Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Mol Ther. Author manuscript; available in PMC 2011 October 14.
Published in final edited form as:
PMCID: PMC3193345

In Vivo Gene Transfer into Adult Stem Cells in Unconditioned Mice by in Situ Delivery of a Lentiviral Vector


The potential of in vivo lentivirus-mediated bone marrow stem cell gene transfer by bone cavity injection, which could take full advantage of any source of stem cells present there, has not been previously explored. Such an approach may avoid several difficulties encountered by ex vivo hematopoietic stem cell (HSC) gene transfer. We sought to determine if efficient gene transfer could be achieved in HSC and mesenchymal stem/progenitor cells (MSC) by intrafemoral injection of a lentivirus vector in mice. Four months after injection, up to 12% GFP-expressing cells were observed in myeloid and lymphoid subpopulations. Significant transduction efficiencies were seen in Linc-kit+Sca1+ HSC/progenitors and CFU with multilineage potential, which were also confirmed by duplex PCR analysis of progenitor-derived colonies. Four months after secondary BMT, we observed 8.1 to 15% vector+ CFU in all recipients. Integration analysis by LAM-PCR demonstrated that multiple transduced clones contributed to hematopoiesis in these animals. We also showed that GFP-expressing MSC retained multilineage differentiation potential, with 2.9 to 8.8% GFP-containing CFU-fibroblasts detected in both injected and BMT recipients. Our data provide evidence that adult stem cells in bone marrow can be efficiently transduced “in situ” by in vivo vector administration without preconditioning. This approach could lead to a novel application for treatment of human diseases.

Keywords: intrafemoral injection, in vivo gene transfer, hematopoietic stem cells, lentiviral vectors, mesenchymal stem/progenitor cells, bone marrow transplantation, hematopoiesis


HIV-based lentiviral vectors (LV) were proven to be capable of transducing a broad spectrum of nondividing cells in multiple mammalian species [1]. Using local injection, LV have been shown to mediate in vivo gene transfer and sustained gene expression in brain neuronal cells [2], retina cells [3], liver hepatocytes [4], rat cardiomyocytes [4], airway epithelium [5], and kidney tissue[6]. In contrast to adenoviral vectors, no pathology that could be specifically attributed to LV administration has been observed in any of these animal studies. Efficient ex vivo transduction of purified hematopoietic stem cells (HSC) by LV has been shown by a number of investigators [7,8]; however, in vivo LV-mediated gene transfer into HSC has not been well studied. Following intravenous administration of a first-generation LV into adult mice, we found that bone marrow exhibited the highest levels of transgene among nine organs examined, with more than 10% green fluorescent protein-positive (GFP+) cells detected in peripheral blood leukocytes (PBL) in these mice [9]. It was also observed by others that a significant transgene signal was detected in the bone marrow (BM) by PCR analysis in adult mice of systemic administration of HIV-biased LV [10]. Recently, we demonstrated detectable levels of transgene (up to 3.9%) in PBL of mice 4 months after secondary bone marrow transplantation (BMT) of HSC transduced by in vivo delivery of LV in newborn pups [11]. These data strongly suggest the possibility that in vivo delivery of LV may provide an alternative approach to transducing HSC.

There is extensive clinical experience with the application of ex vivo gene transfer targeting HSC, mostly with onco-retrovirus vectors (RV). However, the frequency of gene transfer in multipotent HSC and the levels of long-term transgene expression have been variable and generally low. A few exceptions emerged when successful functional correction was demonstrated in children with severe combined immunodeficiency (ADA-SCID and X-SCID) or chronic granulomatous disease following infusion of retrovirally transduced autologous CD34+ cells [1214]. The success in these studies was related in part to the selective growth of genetically modified/corrected progenitors. This positive selection force may compensate for the relatively low numbers of transduced and successfully engrafted HSC [15]. While clinical efficacy has been achieved in these clinical trials using ex vivo gene transfer, a more general clinical application of this approach is still impeded by several difficulties. First, maintaining stem cell properties including in vivo repopulation potential during in vitro culture is a prerequisite of any successful ex vivo gene transfer approach. Second, inadequate engraftment is another unsolved obstacle that affects the outcome of ex vivo HSC gene therapy. Other issues associated with the ex vivo approach include toxicity related to HSC enrichment procedures and cytokine stimulation [15], the potential for proliferation-induced genomic damage, and the contamination risk of multistep in vitro manipulations. As an alternative, in vivo gene transfer into HSC has the potential to overcome these problems.

Bone marrow stem cells from adults have been viewed as the ideal target for gene- and cell-based therapy of genetic diseases, selected malignant diseases, and AIDS. In addition to HSC, bone marrow contains mesenchymal stem/progenitor cells (MSC), which can differentiate into mature cells of multiple mesenchymal tissues including fat, bone, and cartilage. Moreover, a population of highly “plastic” BM cells (copurified initially with HSC from adult marrow of mice, rats, and humans) has also been shown to differentiate into hematopoietic cells, as well as epithelial cells, in liver, lung, and intestine to varying degrees after intravenous injection into immunodeficient mice [16] (see for review [17]). Although controversial in their interpretation as evidence for a common pluripotent precursor cell, the data suggest that adult BM contains cells (in additional to HSC and MSC) that can differentiate into mature, nonhematopoietic cells of multiple tissues including epithelial cells of the liver, kidney, lung, skin, and GI tract and potentially also myocytes of heart and skeletal muscle. Thus, the in vivo approach to bone marrow stem cell gene transfer by intrafemoral delivery of LV studied here could potentially take full advantage of any source of stem cells present in bone cavity. It may also provide an attractive model to study the “true” nature of hematopoiesis and stem cell plasticity without the complications of in vitro manipulation or in vivo preconditioning.

The potential for in vivo RV-mediated HSC gene transfer as a clinically relevant approach has been demonstrated by intrafemoral injection in Jak3 knockout SCID mice pretreated with a sublethal dose of 5-fluorouracil (5-FU) [18]. In this study, we have taken an alternative approach by using LV in unconditioned mice, thus taking full advantage of the capacity of LV to transduce quiescent stem cells compared with RV (which requires at least one cycle of cell division). We demonstrate that efficient transduction of bone marrow HSC could be achieved by in situ delivery of a LV through intra-bone marrow (iBM) injection in mice without any preconditioning. Transgene-expressing MSC were also observed to retain multiple differentiation potential, with significant levels of GFP+ colony-forming-unit fibroblasts (CFU-F) detected in both injected and secondary BMT recipients. Vector integrants and their multiclonality were confirmed by linear amplification-mediated PCR (LAM-PCR). This approach may potentially provide a new technology for disease treatment and represents an interesting new tool to study adult stem cell plasticity and the nature of unperturbed hematopoiesis.


iBM Injection of LV Results in Significant Gene Marking in Lymphoid and Myeloid Cells, as Well as HSC, without Any Detectable Toxicity

To explore the potential of in vivo bone marrow stem cell gene transfer, we intrafemorally injected adult (12–14 weeks of age) immune-competent mice with SIN-LV-CG (in 4 mice) expressing GFP from a CMV promoter, or SIN-LV-EMiG (in 10 mice) expressing GFP in a bicistronic transcript from the elongation factor-α (EF1α) promoter, or buffer as a control (n = 9). Prior to injection, the concentrated vector stock (1–1.2 × 108 TU/ml) was shown to be free of replication-competent lentivirus by p24 ELISA after amplification on C8166 followed by indicator-phase culturing and PCR testing [19]. One control and two LV-injected (Inj0 and Inj8) mice were lost due to bleeding or mishap unrelated to vector administration. We collected organs and tissue samples from LV-injected mice (n = 5) 1 week postinjection for short-term transgene biodistribution analysis. To monitor potential hematologic toxicity, we performed complete blood counts periodically during the 4-month observation period. All blood parameters were normal except that mild monocytosis and leukocytosis (1.2- to2-fold of normal) were detected in both LV-injected (5 of 8 long-term mice) and buffer-injected (6 of 8) mice up to 1 month postinjection, suggesting acute inflammation related to the iBM injection procedure.

We observed GFP-expressing PBL in both myeloid (2.8 ± 2.2%) and lymphoid (3.0 ± 3.9%) subpopulations 3 months postinjection (Fig. 1A). Moreover, we evaluated Linc-kit+Sca1+ cells of BM from femur and tibia of the injected (for LV-EMiG) or both (for LV-CG) hind legs for GFP expression using four-color FACS analysis (Fig. 1B). Up to 4.7% (2.5 ±1.5%) of Linc-kit+Sca1+ cells expressed GFP 4 months postinjection. These results indicated LV-mediated gene transfer and gene expression in HSC/primitive progenitors by iBM injection in mice without any preconditioning.

FIG. 1
GFP+ PBL and HSC were detected for up to 4 months in primary iBM-injected mice. Data shown are representative results. (A) PBL from mice 3 months postinjection were stained with CD11b-APC for myeloid and CD3e-PE (and B220-PE) for lymphoid lineages. We ...

GFP+ CFU with Multilineage Potential Are Detected in Primary Injected Mice

To determine the efficiencies of gene transfer and transgene expression in multipotential progenitor cells, we performed colony-forming-cell capacity (CFC) assays using low-density BM derived from mice 1 week or 4 months postinjection (Fig. 2). We observed no significant difference in either CFU capacity (indicated by total CFU per 105 nucleated cells) or composition of erythroid, granulocytic, and multilineage colonies between buffer-injected and LV-injected mice. Fluorescence microscopy revealed that most of the GFP+ colonies derived from long-term mice were CFU-GM or CFU-GEMM (Fig. 2A), and 2.8–8.9% of colonies (5.5 ± 1.9%) were GFP+ in all tested long-term animals with significantly higher levels (9.4 ± 2.9%) observed in injected femurs of short-term mice (P <0.05) (Fig. 2B).

FIG. 2
GFP+ CFU were demonstrated in primitive progenitors with multilineage potential in primary injected mice. One week or 4 months after injection, CFC assays were performed using low-density BM harvested from femur and tibia of both (for LV-CG injected) ...

One of the limitations of GFP observation using fluorescence microscopy is autofluorescence of senescent colonies, biasing results with false-positive reading. To quantitate more accurately gene transfer efficiency in hematopoietic cells and to verify vector-containing colonies, we established a duplex real-time PCR assay using colonies collected from the CFC assay (Fig. 2C). We validated each colony-direct PCR by coamplification of the murine Apob gene (as quality control for DNA content and PCR amplification). We considered a colony positive for vector sequence only if it (a) yielded log-phase amplification for GFP with Ct-GFP lower than all no-template or background cellulose controls and (b) showed that the difference between Ct-Apob and Ct-GFP was less than 8. The latter criterion allowed us to eliminate the false-positive colonies due to the cross-contamination of cells from GFP+ colonies. Our data demonstrated that gene transfer frequency of up to 9.2% (6.4 ± 1.6%) was achieved in primitive progenitor cells from mice analyzed 4 months postinjection (Fig. 2D). These results further confirmed LV-mediated gene transfer and gene expression in HSC/primitive progenitors in primary injected mice.

GFP+ Cells from Mice Transduced by in Vivo Vector Delivery Gave Rise to Multipotential Progenitors in Secondary Recipient Mice

To demonstrate HSC transduction further in primary injected mice, we harvested BM of the femur and tibia of both legs from two injected mice 4 months after injection and transplanted it into four lethally irradiated secondary recipient C57BL/6 mice. Successful engraftment was demonstrated in all recipients, with 97.5–98.5% donor-derived PBL observed 4 months post-BMT (Fig. 3A). Real-time QPCR analysis revealed that variable levels of GFP+ PBL (mean of 0.52 ± 0.8%) and low-density bone marrow (0.21 ± 0.1%) were found in all 2° BMT recipients (Table 1). We performed the CFC assay and analyzed it by fluorescence microscopy and colony-direct duplex PCR. The frequency of GFP-expressing colonies in all 2° recipients ranged from 8.4 to 17.7% (mean of 13.3 ± 4.9%), with similar levels of transduction frequencies found in the paired animals receiving BM from the same donor (Fig. 3B). The phenotypes of GFP-expressing colonies comprised all lineages, including BFU-E, CFC-G/M, CFU-GM, and CFU-GEMM (Fig. 3C). When further analyzed by duplex PCR assay in randomly collected colonies, transduction frequencies were demonstrated at 8.1 to 15% (mean of 10.5 ± 3.1%) (Fig. 3D). These results suggested direct in vivo gene transfer into primitive HSC in primary iBM-injected mice.

FIG. 3
GFP+ CFU were observed in multilineage hematopoietic progenitors from 2° BMT recipients. (A) Four months after BMT, PBL were analyzed with CD45.1-PE for donor origin and CD45.2-APC for recipient origin. Data shown are representative results. (B) ...
Transduction frequency in hematopoietic and reproductive systems 4 months after iBM injection of LV-CG or 2° BMT

GFP-Expressing Cells Observed in Bone Marrow-Derived, Multipotential Stromal Cells

To evaluate the potential of in vivo gene transfer into MSC, we established long-term stromal cell cultures from BM of primary injected (with LV-CG) and 2° recipient mice. We carried the cultures for 4–5 months with 7 or 8 cell passages. We observed GFP-expressing adherent cells at low frequency (0.1–0.8% by FACS) at multiple passages derived from primary injected mice (Fig. 4A). Furthermore, we detected similar levels of transgene by real-time QPCR in MSC (at 7 or 8 passages) from injected mice (mean of 0.43 ± 0.2%, ranging from 0.21 to 0.67%) and BMT recipients (mean of 0.5 ± 0.2%, ranging from 0.24 to 0.81%) (Table 1). Immunophenotypic characterization by flow cytometry revealed that the cells were negative (<0.7% CD45+) for CD45 (panhematopoietic marker), indicating these cells are not of hematopoietic origin (Fig. 4B). They were also negative for CD117 (c-kit) (<0.4%) and low for CD34 (with 2.6–9% CD34-dim). The cells were strongly positive for Sca1 (mean of 72.9 ± 13%) (an antigen expressed on hematopoietic and mesenchymal stem/progenitor cells), CD44 (83.6 ± 7.7%) (a matrix receptor that mediates cell attachment to hyaluronan and osteopontin), and CD106 (82.9 ± 13%) (VCAM-1, i.e., vascular cell adhesion molecule-1). These results demonstrate that a relatively homogeneous, nonhematopoietic MSC population was successfully isolated and culture-expanded from adult murine BM.

FIG. 4
GFP-expressing cells were observed in long-term culture stromal cells with varied differentiation abilities. (A) GFP+ adherent cells were observed in long-term stromal culture derived from iBM-injected mice. Cells shown (from Inj2) had been in culture ...

We then evaluated the differentiation potential of the isolated MSC at passage 7 or 8. When cultured with osteogenic-inductive medium, the cellular morphology of the MSC changed from the spindle-shape to the cuboidal, and the cells formed a mineralized extracellular matrix as determined by Alizarin Red S staining (Fig. 4C). We induced chondrogenic differentiation using the micromass culture technique in which cellular condensation (a critical first event of chondrogenesis) was imitated. After 3 weeks in culture, MSC nodules were associated with an Alcian blue-positive extracellular matrix (Fig. 4D). This indicates the presence of sulfated proteoglycans, suggesting the capacity of the MSC to differentiate toward the chondrogenic lineage. When cultured with adipocyte-inductive medium, intracellular lipid droplets were noticeable by phase-contrast microscopy (Fig. 4E) and visualized by staining with Oil Red-O (Fig. 4F). When cultured in osteogenic-inductive medium, GFP+ osteoblasts exhibited mineral deposits visualized by Alizarin Red S staining (Fig. 4G). GFP-positive but Alizarin red-negative cells were also detected at the same time. These data demonstrated that stromal cells isolated in this study (including both GFP+ and GFP cells) retained their capacity to differentiate along multiple lineages.

At passage 7–9, we performed a CFU-fibroblast assay for both primary injected and 2° BMT mice, resulting in 5–21 CFU-F/100 cells (data not shown). We observed GFP-expressing CFU-F by fluorescence microscopy (Fig. 4H). Moreover, 2.9 to 8.8% of CFU-F contained GFP transgene as determined by colony-direct duplex PCR in all injected (mean of 4.7 ± 1.6%) and BMT recipients (mean of 6.4 ± 1.9%) (Fig. 4I). These data suggested the likelihood of in vivo gene transfer into mesenchymal stem/progenitor cells in mice.

Vector Integrants and Oligoclonality Confirmed by LAM-PCR in Both Primary Injected and Secondary Recipient Mice

The semi-random integration of the lentiviral provirus in the target cell genome creates a specific junctional fragment that is inherited by all progeny of the initially transduced cells. To confirm further and characterize vector integrants and their adjacent genomic DNA sequences in HSC/progenitors, we performed LAM-PCR with LV-U5-specific primers using genomic DNA from total BM of mice 4 months after primary injection or transplantation (Fig. 5). We confirmed the presence of vector sequence by strong internal vector bands in BM from both primary injected mice and 2° BMT recipients. We also detected multiple viral insertion sites in both primary and recipient mice. These data provided direct molecular evidence for in situ lentivirus-mediated gene transfer into HSC by iBM injection.

FIG. 5
Vector integrants and their multiclonality were confirmed by LAM-PCR in BM of primary injected and 2° recipients. A LAM-PCR assay was performed using 2.5 μg BM genomic DNA with the exception of 10 ng for +C. +C, genomic DNA from a NIH3T3 ...

Transgene Biodistribution in Hematopoietic System and Major Organs by Quantitative Real-Time PCR

To monitor systemic distribution of LV in multiple organs by this route of administration, we isolated genomic DNA from organs of perfused mice 1 week or 4 months after iBM injection of LV-CG (Table 1) or LV-EMiG (Table 2), followed by a real-time QPCR analysis [9,20]. This assay provided quantitation over a 5-log range and detected as low as 1 copy of transgene per 105 (i.e., 0.001% transgene frequency) genome equivalents. To monitor the inherent high risks of false positivity related to the high sensitivity of QPCR, we included mock-injected or transplanted control mice in parallel during all procedures. In short-term animals, PBL exhibited the highest transgene levels (2.7 ± 2.6%), followed by lung (2.1 ± 0.4%), heart (1.8 ± 3.0%), liver (1.3 ± 1.8%), and spleen (1.1 ± 0.7%). GFP marking in total bone marrow (TBM)-inj (0.97 ± 0.9%) was significantly higher (eightfold, P 0.063) than that from uninjected femurs (0.12 ± 0.07%), but similar to that detected in heart, liver, and spleen (P >0.65). Gonads contained low but significantly higher than background levels of transgene (0.24 ± 0.2%). Four months after injection, PBL showed the highest but extremely variable GFP marking (3.7 ± 4.0%, ranging from 0.95 to 13%), followed by TBM-inj (0.8 ± 0.5%) and MSC (0.43 ± 0.2%). Interestingly, transgene levels in uninjected femurs increased significantly (P < 0.05) compared to those in short-term mice, resulting in no significant difference observed between TBM-inj and TBM-un (P = 0.26). Brain contained the lowest, near-background, transgene levels. Moreover, gonads from injected mice exhibited undetectable (<0.0001%) to 0.055% (i.e., up to 55 in 105 cells) levels of transgene, dramatically reduced from those in short-term mice (P < 0.05%). Thus, the hematopoietic system (PBL and BM) remains the main target for in situ LV-mediated gene transfer by iBM injection, although we observed transient elevations of transgene frequencies in lung, liver, and spleen 1 week after injection.

Transgene distribution in mice 1 week or 4 months after iBM injection with LV-EMiG


We have presented data strongly indicating that HSC and MSC can be genetically modified successfully in their natural “niche” in unconditioned mice by LV-mediated in vivo gene transfer. We showed easily detectable levels of GFP+ cells in PBL (mean of 2.8% for myeloid and 3.0% for lymphoid cells) and HSC (Linc-kit+Sca1+, mean of 2.5%) of primary injected mice 4 months after injection. Gene transfer into primitive progenitors with multilineage potential was demonstrated by CFC assay (mean of 5.5%) and was also confirmed by colony-direct PCR (mean of 6.4%). In secondary recipients, we observed even higher levels (mean of 13.5% by microscopy and 10.5% by colony-direct PCR) of transduced primitive progenitors, strongly suggesting transduction of HSC in primary injected mice. Moreover, proviral integration and its multiclonality were confirmed by LAM-PCR in BM of mice 4 months after primary injection or BMT. Transgene biodistribution by real-time QPCR revealed that the hematopoietic system remained the main target for gene transfer by this route of in vivo application. In addition, we demonstrated that GFP-expressing MSC retained multi-differentiation ability after 4–5 months of in vitro expansion. Genetic marking levels were stable in fibroblast CFU derived from MSC 4 months postinjection (mean of 4.7%) or posttransplantation (6.4%), thus implicating MSC gene transfer in primary iBM-injected mice.

Ex vivo HSC/progenitor gene transfer using Moloney-based onco-RV has been applied in many gene therapy clinical trials. However, ex vivo expansion of mostly quiescent HSC with preservation of stem cell pluripotency is still problematic, even though improvements in the “cytokine cocktails” [21,22] and the transduction modalities (protocols) [2325] may better allow transduction in cycling stem cells. In addition, the potential clinical efficacy of ex vivo gene therapy largely relies on the efficiency of the genetically modified HSC to home and engraft to the specialized niches of the bone marrow microenvironment. Therefore, using in vivo HSC gene transfer, as described here, by “in situ” iBM injection of LV has the potential to overcome these hurdles by taking the advantages of both the natural HSC cycling for efficient transduction and the supportive microenvironment in the bone cavity for maintaining stem cell viability and capacities.

The usage of LV for in vivo stem cell gene transfer represents several advantages over onco-RV, although the potential of RV-mediated HSC gene transfer has been demonstrated by iBM injection of RV producer cell lines in macaques [26] and mice [27] or concentrated RV in mice preconditioned with sublethal dose of 5-FU [18]. In addition to the well-documented advantage of transducing nondividing cells, HIV-1-based LV may have several other features making them more suitable vehicles for in vivo gene transfer. First, substantial silencing of transgene expression was observed in embryonic stem cells (ES) and preimplantation embryos transduced by RV [28]. However, in similar settings, LV demonstrated successful transgene expression in both ES cells and various tissues of chimeric animals [29,30]. Second, integration site mapping studies have revealed that murine leukemia virus (MLV) preferentially integrates in and around gene promoters (± 1 kb from CpG islands), whereas HIV-1 integrates mostly within transcriptional units [31,32]. These findings may indicate that the risk factors related to insertional mutagenesis for MLV-based RV could differ from those of HIV-based LV, although additional studies are required to determine fully the relevance of these differences. Finally, the feature of “self-inactivating” LTRs in LV has been found to have significant safety advantages over the intact 3′-LTR used in most RV, which can function as a strong promoter to activate adjacent coding sequences in both directions [33]. Together with the ability of LV to concentrate and the limitation of volume applicable for iBM injection, LV appear to be more suitable for in vivo stem cell gene transfer by in situ vector administration into bone cavity. In fact, the transgene levels observed here in primary injected unconditioned mice 4 months after iBM LV injection were comparable to those found in BMT recipients transplanted with the entire content of the injected femur from 5-FU-conditioned mice 24 h after iBM injection of RV [18].

Mesenchymal stem cells have shown promise for cell and gene therapy applications, although the precise identity of the in vivo MSC remains unclear [34]. In a large number of animal transplantation or implantation studies, MSC expanded ex vivo were able to differentiate into a variety of cell types, including intervertebral disc cartilage [35], bone [36], cardiomyocytes [37], neurons [38], and epithelia [39]. In addition, they were found to support hematopoietic progenitors in vitro and possess potent immunosuppressive properties [40]. In fact, MSC have already demonstrated efficacy in several cell therapy clinical trials, including applications in treating osteogenesis imperfecta [41], bone tissue regeneration strategies [42], and improving hematopoietic recovery in cancer patients [43]. In this study, we obtained a relatively “homogeneous” population of adherent cells after 4–5 months of ex vivo MSC expansion (Fig. 4). These populations were capable of generating CFU fibroblasts, a standard assay that identifies adherent, spindle-shaped cells with proliferating potential [34]. The multilineage differentiation potential of these MSC populations was demonstrated by in vitro inductive cultures. Following inductive culture, we observed GFP+ cells with different differentiation potentials. Significantly higher than background levels of transgene were detected in the “bulk” MSC cultures from all injected and BMT recipients (Table 1). Moreover, transgene-containing CFU fibroblasts were detected in all injected (4.7 ± 1.6%) and BMT recipients (6.4 ± 1.9%). Our data strongly suggest successful transduction of MSC by intrafemoral LV injection in mice. Thus, the in vivo gene transfer of MSC demonstrated herein, together with HSC gene transfer, may provide additional benefits to treatment of a variety of malignant and genetic diseases with abnormal or damaged stroma/MSC, including lysosomal storage diseases, Fanconi anemia, Shwachman–Diamond syndrome, and osteogenesis imperfecta.

Although the route of LV administration studied here was applied in situ into the bone cavity, we still observed limited levels of systemic transgene distribution in other organs (Table 2). Transient elevation of transgene levels was found 1 week postinjection in lung, heart, liver, and spleen, which were comparable to those detected in TBM from injected femur. However, 4 months later, significant reduction in GFP (5- to 10-fold) was observed in all tested systemic organs, while transgene levels remained comparable in TBM-inj. Thus, unlike previous reports by us and others [9,44] on intravenous administration of LV when liver was one of the main targets for transduction, intrafemoral injection of LV did confine most target cells in BM. The risk of inadvertent germ-line gene transfer related to insertional mutagenesis has raised a broad array of ethical issues and safety concerns, especially with an in vivo gene transfer approach [45]. We quantitated transgene frequency in whole gonads of mature mice, with sensitivity of 1 copy in 106 cells (Tables 1 and and2).2). Undetectable to 0.055% levels of transgene were found in injected mice (n = 7). A low level of transgene (0.001–0.01% genomes) was also observed in two of the six control animals. Thus, the possibility of cross-contamination cannot be ruled out. In addition, relatively high levels of transgene (more than 4-log-fold higher than those found in gonads) detected in the PBL of injected mice may contaminate the gonad if not eliminated completely by perfusion. However, to understand fully the risk of germ-line gene transfer by iBM injection, a more thorough evaluation is needed with more animals of both genders and with higher, if possible, dosage of LV applied.

In conclusion, this study demonstrates a novel approach to LV-mediated in vivo stem cell gene transfer that may have a significant impact on disease treatment and stem cell research. First, direct gene delivery would take full advantage of any source of stem cells present in the bone cavity. Second, this approach would avoid the difficulties encountered by ex vivo HSC gene transfer, including maintaining stem cell properties and the loss of engraftment potential. Third, it would also avoid cytokine stimulation, which may activate unwanted signaling pathways that could potentially increase the risk of nonrandom mutagenic events during provirus insertion. Although the iBM injection approach may be limited by the applicable volume per injection and relatively low transduction frequency for clinical application, this study warrants the need for further development of LV-mediated in vivo stem cell gene delivery approach. It is likely that stem cell gene transfer by iBM LV injection could be improved by pretreatment with 5-fluorouracil or sublethal irradiation to reduce the numbers of HSC staying in G0 phase and total numbers of BM cells, by temporarily stopping blood flow (by tourniquet) or by successful repeated LV administration.

Materials and Methods

Generation and intrafemoral injection of LV

A self-inactivating LV containing a CMV-eGFP expression cassette (LV-CG) or EF1α-MGMT-ires-GFP cassette (LV-EMiG) was packaged by three helper plasmids: p2NRF for gagpol, pEF1.Rev for rev, and pMD.G for VSVG env function (kind gifts from Dr. Tal Kafri). Vector was generated and concentrated in a designated BL2+ facility by methods previously described [9]. The infectivity of concentrated vector stocks was determined on 293T cells and scored by FACS analysis for GFP+ cells. Less than 30% of GFP+ cells was considered reliable for calculation to contain <1 copy per target cell. Prior to injection, 0.5% of the total LV production was tested to be “free” of replication-competent lentivirus by the Vector Production Facility at Indiana University.

The intrafemoral injection of LV was accomplished by adopting a previously described intrafemoral cell transplantation procedure [46] with minor modification. In brief, after the mouse was anesthetized, the skin around the right knee joint was sterilized and the hair was shaved. Following a bone marrow aspiration, 20–40 μl of LV was injected through the joint into the right femur using a 28.5-gauge needle insulin syringe to minimize the loss of sample in the needle dead space. Appropriate analgesic (buprenorphine and ibuprofen) was used for pain management.

Immunostaining and flow cytometry analysis

Antibody staining was carried out for 15–30 min on ice with primary antibody typically at 1/50 to 1/100 dilution. Whenever possible, we used 7-aminoactinomycin D (7-AAD) staining to gate out dead cells. Lineage marker antibodies included anti-CD11b (M1/70), anti-Gr-1 (RB6-8C5), anti-B220 (RA3-6B2), anti-CD3 (145-2C11), and anti-Ter119 (Ter119). Other antibodies used in this study included anti-CD45.1 (A20), anti-CD45.2 (104), anti-CD45 (30-F11), anti-Sca-1 (E13-161.7), anti-c-kit (2B8), anti-CD34 (RAM34), anti-CD44 (IM7), and anti-CD106 (429 MVCAM.A). The cells stained with the isotype-matched immunoglobulins served as a negative control. After depletion of red blood cells with ammonium chloride lysing reagent (BD PharMingen), single-cell suspensions were analyzed by FACSCalibur with the CellQuest program.

Bone marrow transplantation

Bone marrow samples extracted by crashing femur and tibia of iBM-injected Boy/J mice (The Jackson Laboratory; Ly5.1) 4 months after injection were single-cell-suspended in Hanks balanced salt solution with 2% fetal bovine serum (FBS). Low-density mononucleated BM cells were obtained after density gradient centrifugation with Histopaque-1083 and injected through the tail vein at 5×106/mouse into a pair of lethally irradiated (with split dosage of 750 and 350 Gy) C57BL/6J (Ly5.2) congenic mice at 7–9 weeks of age.

Long-term stromal culture for MSC Isolation

Marrow-derived MSC cultures were established from BM as described [47] with subtle modifications. In brief, BM cells were cultured at 106 cells/cm2 with the residual humerus pieces in MSC medium consisting of α medium–minimum essential medium with 10% FBS, 2 mM glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin. Media were changed twice a week to remove nonadherent cells until near confluence. Adherent cells were subcultured for four passages with MSC medium and 10% BM conditioned medium, followed by an additional four passages in medium containing 10 ng/ml β-FGF.

In vitro MSC differentiation assays

Osteogenic differentiation was induced by treating cells (at passage 7 or 8) with dexamethasone and recombinant human bone marrow morphogenic protein (R&D Systems, MN, USA) for 3 weeks. To introduce adipocytic differentiation, cells were treated with dexamethasone and insulin for 2 weeks [48]. Chondrogenic differentiation was induced using a micromass technique by insulin and transforming growth factor-β (PeproTech). In addition to morphology, we utilized in situ histochemical and cytological staining to verify differentiation induction [47], including measuring extracellular matrix (ECM) mineralization (using Alizarin Reds staining) to verify osteoblasts, Oil RedO staining for intercellular lipid vacuoles to verify adipocytes, and Alcian blue ECM staining to verify chondrocytic cell nodules.

Quantitative real-time PCR and colony-direct duplex real-time PCR

We quantitated both GFP transgene and endogenous murine Apob in the same 50-μl reaction simultaneously by real-time PCR described previously with minor modification [9]. The duplex reaction contained 0.5–10 μg genomic DNA, 200 nM each GFP primer, 200 nM GFP probe, 40 nM each Apob primer, 200 nM Apob probe, and 25 μl TaqMan 2× Universal Master Mix (Applied Biosystems). Unknown samples were run in triplicate, and standard samples were in duplicate. A standard curve (ranging from 0.001 to 100%) was established from a series of genomic DNA mixtures of a murine myeloid cell line (32Dp210) with a GFP-containing cell line (32Dp210-LNChRGFP) (1 copy per genome as determined by Southern blot analysis).

After being cultured in cytokine-containing medium for 10–14 days, CFU were individually collected into PCR lysis buffer (0.9 mg/ml proteinase K, 0.5% Tween 20, 0.5% NP-40, 1× PCR Buffer; Applied Biosystems), followed by incubation at 60°C for 60 min and heat inactivation of proteinase K at 95°C for 10 min. Each 25-μl duplex reaction contained 5 μl lysed colony, 300 nM each GFP primer, 200 nM GFP probe, 80 nM each Apob primer, 200 nM Apob probe, and 1× TaqMan Universal Master Mix (Applied Biosystems). The amplification conditions were 2 min at 50°C and 10 min at 95°C for the first cycle, followed by 45 cycles of 95°C for 15 s and 60°C for 1 min.


For clonality analysis defining the patterns of lentivirus insertion sites, we performed LAM-PCR consisting of primer extension. Second-strand synthesis, restriction digestion, linker ligation, and exponential amplification were as described [49]. A monoclonal 3T3-LVCG cell line containing three insertion sites was established and used as positive control.


We are grateful for the technical assistance of Jeff Bailey, Victoria Summey-Harner, and Kimberly Bohn. We acknowledge Tal Kafri (University of North Carolina) for LV packaging plasmids and Kenneth Cornetta (Vector Facility at Indiana University) for RCL analysis. David Williams at the Cincinnati Children’s Medical Center is thanked for his helpful discussion and critical appraisal of this article. This work was supported in part by a Translational Research Initiative grant from Cincinnati Children’s Foundation (D.P.) and the National Institutes of Health (AI061703) (D.P.).


1. Kay MA, Glorioso JC, Naldini L. Viral vectors for gene therapy: the art of turning infectious agents into vehicles of therapeutics. Nat Med. 2001;7:33–40. [PubMed]
2. Blomer U, Naldini L, Kafri T, Trono D, Verma IM, Gage FH. Highly efficient and sustained gene transfer in adult neurons with a lentivirus vector. J Virol. 1997;71:6641–6649. [PMC free article] [PubMed]
3. Miyoshi H, Takahashi M, Gage FH, Verma IM. Stable and efficient gene transfer into the retina using an HIV-based lentiviral vector. Proc Natl Acad Sci USA. 1997;94:10319–10323. [PubMed]
4. Kafri T, Blomer U, Peterson DA, Gage FH, Verma IM. Sustained expression of genes delivered directly into liver and muscle by lentiviral vectors. Nat Genet. 1997;17:314–317. [PubMed]
5. Johnson LG, Olsen JC, Naldini L, Boucher RC. Pseudotyped human lentiviral vector-mediated gene transfer to airway epithelia in vivo. Gene Ther. 2000;7:568–574. [PubMed]
6. Gusella GL, Fedorova E, Hanss B, Marras D, Klotman ME, Klotman PE. Lentiviral gene transduction of kidney. Hum Gene Ther. 2002;13:407–414. [PubMed]
7. Miyoshi H, Smith KA, Mosier DE, Verma IM, Torbett BE. Transduction of human CD34+ cells that mediate long-term engraftment of NOD/SCID mice by HIV vectors. Science. 1999;283:682–686. [PubMed]
8. Case SS, et al. Stable transduction of quiescent CD34(+)CD38(−) human hematopoietic cells by HIV-1-based lentiviral vectors. Proc Natl Acad Sci USA. 1999;96:2988–2993. [PubMed]
9. Pan D, et al. Biodistribution and toxicity studies of VSVG-pseudotyped lentiviral vector after intravenous administration in mice with the observation of in vivo transduction of bone marrow. Mol Ther. 2002;6:19–29. [PubMed]
10. Follenzi A, Sabatino G, Lombardo A, Boccaccio C, Naldini L. Efficient gene delivery and targeted expression to hepatocytes in vivo by improved lentiviral vectors. Hum Gene Ther. 2002;13:243–260. [PubMed]
11. Pan D, et al. Metabolic and neuropathologic correction of Hurler syndrome by a single intravenous injection of lentiviral vector to newborn mice. Mol Ther. 2003;7:s76.
12. Hacein-Bey-Abina S, et al. Sustained correction of X-linked severe combined immunodeficiency by ex vivo gene therapy. N Engl J Med. 2002;346:1185–1193. [PubMed]
13. Aiuti A, et al. Correction of ADA-SCID by stem cell gene therapy combined with nonmyeloablative conditioning. Science. 2002;296:2410–2413. [PubMed]
14. Barese CN, Goebel WS, Dinauer MC. Gene therapy for chronic granulomatous disease. Expert Opin Biol Ther. 2004;4:1423–1434. [PubMed]
15. Baum C, et al. Side effects of retroviral gene transfer into hematopoietic stem cells. Blood. 2003;101:2099–2114. [PubMed]
16. Jiang Y, et al. Pluripotency of mesenchymal stem cells derived from adult marrow. Nature. 2002;418:41–49. [PubMed]
17. Herzog EL, Chai L, Krause DS. Plasticity of marrow derived stem cells. Blood. 2003;102:3483–3493. [PubMed]
18. McCauslin CS, et al. In vivo retroviral gene transfer by direct intrafemoral injection results in correction of the SCID phenotype in Jak3 knock-out animals. Blood. 2003;102:843–848. [PubMed]
19. Sastry L, et al. Certification assays for HIV-1-based vectors: frequent passage of gag sequences without evidence of replication-competent viruses. Mol Ther. 2003;8:830–839. [PubMed]
20. Becker K, Pan D, Whitley CB. Real-time quantitative polymerase chain reaction to assess gene transfer. Hum Gene Ther. 1999;10:2559–2566. [PubMed]
21. Hematti P, Sellers SE, Agricola BA, Metzger ME, Donahue RE, Dunbar CE. Retroviral transduction efficiency of G-CSF+ SCF-mobilized peripheral blood CD34+ cells is superior to G-CSF or G-CSF+ Flt3-L-mobilized cells in nonhuman primates. Blood. 2003;101:2199–2205. [PubMed]
22. Li K, et al. Small peptide analog of SDF-1{alpha} supports survival of cord blood CD34+ cells in synergy with other cytokines, enhances their ex vivo expansion and engraftment into NOD/SCID mice. Stem Cells. 2006;24:55–64. [PubMed]
23. Bunnell BA, Muul LM, Donahue RE, Blaese RM, Morgan RA. High-efficiency retroviral-mediated gene transfer into human and nonhuman primate peripheral blood lymphocytes. Proc Natl Acad Sci USA. 1995;92:7739–7743. [PubMed]
24. Pan D, Stroncek DF, Whitley CB. Improved gene transfer and normalized enzyme levels in primitive hematopoietic progenitors from patients with mucopolysaccharidosis type I using a bioreactor. J Gene Med. 2004;6:1293–1303. [PubMed]
25. Hanenberg H, Xiao XL, Dilloo D, Hashino K, Kato I, Williams DA. Colocalization of retrovirus and target cells on specific fibronectin fragments increases genetic transduction of mammalian cells. Nat Med. 1996;2:876–882. [PubMed]
26. Newbound GC, Cooper JR, O’Rourke JP, Baskin CR, Bunnell BA. Analysis of gene transfer efficiency of retrovirus producer cell transplantation for in situ gene transfer to hematopoietic cells. Exp Hematol. 2001;29:163–173. [PubMed]
27. Nelson DM, Metzger ME, Donahue RE, Morgan RA. In vivo retrovirus-mediated gene transfer into multiple hematopoietic lineages in rabbits without preconditioning. Hum Gene Ther. 1997;8:747–754. [PubMed]
28. Cherry SR, Biniszkiewicz D, van Parijs L, Baltimore D, Jaenisch R. Retroviral expression in embryonic stem cells and hematopoietic stem cells. Mol Cell Biol. 2000;20:7419–7426. [PMC free article] [PubMed]
29. Pfeifer A, Ikawa M, Dayn Y, Verma IM. Transgenesis by lentiviral vectors: lack of gene silencing in mammalian embryonic stem cells and preimplantation embryos. Proc Natl Acad Sci USA. 2002;99:2140–2145. [PubMed]
30. Lois C, Hong EJ, Pease S, Brown EJ, Baltimore D. Germline transmission and tissue-specific expression of transgenes delivered by lentiviral vectors. Science. 2002;295:868–872. [PubMed]
31. Schroder AR, Shinn P, Chen H, Berry C, Ecker JR, Bushman F. HIV-1 integration in the human genome favors active genes and local hotspots. Cell. 2002;110:521–529. [PubMed]
32. Wu X, Li Y, Crise B, Burgess SM. Transcription start regions in the human genome are favored targets for MLV integration. Science. 2003;300:1749–1751. [PubMed]
33. Hammond SM, Crable SC, Anderson KP. Negative regulatory elements are present in the human LMO2 oncogene and may contribute to its expression in leukemia. Leuk Res. 2005;29:89–97. [PubMed]
34. Baksh D, Song L, Tuan RS. Adult mesenchymal stem cells: characterization, differentiation, and application in cell and gene therapy. J Cell Mol Med. 2004;8:301–316. [PubMed]
35. Crevensten G, et al. Intervertebral disc cell therapy for regeneration: mesenchymal stem cell implantation in rat intervertebral discs. Ann Biomed Eng. 2004;32:430–434. [PubMed]
36. Chamberlain JR, et al. Gene targeting in stem cells from individuals with osteogenesis imperfecta. Science. 2004;303:1198–1201. [PubMed]
37. Mangi AA, et al. Mesenchymal stem cells modified with Akt prevent remodeling and restore performance of infarcted hearts. Nat Med. 2003;9:1195–1201. [PubMed]
38. Jin HK, Carter JE, Huntley GW, Schuchman EH. Intracerebral transplantation of mesenchymal stem cells into acid sphingomyelinase-deficient mice delays the onset of neurological abnormalities and extends their life span. J Clin Invest. 2002;109:1183–1191. [PMC free article] [PubMed]
39. Ortiz LA, et al. Mesenchymal stem cell engraftment in lung is enhanced in response to bleomycin exposure and ameliorates its fibrotic effects. Proc Natl Acad Sci USA. 2003;100:8407–8411. [PubMed]
40. Glennie S, Soeiro I, Dyson PJ, Lam EW, Dazzi F. Bone marrow mesenchymal stem cells induce division arrest anergy of activated T cells. Blood. 2005;105:2821–2827. [PubMed]
41. Horwitz EM, et al. Isolated allogeneic bone marrow-derived mesenchymal cells engraft and stimulate growth in children with osteogenesis imperfecta: implications for cell therapy of bone. Proc Natl Acad Sci USA. 2002;99:8932–8937. [PubMed]
42. Petite H, et al. Tissue-engineered bone regeneration. Nat Biotechnol. 2000;18:959–963. [PubMed]
43. Koc ON, et al. Rapid hematopoietic recovery after coinfusion of autologous-blood stem cells and culture-expanded marrow mesenchymal stem cells in advanced breast cancer patients receiving high-dose chemotherapy. J Clin Oncol. 2000;18:307–316. [PubMed]
44. VandenDriessche T, et al. Lentiviral vectors containing the human immunodeficiency virus type-1 central polypurine tract can efficiently transduce nondividing hepatocytes and antigen-presenting cells in vivo. Blood. 2002;100:813–822. [PubMed]
45. High KA. The risks of germline gene transfer. Hastings Cent Rep. 2003;33:3. [PubMed]
46. Mazurier F, Doedens M, Gan OI, Dick JE. Rapid myeloerythroid repopulation after intrafemoral transplantation of NOD-SCID mice reveals a new class of human stem cells. Nat Med. 2003;9:959–963. [PubMed]
47. Sun S, et al. Isolation of mouse marrow mesenchymal progenitors by a novel and reliable method. Stem Cells. 2003;21:527–535. [PubMed]
48. Zuk PA, et al. Multilineage cells from human adipose tissue: implications for cell-based therapies. Tissue Eng. 2001;7:211–228. [PubMed]
49. Schmidt M, et al. Clonality analysis after retroviral-mediated gene transfer to CD34+ cells from the cord blood of ADA-deficient SCID neonates. Nat Med. 2003;9:463–468. [PubMed]