NOTES & TIPS 137 Two linker-specific primers, L1 (21 mer) and L2 (20 An Alternative Linker-Mediated Polymerase mer), were designed based on the 5 protruding se- Chain Reaction Method Using a quence as shown in Fig. 1B. Two gene-specific prim- Dideoxynucleotide to Reduce ers, CK1 (24 mer) and CK2 (26 mer including 6 addi- Amplification Background tional nucleotides to provide a BamHI site to facilitate the postamplification cloning), were de- signed based on the 5 sequence of the zebrafish cyto- Ji Liao, Chiew Hua Chan, and Zhiyuan Gong 1 keratin cDNA clone ZF-A39 (7). CK1 was derived School of Biological Sciences, National University of from the sequence immediately following the first Singapore, 10 Kent Ridge Crescent, 119260 Singapore ATG codon, and CK2 was derived from the sequence before the ATG codon (Fig. 1). It is desirable not to Received January 6, 1997 include the ATG codon in the second primer (CK2) since in most cases the analysis of promoter requires A traditional approach to isolate a gene promoter no coding region and thus the final PCR product can is to screen a genomic DNA library by hybridization. be directly cloned into a reporter gene vector. It is The screening and postscreening characterization also advantageous to leave sufficient 5 sequence up- procedures are generally tedious and labor inten- stream of the PCR primers so that the identity of sive. With the advent of polymerase chain reaction the amplified promoter region can be confirmed by technique, several PCR-based methods can be sequence comparison. Two consecutive rounds of adapted to clone a promoter, such as inverse PCR PCR were then performed using two respective sets (1, 2) and linker-mediated PCR (3). However, these of primers, L1 and CK1, and L2 and CK2, to ensure methods are not always effective and usually gener- the high specificity of the PCR product. Because of ate heavy background or multiple PCR products due the presence of ddC in oligo 2, the extension of the to nonspecific amplification. Several improved lower strand, which could happen from intrastrand linker-mediated PCR methods have also been re- annealing of the two linker DNA ends or from hetero- ported by introduction of additional features to the strand annealing of the linker DNA, was blocked. linker, for example, by using a partially mismatched Therefore, during PCR, the two linker-specific prim- linker where PCR primers are designed against the ers, L1 and L2, were annealed only to the strand mismatched region to reduce the nonspecific ampli- extended from gene-specific primers CK1 or CK2 and fication (bubble PCR, 4), or by using a 5 extended thus reduce background amplification. linker whose 3  recessive end nucleotide is either Polymerase chain reaction. PCR was performed overhung (5) or chemically modified by an amine with Advantage Tth polymerase mix (Clontech). The group (6) to prevent the chain extension from non- first round of PCR was performed using primers L1 specific annealing. Here we describe an alternative and CK1. Each reaction (50 ml) contains 5 ml of 101 method by using a dideoxynucleotide at the 3  reces- Tth PCR reaction buffer (11Å 15 mM KOAc, 40 mM sive end of the linker to reduce nonspecific amplifi- Tris, pH 9.3), 2.2 ml of 25 mM Mg(OAc) 2 ,5 ml of 2 mM cation. We demonstrate here the success of the dNTP, 1 ml of L1 (0.2 mg/ ml), 1 ml of CK1 (0.2 mg/ml), method in isolation of a 4.3-kb promoter region from 33.8 ml of H 2 O, and 1 ml (50 ng) of linker-ligated a zebrafish cytokeratin gene. genomic DNA and 1 ml of 501 Tth polymerase mix Strategy of promoter isolation. The strategy of am- (Clontech). The cycling conditions were as follows: plification, including the designing of linker DNA 94°C/1 min, 35 cycles of 94°C/30 s, and 68°C/6 min, and PCR primers, is diagrammed in Fig. 1. The geno- and finally 68°C/8 min. After the primary round of mic DNA was first digested by selected restriction PCR was completed, the products were diluted by enzymes and modified by T4 DNA polymerase to gen- 100-fold. One microliter of diluted PCR product was erate blunt ends (if necessary). Since the distribution used as template for the second round of PCR (nested of restriction sites is more or less random, it is advis- PCR) with primers L2 and CK2, as described for the able to use more than one restriction enzyme in order primary PCR but with the following modification: to isolate a promoter with a reasonable size. The 94°C/1 min, 25 cycles of 94°C/30 s and 68°C/6 min, digested genomic DNA was then ligated with a linker and finally 68°C/8 min. Both the primary and second- DNA which was consisted of a long (oligo 1, 51 mer) ary PCR products were analyzed on a 1% agarose gel and a short oligonucleotide (oligo 2, 10 mer) whose 3  (Fig. 2). end was incorporated with a dideoxycytidine (ddC). 2 Results and discussion. As shown in Fig. 2A, gen- erally there is no obvious amplification background 1 To whom correspondence should be addressed. Fax: 65-7792486. after 35 cycles of the primary PCR from all three E-mail: sbsgzy@nus.sg. 2 Abbreviations used: ddC, dideoxycytidine; CK, cytokeratin. digested genomic DNAs except for a few very faint ANALYTICAL BIOCHEMISTRY 253, 137–139 (1997) ARTICLE NO. AB972201 0003-2697/97 $25.00 Copyright  1997 by Academic Press All rights of reproduction in any form reserved.