Kristine Glunde (Presenter)
The Johns Hopkins University School of Medicine
Bio: Dr. Glunde received her Ph.D. in Chemistry from the University of Bremen in Germany in 2000, followed by post-doctoral training at The Johns Hopkins University School of Medicine. Since joining the Faculty of the Radiology Department at Johns Hopkins in 2003, Dr. Glunde was promoted several times and currently is Associate Professor of Radiology and Oncology. She has been involved in several research studies on multi-modal molecular imaging of cancer as Principal Investigator or Co-Investigator and has mentored more than 25 students and post-doctoral fellows. She has published over 85 publications in the field of molecular and functional imaging of cancer, and has been awarded several NIH R01 and other grants as Principal Investigator. Dr. Glunde is frequently invited to present her work at conferences and seminar series, and serves on grant review panels and editorial boards.
Authorship: Kristine Glunde (1), Ron M. A. Heeren (2)
(1) The Johns Hopkins University School of Medicine, Baltimore, MD (2) Maastricht MultiModal Molecular Imaging (M4I) Institute, Maastricht, The Netherlands
| Short Abstract Breast cancer is a molecularly and spatially heterogeneous disease that continues to escape today’s treatment options. On the road towards personalized medicine, it is paramount to obtain detailed molecular tissue images of cancerous regions that seamlessly integrate with histopathology as well as the clinical workflow of histopathology. We have shown that mass spectrometry imaging (MSI) can be employed to obtain distinct biomolecular signatures, which include metabolites, lipids, and proteins, from aggressive hypoxic and metabolically de-regulated regions of preclinical breast tumor models. Our body of preclinical work has built a framework from which we will be able to translate our preclinical findings directly to large patient cohorts. |
Long Abstract
Introduction
Breast cancer is a molecularly and spatially heterogeneous disease that continues to escape today’s treatment options owing to its genomic plasticity, which results in an enormous ability to adapt to different microenvironments, including the diverse tissue microenvironments at metastatic sites. The emerging availability of personalized, molecularly targeted therapy to treat breast cancer based on its particular molecular subtype necessitates an accompanying ability to precisely characterize its molecular composition for diagnosis, assessment of its aggressiveness, prediction of metastatic risk, and a personalized treatment plan. This combined molecular-spatial tissue heterogeneity of breast cancers is best visible at the cellular and tissue level, and molecular pathology tools provide a first insight into the morphological heterogeneity of the disease. Recent developments allow the direct visualization of this molecular heterogeneity with mass spectrometry imaging (MSI) directly from cells and tissue. Label free imaging with mass spectrometry uses an intrinsic molecular parameter, the molecular mass, to directly visualize the distribution of a wide variety of biomolecules such as proteins, lipids, and drugs at the surface of tissue sections. It is evident that insight into the local molecular heterogeneity is required to determine an optimal treatment strategy. Conventional pathology techniques, however, do not provide this insight. The leap change in molecular imaging and pathology will occur when two elements, high-resolution in vivo molecular morphology and broad multi-scale ex vivo molecular insight in three dimensions can be directly combined. This will result in a new era in clinical pathology where true molecular pathology takes the field beyond ‘just images’ towards personalized diagnostics, prognostics, and ultimately to precision medicine. This philosophy lies at the heart of our research program, which seeks to identify and evaluate molecular markers for precise diagnosis and prognosis of breast cancer with MSI. In our studies, we have combined MSI with other molecular imaging approaches to harness known features of breast tumor aggressiveness such as tumor hypoxia and deregulated tumor metabolism.
Methods
Our research program has developed a number of novel approaches to achieve co-registered 3D multimodal imaging, which combines MSI with magnetic resonance (MR) imaging (I), MR spectroscopic imaging (MRSI), fluorescence imaging, and histology of the same tumor [1-3]. Among these is a novel system of embedding fiducial markers in the gelatin that holds the cancer tissue for cryosectioning [1-4]. These fiducial makers can be detected by optical imaging of fresh tissue sections, histology, and MSI [1,3]. We have also developed MSI of the red fluorescent protein [2], which was used to detect hypoxic regions by means of hypoxia-driven tdTomato red fluorescent protein (RFP) expression in genetically engineered tumor models used in our studies [4]. As alternative approach for clinical samples, we have developed MSI of one of the clinical gold standards for hypoxia imaging, pimonidazole, which is a 2-nitroimidazole used as an exogenous, systemically injected hypoxia marker that it is activated and retained in viable hypoxic cells [5].
Results and Discussion
Leveraging our novel approaches that combine MSI with molecular imaging, we have identified characteristic metabolic, lipid, and protein signatures in hypoxic regions of aggressive breast tumors. We showed that the phospholipid metabolite phosphocholine, which is the largest signal in the non-invasively MRS(I)-detected total choline signal (tCho), is significantly increased in hypoxic tumor regions due to hypoxia-response elements in the promoter regions of choline kinase alpha [6,7]. We proposed that the detection of highly elevated tCho foci in large breast tumors is indicative of tumor hypoxia [6,7]. We demonstrated that secondary ion mass spectrometry (SIMS) and matrix assisted laser desorption ionization (MALDI) MSI are able to detect free choline, phosphocholine, and glycerophosphocholine and thereby identify distinct spatio-molecular signatures in the breast tumor microenvironment of two differentially aggressive breast tumor models [8]. In our subsequent studies, breast tumor xenografts grown from MDA-MB-231-HRE-tdTomato cells, which contain a red fluorescent tdTomato protein construct under the control of a hypoxia response element (HRE)-containing promoter driven by HIF-1α, were used to detect the spatial distribution of hypoxic regions.
Following 3D MRSI of orthotopic MDA-MB-231-HRE-tdTomato breast tumor xenografts in vivo to detect tCho, tumors were cryo-sectioned throughout, followed by on-tissue tryptic digestion and α-cyano-4-hydroxycinnamic-acid matrix deposition, to perform MALDI MSI and detect tryptic peptides to identify MSI-based molecular signatures of the high-tCho regions. We fused MRSI and MSI in 3D as previously described [1,3,4,7,9,10]. The 3D MRSI tCho volume was segmented as high-tCho-containing and low-tCho-containing areas and mapped onto the corresponding MALDI data. This data was not linearly separable, and was therefore analyzed by least absolute shrinkage and selection operator (LASSO) to classify high- and low-tCho-containing voxels by means of the MALDI MSI data [9,10]. Candidate m/z peaks, which mostly contributed to the differentiation, were obtained from the parsimonious sets of features for discriminating between high-tCho and low-tCho generated from LASSO and identified through an in-house built accurate mass and time (AMT) tag peptide/protein database [4]. The 3D tCho distribution detected by in vivo MRSI at 3.2 ppm was heterogeneous [9,10]. A tCho binary image was obtained after image segmentation from each one of four tumors in this study. 4-folds cross validation was applied to all the voxels within the tumors, and an area under the curve (AUC) of 0.8370 under the receiver operating characteristic (ROC) curve was obtained [9,10]. Nineteen candidate tryptic peptides were obtained by LASSO classification of high- versus low-tCho-containing voxels from a total of four tumors [9,10], some of which were also found in our recently published analysis of hypoxic breast tumor regions [4], which is not surprising given the fact that tCho and hypoxia overlap to a large extent [6,7]. By combining MRSI with tryptic on-tissue digestion MALDI MSI, followed by registration and tCho-voxel classification, we identified for the first time some specific proteins that are differentially expressed in aggressive breast tumor regions that contain high tCho.
To further characterize the lipidomic signatures of hypoxic breast tumor regions, we employed the orthotopic MDA-MB-231-HRE-tdTomato breast tumor xenograft model to identify lipids predominantly localizing to regions of normoxia, hypoxia, and necrosis. This was achieved with MALDI MSI with ion mobility separation. This 2D overlay of optical fluorescence and hematoxylin-and-eosin (H&E)-stained images and MSI was enabled by the use of fiducial markers as described above [1]. Palmitoylcarnitine, stearoylcarnitine, and sphingomyelin(SM)(d18:1/16:0) were enriched in hypoxic regions due to inhibition of β-oxidation during hypoxia, which leads to an accumulation of long-chain acylcarnitines in the cytoplasm of hypoxic cells [11]. Hypoxic breast tumor regions also contained elevated sphingolipid levels, which enables signaling-induced hydrolysis of sphingolipids to produce the messenger lipids ceramide and ceramide-1-phosphate as part of signaling cascades impaired in many cancers [11].
Using the same tumor model, we elucidated the 3D spatial relationship between hypoxic regions and the localization of lipids and proteins by using principal component analysis – linear discriminant analysis (PCA-LDA) on 3D rendered MSI volume data from MDA-MB-231-HRE-tdTomato breast tumor xenografts [4]. The discovered hypoxia-regulated proteins were analyzed with the protein-protein interaction database Reactome (http://www.reactome.org/) to generate a functional protein interaction network, which clusters the discovered proteins into distinct biological pathways [4]. Hypoxia-regulated proteins clustered in several distinct pathways such as such as glucose metabolism, regulation of actin cytoskeleton, protein folding, translation/ribosome, splicesome, the PI3K-Akt signaling pathway, hemoglobin chaperone, protein processing in endoplasmic reticulum, detoxification of reactive oxygen species, aurora B signaling/apoptotic execution phase, the RAS signaling pathway, the FAS signaling pathway/caspase cascade in apoptosis, and telomere stress induced senescence [4]. In parallel, we also identified colocalization of hypoxic regions and various lipid species such as PC(16:0/18:0), PC(16:0/18:1), PC(16:0/18:2), PC(16:1/18:4), PC(18:0/18:1), and PC(18:1/18:1), among others. Details on the biological interpretation of individual proteins and protein networks that were identified in the hypoxic breast tumor microenvironment can be found in our recent publication [4]. Our findings shed light on the biomolecular composition of hypoxic tumor regions, which combined are responsible for a given tumor’s resistance to radiation or chemotherapy [4].
Conclusions
In conclusion, we have demonstrated that MALDI MSI has the ability to identify and map biomolecular markers of aggressive breast cancer, which was shown in hypoxic and high tCho containing tumor regions of breast tumor models in preclinical studies. With the recently improved imaging speed of MSI, it is now possible to perform MSI-based molecular pathology of breast cancer tissue sections within 30 minutes at a spatial resolution of about 30-40 micrometer for an area covering 4 cm2. This now enables us to start to translate the results of our preclinical studies on the effects of hypoxia in the breast tumor microenvironment directly to large patient cohorts. The large-scale metabolite, peptide, lipid, and protein signatures that will be obtained from human tissue samples in future studies should be able to serve as molecular pathology tool for identifying, phenotyping, and differentiating breast cancers for molecular diagnosis and prognosis.
References & Acknowledgements:
References
1. Chughtai K, Jiang L, Greenwood TR, Klinkert I, Amstalden van Hove ER, Heeren RM, Glunde K. Fiducial markers for combined 3-dimensional mass spectrometric and optical tissue imaging. Anal Chem 2012;84:1817-23.
2. Chughtai K, Jiang L, Post H, Winnard PT, Jr., Greenwood TR, Raman V, Bhujwalla ZM, Heeren RM, Glunde K. Mass spectrometric imaging of red fluorescent protein in breast tumor xenografts. J Am Soc Mass Spectrom 2013;24:711-7.
3. Jiang L, Greenwood TR, van Hove ER, Chughtai K, Raman V, Winnard PT, Jr., Heeren RM, Artemov D, Glunde K. Combined MR, fluorescence and histology imaging strategy in a human breast tumor xenograft model. NMR Biomed 2013;26:285-98.
4. Jiang L, Chughtai K, Purvine SO, Bhujwalla ZM, Raman V, Pasa-Tolic L, Heeren RM, Glunde K. MALDI-Mass Spectrometric Imaging Revealing Hypoxia-Driven Lipids and Proteins in a Breast Tumor Model. Anal Chem 2015;87:5947-56.
5. Mascini N, Rizwan R, Jiang L, Cheng M, Glunde K, Heeren RM. Mass spectrometric imaging of pimonidazole as a hypoxia marker in breast tumors. Anal Chem 2016, 88(6):3107-3114.
6. Glunde K, Shah T, Winnard PT, Jr., Raman V, Takagi T, Vesuna F, Artemov D, Bhujwalla ZM. Hypoxia regulates choline kinase expression through hypoxia-inducible factor-1 alpha signaling in a human prostate cancer model. Cancer Res 2008;68:172-80.
7. Jiang L, Greenwood TR, Artemov D, Raman V, Winnard PT, Jr., Heeren RM, Bhujwalla ZM, Glunde K. Localized hypoxia results in spatially heterogeneous metabolic signatures in breast tumor models. Neoplasia 2012;14:732-41.
8. Amstalden van Hove ER, Blackwell TR, Klinkert I, Eijkel GB, Heeren RM, Glunde K. Multimodal mass spectrometric imaging of small molecules reveals distinct spatio-molecular signatures in differentially metastatic breast tumor models. Cancer Res 2010;70:9012-21.
9. Jiang L, Chughtai K, Greenwood T, Bhujwalla ZM, Raman V, Eijkel GB, Heeren RM, Glunde K. Combining magnetic resonance spectroscopic imaging and mass spectrometric imaging reveals protein biomarkers of aggressive breast cancer. Proc 62ND ASMS CONFERENCE ON MASS SPECTROMETRY 2014.
10. Jiang L, Chughtai K, Greenwood TR, Bhujwalla ZM, Raman V, Eijkel GB, Heeren RM, Glunde K. Biomarkers of aggressive breast cancer revealed by comgining magnetic resonance spectroscopic imaging and mass spectrometric imaging. Proc ISMRM 23rd Annual Meeting & Exhibition 2015:3811.
11. Chughtai K, Jiang L, Greenwood TR, Glunde K, Heeren RM. Mass spectrometry images acylcarnitines, phosphatidylcholines, and sphingomyelin in MDA-MB-231 breast tumor models. J Lipid Res 2013;54:333-44.
Acknowledgements
We would like to acknowledge the contributing students and post-doctoral fellows Erika R. Amstalden van Hove, Tiffany R. Greenwood, Kamila Chughtai, Lu Jiang, Menglin Cheng, and Nadine Mascini.
| Description | Y/N | Source |
| Grants | no | |
| Salary | no | |
| Board Member | no | |
| Stock | no | |
| Expenses | no |
IP Royalty: no
| Planning to mention or discuss specific products or technology of the company(ies) listed above: | no |