Morphology Markers 101: Enabling Spatial Multiomics
Remarkable advances in spatial multiomics have been made in the last several years, with companies like NanoString introducing automated instruments capable of imaging, quantifying, and localizing RNA and protein expression with spatial context from a single tissue section. The confluence of technologies that make spatial multiomics possible have roots in traditional molecular biology techniques – monoclonal antibodies, nucleic acid hybridization, fluorescent labeling, microscopy, and histology. These technologies, in concert with current advancements in computation and automation enable researchers to study gene and protein expression in situ in a tissue section.
Spatial biology techniques require a consolidation of multiple data inputs to successfully identify, quantify, and localize RNAs and proteins. The individual techniques that make up spatial biology technology, such as nucleic acid hybridization, are well established; indeed, high throughput, high-plex instruments for expression analysis such as NanoString’s nCounter® Pro Analysis System are well-established for direct detection of hundreds of transcripts, successfully combining nucleic acid hybridization, fluorescent labeling, and digital detection.
The added complexity of analyzing gene expression with spatial context requires additional components, including the ability to recognize key architectural features and cell types that delineate key structures with a tissue section. A convenient solution to this issue is to visualize cellular and tissue compartment boundaries with fluorescent labeling of known proteins and/or transcripts, creating a basic map from which certain structures and cell types can be identified for deeper profiling. Morphology markers are designed to enable such precise localization of gene and protein expression.
What are morphology markers?
Morphology markers are biological molecules identified as reliable probes for histological techniques such as immunohistochemistry (IHC) and in situ hybridization (ISH) that allow for microscopic visualization of tissue structures. Most commonly, fluorescently labeled antibodies and oligonucleotides are used as visualization probes to identify the anatomical structures, cell types, and tissue compartments present in each section of interest. Specific landmarks chosen as targets to probe, or mark, orient the observer to the tissue morphology for selecting and segmenting regions of interest (ROIs) for high-plex spatial multiomics of FFPE and fresh frozen tissue sections such as that done with the GeoMx® Digital Spatial Profiler.
Theoretically, any cell- or location-specific protein or RNA can be used as a morphology marker when targeted by the corresponding antibody (for protein markers) or oligonucleotide (for RNA markers). Morphology markers most often tag general cellular structures (such as the nucleus) or proteins specific to different cell types (such as immune cells). NanoString offers pre-made morphology marker kits for GeoMx DSP that contain probes for typically used markers. Custom morphology markers can also be used.
Why use morphology markers?
A well-designed spatial biology experiment starts with an understanding of the morphology of the tissue of interest. The fundamental morphology of individual tissues has traditionally been derived from conventional hematoxylin & eosin (H&E) stains and/or immunofluorescent (IF) microscopy. For a GeoMx experiment, morphology markers are used to identify an ROI within an FFPE or fresh frozen tissue section from which you want to do RNA or protein expression profiling.
Choosing the most relevant morphology markers is extremely important as these markers play a critical role in developing an ROI profiling strategy that helps answer the kinds of biological questions you are posing and maximizes the amount of information you can get from a GeoMx experiment. Using morphology markers as a visual guide, you can select geometric or irregularly shaped ROIs across the tissue from which you want to do expression profiling, or you can separately profile two or more tissue compartments within a given ROI via segmentation. You can even profile an entire population of a given cell type across the tissue section, even if those cells are not contiguous!
What do morphology markers do?
Target proteins or RNAs are selected as morphology markers based on what structures or cell types they represent in the tissue; these same tissue structures and/or cell types can then be used to select ROIs for expression profiling. For instance, target molecules in NanoString’s off-the-shelf GeoMx Morphology Marker Kits include nucleic acids for staining the nucleus and two cell-specific proteins. These morphology marker kits have been validated for an array of GeoMx assays across mouse and human tissues and can profile tissues such as tumor and its associated immune infiltrate or neuronal populations.
Used in conjunction with an ROI profiling strategy, morphology marker kits can answer a variety of biological questions, including assessing tissue heterogeneity, enriching for certain anatomical features of a tissue, evaluating proximity effects on biological response, or revealing the function of distinct cell populations. Each of these profiling strategies reveal the molecular biology of a tissue by using morphology markers to guide ROI selection and compartmentalization.
What targets make for good morphology markers?
Molecules that reliably identify specific cell types or tissue structures make excellent morphology markers, meaning any molecule with the appropriate specificity will work. However, a target protein will make a more reliable morphology marker if the antibodies against it have already been validated for IF and IHC assays by the antibody provider. Similarly, an RNA probe (for example an RNAScope probe) is more reliable if the sequence is unique to the target.
Good morphology markers identify structures that help orient the observer to the architecture of the tissue. For instance, in the Tumor Morphology Marker Kit, two cell types can be identified in solid tumors – epithelial tissue via a pan-cytokeration (PanCK) antibody, and immune cells via an antibody for CD45. Additionally, the nucleic acid stain SYTO 13 is included for marking the nucleus. The Melanoma Morphology Marker Kit also contains markers for CD45 and the nucleus, along with antibodies for melanoma cell markers S100B and Pmel17 SILV.1Henze G, Dummer R, Joller-Jemelka HI, Böni R, Burg G. Serum S100–a marker for disease monitoring in metastatic melanoma. Dermatology. 1997;194(3):208-12. doi: 10.1159/000246103. PMID: 9187834. 2Bakker AB, Schreurs MW, de Boer AJ, Kawakami Y, Rosenberg SA, Adema GJ, Figdor CG. Melanocyte lineage-specific antigen gp100 is recognized by melanoma-derived tumor-infiltrating lymphocytes. J Exp Med. 1994 Mar 1;179(3):1005-9. doi: 10.1084/jem.179.3.1005. PMID: 8113668; PMCID: PMC2191413. Together, S100B and Pmel17 SILV antibodies reliably mark melanoma cells.
Morphology marker targets in the GeoMx Neuroscience Morphology Marker Kits identify disease-specific proteins along with the nucleus again with the stain SYTO 13. The tissue-specific targets in the Alzheimer’s Morphology Kit are Beta amyloid and Iba1. Beta amyloid protein is a peptide of variable length that forms aggregates thought to be the main component of amyloid plaques in the brains of Alzheimer’s patients.3Chen GF, Xu TH, Yan Y, Zhou YR, Jiang Y, Melcher K, Xu HE. (2017) Amyloid beta: structure, biology and structure-based therapeutic development. Acta Pharmacol Sin. 38(9):1205-1235. PMID: 28713158 4De-Paula VJ1, Radanovic M, Diniz BS, Forlenza OV. (2012) Alzheimer’s disease. Subcell Biochem. 65:329-52. PMID: 23225010 Iba1 is a protein widely used as a marker for microglia and macrophages in the brain.5Ito, D. et al. (1998) Brain Res Mol Brain Res 57, 1-9. 6Kanazawa, H. et al. (2002) J Biol Chem 277, 20026-32. The Parkinson’s Morphology Kit uses the morphology markers Map 2 and alpha synuclein along with the nuclear stain SYTO 13. Map 2 is a protein differentially expressed in neurons and some glia.7Dehmelt, L., & Halpain, S. (2005). The MAP2/Tau family of microtubule-associated proteins. Genome Biology. 8Mohan, R., & John, A. (2015). Microtubule-associated proteins as direct crosslinkers of actin filaments and microtubules. IUBMB Life. Alpha synuclein is a protein expressed primarily in the brain that localizes to synapses and nuclei of neurons and aggregates to form Lewy bodies found in Parkinson’s Disease.9Burre J. (2015). The Synaptic Function of alpha-Synuclein. Journal of Parkinson’s disease. 10Lashuel, H. A., Overk, C. R., Oueslati, A., & Masliah, E. (2013). The many faces of alpha-synuclein: from structure and toxicity to therapeutic target. Nature reviews. Neuroscience. 11Rocha, E. M., De Miranda, B., & Sanders, L. H. (2018). Alpha-synuclein: Pathology, mitochondrial dysfunction and neuroinflammation in Parkinson’s disease. Neurobiology of disease. Overall, the GeoMx Morphology Marker kits use markers that identify structures specific to the tissue of interest to spatially orient gene and protein expression.
How are morphology markers used to identify ROIs?
In the GeoMx workflow, fluorescently labeled morphology markers are used to visualize tissue landmarks and cell types. The filter sets used by the GeoMx fluorescent imaging system consist of four standard channels that cover the excitation and emission wavelengths of many commonly used fluorescent dyes such as FITC, Cy3, Texas Red, and Cy5. Each of these filters can be set to visualize a specific morphology marker fluorescently labeling a specific tissue compartment or cell type. In the case of the Morphology Marker Kits, two filters are used to visualize the markers included in the kit, one filter is open for customization, and the last filter is typically reserved to visualize the nucleus.
Once ROIs are selected based on the visualization of tissue landmarks or cell types, the GeoMx DSP profiles the expression of proteins or RNA one ROI at a time using UV light to release oligonucleotide barcodes specific to each protein or RNA that are then counted downstream using either the nCounter Analysis System or an Illumina NGS system. ROIs can be further segmented based on morphology markers into discrete biological compartments or areas of illumination (AOI). The high-resolution visualization and quantification performed by GeoMx allows for biology-driven profiling of FFPE or fresh frozen tissue sections.
NanoString gives customers the ability to identify relevant morphology markers for their tissue through small trial projects run in the Technology Access Program (TAP). Through the TAP, NanoString has identified a long list of previously characterized morphology markers, either antibodies or RNAScope probes.
Custom morphology markers
Although NanoString supplies off-the-shelf morphology marker kits, some GeoMx projects may require custom markers. Custom morphology markers can be identified by choosing the appropriate protein or RNA targets that are expressed in tissue compartments or cell types of interest. Understanding the biological questions being asked and the ROI selection strategy naturally guides the morphology marker selection process. For example, GeoMx can be used to identify expression changes in the tumor versus tumor microenvironment (TME) in relation to therapeutic response. Use of morphology markers helps answer this question by analyzing tumor regions (PanCk-labeled) for enrichment or depletion of immune cells (CD45-labeled). Selected regions are then segmented into tumor (PanCk+) and the TME (PanCk-, DNA+) for molecular profiling across different clinical response groups.
Further hypotheses about the influence of specific immune cells on therapeutic response can be tested using additional markers, such as CD68 for macrophages, to further segment the tissue (PanCK-, CD68+, DNA+). In this way, a custom morphology marker kit can be developed to further understanding of tissue morphology and key histological structures, giving you the power to unlock spatial expression patterns hidden in a standard bind and grind RNA-Seq experiment.
NanoString’s Community Verified Morphology Markers can be used as a shortcut to select custom GeoMx morphology markers. The GeoMx user community has worked with and characterized a wide range of morphology markers that work on different tissue types. Users are encouraged to share markers used successfully in a GeoMx experiment for inclusion in the list of Community Verified Morphology Markers. Additionally, verified third-party GeoMx morphology markers are available through our partners Advanced Cell Diagnostics, ACD (for RNA morphology targets) or Abcam and Biolegend (for antibody morphology markers). Please refer to the Morphology Marker Guidelines for further details. In addition, morphology marker datasheets can be found on the Morphology Marker webpage that show a GeoMx image taken with a given morphology marker and include details on the Clone ID, concentration used, fluorophore, and antibody host and isotype.
Morphology markers are a key component of spatial biology studies, providing landmarks for orienting gene and protein expression within the architecture of the tissue of interest. Whether you are using NanoString’s off-the-shelf morphology marker kits or custom options, morphology markers play a critical role in designing your GeoMx experiment and ensure that you gain the maximal biological information out of a given DSP experiment.
- 1Henze G, Dummer R, Joller-Jemelka HI, Böni R, Burg G. Serum S100–a marker for disease monitoring in metastatic melanoma. Dermatology. 1997;194(3):208-12. doi: 10.1159/000246103. PMID: 9187834.
- 2Bakker AB, Schreurs MW, de Boer AJ, Kawakami Y, Rosenberg SA, Adema GJ, Figdor CG. Melanocyte lineage-specific antigen gp100 is recognized by melanoma-derived tumor-infiltrating lymphocytes. J Exp Med. 1994 Mar 1;179(3):1005-9. doi: 10.1084/jem.179.3.1005. PMID: 8113668; PMCID: PMC2191413.
- 3Chen GF, Xu TH, Yan Y, Zhou YR, Jiang Y, Melcher K, Xu HE. (2017) Amyloid beta: structure, biology and structure-based therapeutic development. Acta Pharmacol Sin. 38(9):1205-1235. PMID: 28713158
- 4De-Paula VJ1, Radanovic M, Diniz BS, Forlenza OV. (2012) Alzheimer’s disease. Subcell Biochem. 65:329-52. PMID: 23225010