$$\rightleftharpoonup{xx}$$
$$\longleftharp{xx}$$,
$$\longrightharp{xx}$$,
Over the past decade, next-generation sequencing and high-resolution imaging have converged to advance spatial transcriptomics, a method of whole transcriptome quantification while preserving each transcript's precise location in tissue sections1,2. When paired with the rapidly maturing single-cell RNA-sequencing (scRNA-seq) field, spatial approaches provide an efficient route to link gene signatures with specific cell types, states, and niches within a tissue or organ, greatly enriching our understanding of disease mechanisms3,4,5,6,7. Multiple commercial platforms are being developed, such as 10X Visium, NanoString GeoMx and CosMx, Slide-seq, and DBiT-seq, each supported by protocols that are optimized for soft, highly cellular tissues such as tumors and brain tissue. The field is experiencing explosive growth, with PubMed indexing approximately 150 publications in 2023, nearly 300 in 2024, and on pace to surpass 500 in 2025. Despite this surge, vascular biology remains underrepresented compared with cancer, neuroscience, and immunology.
Hard or mineralized tissues, such as bone, cartilage, teeth, and plaque-laden, calcified arteries or veins, have proved far less amenable. The technical hurdles are multifaceted. First, the tissues with dense mineral matrices demand chelation or acid-based removal, aggressive treatments that may fragment RNA and distort tissue morphology. Second, diseased tissues, either injured or fibrotic, often contain few living cells embedded in copious extracellular matrix (ECM), reducing transcript yield and complicating data normalization. In addition, vascular tissues possess a thin, concentric architecture (intima, media, adventitia) wrapped around a hollow lumen. Limited wall thickness, especially the single-cell-layer intima, means capture areas may hold scant tissues, and sections can tear or detach during multi-hour staining and hybridization steps. As a result, spatial transcriptomic studies of vascular tissue remain scarce, even though cardiovascular diseases consistently rank among the leading causes of death and serious morbidity worldwide. Protocol papers that do exist overwhelmingly emphasize data processing, visualization, or platforms other than NanoString GeoMx8,9,10,11,12,13. Among these papers, few describe bench-level methods for processing diseased tibial arteries, a clinically relevant vessel in peripheral artery disease (PAD), especially for diabetic patients who make up an increasing proportion of the affected population.
To address this gap, we present a comprehensive workflow for calcified human tibial arteries obtained from PAD patients who underwent amputation. The protocol spans: (1) tissue trimming and dissection immediately after operating-room retrieval; (2) proper tissue fixation and decalcification that maintain integrity of tissue morphology and RNA; (3) construction of cost-saving tissue microarrays (TMAs) to minimize batch effects; (4) RNAscope in situ hybridization and histological calcium staining for quality control; and (5) Region of interest (ROI) selection strategy specifically designed for calcified tibial arteries on the GeoMx Digital Spatial Profiler (DSP) platform. These steps include all procedures prior to sample collection on the DSP instrument, library preparation, sequencing, and data analysis.
Although demonstrated on the referenced DSP platform, the principles can guide tissue preparation for 10X Visium or other platforms, and the methodology is readily adaptable to additional vascular beds or animal models. This protocol may be adapted for biobank and post-mortem tissues, but its success depends critically on tissue preservation, fixation, and storage conditions. Tissues should be fixed immediately after harvest to minimize ischemic time (ideally within 1 h). Standard 10% neutral-buffered formalin (NBF) is recommended for at least 16 h at room temperature (RT). Tissues should be embedded promptly after fixation and should not be stored in ethanol for more than 3 days. Formalin-fixed paraffin-embedded (FFPE) blocks processed and stored under recommended conditions remain suitable for spatial profiling for approximately 3 years, although RNA quality declines with block age. Accordingly, RNA quality assessment is recommended for low-cellularity samples, blocks that are stored over 6 months, biobank specimens, and post-mortem tissues. The minimum acceptable RNA quality depends on the assay selected, ROI selection, desired resolution, and core facility recommendations. In practice, approximately 50-100 nuclei per segment are commonly recommended for robust results. The DSP platform supports both FFPE and fresh frozen tissues. In this protocol, we focus on FFPE tissues.
While this study focuses on spatial transcriptomics, it is readily integrated with scRNA-seq. Using published scRNA-seq datasets and computational spatial deconvolution, cell-type proportions can be estimated for each spatial spot, and scRNA-seq-defined cell states can be projected onto tissue architecture. By maintaining both structure and RNA integrity, our protocol enables downstream single-cell or single-molecule validation, maximizing biological insight from precious vascular specimens and wider adoption of spatial transcriptomics in vascular biology research.