Biochemical and Biophysical Research Communications 271, 534 –536 (2000)
doi:10.1006/bbrc.2000.2668, available online at http://www.idealibrary.com on
Vectors with Hidden Cloning Sites
Ervin Welker* ,† ,1 and András Váradi*
*Institute of Enzymology, Hungarian Academy of Sciences, P.O. Box 7, H-1518 Budapest, Hungary; and †National Institute
of Haematology and Immunology, Research Group of the Hungarian Academy of Sciences, H-1113 Budapest, Hungary
Received April 10, 2000
A general strategy is described for using the cleavage site of restriction enzymes in vectors for cloning
regardless of how many sites the given enzymes have
in the vector. The application of this method allows
one to open any vector at its cloning site with protruding ends which can be compatible with almost every
commercially available Class II restriction enzyme. By
employing this method, the laborious construction of
new vectors can be simplified considerably. This general strategy is based on the known ability of Class IIS
restriction enzymes to cut any sequence located outside of their recognition site; the introduction of a
linker containing recognition site(s) for Class IIS restriction enzyme(s), not present originally in the vector, gives rise to the possibility of opening the vector
so as to produce overhangs of arbitrary sequence. In
particular, when a symmetrical short sequence representing the protruding end of any Class II enzyme is
situated at the cutting position of the Class IIS enzyme, cleavage with the Class IIS enzyme exposes the
hitherto hidden, “unique” cloning site. This technique
is demonstrated by cloning the cDNA of the multidrug
resistance protein to an expression vector. © 2000
Academic Press
A typical cloning vector harbors multiple cloning
sites (MCS) with 10 –15 restriction– enzyme recognition sequences. These MCS’s ensure the convenient
ligation of inserts generated with many of the most
frequently used restriction enzymes. However, these
MCS’s can be troublesome when large vectors with
complex structure are to be used, e.g., eukaryotic expression vectors with a selection marker and/or a reporter gene; in such vectors, the freedom of choice is
rather limited since they contain either a single cloning
site or MCS with few restriction sites. The creation of a
1
To whom correspondence should be addressed at present address: P.O. Box 294, Baker Laboratory of Chemistry and Chemical
Biology, Cornell University, Ithaca, NY 14853-1301. Fax: (607) 2554137. E-mail: ew31@cornell.edu.
0006-291X/00 $35.00
Copyright © 2000 by Academic Press
All rights of reproduction in any form reserved.
suitable cloning site in such a vector can be laborious
since it often requires the elimination of the existing
restriction sites by mutagenesis. This article describes
a simple, general strategy to solve this problem using
Class IIS restriction enzymes.
In the method proposed here, a linker containing a
recognition site for Class IIS restriction enzyme, not
present originally in the vector, is inserted into the
cloning site of the vector. Recognition sites of Class IIS
enzymes are 1–20 bases away from the cleavage sites
and the base sequence of the latter is not restricted in
any form (see reference 1 for a review). In this way, a
unique Class IIS site is generated at the cloning site in
the vector. In the linker at the cutting position of the
Class IIS enzyme, a symmetrical short sequence is
designed that represents the protruding end of any
Class II enzyme (e.g., EcoRI, HindIII, BamHI, and so
on). The only requirement is that, after digestion, the
Class IIS and the Class II restriction enzymes have to
leave compatible protruding DNA ends regarding of
the type of the overhang (5⬘ or 3⬘) and it’s length. Thus,
cleavage with the Class IIS enzyme exposes the hitherto hidden, “unique” cloning site. By varying the sequences of the linker at the cutting position of the
Class IIS restriction enzyme, a series of vectors can be
generated, each with a different linker for the hidden
cloning site of different Class II enzymes. This series of
vectors allows one to use almost any commercially
available Class II restriction enzymes without destroying the existing restriction sites.
MATERIALS AND METHODS
All restriction enzymes used in this study were found from Biolabs.
Plasmid pvl941mdr1 was constructed as described previously (2).
Human MDR1 cDNA has been removed from this vector by BamHI
digestion and the cDNA insert was purified from agarose gel. Expression vector RSV (Invitrogen) was digested at the multiple cloning site of the vector. An oligonucleotide linker containing two
BsmBI sites and a BamHI site (see Fig. 1) was ligated into the
multiple cloning site of RSV yielding RSV-B. Vector RSV-B was then
cleaved with BsmBI (which opens up the BamHI overhangs situated
in the linker) and the purified MDR1 cDNA with BamHI ends was
534
Vol. 271, No. 2, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
the baculovirus transfer vector pvl941 into the pRSV-B
by this procedure.
DISCUSSION
FIG. 1. Sequence of the linker for creating a hidden BamHI site.
The BsmBI recognition sites are indicated with capital letters; the
BamHI overhangs produced by BsmBI digestion are indicated with
bold lowercase letters. A vertical bar (兩) shows cleavage with BmsBI
(or with BamHI). The presence of two BsmBI sites ensures that the
cloned insert can be removed later with BsmBI produced BamHI
overhangs at both ends.
ligated into RSV-B. The direction of the cDNA insert was checked by
restriction mapping.
RESULTS
We have utilized this “hidden cloning site” strategy
to facilitate the shuttle between the mammalian expression vector pRSV and the baculovirus expression
vector pvl941, which contains the cDNA of the human
multidrug resistance protein as well as that of its mutant variants at the single BamHI cloning site (2, 3).
The pRSV vector does not have a unique BamHI site in
its MCS. On the basis of the above considerations, a
linker has been designed in which two BsmBI (a Class
IIS enzyme) recognition sites surround a BamHI site
(Fig. 1). The latter is situated at a distance from the
BsmBI recognition sites which ensures cleavage
by BsmBI, leaving BamHI-compatible 5⬘ four base
(GATC) overhangs. After BsmBI cleavage, this linkervector construct allows the insertion of fragments removed by BamHI from any other vector and later removal of this fragment with BamHI overhangs via
BsmBI digestion. The linker was inserted into the multiple cloning site of expression vector pRSV. This vector
possesses three BamHI sites, but has no BsmBI site.
The resulting RSV-B vector was then digested with
BsmBI (Fig. 2), which uncovers the unique, hidden
BamHI site. The cDNA of the human multidrug resistance protein (MDR1)—previously removed from the
pvl941 vector by BamHI cleavage—was than cloned
into this hidden, unique BamHI cloning site of RSV-B,
yielding plasmid RSV-B-mdr1. Digestion with either
BsmBI or with BamHI removes the insert from this
recombinant vector (Fig. 2). Several MDR1 cDNA variants harboring mutations in the ABC-signature region
of the first ATP-binding cassette (3) were cloned from
Our approach offer a versatile way in inserting DNA
fragments into even very large vectors. A great number
of Class IIS enzymes with recognition sequences of 6 or
more base pairs are known today, which holds out the
possibility of applying the method proposed here to a
wide variety of vectors. Enzymes which produce fourbase overhangs can be particularly useful in addition
to BsmBI, most notably BbsI, BsaI, BsmAI, and
BspMI. However, enzymes that do not cleave in a predictable manner or those leaving protruding ends that
difficult to ligate are not suitable for creating such
hidden cloning sites.
Several other methods have been published (for review see 1) that rely on the ability of the Class IIS
restriction enzymes to cut any sequence found at a
given distance from their recognition site. In all of
these earlier methods, a unique not-palindromic protruding end was created by the Class IIS enzyme. Most
relevant for the present work is the early elegant
method of Podhajska and Szybalski (4, 5) which uses
the Fok I Class IIS enzyme with a tailored oligonucleotide adapter as a “universal restriction enzyme” to
cleave single-stranded DNA at any predetermined position (in principle). By combining this approach with
the Polk I polymerase, a double-stranded DNA fragment was generated from the single-stranded template
(6). Thus, this method can be used to produce a doublestranded DNA fragment terminated at arbitrary end
points. During the synthesis of the second strand, the
overhangs generated by the universal restriction en-
FIG. 2. Digestion of vector constructs harboring hidden restriction sites. RSV-B plasmid digested with BsmBI (lane A) or
with BamHI (lane B); RSV-B-mdr1 plasmid digested with BsmBI
(lane C) or with BamHI (lane D). On the left side, the size (in kb)
of the fragments of the DNA marker is indicated. The small
fragments generated by BamHI digestion are not seen on the gel
in lanes A and B.
535
Vol. 271, No. 2, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
zyme were filled up by the Polk I polymerase, rendering the double-stranded fragment suitable for subsequent cloning. However, the authors did not consider
the possibility of creating palindromic overhangs by
this method, presumably because it would decrease the
generality of their method and would likely have fewer
useful applications since the single-stranded DNA
would only be cut at pre-existing palindromic sequences. Moreover, their method is technically demanding and applicable only to single-stranded DNA.
By contrast, the present approach is simple to perform,
involves double-stranded DNA vectors, and does not
require a preexisting palindromic sequence at the
cleavage position.
The strategy of “hidden cloning sites” described here
provides an economical alternative for constructing
new vectors without abolishing unwanted restriction
sites by laborious mutagenesis. To our knowledge, the
present paper gives the first description of using Class
IIS restriction enzyme on DNA vector for generating
hidden cloning sites.
ACKNOWLEDGMENT
We thank William J. Wedemeyer for helpful comments on the final
draft.
REFERENCES
536
1. Szybalsky, W., Kim, S. C., Hasan, N., and Podhajska, A. J. (1991)
Gene 100, 13–26.
2. Müller, M., Bakos, É., Welker, E., Váradi, A., Germann, U. A.,
Gottesman, M. M., Morse, B. S., Roninson, I. B., and Sarkadi, B.
(1996) J. Biol. Chem. 271, 1877–1883.
3. Bakos, É., Klein, I., Welker, E., Szabó, K., Müller, M., Sarkadi,
B., and Váradi, A. (1997) Biochem. J. 323, 777–783.
4. Podhajska, A. J., and Szybalsky, W. (1985) Gene 40, 175–182.
5. Szybalsky, W. (1985) Gene 40, 169 –173.
6. Kim, S. C., Podhajska, A. J., and Szybalsky, W. (1988) Science
240, 504 –506.