| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
4 integrin expressionDepartment of Pathology, The University of Alabama at Birmingham, Birmingham, Alabama, USA
1Correspondence: Department of Pathology, 701, 19th St. South, LHRB 513, The University of Alabama at Birmingham, Birmingham, AL 35294-0007, USA. E-mail: pons{at}uab.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
4 integrin (CD49d) on MSC greatly increases bone homing in an immunocompetent mouse model. Heterodimerization of the 4 integrin with endogenous β1 integrin (CD29) was confirmed to influence this targeting. In addition to retaining their stem cell property, the engrafted MSC were also found to form osteoblasts and osteocytes in the growth plate of recipient mouse limb bones (femur/tibia) in vivo. These findings provide evidence for a novel strategy to achieve bone homing of genetically engineered MSC, which may broadly benefit in targeted therapies for osteopenic bone defects and cancer bone metastasis.—Kumar, S., Ponnazhagan, S. Bone homing of mesenchymal stem cells by ectopic 4 integrin expression.
Key Words: tissue regeneration • cell therapy • bone targeting
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Despite their potential in regenerative medicine, a main challenge in therapeutic MSC transplantation remains in the delivery of ex vivo modified cells efficiently to the sites of intended correction. In particular, for diseases affecting bone, homing of ex vivo modified MSC necessitates distribution of the cells to entire skeleton. Although systemic administration of MSC appears to be the ideal route to achieve this, a majority of systemically administered MSC target lungs (2)
. Thus, development of methods to overcome ineffective bone targeting of MSC should greatly increase their application in long-term correction of osteopenic bone defects. Such genetically engineered MSC, capable of homing to bone, can also be used as cell therapy vehicles in pathological conditions such as cancer bone metastasis and primary osteosarcoma.
To overcome this limitation, we reasoned that transient, ectopic expression of 4β1 integrin on MSC surface should facilitate their bone homing and engraftment on in vivo transfer. Integrins are known to play an important role in cell migration, invasion, and metastasis during tumor progression (3
, 4)
. Among the family of integrins, 4β1 integrin is a cell surface heterodimer, which mediates cell-cell and cell-extracellular matrix interactions through adhesion to the vascular cell adhesion molecule (VCAM)-1 and to the IIICS region of fibronectin (5)
. Mobilization, homing, and early engraftment of peripheral blood progenitors are also related to reversible 4 integrin expression and function (6)
. It is becoming increasingly evident that homing of hematopoietic stem cells to bone is critically dependent on the expression of matrix-associated proteins, which interact with receptors on cells in the bone marrow (7
, 8)
. Evidence for the requirement of 4 integrin for bone marrow engraftment of stem cells has been substantiated in studies where, when murine or sheep bone marrow was incubated with anti-4 integrin antibody, the repopulating stem cells failed to engraft in bone (9)
. The 4 integrin has been shown to affect homing of not only the hematopoietic stem cells but also of metastatic tumor cells to bone. Cell adhesion to the extracellular matrix is an important process that controls cell migration, proliferation, survival, and differentiation (10)
. The cellular expression of adhesion molecules is critically important for cell-based therapy to mediate adhesion of the implanted cells to the extracellular matrix of targeted tissue in host animals (11)
. The increase in 4β1 integrin expression on primary melanomas has been reported to significantly correlate with the development of bone metastasis (12)
. Therefore, we hypothesized that enrichment of progenitor phenotype and bone homing signal on ex vivo cultured MSC should enhance reconstitution in bone on systemic transplantation. Biologically active integrins are formed by noncovalent association of an subunit with a β subunit. Several studies (13)
have described that the β subunit of integrin is synthesized in excess, suggesting that the availability of subunit might be rate-limiting for the maturation of β subunit. Thus, we exploited the natural abundance of endogenous β subunit in MSC to heterodimerize with subunit by overexpression of the latter by gene transfer and demonstrate enriched bone homing of pluripotent MSC in vivo.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
4 and β1 integrin antibodies were purchased from e-biosciences (San Diego, CA, USA) and Santa Cruz Biotechnology (Santa Cruz, CA, USA), respectively. Recombinant VACM-1 was purchased from R&D Systems (Minneapolis, MN, USA), and purified fibronectin, plasminogen, and BSA were purchased from Sigma (St. Louis, MO, USA).
Construction of plasmids and production of recombinant adeno-associated virus (rAAV)
All AAV plasmids were constructed using pSub201 as the backbone. cDNA encoding mouse 4 integrin was kindly provided by Dr. David Rowe (University of Connecticut Health Science Center, Farmington, CT, USA). The coding sequence of mouse 4 integrin was subcloned in rAAV plasmid under the control of CMV promoter. Recombinant AAV2 encoding GFP was described previously (14)
. Packaging of rAAV2 encoding 4 integrin was performed in an adenovirus-free system as described previously (15)
. Purification of the virions was done in discontinuous iodixanol gradient centrifugation followed by heparin affinity chromatography. Particle titers of the purified virions were determined by quantitative slot blot analysis as described previously (15)
.
Primary mouse MSC culture, gene transfer, and enrichment of 4 integrin-positive cells
C57BL/6 mice were purchased from the National Cancer Institute-Frederick Cancer Research Facility (Frederick, MD, USA), and the GFP transgenic mice [C57BL/6-Tg(ACTbEGF)1Osb/J] were purchased from the Jackson Laboratories (Barr Harbor, ME, USA). All animal protocols were performed following the guidelines of the Institutional Animal Care and Use Committee. For preparation of bone marrow conditioned media, the femur and tibia bone fragments flushed off marrow were cultured in alpha-modified minimum essential medium (-MEM) supplemented with 20% fetal bovine serum (FBS). Three days later, the medium was collected by centrifugation, diluted with equal volume of -MEM media to get the 10% FBS concentration, and filtered through disposable 0.2 micron tissue culture filtration unit. To obtain bone marrow stromal cells, 6- to 8-wk-old male mice were killed, bone marrow was flushed from the femur and tibia, and the marrow mononuclear cells were purified by ficoll gradient. Bone marrow stromal cells were grown in stromal cell culture conditioned medium (-MEM containing 10% FBS) supplemented with 10–9 M FGF2 to maintain cells in pluripotent and undifferentiated state (16)
. Periodically, floating cells were removed and fresh medium replenished. Residual macrophages from the MSC culture were removed by IMAC using anti-mouse CD11b beads (cat# 558013 BD IMag, BD Biosciences, San Diego, CA, USA). After 14 days, the adherent stromal cells were split before attaining confluence to avoid possible onset of differentiation. The cells were routinely prepared and used for in vitro and in vivo studies as low passage cultures (passage 4–8).
Undifferentiated MSC were transduced with 100 multiplicity of infection (MOI) of rAAV2-4 integrin or rAAV2-GFP. Before transduction, cells were removed of the growth medium and washed once with serum-free Opti-MEM (Gibco-BRL, Gaithersburg, MD, USA). Virus infection was performed in 125 µl of Opti-MEM for 2 h at 37°C after which complete medium with FGF2 was added. Seventy-two hours after transduction, cells were dislodged by gentle treatment with cell stripper solution (cat# 25–056-Cl, Mediatech, CELLGRO, Herndon, VA, USA) and sorted by magnetic cell separator (IMAC) in sterile condition based on 4-integrin positivity. The 4 integrin positive cell population was used in in vitro studies and transplantation experiments in vivo in mice.
Heterodimerization of 4 with endogenous β1 integrin
Heterodimerization of 4 integrin with endogenous β1 integrin was detected by immunoprecipitation. MSC were transduced with rAAV encoding 4 integrin or GFP and maintained in culture for 10 days. Cell lysates were prepared in lysis buffer (25 mM Tris Phosphate, pH7.8, 2 mM DTT, 2 mM 1,2-diaminocyclohexane N,N,N',N'-tetraacetic acid, 10% glycerol, 1% Triton X-100, and protease inhibitor cocktail) and immunoprecipitation was done using an 4 integrin antibody bound to the proteinA/G agarose beads (Santa Cruz Biotechnology). Proteins bound to the beads were denatured and electrophoretically separated in 10% SDS-PAGE and transferred to PVDF membrane. Immunodetection of the blot was performed using β1 integrin antibody. The blot was developed with ECL chemiluminescence substrate.
For immunofluorescence analysis and colocalization of 4 and β1 integrins in MSC in situ, mock-transduced and rAAV2-4 integrin-transduced MSC were harvested using cell stripper solution. Staining was performed with phycoerythrin (PE)-conjugated 4 integrin antibody and fluorescein isothyocyanate (FITC)-conjugated β1 integrin antibody. After being washed the unbound antibodies, the cells were analyzed by digital confocal microscopy (Olympus IX70 confocal microscope).
Adhesion assay
The MSC were mock-transduced or transduced with rAAV2-4 integrin or rAAV2-GFP at an MOI of 100. After 1 wk, MSC were transfected with a plasmid encoding luciferase using lipofectamine 2000 (Gibco-BRL). Forty-eight hours later, the cells were gently dislodged using cell stripper solution and 1 x 106 cells were added in triplicate to a 48-well tissue culture plate; coated with 10 µg/ml recombinant, soluble VCAM-1, 20 µg/ml fibronectin, plasminogen, or BSA in PBS at 4°C overnight; and then blocked with 1% BSA for 1 h at room temperature per well to VCAM-1, fibronectin, plasminogen or BSA coated wells. Unbound cells were washed immediately with PBS three times after which bound cells were lysed in 1x lysis buffer. Luciferase activity was measured in a Berthold Sirius Luminometer.
Cell cycle analysis
To determine whether rAAV transduction alters growth kinetics in mouse MSC, cells were mock-transduced or transduced with rAAV2-4 integrin and maintained in culture for 20 days after which the cells were trypsinized and fixed with cold ethanol and stained with propidium iodide (PI) and analyzed by FACS Aria flow cytometer (BD Biosciences, San Diego, CA, USA). The cell cycle fractions were quantified using the ModFit LT 3.0 software (BD Biosciences).
Cell growth analysis
Cultured MSC were mock-transduced or transduced with rAAV2 encoding 4 integrin in triplicates and cultured for five passages. After each passage, cells were trypsinized and the total number of cells was counted after trypan blue staining in a Olympus (CK40) microscope.
Apoptosis assay
Apoptosis was determined by staining cells with fluorescein isothiocyanate-conjugated annexin V (FITC-annexin V) and PI; annexin V detects early apoptotic cells based on externalized phosphatidylserine and PI detects the late apoptosis rate. Trypsinized cells were stained with FITC-annexin V and PI using the annexin V-FITC apoptosis detection kit (BD Biosciences) and then analyzed by FACS Aria flow cytometer.
Multilineage differentiation studies of cultured mouse MSC after rAAV-4 integrin transduction
Osteoblast differentiation was induced by culture medium containing 10% FBS, 0.1 µM dexamethazone, 2 mM β-glycerophosphate, and 150 µM ascorbate-2-phosphate (17)
. Cells were seeded at 10,000 cell/cm2 and incubated for 28 days at 37°C. Medium was changed every 3 days. Adipogenic differentiation was induced by culturing in medium with 20% FBS, 1 µM dexamethazone, 0.35 µM hydrocortisone, 0.5 mM isobutyl-methylxanthine (IBMX), 100 ng/ml insulin, and 60 µM indomethacin (17)
. Cells were seeded at 20,000 cells/cm2 and incubated for 21 days at 37°C. Medium was changed every alternate day. Chondrogenic differentiation was obtained in micropellets (3x105 cells/pellet) incubated at 37°C for 25 days in chondrogenic medium containing 0.1 µM dexamethazone, 1 mM sodium pyruvate, 170 µM ascorbic acid-2-phosphate, 350 µM praline, 1x insulin-transferrin-selenium, and 10 ng/ml TGF-β (17)
. Medium was changed every 3 days. Vascular smooth muscle differentiation was obtained in long-term culture medium with 12.5% screened horse serum, 12.5% FBS, 20 µM L-glutamine, 0.8 mM L-serine, 0.15 mM L-asparagine, 1 mM sodium pyruvate, 5 mM sodium bicarbonate, 1 µM hydrocortisone, and antibiotics (17)
. Medium was changed every 3 days after the end of the first week of passage. Neurogenic differentiation was induced by the incubation of subconfluent MSC in neurogenic medium supplemented with 100 µM CoCl2 (Sigma) and 5 ng/ml basic FGF-2 (18)
.
For evaluation of mineralized matrix, the cell layer was fixed in 10% buffered-formalin, then stained by von Kossa stain using 5% (w/v) silver nitrate (Sigma) under ultraviolet light for 30 min, followed by 5% (w/v) sodium thiosulphate (Sigma) for 2 min. For Oil red-O staining, cells were fixed in formalin and stained for 1 h with oil red-O (Sigma). For toluidine blue-O staining, cells were fixed in formalin for 24 h and stained with toluidine blue-O. For immunofluorescence detection, confluent layers of culture were fixed and permeabilized with methanol for 10 min. Slides were incubated for 1 h with smooth muscle actin antibody (Neomarkers, Fremont, CA, USA) for myogenic differentiation or nestin antibody for neurogenic differentiation, followed by alexafluor-594 conjugated goat anti-mouse secondary antibody (Molecular Probes, Eugene, OR, USA) and examined with a Leica fluorescence microscope (Leica Microsystems Wetzlar GmbH, Wetzlar, Germany) after thorough washing and mounting in antifade fluoromount medium.
Transplantation of marked MSC into syngeneic female mice and analysis of donor cell engraftment
The recipient animals were 6-wk-old female C57BL/6 mice. A total of eight mice was included in each group. Mock-transduced or rAAV2 (GFP or 4 integrin)-transduced MSC were resuspended in a volume of 100 µl normal saline and systemically administered into recipient mice through tail vein. Cohorts of mice received 1 x 106 cells each of untransduced, rAAV2-GFP transduced, or rAAV2-4 integrin transduced MSC.
Four weeks after transplantation, animals were killed. Identification of homed MSC from the donor male mice was done by in situ hybridization using Y chromosome-specific probe. A digoxigenin (DIG)-labeled mouse Y-chromosome specific probe (19)
was generated by polymerase chain reaction (PCR) using DIG-labeling mixPlus (Molecular Biochemicals, Mannheim, Germany) following the manufacturer’s protocol. Formalin-fixed and decalcified bone tissue sections were deparaffinized in xylene and rehydrated through a series of graded-ethanol and PBS. Slides were then treated with 0.01 M citrate buffer, pH 6.0 at 42°C for 3 h. Prehybridization was performed at 65°C for 2 h in ULTRAhyb hybridization solution (Ambion, Austin, TX, USA), and hybridization reaction was performed with 
400 ng/ml DIG-labeled, Y-chromosome-specific probe in same solution at 65°C overnight. After the excess probe was thoroughly washed, detection of hybridized probe signal was performed using the DIG nucleic acid detection kit. Counterstaining of the slide was done with diluted eosin solution for 1–2 min and then mounted with Immuno-mount (Thermo Shandon, CA, USA). A region of interest (ROI) was selected just below the growth plate area of bone marrow section. Relative percentage of homed MSC, as evidenced by positive-signal, was enumerated by microscopy. For quantitation, at least five random fields from each slide were counted in an Olympus BX51 microscope.
Identification of GFP-positive donor MSC, isolated from syngeneic, GFP transgenic mice in the recipient mouse bone sections was performed by immunohistochemistry using anti-mouse CD44 and anti-GFP primary antibodies and developed with goat anti-mouse labeled with alexafluor 594 (Molecular Probes) and donkey anti-mouse alexafluor 488 (Molecular Probes) secondary antibodies, respectively.
To determine the percentage of engrafted MSC in the recipient bone marrow, stromal cells were isolated 2 months after implantation and cultured for 3 wk in vitro. The cells were trypsinized and analyzed by flow cytometry based on GFP fluorescence.
Quantitation of repopulated donor MSC in different tissues
Identification and quantitation of the donor cells that homed to different organs were determined by Southern blot hybridization and real-time PCR, respectively, using genomic DNA isolated from bone marrow, lung, liver, adipose tissue, muscle, lymph node, and kidney. For Southern blot analysis, 5 µg of EcoRI digested genomic DNA from the bone marrow of each animal were separated by agarose gel electrophoresis and transferred to nylon membrane. Detection of donor cell DNA was performed as described earlier with P32-labeled, Y-chromosome specific probe (19)
. After detection, the blot was stripped off the Y-chromosome probe and rehybridized with P32-labeled IL-2 cDNA probe. Quantitative real-time PCR was carried out with DNA isolated from bone marrow, lung, liver, adipose tissue, muscle, lymph node, and kidney using Y chromosome-specific primers in a Bio-Rad icycler (Optical Module). Reactions were performed using the LightCycler-FastStart DNA Master SYBR Green system (Roche Molecular Biochemicals, Indianapolis, IN, USA) in a final volume of 25 µl, consisting of 0.5 µl of each primer (0.5 pM), 12.5 µl of 2x supermix containing reaction buffer, Fast-start TaqDNA double strand-specified SYBR Green I dye, 5.5 µl H2O, and 5 µl (0.5 µg) of template DNA. PCR was performed with a 3 min preincubation at 95°C followed by 45 cycles of 15 s denaturation at 95°C, and 30 s annealing and extension at 57°C. PCR products were subjected to melting curve analysis using the light cycler system to exclude amplification of nonspecific sequences. Values obtained from amplification of donor-specific sequences from each sample were normalized to copy number of GAPDH gene amplification from the same sample to derive the relative number of donor cells from each sample.
Statistical analysis
Results are presented as mean ± SD. Comparisons between individual data points were performed using the Student’s t test. Results were considered significant if P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
4 integrin in MSC leads to successful heterodimerization with endogenous β1 integrin and localization on cell membrane4 integrin in mouse MSC engraftment to bone, a 3.5 kb full-length cDNA fragment of the murine 4 integrin subunit was cloned into an AAV2 vector under the transcriptional control of CMV promoter. The AAV2-4 integrin virus was transduced to cultured mouse MSC, and expression of the 4 integrin was confirmed by flow cytometry and >80% of the cells were positive for transgene expression (Fig. 1
A). Ectopic expression of 4 was confirmed by staining the cells using PE-conjugated 4 integrin antibody and FITC-conjugated β1 integrin antibody. Colocalization of the two subunits on cell membrane was demonstrated by merging the images. Once again, >80% of the cells that were positive for β1 integrin antibody were also found to be positive for 4 integrin antibody indicating high-efficiency gene transfer (Fig. 1A
).
|
To demonstrate that the vector-expressed 4 integrin heterodimerizes with endogenous β1 integrin, immunoprecipitation was performed. Cell lysates from rAAV2-4 integrin transduced, rAAV2-GFP transduced, or mock transduced MSC were precipitated with an antibody for 4 integrin, and the immunoprecipitated protein was electrophoretically separated on denaturing SDS-PAGE, transferred to a PVDF membrane and detected with an antibody specific for β1 integrin. Results, shown in Fig. 1B
, indicate heterodimerization of vector-expressed 4 with endogenous β1 integrin. Western blot analysis of lysate from untransduced MSC did not show detectable levels of endogenous 4 expression.
Expression of 4 integrin on MSC does not alter cell cycle pattern or rate of apoptosis
Growth curve analysis of MSC after transduction with rAAV-4 integrin did not show any significant difference compared to untransduced MSC (Fig. 2
A). Cell cycle stages were analyzed by PI staining and FACS analysis in mock-transduced or rAAV transduced MSC after 20 days of culture in vitro. Results of this study did not show any significant change in cell cycle profiles (Fig. 2B
). Whereas mock transduced MSC showed 78.65% of cells in G1/0, 10.73% of cells in G2/M, and 10.61% of cells in S phase, rAAV-4 integrin transduced MSC showed 78.88% of cells in G1/0, 11.20% of cells in G2/M and 9.92% of cells in S phase of growth (P>0.05). Similarly, scoring the early apoptosis rate by FITC-annexin V staining and late apoptosis rate by PI staining in FACS based assay did not show significant changes in the early apoptosis or late apoptosis rate. In mock-transduced MSC, 4.5% of cells were annexin V positive and 0.2% of cells were PI positive and rAAV-4 integrin transduced MSC, 4.7% of cells were annexin V positive, and 0.2% of cells were PI positive (Fig. 2C
).
|
Transduction of rAAV and expression of 4 integrin does not affect multipotency of MSC
Results shown in Fig. 3
indicate that under the appropriate culture conditions rAAV-4 integrin transduced MSC could be differentiated into adipocytes 3 wk later, mineralized bone 4 wk later, and chondrocytes 25 days of culturing. After 3 wk of culture in adipogenic medium, rAAV-4 integrin transduced MSC contained large oil red O positive vesicles. After 21 days in long-term culture medium inducing vascular smooth muscle differentiation, immunofluorescence studies with smooth muscle actin antibody indicated the presence of microfilaments containing smooth muscle actin. Similarly, after 21 days of culture in neurogenic medium, the rAAV-4 integrin transduced MSC exhibited neuronal morphology and immunofluorescence staining with neuron specific nestin antibody confirmed the neurogenic differentiation of MSC (Fig. 3)
.
|
4β1 integrin expressing MSC demonstrate target specific binding
The 4β1 integrin is a member of very late antigen (VLA) protein subfamily and has been implicated in triggering intercellular and homotypic cell adhesion (20
, 21)
. To determine whether the 4 subunit-expressing MSC retain biological activity of VLA-4 specificity on transduced MSC, adhesion to soluble recombinant VCAM-1 and fibronectin and a non-4 integrin binding ligand, plasminogen, was performed. Specific adhesion to VCAM-1 and fibronectin was observed from the rAAV-4 integrin transduced cells compared to the control confirming that the transgenic protein is functionally active with high avidity for VCAM-1 and fibronectin (Fig. 4
). There was no significant binding of 4 integrin-expressing or control MSC in BSA-coated or plasminogen-coated wells confirming the specificity of binding.
|
Expression of 4 integrin on MSC allows enhanced homing of the cells to bone on autologous transfer
To determine the efficacy of MSC expressing 4 integrin in bone homing in vivo, the cells were systemically injected through tail vein in C57BL/6 mice. Before in vivo injection, the AAV-4 integrin-transduced MSC were sorted by a magnetic cell separator using 4 integrin antibody to enrich cells positive for surface expression of 4 integrin. The sorted cells were washed with PBS and injected in vivo. To distinguish the transplanted MSC from endogenous MSC in target tissues, donor cells were derived from male mice and were transplanted into female recipients. A dose of 1 x 106 MSC was administered per mouse in 5 consecutive days of intravenous injection. All animals survived after the MSC infusion. Two weeks later, mice were killed and tissues including bone marrow, adipose tissue, muscle, lung, liver, kidney and lymph node were collected. Portions of the tissue were embedded in paraffin or used for genomic DNA isolation. In situ hybridization was performed in decalcified bone sections of tibia, femora, and vertebrae and other tissues with Y-chromosome-specific probe. For quantitation of positive signal in the ROI, below the growth plate of each femur section, at least five fields were counted in a microscope. Results shown in Fig. 5
indicated a significant enhancement of bone homing of MSC after rAAV-4 integrin transduction before transplantation. There was nearly a 10-fold increase in the number of donor cells after 4 expression. Thus far, we were able to record retention of bone marrow homed MSC for 4 months without significant diminution (data not shown). There was no significant difference in bone homing between untransduced MSC and rAAV-GFP transduced MSC injection (Fig. 5)
|
Southern blot and semiquantitative real-time PCR analyses were performed with genomic DNA obtained from the tissues to determine approximate copy number of repopulated donor cells. Results shown in Fig. 6
A, B also corroborate the in situ data that there was a significant enhancement in bone homing of MSC after surface expression of 4β1 integrin and a significant reduction in homing to other organs, especially to the lungs. Here again there was no significant difference between untransduced MSC and rAAV2-GFP transduced MSC groups (data not shown).
|
Quantitative analysis of donor-derived MSC in recipient bone marrow on repopulation in vivo
To determine the proportion of repopulated MSC from infused donor cells in the bone, low-density bone marrow cells, obtained from mice that received donor MSC derived from GFP transgenic mice were cultured for 21 days, and adherent stromal cell population was recovered. The cells were sorted by flow cytometry to enumerate the percentage of repopulation. Results, shown in Fig. 7
, indicate that when MSC from GFP mice were transplanted without prior transduction by rAAV-4 integrin, there was only 2.35% of MSC of donor origin; however, when 4 integrin expressing MSC were transduced, 27.19% of the cells in the bone marrow of recipient mice were found to be of donor origin.
|
Genetically modified MSC expressing 4 integrin are viable for both retaining pluripotency and osteogenesis in vivo
To determine whether bone-homed MSC have osteogenic potential, we derived MSC from syngeneic, GFP transgenic mice [C57BL/6-Tg(ACTbEGF)1Osb/J] and used for the transplantation in C57BL/6 recipients after rAAV-4 integrin transduction. Four weeks later, mice were killed and bone sections were immunohistochemically stained with GFP antibody to localize repopulated donor cells. Results indicated significant number of GFP-positive osteoblasts, suggesting their origin from donor MSC. Further, retention of stem cell property of the transplanted MSC was also evident by staining the sections with antibody for CD44, a marker exclusively present in undifferentiated MSC. Trichrome staining confirmed the distribution of reconstituted donor MSC in both the epiphyseal region and growth plate of femur and tibia (Fig. 8
). These data provide evidence for not only the osteogenic potential of bone-targeted MSC but also retention of their stem cell property, suggesting their utility in cell therapy and tissue regeneration.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
or when donor marrow cells were exposed to a mixture of cytokines (23)
. In addition to the failure to engraft, improper culture conditions also result in lineage commitment of MSC (23)
. Thus, it is imperative that appropriate culture conditions are maintained to avoid lineage commitment of MSC in vitro. We have optimized our MSC culture conditions by using 10–9 M FGF-2, which prevents differentiation of the stem cells and maintains their pluripotent nature (16)
. Further, we have shown earlier that protocols involved in transduction of mouse MSC with rAAV do not alter key gene expression pattern or induce differentiation as these cells could be differentiated into osteoblasts, chondrocytes, or adipocytes with appropriate stimuli (14)
. Growth characteristics of multipotent MSC were not found to be affected after transduction with rAAV2-4 integrin as determined by cell cycle, proliferation and apoptosis analyses. Similarly, after rAAV2-4 integrin transduction, the ability of multilineage differentiation of MSC was confirmed in vitro and the cells were differentiated into osteogenic, chondrogenic, adipogenic, neurogenic, and vascular smooth muscle lineages with appropriate stimuli.
The 4β1 integrin is one of the integrins that can mediate initial capture, rolling, and firm attachment of cells to the bone marrow and endothelial cells (10
, 11
, 24)
. The 4β1 integrin has been shown to promote homing of myeloid cells/macrophages to tumor tissues (25)
. Analysis of homing showed that these cells travel throughout the vessels of the tumor but only attached within the tumor vessels at the periphery of the tumor. The 4β1 integrin-VCAM interaction also regulates T cell and natural killer cell trafficking (26)
. It is also known that although most of the versatile functions of 4 integrin are likely through interaction with its primary ligand VCAM-1, binding to other ligands cannot be excluded. Several VCAM-1 independent functions of 4β1 integrin have been reported (27
, 28)
. Thus, in addition to interaction with VACM-1, we believe upon in vivo infusion, surface expression of 4β1 integrin on MSC may further utilize other ligands present endogenously for preferential homing to bone.
The involvement of specific cell-surface molecules in selective association of cells is very critical in biological processes. Intravenous injection of cells mostly results in lung homing in mice, and 4β1 expressing melanoma cells have been reported to metastasize to lung (29)
. Interestingly, in our studies with MSC, we observed less number of cells that homed to lung after 4 integrin expression but an increase in bone homing of these MSC. Several different proteins orchestrate the adhesion of cells to exposed extracellular matrices, and there may be unique changes in rAAV2-4 integrin transduced MSC that are associated with reduced homing in lung. One possibility is that the expression of subunit of integrin in MSC may alter the expression levels of other adhesion molecules, including other and β integrins and change the overall ligand binding capacity so that frequency of early emboli of MSC trapped in lung tissues after transplantation is minimized. Because of the broad range of proposed roles ascribed to 4β1 in cellular function and its unique structural attributes, systematic analyses of the 4β1 integrin expressing MSC by biosynthetic and functional approaches are essential to completely understand the biological properties of these cells. Melanoma cells expressing the 4β1 integrin have been shown to metastasize to lung (29)
. Most of these studies were done with transformed cancer cells, which are known to up-regulate several other molecules involved in cell migration and attachment compared to primary MSC. Thus, transient expression 4β1 integrin on MSC surface may preferentially increase the homing and retention of these cells in the bone marrow, which is the natural niche for MSC.
Identification of donor MSC with stem cell marker in the epiphyseal region of femur and tibia as well as participation of donor-derived cells at the growth plate region in our studies demonstrates the osteogenic potential of these cells. We employed rAAV to express murine 4 integrin to transduce mouse MSC before in vivo transfer. Several reasons led us to choose rAAV for this purpose. AAV is a nonpathogenic vector with very poor integration into cellular genomic DNA. We recently established high-efficiency gene transfer of rAAV in MSC and that rAAV transduction of MSC does not alter their pluripotent nature (14)
. Although other nonintegrating viral and nonviral vectors can be used as alternatives for transient expression of 4 integrin, retention of pluripotent status of MSC after gene transfer should be carefully monitored.
Retention of stem cell property of genetically modified MSC in our studies indicates the potential of this strategy for not only defects in osteogenesis but also other pathological situations such as cancer bone metastasis, frequently seen in patients with breast and prostate cancers. By stably transducing genes for factors that are inhibitory to osteoclastogenesis, such as osteoprotegerin (30)
, or inhibitory to osteoblastogenesis, such as noggin (31)
, it is indeed possible to alleviate osteolytic and osteoblastic bone lesions commonly encountered in patients with breast and prostate cancers respectively. However, further studies on long-term repopulation may be needed to establish whether 4 integrin expressing cells may undergo cellular transformation or augment the metastatic potential of primary tumors.
Collectively, these studies provide evidence for a novel strategy to enrich bone homing of MSC for ex vivo therapy. By using other cell-specific ligands, substituting the 4 integrin, the strategy described here should be adapted to target multiple tissue types utilizing MSC as either cellular vehicles expressing therapeutic proteins or as a stem cell source for tissue regeneration. The multilineage potential of MSC identified recently can largely benefit by this targeting approach to achieve not only bone homing of MSC but also to other tissues including heart, muscle, lung, liver, and spleen for therapeutic function or tissue regeneration.
|
ACKNOWLEDGMENTS |
|---|
Received for publication February 2, 2007. Accepted for publication June 7, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
4β1 integrin expression and function. J. Clin. Invest. 101,2456-2467[Medline]4β1 into tumor cells. Am. J. Pathol. 148,55-61[Abstract]4β1 on melanoma cells can inhibit the invasive stage of metastasis formation. Cell 77,335-347[CrossRef][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
