In the summer of 1984, Henry Erlich, Glenn Horn, Randy Saiki, and myself were working on developing a DNA test for sickle cell anemia (SCA) at Cetus Corp. SCA is an autosomal dominant disease caused by a single base mutation in the β-globin subunit of the hemoglobin gene.
Randy had developed the detection methodology for the sickle cell mutation using radiolabeled oligonucleotide probes that hybridized specifically to β-globin. These probes were cut using restriction enzymes, in a process called oligomer restriction, into two small specifically sized DNA fragments indicating the presence of normal or sickle cell alleles in the sample. These small, radioactive restriction fragments were then separated on thin-layer chromatography (TLC) plates and detected by autoradiography.
While oligomer restriction was conceptually elegant, from a practical perspective we simply could not get our probes hot enough to see a statistically significant signal above background. So, the problem we were dealing with was a classic one: how can we amplify the signal above the noise?
At that time, Kary Mullis, who conceived the idea of PCR in 1983, was running the lab that synthesized our oligo probes and was aware of the signal-to-noise problem we were having. Glenn, Randy, and I brainstormed several ideas of amplifying the signal, including using a Qβ Replicase, but none of the ideas stuck.
I remember the day in the summer of 1984 when Kary came to our lab and suggested that we try PCR to amplify the signal of the desired genes above the background of the genome.
It wasn’t until a few months later, in November 1984, that Thomas White, our vp of research, suggested to Henry that I work on reducing the PCR invention to practice as the amplification “front end” for the sickle cell anemia test.
Back then there were no thermal stable polymerases available so I started on the task of developing and characterizing the PCR methodology utilizing a fragment of E. coli DNA polymerase I (called Klenow fragment) to mediate the polymerase chain reaction. As there also were no thermal cyclers at the time, the enzyme had to be added manually at each cycle, moving the reaction tubes by hand between temperature-controlled hot plates set at the denaturation, annealing, and extension temperatures.
We had also revised the oligomer restriction assay to utilize 30% acrylamide gels instead of TLC plates to separate the now-amplified wild-type and mutant DNA fragments. The gels were dried and detected using autoradiography.
Imagine my surprise when I went to the x-ray film processor and pulled out the film of the first experiment I did in conjunction with Kary’s technician, Fred Faloona. There were two strong bands indicating the presence of the normal and mutant alleles, and virtually no noise! It had worked!
I ran to the lab and showed the results to Henry. We were very encouraged and embarked on making the PCR technology into a robust, reproducible, and repeatable methodology for use in the SCA test.
One of the most important things we did during this initial phase of PCR development was to form a team composed of Norm Arnheim, Kary, Henry, Glenn, Fred, Randy, and myself. We held meetings every Friday afternoon to review the week’s results, derive insight from the data, and decide what experiments to do next.