Dipeptide species regulate p38MAPK-Smad3 signalling to maintain chronic myelogenous leukaemia stem cells

Abstract

Understanding the specific survival of the rare chronic myelogenous leukaemia (CML) stem cell population could provide a target for therapeutics aimed at eradicating these cells. However, little is known about how survival signalling is regulated in CML stem cells. In this study, we survey global metabolic differences between murine normal haematopoietic stem cells (HSCs) and CML stem cells using metabolomics techniques. Strikingly, we show that CML stem cells accumulate significantly higher levels of certain dipeptide species than normal HSCs. Once internalized, these dipeptide species activate amino-acid signalling via a pathway involving p38MAPK and the stemness transcription factor Smad3, which promotes CML stem cell maintenance. Importantly, pharmacological inhibition of dipeptide uptake inhibits CML stem cell activity in vivo. Our results demonstrate that dipeptide species support CML stem cell maintenance by activating p38MAPK-Smad3 signalling in vivo, and thus point towards a potential therapeutic target for CML treatment.

Figure 1. CML stem cells show accumulations of dipeptides not found in normal HSCs.

Metabolomic analyses of KLS⁺, KLS⁻ and Lin⁺ cells from CML-affected (Tal1-tTA⁺TRE-BCR-ABL1⁺) mice (male, n=6; female, n=6; three experiments) and normal healthy (Tal1-tTA⁺) littermates (male, n=6; female, n=6; two experiments) at 5 weeks post DOX withdrawal. Data are representative of three independent trials. (a–c) Amounts of (a) metabolites related to glycolysis, (b) amino acids and (c) dipeptides in KLS⁺ cells were plotted in whisker boxes. Cross, mean value; horizontal line across box, median value; error bars, maximum and minimum of distribution; dot, extreme data point. P value indicates the statistical significance among normal KLS⁺ versus CML-KLS⁺ as measured by Welch's t-test. (d) Ratios of the indicated dipeptide levels in CML versus normal haematopoietic cells of the indicated subsets. Red-shaded values indicate an increase in expression in CML cells of >10-fold; light red, >5-fold; light blue, <0.5-fold (*P<0.05; normal versus CML; Welch's t-test).

Figure 2. CML stem cells internalize dipeptides via the Slc15A2 dipeptide transporter.

(a) qRT–PCR determination of relative Slc15A2 mRNA levels in LT stem, ST stem, CD48⁺, MPP and KLS⁻ cells from CML-affected (Tal1-tTA⁺TRE-BCR-ABL1⁺) mice (male, n=1; female, n=5) and normal littermate (Tal1-tTA⁺) mice (male, n=4; female, n=4) at 5 weeks post DOX withdrawal. Data are the mean ratio±s.d. of transcript levels normalized to Actb (n=3) (P value, LT-CML stem cells versus normal LT-HSCs; Student's t-test). (b) Uptake of [³H]GlySar by normal-KLS⁺cells, and CML-KLS⁺cells. Cefadroxil (Cefa), Slc15A2 chemical competitor. Data are the mean d.p.m.±s.d. (n=3) (P value, Student's t-test). (c) Uptake of the dipeptide Ser–Leu (S-L) by CML-KLS⁺ cells incubated for 2 h. Internalized S–L and hydrolysed Ser and Leu were determined by metabolomics. Amounts of dipeptides or amino acids were plotted in whisker boxes. Cross, mean value; horizontal line across box, median value; error bars, maximum and minimum of distribution (P value, Cont versus Cefa; Welch's t-test). (d) Metabolomic analyses of dipeptide species in normal-KLS⁺ and CML-KLS⁺ cells isolated from normal littermates (male, n=6; female, n=6; two experiments), and CML-affected mice that received either vehicle (Cont; male, n=2; female, n=4; three experiments) or Cefa (male, n=0; female, n=4; three experiments) for 30 days (P value, Cont versus Cefa; Welch's t-test). (e) Quantification of colony-forming capacity of CML-KLS⁺ and CML-KLS⁻ cells lentivirally transduced to express scrambled shRNA, or shRNA targetting Slc15A2 (shRNA B and D). GFP⁺ cells were co-cultured on OP-9 stromal cells (3% O₂) for 5 days. Data are the mean colony number±s.d. (n=3) and are representative of three experiments (P value, Student's t-test).

Figure 3. Non-canonical Smad3–Ser208 phosphorylation supports LT-CML stem cells in vivo.

(a) Duolink in situ PLA (D-PLA) imaging of (top) Smad2/3 C-terminal phosphorylation and (bottom) interaction between Foxo3a and Smad2 or Smad3 in LT-CML stem cells. The combinations of primary antibodies used are listed in Supplementary Table 1. Ab(−), technical negative control using a single mouse anti-Smad3 antibody. Nuclei were visualized using DAPI. Results are representative of three trials. Scale bar, 10 μm. (b) Quantification of dot number per single LT-CML stem cell in the bottom panel of a from the three experiments. The mean dot number (red line) appears under total cells number (n). P value was measured by the Student's t-test. (c,d) D-PLA imaging and quantification of Smad3 phosphorylation at the indicated sites in freshly isolated LT-CML stem cells from the three experiments. Scale bar, 10μm. (*P<0.0005 compared with Ab(−); Student's t-test; NS, not significant). (e–g) D-PLA imaging and quantification of Smad3–Ser423/425 and Smad3–Ser208 phosphorylation, and Foxo3a–Smad3 interaction, in the indicated CML cell subsets, from the three experiments. Scale bar, 10μm. (*P<0.0005 compared with LT-CML stem cell; Student's t-test; NS). (h,i) Flow cytometric quantification of (h) the indicated GFP⁺CML-KLS⁺ cell subpopulations (red rectangles) among total GFP⁺CML cells, and (i) the most primitive GFP⁺LT-CML stem cells (red rectangles) among GFP⁺CML-KLS⁺ cells, isolated from mice transplanted with CML-KLS⁺ cells bearing retroviral vectors expressing GFP, Smad3-WT, Smad3–3SA or Smad3–S208A. Results are representative of three trials. (j) Quantification of the frequency of GFP⁺LT-CML stem cells among the GFP⁺CML-KLS⁺ cells analysed in i. Data are the mean percentage of GFP⁺ LT-CML stem cells±s.d. (n=3) (P value, Student's t-test; NS).

Figure 4. Dipeptide-induced nutrient signalling regulates CML stem cell activity through a p38MAPK/Smad3–Ser208/Foxo3a axis.

(a–c) D-PLA imaging and quantification of (a,b) Smad3–Ser423/425 and (a,c) Smad3–Ser208 phosphorylation in LT-CML stem cells treated for 30 min with dimethyl sulfoxide (Cont), Ly364947 (TGF-β type I receptor kinase inhibitor; 5 μM) or SB203580 (p38MAPK inhibitor; 5 μM). Nuclei were visualized using DAPI. Results are representative of three trials. The combinations of primary antibodies used are listed in Supplementary Table 1. Scale bar, 10 μm. (P value, Student's t-test; NS, not significant.) (d–f) D-PLA imaging and quantification of (d,e) phospho-p38MAPK and (d,f) phospho-Smad3–Ser208 in LT-CML stem cells treated for 30 min with vehicle (Cont), GlySar (5 μM) or cefadroxil (Cefa; 5 μM). Scale bar, 10 μm. (*P<0.00005 compared with cont; Student's t-test; NS). (g) Quantification of in vitro colony-forming capacity of Foxo3a^+/+LT-CML stem cells and Foxo3a^−/−LT-CML stem cells. Vehicle or Cefa (5 μM). Data are the mean colony number±s.d. (n=3) and are representative of three experiments (P value, Student's t-test). (h,i) D-PLA imaging and quantification of the interaction between phospho-Smad3–Ser208 and Foxo3a in Foxo3a^+/+LT-CML stem cells. Foxo3a^−/−LT-CML stem cells were analysed as a negative control. Scale bar, 10μm. (*P<0.00005; Student's t-test).

Figure 5. Dipeptide transporter chemical competitor does not inhibit normal HSCs in vivo.

(a,b) D-PLA imaging and quantification of p38MAPK–Thr180/Tyr182, Smad3–Ser208 and Smad3–Ser423/425 phosphorylation in LT-CML stem cells and normal LT-HSCs. Nuclei were visualized using DAPI. Results are representative of three trials. The combinations of primary antibodies used are listed in Supplementary Table 1. Scale bar, 10 μm. (P value, Student's t-test.) (c) Quantification of colony-forming capacity of LT-CML stem cells and normal LT-HSCs treated with vehicle (Cont) or Cefa (5 μM) on OP-9 stromal cells (3% O₂) for 5 days. Data are the mean colony number±s.d. (n=3) and are representative of three experiments (P value, Student's t-test). (d,e) Competitive reconstitution assay to determine the effects of Cefa on normal HSCs. (d) Lethally irradiated (9 Gy) C57BL/6 (CD45.2) recipient mice were reconstituted with 1 × 10⁴ normal KLS⁺ cells (CD45.1) plus 5 × 10⁵ unfractionated BM MNCs (CD45.2). Recipients were treated with vehicle (female, n=4) or Cefa (36 mg kg⁻¹ per day; female, n=4) from day 0 to 12 weeks (wks) post transplantation. Reconstitution by donor-derived cells (CD45.1⁺) was monitored at 4, 8 and 12 wks by flow cytometry. Data are the mean (%) of donor-derived CD45.1⁺CD45.2⁻ cells among PBMCs±s.d. (n=4) (P value, Student's t-test.) (e) Representative flow cytometric analyses of donor-derived CD45.1⁺CD45.2⁻cells among the PBMCs at 8 wks in d stained with anti-CD45.1 and anti-CD45.2 mAb. Cells from littermate C57BL6 (CD45.2) mice were stained as a negative control.

Figure 6. Inhibition of dipeptide uptake in combination with TKI therapy eradicates CML stem cells in vivo.

(a) Relative SLC15A2 mRNA expression in human CML patients as determined by microarray analysis of nine healthy donors and nine CML patients pre- and post IM therapy. Data are from a public database (GEO, GSE33075) (P value, one-sided t-test). (b) Metabolomic analyses of dipeptides in CML-KLS⁺ cells from CML-affected mice that received either vehicle (male, n=2; female, n=4; three experiments) or IM (100 mg kg⁻¹ per day; male, n=2; female, n=6; three experiments) for 30 days. Amounts of dipeptides were plotted in whisker boxes. Cross, mean value; horizontal line across box, median value; error bars, maximum and minimum of distribution (P value, vehicle versus Cefa; Welch's t-test). (c) Quantification of colony-forming capacity of LT-CML stem cells (3% O₂) with either vehicle (−) or 5 μM Cefa (+) for 5 days in the absence or presence of 1 μM IM. Data are the mean colony number±s.d. (n=3) and are representative of three experiments (P value, Student's t-test). (d) Survival rates of CML-affected mice (female) that received vehicle, IM (100 mg kg⁻¹ per day) and/or Cefa (36 mg kg⁻¹ per day) for days 8–90 post transplantation. Results shown are cumulative data obtained from three independent experiments. Statistical differences were determined using the log-rank non-parametric test. (e,f) Eradication of CML stem cells in vivo by cefadroxil. CML-affected mice received vehicle (−), IM and/or Cefa (+) daily for 30 days post transplantation as in d. (e) Mean frequency±s.d. of GFP/BCR-ABL1⁺CML-KLS⁺cells among total GFP/BCR-ABL1⁺CML cells (n=3) (P value, Student's t-test). (f) Survival rate of new recipient mice (female) that received serial transplantation of GFP/BCR-ABL1⁺CML-KLS⁺cells (3 × 10⁴ cells) from the CML-affected mice in e that had been treated with vehicle (female, n=4) or Cefa (female, n=5). Mouse survival was monitored for 90 days (P value, log-rank non-parametric test). (g,h) Quantification of colony-forming capacity of human CD34⁺CD38⁻Lin⁻CML-LICs that were treated in vitro: (g) with either vehicle (−) or 5 μM Cefa (+) for 5 days, or (h) with vehicle (−) or 5 μM Cefa (+) in the absence or presence of 1 μM IM or 500 nM dasatinib (Dasa) for 3 days. Data shown are the mean colony number±s.d. (n=3) (P value, Student's t-test).

Figure 7. p38MAPK inhibitor depletes CML in vivo.

(a) Quantification of colony-forming capacity of murine LT-CML stem cells treated with DMSO (Cont), the p38MAPK inhibitors Ly2228820 (5 μM), VX-702 (5 μM) or BIRB796 (5 μM) in the absence or presence of IM (1 μM) on OP-9 stromal cells (3% O₂) for 5 days. Data are the mean colony number±s.d. (n=3) (P value, Student's t-test). (b) Quantification of colony-forming capacity of human CML-LICs treated with DMSO (Cont) or Ly2228820 (5 μM) in the absence or presence of dasatinib (500 nM) on OP-9 stromal cells (3% O₂) for 3 days. Data are the mean colony number±s.d. (n=3) (P value, Student's t-test). (c) Survival curve of CML-affected mice (BCR-ABL1-CML-affected mice) receiving Ly2228820 alone. CML-affected mice received vehicle (female, n=16) or Ly2228820 (2.5 mg kg⁻¹ every 3rd day; female, n=16). Mouse survival was monitored for up to 60 days. Results shown are cumulative data obtained from two independent experiments (P value, log-rank non-parametric test). (d) Survival curve of tet-inducible CML-affected mice receiving Ly2228820 plus dasatinib. At 1 day post DOX withdrawal, tet-inducible CML-affected mice received either vehicle or dasatinib (5 mg kg⁻¹ per day). At 8 days post DOX withdrawal, these animals received additional vehicle alone (male, n=15; female, n=16), dasatinib plus vehicle (male, n=19; female, n=16) or dasatinib plus Ly2228820 (2.5 mg kg⁻¹ every 3rd day; male, n=12; female, n=19). Survival was monitored for up to 30 days. Results shown are cumulative data obtained from five independent experiments (P value, log-rank non-parametric test).

Figure 8. Therapeutic approach based on disruption of a nutrient supply essential for CML stem cells.

Diagram outlining the proposed role of dipeptide uptake in LT-CML stem cell maintenance. Dipeptide species internalized by the Slc15a2 dipeptide transporter may initiate nutrient signalling activating p38MAPK. This p38MAPK activation mediates non-canonical Smad3–Ser208 phosphorylation that allows the LT-CML stem cells to activate Foxo3a and is critical for LT-CML stem cell maintenance. Because this mechanism is not important for normal HSCs, interference with dipeptide uptake nutrient signalling may represent a novel therapeutic approach for CML patients.