Landmine Detection

Landmine Detection
by Robert S. Burlage

Using bacteria to detect the presence of landmines capitalizes on their response to explosive compounds. This research involves engineering, which requires a knowledge of the bacterial genes, and sensitive detection equipment. The bacterial chromosome may contain several thousand genes, some of which encode enzymes that make organic compounds, a useable food source. Bacteria have evolved control procedures to express these genes only when the food source is available, thus making much more efficient use of resources. Some bacteria have genes that are responsive to trinitrotoluene (TNT) or other explosive compounds. We can exploit this ability by attaching another gene to the TNT gene, effectively making a “bioreporter” bacterial strain that translates the compound the bacterium detects (TNT) into a signal we can detect (fluorescence). Our system uses the green fluorescent protein (GFP) gene or one of its derivatives to allow the cells to produce a bright fluorescence. Using conventional techniques of molecular biology, recombinant microorganisms can be constructed to produce the bioreporter signal when desired. GFP emits extremely bright fluorescence on exposure to ultraviolet light, which can be detected using optical or electronic means. The protein is extremely stable after it is formed, allowing detection for at least two days after application.

These bacteria can be an advantage with the detection of unexploded ordnance, particularly landmines. Our microorganisms have been designed to respond to TNT by expressing the GFP gene, thus becoming fluorescent. This process of landmine detection, which is summarized in Figure 1, involves spreading large numbers of the bacteria (e.g., by crop duster) over a region that is believed to contain landmines. Landmines are known to leak small amounts of TNT over time, so the bacteria will scavenge the TNT as a food source and express the appropriate genes. Expressing the GFP gene at that time will make the bacteria become fluorescent only where they detect the leaking TNT. Therefore, “hotspots” of fluorescent bacteria are apparent on the ground. An ultraviolet light is used to scan the ground for fluorescent hotspots, locating the leaking landmines. Ideally, the spraying would be done in the evening, and the detection would take place in the middle of the night when the fluorescence can easily be seen. However, more advanced photo detectors are envisioned which can block out sunlight and detect only the fluorescent signal.

In October 1998 we demonstrated this technology at a field site in South Carolina that contained buried sources of TNT. The bacteria located all five targets in a 300-m test plot. The explosive sources were buried about 15cm deep and ranged in size from 8g to 5kg. The explosives were buried approximately six weeks before the demonstration. The success of this work has encouraged us to refine the methods for field application and detection prior to further field tests. In addition, we will create another bacterial strain that is responsive to RDX.

This method is safer than the current methods of landmine detection and has the potential to be far cheaper as well. The reduced cost of finding landmines may ultimately make landmine usage less attractive as a weapon of war, since one of the present advantages of landmines is the substantially higher cost of finding a landmine versus placing a landmine, which is an advantage to a poorer country.