Vanderbilt University



Discoveries Featured

Visualizing Infection

By: Carol A. Rouzer, VICB Communications
Published: July 18, 2012

The combination of imaging mass spectrometry and magnetic resonance imaging provides a three-dimensional picture of the response to infection.

When pathogenic bacteria invade the body, they trigger a complex panoply of cellular and molecular defenses, which comprise the inflammatory response (Figure 1). For decades, investigators have worked to define the multitude of players in this response and to understand how they interact to defend the body against infection. Much progress has been made; however, even the best technological advances have failed to provide a three-dimensional picture of the host-pathogen interaction in real time. Now, Vanderbilt Institute of Chemical Biology investigators Eric Skaar and Richard Caprioli along with Vanderbilt Institute for Imaging Science investigator John Gore provide a new approach to address this important question [A. S. Attia et al., (2012) Cell Host & Microbe, 11, 664].

Figure 1. Scanning electron micrograph of Staphylococcus aureus (green) being engulfed by a leukocyte, one of the earliest steps in the response to infection. Image reproduced through the courtesy of Wikimedia Commons under the GNU Free Documentation License.

Numerous methods are available to investigators interested in understanding host defense mechanisms. Fluorescence activated cell sorting (FACS) provides the means to identify and quantify inflammatory cell populations. Expression profiling measures changes in transcription levels for cytokines and other effector proteins. Mass spectrometry-based metabolomics detects the production of small molecule mediators. All of these approaches provide valuable quantitative data. However, because they require tissue homogenization, the spatial context is lost. Immunohistochemistry addresses the need for spatial information by delineating the location of specific cells or molecules within a two-dimensional tissue section, but it cannot be used to monitor the response in real time or in the context of the whole organism. An additional disadvantage to these traditional methods is that they require prior knowledge of the cells or molecules to be studied, and most also require a specific probe (e.g., antibody, oligonucleotide) for target detection.

A recent development in the study of host defense is the application of magnetic resonance imaging (MRI), which allows noninvasive real-time monitoring of changes in infected organs. Through its high soft tissue contrast, MRI can detect swelling, abscess formation, and alterations in organ architecture as infection progresses. However, MRI provides no information on the cell populations or effector molecules present at the affected site. To address this need, Attia et al. applied matrix-assisted laser desorption/ionization imaging mass spectrometry (MALDI IMS) to directly detect the distribution of macromolecules in tissue sections. Unlike immunohistochemistry, MALDI IMS does not require prior knowledge of the molecules to be imaged, and probes directed to the target molecules are also not required. Like immunohistochemistry, however, MALDI IMS provides a static, two dimensional picture of an infected site. By melding MALDI IMS with MRI, Attia et al. were able to obtain a three-dimensional view of the response to infection.

Staphylococcus aureus (Figure 1) is noted for its ability to infect almost any tissue, thereby causing a diverse array of diseases. The recent emergence of methicillin resistant S. aureus (MRSA), which is resistant to all first-line antibiotics, has made this pathogen a target of interest for new therapeutic approaches. Intravenous inoculation of mice with S. aureus results in systemic colonization with the bacteria, which can be cultured from nearly every organ. However, only the kidneys display abscess formation, which is the hallmark of S. aureus-mediated disease in humans. Attia et al. used this model of S. aureus infection to show that kidney abscesses could be visualized with MRI as early as 96 hours after inoculation (Figure 2), consistent with a bacterial count of 1 x 107 CFU in the infected kidneys. Treatment of 96 hour-infected mice for an additional 96 hours with the anti-inflammatory antibiotic linezolid reduced the bacterial burden as well as the number of inflammatory neutrophils and macrophages in the kidneys. MRI of linezolid-treated mice indicated smaller, more organized abscesses at the end of the four day therapy. In contrast, the kidneys of untreated mice revealed an increase in the size and number of abscesses with destruction of normal organ architecture following the additional 96 hours of infection.

Figure 2. MRI image of the kidneys of a mouse infected with S. aureus. The arrows show the locations of abscesses. Serial images such as this can be joined to provide a three-dimensional view of the entire kidney. Since this technique uses live animals, the progression of infection can be monitored over time. Image kindly provided by the Skaar, Caprioli, and Gore laboratories. Copyright 2012.


To learn more about the response to S. aureus infection, the investigators sacrificed mice 96 hours after pathogen inoculation and prepared serial sagittal sections for MALDI IMS analysis. The results revealed masses for proteins that were present in selected organs and proteins that were present throughout the body in both infected and control animals (Figure 3). They also revealed proteins that were present only in infected mice. Of particular interest were the signals at m/z 5,020, which the investigators identified as a truncated form of α-globin, and at m/z 10,165, identified as the calgranulin A subunit of calprotectin. Although both were found primarily in the kidney, α-globin was present in infected and control mice, while calprotectin was strongly associated with abscesses and found in infected mice only (Figure 4). Thus, these two proteins provided markers that differentiated diseased and healthy kidneys.

Figure 3. ALDI-IMS images of a transverse section of a mouse.  Each image shows the location of a different protein, denoted by its mass to charge ratio (m/z).  Note that some proteins are localized to only one tissue, such as m/z 6714 and m/z 11451, which are found in the brain and skin, respectively.  Other proteins exhibit a broader distribution, such as m/z 7932 and m/z 9975. Images kindly provided by the Skaar, Caprioli, and Gore laboratories. Copyright 2012.

To obtain a three-dimensional view of S. aureus-mediated pathology, the investigators subjected infected mice to MRI, and then sacrificed them for the preparation of serial sagittal sections, which were photographed and analyzed by MALDI IMS. Reconstruction of the images into a continuous blockface volume through serial registration provided the foundation for tying together the MALDI IMS and MRI data. The results showed that the kidneys of mice that had been infected with S. aureus for eight days were severely deformed by the presence of large abscesses. The abscesses lacked distinct architecture, and a strong calprotectin signal could be detected throughout the kidney. In contrast, the kidneys of mice that had been treated for four days with linezolid retained normal tissue architecture in the face of much smaller abscesses. Calprotectin was associated only with the abscesses in these kidneys. In both cases, α-globin was confined to the kidney cortex, and could not be observed in abscesses.

Figure 4. MALDI-IMS readily detects calprotectin in infected animals, where it is primarily localized to kidney abscesses.  The location of abscesses as seen by light microscopy on the right correlates exactly with the location of the MALDI-IMS signal for calprotectin as seen on the left.  Image kindly provided by the Skaar, Caprioli, and Gore laboratories.  Copyright 2012.


The results confirm the potential value of MALDI IMS for detecting infection-specific proteins within a spatial context. This method requires no prior knowledge of or specific probes against the target molecules. However, considerable effort will be required to identify all of the signals that can be obtained from this type of imaging study. More importantly, these experiments illustrate how the sensitivity of MALDI IMS can be combined with the structural resolution of MRI to obtain a three-dimensional picture of the response to infection. Application of this approach provides the foundation for a systems-biology approach to study the host-pathogen interaction.











Vanderbilt University School of Medicine | Vanderbilt University Medical Center | Vanderbilt University | Eskind Biomedical Library

The Vanderbilt Institute of Chemical Biology 896 Preston Building, Nashville, TN 37232-6304 866.303 VICB (8422) fax 615 936 3884
Vanderbilt University is committed to principles of equal opportunity and affirmative action. Copyright © 2013 by Vanderbilt University Medical Center