Stem Cells Express, published online September 1, 2005; doi:10.1634/stemcells.

2005-0021

Original Article

A Novel Human Artificial Chromosome Vector Provides Effective Cell Lineage-Specific

Transgene Expression in Human Mesenchymal Stem Cells

Running Title: Cell Lineage-Specific Transgene Expression by HAC

Xianying Ren1, Motonobu Katoh2, Hedetoshi Hoshiya3, Akihiro Kurimasa3, Toshiaki Inoue2,

Fumiaki Ayabe3, Kotaro Shibata4, Junya Toguchida 4 and Mitsuo Oshimura1,2,3

Downloaded from www.StemCells.com by on September 14, 2009

1
Department of Molecular and Cell Genetics, Graduate School of Medical Science, Tottori

University, 86 Nishi-cho, Yonago, Tottori 683-8503, Japan.

2
Department of Human Genome Sciences (Kirin Brewery), Graduate School of Medical

Science, Tottori University, 86 Nishi-cho, Yonago, Tottori 683-8503, Japan.

3
Department of Biomedical Science, Institute of Regenerative Medicine and Biofunction,

Graduate School of Medicine, Tottori University, 86 Nishi-cho, Yonago, Tottori 683-8503,

Japan.
4
Department of Tissue Regeneration, Institute for Frontier Medical Science, Kyoto University,

53 Kawahara-cho, Shogoin, Sakyo-ku, Kyoto 606-8057, Japan.

Correspondence: Professor Mitsuo Oshimura Ph.D., Department of Biomedical Science,

Institute of Regenerative Medicine and Biofunction, Graduate School of Medical Science,

Tottori University, 86 Nishi-cho, Yonago, Tottori 683-8503, Japan.

Telephone: 81- 859-348261; FAX: 81-859-348134;

e-mail: oshimura@grape.med.tottori-u.ac.jp

Copyright © 2005 AlphaMed Press

Received January 15, 2005; accepted for publication June 7, 2005. ©AlphaMed Press

1066-5099 doi: 10.1634/stemcells.2005-0021

Downloaded from www.StemCells.com by on September 14, 2009

2
ABSTRACT
Mesenchymal stem cells (MSC) hold promise for use in adult stem cell-mediated gene

therapy. One of the major aims of stem cell-mediated gene therapy is to develop vectors

that will allow appropriate levels of expression of therapeutic genes along differentiation

under physiological regulation of the specialized cells. Human artificial chromosomes

(HACs) are stably maintained as independent chromosomes in host cells and should be free

from potential insertional mutagenesis problems of conventional transgenes. Therefore,

HACs have been proposed as alternative implements to cell-mediated gene therapy.

Previously, we constructed a novel HAC, termed as 21 ∆pq HAC, with a loxP site in which

Downloaded from www.StemCells.com by on September 14, 2009

circular DNA can be reproducibly inserted by the Cre/loxP system. We here assessed

feasibility of lineage-specific transgene expression by the 21∆pq HAC vector, utilizing in

vitro differentiation system with a MSC cell line, hiMSC, which has potential for osteogenic,

chondrogenic, and adipogenic differentiation. An EGFP gene driven by a promoter for

osteogenic lineage-specific osteopontin (OPN) gene was inserted onto the 21 ∆pq HAC, and

then was transferred into hiMSC. The expression cassette was flanked by the chicken HS4

insulators to block promoter interference from adjacent drug resistant genes. The EGFP

gene was specifically expressed in the hiMSC that differentiated into osteocytes in

coordination with the transcription of endogenous OPN gene, but was not expressed after

adipogenic differentiation induction or in non-induction culture. These results suggest that

use of the HACs vector is suitable for regulated expression of transgenes in stem

cell-mediated gene therapy.

Key words. Human artificial chromosome vector ⋅ Insulator⋅ Mesenchymal stem cells ⋅ Cell

lineage-specific transgene expression ⋅ Microcell mediated chromosome transfer ⋅

Differentiation

3
INTRODUCTION

In cell-based gene therapy, human bone marrow-derived mesenchymal stem cells (hiMSC)

draw attention as a potential progenitor cell source for repair and regeneration of diverse

adult tissues, including fat, cartilage, bone, marrow stroma and skeletal muscle [1-3].

Recent studies have shown that the MSC can regenerate myocardium, liver, neural tissue and

hepatocytes by cell fusions [4-10]. Their accessibility, ease of expansion and manipulation

in vitro has made MSC as an attractive delivery vehicle for applications of cell-based gene

therapy.

Downloaded from www.StemCells.com by on September 14, 2009

Numerous studies have been performed utilizing MSC as platforms for the systemic

delivery of therapeutic genes in vivo by viral vector systems [11-17]. Transgene production

mainly achieved by the use of viral promoters evokes high-level gene expression in all cell

types transduced. However, achieving high expression is not the only goal of gene therapy

including stem cell-based gene therapy. Since inappropriate doses, timing or localization of

transgene expression can result in potentially harmful effects, providing appropriate levels of

expression of therapeutic genes along differentiation under physiological regulation in the

specialized cells is particularly important. One of the major challenges is to develop vectors

that will allow appropriate levels of spatial and temporal expression of therapeutic genes.

Human artificial chromosomes (HACs) that rely on the mechanisms governing replication

and accurate segregation of natural chromosomes, offers a new approach for creating gene

delivery vectors with potential therapeutic applications [18-20]. Since they are maintained

as independent chromosomes in host cells, they should be free from potential random

transgene integration into the host genome and silencing of transgene expression caused by

chromosomal position effect. In a previous study, we constructed a novel HAC from human

chromosome 21 by telomere-directed chromosome breakage [21]. It contains a single loxP

4
sequence for site-specific insertion of circular DNA by the Cre/loxP system. The HAC,

designated 21∆pq HAC vector, was mitotically stable in human cells and achieved

reproducible introduction and expression of EGFP gene driven by hCMV promoter. It was

proved that introduction of transgenes into specific, single, defined sites within the 21∆pq

HAC vector provides competence to create a genetic environment for reproducible transgene

expression. From the features mentioned above, the 21∆pq HAC vector may be feasible to

be developed for transgene regulation system.

Here we addressed whether the 21∆pq HAC vector can provide inducible transgene

expression, utilizing an in vitro differentiation system with a human bone marrow derived

Downloaded from www.StemCells.com by on September 14, 2009

MSC cell line. To this end, we made reporter gene constructs with EGFP gene driven by

the promoter from OPN gene [22], whose expression is upregulated along differentiation of

MSC into osteoblastic lineage [23]. Two types of 21∆pq HAC vectors, with or without

insulators in both sides of OPN-EGFP expression units were constructed and transferred into

a MSC cell line, hiMSC [24] that are capable of in vitro tri-directional differentiation into

chondrocytes, adipocytes and osteocytes in response to appropriate culture stimuli.

Expression status of the reporter gene in the hiMSC hybrids containing these HAC vectors

were tested before and after induction of osteogenic- and adipogenic -differentiation.

MATEREALS AND METHODS

Cell culture

A CHO hybrid cell lines containing the 21∆pq HAC vector [21], H8 were maintained in

Ham’s F-12 nutrient mixture (Invitrogen) supplemented with 10% calf serum (HyClone,

Logan, UT) and 8µg/ml blasticidin S hydrochloride (Funakoshi, Japan). A human

immortalized mesenchymal stem cell line (hiMSC) [24] was maintained in DMEM (Sigma,

5
St. Louis, MO) supplemented with 10% fetal bovine serum (JRH Biosciences) and incubated

at 37oC with 5% CO2.

Construction and transfection of OPN-EGFP reporter construct

The plasmid pSVmCAT-124, which contains the OPN gene promoter [22], was kindly

provided by S. Yamamoto (Oita Medical University, Japan) and pBS226 vector containing

hCMV-loxP-5’neo sequence [25] was purchased from Invitrogen.

The 250-bp insulator core of chicken β-globin 5’cHS4 region containing a CTCF binding

Downloaded from www.StemCells.com by on September 14, 2009

site was obtained from DT40 genomic DNA by a two-step PCR and three copies of the

insulator core were ligated and cloned into pCR4-TOPO (Invitrogen) to generate pIN-J2 [26].

pBS226/In-OPN-EGFP reporter construct with insulator was constructed as follow. A

750bp EcoRI fragment of pIN-J2 carrying the insulator was inserted into the EcoRI site of

pBS226, followed by changing the BamHI site into NheI site to generate pBS226/In. A

XhoI/SalI fragment carrying the 750bp insulator from pIN-J2 was inserted into the XhoI site

of pEGFP-N1 (Clontech), followed by changing the DraIII site into the SalI site to generate

pEGFP-N1/I. Then, a SalI/NheI fragment of pEGFP-N1/I was inserted into

SalI/NheI-digested pBS226/In to generate pBS226/In-EGFP. A 214-bp fragment containing

the OPN promoter was amplified from pSVmCAT-124 by PCR with the primer pair

5’-ATACCGCGGGGGGAAG- -TGTGGGAGCAG and

5’-AAAGATCTCCTTGGTCGGCGTTTGGC containing SacII and BglII restriction sites.

The PCR products were then digested with these enzymes and cloned into

SacII/BamHI-digested pBS226/EGFP-In plasmid to generate pBS226/In-OPN-EGFP, which

contains an OPN-EGFP expression cassette flanked by three copies of the 250bp core

element from cHS4 region at 5’upstream of the chicken β-globin gene.

pBS226/OPN-EGFP reporter construct without insulator was constructed as follow. A

6
1.2-kb fragment containing OPN-EGFP sequence was amplified from pBS226/

In-OPN-EGFP plasmid by PCR with the primer pair 5’-aaaggatccggggaagtgtgggagcaggt and

3’-tttgaattccctgatagacggtttttcgc containing EcoRI and BamHI restriction sites. The PCR

products were then digested with these enzymes and cloned into EcoRI/BamHI-digested

pBS226 plasmid.

4 x 105 of H8 cells were co-transfected with 2µg of reporter construct plasmids and 1µg of

Cre-expression vector pBS185 plasmid using 7.5 µl of Lipofectamine 2000 reagent according

to the supplier’s protocol (Invitrogen). After 24h of culture in basic medium, cells were

trypsinized and were cultured in the presence of G418 (800µg/ml). Drug-resistant colonies

Downloaded from www.StemCells.com by on September 14, 2009

were isolated and expanded for further analysis.

Microcell-mediated chromosome transfer

Introduction of empty, 21∆pq HAC/OPN-EGFP or 21∆pq HAC/In-OPN-EGFP into

hiMSC was performed by microcell-mediated chromosome transfer (MMCT) as described

[27] with the following modifications. Briefly, CHO hybrid cells containing 21∆pq HAC

vector were treated with 0.1 µg/ml colcemid to induce the formation of micronuclei and

centrifuged in culture flasks filled with medium containing cytochalasin B (10 µg/ml).

Recovered microcells were loaded on hiMSC and treated with 47% polyethylene glycol (MW

1000) for 1min, followed by extensive washing in serum-free DMEM. After 24 h of culture

in DMEM supplemented with 10% FBS, cells were trypsinized, and 4x106 cells were plated

into ten 90-mm dishes with selection medium containing blasticidin S hydrochloride

(4µg/ml).

7
Fluorescence in situ hybridization (FISH)

FISH was carried out according to standard protocol [27]. In single color FISH,

hChr21-derived alphoid DNA p11-4 [28] were labeled with digoxigenin-dUTP for detection

of the 21∆pq HAC vector in CHO or hiMSC hybrids. In two-color FISH, in addition to

digoxigenin-dUTP-labeled p11-4 probe (gift from H. Masumoto, Nagoya University, Japan),

pBS226/In-OPN-EFGP probe was labeled with biotin-16-dUTP for detection of the

OPN-EGFP sequence on 21∆pq HAC vector in CHO or hiMSC hybrids. Images were

captured using a fluorescence microscope (Nikon, Japan) equipped with a photometric CCD

camera, and digitally visualized with an Argus system (Hamamatsu Photonics, Japan).

Downloaded from www.StemCells.com by on September 14, 2009

Mitotic stability assay

To test the mitotic stability of 21∆pq HAC vector, the hiMSC hybrid cells were divided

into two sublines which were maintained independently either with blasticidin S

hydrochloride or without selective agent. Cells were plated 1:8 every 3 days up to

population doubling levels (PDLs) 100. At various time points, a portion of the culture was

harvested for FISH and analyzed for the presence of the 21∆pq HAC vector. Fifty

metaphases were scored at each time point.

Induction of differentiation and histochemical staining

The multi-directional differentiation experiments and histochemical staining for osteocytes,

chondrocytes and adipocytes were carried out as described [24] according to the supplier’s

protocols (Cambrex Corporation, USA).

Osteogenic differentiation: 3 x 104 cells were seeded in a 6-well plate and cultured with

2ml osteogenesis induction medium. The cultures were maintained for 3 weeks and the

culture medium was replaced every three days.

8
 Adipogenic differentiation: 2 x 105 cells were seeded in 2 ml medium of a 6-well plate.

When the cells reach confluence, three cycles of induction/maintenance culture stimulate

optimal adipogenic differentiation. Each cycle consists of 3 days of adipogenesis induction

culture, followed by 1-3 days of culture in maintenance medium. After 3 complete cycles of

induction/maintenance, the cells cultured more 7days in adipogenic maintenance medium.

Chondrogenic differentiation: Cells (2.5x105) were placed in a 15ml polypropylene tube,

centrifuged at 800rpm for 5min at room temperature, and re-suspended in 0.5 ml of

chondrogenic differentiation medium. Cells were re-centrifuged and maintained as a small

pellet for 21 days. The chondrogenesis induction medium was replaced every two days.

Downloaded from www.StemCells.com by on September 14, 2009

After 3 weeks of osteogenic-, chondrogenic- or adipogenic–induction culture, cells were

rinsed twice with PBS, and fixed with 4% paraformaldehyde for 10 min at room temperature.

Fixed cells were treated with alkaline phosphatase substrate (Vector Laboratories, INC) for

osteogenic differentiation assay, or stained with 0.3% Oil-red-O (Nakalai Tesque, Kyoto,

Japan) for the adipogenic differentiation assay. In the chondrogenic-differentiation assay,

the fixed pellet was dehydrated with a series of graded alcohol, cleaned by treatment with

xylene, and infiltrated with paraffin. Paraffin-embedded sections were stained with Alcian

blue.

Reverse transcription PCR

Total RNA was extracted by Isogen reagent (Nippon Gene, Japan). RT-PCR for detection

of OPN and PPARγ (peroxisome proliferator-activated receptor gamma) were carried out

according to a standard protocol using primers: sense: 5’-ACCTGCCAGCAACCGAAGTT,

antisense: 5’-TGGCTGTGGGTTTCAGCACT; sense: 5’-GAGCCCAAGTTTGAGTTTGC,

and antisense: 5’-CTGTGAGGACTCA -GGGTGGT.

9
RESULTS

21∆ pq HAC is stably maintained in hiMSC hybrid cells

The 21∆pq HAC vector is maintained in CHO hybrids [21] (Fig.1a). To determine

whether the 21∆pq HAC can be stably maintained in host cells throughout mitotic divisions,

we transferred the HAC from the CHO hybrids into the hiMSC by means of MMCT. Four

drug resistant clones (M8#8-1, M8#1, M8#3 and M8#4) were obtained by selection with

blasticidin S hydrochloride. In these clones, retention of the 21∆pq HAC was confirmed by

PCR amplifying the blasticidin resistant gene (data not shown). FISH analysis was

Downloaded from www.StemCells.com by on September 14, 2009

performed to test the transfer of the HAC. A single copy of 21∆pq HAC was detected in all

metaphases observed (Fig.1b). Neither insertion into host chromosome nor apparent

amplification of the 21∆pq HAC was observed. In addition, these clones showed normal

karyotype. These data suggests that 21∆pq HAC vector had been transferred into hiMSC

without introducing lesions in the host cells’ chromosomes.

To further investigate mitotic stability of the 21∆pq HAC vector in hiMSC, two sublines

from four hybrid clones were independently maintained either in the presence or absence of

blasticidin up to PDLs 100. Metaphase chromosomes were prepared at PDLs 10, 23, 49 and

100. Retention rate of the HAC vector was then analyzed by FISH. The results are

summarized in table 1. Single copy of the 21∆pq HAC vector was constantly observed in

most metaphase spreads. Loss rate of the 21∆pq HAC vector was very low at any time

point, regardless of the absence or presence of selective pressure. These results suggested

that the 21∆pq HAC vector was mitotically stable in hiMSC.

21∆pq HAC vector does not affect multipotent differentiation ability of hiMSC

The original hiMSC cell line had the ability to differentiate into adipocytes, chondrocytes

10
and osteocyte [24], but there was a possibility that in vitro serial culture and subcloning

procedures by MMCT could prevent differentiation ability of the hiMSC. To evaluate

whether the hiMSC hybrids containing the 21∆pq HAC vector still maintained its

differentiation potential, four hybrid clones (M8#8-1, M8#1, M8#3 and M8#4) at PDLs 10

were cultured in differentiation induction medium. Histochemical staining assays revealed

that these clones differentiated into adipocytes, chondrocytes and osteocytes (Fig.2),

indicating that the presence of 21∆pq HAC vector did not affect the tri-directional

differentiation potential of the hiMSC.

Downloaded from www.StemCells.com by on September 14, 2009

Site-specific insertion of the OPN-GFP reporter gene into the 21∆pq HAC and transfer

into hiMSC

To investigate whether the 21∆pq HAC vector implements regulation of a transgene

expression in differentiated MSC, we constructed an EGFP reporter gene under the control of

lineage-specific transcriptional regulatory elements. For this purpose, we chose a 0.2 kb

fragment from upstream of the human OPN gene, which showed promoter activity [22].

On the 21∆pq HAC vector, acceptor loxP site is surrounded by viral promoters for driving

drug resistant genes (Fig.3). Although the EGFP reporter gene was efficiently expressed

from the 21∆pq HAC vector housed in an HT1080 hybrid [21], there was still a possibility

that surrounding sequences may interfere with lineage-specific transcriptional regulation [29].

To prevent such interferences, in reporter construct pBS226/In-OPN-EGFP, three copies of

250bp insulator sequences from cHS4 region at 5’-upstream of chicken β-globin gene [29]

were positioned at both sides of the transcriptional units of OPN-EGFP. Another reporter

construct, pBS226/OPN-EGFP without insulator, was used as a control.

Since the hiMSC were tagged with a neo gene during the immortalization process [24],

11
G418 resistance was not applicable as a selection marker for hiMSC that underwent

site-directed insertion of the gene of interest into the 21∆pq HAC vector. We therefore

inserted the pBS226/OPN-EGFP or pBS226/In-OPN-EGFP reporter constructs into the

21∆pq HAC vector first in CHO hybrid H8, and then transferred these HAC vectors into

hiMSC by MMCT. The H8 cells were co-transfected with these reporter constructs and

Cre-recombinase expression vector (Fig.3).

When transfect the H8 cells by pBS226/In-OPN-EGFP vector, eight G418-resistant

hybrids (H8-OPN#1, #3, #4, #5, #6, #7, #8 and #9) were obtained. PCR amplifying the

OPN-EGFP expression unit was performed for identifying the insertion event on the 21∆pq

Downloaded from www.StemCells.com by on September 14, 2009

HAC vector. Three of eight clones (H8-OPN#3, #7 and #9) showed the expected size bands

(data not shown). Southern blot using a GFP probe revealed correct insertion in these three

clones (Fig.4a). In two color FISH analysis, a single independent 21∆pq

HAC/In-OPN-EGFP vector was observed (Fig.1c), suggesting that the 21∆pq HAC vector

was successfully transferred.

Among these three clones, we arbitrarily chose clone H8-In-OPN#9 as the donor of the

21∆pq HAC/In-OPN-EGFP vector and transfer this vector to hiMSC. Sixteen resistant

clones (M8-In-OPN#B-11, B-12, B-13, B-42, B-43, B-51, B-52, B-61, B-71, B-72, B-92,

B-101, C-11, E-3, E-4 and E-6) were obtained by selection with blasticidin S hydrochloride

and no clones express EGFP in maintenance culture. PCR was performed to verify intact

OPN-EGFP gene expression unit. PCR results revealed that the OPN-EGFP expression unit

was retained in all clones (data not shown). Southern blot with the EGFP probe confirmed

the presence of the In-OPN-EGFP construct in all but one clone (Fig.4b). FISH analysis

showed the presence of a single, independent 21∆pq HAC vector in the hiMSC hybrid cells

(Fig.1d). These data indicates that the intact 21∆pq HAC/In-OPN-EGFP vector was

12
successfully transferred into the hiMSC. It was noted that EGFP expression was not

observed in maintenance culture (data not shown).

On the other hand, we prepared the HAC vector without insulator. Transfection of the H8

cells with pBS226/OPN-EGFP vector yielded eight G418-resistant clones. Presence of

OPN-EGFP expression unit was confirmed by PCR (data not shown). One of these clones

was arbitrarily chosen as the donor of the 21∆pq HAC/OPN-EGFP vector and MMCT was

performed. Four drug-resistant hiMSC clones were obtained (M8-OPN#A7, #C1, #C2,

#C4). Among them, two clones (M8-OPN#C1 and #C4) expressed EGFP, even in

non-induction culture (supplementary Fig.1). The EGFP expression in clone M8-OPN#C1

Downloaded from www.StemCells.com by on September 14, 2009

and C4 may result from interference of hCMV in adjacent gene on the same HAC vector.

Thus, we used only the hiMSC hybrids, carrying the 21∆pq HAC/In-OPN-EGFP for the

following studies.

Osteocyte-specific transgene expression after differentiation of hiMSC hybrids

As described above, the hiMSC hybrids containing 21∆pq HAC vector retained the ability

to differentiate into multiple lineages. To investigate whether 21∆pq HAC vector can

mediate cell lineage-specific transgene expression, the hiMSC hybrids containing 21∆pq

HAC/In-OPN-EGFP vector were induced to differentiate into osteocytes and adipocytes.

Four hiMSC hybrid clones (M8-In-OPN#B-51, B-61, B-72 and B-92) were arbitrarily chosen,

split into three sublines and cultured independently in osteogenic-, adipogenic-induction or

non-induction medium. Evidence for osteogenic- and adipogenic differentiation was

obtained, by lineage-specific histochemical staining for alkaline phosphatase activity and

Oil-red-O staining respectively. Notably, EGFP fluorescence was only observed in the cells

cultured in osteogenic–induction culture (Fig.5a-ii). The EGFP expression level increased

13
in a time-dependent manner, and EGFP fluorescence was observed in the cell fraction

expressing alkaline phosphatase (Fig.5a-i and ii). At three weeks post-induction over 90%

of the cells expressed EGFP (Fig.5a-ii). For adipogenic differentiation, the accumulation of

Oil-red-O-positive lipid vesicles was used as a marker for adipogenic differentiation

(Fig.5c-iii). In contrast to osteogenic-induction culture, no expression of EGFP was

observed in the cells cultured with either adipogenic-induction medium (Fig.5c-ii) or

non-induction medium (Fig.5b-ii). This indicates that the EGFP reporter gene in the 21∆pq

HAC vector exhibits lineage-specific regulation.

We next investigated expression status of the endogenous OPN gene in differentiation

Downloaded from www.StemCells.com by on September 14, 2009

induction culture. RT-PCR detected OPN transcripts exclusively in osteogenic -induced

cells, in which the EGFP reporter gene was driven by an OPN promoter (Fig.6). Taken

together, our results indicate that the 21∆pq HAC vector is capable of mediating

lineage-specific transgene expression in hiMSC following in vitro differentiation.

DISCUSSION

One of the major goals of stem cell-based gene therapy is to develop vectors that will allow

appropriate levels of expression of therapeutic genes in defined target cells. In the present

study, we aimed at addressing whether the 21∆pq HAC vector could provide lineage-specific

expression of a transgene, utilizing in vitro differentiation system in a MSC cell line. Our

results demonstrated that the 21∆pq HAC vector allowed inducible expression of EGFP gene

driven by OPN promoter in hiMSC differentiated into osteogenic lineage.

The hiMSC can differentiate into osteoblasts or osteocytes without transfer of the HAC

vector in osteogenic-induction culture conditions, because the hiMSC is an established

mesenchymal stem cells line, which can differentiate into osteoblasts, chondrocytes or

14
adipocytes [24]. As shown in figure 5, almost all cells were alkaline phosphatase- and

EGFP positive, suggesting that osteogenic differentiation was induced in a high rate.

However, targets in the stem cell-based ex vivo gene therapy will be the primary culture of

bone marrow stroma-derived MSC, that are heterogeneous population constituted by a group

of cells with non-differentiation, or uni-, bi-, or tri-directional differentiation potential [24].

When a gene of interest is transduced into the primary culture MSC, there is a possibility that

the transgene does not contribute to lineage specific expression, because limited cell

population can differentiate into osteocytes. Thus, in this case, we can easily concentrate

precursor cells that differentiate into a desired direction, utilizing the EGFP tag.

Downloaded from www.StemCells.com by on September 14, 2009

When considering the stem cell-based ex vivo gene therapy, the recipient stem cells

transduced with therapeutic genes ex vivo should be propagated before autologous

transplantation into the patient. Stable maintenance of the HAC in undifferentiated MSC

during long term culture could promise substantial supply of treated cells that are competent

to express therapeutic genes after induction of appropriate differentiation. In near future

study, our aim is to harbor a tissue-specific therapeutic gene into HAC vector to achieve more

safe and effective therapy.

Recently, Vanderbyl et al. reported transfection of a mammalian artificial chromosome

(ACE) derived from mouse satellite DNA [30], carrying multiple copies of the red fluorescent

protein (RFP) reporter gene, into MSC [31]. The ACE was maintained as a single

chromosome in MSC and provided stable expression of the RFP transgene along

differentiation into adipogenic or osteogenic lineages. These results supported the possible

application of artificial chromosomes for ex vivo stem cell therapy. However, the

lineage-specific transgene expression along with differentiation was not intended in their

study. In this context, our result is the first demonstration of lineage-specific expression

induction of transgene by artificial chromosome vector in MSC.

15
 The 21∆pq HAC vector could be transduced from CHO donor hybrids into hiMSC by

MMCT. Microcell mediated chromosome transfer (MMCT) is a powerful tool for transfer

HAC into recipient cells, but also there is a possibility that the non-human genomic DNA

derived from CHO cells is concomitantly transferred into hiMSC. One supposed case is an

introduction of HAC-integrated hamster chromosome; the other is carry over of an

independent hamster chromosome with HAC. To address these issues, we first analyzed the

CHO hybrid by FISH and confirmed that the HAC was remained as an intact episomal

chromosome and that no translocation took place. After transfer of HAC into hiMSC,

obtained hybrid cells were analyzed by FISH again and were confirmed that the hiMSC

Downloaded from www.StemCells.com by on September 14, 2009

showed normal karyotype (46, x y) plus an extra HAC. These results indicated that the

carry over of hamster chromosomes was not concern.

In transgenesis, it is known that two transgenes in a single construct often interfere

mutually, when they are driven by different transcriptional elements [29]. Insulators are

endogenous cis-acting boundary elements that can protect chromatin domains from the

non-specific effects of the surrounding chromatin, by blocking the passage of regulatory

signals from adjacent loci and chromatin domains [32-35]. Hasegawa et al tested the effect

of cHS4 insulator to prevent transcriptional interference between ubiquitous and

tissue-specific transcriptional regulatory elements in the transgene study [29]. Another

study utilizing the 21∆q HAC vector showed that highly reproducible tetracycline-regulating

system was established with insulator [26]. In our study, the hiMSC hybrids containing

21∆pq HAC/OPN-EGFP expressed EGFP in non-induction culture, indicating that

transcriptional interference caused leaky transcription from the promoters that should not be

activated. In contrast, the hiMSC hybrid containing the HAC vector with insulator achieved

strict cell lineage-specific expression along with MSC differentiation. Our results support

the significance of placing insulators on the HAC vector for the precise regulation of

16
transgene expression.

For future application of the 21∆pq HAC vector for stem cell-based gene therapy, an issue

to be addressed is improvement in efficiency of transferring the HAC. The MMCT has been

adopted to transfer a single, intact chromosome from donor to recipient cells, since the risk of

truncation or rearrangement of the transferred chromosome was relatively low, compared to

other method [36, 37] such as transfection of flow-sorted chromosomes [31]. Most of

chromosome transfer experiments so far have been applied for cell lines with unlimited

replication potential. It is more recently that chromosome transfer has begun to be applied

for primary cells with limited lifespan. Therefore, experimental protocol has not been fully

Downloaded from www.StemCells.com by on September 14, 2009

optimized. In our previous study, the 21∆pq HAC vector carrying human erythropoietin

gene has been successfully transferred into primary human fibroblast [38]. Modification of

microcell fusion protocol achieved transfer efficiency up to ~10-4 order. Further efforts

have been made to optimize yet to be examined factors.

In summary, we have demonstrated the successful transcriptional regulation by the 21∆pq

HAC vector in hiMSC along differentiation. A remaining challenge is developing methods

that will allow efficient delivery of these very large molecules into primary human cells with

limited life span. Although substantial improvement in delivery method is indispensable,

the demonstration of HAC-mediated cell lineage-specific transgene expression is first step

towards the prospective use of the HACs vector in stem cell-based gene therapy.

ACKNOWLOGEMENT

We thank Hiroyuki Kugoh, Yasuaki Shirayoshi for valuable discussions. We also thank for

Hidetoshi Yamazaki for technical advice, Satoko Norikane for technical support and Candice

Ginn T. Tahimic for critical reading of the manuscript. This study was supported in part by

a Health and Labour Sciences Research Grant for Research on Human Genome, Tissue

17
Engineering from the Ministry of Health, Labour and Welfare, Japan and by the 21st Century

Program: The Research Core for Chromosome Engineering Technology.

RERERENCES

1. Jiang Y, Jahagirdar BN, Reinhardt RL et al. Pluripotency of mesenchymal stem cells

derived from adult marrow. Nature 2002; 418: 41-49.

2. Tuan RS, Boland G,Tuli R. Adult mesenchymal stem cells and cell-based tissue

engineering. Arthritis Res Ther 2003; 5: 32-45.

Downloaded from www.StemCells.com by on September 14, 2009

3. Herzog EL, Chai L,Krause DS. Plasticity of marrow-derived stem cells. Blood 2003;

102: 3483-3493.

4. Rodic N, Rutenberg MS,Terada N. Cell fusion and reprogramming: resolving our

transdifferences. Trends Mol Med 2004; 10: 93-96.

5. Terada N, Hamazaki T, Oka M et al. Bone marrow cells adopt the phenotype of other

cells by spontaneous cell fusion. Nature 2002; 416: 542-545.

6. Vassilopoulos G, Wang PR,Russell DW. Transplanted bone marrow regenerates liver by

cell fusion. Nature 2003; 422: 901-904.

7. Orlic D, Kajstura J, Chimenti S et al. Bone marrow cells regenerate infarcted

myocardium. Nature 2001; 410: 701-705.

8. Orlic D, Kajstura J, Chimenti S et al. Bone marrow stem cells regenerate infarcted

myocardium. Pediatr Transplant 2003; 7 Suppl 3: 86-88.

9. Alvarez-Dolado M, Pardal R, Garcia-Verdugo JM et al. Fusion of bone-marrow-derived

18
 cells with Purkinje neurons, cardiomyocytes and hepatocytes. Nature 2003; 425:

968-973.

10. Wang X, Willenbring H, Akkari Y et al. Cell fusion is the principal source of

bone-marrow-derived hepatocytes. Nature 2003; 422: 897-901.

11. Lee K, Majumdar MK, Buyaner D et al. Human mesenchymal stem cells maintain

transgene expression during expansion and differentiation. Mol Ther 2001; 3: 857-866.

12. Bartholomew A, Patil S, Mackay A et al. Baboon mesenchymal stem cells can be

Downloaded from www.StemCells.com by on September 14, 2009

genetically modified to secrete human erythropoietin in vivo. Hum Gene Ther 2001; 12:

1527-1541.

13. Zhang XS, Linkhart TA, Chen ST et al. Local ex vivo gene therapy with bone marrow

stromal cells expressing human BMP4 promotes endosteal bone formation in mice. J

Gene Med 2004; 6: 4-15.

14. Tsuda H, Wada T, Ito Y et al. Efficient BMP2 gene transfer and bone formation of

mesenchymal stem cells by a fiber-mutant adenoviral vector. Mol Ther 2003; 7: 354-365.

15. Chang SC, Chuang H, Chen YR et al. Cranial repair using BMP-2 gene engineered bone

marrow stromal cells. J Surg Res 2004; 119: 85-91.

16. Partridge K, Yang X, Clarke NM et al. Adenoviral BMP-2 gene transfer in mesenchymal

stem cells: in vitro and in vivo bone formation on biodegradable polymer scaffolds.

Biochem Biophys Res Commun 2002; 292: 144-152.

17. Dong WJ, Wu XB, Liu DP et al. Analysis of adeno-associated virus-mediated ex vivo

19
 transferred human beta-globin gene in bone marrow engrafted mice. J Biomed Sci 2002;

9: 253-260.

18. Mejia JE, Willmott A, Levy E et al. Functional complementation of a genetic deficiency

with human artificial chromosomes. Am J Hum Genet 2001; 69: 315-326.

19. Grimes BR, Schindelhauer D, McGill NI et al. Stable gene expression from a

mammalian artificial chromosome. EMBO Rep 2001; 2: 910-914.

20. Ikeno M, Inagaki H, Nagata K et al. Generation of human artificial chromosomes

Downloaded from www.StemCells.com by on September 14, 2009

expressing naturally controlled guanosine triphosphate cyclohydrolase I gene. Genes

Cells 2002; 7: 1021-1032.

21. Katoh M, Ayabe F, Norikane S et al. Construction of a novel human artificial

chromosome vector for gene delivery. Biochem Biophys Res Commun 2004; 321:

280-290.

22. Hijiya N, Setoguchi M, Matsuura K et al. Cloning and characterization of the human

osteopontin gene and its promoter. Biochem J 1994; 303 ( Pt 1): 255-262.

23. Zohar R, Cheifetz S, McCulloch CA et al. Analysis of intracellular osteopontin as a

marker of osteoblastic cell differentiation and mesenchymal cell migration. Eur J Oral

Sci 1998; 106 Suppl 1: 401-407.

24. Okamoto T, Aoyama T, Nakayama T et al. Clonal heterogeneity in differentiation

potential of immortalized human mesenchymal stem cells. Biochem Biophys Res

Commun 2002; 295: 354-361.

25. Fukushige S,Sauer B. Genomic targeting with a positive-selection lox integration vector

20
 allows highly reproducible gene expression in mammalian cells. Proc Natl Acad Sci U S

A 1992; 89: 7905-7909.

26. Otuki A, Tahimic C, Tomimatsu N et al. Construction of a novel expression system on a

human artificial chromosome. Biochem Biophys Res Commun 2005; 329: 1018-1025.

27. Kugoh H, Mitsuya K, Meguro M et al. Mouse A9 cells containing single human

chromosomes for analysis of genomic imprinting. DNA Res 1999; 6: 165-172.

28. Ikeno M, Masumoto H,Okazaki T. Distribution of CENP-B boxes reflected in CREST

centromere antigenic sites on long-range alpha-satellite DNA arrays of human

Downloaded from www.StemCells.com by on September 14, 2009

chromosome 21. Hum Mol Genet 1994; 3: 1245-1257.

29. Hasegawa K, Nakatsuji N. Insulators prevent transcriptional interference between two

promoters in a double gene construct for transgenesis. FEBS Lett 2002; 520: 47-52.

30. Stewart S, MacDonald N, Perkins E et al. Retrofitting of a satellite repeat DNA-based

murine artificial chromosome (ACes) to contain loxP recombination sites. Gene Ther

2002; 9: 719-723.

31. Vanderbyl S, MacDonald GN, Sidhu S et al. Transfer and stable transgene expression of

a mammalian artificial chromosome into bone marrow-derived human mesenchymal

stem cells. Stem Cells 2004; 22: 324-333.

32. Recillas-Targa F, Pikaart MJ, Burgess-Beusse B et al. Position-effect protection and

enhancer blocking by the chicken beta-globin insulator are separable activities. Proc Natl

Acad Sci U S A 2002; 99: 6883-6888.

33. Razin SV, Farrell CM,Recillas-Targa F. Genomic domains and regulatory elements

21
 operating at the domain level. Int Rev Cytol 2003; 226: 63-125.

34. West AG, Gaszner M,Felsenfeld G. Insulators: many functions, many mechanisms.

Genes Dev 2002; 16: 271-288.

35. Burgess-Beusse B, Farrell C, Gaszner M et al. The insulation of genes from external

enhancers and silencing chromatin. Proc Natl Acad Sci U S A 2002; 99 Suppl 4:

16433-16437.

36. Klobutcher LA, Miller CL,Ruddle FH. Chromosome-mediated gene transfer results in

Downloaded from www.StemCells.com by on September 14, 2009

two classes of unstable transformants. Proc Natl Acad Sci USA 1980; 77: 3610-3614.

37. de Jong G, Telenius A, Vanderbyl S et al. Efficient in-vitro transfer of a 60-Mb

mammalian artificial chromosome into murine and hamster cells using cationic lipids and

dendrimers. Chromosome Res 2001; 9: 475-485.

38. Kaketa M, Tomizuka K, Nagata K et al. Human artificial chromosome (HAC) vector

provides long-term therapeutic transgene expression in normal human primary

fibroblasts. Gene Ther 2005( in press).

Figure Legends

Fig.1. FISH analysis of CHO and hiMSC hybrid cells carrying the 21∆pq HAC vectors or

21∆pq HAC/In-OPN-EGFP vector.

Human Chromosome 21-derived alphoid DNA probe was hybridized to the 21∆pq HAC

vector in (a) H8 and (b) hiMSC hybrid cell-M8#1(red signals with arrows). In the hiMSC

hybrid cell, 21∆pq HAC vector (red signal with arrow) was identified as a chromosome

22
fragment whose size was reduced compared to intact chromosome 21. Additional red

signals were detected on the centromere region of endogenous human chromosome 21 and

chromosome 13, which possess high homology to alphoid satellite on chromosome 21.

pBS226/In-OPN-EFGP plasmid probe (green signal) was hybridized to 21∆pq

HAC/In-OPN-EGFP vector in (c) CHO hybrid cell H8-In-OPN#9 and (d) hiMSC hybrid cell

M8-In-OPN#B72. Chromosomal DNA was counterstained with DAPI (blue). The inset

showed enlargement of the 21∆pq HAC vector with (c and d) or without (a and b) the

OPN-EGFP insert. Note that the single copy of 21∆pq HAC vector was maintained

Downloaded from www.StemCells.com by on September 14, 2009

independently in CHO hybrid cells and hiMSC hybrid cell, without any translocation or

insertion to host chromosomes. Background karyotype of the host hiMSC was apparently

normal. The images representative of the results from the other hiMSC hybrid cell lines

containing 21∆pq HAC vector. Original magnification: x1600.

Fig.2. Differentiation of hiMSC hybrids containing 21∆pq HAC vector. The cells were

induced to differentiate into adipocytes, chondrocytes and osteocytes. Adipogenesis (a),

chondrogenesis (b), and osteogenesis (c) were indicated by the histochemical staining with

oil Red O, alician blue, and by the detection of alkaline phosphatase activity, respectively. It

was suggested that the introduction of the 21∆pq HAC vector did not affect the pluripotent

stem cell phenotype of hiMSC. The images show details of histochemical staining from

M8#1, representative of the results from three other hiMSC hybrid cell lines containing

21∆pq HAC vector. Original magnification: x200.

Fig.3. Site-specific insertion of pBS226/OPN-EGFP reporter construct or

pBS226/In-OPN-EGFP reporter construct into the loxP site on the 21∆pq HAC vector. In

23
pBS226/OPN-EGFP, a circular targeting construct carries EGFP gene driven by OPN

promoter; in pBS226/In-OPN-EGFP, the OPN-EGFP expression unit was flanked on both

sides with the insulators, hCMV promoter for neo gene, and the loxP sequence (top).

Cre-recombinase-mediated site-specific integration of the targeting construct into the loxP

site on the HAC vector (middle) regenerates a functional neo gene on the HAC vector. The

resulting inserted allele is shown at the bottom. Co-transfection of CHO hybrids harboring

21∆pq HAC vector with the reporter construct and Cre recombinase expression vector

yielded G418 resistant transfectants.

Downloaded from www.StemCells.com by on September 14, 2009

Fig.4. Southern analysis for site-specific insertion of pBS226/In-OPN-EGFP constructs into

the loxP site on the 21∆pq HAC in H8 cells (a) and detection of the 21∆pq

HAC/In-OPN-EGFP vector in hiMSC hybrid cells after MMCT (b). A neo probe was

hybridized to BamHI digested genomic DNA. A 3.3 kb fragment from the wild type allele

was replaced with 9.6 kb fragment in successfully targeted transfectants H8-In-OPN#3, #7

and #9 (a). An EGFP probe was hybridized to Bst XI digested genomic DNA. A 9.7 kb

fragment is detected in all successfully transferred M8-In-OPN clones except

M8-In-OPN#E-4(b).

Fig.5. Lineage-specific EGFP expression in hiMSC hybrids containing 21∆pq

HAC/In-OPN-EGFP vector after osteogenic differentiation. Representative fluorescence

and phase contrast microscopic view of hiMSC hybrids; a) after osteogenic differentiation, b)

after adipogenic differentiation, c) control culture without differentiation induction. Panel

(i): detection of red fluorescence produced by alkaline phosphatase activity. Panel (ii):

detection of green fluorescence of EGFP. Panel (iii): Phase contrast microscopic view of the

identical field as depicted in panel (i) and (ii). Adipogenic differentiation was indicated by

24
the accumulation of lipid vacuoles stained by Oil Red (b, iii). EGFP was exclusively

expressed at 3 weeks post induction of osteogenic differentiation in hiMSC hybrid cells (a-ii),

but not in undifferentiated cells (c) or cells at post induction of adipogenic differentiation

(b-ii). The images shows details of histochemical staining from M8-In-OPN#B-72,

representative of the results from other hiMSC hybrid cell lines (M8-In-OPN#B-51, B-61,

and B-92) containing 21∆pq HAC /In-OPN-EGFP vector. Original magnification: x200.

Fig.6. Detection of lineage-specific transcription of endogenous marker genes in hiMSC

hybrids by RT-PCR. OPN and PPARγ were tested as markers for osteogenic- and

Downloaded from www.StemCells.com by on September 14, 2009

adipogenic-differentiation, respectively. Three weeks post induction of osteogenic and

adipogenic differentiation, total RNA was extracted and analyzed for OPN, PPARγ, and

GAPDH expression. Lineage-specific marker genes were upregulated at transcriptional

level in the hiMSC hybrids along differentiation induction. Lane 1-4 (M8-In-OPN#B-51,

B-61, B-72 and B-92, respectively), representative of the hiMSC hybrid cells containing

21∆pq HAC/In-OPN-EGFP vector.

Supplementary Fig.1. EGFP expression in hiMSC hybrids containing 21∆pq

HAC/OPN-EGFP vector in non-induction culture. Representative result from the clone
M8-OPN#C1 is shown. (i): detection of green fluorescence of EGFP. (ii): Phase contrast
microscopic view of the identical field as depicted in (i). Original magnification: x200.
Detection of EGFP expression in non-induction culture suggested that strict lineage-specific
transgene expression was not achieved without insulator.

25
Downloaded from www.StemCells.com by on September 14, 2009

26
Downloaded from www.StemCells.com by on September 14, 2009

27
Downloaded from www.StemCells.com by on September 14, 2009

28
 Downloaded from www.StemCells.com by on September 14, 2009
Oshimura Fig.4

29
Downloaded from www.StemCells.com by on September 14, 2009

30
 Downloaded from www.StemCells.com by on September 14, 2009
Oshimura Fig.6

31
Downloaded from www.StemCells.com by on September 14, 2009

32
Table 1.
21∆pq HAC stability in the absence of selective pressure in hiMSC

21∆pq HAC in Cells

Cell line PDLs Under / Off Selection (% of Spreads)a Loss rate (%)b

# 8-1 23 82 / 80 0.1
46 82 / 78 0.1
100 80 / 80 0.0
#1 23 98 / 76 1.1
46 98 / 74 0.6

Downloaded from www.StemCells.com by on September 14, 2009

100 94 / 84 0.1
#3 23 90 / 86 0.2
46 94 / 88 0.1
100 90 / 80 0.1
#4 23 87 / 76 0.6
46 88 / 78 0.3
100 90 / 80 0.1
a
FISH analysis was performed with a hChr21-derived alphoid DNA probe p11-4; on average,
50 metaphase chromosome spreads from each clone were examined at the time points
indicated.
b
Calculated by the formula Nn=N0 x (1-R)n, where N0 is the number of metaphase spreads
containing artificial chromosomes in the cells cultured under selection, Nn is the number of
such spreads after n PDLs of culture in the absence of selection, and R is the loss rate.

33
 A Novel Human Artificial Chromosome Vector Provides Effective Cell
Lineage-Specific Transgene Expression in Human Mesenchymal Stem Cells
Xianying Ren, Motonobu Katoh, Hedetoshi Hoshiya, Akihiro Kurimasa, Toshiaki
Inoue, Fumiaki Ayabe, Kotaro Shibata, Junya Toguchida and Mitsuo Oshimura
Stem Cells published online Sep 1, 2005;
DOI: 10.1634/stemcells.2005-0021
This information is current as of September 14, 2009

Updated Information including high-resolution figures, can be found at:

& Services http://www.StemCells.com

Downloaded from www.StemCells.com by on September 14, 2009