Imaging Malaria-Infected Red Blood Cells with AFM-IR

David Perez-Guaita

In biology, the study of intracellular structures is important and requires analytical techniques with submicrometer resolution. Atomic force microscopy-infrared (AFM-IR) spectroscopy is one technique that has the required lateral spatial resolution to observe such structures. David Perez-Guaita, PhD, a research fellow in the group of Associate Professor Bayden R. Wood at the Centre for Biospectroscopy at Monash University in Australia, is pioneering work applying AFM-IR to the study of red blood cells infected with the malaria parasite. The goal is to obtain information about the phenotype of the parasite, particularly of drug-resistant strains, which will contribute to the development of diagnostic and control measures. He recently spoke to us about this work.

You have recently used AFM-FT-IR to acquire nanoscale images of the intracellular structures of malaria-infected red blood cells (1). Can you briefly describe the steps in this approach?

The procedure can be divided in three phases. The first is sample preparation. The approach we applied is a standard procedure in parasitology, which involves the culture and synchronization of parasites and preparation of a smear. However, instead of using a glass slide, the infected red blood cells are smeared on a CaF₂ window, which is transparent to infrared light.

The second phase is generating AFM-IR multispectral images. The first stage of this second phase is to record an AFM image of a red blood cell. If the red blood cell is infected, it will exhibit topographical features caused by the presence of the parasite. The second stage is generating AFM-IR maps at different wavenumber values based on parasite bands identified in the AFM-IR spectra. Each map is a measurement of the thermal expansion produced by the laser, which is related to the absorbance of the material under the AFM tip. The wavenumber values measured are selected taking into account the molecules targeted by the analysis.

The third phase is statistical analysis, which requires coregistering the individual maps and applying multivariate data analysis to extract information about the composition of the different topographical features.

What was the role of multispectral image analysis in this work?

This critical step is to extract useful chemical information from the multispectral images. First, the different maps have to be registered to create a spectral hypercube containing a set of intensity values (one for each wavenumber) for each pixel. Then multivariate data analysis is applied using k-Means Cluster Analysis (kMCA). For example, pixels are grouped accordingly to their similarities. This approach makes it possible to identify different subcellular structures such as a nucleolus or food vacuole. Using kMCA also provides the average values for the different classes, which can be used to study the composition of the different substructures identified.

What was the role of Raman confocal microscopy in the study?

Raman was used to confirm the presence of hemoglobin (Hb) and hemozoin (Hz) in the locations identified by AFM-IR. However, the spatial resolution of Raman spectroscopy is limited by the laser diffraction limit and hence is much lower than the lateral resolution achieved using the AFM tip.

What challenges did you face in developing the approach?

One of the main challenges was to select the best conditions for image acquisition. Parameters such as laser power, acquisition time, and the number of coadded scans must be tuned to obtain high quality images. There is a trade-off between the signal-to-noise ratio and the acquisition time. Furthermore, cells can be damaged if the laser power is too high or the exposure time is too long. In fact, we burned several samples before we were able to find suitable conditions for the measurements.

What were you able to see with this technique that was not possible to see with other techniques?

At the Monash Center for Biospectroscopy, we have been trying to develop IR- and Raman-based technologies to study malaria parasites for several years. The use of AFM-IR has two main advantages compared with other molecular imaging approaches. First, the lateral resolution of techniques such as Raman and IR spectroscopy is diffraction limited. In the case of Raman, Hz and Hb are strong Raman scatterers and hence it is difficult to investigate other molecules in their vicinity. FT-IR spectroscopy, on the other hand, is able to detect changes in lipid and DNA composition, but the wavelength applied is in the size range of the parasites, making it impossible to identify subcellular structures. Our study shows that AFM-IR was able to distinguish and chemically characterize subcellular structures with submicrometer resolution. Secondly, AFM-IR not only offers information about the chemical composition of the subcellular structures, but also provides topographical information from the AFM maps. Unfortunately, the approach has also some disadvantages. The main disadvantage is that the penetration depth of the laser is not known with certainty and hence it is hard to know to what extent we are measuring the whole cell or just the surface.

How can these images help in malaria research or other biology research? What are your next steps in this work?

This paper described is a proof of concept study that demonstrates the capabilities of AFM-IR in the study of the Plasmodium sp. parasites. We are currently trying to exploit these capabilities in the study of strains of the Plasmodium sp. parasites resistant to artemisinin. The use of artemisinin derivatives is threatened by the rapid spread of drug resistance in Southeast Asia. The World Health Organization (WHO) has warned that if the resistance reaches Africa or India, “There is a limited window of opportunity to avert a regional public health disaster, which could have severe global consequences.” It is known that artemisinin and artemisinin derivatives are activated by Fe-containing compounds present in the Plasmodium parasite to form carbon-centered free radicals, which damage the proteins of the parasite. Some strains have developed resistance to this mechanism and we want to know why. So our next steps in this work will include  experiments comparing the chemical composition of resistant and susceptible strains using AFM-IR, and  inoculation studies. By investigating the spectral differences between untreated and treated samples, one can identify the phenotypic changes induced by a drug. The study of the IR spectra of a treated susceptible strain of P. falciparum will potentially increase our knowledge of the mode of action of artemisinin. It will indicate the compositional changes induced by the presence of the reactive oxygen species from the activated artemisinin. The discovery of spectral markers associated with resistance can assist in the development of new drugs and increase the effectiveness of treatments.                                                                  

Reference

• D. Perez-Guaita, K. Kochan, M. Batty, C. Doerig, J. Garcia-Bustos, S. Espinoza, D. McNaughton, P. Heraud, and B.R. Wood, Anal. Chem., in press. DOI: 10.1021/acs.analchem.7b04318