Low-dose insulin-like growth factor binding proteins 1 and 2 and angiopoietin-like protein 3 coordinately stimulate ex vivo expansion of human umbilical cord blood hematopoietic stem cells as assayed in NOD/SCID gamma null mice

Abstract

Introduction: Insulin-like growth factors (IGFs), IGF binding proteins (IGFBPs) and angiopoietin-like proteins (ANGPTLs) can enhance the ex vivo expansion of hematopoietic stem cells (HSCs) when used with a standard cytokine cocktail of stem cell factor (SCF), thrombopoietin (TPO) and FLT3 ligand (FL). In order to determine the optimal dose and combination of IGFs, IGFBPs and ANGPTLs, serial dilution and full permutation of IGFBP1, IGFBP2, IGF2 and ANGPTL3 were applied on a cryopreserved umbilical cord blood mononuclear cell (UCB-MNC) ex vivo expansion system.

Methods: In this system, 4 × 105 cells/ml of UCB-MNCs were inoculated in serum-free Stemspan® medium (Stemcell technologies, vancouver, BC, Canada) supplied with standard basal cytokine combination of 100 ng/ml SCF, 50 ng/ml FL and 100 ng/ml TPO and supported by a bone marrow mesenchymal stromal cell layer.

Results: Paradoxically, experiment results showed that the highest expansion of CD34+CD38-CD90+ primitive progenitor was stimulated by cytokine combination of SCF + TPO + FL + IGFBP1 + IGFBP2 + ANGPTL3 at a low dose of 15 ng/ml IGFBP1 and 20 ng/ml IGFBP2 and ANGPTL3. This ex vivo expansion was further validated in 8-week-old to 10-week-old nonobese diabetic/severe combined immunodeficiency interleukin 2 gamma chain null (NOD/SCID-IL2Rγ-/-) mice. Limiting dilution assay showed excellent correlation between the HSC ex vivo surface marker of CD34+CD38-CD90+ and the in vivo competitive repopulating unit (CRU) functional assay.

Conclusion: IGFBP1, IGFBP2, IGF2 and ANGPTL3 can stimulate the expansion of CD34+CD38-CD90+ primitive progenitor at low dose. The optimal combination comprises IGFBP1, IGFBP2 and ANGPTL3 together with the standard cytokine cocktail of SCF, FL and TPO. The CD34+CD38-CD90+ phenotype can serve as a surrogate ex vivo surface marker for HSCs due to consistency with the in vivo CRU functional assay.

Figure 1

IGFBP1, IGFBP2, IGF2 and ANGPTL3 stimulate the ex vivo expansion of CD34⁺CD38⁻CD90⁺ primitive progenitors at low dose. Post-thaw umbilical cord blood cells (4 × 10⁵ cells/ml) were suspended in serum-free Stemspan® media (Stemcell technologies, vancouver, BC, Canada) supplied with different doses of indicated cytokines in the range 0 to 200 ng/ml and standard cytokine combinations of 100 ng/ml stem cell factor, 50 ng/ml FLT3 ligand and 100 ng/ml thrombopoietin, inoculated on a passage 3 to 5 bone marrow-derived mesenchymal stromal cell layer in a 24-well plate and cultured for 12 days. (A) Expansion of viable total nucleated cells (TNCs). (B) Expansion of CD34⁺ cells. (C) Expansion of granulocyte–macrophage colony-forming units (CFU-GM). (D) Expansion of CD34⁺CD38⁻CD90⁺ cells. Results expressed as mean ± standard deviation. For each novel cytokine, a t test was performed between the variant cytokine dosage group and baseline (standard cytokine combination). **P < 0.01. ANGPTL, angiopoietin-like protein; IGF, insulin-like growth factor; IGFBP, insulin-like growth factor binding protein.

Figure 2

Optimal dose of IGFBP1, IGFBP2, IGF2 and ANGPTL3 was established from 10 to 20 ng/ml. Post-thaw umbilical cord blood cells (4 × 10⁵ cells/ml) were suspended in serum-free Stemspan® media (Stemcell technologies, vancouver, BC, Canada) supplied with different doses of indicated cytokines in the range 0 to 50 ng/ml and standard cytokine combinations of 100 ng/ml stem cell factor, 50 ng/ml FLT3 ligand and 100 ng/ml thrombopoietin, inoculated on a passage 3 to 5 bone marrow-derived mesenchymal stromal cell layer in a 24-well plate and cultured for 12 days. (A) Expansion of viable total nucleated cells (TNCs). (B) Expansion of CD34⁺ cells. (C) Expansion of granulocyte–macrophage colony-forming units (CFU-GM). (D) Expansion of CD34⁺CD38⁻CD90⁺ cells. Results expressed as mean ± standard deviation. For each novel cytokine, the t test was performed between the variant cytokine dosage group and baseline (standard cytokine combination). **P < 0.01. ANGPTL, angiopoietin-like protein; IGF, insulin-like growth factor; IGFBP, insulin-like growth factor binding protein.

Figure 3

‘SCF + TPO + FL + IGFBP1 + IGFBP2 + ANGPTL3’ is the optimal cytokine combination to enhance ex vivo expansion of CD34⁺CD38⁻CD90⁺ primitive progenitor cells. Post-thaw umbilical cord blood cells (4 × 10⁵ cells/ml) were suspended in serum-free Stemspan® media (Stemcell technologies, vancouver, BC, Canada) supplied with different cytokine combinations with optimal doses and standard cytokine combinations of 100 ng/ml stem cell factor (S), 50 ng/ml FLT3 ligand (F) and 100 ng/ml thrombopoietin (T), inoculated on a passage 3 to 5 bone marrow-derived mesenchymal stromal cell layer in a 24-well plate and cultured for 12 days. Here, the optimal dose of IGFBP1 (A), IGFBP2 (B), IGF2 (C) and ANGPTL3 (D) was 15 ng/ml, 20 ng/ml, 10 ng/ml and 20 ng/ml respectively. (A) Expansion of viable total nucleated cells (TNCs). (B) Expansion of CD34⁺ cells. (C) Expansion of granulocyte–macrophage colony-forming units (CFU-GM). (D) Expansion of CD34⁺CD38⁻CD90⁺ cells. Results expressed as mean ± standard error (cord blood unit number, n = 6). ANGPTL, angiopoietin-like protein; IGF, insulin-like growth factor; IGFBP, insulin-like growth factor binding protein.

Figure 4

Human cell engraftment and multi-lineage reconstitution profile. (A) Amount of human chimerism in the bone marrow of NOD/SCID-IL2Rγ^−/−mice that received a transplant of 5 × 10⁵ and 1 × 10⁶ unexpanded human mononuclear cord blood cells and 1 × 10⁶ and 2 × 10⁶ expanded progeny cells. Each symbol represents the engraftment of a single mouse that underwent transplantation assayed at 4 months post transplantation. (B) Summary of multi-lineage reconstitution from mice bone marrow transplanted 5 × 10⁵ unexpanded cells in lane 1 and transplanted 2 × 10⁶ expanded cells with the cytokine combination ‘S + F + T + ABD’ in lane 10 of (A). S, stem cell factor; F, FLT3 ligand; T, thrombopoietin. A, IGFBP1; B, IGFBP2; C, IGF2; D, ANGPTL3. ANGPTL, angiopoietin-like protein; IGF, insulin-like growth factor; IGFBP, insulin-like growth factor binding protein.

Figure 5

Cytokine combination of ‘SCF + TPO + FL + IGFBP1 + IGFBP2 + ANGPTL3’ stimulates highest ex vivo expansion of primitive progenitors as assayed in NOD/SCID-IL2Rγ^−/−mice. (A) Limiting dilution assay for the unexpanded and expanded cells. Negative engraftment was defined by less than 0.5% of human CD45⁺ cell engraftment in mice bone marrow (P < 0.05). (B)Ex vivo expansion of umbilical cord blood (UCB) cells. Post-thaw UCB cells (4 × 10⁵ cells/ml) were suspended in serum-free Stemspan® media (Stemcell technologies, vancouver, BC, Canada) supplied with cytokine combinations of ‘S + T + F’,’ S + T + F + B’,’ S + T + F + BCD’ and ‘S + T + F + ABD’ respectively, inoculated on a passage 3 to 5 bone marrow-derived mesenchymal stromal cell layer in a T₁₇₅ flask and cultured for 12 days. Here, the concentration of stem cell factor (S), FLT3 ligand (F), thrombopoietin (T), IGFBP1 (A), IGFBP2 (B), IGF2 (C) and ANGPTL3 (D) was 100 ng/ml, 50 ng/ml, 100 ng/ml, 15 ng/ml, 20 ng/ml, 10 ng/ml and 20 ng/ml respectively. (C) Correlation of in vivo competitive repopulating unit (CRU) functional assay and ex vivo CD34⁺CD38⁻CD90⁺ cell surface marker. Results expressed as mean ± standard deviation. For multiple comparisons, Bonferroni’s test was used to correct the P value for the t test. P < 0.05 → P_corrected < 0.017 and P < 0.01 → P_corrected < 0.003 when n = 3. *P < 0.05, **P < 0.01. ANGPTL, angiopoietin-like protein; CFU-GM, granulocyte–macrophage colony-forming units; IGF, insulin-like growth factor; IGFBP, insulin-like growth factor binding protein; TNC, total nucleated cell.