Conditionally immortalized mouse embryonic fibroblasts retain proliferative activity without compromising multipotent differentiation potential

Abstract

Mesenchymal stem cells (MSCs) are multipotent cells which reside in many tissues and can give rise to multiple lineages including bone, cartilage and adipose. Although MSCs have attracted significant attention for basic and translational research, primary MSCs have limited life span in culture which hampers MSCs' broader applications. Here, we investigate if mouse mesenchymal progenitors can be conditionally immortalized with SV40 large T antigen and maintain long-term cell proliferation without compromising their multipotency. Using the system which expresses SV40 large T antigen flanked with Cre/loxP sites, we demonstrate that mouse embryonic fibroblasts (MEFs) can be efficiently immortalized by SV40 large T antigen. The conditionally immortalized MEFs (iMEFs) exhibit an enhanced proliferative activity and maintain long-term cell proliferation, which can be reversed by Cre recombinase. The iMEFs express most MSC markers and retain multipotency as they can differentiate into osteogenic, chondrogenic and adipogenic lineages under appropriate differentiation conditions in vitro and in vivo. The removal of SV40 large T reduces the differentiation potential of iMEFs possibly due to the decreased progenitor expansion. Furthermore, the iMEFs are apparently not tumorigenic when they are subcutaneously injected into athymic nude mice. Thus, the conditionally immortalized iMEFs not only maintain long-term cell proliferation but also retain the ability to differentiate into multiple lineages. Our results suggest that the reversible immortalization strategy using SV40 large T antigen may be an efficient and safe approach to establishing long-term cell culture of primary mesenchymal progenitors for basic and translational research, as well as for potential clinical applications.

Figure 1. Morphology of primary and immortalized mouse embryonic fibroblasts (MEFs).

(A) Schematic representation of the reversible immortalization vector SSR #69 . This retroviral vector contains the hygromycin and SV40 T antigen expression cassette flanked by loxP sites. (B) Morphology of primary MEFs in cell culture. Primary MEFs were seeded at 20–30% confluence and passed consecutively for five passages (P5). (C) Morphology of the reversibly immortalized MEFs (iMEFs) in cell culture. The iMEF cells were seeded at low density and passed consecutively for 15 passages (P15).

Figure 2. The iMEFs exhibit higher proliferative activity than that of primary MEFs.

(A) and (B) Cell viability and proliferation assay by crystal violet staining assay. The same number of primary MEF and iMEF cells was seeded at a low density. Cells were stained with crystal violet at the indicated time points (A) and the viable and stained cells were dissolved for OD reading as previously described (B) . (C) Cell proliferation assessed with MTT assay. The same cell number of primary MEFs and iMEFs was seeded at a low density. Cells were collected for MTT assay at the indicated time points. (D) Cell counting assay. The same number of primary MEF and iMEF cells was seeded at 20% confluence. Cells were trypsinized, stained with Trypan blue, and counted at the indicated time points. For all of the above assays, each assay condition was done in triplicate. The assays were repeated in at least two independent batches. Representative results are shown.

Figure 3. The iMEFs express most mesenchymal stem cell markers.

The iMEF cells were seeded at subconfluency for 24 h and stained with MSC marker antibodies as indicated. The consensus MSC markers include CD44, CD90/Thy-1, CD73, CD105/Endoglin and CD166/ALCAM . Cell nuclei were stained with DAPI. Respective IgG Isotypes were used as immunostaining control.

Figure 4. Induction of osteogenic, chondrogenic, and adipogenic lineage markers in iMEFs.

(A) Expression of lineage-specific regulators in iMEFs stimulated by BMP9. Subconfluent iMEF cells were infected with AdBMP9 or AdGFP. Total RNA was isolated at the indicated time points and subjected to RT-PCR reactions. The cDNA products were used as templates for semi-quantitative amplification of mouse Runx2, Sox9 and PPARγ2 transcripts. All samples were normalized with GAPDH expression levels. (B) and (C) Induction of early osteogenic marker alkaline phosphatase (ALP) in primary MEFs and iMEFs. Subconfluent primary MEFs and iMEFs were infected with AdBMP9 or AdGFP. ALP activity was histochemically stained on day 5 (B) or quantitatively determined at days 3, 5 and 7 (C). The ALP activity was normalized with total cellular protein (TCP) (C). (D) Matrix mineralization assessed with Alizarin Red S staining. Subconfluent MEFs and iMEFs were infected with AdBMP9 or AdGFP for 14 days. Cells were fixed and stained with Alizarin Red S. (E) Adipogenic differentiation assessed with Oil Red O staining. Subconfluent iMEFs were infected with AdBMP9, AdPPARγ2, or AdGFP for 10 days. Cells were fixed and stained with Oil Red O staining (panels a and b), or stained for ALP activity, followed by Oil-Red O staining (panels c and d). (F) Osteogenic and adipogenic differentiation of immortalized human bone marrow stromal stem cells. Subconfluent cells were infected with AdBMP9 or AdGFP. ALP staining was carried out at day 7 (a) while Oil-red O staining was done at day 14 (b). Each assay condition was done in triplicate. The assays were repeated in at least two independent batches. Representative results are shown.

Figure 5. The proliferation properties of the iMEFs can be reversed by Cre.

(A) Efficient transduction of iMEFs by adenoviral vectors. Subconfluent iMEFs were infected with AdCre (a) or AdGFP (b) for 24 h. GFP signal was recorded under fluorescence microscopy. (c) Cre-mediated efficient removal of SV40 T Ag detected by Western blotting. Subconfluent iMEFs were infected with AdCre or AdGFP for 36 h. Cells were lysed, and cell lysate was subjected to Western blotting using anti-SV40 T Ag antibody. Anti-ß-actin Western blotting confirms equal loading of the samples. (B) Cell proliferation by viable cell counting. AdCre or AdGFP-transduced iMEFs were counted at indicated time points as described in Figure 2D . (C) Cell proliferation assessed by crystal violet staining. AdCre or AdGFP-transduced iMEFs were stained with crystal violet at indicated time points as described in Figure 2A . (D) Effect on ALP activity in iMEFs by Cre. AdCre or AdGFP-transduced iMEFs were stained for ALP activity at indicated time points. (E) The effect of Cre-mediated reversal on late stage differentiation of the iMEFs. Subconfluent iMEFs were infected with AdCre or AdGFP and maintained for 14 days. Cells were processed with Alizarin Red S staining (a and b) and Oil Red-O staining (c and d), or immunostained against osteogenic late markers osteopontin (e and f) and osteocalcin (g and h). See Methods.

Figure 6. The iMEFs can effectively induce ectopic bone formation, chondrogenesis, and adipogenesis upon BMP9 stimulation in vivo; and yet the iMEFs are non-tumorigenic.

(A) The iMEFs without the removal of SV40 T Ag form larger ectopic bone masses. Subconfluent iMEFs were co-infected with BMP9, GFP and/or Cre for 16 h. Cells were collected and injected into the flanks of athymic mice subcutaneously. Bony masses were retrieved from mice after 4 weeks. No masses were formed in the cells transduced with AdGFP or AdCre alone. (B) H & E staining. The retrieved bony masses were fixed, decalcified, and subjected to H & E staining. (C) Trichrome, Alcian Blue, and Oil Red-O staining of the retrieved samples. AC, adipocyte; CC, chondrocyte; CM, chondroid matrix; MOM, mineralized osteoid matrix; OM, osteoid matrix. (D) Potential tumorigenicity of the iMEFs. The iMEFs and human osteosarcoma line 143B cells were stably labeled with firefly luciferase and injected into the flanks of athymic nude mice (2×10⁶ cells/injection). At the indicated time points, animal were anesthetized with isoflurane and injected (i.p.) with D-Luciferin sodium salt at 100 mg/kg in 0.1 ml sterile saline. Bioluminescence imaging was conducted with the Xenogen IVIS 200 imaging system. The pseudoimages were obtained by superimposing the emitted light over the gray-scale photographs of the animals. Representative results are shown.