Retrovirus vectors and in vitro expansion of transduced Fancc^−/− BM cells. (A) Retroviral vectors used in this study. SF indicates enhancer/promoter from the spleen focus forming virus long terminal repeat; FANCC, cDNA encoding the human FA complementation group C protein; HOXB4, cDNA encoding the human homeobox cluster B paralog group 4 protein; IRES, the internal ribososome entry site from encephalomyocarditis virus; eGFP, enhanced green fluorescent protein; Venus, modified version of yellow fluorescent protein; Pre, truncated version of the woodchuck hepatitis virus posttranscriptional regulatory element, which lacks any X protein coding sequence; Ψ, packaging signal. (B) Preservation of colony forming activity after in vitro culture. C57B6 LSK BM cells were transduced twice with either SF-HOXB4-IG or SF-IG at an MOI of 6 and cultured for an additional 14 days. The initial transduction frequency was 48.2% plus or minus 0.6% for SF-HOXB4-IG versus 55.1% plus or minus 1.5% for SF-IG–transduced LSK cells based on fluorescence at day 2 after transduction. BM was then plated out in methylcellulose, as described in “Colony-forming assays” and colonies were scored 7 days later. Data represent the mean of 3 independent experiments plus or minus SEM. □ represents SF-IG; ■, SF-HOXB4-IG. **P < .01 compared with SF-IG–transduced group by Student t test. (C) In vitro expansion of BM cells transduced with SF-HOXB4-IG. LSK BM cells from either C57B6 or Fancc^−/− mice were transduced with either SF-HOXB4-IG or SF-IG and were then cultured for an additional 14 days. The percentage of eGFP⁺ cells in culture was determined at days 4, 9, and 14 after transduction. Data represent the mean of 5 or 6 independent experiments plus or minus SEM. □ represents C57B6 BM; Δ, Fancc^−/− BM; —, SF-IG; , SF-HOXB4-IG. **P < .01 compared with corresponding SF-IG transduced group by Student t test.

Effect of HOXB4 expression on engraftment, survival, and ROS. (A) Schematic representation of competitive repopulation assay used to assess relative engraftment potential of gene-modified HSC populations. (B) Competitive repopulation of transduced BM as measured by the percentage fluorescent-positive cells in peripheral blood. At the indicated time points after transplantation, the contribution of gene modified cells to the production of peripheral blood leukocytes was analyzed by flow analysis. Percentages: □ represents Fancc^−/− + SF-IG; , Fancc^−/− + SF-FAC-IG; ■, Fancc^−/− + SF-HOXB4-IV; ▨, Fancc^−/− + SF-HOXB4-IV + SF-FAC-IG; , WT + SF-IG–transduced cells in the peripheral blood. Data represent the mean plus or minus SEM of 2 independent experiments incorporating 5 to 14 mice per experimental group. *P < .05, **P < .01 compared with Fancc^−/− + SF-IG group (Wilcoxon rank sum test). NS indicates not significant. (C,D) Apoptosis and ROS levels in transduced BM. LSK BM from C57B6 or Fancc^−/− mice were transduced with either SF-IV or SF-HOXB4-IV. At 2 days after transduction, BM was stained with antibodies directed against lineage markers, c-Kit and Sca-1, in addition to staining with either (C) annexin V or (D) H₂DCFDA. Data represent the least squares mean estimates plus or minus SEM from mixed-effects model analysis of 5 or 6 independent experiments. □ represents Fancc^−/− + SF-IV; ■, Fancc^−/− + SF-HOXB4-IV; , WT + SF-IV; and ▩, WT + SF-HOXB4-IV. *Significant after multiple comparison adjustment (P < .05) by Tukey method.

Determination of TNF-α receptor expression levels in transduced LSK BM by flow. Transduced murine BM cells were isolated by flow sort and were cultured for 7 days. Cells were then stained with antibodies directed against lineage markers, c-Kit, Sca-1, and either TNFR1 or TNFR2. (A) The mean percentage of TNFR1⁺ or TNFR2⁺ LSK cells and (B) the mean fluorescent intensity of staining for TNFR1 or TNFR2 within different retroviral transduced populations. Data represent the least squares mean estimates plus or minus SEM from mixed-effects model analysis of 3 or 4 independent experiments. □ represents Fancc^−/− + SF-IG; , Fancc^−/− + SF-FAC-IG; ■, Fancc^−/− + SF-HOXB4-IG; , WT + SF-IG. *Significant after multiple comparison adjustment (P < .05) by Tukey method. (C,D) 32D cells were transduced with either SF-IV or SF-HOXB4-IV, and Venus⁺ cells were isolated by flow sorting 2 days after transduction and expanded for 4 days in vitro. Expanded cells were harvested and treated either with 20 ng/mL TNF-α (+) or with an equivalent volume of PBS (−). After 5 minutes of incubation at 37°C, cells were harvested and lysed. The lysate from 2 × 10⁵ cells was probed by immunoblot using an antibody specific for (C) NF-κB p65 phosphorylated at the Ser536 residue or (D) SAPK/JNK phosphorylated at the Thr183/Tyr185 residues. Membranes were then sequentially stripped of antibody and reprobed with antibodies against either total NF-κB p65 and then β-actin; or total SAPK/JNK and then β-actin.

Ectopic HOXB4 protects HSC/P from the inhibitory effects of TNF-α. (A) Survival of transduced progenitor cells with increasing concentrations of TNF-α. Murine LSK cells were transduced with the indicated retroviral vectors, and sorted transduced cells were then plated in methylcellulose supplemented with 1 ng/mL, 10 ng/mL, or no (NTX) TNF-α. The frequency of surviving day 7 colonies, expressed as a percentage of nontreated controls, is shown. Data represent the least squares mean estimates plus or minus SEM from mixed-effects model analysis of 3 or 4 independent experiments. □ represents Fancc^−/− + SF-IG; , Fancc^−/− + SF-FAC-IG; ■, Fancc^−/− + SF-HOXB4-IG; , WT + SF-IG. *Significant compared with Fancc^−/− + SF-IG group or as indicated by bars after multiple comparison adjustment (P < .003) by Bonferroni method. NS indicates not significant. (B,C) Immunophenotypic analysis of percentage HSC/P in culture after TNF-α treatment. LSK BM from C57B6 or Fancc^−/− mice were transduced with either SF-IV or SF-HOXB4-IV. At 36 hours after transduction, cells were treated with the indicated dose of TNF-α as described in “Analysis of G₂/M arrest and FAN CD2 mono-ubiquitination in human FA lymphoblast cell lines” and a further 24 hours later BM was stained with antibodies directed against lineage markers, c-Kit and Sca-1. The percentage of BM cells corresponding to either (B) lin⁻ c-Kit⁺ or (C) lin⁻, c-Kit⁺, Sca1⁺ is shown. Data represent the least squares mean estimates plus or minus SEM from mixed-effects model analysis of 5 or 6 independent experiments. □ represents Fancc^−/− + SF-IV; ■, Fancc^−/− + SF-HOXB4-IV; , WT + SF-IV; ▩, WT + SF-HOXB4-IV. *Significant compared with nontreated control after multiple comparison adjustment (P < .05) by Tukey method.

Characterization of the FA cellular phenotype in transduced hematopoietic cells. (A) G₂/M arrest of transduced LCLs after treatment with melphalan. The human FANCC^−/− lymphoblast cell line HSC536-FAC was transduced with the indicated retroviral vectors. The transduced cells were then treated with melphalan as described in “Methods analysis of G2/M arrest and FANCD2 monoubiquitination in human LCLs.” The proportion of eGFP⁺ cells in G₂/M after melphalan treatment is indicated. NTX indicates nontreated. ▩ represents G₀/G₁; ▨, S; and ■, G₂/M. (B) FANCD2 mono-ubiquitination in transduced LCLs after treatment with hydroxyurea. HSC526-FAC cells were transduced with the indicated retroviral vectors and were then either subject to treatment with hydroxyurea (HU) before cell lysis (+) or were directly lysed without treatment (−). Immunoblot was then performed to visualize the relative abundance of long-form mono-ubiquitinated FANCD2 (L) and short-form nonubiquitinated FANCD2 (S).

Inducible activation of HOXB4 in vivo. (A) Retroviral vector used in this study. ERT indicates mutated ligand binding domain of murine estrogen receptor-α. (B) Schematic representation of competitive repopulation assay used to assess relative engraftment potential of gene-modified HSC populations in the presence or absence of active HOXB4 as induced by tamoxifen citrate treatment. (C) Competitive repopulation of transduced BM as measured by the percentage of fluorescent-positive cells in peripheral blood. At the indicated time points after transplantation, the contribution of gene modified cells to the production of peripheral blood leukocytes was analyzed by flow analysis. The mean percentage of WT + SF-HOXB4-ERT-IG–transduced cells in the peripheral blood of transplanted recipients is shown plus or minus SEM. ■ represents recipients of 100 μg/mL tamoxifen citrate in drinking water; and □, nonrecipients of tamoxifen citrate. **P < .01, Student t test; n = 5 mice per group.