Reduction in levels of the cyclin-dependent kinase inhibitor p27(kip-1) coupled with transforming growth factor beta neutralization induces cell-cycle entry and increases retroviral transduction of primitive human hematopoietic cells

Abstract

Successful gene therapy depends on stable transduction of hematopoietic stem cells. Target cells must cycle to allow integration of Moloney-based retroviral vectors, yet hematopoietic stem cells are quiescent. Cells can be held in quiescence by intracellular cyclin-dependent kinase inhibitors. The cyclin-dependent kinase inhibitor p15(INK4B) blocks association of cyclin-dependent kinase (CDK)4/cyclin D and p27(kip-1) blocks activity of CDK2/cyclin A and CDK2/cyclin E, complexes that are mandatory for cell-cycle progression. Antibody neutralization of beta transforming growth factor (TGFbeta) in serum-free medium decreased levels of p15(INK4B) and increased colony formation and retroviral-mediated transduction of primary human CD34(+) cells. Although TGFbeta neutralization increased colony formation from more primitive, noncycling hematopoietic progenitors, no increase in M-phase-dependent, retroviral-mediated transduction was observed. Transduction of the primitive cells was augmented by culture in the presence of antisense oligonucleotides to p27(kip-1) coupled with TGFbeta-neutralizing antibodies. The transduced cells engrafted immune-deficient mice with no alteration in human hematopoietic lineage development. We conclude that neutralization of TGFbeta, plus reduction in levels of the cyclin-dependent kinase inhibitor p27, allows transduction of primitive and quiescent hematopoietic progenitor populations.

Figure 1

Immunoblotting analysis of CD34⁺ cells from normal human bone marrow cultured with anti-TGFβ Ab showed a selective reduction in the levels of the CKI p15. Marrow-derived CD34⁺ cells were cultured in serum-free medium with IL-3, IL-6, and SCF for 12 hr, with addition of TGFβ or neutralizing Ab to TGFβ. After 12 hr, cells were collected and immunoblot analysis was done by using Abs against p15 and ERK2. Lane 1, medium alone; lane 2, medium + 5 ng/ml TGFβ; and lane 3, medium + anti-TGFβ Ab.

Figure 2

The combination of anti-TGFβ and antisense ON to p27^kip-1 increased the percentage of 4-HC-resistant CD34⁺ cells exiting G₀. 4-HC-resistant CD34⁺ cells were incubated in serum-free medium with cytokines, ±anti-TGFβ Ab, ±antisense ON to p27 for 18 hr. (A) Levels of p27 protein were tested by immunoblotting. Lane 1, cells incubated in medium alone; lane 2, medium + anti-TGFβ Ab; lane 3, medium plus anti-p27 ON; and lane 4, medium with anti-TGFβ Ab + anti-p27 ON. (B) Levels of the Ki67 protein were assessed by immunohistochemistry. (1) Cells incubated with anti-TGFβ Ab alone. (×8.5.). (2) Cells incubated with anti-TGFβ Ab + anti-p27 ON (×8.5.). (3) Cells incubated with anti-TGFβ Ab alone (×85.). (4) Cells incubated with anti-TGFβ Ab + anti-p27 ON (×85.).

Figure 3

Addition of anti-TGFβ and antisense ON to p27^kip-1 increased both total and G418-resistant CFU production from 4-HC-resistant CD34⁺ cells. Cells were incubated for 12 hr in serum-free medium with cytokines on FN, with and without anti-TGFβ Ab, and with and without addition of missense or anti-p27 ON. After the preincubation period, supernatant from the PG13/LN cell line was added, and the cells were incubated for an additional 12 hr. A colony-forming assay was plated ±G418, and CFU were counted on day 21. The average of eight experiments is shown.

Figure 4

Cyclin A levels are increased in primary human CD34⁺ cells cultured with anti-TGFβ Ab and anti-p27 ONs. Marrow-derived CD34⁺ cells were cultured in serum-free medium with IL-3, IL-6, and SCF for 16 hr, with and without anti-TGFβ Ab and anti-p27 ON. Cells were collected and lysed, and the lysates were run on SDS/PAGE, transferred to a membrane, and incubated with anti-p27 and anti-cyclin A Abs. The level of total cellular protein in each lane was assessed by immunoblotting with an anti-actin Ab.

Figure 5

PCR for the neo gene in tissues from recipients of human bone marrow-derived CD34⁺/38⁻ cells subjected to a 24 hr ex vivo transduction period with and without anti-p27 ON. Human CD34⁺/38⁻ cells from human bone marrow were precultured on FN in serum-free medium with (A) cytokines, anti-TGFβ Abs, and antisense p27 ON, or (B) cytokines, anti-TGFβ Abs, and missense p27 ON for 12 hr. For each group, 1 and 2 refer to individual mice. LN vector-containing supernatant was then added for an additional 12 hr. The cells were transplanted into immune-deficient mice. Tissues were harvested 12 mos after transplantation, genomic DNA was isolated, and PCR for the neo gene carried by the LN vector was done. BM, bone marrow; K, kidney; Li, liver; Lu, lung; Sp, spleen.

Figure 6

Clonal integration analysis of individual human T cell clones analyzed by inverse PCR. Individual human T lymphoid (CD45⁺/3⁺) cells recovered from a mouse transplanted with human CD34⁺/38⁻ cells transduced for 24 hr in the presence of anti-TGFβ Abs and anti-p27 ON were sorted by automated cell deposition into 96-well plates. Clones were stimulated with nontransduced, irradiated allogeneic feeder cells, phytohemagglutinin, and recombinant human IL-2, as described (25). Human T lymphocytes were expanded individually, then genomic DNA was isolated, and a number of the clones were demonstrated by PCR to contain the neo gene. Clonal integration analysis was then performed (2, 25). Nine unique patterns are shown, indicating that nine lymphoid progenitors capable of long-term reconstitution in the mice were initially transduced within the CD34⁺/38⁻ population.