Aryl hydrocarbon receptor inhibition promotes hematolymphoid development from human pluripotent stem cells

Abstract

The aryl hydrocarbon receptor (AHR) plays an important physiological role in hematopoiesis. AHR is highly expressed in hematopoietic stem and progenitor cells (HSPCs) and inhibition of AHR results in a marked expansion of human umbilical cord blood-derived HSPCs following cytokine stimulation. It is unknown whether AHR also contributes earlier in human hematopoietic development. To model hematopoiesis, human embryonic stem cells (hESCs) were allowed to differentiate in defined conditions in the presence of the AHR antagonist StemReginin-1 (SR-1) or the AHR agonist 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). We demonstrate a significant increase in CD34⁺CD31⁺ hematoendothelial cells in SR-1-treated hESCs, as well as a twofold expansion of CD34⁺CD45⁺ hematopoietic progenitor cells. Hematopoietic progenitor cells were also significantly increased by SR-1 as quantified by standard hematopoietic colony-forming assays. Using a clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated protein 9 (Cas9)-engineered hESC-RUNX1c-tdTomato reporter cell line with AHR deletion, we further demonstrate a marked enhancement of hematopoietic differentiation relative to wild-type hESCs. We also evaluated whether AHR antagonism could promote innate lymphoid cell differentiation from hESCs. SR-1 increased conventional natural killer (cNK) cell differentiation, whereas TCDD treatment blocked cNK development and supported group 3 innate lymphoid cell (ILC3) differentiation. Collectively, these results demonstrate that AHR regulates early human hematolymphoid cell development and may be targeted to enhance production of specific cell populations derived from human pluripotent stem cells.

Figure 1.

Small-molecule antagonism of AHR enhances early hematoendothelial development from hESCs. (A) Schema of hESC differentiation into early hematoendothelial cells as spin EBs. hESCs are made into spin EBs at day 0 and conditioned into mesoderm lineages for 6 days using defined cytokines (stage 1). At day 6, spin EBs are transferred into hematoendothelial culture media (stage 2) to promote endothelial and hematopoietic cell differentiation. For these studies, cells are treated with either 1 μM SR-1, 10 nM TCDD, or DMSO vehicle control beginning at day 6+0 with media exchanges and/or harvesting performed at day 6+3, day 6+6, and day 6+9. (B) Representative flow cytometry plots of 1 hESC differentiation. Both adherent and nonadherent cell fractions are harvested at day 6+3, day 6+6, and day 6+9 and assessed for endothelial cell (CD34⁺CD31⁺, CD34⁺CD144⁺) and hematopoietic progenitor cell (CD34⁺CD43⁺, CD34⁺CD45⁺) phenotype. (C) Fold change of the total percentage of each hematoendothelial phenotype for SR-1– and TCDD-treated hESCs normalized to matched DMSO-treated controls; n = 4-6, error bars represent standard error of the mean (SEM). *P < .05 as compared with DMSO-treated controls by the Student t test. N/A, not applicable due to absence appreciable of CD34⁺CD45⁺ populations at day 6+3 time point; TPO, thrombopoietin.

Figure 2.

SR-1–treated hESCs demonstrate increased multilineage hematopoietic development. (A) Bulk, nonadherent hematopoietic progenitor cells and (B) sorted CD34⁺CD45⁺ cells derived from hESCs differentiated in the presence of SR-1, TCDD, or DMSO controls were harvested at day 6+5 and seeded at 50 000 cells per dish in a standard methylcellulose CFU assay. Colonies were counted for each treatment group following 2 weeks of culture and scored for the following morphological subsets: burst-forming unit-erythroid (BFU-E); CFU-E; CFU–granulocyte, macrophage (CFU-GM); CFU–granulocyte, erythroid, macrophage, megakaryocyte (CFU-GEMM); CFU-M; n = 3, error bars represent SEM of the total number of colonies per 50 000 cells seeded. *P < .05, **P < .01 as assessed with 1-way analysis of variance (ANOVA) + Tukey-Kramer multiple comparisons post hoc test. (C) Representative BFU-E, CFU-E, CFU-GEMM, CFU-M, and CFU-GM as scored in panels A and B are shown. Scale bar, 100 μm. (D) Nonadherent hematopoietic progenitor cells derived from hESCs differentiated in the presence of SR-1, TCDD, or DMSO controls were harvested at day 6+3, day 6+6, and day 6+9 time points and probed for gene expression by qRT-PCR. For each gene, cycle threshold (C_t) values were normalized to GAPDH at each time point and data are presented as relative fold change to DMSO-treated controls; n = 3, error bars represent SEM; *P < .05, **P < .01 using the Student t test.

Figure 3.

CRISPR/Cas9-engineered hESCs with AHR deletion demonstrate increased early hematoendothelial cell development. (A) gRNA cassette design targeting AHR. gRNA exon 1 indicates 22-nt gRNA specific to AHR exon 1. (B) Gel electrophoresis of PCR products produced from clonally derived hESC-RUNx1c-tdTomato cells nucleofected with AHR gRNA cassette. Genomic DNA was harvested and primers flanking the AHR exon 1 locus were used to generate a PCR product with predicted full-length of 718 bp. WT indicates negatively selected nucleofected hESC-RUNX1c-tdTomato hESCs; +/− indicates individual clones with AHR heterozygous deletion (AHR^+/−); −/− indicates individual clones with AHR homozygous deletion (AHR^−/−). (C) Immunoblot of protein lysate harvested from K562 cells (positive control), NK92 NK cells (positive control), WT hESC-RUNX1c-tdTomato (hESC-R1c-tdTom), heterozygous AHR-deleted hESC-RUNX1c-tdTomato (+/−), and homozygous AHR deleted hESC-RUNX1c-tdTomato (−/−). (D) Representative flow cytometry plots at day 6+3, day 6+6, and day 6+9 from 1 differentiation of WT hESC-RUNX1c-tdTomato (WT), heterozygous AHR hESC-RUNX1c-tdTomato deletion (AHR^+/−), and homozygous AHR hESC-RUNX1c-tdTomato deletion (AHR^−/−). Both adherent and nonadherent cell fractions are harvested at day 6+3, day 6+6, and day 6+9 and assessed for endothelial (CD31, CD144), and hematopoietic (CD33, CD41a, CD43, CD45) phenotype. (E) Representative flow cytometry plots at day 6+3 and day 6+6 from 1 differentiation assessing for RUNX1c expression based on tdTomato fluorescent reporter protein. (F) Nonadherent hematopoietic progenitor cells derived from WT hESC-RUNX1c-tdTomato, AHR^+/− hESC-RUNX1c-tdTomato, or AHR^−/− hESC-RUNX1c-tdTomato were harvested at day 6+5 and seeded at 50 000 cells per dish in a standard methylcellulose CFU assay (CFU). Colonies were counted for each treatment group following 2 weeks of culture and scored for the following morphological subsets, as previously described; n = 3, error bars represent SEM of the total number of colonies per 50 000 cells seeded. *P < .05 as assessed with 1-way ANOVA + Tukey-Kramer multiple comparisons post hoc test. IS, insertion sequence; Term, termination sequence. *718-bp amplicon; ^571-bp amplicon.

Figure 4.

hESCs differentiated in the presence of SR-1 promote the development of functional NK cells. (A) Schema of hESC differentiation into lymphoid cells as spin EBs. hESCs are made into spin EBs at day 0 and cultured in stage 1 conditions with defined cytokines to promote mesoderm development for 11 days. At day 11, spin EBs are transferred onto OP9-DL1 in the presence of NKDM to promote lymphoid differentiation. Cells are treated beginning at day 11+0 with either 1 μM SR-1, 10 nM TCDD, or DMSO vehicle control with media exchanges and/or harvesting performed every week for up to 4 weeks. (B) At day 11, differentiated spin EBs (original magnification ×40) are phenotyped for CD34⁺CD45⁺ expression and transferred to OP9-DL1 stroma in NKDM. Nonadherent hematopoietic cells cultured either in the presence of DMSO, SR-1, or TCDD were assessed for developing NK-cell immunophenotype based on CD56⁺CD45⁺ expression at days 11+21, and 11+28; representative flow cytometry plots from 1 differentiation are shown. (C) Quantification of fold change in total percentage of CD56⁺CD45⁺ cells at both day 11+21 and day 11+28 when treated with DMSO, SR-1, or TCDD. SR-1 and TCDD treatments for each differentiation are normalized to DMSO controls; n = 3 independent differentiation experiments, error bars represent SEM. *P < .05 as assessed with 2-way ANOVA + Tukey-Kramer multiple comparisons post hoc test. (D) Nonadherent hematopoietic progenitor cells derived from hESCs differentiated in the presence of SR-1, TCDD, or DMSO controls were harvested at day 11+28 and probed for gene expression by qRT-PCR. For each gene, C_t values were normalized to GAPDH at each time point and data are presented as relative fold change to DMSO-treated controls; n = 3, error bars represent SEM. *P < .05; #P < .01 using the Student t test. (E) Nonadherent hematopoietic progenitor cells derived from hESCs differentiated in the presence of SR-1, TCDD, or DMSO controls were harvested at day 11+28 and assessed for CD107a expression following 4 hours of coculture with K562 target cells at a 2:1 effector:target ratio. Representative flow cytometry plots are shown from 1 experiment. (F) Quantification of percentage of CD107a⁺ cells when treated with DMSO, SR-1, or TCDD at day 11+28; n = 2-3 replicates. SSC, side scatter.

Figure 5.

hESCs differentiated in the presence of SR-1 skews development toward cNK cells whereas TCDD supports the development of an ILC phenotype. (A) Representative flow cytometry profile of nonadherent hematopoietic cells differentiated from hESCs in the presence of DMSO, SR-1, or TCDD at day 11+28. (B) cNK, NKP, and NKP/ILC subpopulations from day 11+28 DMSO-, SR-1–, and TCDD-differentiated hESCs assessed for CD56 and LFA (CD11a/CD18) surface antigen expression. Representative flow cytometry plots are shown; n = 3. (C) Total percentage of cNK, NKP, and NKP/ILCs present in the nonadherent fraction of differentiating hESCs in the presence of DMSO, SR-1, or TCDD at day 11+28; n = 3, error bars represent SEM. *P < .05, **P < .01, ***P < .001 as compared with DMSO-treated controls and assessed by 2-way ANOVA + Tukey-Kramer multiple comparisons post hoc test. (D) CD94⁻CD117⁺ subpopulations were further quantified for expression of LFA⁺ (NKP) and LFA⁻ (ILC) by flow cytometry; n = 3, error bars represent SEM. *P < .05, **P < .01 as compared with DMSO-treated controls and assessed by 2-way ANOVA + the Tukey-Kramer multiple comparisons post hoc test.

Figure 6.

Model of AHR activity in human developmental hematopoiesis. AHR inhibition mediated by SR-1 (blue arrow) enhances the differentiation of both endothelial cells (ECs) and CD34⁺ hematopoietic progenitor cells. AHR hyperactivation mediated by TCDD (orange arrow) reciprocally acts to attenuate both ECs and CD34⁺ hematopoietic progenitor cells. Once CD34⁺ has been differentiated, AHR inhibition deters further differentiation into CD34⁻ terminally matured hematopoietic cells, whereas AHR hyperactivation supports this process. Upon production of NKP cells, AHR inhibition promotes cNK cell differentiation (cNK), whereas AHR hyperactivation promotes ILC3 differentiation.