Short Communication

Cre-mediated Site-specific
Cassette Exchange in Erythroid Cell

LI Xing-Guo, YAN Hao-Heng, LIU De-Pei*, HAO De-Long, LIANG Chih-Chuan

( National Laboratory of Medical Molecular
Biology, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences
& Peking Union Medical College, Beijing
100005, China )

Abstract        Cre-mediated
cassette exchange has been developed to perform site-specific chromosomal
integration using Cre recombinase. Here, site-specific integration with
inverted Lox sites was used to investigate the erythroid cis-acting DNA element
in specific chromatin contexts in mouse erythroleukemia cells. Single
hygromycin-resistant clones were obtained from the selective semi-solid medium
containing hygromycin post-electroporation. PCR and Southern blotting analysis
showed single-copy integration of target vector in clones A, B and D.
Site-specific cassette exchange was performed in clone A with exchange vector
and Cre expression plasmid, followed by gancyclovir selection. Flow cytometry
was used for analysis of EGFP gene expression. A 732-bp fragment of human
β-globin gene cluster 5′ DNase I hypersensitive site 2(HS2) was exchanged and integrated
into clone A in an anti-genomic orientation. The low EGFP expression in clone
A-HS may be due to the orientation-dependent gene silencing caused by
integration of HS2 in a non-permissive orientation.

Key
words     chromosomal integration; site-specific recombination; Cre/Lox;
globin gene; gene silencing

Many techniques
available for stable integration of transgenes in mammalian cells often result
in integration at random chromosomal locations of multiple copies of transgenes
that express at levels that are difficult to be predicted or reproduced because
of position effects. While homologous recombination is the common solution for
embryonic stem cells, the efficiency of stable integration of transgenes is
largely masked in permanent cell line by massive illegitimate recombination
event[1]. An alternative way was to utilize the site-specific recombinases
adapted from phages or yeast (Cre or Flp, respectively)[2]. The target site of
the Cre recombinase (Lox site) is a 34-bp sequence that consists of two
inverted 13-bp Cre-binding sites linked by an 8-bp spacer within which the
recombination occurs[3] (Fig.1). Two recombination target sites will not
recombine with different central 8-bp-spacer region, whereas recombine
efficiently with same spacer region. A chromosomal cassette flanked by two
hetero-specific Lox sites can therefore be replaced by another cassette
(located on a plasmid) flanked by such mutually incompatible Lox sites through
a double reciprocal recombination[4]. Cre-mediated site-specific chromosomal
integration in mammalian cells has been previously shown feasible, although
with relatively low efficiency[5]. Recent reports have demonstrated that the
mutated hetero-specific Lox sites (LoxP 511 and LoxP, for example) are not
entirely incompatible, and that an unexpected excision reaction occurred
between such Lox sites due to recombination between them[6].

Fig.1       Structure
of a Lox site

The target site of the Cre recombinase is
a 34 bp DNA sequence that consists of two inverted 13 bp Cre-binding sites
linked by an 8-bp spacer within which recombination occurs.

In the present
study, a modified strategy based on inverted Lox sites[L1 (LoxP 511) and 1L (inverted
LoxP 511)] was employed to investigate the erythroid cis-acting DNA element, a
732-bp fragment of human β-globin gene cluster 5′ DNase I hypersensitive site 2(HS2),  in specific chromatin contexts of mouse erythroleukemia
cells (MELc). Intra-chromosomal recombination between two inverted Lox sites
would lead to inversion rather than excision, and would therefore maintain the
high efficiency of recombination[6] (Fig.2). The positive selection,
negative selection and orientation of cassette exchange were discussed.

Fig.2       Cre-mediated
cassette exchange

The classical RMCE event generally
goes as the following steps. Firstly, the target vector is randomly integrated
into the cellular genome. Both the target sequence and the exchange sequence
are flanked by two inverted Lox sites (LoxP 511 and inverted LoxP 511,
designated as L1 and 1L respectively). After co-transfection of the exchange
vector and Cre expression vector, the target sequence between inverted Lox
sites can be readily recombined to the exchange vector containing the same pair
of inverted Lox sites.

1    Materials and Methods

1.1   Plasmids

Plasmid
construction was performed according to the conventional techniques[7]. The
target vector p1L-HyTk-L1-β-EGFP-IRES-neo (designated as β-EGFP-neo thereafter)
was constructed in three steps. First, a PCR product of human β-globin promoter
(309 bp, HUMHBB positions 61 921 to 62 230) was inserted into the SmaI site of
plasmid pEGFP-1 (Clontech, USA)
to create pβEGFP-1. Second, a 4.4-kb SalI / HindIII fragment (with HindIII end
filled in with Klenow fragment) from pβEGFP-1 was ligated with a 3.3-kb XhoI /
PvuII fragment from pL1-HyTk-1L (a generous gift of Dr. Eric E. Bouhassira,
Bronx, USA) to create p1L-HyTk-L1-β-EGFP. Finally, a NruI / NotI fragment
containing IRES-neo sequence from pIRES-neo (Clontech,
USA) was inserted into the
Eco47III /NotI sites of p1L-HyTk-L1-β-EGFP to form β-EGFP-neo. Restriction
digestion and DNA sequencing confirmed the proper content and orientation of
plasmids.

pL1-HS2-1L was
constructed from pL1-HyTk-1L by replacing the HyTk sequence with HS2 (732 bp,
HUMHBB positions 8486 to 9218) which was a HindIII/BglII core fragment of 5′ HS2 of human β-globin gene locus
control region.

1.2   Cell culture and creation of
hygromycin-resistant clones

MELc were grown
in Dulbecco’s modified
Eagle’s medium
(DMEM) containing 10% fetal bovine serum. Electroporation was performed as
follows: 5×106 MELc in
the mid-to-late logarithmic phase of growth in 400 μL of PBS was mixed with 10 μg of Eco31I(MBI Fermentas,
USA)-linearized β-EGFP-neo, and electroporated at 280 V and 950 μF in a gene pulser (Bio-Rad, USA). The
selective semi-solid medium containing 2% methyl cellulose (Sigma,
USA) and 0.8 g/L
hygromycin B (Hyg) (Roche, Germany)
was feed to transfected MELc 48 h post-electroporation. After 14 d of positive
selection, single HygR clones were isolated, and single-copy integration events
were identified by PCR and Southern blotting after digestion with XbaI, which
cuts only once in β-EGFP-neo.

1.3   Cre-mediated site-specific cassette
exchange

Clone A was
co-electroporated with 25 μg pL1-HS2-1L and 15 μg Cre expression vector pBS185 (Life Technologies, USA). Negative
selection with 3.0 g/L gancyclovir (Gcv) (Sigma,
USA) was applied 24 h
post-transfection. Single gcvR cell clone was picked after about 14 d of
negative selection, and then tested for the exchange events by PCR.

1.4   Flow cytometry analysis

Flow cytometry
was performed on a Becton Dickinson instrument (FACS Calibur, Becton
Dickinson, USA).
EGFP was quantitated under standardized conditions using normal MELc as
standards. Three independent subclones from each single clone were assayed as
pools to minimize the extent of chromosomal loss and clonal variations.

2    Results

In target vector the β-globin gene
promoter is located upstream of the coding sequence of two reporter genes: EGFP
and neo were interposed by internal ribosome entry site (IRES) sequence. The
transcriptional direction of reporter genes is opposite to that of selectable
marker HyTk gene (Fig.3).

Fig.3       Target
vector pβ-EGFP-neo and exchange vector pL1-HS2-1L

The target vector pβ-EGFP-neo includes
the CMV-HyTk gene and can be selected positively by Hyg and negatively by Gcv.
The arrow indicates the transcriptional orientation of reporter genes driven by
a 309-bp human β-globin gene promoter. The thick line below stands for 309-bp
human β-globin gene promoter as the probe used in Southern blotting analysis of
XbaI-digested genomic DNA.

PCR and Southern
blotting analysis indicated that three hygR clones (clone A, B and D) contained
single copy of target vector pβ-EGFP-neo (Fig.4, 5). For each of these three
clones, flow cytometry analysis showed no significant differences among
subclones both in the percentage and in the mean fluorescence intensity (MFI)
of EGFP-expressing cells (data not shown).

Fig.4       PCR
analysis of genomic DNA of four hygR single clones with upstream primer of
human β-globin gene promoter and downstream primer of EGFP gene (PCR product:
779 bp) and specific Neo gene primers (PCR product: 432 bp)

M, 100 bp DNA ladder; N, negative
control; P, plasmid pβ-EGFP-neo as the positive control; H, human genomic DNA;
Mc, MELc; A, clone A; B, clone B; C, clone C; D, clone D.

Fig.5       Southern
blotting analysis of XbaI digested genomic DNA of three hygR single clones with
the 309-bp of human β-globin gene promoter as the probe

P, plasmid pβ-EGFP-neo as the positive
control; N, negative control; Mc, MELc; A, clone A; B, clone B; D, clone D.

PCR analysis
also indicated that the HS2 fragment was exchanged and integrated into gcvR
clone A-HS in an anti-genomic orientation (Fig.6). There were no significant
differences among the subclones of clone A-HS2 (data not shown) while the mean
percentage of EGFP-expressing cells was much lower (2.42%) than that in clone A
(35.94%) (Table 1). 

Fig.6       PCR
products amplified by primers of β-globin promoter and HS2 core primers

M, 100 bp DNA ladder; N1, negative control (β-globin promoter primers);
P1, plasmid pβ-EGFP-neo (PCR product: 309 bp of human β-globin gene promoter
sequence); A1, clone A; A-HS1, clone A-HS; N2, negative control (HS2 core
primers); P2, plasmid pL1-HS2-1L (PCR product: 247 bp of 5′ HS2 core
sequence); A2, clone A; A-HS2, clone A-HS; N3, negative control (downstream
primer of HS2 & downstream primer of β-globin promoter); A3, clone A;
A-HS3, clone A-HS (PCR product: 748 bp); N4, negative control (upstream primer
of HS2 & downstream primer of β-globin promoter); A4, clone A; A-HS4, clone
A-HS.

 

Table 1   Comparison
of EGFP expression in MELc, clone A and clone A-HS2 (un-induced and induced by
2% DMSO or 75 μmol/L hemin)

3    Discussion

The Cre-mediated
cassette exchange strategy uses site-specific recombinase to integrate
single-copy transgene without selective markers into previously tagged sites in
mammalian cells[8]. In this study we described a simple method for
site-specific integration in erythroid cells based on the use of inverted Lox
sites to investigate the erythroid cis-acting DNA element in specific chromatin
contexts. PCR and Southern blotting analysis showed single-copy integration of
target vector pβ-EGFP-neo in HygR clones A, B and D. A 732-bp fragment of 5′HS2 was exchanged and integrated
into clone A in an anti-genomic orientation. The low EGFP expression in clone
A-HS may be due to the orientation-dependent gene silencing caused by
integration of HS2 in a non-permissive orientation.

The target
vector pβ-EGFP-neo includes the CMV-HyTk gene and can be selected positively by
Hyg and negatively by Gcv. Gcv, a non-toxic pro-drug, is converted to a
phosphorylated active analog and is incorporated into the DNA of replicating
eukaryotic cells, causing death of dividing cells, upon expression of a viral
suicide gene encoding thymidine kinase (Tk). It has been widely used in
molecular biology for selection against random recombination events when
homologous recombination of a gene of interest is required. There have been
several reports describing the strong sensitivity of HyTk-expressing cells to
Gcv (>99% cell death in 48 h in 10 μmol/L Gcv)[9]. Our results indicated that hygR clones A, B, and D
showed differential sensitivity to Gcv, with lethal concentration of Gcv in 48
h at 10 μmol/L, 20 μmol/L, and 30 μmol/L respectively (data not shown). Due to
the limited number of Gcv clones in this study, we still cannot rule out the
possible relationship between Gcv sensitivity and exchange efficiency at
different chromosomal sites of integration[10].

In contrast to
recessive markers for negative selection, many dominant markers for positive
selection were isolated from bacteria as autonomously replicating plasmid DNA
molecules carrying genes for resistance to antibiotics. G418 and Hyg are two
widely used selective reagents produced by streptomycetes[11]. As this kind of
antibiotics exhibit broad specificity in prokaryotes and in eukaryotes as
diverse as yeast and mammals, it makes positive selection of stably transfected
suspension cell clones from semi-solid medium containing appropriate
antibiotics an especially simple and efficient method (for example, selective
semi-solid medium containing 2% methyl cellulose in the selection of hygR
clones in this study). As for negative selection, however, semi-solid medium
selection didn’t show any obvious advantages over other routine clone expansion
methods. In general, stably transfected cells selected for either G418 or Hyg
resistance will maintain the plasmid sequences in the absence of drug selection
for as many as 50－75 cell
doublings[12]. Prior to Cre-mediated cassette exchange, however, it is
necessary to select HyTk-expressing cell clones in selective medium for 2－3 weeks to ensure that all cells
express the HyTk gene[13].

5′-HS2 is a
well-characterized erythroid enhancer. A 732-bp HindIII/BglII core fragment
showed enhancer activity that was erythroid-specific and developmental-stage
nonspecific[14]. DNA fragments containing HS2 will confer position-independent,
high level expression on globin genes in transgenic mice. When tested in cell
transfection assays, HS2 greatly enhances expression of β-like globin genes in
an erythroid-specific manner, both in transient expression from un-integrated
constructs and after stable integration into a chromosome[15]. Transgenic
studies suggested that HS2 linked to the β-globin gene in the genomic orientation can activate high-level
human β-globin gene expression in the erythroid tissues of transgenic mice,
whereas HS2 in the anti-genomic orientation appears to predispose the construct
to significant rearrangements[16]. A recent report demonstrated
orientation-specific silencing of 5′-HS 234 fragment of human β-globin locus control region occurred in
the two distinct orientations at the same integrated locus in MELc[9].

Due to the
nature of inverted Lox sites, the cassette exchange will occur in an unexpected
manner: in one orientation in half of the clones and the reverse one in the
other half. In our study, PCR analysis showed the anti-genomic integration of
HS2 fragment in clone A-HS2. Similarly, clone A-HS2 was also not sensitive to
DMSO or hemin induction, in contrast to the regular induction patterns of these
reagents (Table 1). The reporter gene suppression in A-HS2 may be due to the
orientation-specific gene silencing caused by non-permissive integration.
Further study on the reversion from non-permissive orientation to permissive
orientation of HS2 in A-HS may provide clue to understanding the mechanisms
underlying orientation-dependent gene silencing and effect of integration site
on gene expression and regulation. The cell clones A, B and D contain inverted
Lox sites in the artificial genomic loci and may provide the source of
recipient cell strains for further steps toward systematic screening and
functional elucidation of cis-acting elements in the specific chromatin
contexts[17].

References

1     Capecchi
M. Gene targeting: How efficient can you get? Nature. 1990, 348: 109

2     Bode
J, Schlake T, Iber M, Schubeler D, Seibler J, Snezhkov E, Nikolaev L. The
transgeneticist’s toolbox: Novel methods for the targeted modification of
eukaryotic genomes. Biol Chem, 2000, 381: 801－813

3     Hoess
RH, Abremski K. Interaction of the bacteriophage P1 recombinase Cre with the
recombining site loxP. Proc Natl Acad Sci USA, 1984, 81: 1026－1029

4     Hoess
RH, Wierzbicki A, Abremski K. The role of the loxP spacer region in P1 site-specific
recombination. Nucleic Acids Res, 1986, 14: 2287－2300

5     Kilby
NJ, Snaith MR, Murray JA. Site-specific recombinases: Tools for genome
engineering. Trends Genet, 1993, 9: 413－421

6     Feng
YQ, Seibler J, Alami R, Eisen A, Westerman KA, Leboulch P, Fiering S et al.
Site-specific chromosomal integration in mammalian cells: Highly efficient CRE
recombinase-mediated cassette exchange. J Mol Biol, 1999, 292: 779－785

7     Sambrook
J, Russsell DW. Molecular Cloning: A Laboratory Manual, 3rd ed, NY: Cold Spring
Harbor Laboratory Press, 2001

8     Baer
A, Bode J. Coping with kinetic and thermodynamic barriers: RMCE, an efficient
strategy for the targeted integration of transgenes. Curr Opin Biotechnol,
2001, 12: 473－480

9     Feng
YQ, Lorincz MC, Fiering S, Greally JM, Bouhassira EE. Position effects are
influenced by the orientation of a transgene with respect to flanking
chromatin. Mol Cell Biol, 2001, 21: 298－309

10    Lauth
M, Moerl K, Barski JJ, Meyer M. Characterization of Cre-mediated cassette
exchange after plasmid microinjection in fertilized mouse oocytes. Genesis,
2000, 27: 153－158

11    Santerre
RF, Allen NE, Hobbs JN Jr, Rao RN,
Schmidt RJ. Expression of prokaryotic genes for hygromycin B and G418
resistance as dominant-selection markers in mouse L cells. Gene, 1984, 30: 147－156

12    Santerre
RF, Walls JD, Grinnell BW. Use of vectors to confer resistance to antibiotics
G418 and hygromycin in stably transfected cell lines. Methods in Mol Biol,
1991, 7: 245－256

13    Schubeler
D, Lorincz MC, Groudine M. Targeting silence: The use of site-specific
recombination to introduce in vitro methylated DNA into the genome. Sci STKE,
2001, 2001(83): PL1

14    Tuan
DY, Solomon WB, London IM, Lee DP.
An erythroid-specific, developmental-stage-independent enhancer far upstream of
the human “β-like globin” genes. Proc Natl Acad Sci USA,
1989, 86: 2554－2558

15    Hardison
R, Slightom JL, Gumucio DL, Goodman M, Stojanovic N, Miller W. Locus control
regions of mammalian β-globin gene clusters: Combining phylogenetic analyses
and experimental results to gain functional insights. Gene, 1997, 205: 73－94

16    Curtin
PT, Liu DP, Liu W, Chang JC, Kan YW. Human β-globin gene expression in transgenic mice is enhanced by a distant
DNase I hypersensitive site. Proc Natl Acad Sci USA, 1989, 86: 7082－7086

17    Zhang
LH, Liu DP, Liang CC. Finding regulatory sequences. Int J Biochem Cell Biol,
2003, 35: 95－103

_________________________________________________

Received: April
24, 2003 Accepted: June
30, 2003

This work was supported by the grants from
the National Natural Science Foundation of China (No. 30393110) and the Science
Foundation for Chinese Outstanding Youth (No. 3952006)

*Corresponding author: Tel,
86-10-85116436-8514; Fax, 86-10-65133086; e-mail, [email protected]

Updated at: 2003-10-09

红系细胞中Cre重组酶介导的位点特异性片段交换

李兴国     严皓珩     刘德培*    郝德龙     梁植权

( 中国医学科学院基础医学研究所， 中国协和医科大学基础医学院医学分子生物学国家重点实验室，
北京 100005 )

摘要       re介导的片段交换技术利用重组酶Cre的位点特异性重组特性， 在基因组的特定位点进行靶片段与目的片段的交换。 运用互为反向的Lox位点， 在鼠红白血病MEL细胞中进行靶载体的整合和交换载体的交换， 探讨在特定的染色质环境下红系特异性顺式作用元件的功能。 电穿孔转染MEL细胞后从含有潮霉素（hygromycin）的选择性半固体培养基中挑取MEL细胞单克隆， 通过PCR和Southern杂交鉴定整合完整性和拷贝数， 获得三种整合有靶载体p1L-HyTk-L1-β-EGFP-neo的细胞株A， B和D。
交换载体pL1-HS2-1L（含有732-bp的人β-珠蛋白基因簇5′ DNase I 高敏位点2核心片段）和Cre表达载体pBS185共转染细胞株A， 9-(1,3-二羟-2丙氧甲基)鸟嘌呤(gancyclovir)负筛选后挑取单细胞克隆A-HS。 PCR检测显示HS2片段以反方向进行了交换。 流式细胞仪分析显示平均的荧光细胞百分比（2.42 %）低于未交换的细胞株A （35.94 %）。 A-HS中EGFP的低表达可能是处于非容许方向的HS2片段出现方向依赖性基因沉默所致。

关键词   色体整合； 位点特异性重组； Cre/Lox； 珠蛋白基因； 基因沉默