Only mice with >1% donor chimerism were considered reconstituted. Recipients transplanted; ZsGr; n=12, 8-TF; n=6, 8TF^Poly; n=14. (C) Representative flow cytometry plots showing donor reconstitution and lineage contribution of mice transplanted with control (ZsGr), 8-TF, or 8TF^Poly transduced Pro/Pre B cells 16-weeks post-transplant. Lineage contribution to Mac1+GR1- myeloid cells, Mac+GR1+ granulocytes, B220+ B-cells, and CD3/4/8+ T-cells is shown. (D) Reconstitution of indicated peripheral blood cell lineages of individual recipients showing >1% donor chimerism presented as % of donor. (E) PCR analysis of immunoglobulin heavy (J_H) chain rearrangement in recipient (Recip), and donor (Donor) myeloid cells. Loading control; genomic PCR for CD45. See also Figures S2, S3, Table S2.

Figure 5. Reprogrammed cells engraft secondary hematopoietic organs, bone marrow progenitor compartments and reconstitute secondary recipients

(A) Donor reconstitution of peripheral blood (PB), bone marrow (BM), spleen, and thymus of mice transplanted with 8-TF, or 8-TF^Poly transduced Pro/Pre B cells 18–20 weeks post-transplantation. (B) PCR analysis of immunoglobulin heavy (J_H) chain rearrangement in recipient (R), and donor (D) cells. Cell types analyzed include Mac1+ myeloid cells (M), Mac1+GR1+ granulocytes (G), and T-cells (T). Loading control; genomic PCR for CD45. (C) Representative bone marrow stem and progenitor analysis of a recipient transplanted with 8-TF^Poly transduced Pro/Pre B cells 18-weeks post-transplantation showing donor-reconstitution of myeloid progenitors (Myl Pro), megarkaryocyte/erythrocyte progenitors (MEP), granulocyte/monocyte progenitors (GMP), common myeloid progenitors (CMP), megakaryocyte progenitors (MkP), erythroid progenitors (EP), common lymphoid progenitors (CLP), Lineage-negative Sca1+ckit+ multipotent progenitors (LSK), multipotent progenitors (MPP1, MPP2), and hematopoietic stem cells (HSC). All cells were pre-gated through doublet-discriminated, live (propidium iodide negative), and lineage negative cells. (D) Total donor reconstitution of the indicated populations in mice analyzed in (A). (E–F) Reconstitution of the indicated myeloid progenitor (E) and primitive multi-potent and stem cell (F) populations in mice analyzed in (A) presented as percentage of donor. (G) PCR analysis of immunoglobulin heavy (J_H) chain rearrangement in the indicated recipient and donor populations. Loading control; genomic PCR for CD45. (H) Donor reconstitution of secondary recipient mice transplanted with whole bone marrow (WBM) or c-Kit positive bone marrow cells derived from primary transplants of 8-TF transduced Pro/Pre B cells analyzed at 16–22 weeks. Number of recipients transplanted; WBM; n=5, c-Kit+; n=4. Grey bar indicates donor reconstitution level of primary recipient. (I) Reconstitution of indicated peripheral blood cell lineages of individual recipients presented as % of donor. See also Figures S2, S4, S5.

Figure 6. Transient expression of defined transcription factors in myeloid effector cells is sufficient instill them with progenitor activity in vitro, and long-term multi-lineage transplantation potential in vivo

(A) Schematic representation of experimental strategy for assaying the colony forming potential of 8-TF transduced peripheral blood cells. (B) Colony number and type arising in methylcellulose from peripheral blood cells from recipient (left-most lanes) or donor-derived cells from a recipient transplanted with Pro/Pre B cells transduced with 8-TF or 8-TF^Poly cocktail, plus (+) or minus (−) exposure to doxycycline. Results from individual mouse performed in triplicate are shown. (C) Colony number and type arising in methylcellulose from plated granulocytes, macrophages/monocytes (Myl), B-cells, and T-cells purified from the peripheral blood of cells pooled recipients transplanted with Pro/Pre B cells transduced with 8-TF^Poly cocktail plus (+) or minus (−) exposure to doxycycline. n=4 biological replicates per cell type, per condition. (D) Representative colony types and cytospins stained with May Grunwald of colonies derived in (C). (E) Donor reconstitution of mice transplanted with ZsGr, 6-TF^Poly, 8-TF or 8-TF^Poly transduced Mac1+cKit- myeloid effector cells at indicated time points post-transplantation. Only mice with >1% donor chimerism were considered reconstituted. Recipients transplanted; ZsGr; n=6, 6-TF^Poly; n=7, 8-TF; n=6, and 8-TF^Poly; n=8. (F) Reconstitution of indicated peripheral blood cell lineages of mice showing >1% donor chimerism presented as % of donor. (G) Donor reconstitution 16 weeks post-transplant of secondary recipient mice transplanted non-competitively with 5x10⁶ donor-derived (CD45.2+) bone marrow cells derived from primary recipients of 6-TF^Poly, 8-TF or 8-TF^Poly transduced Mac1+cKit- myeloid effector cells. Cells from individual primary donor mice (indicated by ID) were transplanted into N=5 secondary recipients each. (H) Average reconstitution of indicated peripheral blood cell lineages presented as % of donor. n=5 recipients per group. See also Figures S2, S6.

Figure 7. Single cell expression profiling of iHSCs reveals evidence of partial and full reprogramming

(A) Principal component analysis of single cell expression data of iHSCs and the indicated control cell types. iHSCs were derived from experiments in which Pro/Pre B cells were transduced with the 8 identified transcription factors via single (iHSC 8-TF) or polycistronic viruses (iHSC 8-TF^Poly). Data for individual cells of given type indicated in the legend. (B) Dendrograms showing unsupervised hierarchical clustering of single cell expression data of representative control cells, and all iHSCs generated using single viruses (top panel), or polycistronic viruses (lower panel) as described in (A). Dendrogram branches are color-coded according to cell types indicated in the legend. (C) Violin plots and the correlation heatmaps showing single cell expression data of the indicated genes. Expression levels are shown with high expression in red, and low expression in blue. See also Figure S7, Table S3.