Editor’s Note: This is an expanded version of the article that appeared in the November 2008 issue of Law Officer.
If only we had access to the world described by television. Phaser pistols would disable bad guys silently, bloodlessly, and unfailingly, and they wouldn’t even bump their heads when they dropped. The most complex case could be solved in an hour, and we could locate and bring in any suspect or witness on a moment’s notice. Nearly all cops and the people they worked with would be quick of wit and drop dead gorgeous. And identification via DNA analysis would require no more time or money than making a photocopy of a single piece of paper. On that last one, we’re making a little progress–but it’s going to be a while before you see self-contained DNA test kits in your equipment bag.
Forensic DNA examinations have proven their worth time and again in heavyweight crimes like rape and murder, and they’re now being extended to lesser offenses, such as burglary. This is partly because the facilities that perform the tests have expanded in number and capacity, and partly because the amount of DNA needed to perform an analysis has decreased markedly. Some agencies are now routinely swabbing the area of a latent fingerprint after the print has been developed with fingerprint powder and lifted. Conventional wisdom held that the powder would contaminate the sample excessively, and the lift tape would peel the DNA off with the print powder. We now know that there is usually residual biological material left behind on the touched surface, and techniques have been refined enough to make use of it.
Resistance to expanded use of DNA analysis comes about via two traditional limiting factors: time and money. Most law enforcement agencies don’t have their own in-house forensic labs, and have to ship their samples to state or regional lab. These labs are often overwhelmed with evidence to analyze, and barring special circumstances, samples are placed in queue to wait their turn at the lab bench. If the lab runs over their budget for these tests, the samples are held until the next funding cycle, while more accumulate at the end of the line. Meanwhile, a suspect is still loose and unidentified, even though they might be quickly arrested if the local cops knew who he or she was.
The Lab-on-a-Chip Concept
Below the evidentiary standard of conclusive examinations are presumptive tests. We already use presumptive tests in other areas of law enforcement like DUI and narcotics investigations. An officer at the side of the road might have a handheld preliminary breath tester that will provide a measure of a violator’s blood alcohol content, typically accurate to 10%. Self-contained presumptive narcotics test kits reveal the presence of contraband drugs via a color indicator. Either of these, with or without additional evidence, contribute to the standard of “probable cause” sufficient to support a warrantless arrest, or to obtain an arrest warrant from a judge.
A “lab on a chip” would be a form of presumptive test, not intended to be conclusive, but sufficient to establish probable cause for an arrest or search warrant, with or without other evidence. The technology would allow investigators to introduce a DNA-bearing specimen into a handheld analyzer, allow the analyzer to process the sample, and then compare the results to those in a database such as CODIS (Combined DNA Index System). The television version will do this with any biological specimen, produce results in under five seconds, and display the owner’s name, mug shot, and present location on an integrated panel in full color. The reality is likely to be something a bit less snazzy.
Conventional analysis
Depending on how sophisticated your crime lab is and how recently they obtained their hardware, the existing analysis process requires 24 hours to several days to complete, and an analyst may be able to work on only one sample at a time. Much of the processing has been automated, allowing for faster throughput and analysis of multiple samples in a single “run.” A question like, “How many samples can a lab run at one time, and how fast can they run them?” is answered by another question: “How much money are you willing to spend?” New technology emerges constantly, but with price tags often going well into six figures.
There are three principal steps in DNA analysis:
- DNA extraction from the collected specimen, leaving behind tissue, proteins, and other unwanted material mixed with the specimen
- Polymerase chain reaction (PCR) amplification of the DNA, replicating the DNA strands of interest millions of times and allowing for analysis of a tiny bit of DNA, and
- Separation of short tandem repeat (STR) fragments of the amplified strands into markers that can be compared to other samples.
The output of this last step is usually represented as a graphic of several vertical rows of widely-spaced horizontal lines of varying darkness and thickness, visually illustrating the markers. This is produced by applying the output from the STR fragment separation onto a gel substrate, then applying an electric current through the gel, or by using capillary electrophoresis. In gel electrophoresis, the phosphate groups in the DNA molecule have a negative charge, so they will move through the gel towards the positive electrode, or anode. Larger molecules move through the gel more slowly than the smaller ones, so the molecules separate into bands. We see these as the short horizontal lines on the gel. The presence of a band indicates a specific allele, or one half of a gene pair. It is the combination of these alleles that make up our genetic “fingerprint.”
An alternative–and these days, more common–method of separating and identifying the alleles is by using a capillary system, rather than a gel. Capillary electrophoresis (CE) uses tiny tubes to contain the DNA material and has several advantages over gel electrophoresis. A capillary system allows for higher voltages in the CE process, completing the job in minutes, rather than hours. Higher voltage means more heat, and too much can “cook” the gel. Automated CE processors run multiple samples in one pass, so throughput is improved considerably. CE also requires a smaller volume of sample than gel electrophoresis.
Prototype devices
A lab-on-a-chip consolidates most or all of the DNA analysis process into a single portable device. There are already some lab-on-a-chip products in production, but none designed specifically for the forensics market. The existing devices are mostly for screening samples for a particular allele or group of alleles, such as might be contained in a bacterium responsible for causing a disease. A self-contained test can determine if someone is suffering from a specific infection so that they can be immediately quarantined and/or treated when time is critical. These devices are pre-programmed for that profile, so the result is a simple “positive” or “negative.”
Forensic requirements are more complex, because the end result will be a profile not anticipated. The memory capacity of a lab-on-a-chip isn’t nearly large enough to contain a database like CODIS, so the result will always have to be exported for comparison.
A lab-on-a-chip is also going to be limited in what type of sample it can accept for analysis. DNA specimens arrive at the lab in any number of forms: cotton swabs of buccal cells taken from inside the mouth, blood in sealed test tubes, fabric with dried blood, semen or other biological material on it, cigarette butts, soda cans, condoms, and anything else. Criminalists assigned to process these use a variety of procedures to isolate the DNA from its carrier, involving various solvents, washes, test tubes and disposable plastic “wells,” centrifuges, and heating blocks. Not every specimen collected in the field is going to lend itself to lab-on-a-chip analysis.
There have been at least three proof-of-concept lab-on-a-chip devices developed for forensic use. One of these requires a specimen that has already been subjected to traditional DNA extraction procedures; another will accept extracted DNA or untreated whole blood. One device performs the STR portion of the analysis only, and the other performs the extraction and PCR amplification stages, producing a sample for gel or capillary electrophoresis analysis.
Complex process
Following is a description of the process required for the first device described above. It’s almost certainly more detail than you want, but I’ve included it for a reason. The test was conducted at a mock crime scene in cooperation with the Palm Beach County Sheriff’s Office (PBSO) in Florida, and reported in a paper by Peng Liu et al. Blood specimens were left on paper towels and a blue fabric shirt, which were in turn on top of a cardboard “victim” at the crime scene. The PBSO Forensic Biology Unit deployed a mobile command unit equipped with a power generator, air conditioning, and a satellite-based Internet link, along with the gear necessary to perform the DNA extraction step. The lab-on-a-chip device was contained in a separate room of the mobile command unit so prevent contamination of samples.
The samples were first extracted using a Maxwell 16 and a DNA IQ Sample Kit, both products of the Promega Corporation. Before the samples could be introduced into the Maxwell 16, they had to be first placed into a microcentrifuge tube with three buffer reagents and heated for 30 minutes in a heat block. Lysis (dissolving) buffer was added, the tubes were vortexed and centrifuged with more specialized lab equipment, and the DNA-containing lysates was separated from the substrates (the paper and fabric) with a spin basket. The lysates were then introduced into DNA IQ cartridges for processing in the Maxwell 16. The Maxwell 16 processing required 30 minutes. The last step before introducing the samples into the prototype CE device was to centrifuge them again to concentrate the DNA extracts.
The extracts were loaded into the CE device. A different device was used for each sample. The PCR step required two hours, and then CE separations were performed three times, each requiring 30 minutes. The result of the CE separation steps was sent to a mock CODIS database set up by PBSO and matched with the profiles of the original donors via the wireless Internet link. Total time required for the analysis: six hours.
The other two devices use processes that are similarly painful to describe and even more so to read. It’s important to have some idea of what goes on here to appreciate the level of expertise required to conduct these analyses, and the technological accomplishment that these devices represent. I was tempted to compare this with the shrinking of a room-size Univac computer (required over 6,000 square feet of space, 160 kilowatts of power, and cost $1.9 million in 1958 dollars) down to a pocket calculator (easily fits in a watch, runs on solar power, so cheap they’re given away), but those devices just managed electrons and magnetic charges. Managing wet chemistry precise to a molecular level, and reducing the size of the lab to something you can hold in your hand is even more impressive, especially if you’ve ever performed analyses requiring this kind of tolerance.
Understanding the process is also important to know why these devices won’t be available at any kind of reasonable cost anytime soon. Mass production is required to lower costs, and the practical application for these tools at this stage of development is limited. Even if a single device were capable of performing the entire analysis process–which none of these are–I can’t think of a practical scenario where having a result available in six hours without returning to a lab would be all that useful. Time and technology will see refinements that will get the equipment down to something the size of a breath alcohol tester, which are affordable and routinely taken into the field for use. I expect to see a device like this within five to ten years, maybe sooner.
In the meantime, perhaps you will now have more respect for the criminalists that perform your DNA work for you. And if anyone knows where that TV-land technology is available, I want one of those tabletop 3-D computers they have on CSI: Miami, and Calleigh Duquesne to show me how to operate it.
References:
1. Wen, J., Guillo, C., Ferrance, J., Landers, J. Microfluidic-Based DNA Purification in a Two-Stage, Dual-Phase Microchip Containing a Reversed-Phase and a Photopolymerized Monolith. Analytical Chemistry 2007, 79 (16), 6135-6142.
2. Wen, J., Legendre, L., Bienvenue, J., Landers, J. Purification of Nucleic Acids in Microfluidic Devices. Analytical Chemistry 2008, 80 (17), 6472-6479.
3. Ausserer, W.A., Bousse, L., Gallagher, S.J., Kennedy, C.B., Phan, H.L. Automated Lab-on-a-Chip Analysis of DNA Fragments. JALA – Journal of the Association for Laboratory Automation 2001, 6 (3), XIV-XV.
4. Liu, P., Yeung, S.H.I., Crenshaw, K.A., Crouse, C.A., Sherer, J.R., Mathies, R.A. Real-time forensic DNA analysis at a crime scene using a portable microchip analyzer. Forensic Science International: Genetics 2008, 2, 301-309.
5. Teles, F.R.R., Fonseca, L.P., Trends in DNA biosensors. Talanta 2008, doi:10.1016/j.talanta.2008.07.024
6. Butler, J.M. Forensic DNA Typing: Biology, Technology and Genetics of STR Markers, Second Edition 2005. Elsevier.