We describe a simple, quantitative assay for any amplifiable DNA sequence that uses a video camera to monitor multiple polymerase chain reactions (PCRs) simultaneously over the course of thermocycling. The video camera detects the accumulation of double-stranded DNA (dsDNA) in each PCR using the increase in the fluorescence of ethidium bromide (EtBr) that results from its binding duplex DNA. The kinetics of fluorescence accumulation during thermocycling are directly related to the starting number of DNA copies. The fewer cycles necessary to produce a detectable fluorescence, the greater the number of target sequences. Results obtained with this approach indicate that a kinetic approach to PCR analysis can quantitate DNA sensitively, selectively and over a large dynamic range. This approach also provides a means of determining the effect of different reaction conditions on the efficacy of the amplification and so can provide insight into fundamental PCR processes.

We have used the polymerase chain reaction (PCR) to amplify up to 22 kb of the 3-globin gene cluster from human genomic DNA and up to 42 kb from phage A DNA. We have also amplified 91 human genomic inserts of 9-23 kb directly from recombinant A plaques. To do this, we increased pH, added glycerol and dimethyl sulfoxide, decreased denaturation times, increased extension times, and used a secondary thermostable DNA polymerase that possesses a 3'-to-5'-exonuclease, or "proofreading," activity. Our "long PCR" protocols maintain the specificity required for targets in genomic DNA by using lower levels of polymerase and temperature and salt conditions for specific primer annealing. The ability to amplify DNA sequences of 10-40 kb will bring the speed and simplicity of PCR to genomic mapping and sequencing and facilitate studies in molecular genetics.PCR (1, 2) and molecular cloning are powerful tools for the amplification of genetic sequences; yet PCR can be quicker, simpler, and less costly to perform. As a result, PCR has been widely applied in molecular biology, molecular evolution, genetics, and forensic biology (3). PCR has also had broad impact on genome mapping and sequencing projects (4, 5). PCR would have an even greater role, however, if sequences longer than 10 kb-sequences currently cloned with phage A or cosmid vectors-could be amplified reliably.Although recent reports (6-11) described amplifications of 5-15 kb, reported yields were low. Our goals were to amplify targets of at least 20 kb with high yields, even from singlecopy genes within complex genomes, and to better understand the most critical parameters for longer amplifications.We surveyed various thermostable DNA polymerases, reaction buffers and additives, and thermal cycling profiles, guided by the following likely requirements for a reliable "long PCR": (i) complete denaturation of target sequences, as longer targets may become increasingly difficult to denature; (ii) extension times sufficient for complete strand synthesis in each PCR cycle; (iii) protection of template DNA against damage [e.g., depurination (12)] during thermal cycling; and (iv) retention of specificity necessary for singlecopy gene amplifications from genomic DNA.One of us (W.M.B.) has also hypothesized that misincorporated nucleotides reduce the efficiency of amplifying long targets. A mismatched 3'-terminal base may cause prematurely terminated strand synthesis (13). Even the low levels of nucleotide misincorporation estimated for Taq DNA polymerase [<1 in 10-50,000 bases (14)] will affect sequences longer than 10 kb. A small amount of thermostable 3'-to-5'-exonuclease activity removes such mismatched nucleotides and permits the predominant polymerase activity to complete strand synthesis (15). The use of such proofreading activity, combined with conditions identified in our survey, resulted in the highest yields of the longest products. The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked...