:,/(<$87+25',6&2817&$5'




$V D KLJKO\ YDOXHG FRQWULEXWRU WR :LOH\¶V SXEOLFDWLRQV ZH ZRXOG OLNH WR VKRZ RXU
DSSUHFLDWLRQWR\RXE\RIIHULQJDXQLTXHGLVFRXQWRIIWKHSXEOLVKHGSULFHRIDQ\RI
RXUERRNV 

7R WDNH DGYDQWDJH RI WKLV RIIHU DOO \RX QHHG WR GR LV DSSO\ IRU WKH :LOH\ $XWKRU
'LVFRXQW &DUG E\ FRPSOHWLQJ WKH DWWDFKHG IRUP DQG UHWXUQLQJ LW WR XV DW WKH IROORZLQJ
DGGUHVV

7KH'DWDEDVH*URXS
-RKQ:LOH\ 6RQV/WG
7KH$WULXP
6RXWKHUQ*DWH
&KLFKHVWHU

FS
:HVW6XVVH[3264
8.


O
,Q WKH PHDQWLPH ZKHQHYHU \RX RUGHU ERRNV GLUHFW IURP XV VLPSO\ TXRWH SURPRWLRQDO
FRGH6:WRWDNHDGYDQWDJHRIWKHGLVFRXQW


O
7KHQHZHVWDQGTXLFNHVWZD\WRRUGHU\RXUERRNVIURPXVLVYLDRXUQHZ(XURSHDQZHEVLWH
DW

PR


KWWSZZZZLOH\HXURSHFRP

.H\EHQHILWVWRXVLQJWKHVLWHDQGRUGHULQJRQOLQHLQFOXGH
D
• 5HDOWLPH6(&85(RQOLQHRUGHULQJ
• 7KHPRVWXSWRGDWHVHDUFKIXQFWLRQDOLW\WRPDNHEURZVLQJWKHFDWDORJXHHDVLHU
• 'HGLFDWHG$XWKRUUHVRXUFHFHQWUH
TE

• (PDLODIULHQG
• (DV\WRXVHQDYLJDWLRQ
• 5HJXODUVSHFLDORIIHUV
• 6LJQXSIRUVXEMHFWRULHQWDWHGHPDLODOHUWV
EC


6R WDNH DGYDQWDJH RI WKLV JUHDW RIIHU UHWXUQ \RXU FRPSOHWHG IRUP WRGD\ WR UHFHLYH \RXU
GLVFRXQWFDUG

R

<RXUVVLQFHUHO\

R
O


C

9HULW\/HDYHU
(PDUNHWLQJDQG'DWDEDVH0DQDJHU
N




U



7(506$1'&21',7,216
7KLVRIIHULVH[FOXVLYHWR:LOH\$XWKRUV(GLWRUV&RQWULEXWRUVDQG(GLWRULDO%RDUG0HPEHUVLQDFTXLULQJERRNVH[FOXGLQJHQF\FORSDHGLDV
DQG PDMRU UHIHUHQFH ZRUNV IRU WKHLU SHUVRQDO XVH 7KHUH PXVW EH QR UHVDOH WKURXJK DQ\ FKDQQHO 7KH RIIHU LV VXEMHFW WR VWRFN
DYDLODELOLW\ DQG FDQQRW EH DSSOLHG UHWURVSHFWLYHO\ 7KLV HQWLWOHPHQW FDQQRW EH XVHG LQ FRQMXQFWLRQ ZLWK DQ\ RWKHU VSHFLDO RIIHU :LOH\
UHVHUYHVWKHULJKWWRDPHQGWKHWHUPVRIWKHRIIHUDWDQ\WLPH






 5(*,675$7,21)250



)25%22.',6&2817&$5'

7RHQMR\\RXUVSHFLDOGLVFRXQWWHOOXV\RXUDUHDVRILQWHUHVWDQG\RXZLOOUHFHLYHUHOHYDQWFDWDORJXHVRUOHDIOHWV
IURPZKLFKWRVHOHFW\RXUERRNV3OHDVHLQGLFDWH\RXUVSHFLILFVXEMHFWDUHDVEHORZ


$FFRXQWLQJ >@ $UFKLWHFWXUH >@

• 3XEOLF >@  
• &RUSRUDWH >@ %XVLQHVV0DQDJHPHQW >@

&KHPLVWU\ >@ &RPSXWHU6FLHQFH >@
• $QDO\WLFDO >@ • 'DWDEDVH'DWD:DUHKRXVH >@
• ,QGXVWULDO6DIHW\ >@ • ,QWHUQHW%XVLQHVV >@
• 2UJDQLF >@ • 1HWZRUNLQJ >@
• ,QRUJDQLF >@ • 3URJUDPPLQJ6RIWZDUH >@
• 3RO\PHU >@ 'HYHORSPHQW >@
• 6SHFWURVFRS\ >@ • 2EMHFW7HFKQRORJ\ >@

FS

(QF\FORSHGLD5HIHUHQFH >@ (QJLQHHULQJ >@
• %XVLQHVV)LQDQFH >@ • &LYLO >@
• /LIH6FLHQFHV >@ • &RPPXQLFDWLRQV7HFKQRORJ\ >@

O
• 0HGLFDO6FLHQFHV >@ • (OHFWURQLF >@
• 3K\VLFDO6FLHQFHV >@ • (QYLURQPHQWDO >@

O
• 7HFKQRORJ\ >@ • ,QGXVWULDO >@
 • 0HFKDQLFDO >@

PR

(DUWK (QYLURQPHQWDO6FLHQFH >@ )LQDQFH,QYHVWLQJ >@
  • (FRQRPLFV >@
+RVSLWDOLW\ >@ • ,QVWLWXWLRQDO >@
• 3HUVRQDO)LQDQFH >@

D
*HQHWLFV >@ /LIH6FLHQFH >@
• %LRLQIRUPDWLFV&RPSXWDWLRQDO >@  
TE

%LRORJ\ >@ /DQGVFDSH$UFKLWHFWXUH >@

• 3URWHRPLFV >@  
• *HQRPLFV >@ 0DWKHPDWLFV6WDWLVWLFV >@
• *HQH0DSSLQJ >@  
EC

• &OLQLFDO*HQHWLFV >@ 0DQXIDFWXULQJ >@

  
0DWHULDO6FLHQFH >@

R

0HGLFDO6FLHQFH >@ 3V\FKRORJ\ >@

• &DUGLRYDVFXODU >@ • &OLQLFDO >@
R

• 'LDEHWHV >@ • )RUHQVLF >@

• (QGRFULQRORJ\ >@ • 6RFLDO 3HUVRQDOLW\ >@
• ,PDJLQJ >@ • +HDOWK 6SRUW >@
O

• 2EVWHWULFV*\QDHFRORJ\ >@ • &RJQLWLYH >@

• 2QFRORJ\ >@ • 2UJDQL]DWLRQDO >@
C

• 3KDUPDFRORJ\ >@ • 'HYHORSPHQWDODQG6SHFLDO(G >@

• 3V\FKLDWU\ >@ • &KLOG:HOIDUH >@
N

 • 6HOI+HOS >@

1RQ3URILW >@ 3K\VLFV3K\VLFDO6FLHQFH >@
U




>@,FRQILUPWKDW,DPD:LOH\$XWKRU(GLWRU&RQWULEXWRU(GLWRULDO%RDUG0HPEHURIWKHIROORZLQJ
SXEOLFDWLRQV






6,*1$785(«««««««««««««««««««««««««««««««««««««««««««««««««««««««

3/($6(&203/(7(7+()2//2:,1*'(7$,/6,1%/2&.&$3,7$/6

7,7/($1'1$0(HJ0U0UV'U««««««««««««««««««««««««««««««««««««««««

-2%7,7/(««««««««««««««««««««««««««««««««««««««««««««««««««««««««

'(3$570(17«««««««««««««««««««««««««««««««««««««««««««««««««««««

&203$1<,167,787,21«««««««««««««««««««««««««««««««««««««««««««««««

FFSS

$''5(66««««««««««««««««««««««««««««««««««««««««««««««««««««««««

«««««««««««««««««««««««««««««««««««««««««««««««««««««««««««««««

OO

«««««««««««««««««««««««««««««««««««««««««««««««««««««««««««««««

OO

«««««««««««««««««««««««««««««««««««««««««««««««««««««««««««««««

PPRR

72:1&,7<«««««««««««««««««««««««««««««««««««««««««««««««««««««««

&2817<67$7(««««««««««««««««««««««««««««««««««««««««««««««««««««

DD
&28175<««««««««««««««««««««««««««««««««««««««««««««««««««««««««

TTEE

3267&2'(=,3&2'(«««««««««««««««««««««««««««««««««««««««««««««««««

'$<7,0(7(/««««««««««««««««««««««««««««««««««««««««««««««««««««««

EECC

)$;««««««««««««««««««««««««««««««««««««««««««««««««««««««««««««

(0$,/««««««««««««««««««««««««««««««««««««««««««««««««««««««««««
RR


<2853(5621$/'$7$
:H-RKQ:LOH\ 6RQV/WGZLOOXVHWKHLQIRUPDWLRQ\RXKDYHSURYLGHGWRIXOILO\RXUUHTXHVW,QDGGLWLRQZHZRXOGOLNHWR
RR


 8VH\RXULQIRUPDWLRQWRNHHS\RXLQIRUPHGE\SRVWHPDLORUWHOHSKRQHRIWLWOHVDQGRIIHUVRILQWHUHVWWR\RXDQGDYDLODEOHIURPXVRU
RWKHU:LOH\*URXSFRPSDQLHVZRUOGZLGHDQGPD\VXSSO\\RXUGHWDLOVWRPHPEHUVRIWKH:LOH\*URXSIRUWKLVSXUSRVH
OO


>@3OHDVHWLFNWKHER[LI\RXGRQRWZLVKWRUHFHLYHWKLVLQIRUPDWLRQ
CC

 6KDUH\RXULQIRUPDWLRQZLWKRWKHUFDUHIXOO\VHOHFWHGFRPSDQLHVVRWKDWWKH\PD\FRQWDFW\RXE\SRVWID[RUHPDLOZLWKGHWDLOVRIWLWOHV
DQGRIIHUVWKDWPD\EHRILQWHUHVWWR\RX

NN

>@3OHDVHWLFNWKHER[LI\RXGRQRWZLVKWRUHFHLYHWKLVLQIRUPDWLRQ

,IDWDQ\WLPH\RXZLVKWRVWRSUHFHLYLQJLQIRUPDWLRQSOHDVHFRQWDFWWKH'DWDEDVH*URXSGDWDEDVHJURXS#ZLOH\FRXNDW-RKQ:LOH\ 6RQV/WG
UU

7KH$WULXP6RXWKHUQ*DWH&KLFKHVWHU:HVW6XVVH[32648.

(0$,/$/(57,1*6(59,&(
:HRIIHUDQLQIRUPDWLRQVHUYLFHRQRXUSURGXFWUDQJHVYLDHPDLO,I\RXGRQRWZLVKWRUHFHLYHLQIRUPDWLRQDQGRIIHUVIURP-RKQ:LOH\FRPSDQLHV
ZRUOGZLGHYLDHPDLOSOHDVHWLFNWKHER[>@

7KLVRIIHULVH[FOXVLYHWR:LOH\$XWKRUV(GLWRUV&RQWULEXWRUVDQG(GLWRULDO%RDUG0HPEHUVLQDFTXLULQJERRNVH[FOXGLQJHQF\FORSDHGLDVDQGPDMRU
UHIHUHQFHZRUNVIRUWKHLUSHUVRQDOXVH7KHUHVKRXOGEHQRUHVDOHWKURXJKDQ\FKDQQHO7KHRIIHULVVXEMHFWWRVWRFNDYDLODELOLW\DQGPD\QRWEH
DSSOLHGUHWURVSHFWLYHO\7KLVHQWLWOHPHQWFDQQRWEHXVHGLQFRQMXQFWLRQZLWKDQ\RWKHUVSHFLDORIIHU:LOH\UHVHUYHVWKHULJKWWRYDU\WKHWHUPVRI
WKHRIIHUDWDQ\WLPH

5HI6:

Offprint order form

Journal title

Article reference number

Twenty-five offprints are supplied free of charge to the first named author, or the author named for correspondence. If
you would like to purchase further offprints of your paper and journal copies, these are available at pre-publication prices
in the quantities below. Only one contributor receives this form, so please ensure that you include the offprint and journal
requirements for all the contributors. Offprint orders received after your corrected proofs have been returned, approved
for publication, may be subject to a special reprint quotation as they may have to be printed separately and thus incur
additional print set-up costs.

Article title Billing address (if different from delivery address)

FS
Name

Address

O
O
PR
Postcode/Zip Code Postcode/Zip Code

Telephone Telephone

Fax Fax
D
E-mail E-mail
TE

Note for USA authors: please do not use PO Box numbers as delivery addresses, as we cannot guarantee delivery.

Offprint price list (based on cash with order)

EC

Number of copies 1-4 pages 5-8 pages 9-12 pages 13-16 pages 17 pages
and over
25 £75/$125 £135/$230 £195/$330 £255/$430 £315/$535
R

50 £100/$170 £180/$305 £260/$445 £340/$580 £420/$715

R

100 £130/$220 £210/$360 £290/$495 £370/$630 £460/$780

O

200 £190/$325 £290/$495 £390/$665 £490/$835 £590/$1000

300 £250/$425 £370/$630 £490/$835 £610/$1040 £730/$1240
C

400 £310/$530 £450/$765 £590/$1005 £730/$1240 £870/$1480

N

per extra 100 £60/$105 £80/$135 £100/$170 £120/$205 £140/$235

Please add 10% to prices if payment is to be made later by invoice or purchase order
U

Number of pages

Number of offprints required 25 Free+

Price (add 10% if payment is made

by invoice or purchase order)
Journal copies price list (based on cash with order)
Number of copies
1 2 3 4 5 6 7 8 9 10
£12/$20 £20/$34 £28/$48 £36/$60 £44/$75 £52/$88 £60/$102 £68/$115 £76/$130 £84/$142
Please add 10% to prices if payment is made by invoice or purchase order.

Number of journal copies required

Price (add 10% if payment is made

by invoice or purchase order)

Please ask for a special quote from Advertising Sales (contact details overleaf ) if more than 10 journal copies are required.

Payment
Please indicate the method by which you are paying. Payment should be made in sterling or another negotiable currency
at the present rate of exchange. Prices are exclusive of VAT and other taxes.

Special notes:
Canadian residents: please add 7% GST
VAT: If the institution/company/individual paying the invoice is VAT registered within the EU please supply the

FS
appropriate VAT number. If exempt from VAT please provide proof of exemption.

Method of payment

O
Paying now
■ Credit Card ■ American Express ■ Diners Club ■ JCB ■ Barclaycard/Visa ■ Mastercard/Access

O
Please indicate type of card and complete all details accurately.

PR
Card number Security No.

Expiry Date Signature _________________________________ Date _________________

D
■ Cheque, please make payable to John Wiley & Sons, Ltd.
TE

Paying later
Send Invoice ■ Purchase order number (please enclose)
EC

Amount due
£Sterling $US

Offprints
R

Journal copies
R

Tax (VAT/GST)
O

Total
C

Delivery
N

Offprints and journal copies are despatched by second class post within the UK and by accelerated despatch elsewhere.
Therefore, please allow 6 weeks from publication for delivery.
U

Where to send your order

Please return your completed order with your corrected proofs to: Journals Production Department,
John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK
or by fax to: +44 (0)1243 770379;
or to the journal Editor for the following journals:
International Journal of Climatology
Aquatic Conservation: Marine and Freshwater Ecosystems
Journal of Applied Econometrics
Regulated Rivers
Journal of Quaternary Science
 jgm706

THE JOURNAL OF GENE MEDICINE RESEARCH ARTICLE

J Gene Med 2004; 6: 000–000.
1 Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jgm.706 60
2 61
3 62
4 63
5 64
6
7
Improving gene replacement by intracellular 65
66
8
9
formation of linear homologous DNA 67
68
10 69
11 70
12 71
13 72
14 73
G. de Piédoue1 Abstract
15 74
16
R. Maurisse1,4 75
17 I. Kuzniak1 Background Gene targeting is a potential tool for gene therapy but is limited 76
18 B. Lopez2 by the low rate of homologous recombination. Using highly homologous 77

FS
19 A. Perrin3 linear DNA improves gene targeting frequency but requires microinjection 78
20 into nuclear cells to be effective. Because transfection of circular DNA is
O. Nègre1 79
21 more efficient than transfection of linear DNA and adaptable to viral vectors, 80
P. Leboulch1 we developed a system for the intracellular release of linear fragments from
22 81

O
23
J.-P. Feugeas1 * circular plasmids. 82
24 83

O
1
INSERM emi 0111, laboratoire de
25 Methods Only one cutting site inside the ‘‘donor’’ DNA was not convenient 84
Thérapie Génique Hématopoı̈étique,
26 Institut Universitaire d’Hématologie, because it led to integration of exogenous sequences into the target. So we 85

PR
27 Hôpital Saint-Louis, 1 Av. C. constructed several ‘‘donor’’ plasmids containing the homologous sequences 86
28 Vellefaux, 75010 Paris, France flanked by two I-Sce I recognition sites. Expression of I-Sce I allowed 87
29 2 intracellular delivery of ‘‘ends-out’’ (replacement) vectors. We compared 88
UMR CEA/CNRS 217, CEA, Div des
30 Sciences du Vivant, DRR, 60- 68 the efficiency of different constructions to correct a mutated gfp target. 89
31 Avenue du General Leclerc, 92265, 90
D
32 Fontenay-aux-Roses Cedex, France Results Co-transfection of ‘‘donor’’ plasmids and an I-Sce I expression vector 91
33 3 into CHO cells enhanced the correction of an extrachromosomal mutated gfp 92
TE

Direction Scientifique, Institut

34 Pasteur, 25 rue du Docteur Roux, target by at least 10 times. Maximum correction was observed with the 93
35 75724 Paris, France greatest homology size and maximum effect of I-Sce I was obtained when 94
36 4 the long hemi-sites of the duplicated I-Sce I sites were contiguous to the 95
Laboratoire de Biophysique,
EC

37 UMR8646 CNRS-Museum National homologous sequence. Unexpectedly, the reverse orientation of I-Sce I sites 96
38 d’Histoire Naturelle, U201 INSERM, provided little or no effect, probably due to the asymmetrical activity of the 97
39 43 rue Cuvier, 75231 Paris cedex 05, I-Sce I meganuclease. 98
40 France 99
R

41 Conclusions Releasing homologous DNA fragments with I-Sce I enhances 100

*Correspondence to: J.-P. Feugeas,
42 INSERM emi 0111, laboratoire de gene replacement. This work provides the basis for the future design of viral 101
R

43 Thérapie Génique Hématopoı̈ étique, vectors for gene replacement. Copyright  2004 John Wiley & Sons, Ltd. 102
44 Institut Universitaire d’Hématologie, 103
O

45 Hôpital Saint-Louis, 1 Av. C. 104

Keywords gene targeting; meganuclease; homologous recombination
46 Vellefaux, 75010 Paris, France. 105
47 E-mail: jp.feugeas@chu-stlouis.fr 106
C

48 Introduction 107
49 108
N

50 Gene targeting can be used to correct a mutation by homologous recombi- 109

51 110
U

nation between an exogenic DNA (‘‘donor’’ DNA) and the mutated locus
52 (target). It is an interesting approach for gene therapy because it can 111
53 permanently reverse genetic disorders by restoring normal protein expres- 112
54 sion with its physiological regulation. Moreover, targeting specific DNA 113
55 sequences could prevent harmful insertional mutagenesis that can result 114
56 Received: 21 May 2004 in leukemia [1,2]. Until now, the limited efficiency of homologous recombi- 115
57 Revised: 17 September 2004 nation has made clinical application impossible. Without a specific design, 116
58 Accepted: 24 September 2004 the proportion of corrected somatic cells ranges from 10−7 to 10−5 per 117
59 transfected cell [3]. However, two conditions could very much increase 118

Copyright  2004 John Wiley & Sons, Ltd.

 2 G. de Piédoue et al.

1 efficiency: (i) making a DNA strand break in the target an expression vector encoding the I-Sce I endonuclease to 41
2 DNA [4–10] and (ii) introducing a highly homologous cut the duplicated I-Sce I sites and to free the homologous 42
3 linear ‘‘donor’’ DNA into cells [11–13]. sequence inside the cells (Figure 1). This system has 43
4 We have designed a system for the intracellular not previously been described for gene targeting in 44
5 production of linear ‘‘donor’’ fragments by producing a mammalian cells, even if duplicated I-Sce I sites have 45
6 double-strand break (DSB) at each end of the homologous been used for transgenesis in fish [21], Drosophila [14], 46
7 sequence. These linear fragments are ‘‘ends-out’’ vectors and also to study end-joining [22]. Orientation of the 47
8 that can be used for gene replacement [14]. On the duplicated sites could modulate the effect of I-Sce I 48
9 other hand, only one DSB inside the homologous DNA because I-Sce I has asymmetrical recognition and activity 49
10 provides ‘‘ends-in’’ (insertional) vectors [12]. Insertional [23]. In this study, we tested two configurations with 50
11 gene targeting results in integration of the whole vector an extrachrosomal GFP target and found that one was 51
12 and generates duplication of the targeted sequence significantly more effective than the other. 52
13 [15]. These insertional vectors are extensively used to 53
14 knock out genes in mice but genomic insertion of long 54
15 stretches of heterologous sequences from the vector 55
16 backbone can interfere with normal cellular functions.
Materials and methods 56
17 Gene replacement is better adapted for gene therapy than 57
18 gene insertion and can be performed with DNA fragments Plasmids 58

FS
19 produced by PCR (polymerase chain reaction), a strategy 59
20 called SFHR (small fragment homologous replacement) The target plasmid pEGFPstop had a stop mutation 60
21 or SDF (small DNA fragment) [16]. We used these at position 810 (TAC/TGA) and was constructed by 61
22 PCR products combined with TFOs (triple helix forming PCR-mediated site-directed mutagenesis of pEGFP-C1 62

O
23 oligonucleotides) to induce homologous recombination (Clontech Inc., Palo Alto, CA, USA). 63
24 [17]. Except for small oligonucleotides, introducing a The ‘‘donor’’ plasmids pSc100, pSc200, pSc350, 64

O
25 linear ‘‘donor’’ DNA by a chemical method is much pSc500, pSc850, and pSc1400 were engineered by 65
26 less effective than transfecting circular plasmids [18] or cloning PCR products into pGEM-T (Promega, Madi- 66

PR
27 transducing cells with viral vectors [19]. Conversion of son, USA). Six pairs of primers (Table 1) amplified 100, 67
28 the normal β-globin locus to a sickle cell disease genotype 200, 350, 500, 850 and 1400 bp wild-type fragments 68
29 of progenitor hematopoietic cells (CD34+) has been centered on nucleotide 810. To directly generate PCR 69
30 successfully performed by microinjection of small DNA fragments with I-Sce I sites, synthetic I-Sce I recog- 70
31 fragments [13]. However, automation of microinjection nition sites (5
-tagggataacagggtaat-3
) were linked to 71
D
32 is difficult for routine use and methods to obtain in vivo the 5
ends of the upstream and downstream primers, 72
33 linear DNA delivery are needed. We have introduced I-Sce giving two I-Sce I sites at both ends after PCR ampli- 73
TE

34 I recognition sites into ‘‘donor’’ plasmids on both sides of fication. A KpnI site was also introduced downstream 74
35 homologous sequences. I-Sce I is a mitochondrial intron- of I-Sce I sites to facilitate cloning procedures. Two 75
36 homing meganuclease from Saccharomyces cerevisiae that synthetic oligonucleotides (Proligo Inc., France) 5
- 76
CCGTGACCACCCTGACCTACGGCGTGCAGTGCGGTAC-3

EC

37 cleaves a 18-bp site not present in the human genome 77

38 [20]. Due to its high affinity for its recognition site, I-Sce I and 5
-CGCACTGCACGCCGTAGGTCAGGGTGGTCACGG- 78
39 can selectively produce broken-ended extrachromosomal GTAC-3
were hybridized and cloned into pSK850 79
40 DNA molecules. We co-transfected ‘‘donor’’ plasmids with digested by KpnI to construct pSc30. The double cleavage 80
R

“target” plasmid “corrected” plasmid

R

pCMV-Sce pEGFPstop
O
C

fluor escence
I-Sce I meganuclease 3’
3’
N

EGFP sequence
I-Sce I site I-Sce I site
U

cleavage
3’
donor plasmid 3’

Figure 1. I-Sce I system used to free the linear homologous DNA fragment. pCMV-Sce is an expression vector of I-Sce I and is
co-transfected together with ‘‘donor’’ plasmids and pEGFPstop. I-Sce I cuts the duplicated I-Sce I sites, releasing homologous DNA
that can replace the mutated sequence on pEGFPstop. Fluorescent cells were seen with epifluorescence microscopy and quantified
by FACS. No fluorescence was observed in the absence of pEGFPstop or of ‘‘donor’’ plasmids

Copyright  2004 John Wiley & Sons, Ltd. J Gene Med 2004; 6: 000–000.
 Gene Targeting by In Vivo Cutting DNA 3

Table 1. Primers used to amplify and clone the homologous GFP sequence into
pGEM-T and to obtain ‘‘donor’’ pSc and pScRev plasmids

Plasmids Primers Primer sequences

pSc100 SKL100 5
-tagggataacagggtaatggtaccccaccggcaagctgcccgtg-3

SKR100 5
-tagggataacagggtaatggtacctcgtgctgctt catgtggtc-3

pSc200 SKL200 5
-tagggataacagggtaatggtaccgggcgagggcg atgccacct-3

SKR200 5
-tagggataacagggtaatggtaccagatggtgcgc tcctggacg-3

pSc350 SKL350 5
-tagggataacagggtaatggtaccgggcgagggcg atgccacct-3

SKR350 5
-tagggataacagggtaatggtacctgtggctgttgtagttgtac-3

pSc500 SKL500 5
-tagggataacagggtaatggtacctagggataacagggta-3

SKR500 5
-tagggataacagggtaatggtacctgtggctgttgtagttgtac-3

pSc850 SKL850 5
-tagggataacagggtaatggtacccggtgggaggtctatataagc-3

SKR850 5
-tagggataacagggtaatggtaccccgggcccgcggtaccgtcg-3

p850 L850 5
-cggtgggaggtctatataagc-3

R850 5
-ccgggcccgcggtaccgtcg-3

pScRev850 SKRevL850 5
-attaccctgttatccctaggtacccggtgggaggtctatataagc- 3

SKRevR850 5
-attaccctgttatccctaggtaccccgggcccgcggtaccgtcg- 3

p1400 L1400 5
-gggcgagggcgatgccacct- 3

R1400 5
-tacatatttgaatgtatttag- 3

pSc1400 SKL1400 5
-tagggataacagggtaatggtaccgggcgagggcgatgccacct- 3

SKR1400 5
-tagggataacagggtaatggtacctacatatttgaatgtatttag-3

5
-attaccctgttatccctaggtaccgggcgagggcg atgccacct-3

FS
pScRev1400 SKRevL1400
SKRevR1400 5
-attaccctgttatccctaggtacctacatatttgaatgtatttag-3

O
1 of the pSK plasmids by I-Sce I was tested by in vitro diges- pEGFPstop) with 2 µg of polyethyleneimine (PEI Exgen 38
2 tions with I-Sce I (New England Biolabs Inc., Beverly, MA, 500; Euromedex, Souffelweyersheim, France). Overall 39

O
3 USA) followed by agarose gel electrophoresis (Figure 3). DNA quantities were kept constant with an empty 40
4 All plasmids were sequenced to check the absence of plasmid (pBluescript II KS(+); Stratagene). After 2 h 41
incubation at 37 ◦ C with DNA and PEI in opti-MEM

PR
5 artifactual mutations due to PCR amplifications or recom- 42
6 binations (sequencing was carried out by Genome Express 1 (Invitrogen, Paisley, UK) the cells were washed 43
7 Inc., Paris, France). and 1 ml of fresh medium added. Then, 48 h after 44
8 The pCMV-Sce expression vector has been described transfection, the efficiency of extrachromosomal gene 45
9 elsewhere [4] and was a gift from B. Lopez (CEA, correction was assessed by dividing the percentage of 46
D
10 Fontenay-aux-roses, France). It contained a modified fluorescent cells with pEGFPstop in the presence of 47
11 form of yeast I-Sce I [24], a simian virus 40 (SV40) the ‘‘donor’’ DNA by the percentage of fluorescent cells 48
TE

12 nuclear localization signal (NLS), and an HA epitope obtained with pEGFP-C1 under the same conditions. We 49
13 tag. Expression of I-Sce I after transfection of CHO-K1 did not take into consideration the loss of cells due to 50
14 cells was controlled by Western blotting according to the transfection procedure, assuming that the loss was 51
15 standard procedures (Western blotting kit; Lumi-Light similar with pEGFPstop and pEGFP-C1. Fluorescent cells 52
EC

16 Plus, Boehringer, Mannheim, Germany). I-Sce I cleavage were quantified by fluorescence-activated cell sorting 53
17 was controlled by Southern blot after DNA extraction (FACS) (Becton-Dickinson) (Figure 2). Epifluorescence 54
18 from CHO cells by a modified alkaline lysis procedure microscopy (DM-IL; Leica Microsystems) was used to 55
19 as described by Wang et al. [25]. A 32 P DNA probe rapidly control transfection and to confirm the absence 56
R

20 was produced by PCR from pEGFP-C1 with primers of fluorescent cells in negative controls (transfections 57
21 L850/R850 and labeled with a random primed DNA without target plasmids). Each point was performed in 58
R

22 labeling kit (Roche Diagnostics, Meylan, France). triplicate. 59

23 In some cases we performed plasmid extractions as 60
O

24 described by Wang et al. [25], followed by enzymatic 61

Cell cultures, transfections and
25 digestion with Dde I endonuclease (that did not cut the 62
correction efficiency
C

26 corrected sequence), PCR amplification and cloning of 63

27
CHO-K1 cells (gift from B. Lopez, CEA, Fontenay-aux- PCR products into pGEMT . We sequenced some of the 64
N

28 positive clones to confirm correction (Genome Express 65

roses, France) were grown in minimum essential medium-
29 Inc.). 66
alpha (Gibco BRL) supplemented with 10% fetal bovine
U

30 67
serum (Hy Clone; Perbio Science, South Logan, UT, USA),
31 68
100 U/ml streptomycin, 100 U/ml penicillin and 2 mM
32
glutamine. Results 69
33 70
Transfections and evaluation of correction efficiency
34 71
were carried out as previously described [17]. Briefly, Experimental design
35 2 × 105 cells were seeded per well in a 24-well 72
36 plate 24 h before transfection. Cells were transfected Co-transfections of ‘‘donor’’ DNA and ‘‘target’’ DNA 73
37 by 1.2 µg of DNA (pCMV-I-Sce-I, ‘‘donor’’ DNA and (pEGFPstop) into CHO cells provided an episomal assay 74

Copyright  2004 John Wiley & Sons, Ltd. J Gene Med 2004; 6: 000–000.
 4 G. de Piédoue et al.

I allowed intracellular delivery of linear homologous DNA 34

after cleavage of two I-Sce I recognition sites at both sides 35
A M1
of the homologous sequence (Figure 1). Forty-eight hours 36
after transfection, the ‘‘correction’’ was estimated as the 37
percentage of fluorescent cells divided by the percentage 38
of transfected cells (Figure 2). The pCMV-Sce expression 39
cell number

B 40
vector had an NLS sequence and we previously used
41
it for the in vivo cleavage of one I-Sce I site that was
42
artificially introduced into genomic or extrachromosomal
43
DNA [26]. We developed a Southern-blot-based strategy 44
to show that I-Sce I actually cleaved the ‘‘donor’’ plasmid 45
inside the cells. In vitro digestion of large amounts of 46
‘‘donor’’ plasmid by I-Sce I resulted in partial degradation 47
C
(Figure 3) but, 24 h after co-transfection, the quantities 48
of intact plasmid were very low (while quantities of intact 49
101 102 103 104
plasmid were high after transfection without pCMV-I- 50
FL-1
Sce I) (Figure 4). The linear fragments were present 51

FS
Figure 2. Example of FACS data obtained with pEGFPstop in the cells although we could not accurately quantify 52
(A), pEGFP-C1 (B) and after correction of p-EGFPstop (C). The the cumulative amounts of linear ‘‘donor’’ fragments due 53
superimposed histograms show cell number vs. GFP fluorescence 54
to a progressive release concomitant with a progressive
intensity (FL1). ‘‘Correction’’ was obtained by the ratio of the 55

O
number of fluorescent cells (in the M1 region) obtained after degradation. The effect of I-Sce I was estimated by
56
correction (C) and the number of fluorescent cells obtained the ratio of ‘‘corrected’’ cells with or without I-Sce I.
with pEGFPC1 (B) transfected under the same conditions. This 57

O
Several constructions were engineered that varied the
example is one of the results obtained with pSc1400 and 58
pCMV-I-Sce I size of the homologous sequence and the orientation
59

PR
of the I-Sce I sites (Figure 3). The effect of I-Sce I is 60
shown in Figure 5 with ‘‘donor’’ plasmids containing 850 61
1 for gene targeting [17]. The ‘‘donor’’ DNA did not bp identical to the eGFP sequence. The percentage of 62
2 have promoter sequences and could not express green ‘‘corrected’’ cells was 12.6 ± 1.5 times greater with double 63
3 fluorescent protein (GFP). The ‘‘target’’ DNA had a stop I-Sce I cleavage (pSc850 with pCMV-Sce) than with no 64
D
4 mutation and could provide a functional GFP only if there cleavage (pSc850 without pCMV-Sce or p850 with or 65
5 was recombination with ‘‘donor’’ DNA. Expression of I-Sce without pCMV-Sce). 66
TE

6
7
a b “donor” plasmi
8
EC

9
10 pGEM-T + “donor” DNA
11 p1400
12 pScRev 1400
pSc1400
R

13
14 pScRev 850
pSc850
R

15
16 pSc500
17
O

pSc350
18
pSc200
19
C

pSc100
20
pSc30
21
N

22 1 2 3 4 5 6 7
23 pEGFP-stop
U

24 “target” plasmid
25
26 Figure 3. ‘‘Donor’’ plasmids constructed for gene targeting. (a) Gel electrophoresis after in vitro digestion of ‘‘donor’’ plasmids
27 with I-Sce I. Lane 1: pSc30; lane 2: pSc100; lane 3: pSc350; lane 4: pSc500; lane 5: pSc850 or pScRev850; lane 6: pSc1400 or
pScRev1400. pSc and pScRev plasmids provided the same fragments after I-Sce I digestion. (b) Size of the sequences homologous to
28
egfp for the different ‘‘donor’’ plasmids. Egfp fragments were produced by PCR except for pSc30 which was obtained with synthetic
29 oligonucleotides. Gfp fragments were cloned into pGEMT to construct ‘‘donor’’ plasmids. stop mutation. I-Sce I recognition
30 site with the long hemi-site on the right side. : I-Sce I recognition site with the long hemi-site on the left side. egfp gene.
31 : backbone sequences of pEGFPC1
32
33 Copyright  2004 John Wiley & Sons, Ltd. J Gene Med 2004; 6: 000–000.
 Gene Targeting by In Vivo Cutting DNA 5

numerous data showing enhancement of homologous 60

recombination after I-Sce I double-strand breaks from 61
yeast to humans [8,27,28]. 62
When the amounts of ‘‘donor’’ (Figure 6b) or target 63
plasmid (Figure 6c) were increased, the percentage of 64
fluorescent cells also increased. I-Sce I was apparently 65
more effective when the target plasmid concentrations 66
were low (Figures 5 and 6c). When pEGFPstop quantities 67
were very low (data not shown), the percentage of 68
fluorescent cells reached detection limits although the I- 69
Sce I effect increased. When there were large quantities of 70
target plasmids (Figure 6c), the percentage of ‘‘corrected’’ 71
cells was high and the effect of I-Sce I effect seemed to 72
be reduced (4× or less). However, the assay was not 73
quantitative with large amounts of the target plasmid. 74
Indeed, in this case, the number of extrachromosomal 75
76
Figure 4. Cleavage of ‘‘donor’’ plasmids by I-Sce I. Southern blot
Correction 77

FS
25% 78
was performed with a 32 P gfp probe produced by PCR. Lane
1: pSc850; lane 2: pSc850 after in vitro cleavage by I-Sce I; 20% 79
lane 3: plasmid extracts from CHO cells 24 h after transfection 80
with pSc850; lane 4: plasmid extracts from CHO cells 24 h after 15% 81

O
co-transfection with pSc850 and pCMV-I-Sce I
10% 82
83

O
2.0% 5%
With I-Sce I 84
1.8% Without I-Sce I
0% 85

PR
1.6% 0 0.01 0.1 0.2 0.3 0.4 µg
86
1.4% a Variable quantities ofp CMV-Sce
0.3 µg pSc850 and 0.3 µgpEGFPstop
87
1.2%
Correction

88
1.0% 89
With I-Sce I
0.8% 14% Without I-Sce I 90
D
0.6% 12% 91
0.4% 10% 92
TE

0.2% 8% 93
0.0% 6% 94
p850 pSc850
4% 95
(0.05 µg of “target” plasmid and 0.4 µg of “donor” plasmid)
EC

2% 96
Figure 5. Percentage of corrected cells after transfections with 0%
97
(
) and without () pCMV-I-Sce I. ‘‘Correction’’ was obtained by 0.2 µg 0.4 µg 0.7 µg 1.3 µg 98
the ratio of the number of fluorescent cells obtained after trans- b Variable quantities of donor plasmid pSc850 99
fection with (pEGFPstop + donor plasmids ± pCMV-Sce) and the 0.15 µg of pEGFPstop and 0.1 µgpCMV-Sce
R

100
number of fluorescent cells obtained with (pEGFPC1 + donor 3.5%
plasmid ± pCMV-Sce). pEGFPstop was the target. p850 and With I-Sce I 101
R

pSc850 were the ‘‘donor’’ plasmids and had 850 bp homolo- 3.0% Without I-Sce I 102
gous to pEGFPstop. p850 had no I-Sce I recognition site. pSc850 103
had two I-Sce I recognition sites at both sides of the homolo- 2.5%
O

104
gous sequence 2.0% 105
106
C

1.5%
1 Relative concentrations of I-Sce I, 107
2 ‘‘donor’’ and target plasmids 1.0%
108
N

3 0.5% 109
4 We performed transfections with different amounts of the 110
U

5 0.0%
I-Sce I expression vector (Figure 6a). The maximal effect 0.02 µg 0.05 µg 0.1 µg 0.2 µg 111
6 was obtained with small amounts of pCMV-Sce which c 112
Variable quantities of the “target” pEGFPstop
7 is consistent with the high affinity of this endonuclease 0.4 µg of pSc850 and 0.1 µg ofp CMV-Sce 113
8 for its target site [23]. With high amounts of pCMV-Sce 114
9 there was no stimulation perhaps because high doses of Figure 6. Correction with variable quantities of pCMV-Sce (a),
pSc850 (b) and pEGFPstop (c). In each case, co-transfections
115
10 pCMV-Sce have an indirect effect and lower expression of 116
of CHO cells were performed with different concentrations
11 other genes (including the target gene). A direct negative of only one plasmid, the two other plasmid concentrations 117
12 interaction could not be excluded but does not support remained constant 118
13
14 Copyright  2004 John Wiley & Sons, Ltd. J Gene Med 2004; 6: 000–000.
15
16
17
18
 6 G. de Piédoue et al.

1 targets in each transfected cell was high and probably a I-Sce I recognition site
2 more than one target per cell was corrected. As the
3 number of fluorescent cells did not correlate with the
5' −TAGGGATAACAGGGTAAT−3'
4 number of corrected targets per cell, correction could
3' −ATCCCTATTGTCCCATTA−5'
5 not be quantified when high target values were used.
6 With the highest amounts of targets, the percentage of Long hemi-site
7 corrected cells even reached 100% (data not shown)
8 and no comparative analysis could be performed. For b pSc and pScRev plasmids
9 subsequent tests we limited doses of targets to obtain
10 correction below 10%. pSc
11
12
13 Size of the homologous sequence of
14 ‘‘donor’’ DNA pScRev
15
16 Seven EGFP fragments flanked by I-Sce I sites were cloned
17 into pGEM-T to construct seven ‘‘donor’’ plasmids called c I-Sce I effect
18 pSc (from pSc30 to pSc1400) (Figure 3). The homologous 8

FS
19 fragment in pSc1400 had 1400 bp identical to pEGFP- 7
20 C1 including the EGFP gene (without the 100 first 6
21 nucleotides) and 765 nucleotides downstream from that 5
22 gene. The smallest ‘‘donor’’ plasmid, pSc30, had only 30 4

O
23 bp identical to pEGFP-C1 including the site of the opal 3

24 mutation of pEGFPstop. Correction increased with the 2

O
1
25 size of the ‘‘donor’’ fragment (Figure 7). The effect of I-Sce
0
26 I was also more significant with large than with small pSc1400 pScRev1400 p1400 pSc850 pScRev850 p850

PR
27 sizes where there was little correction.
28 Figure 8. Role of the orientation of I-Sce I recognition sites
in the ‘‘donor’’ plasmids. (a) I-Sce I recognition site. The two
29
30
Orientation of the I-Sce I sites in the vertical arrows indicate positions of cleavage. (b) Orientation of
I-Sce I sites in pSc and pScRev plasmids. (c) I-Sce I effect with
31 ‘‘donor’’ plasmid
pSc and pScRev plasmids. I-Sce I effect was the ratio between
D
32 correction with pCMV-Sce and correction without pCMV-Sce.
33 The I-Sce I recognition site is asymmetrical (Figure 8a) pSc1400 and pSc850 had two I-Sce I recognition sites with
TE

34 and duplicated sites can form direct (DR) or inverted the long hemi-site contiguous to the homologous sequence.
repeats (IR) in ‘‘donor’’ plasmid. We did not construct pScRev1400 and pScRev850 had the reverse configuration with
35 the long hemi-site contiguous to the vector backbone. p1400
36 ‘‘donor’’ plasmids with DR sites because I-Sce I cutting
and p800 had no I-Sce I recognition site
of these constructions would provide linear fragments
EC

37
38 with cohesive ends that favor concatemerization. Indeed,
other authors have found that DR I-Sce I sites result in integration of concatemers into genomic DNA when 60
39
integration loci with IR constructions contained only one 61
40
or a few copies of the transgene in tandem ( [21] and J.-S. 62
R

41 Correction
14% Joly, personal communication). After I-Sce I cutting, one 63
42 with I-Sce I
of the cleavage products is longer than the other and has 64
R

43 12% without I-Sce I

a high affinity for I-Sce I [23,29] (Figure 8a). In ‘‘donor’’ 65
44 10% plasmids, this ‘‘long’’ hemi-site was contiguous to the 66
O

45
8% homologous sequence (pSc plasmids) or to the plasmid 67
46
backbone (pScRev plasmids) (Figure 8b). We observed 68
47
C

6%
that the effect of I-Sce I was more marked with pSc1400 69
48 4% than with pScRev1400 plasmids (Figure 8c) and that no 70
49
N

2% effect was obtained with pScRev850. We concluded that 71

50
the configuration of I-Sce I sites in pScRev plasmids was 72
51 0%
U

pSc30 pSc100 pSc200 pSc350 pSc500 pSc850 pSc1400 not effective for gene targeting. 73
52
74
53 Figure 7. Correction of pEGFPstop (0.10 µg) with (
) and
75
54 without () pCMV-I-Sce I (0.01 µg) with different sizes of
the ‘‘donor’’ sequence. pSc plasmids (1.3 µg) had two I-Sce I Discussion 76
55
recognition sites on both sides of the ‘‘donor’’ sequence. The long 77
56
hemi-site was contiguous to the homologous sequence. pSc30, 78
57 pSc100, pSc200, pSc350, pSc500, pSc850 and pSc1400 had 30,
In this study, ‘‘donor’’ plasmids (pSc plasmids) were
79
58 100, 200, 350, 500, 850, and 1400 base pairs homologous to designed with homologous sequences bordered by two I-
80
59 pEGFP, respectively Sce I recognition sites to release linear substrates into cells
81
82
Copyright  2004 John Wiley & Sons, Ltd. J Gene Med 2004; 6: 000–000.
83
84
85
86
 Gene Targeting by In Vivo Cutting DNA 7

1 for gene targeting after cleavage by I-Sce I meganuclease. vector (pScRev plasmids). No effect was observed with 60
2 The percentage of fluorescent cells after co-transfection pScRev850 and there was little stimulation (2×) with 61
3 of CHO-cells with pSc plasmids and pEGFPstop (target) pScRev1400. This was unexpected because I-Sce I cutting 62
4 provided a semi-quantitative assay of episomal gene of both pSc and pScRev plasmids was effective. This 63
5 targeting. The effect of intracellular linearization of discrepancy might be explained by the asymmetric activity 64
6 the ‘‘donor’’ sequence was estimated by ratios of the of I-Sce I. After cleavage of pSc plasmids, I-Sce I might 65
7 number of fluorescent cells with or without the pCMV- stay on the high affinity hemi-site (Figure 8) at the end of 66
8 Sce expression vector. I-Sce I expression increased the the ‘‘donor’’ sequence [23]. For pScRev plasmids, I-Sce I 67
9 percentage of corrected cells by at least 10 times when the might remain bound to the linearized backbone vector but 68
10 concentration of extrachromosomal targets (pEGFPstop) not to the ‘‘donor’’ fragment. Binding of I-Sce I probably 69
11 was low. The efficiency of low concentrations of target does not prevent homologous replacement because even 70
12 was encouraging for use to target unique chromosomal a covalent binding of proteins such as spo11 at the 71
13 genes. The homologous sequence had to be more than DNA termini in meı̈ osis does not prevent homologous 72
14 100 bp long for a clear-cut improvement by I-Sce I recombination [32]. Conversely, I-Sce I could partially 73
15 (Figure 7). The effect of I-Sce I was maximal with pSc500 protect linear ‘‘donor’’ fragments against degradation 74
16 and did not significantly vary above a homology of 500 and/or religation by non-homologous end-joining (NHEJ) 75
17 bp. Gene correction (with or without I-Sce I) continued to [21]. In Drosophila, Gong and Golic [14] have developed 76
18 increase with the size of homology, and was most effective a similar cleavage system to perform gene replacement 77

FS
19 with the pSc1400 plasmid. The increased efficiency of by transgenic expression of FLP site-specific recombinase 78
20 gene targeting with long fragments is in support of and I-Sce I endonuclease. They observed that I-Sce I alone 79
21 the intracellular release of linear DNA because chemical could not excise the DNA ‘‘donor’’ from its chromosomal 80
22 transfection of long linear fragments is not efficient [30]. location when ‘‘donor’’ sequences were flanked by two 81

O
23 Microinjection avoids this limitation, providing highly I-Sce I sites. This apparently conflicting result could 82
24 effective gene targeting [13], but is well adapted only be due to an inherent difference between plasmid and 83

O
25 for transfecting individual adherent cells. Anyway, the chromosomal events or the use of different configurations 84
26 use of the I-Sce I system could still be beneficial with of the duplicated I-Sce I sites. Gong and Golic engineered 85

PR
27 microinjection because it may reduce the copy number IR constructions similar to our pScRev plasmids but 86
28 of illegitimate integrated concatemers, as shown by Joly not to the pSc plasmids (Gong and Golic, personal 87
29 and co-workers after microinjection into fish embryos communication); they also used DR constructions that 88
30 [21]. Those authors suggested that the co-injected I- we did not test but that might produce concatemers [21]. 89
31 Sce I meganuclease counteracted the endogenous ligase Taken together, these data suggest that the effect of I-Sce 90
D
32 activity, preventing the generation of long concatemers I depends on the orientation of the I-Sce I recognition 91
33 found upon injection of circular or linear DNA. sites. Other studies have to be performed to determine if 92
TE

34 The linearized ‘‘donor’’ fragments produced by I- a double action of I-Sce I and FLP could be advantageous 93
35 Sce I were identical to their targets except for two in our system. 94
36 nucleotides in the center of the fragment (corresponding Our results show that releasing homologous ‘‘donor’’ 95
EC

37 to the stop mutation) and 14 nucleotides at the DNA fragments with I-Sce I enhances extrachromosomal 96
38 DNA ends corresponding to the I-Sce I hemi-sites gene replacement. It remains to be determined if it could 97
39 (Figure 8a). I-Sce I generated 4-bp staggered cuts with also be effective for targeting genomic DNA into primary 98
40 3’OH overhangs and the terminii were not homologous cells. This implies adapting this system to viral vectors. 99
R

41 to the target but were probably degraded before or Anglana and Bacchetti have previously constructed a 100
42 during replacement. Elimination of heterologous ends recombinant adenovirus for efficient delivery of I-Sce I to 101
R

43 has been shown with ‘‘ends-out’’ vectors designed for human cells [33]. Such a vector could be co-transferred 102
44 use in positive-negative selection schemes [31]. These with another viral vector that would bring the ‘‘donor’’ 103
vectors terminated with long heterologous sequences that sequence (flanked by the two I-Sce I sites). Adenoviral
O

45 104
46 encoded a counterselectable marker that was removed vectors are good candidates for gene targeting because 105
47 during homologous recombination. Here, heterology only they may provide great quantities of extrachromosomal 106
C

48 corresponded to 14 nucleotides and the effect of the I-Sce ‘‘donor’’ DNA. Alternatively, lentiviral vectors are actually 107
49 I effect was not less effective for pSc500 (DNA terminii among the best vectors for transferring DNA into 108
N

50 inside the coding region of egfp) than for pSc850 (DNA hematopoietic progenitor cells [19]. As integrase defective 109
51 terminii outside the coding region of egfp). Therefore, lentiviral vectors provide episomal DNA [34], it would 110
U

52 even if linear fragments with completely homologous be possible to use them to deliver ‘‘donor’’ DNA into 111
53 ends could be delivered using meganucleases that cleave nuclears. For optimal efficiency, this system should be 112
54 DNA outside the enzyme recognition sequences, it does coupled to specific modifications of the target DNA such 113
55 not seem to be necessary. as double-strand breaks (DSBs) [4,5,7,8] or triple helix 114
56 The effect of I-Sce I was more significant when the formations with TFOs (triplex forming oligonucleotides). 115
57 ‘‘long’’ high affinity hemi-sites of the duplicated I-Sce TFOs can also be tethered to a cutting molecule in order 116
58 I sites were contiguous to the homologous sequence to induce DSBs [35]. TFOs should be simultaneously 117
59 (pSc plasmids) and not contiguous to the backbone transferred with adenoviral vectors by PEI adenofection. 118

Copyright  2004 John Wiley & Sons, Ltd. J Gene Med 2004; 6: 000–000.
 8 G. de Piédoue et al.

1 [36]. In the case of lentiviral vectors, transfection of 18. Cherng JY, Schuurmans-Nieuwenbroek NM, Jiskoot W, et al. 60
2 Effect of DNA topology on the transfection efficiency of poly((2- 61
oligonucleotides should be carried out 24 h after lentiviral
dimethylamino)ethyl methacrylate)-plasmid complexes. J
3 transduction that permeabilizes the nuclear membrane Control Release 1999; 60: 343–353. 62
4 [37]. So, adaptation of our system to viral vectors 19. Pawliuk R, Westerman KA, Fabry ME, et al. Correction of sickle 63
5 cell disease in transgenic mouse models by gene therapy. Science 64
would still permit the combination of an intracellular
2001; 294: 2368–2371.
6 linearization of the ‘‘donor’’ DNA with specific lesions of 20. Dujon B. Sequence of the intron and flanking exons of the 65
7 the chromosomal target. mitochondrial 21S rRNA gene of yeast strains having different 66
8 alleles at the omega and rib-1 loci. Cell 1980; 20: 185–197. 67
21. Thermes V, Grabher C, Ristoratore F, et al. I-SceI meganuclease
9 mediates highly efficient transgenesis in fish. Mech Dev 2002; 68
Acknowledgements
10 118: 91–98. 69
11 22. Ricchetti M, Dujon B, Fairhead C. Distance from the 70
This work was supported by INSERM (Institut National de la chromosome end determines the efficiency of double strand
12 Santé et de la Recherche Médicale), AFM (Association Française break repair in subtelomeres of haploid yeast. J Mol Biol 2003; 71
13 contre les Myopathies), ARC (Association pour la Recherche 328: 847–862. 72
14 contre le Cancer) and Ligue contre le cancer. 23. Perrin A, Buckle M, Dujon B. Asymmetrical recognition and 73
activity of the I-SceI endonuclease on its site and on intron-exon
15 junctions. EMBO J 1993; 12: 2939–2947. 74
16 24. Choulika A, Perrin A, Dujon B, et al. The yeast I- 75
Sce I meganuclease induces site-directed chromosomal
17 References recombination in mammalian cells. Mol Cell Biol 1994; 317:
76
18 1013–1019. 77

FS
19 1. Check E. Second cancer case halts gene-therapy trials. Nature 25. Wang G, Levy DD, Seidman MM, et al. Targeted mutagenesis 78
20 2003; 42: 305. in mammalian cells mediated by intracellular triple helix 79
2. Li Z, Dullmann J, Schiedlmeier B, et al. Murine leukemia formation. Mol Cell Biol 1995; 15: 1759–1768.
21 induced by retroviral gene marking. Science 2002; 296: 497. 26. Lambert S, Lopez BS. Characterization of mammalian RAD51 80
22 double strand break repair using non-lethal dominant-negative 81

O
3. Yanez RJ, Porter AC. Therapeutic gene targeting. Gene Ther
23 1998; 5: 149–159. forms. Embo J 2000; 19: 3090–3099. 82
4. Rouet P, Smih F, Jasin M. Expression of a site-specific 27. Plessis A, Perrin A, Haber JA, et al. Site-specific recombination
24 determined by I-Sce I, a mitochondrial group I intron-encoded 83

O
endonuclease stimulates homologous recombination in
25 mammalian cells. Proc Natl Acad Sci U S A 1994; 91: 6064–6068. endonuclease expressed in yeast nucleus. Genetics 1992; 130: 84
26 5. Rouet P, Smih F, Jasin M. Introduction of double-strand breaks 451–460. 85

PR
into the genome of mouse cells by expression of a rare cutting- 28. Cohen-Tannoudji M, Robine S, Choulika A, et al. I-Sce I induced
27 endonuclease. Mol Cell Biol 1994; 14: 8096–8106. gene replacement at a natural locus in embryonic stem cells. 86
28 6. Rouet P, Smih F, Jasin M. Induction of homologous Mol Cell Biol 1998; 18: 1444–1448. 87
29 recombination in mammalian chromosomes by using the I-Sce 29. Beylot B, Spassky A. Chemical probing shows that the intron- 88
I system of Saccharomyces cerevisiae. Mol Cell Biol 1995; 15: encoded endonuclease I-SceI distorts DNA through binding in
30 1968–1973. monomeric form to its homing site. J Biol Chem 2001; 276: 89
31 7. Miller G, Petek LM, Russel DW. Human gene targeting by adeno- 25 243–25 253. 90
D
32 associated virus vectors is enhanced by DNA double-strand 30. Chen C, Okayama H. High-efficiency transformation of 91
breaks. Mol Cell Biol 2003; 23: 3550–3557. mammalian cells by plasmid DNA. Mol Cell Biol 1987; 7:
33 92
TE

8. Porteus MH, Baltimore D. Chimeric nucleases stimulate gene 2745–2752.

34 targeting in human cells. Science 2003; 300: 763. 31. Mansour SL, Thomas KR, Deng CX, et al. Introduction of a 93
35 9. Porteus MH, Cathomen T, Weitzman MD, et al. Efficient gene lacZ reporter gene into the mouse int-2 locus by homologous 94
targeting mediated by adeno-associated virus and DNA double- recombination. Proc Natl Acad Sci U S A 1990; 87: 7688–7692.
36 strand breaks. Mol Cell Biol 2003; 23: 3558–3565. 32. Keeney S, Giroux CN, Kleckner N. Meiosis-specific DNA double- 95
EC

37 10. Bibikova M, Beumer K, Trautman JK, et al. Enhancing gene strand breaks are catalyzed by Spo11, a member of a widely 96
38 targeting with designed zinc finger nucleases. Science 2003; conserved protein family. Cell 1997; 88: 375–384. 97
300: 764. 33. Anglana M, Bacchetti S. Construction of a recombinant
39 11. Capecchi MR. High efficiency transformation by direct adenovirus for efficient delivery of the I-Sce I yeast endonuclease 98
40 microinjection of DNA into cultured mammalian cells. Cell 1980; to human cells and its application in the in vivo cleavage of 99
22: 479–488. chromosomes to expose new potential telomeres. Nucleic Acids
R

41 100
12. Smithies O, Gregg RG, Boggs SS, et al. Insertion of DNA Res 1999; 27: 4276–4281.
42 sequences into human chromosomal β-globin locus by 34. Vargas JJ, Gusella GL, Najfeld V, et al. Novel integrase-defective 101
R

43 homologous recombination. Nature 1985; 317: 230–234. lentiviral episomal vectors for gene transfer. Hum Gene Ther 102
44 13. Goncz KK, Prokopishyn NL, Chow BL, et al. Application of SFHR 2004; 15: 361–372. 103
to gene therapy of monogenic disorders. Gene Ther 2002; 9: 35. Arimondo PB, Boutorine A, Baldeyrou B, et al. Design and
O

45 691–694. optimization of camptothecin conjugates of triple helix- 104

46 14. Gong WJ, Golic KG. Ends-out, or replacement, gene targeting in forming oligonucleotides for sequence-specific DNA cleavage 105
47 Drosophila. Proc Natl Acad Sci U S A 2003; 100: 2556–2561. by topoisomerase I. J Biol Chem 2002; 277: 3132–3140. 106
C

15. Rong YS, Golic KG. Gene targeting by homologous 36. Meunier-Durmort C, Grimal H, Sachs LM, et al. Adenovirus
48 recombination in Drosophila. Science 2000; 288: 2013–2018. enhancement of polyethyleneimine-mediated transfer of 107
49 16. Kunzelmann K, Legendre JY, Knoell DL, et al. Gene targeting of regulated genes in differentiated cells. Gene Ther 1997; 4: 108
N

50 CFTR DNA in CF epithelial cells. Gene Ther 1996; 3: 859–867. 808–814. 109
17. Maurisse R, Feugeas JP, Biet E, et al. A new method (GOREC) 37. Morris KV, Chan SWL, Jacobsen SE, et al. Small interfering RNA-
51 110
U

for directed mutagenesis and gene repair by homologous induced transcriptional gene silencing in human cells. Science
52 recombination. Gene Ther 2002; 9: 703–707. 2004; 305: 1289–1292. 111
53 112
54 113
55 114
56 115
57 116
58 117
59 118

Copyright  2004 John Wiley & Sons, Ltd. J Gene Med 2004; 6: 000–000.