An improved medium formulation for efficient ex vivo gene editing, expansion and engraftment of hematopoietic stem and progenitor cells

Abstract

Gene editing has emerged as a powerful tool for the therapeutic correction of monogenic diseases. CRISPR-Cas9 applied to hematopoietic stem and progenitor cells (HSPCs) has shown great promise in proof-of-principle preclinical studies to treat hematological disorders, and clinical trials using these tools are now under way. Nonetheless, there remain important challenges that need to be addressed, such as the efficiency of targeting primitive, long-term repopulating HSPCs and their in vitro expansion for clinical application. In this study, we assessed the effect of different culture medium compositions on the ability of HSPCs to proliferate and undergo homology-directed repair-mediated knock-in of a reporter gene, while preserving their stemness features during ex vivo culture. We demonstrated that by supplementing the culture medium with stem cell agonists and by fine-tuning its cytokine composition it is possible to achieve high levels of gene targeting in long-term repopulating HSPCs both in vitro and in vivo, with a beneficial balance between preservation of stemness and cell expansion. Overall, the implementation of this optimized ex vivo HSPC culture protocol can improve the efficacy, feasibility, and applicability of gene editing as a key step to unlocking the full therapeutic potential of this powerful technology.

Keywords: cell culture; culture medium; gene editing; hematopoietic stem cells.

Graphical abstract

Figure 1

Comparison of ex vivo culture and manipulation of HSPCs grown in media supplemented with either IL-3 or IL-6+SR1+UM171 Frequency of (A) NHEJ-mediated repair (indels) and (B) HDR-mediated knock-in of a PGK-GFP reporter cassette at the WAS locus in CD34+ HSPCs cultured in medium A or medium B. (C) Total number of cells retrieved after 2, 4, and 6 days of culture in the two media. The dotted line represents the starting cell number (100,000 cells). (D) Total number of edited cells retrieved in either medium after 6 days of culture. (E) Frequency of HSCs (CD34+ CD38- CD90+ CD45RA- cells) and (F) MPPs (CD34+ CD38- CD90- CD45RA- cells) detected in the CD34+ bulk cultured in either medium 4 and 7 days after gene editing (6 and 9 days of culture, respectively). HDR, homology-directed repair; HSC, hematopoietic stem cell; MPP, multipotent progenitor; NHEJ, non-homologous end-joining. Data in Figure 1 are presented as mean ± SD, with n = 4 biological replicates; p values were calculated using one-way ANOVA with Tukey’s comparison test (E, F) or two-tailed unpaired Student’s t test (A–D) (∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.005; ∗∗∗∗p < 0.001; no asterisk = nonsignificant).

Figure 2

Evaluating the frequency of reporter gene knock-in and preservation of stemness in HSPC subpopulations (A) Schematic representation of the experimental procedure. CD34+ cells were thawed and FACS sorted into primitive HSCs (CD34+ CD38- CD90+ CD45RA-), MPPs (CD34+ CD38- CD90- CD45RA-) and CD38+ committed progenitors (CD34+ CD38+). HSCs, MPPs, CD38+, and CD34+ unsorted cells were then cultured into four different media (A–D) containing a distinct cocktail of cytokines and stem cell factors. After 2 days of culture, sorted and unsorted cells were gene-edited with the CRISPR-Cas9 platform and the AAV6 donor vector containing a PGK-GFP reporter cassette. After 4 additional days of culture, cells were harvested and analyzed. (B) Expansion rate (expressed as fold increase) of the different populations after 6 days of culture post thawing compared with day 1. (C) Phenotyping of the sorted HSC population at day 6 of culture. The plot shows the percentage of cells in the HSC-sorted population that has retained a HSC phenotype after cell culture, or has differentiated into MPPs, other CD38- progenitors (Other: CD34+ CD38- CD90-/low CD45RA+; this population may include LMPPs and MLPs), or to CD38+ committed progenitors (CD34+ CD38+), when cultured in the four different media. (D) Plots representing the number of white and red colonies formed in semi-solid culture media by sorted and gene-edited HSC and MPP populations and unsorted CD34+ cells cultured in the different media. (E) Percentage of cells harboring a GFP reporter cassette knocked in in the WAS locus in the sorted HSC and MPP populations or unsorted CD34+ cells cultured in the different media. (F) Number of cells harboring a GFP reporter cassette knocked in in the WAS locus in the sorted HSC and MPP populations cultured in the different media. (G) Number of edited repopulating CD34+ CD38- progenitors detected in the unsorted CD34+ samples cultured in the four different media. (H) Plots representing the percentage of GFP-positive colonies formed in methylcellulose by sorted and edited HSC and MPP populations or unsorted and edited CD34+ HSPCs cultured in the different media. HSC, hematopoietic stem cell; LMPP, lympho-myeloid primed progenitor; MLP, multi-lymphoid progenitor; MPP, multipotent progenitor. Data in Figure 2 are presented as mean ± SD, with n = 4 biological replicates in all panels except for (H), where n = 3. p values were calculated using one-way ANOVA with Tukey’s comparison test (∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.005; ∗∗∗∗p < 0.001; no asterisk = nonsignificant).

Figure 3

Evaluating the frequency of gene knock-in and hematopoietic reconstitution by gene-edited HSPCs in vivo A) Schematic representation of the experimental procedure. CD34+ HSPCs from three different healthy donors were thawed, cultured in media (A–D) and gene-edited at day 2. Two days later, cells were harvested and transplanted into sub-lethally irradiated NSG mice. Twelve to 16 weeks after transplantation, the experiment was terminated and the mice analyzed to detect human cell engraftment. (B) Rates of targeted integration (GFP+ cells) achieved in vitro in HSPCs pre-transplant. (C) Engraftment of human cells (CD45+) in the PB of NSG mice at 8 weeks post-transplant and (D) in the BM at 12–16 weeks post-transplant. (E) percentages of B (CD19+), myeloid (CD33+), and T (CD3+) human CD45+ cells retrieved from the BM of transplanted mice at termination. F) Number of human HSCs (CD34+ CD38- CD90+ CD45RA-) and (G) MPPs (CD34+ CD38- CD90- CD45RA-) detected in the BM of transplanted mice. (H) Percentage of human cells with a GFP reporter gene knocked in in the WAS locus (CD45+ GFP+) in the BM of experimental mice at 12–16 weeks post-transplant, respectively. (H) Percentage of human cells with a GFP reporter gene knocked in in the WAS locus (CD45+ GFP+) in the BM of experimental mice at 12–16 weeks post-transplant, respectively. (I) Percentage of human cells with a GFP reporter gene knocked in the WAS locus (CD45+ GFP+) in B (CD19+), myeloid (CD33+), and T cells (CD3+) retrieved from the BM of experimental mice at 12–16 weeks post-transplant. (J) Measurement of the reduction in the percentage of GFP-expressing cells (expressed as fold decrease) in the BM of mice at 12–16 weeks post-transplant compared with pre-transplant. (K) Number of GFP-expressing HSCs, (L) MPPs, and (M) CD34+ CD38− cells in the BM of mice at 12–16 weeks post-transplant. BM, bone marrow; HSC, hematopoietic stem cell; MPP, multipotent progenitor; PB, peripheral blood. Data in Figure 3 are presented as mean ± SD, with n = 10 mice transplanted with HSPCs from two different biological donors in all panels except for (B) where n = 3 biological replicates. p values were calculated using one-way ANOVA with Tukey’s comparison test, except for (I), where a two-way ANOVA with Bonferroni post-test was used (∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.005; ∗∗∗∗p < 0.001; no asterisk = nonsignificant).