Most people know that the instructions for producing a living organism are encoded in the sequence of four bases (A,T, G and C) in a long molecule called deoxyribonucleic acid (DNA). Every human cell has 46 chromosomes composed of very long DNA molecules stabilized by their interactions with specialized proteins called histones. The DNA content of a cell totals 3 billion base pairs, which would be over a meter in length if lined up end to end. Determining the base sequence of the human genome required an investment of over $3 billion, and took hundreds of scientists nearly ten years to complete. Today there are instruments that can do the same analysis in a month at a cost of $100,000. This is still too costly for ordinary applications, so the race is on to find much less expensive approaches.
In 1996, I was a member of a team who published a new idea that promised cheap sequencing. We knew that a bacterial toxin called alpha hemolysin spontaneously inserts into a lipid bilayer membrane to produce an ion-conducting channel, now called a nanopore. Alpha hemolysin happens to have a pore precisely the diameter of a DNA molecule in cross section, and we discovered that if a small voltage was applied across the membrane, molecules of DNA in solution were captured by the electrical field in the pore and pulled through at half a million bases per second. While passing through the pore, each DNA molecule partially blocked ionic current for a few milliseconds, which meant that we could detect single molecules of DNA by monitoring electrical signals from the pore. The technique is not limited to DNA, by the way. Other linear polymers can also be detected as single molecules, including RNA and certain proteins.
In the paper, we proposed that the sequence of bases in DNA could be determined if each base produced a distinctive signal as it passed through the pore. This was followed in 1999 by a second paper in which we demonstrated that a nanopore signal had two distinct levels, on for Cs and one for As when a 100mer of RNA composed of 70 cytosine bases and 30 adenines passed through. Those two papers revealed the power of single molecule analysis by a nanopore. Now, fifteen years later, a search in Google Scholar using the terms 'nanopore sequencing' reveals 5900 related papers. A British company - Oxford Nanopore Technologies Ltd - has licensed the patents, and expects that their instrument, which incorporates the hemolysin pore, will sequence an entire human genome of 3 billion bases in less than a day, for well under $1000. IBM also put out a press release in October 2009, in which they described their progress toward nanopore sequencing using a synthetic nanopore.
It seems inevitable that some version of nanopore sequencing will become commercially available in the next few years. What will that mean for the future? I like to compare it to computing. In the 1970s, computing consisted of a terminal linked to a very expensive mainframe computer, and the hassle of working this way markedly limited ready access to computing power. In the late 1970s Steve Jobs and Steve Wozniac initiated a different approach to computing. I wish I had been clever enough to purchase some of the original Apple stocks, but instead I bought an Apple II Plus in 1981, with an amazing 64 kilobytes of memory. Then IBM came out with their first desk top computer, and a young guy named Bill Gates bought an operating system for $50,000, then licensed it to IBM as MS DOS. Suddenly, everyone had their own personal computer.
I think something similar will happen when desk top sequencing becomes available. The way it is done now is to send DNA samples to a central facility where the DNA will be analyzed by sequencing instruments costing over half a million dollars each, and a week or so later the sequence is sent back. But just as mainframes became obsolete with the advent of desk top computers, my guess is that in the next few years desk top sequencers will replace central facilities. You will be able to inject a few micrograms of viral or bacterial DNA into a port and couple of minutes later the sequence will be up on the screen. A sample of human DNA will take a little longer, perhaps a few hours.
Because DNA is at the core of all life processes, the implications of desk top sequencing are immense. Let's think about applications in medicine first. I am willing to predict that by 2020, anyone who wants to can own a CD containing their personal genome. Patients visiting their doctor will hand over their genome to help with diagnosis and treatment. For instance, the base sequence in DNA contains information that can help a physician predict a patient's reaction to pharmaceutical agents, so instead of simply treating everyone as an average patient, we will enter the era of personalized medicine so that adverse drug reactions will become increasingly rare.
A personal genome will also indicate the odds that someone will develop common diseases such as cancer, heart disease, or diabetes during their lifetime, so that they can adjust lifestyle and diet to minimize risk factors. Furthermore, if cancer does develop, a biopsy will be taken, the tumor genome sequenced, and the genetic information will be used to choose the best chemotheraputic agent.
Well, that's nice, but what else will be possible? The fact is that genes contain a history of evolution that has never before been available. Carl Woese pioneered this approach, and in 1990 published a paper based on ribosomal RNA sequences that convincingly divided life on Earth into three domains: Archaea, Bacteria, and Eukarya.
The three domains are rooted in an original microorganism called the Last Universal Common Ancestor (LUCA). After this bold beginning, the emerging science of bioinformatics has discovered much more. For instance, the human genome is riddled with fossil viral genes. We can reproduce the genome of a mammoth, compare our genome to that of neanderthals, and deduce the base sequence of a protein that functioned in an ancient mammal 100 million years ago, during the time that our mammalian ancestors led precarious lives in a world ruled by dinosaurs. When every evolutionary biologist has access to inexpensive sequencing, our knowledge of the biosphere is sure to explode, and we will understand human evolutionary history in unprecedented detail.
Then there is the question of life elsewhere. Ten years from now, NASA plans to have sophisticated rovers looking for evidence of life on Mars, targeting regions that look like they were once watery. There is no doubt that water exists on Mars, first as shallow seas three billion years ago, and now as extensive ice deposits covered by wind blown dust. If microbial life ever did exist on Mars, it needed to have liquid water as a medium, which is now ice. There are two ways that life could leave traces in ice. First, when living cells die they release polymers into the surrounding solution that would become embedded in the ice. A single cell might release millions of polymers which, if frozen, may survive as intact fragments even 3 billion years later. Second, where there are bacteria there are bacteriophages, virus-like organisms that feast on bacteria. On Earth, bacteriophages represent a substantial fraction of the marine biomass, and on Mars they would also leave fragments of polymers.
I am a member of a team that is proposing a set of experiments for a future Mars lander. One of the instruments will be an advanced nanopore device designed to search for single molecules of biological polymers in a melted sample of Mars ice. It would be truly wonderful if the little group of nanopore researchers turn out to be the prospectors who are lucky enough to strike the mother lode, in this case discovering convincing evidence that life can exist beyond the small blue dot of our planet.
- DNA Sieve-- Nanoscale Pores Can Be Tiny Analysis Labs
- Origin Of Life Funding: Benefactor Offers Hard Cash For A Good Research Idea
- Roche Licenses Nanopore Sequencing Technologies From Arizona State University And Columbia University For Rapid, Affordable DNA Sequencing
- Multi-Threading Genome Sequencing Could Mean Sub $1000 Cost
- Graphene: 'Wonder' Material May Revolutionize DNA Sequencing