Osteoinduction by Ex Vivo Adenovirus-Mediated BMP2 Delivery Is Independent of Cell Type
*Gugala, Z.; *Olmsted-Davis, E.A.; **Gannon, F.H.; *Lindsey, W.L.; *Davis, A.R.
*Baylor College of Medicine, Houston, TX
INTRODUCTION: Localized delivery of cells transduced ex vivo with
adenoviral vectors encoding bone morphogenetic proteins (BMPs) has
proven successful in induction of new bone formation in hetero- and
orthotopic locations.[1-3] The adenoviral delivery of BMP offers the
potential of rapidly achieving local effective concentrations of the protein
to induce bone formation, although safe and efficacious gene-based bone
induction methods have yet to be developed. Systems utilizing ex vivo
transduction with adenovirus type 5 encoding BMP2 (Ad5-BMP2) have
been employed. Recently a novel chimeric adenoviral vector consisting of
Ad5 and a fiber gene of Ad35 has been developed and demonstrated
significantly higher expression of BMP2 in human bone marrow cells as
compared to its Ad5-BMP2 counterpart.[4] However, the optimal target
cell type for the ex vivo adenoviral BMP2 gene delivery remains
undetermined, although the osteogenic potential of the target cells is
postulated as advantageous.
OBJECTIVES: To analyze and compare the abilities of various human
cell types with inherently dissimilar osteogenic potentials to induce
heterotopic bone formation in vivo following ex vivo transduction with
Ad5-BMP2 or Ad5F35-BMP2 representing low- and high BMP2
expression, respectively.
MATERIAL & METHODS: Three cell types were studied: primary
bone marrow mesenchymal stem cells (BM-MSCs); primary skin
fibroblasts (SFs); and a diploid fetal lung cell line (MRC-5). The vectors
included Ad5-BMP2 or a new chimeric Ad5F35-BMP2, both
demonstrating significantly different expression of BMP2 in vitro. The
experimental groups combined all three human cell types each transduced
with one of the two adenoviral vectors. The BMP2 expression was
identified in vitro, quantitated, and compared. The osteoinduction was
assessed heterotopically in vivo using non-obese diabetic severe
combined immunodeficiency (NOD/SCID) mice. Animals were injected
intramuscularly in both hind limbs with a solution containing specific cell
type transduced ex vivo with the specific adenovirus (n=6). Animals were
euthanized 2 weeks post-injection. The nature and extent of heterotopic
bone formation were analyzed radiographically and histologically. The
human origin of the cells in the injection site was traced using anti-human
mitochondrial protein assay.
observed using Ad5F35-BMP2 transduced MRC-5 cells. In contrast, all
cell types transduced with Ad5-BMP2 produced little or no bone. As
compared with the other cell types, no significant beneficial effect of
using transduced BM-MSCs in terms of the extent and maturation of the
heterotopic bone could be demonstrated for either low- or high BMP2
expressing cells. The presence of human cells at the site could be
identified at 2 days post injection; however at 12 days post injection, no
human cells could be identified within the muscle nor within the
heterotopically formed bone suggesting the new bone to be entirely
murine origin.
DISCUSSION: In contrary to previous reports examining similar
osteoinductive systems [5] the results of the present study demonstrate no
significant effect of the cell type in the ability to induce heterotopic bone
formation following ex vivo transduction with Ad5-BMP2 or Ad5F35BMP2. In vitro low- or high levels of BMP2 expression resulted,
depending upon whether Ad5-BMP2 or Ad5F35-BMP2 was used,
irrespective of the cell type. All cell types transduced with Ad5F35BMP2 efficiently induced heterotopic bone formation compared to the
same cells transduced with Ad5-BMP2. No beneficial effect of the
osteogenic potential of BM-MSCs could be noticed in the experiment in
terms of their osteoinductivity as compared with SFs or MRC-5. This
suggests that the level of BMP2 expression appears to be the determining
factor in the efficacy of heterotopic bone formation, rather than the cell
type. This is corroborated by the fact that the newly formed bone was
identified as entirely of host origin. The autocrine effect of BMP2 on the
target cells [6] could not be supported by this experiment; recruitment of
the host stems cells and triggering their osteoblastic differentiation is a
hypothetical mode of Ad-BMP2 action which requires validation.
REFERENCES: 1. Lieberman JR et al. J Ortho Res 16:330-339, 1998;
2. Alden TD et al. Hum Gene Ther 10:2245-2253, 1999; 3. Okubo Y et
al. JBJS 83A:S99-104 2001; 4. Olmsted-Davis EA et al. Hum Gene Ther
13:1337-1347, 2002; 5. Musgrave DS et al. Clin Orthop 378:290-305,
2000; 6. Gazit D et al. J Gene Med 1:121-133, 1999.
**Armed Forces Institute of Pathology; Washington, DC
RESULTS: BM-MSCs transduced with Ad5F35-BMP2 synthesized
considerably more BMP2 in vitro than the same cells transduced with
Ad5-BMP2. No significant differences were observed between the levels
of BMP2 produced by transduction of BM-MSCs, SFs, or MRC-5 cells
with Ad5F3-5BMP2 (Fig. 1). Consistently, all three types of cells
transduced with Ad5-BMP2 demonstrated undetectable levels of BMP2
expression. None of these cells produced detectable levels of endogenous
BMP2, and BMP2 was not detected in cells transduced with control
adenoviral vectors (Ad5-lacZ or Ad5F35-empty).
At fourteen days substantial amounts of heterotopic bone were formed in
the muscle of the limbs injected with BM-MSCs, SFs, and MRC-5 cells
transduced with Ad5F35-BMP2 (Fig. 2). The same cell populations
transduced with Ad5-BMP2 induced either no bone or significantly less.
None of the cells transduced with the control adenoviruses produced
radiologically detectable bone. The surface area of the heterotopic bone
as assessed from biplanar plain radiography, demonstrated no statistically
significant difference (p value range = 0.17-0.41) in the amounts of bone
formed between the three cell types transduced with the same adenovirus.
However, statistically significant (p value range = 0.0002-0.003) more
bone was formed using each of the three cell types transduced with
Ad5F35-BMP2 as compared with Ad5-BMP2.
Fig. 1.
Fig. 2.
Histological evaluations corroborated the radiographic results. Significant
amounts of new bone were formed when Ad5F35-BMP2 was used
irrespective of the cell type. The heterotopic bone had a normal, woven
appearance, and was well demarcated from the adjacent muscle with an
irregular oval shape. The new bone formed consistently in all animals that
received cells transduced with Ad5F35-BMP2. The heterotopic bone was
not contiguous with the periosteum of the femur but originated from the
space in between muscle fibers radiating outward and/or splitting muscle
layers. Histological analysis also confirmed that the bone was formed by
an enchondral process. The most extensive cartilage formation was
50th Annual Meeting of the Orthopaedic Research Society
Paper No: 0009