Efficient and sensitive methods to determine whether, and to what extent, a person is infected with malaria should help to improve treatment. A high-tech approach, using mass spectrometry, may be the answer.
Years ago my PhD adviser, John Fenn of Yale University, told me that he wanted to develop a mass spectrometer — the well-known tool for identifying and quantifying compounds and for determining their structure — that could be used in the doctor's office. People would deposit their sample in one end of the machine, and out from the other end would come a list of diagnoses based on unambiguous 'digital' mass-spectrometry data. Of course, this didn't actually happen; instead, Fenn invented electrospray mass spectrometry1, which, together with matrix-assisted laser-desorption (MALDI) mass spectrometry2, laid the foundation for the current boom in mass-spectrometry-based proteomics, one of the hottest fields in functional genomics3.
Out of the limelight, however, mass spectrometrists are busily inventing technologies that bring us ever closer to the routine use of mass spectrometry in diagnosis. In a paper just out in Analytical Chemistry, for instance, Demirev and colleagues4 describe how they devised an ingenious method to detect the presence of the malaria parasite at a sensitivity of just ten parasites per microlitre of human blood. The method, which was demonstrated with cultured parasites, addresses one of the most serious worldwide health problems5. It is clinically important to determine not only whether patients are infected with the malaria parasite, but also how heavily they are infected; that makes it easier, for example, to determine how effective a treatment is.
Demirev and colleagues' approach4 takes advantage of a peculiarity in the parasite's life cycle: while circulating in the human blood stream, the parasite grows inside red blood cells (Fig. 1). It builds up elaborate structures, including inner membranes, inside these cells, before bursting them to release progeny parasites. The material for this build-up comes almost entirely from the breakdown of haemoglobin, which constitutes 95% of the red blood cell. After haemoglobin has been digested by the parasite, the oxygen-carrying haem group remains and the parasite stores it in an almost crystalline form in special vacuoles. The haem group is photoactive and turns out to be easily detectable by direct laser-desorption mass spectrometry. As it builds up to very high levels inside the parasite, it is a promising diagnostic marker.
In their technique, Demirev et al. exploit the fact that the membranes of red blood cells contain cholesterol, so as to lyse the cells while keeping the inner, cholesterol-lacking, membranes of the parasite intact. They then isolate the parasite and apply its contents to a metal target, on which many samples can be arrayed. The target is transferred into the vacuum system of a MALDI instrument and illuminated by short laser pulses, which volatilize and ionize the haem (assisted not by a matrix as in normal MALDI, but by the photoactive properties of the haem). The haem ions are then accelerated and shot down a flight tube under vacuum, and their mass is determined by the arrival time relative to the laser pulse — a technique called time-of-flight mass spectrometry. Because of the high concentration of haem in the parasites, and because the normal red-blood-cell contents have been removed, excellent sensitivity and discrimination between infected and uninfected blood can be obtained.
The authors show that as little as one parasite per microlitre should be detectable with more specialized equipment and optimized procedures. Moreover, characteristic fragmentation of the haem is observed under certain conditions, increasing the specificity of the measurement. The method also seems to be quantitative — traditionally a weak point with mass spectrometry, especially MALDI mass spectrometry, because of the many factors that influence peak height in a mass spectrum. Even if the quantity of haem could not robustly be determined from a normal mass spectrum, there are now many variants of stable-isotope methods that could provide effective quantification by simple matching of peaks to an internal standard.
How does this new method stack up against existing procedures? That depends on which parameter we look at. It seems clear that measuring the haem within parasites in blood is a clever way to diagnose malaria, and the method already compares well with current techniques in terms of sensitivity. History shows that once it gets a foothold into an analytical field, mass spectrometry usually conquers it — witness such disparate recent developments as the shift to the use of mass spectrometry rather than two-dimensional gel electrophoresis in proteomics, and the routine analysis of single-nucleotide repeats in nucleic acids. Mass spectrometry's advantages are its precise, unambiguous and machine-readable output and the fact that measurements can be performed in a very short time if there are standard conditions. For malaria diagnosis, many samples could be prepared in parallel and measurement per sample should take no longer than a second or so.
However, malaria diagnosis is most needed in remote rural areas, often with no electricity. Today's 'gold standard' for diagnosis is the thick and thin blood smear, which consists of counting parasites in stained blood smears by microscopy. That method is decidedly low-tech, having been developed a hundred years ago6; it is difficult to perform; and in practice its sensitivity is often less than the World Health Organization's recommendation of 100 parasites per microlitre of blood. But it does have the great advantage of being deployable in the places where malaria is endemic. These are not hospitable environments for today's high-tech mass spectrometers. That said, different types of mass spectrometer have been used in the field for more than ten years. For instance, battery-operated ones are used by the armed forces and in environmental work. Moreover, small mass spectrometers can operate in challenging environments (one was delivered to Mars by the Viking space probe), and even the time-of-flight spectrometer used with MALDI can in principle be miniaturized tremendously, possibly even to palm size7.
The biochemical purification steps of Demirev and colleagues' technique, including centrifugation, would also have to be compatible with tough environments. So it is interesting that analysis of whole blood by direct laser desorption does not result in an appreciable haem signal, presumably because the haem in normal red blood cells is tightly bound to haemoglobin (P. A. Demirev, personal communication). This suggests that much simpler preparation methods, without centrifugation steps to remove normal red blood cells, might be feasible. Finally, the method as described does not distinguish between different types of malaria parasite. But this could in principle be done by proteomics technology using mass spectrometry.
So the technical potential for a routine and relevant malaria-detection assay is certainly here, and further developments will — as usual — depend to a certain extent on the political will and economic or philanthropic interest in better methods to diagnose this predominantly tropical disease. We can hope that the current push to detect biological-warfare agents, in which mass spectrometry has a prominent role, will aid diagnostic efforts as a side effect. There are also many, equally promising, efforts to detect proteins that are characteristic of other diseases in blood or other bodily fluids — all fuelled by the rapid technical developments in mass spectrometry. Perhaps we are coming full circle to realize Fenn's dream of the mass spectrometer as a universal diagnostic machine.
Fenn, J. B., Mann, M., Meng, C. K., Wong, S. F. & Whitehouse, C. M. Science 246, 64–71 (1989).
Hillenkamp, F., Karas, M., Beavis, R. C. & Chait, B. T. Anal. Chem. 63, 1193A–1202A (1991).
Mann, M., Hendrickson, R. C. & Pandey, A. Annu. Rev. Biochem. 70, 437–473 (2001).
Demirev, P. A. et al. Anal. Chem. 74, 3262–3266 (2002).
Nature Insight: Malaria. Nature 415, 669–715 (2002).
Moody, A. Clin. Microbiol. Rev. 15, 66–78 (2002).
Cotter, R. J., Fancher, C. & Cornish, T. J. J. Mass Spectrom. 34, 1368–1372 (1999).
About this article
Travel Medicine and Infectious Disease (2008)
How to make big molecules fly out of liquid water: applications, features and physics of laser assisted liquid phase dispersion mass spectrometry
Physical Chemistry Chemical Physics (2007)
Rapid Quantitative Detection of Bovine Serum Albumin in Blood Serum with Seeded Liquid Beam Desorption Mass Spectrometry
Analytical Letters (2006)
Time-Resolved Micro Liquid Desorption Mass Spectrometry: Mechanism, Features, and Kinetic Applications
Australian Journal of Chemistry (2006)
Neural Networks (2005)