Dissection of Cellular Senescence in Aging by Stereo-seq at Single Cell Resolution

Aging characterized by a progressive deterioration in functions, tissue structural integrity and physiological regulations across multiple organs is the dominant risk factor of cancer, cardiovascular disease, diabetes, and neurodegenerative disorders. Senescence is a cellular response featured by a permanent proliferation arrest via activation of p16INK4a/Rb and p53/p21CIP1 pathways and other phenotypic alterations, including activation of a so-called senescence-associated secretory phenotype (SASP) with a complexed proinflammatory molecules (Di Micco et al., 2021; Gorgoulis et al., 2019). On the one hand, cellular senescence is involved in diverse biological processes, such as embryonic development, tissue homeostasis, and suppression of tumor progression. On the other hand, senescence has also been revealed as a major cause of aging and age-related diseases. Although it is still controversial on the robustness of using a single and universal biomarkers to identify senescent cells due to heterogeneity of cell and tissue types, β-galactosidase (β-gal) positive staining, lack of proliferative marker Ki-67 and p21 expression, when present simultaneously, have been implicated as the common combinatory method to distinguish permanently arrested senescent cells from those pre-senescent, quiescent and differentiated cells. Currently, one major challenge in the field of senescence, especially in aging tissues, is the complexity of geographically uneven distribution of senescent cells and their variable molecular characteristics within and across entropic organ areas. Thus, a potent spatial-transcriptomics method that associates gene expression with cellular spatial-temporal information is essentially needed to dissect the molecular dynamics of senescence cells within their niche and their interactions with the neighboring environment for deepening our understanding and discovering novel therapeutic targets to mitigate the progress of senescence-related aging.

Stereo-seq empowered by MGI DNA nanoball (DNB) sequencing technology is an in-situ capture-based spatial multi-omic solution. It provides unparalleled subcellular resolution at nm scale and large cm field-of-view to realize high-throughput and simultaneous analysis between gene expression, cellular morphology, microenvironments and protein makers on single cell level. Moreover, a comprehensive benchmarking study on sequencing-based spatial transcriptomic methods (sST) using a spectrum of mouse tissues and regions with the well-defined histological architectures was published recently in Nature methods (You et al., 2024). The study concludes that Stereo-seq is technically the top-ranked spatial solution when systematically compared with other sST methods, by means of spatial resolution, capture efficiency, molecular diffusion and gene detection in both all reads and downsampled data (Fig. 1). Especially, the spatial gene expression profiles generated by Stereo-seq are much more refined and superior to other methods due to its the sub-micrometer spot size, which is much smaller than a single cell. Additionally, Stereo-seq exhibited way lower level of blood contamination in comparison with all Visium and DynaSpatial spots, of which majority express notable level of Hba-a1.

Figure 1 (adapted from You et al., 2024). Ranking of the 11 sST methods based on their performance in the specified categories with the highest performing methods positioned at the top. Methods that offer resolution levels below 20 μm have been given higher preference. In the right panel,essential characteristics of the sST methods examined are outlined. The analyzed issues include mouse embryonic eyes, hippocampal regions of the mouse brain and mouse olfactory bulbs

By utilizing Stereo-seq’s unprecedent quality and throughput capacity, a fresh paper, spatial transcriptomic landscape unveils immunoglobin-associated senescence as a hallmark of aging, has been published in Cell (Ma, et al., 2024) identifying senescence-sensitive spots (SSSs) and accumulation of immunoglobulin G (IgG) as features of tissue aging and entropy. Nine tissues from both young (2-month) and aged male (25-month) mice were included in this study to generate a systematic and in-depth spatial transcriptomic atlas as the cornerstone to unveil characteristic drivers for aging-related loss of tissue integrity and organ dysfunction. 1,535,191 high-resolution spots from 103 tissue samples with an average of 1,450 genes per spot were obtained using Stereo-seq along with the confirmation of senescent cells in aged mice by the β-gal staining (Fig. 2). Moreover, 71 distinct spot types matching with known cellular components in different tissues and organs were identified after annotation with marker genes. In addition, a stack of aged-related features, including chronic inflammation revealed by CD45-positive immune cells, enlarged inflammatory areas, increased fibrosis and morphological changes were clearly spotted in the issues from aged mice. Overall, a high-resolution spatial expression platform with stringent controls and validation in aging and senescence has been established in this study for subsequent dissection of underlying molecular mechanisms of aging.

Figure 2 (from Ma et al., 2024). Building a spatial transcriptomic atlas for young and aged male mice across multiple tissues (A) Schematic diagram of tissue sampling, histological analysis, and spatial transcriptomics sequencing in young and old mice. (B) SA-b-Gal analysis of multiple tissues from young (2-month-old) and old (25-month-old) mice (n = 5 per group). Data are presented as the mean ± SEM. Mann-Whitney test. Scale bars, 50 mm. (C) Spatial transcriptomic atlas of young mouse tissues. The central UMAP plot shows the distribution of spots across various mouse tissues, with surrounding diagrams indicating the respective tissues. Refer to Table S1 for the abbreviation keys. Scale bars in inner plots, 200 mm (lymph node), 500 mm (hippocampus and spinal cord), 1 mm (others). Scale bars in the outer plots, 100 mm (lymph node), 200 mm (others). (D) The proportion of spot types in each tissue from young samples. The dendrogram shows the positional similarity between spot types. The bar plot shows the proportion of different spot types in each tissue, with each dot representing one sample.

Furthermore, a novel organizational structure entropy (OSE) method developed from those refined granular tissue spatial pictures by Stereo-seq was proposed to quantify aging-related tissue deterioration in this paper, showing scoring is higher in aged male mice in the examined issues and increases with age (4, 13, 19 months) when compared with younger mice (Fig.3). Sub-cellular spatial-solution offered Stereo-seq also provides a possibility to this study for in-depth analysis of differentially expressed genes (DEGs) along age and across different spatial locations, demonstrating DEGs are closely associated with chronic inflammation, mitochondrial dysfunction, disrupted nutrient and apoptosis. Moreover, both spatial and cell-type-specific expression patterns are revealed in a large proportion of aging-related DEGs.

Figure 3 (from Ma et al., 2024). Exploration of structural deterioration and cellular identity loss in aged male mice. (A) Schematic illustrating the organizational structure entropy (OSE) analysis. (B) Violin plots demonstrating the increased OSE score during aging. Wilcoxon rank-sum test. (C) The spatial mappings showing the OSE score in young (2-month-old) and old (25-month-old) mice tissues. Scale bars, 500 and 100 mm (zoom-in) for hippocampus and lymph node, 1 mm and 200 mm (zoom-in) for the spleen. (D) Heatmaps display cell identity scores in aged tissues and regions with high OSE scores. (E) The spatial mappings showing the decreased cell identity scores during aging. Scale bars, 200 mm. (F) Schematic diagram of tissue sampling and spatial transcriptomics sequencing of 4, 13, 19-month-old mice. (G) Spatial mappings showing spot types of 4-month-old mice. Scale bars, 200 and 100 mm (zoom-in) for hippocampus and lymph node, 1 mm and 200 mm (zoom-in) for liver and spleen. (H) Boxplots showing the change of OSE during aging, with each point indicating a superpixel unit. Wilcoxon rank-sum test. (I) Heatmap showing the decreased cell identity scores during aging. (J) Spatial mappings showing the decreased cell identity scores during aging. Scale bars, 200 mm.

To dissect senescent cells and their interactions with the neighboring environment, the study proposed the concept of senescence sensitive spot (SSS). A careful characterization of SSS cellular and molecular signatures unveiled that immunoglobulin-producing cells (Ighigh) in various aged tissues preferentially locate in the vicinity of SSS, thus creating a pro-inflammatory niche. In addition, OSE scoring declines with increasing distance from those SSS spots. Intriguingly, the study discovered IgG, a key immune factor increases with age in both male and female mice tissues at different life stages from young to aged, indicating IgG accumulation during aging is a sex-independent universal fact. More importantly, the same findings have been also demonstrated in human issues, including lymph node, liver, spleen, and brain. Consequently, a consistent pattern of intra-tissue IgG elevation observed across multiple life stages, sexes, and species, supporting IgG could be an evolutionarily conserved marker in human and mammals aging. These findings are further reinforced by subsequent in vivo and in vitro physiological studies, showing loss-of-function in terms of clearance of antibody-producing cells, mediated FcRn ablation, or neutralizing IgG’s pro-senescent effects alleviates aging-related tissue deformation and senescence and could even partially revert the aging phenotype in mice. Conversely, in vivo gain-of-function study by long-term IgG administration showed accelerated aging-related defects and senescence in mice, indicating IgG acts as a senescence-promoting factor.

To conclude, a high-precision spatiotemporal profile achieved by Stereo-seq followed by MGI DNB sequencing on DNBSEQ-T7 delivers unsurpassed spatiotemporal transcriptomic profiles of murine aging, covering various tissues across different chronological stages and forms a concrete foundation forsimilar studies in the future. A thorough spatial transcriptomic landscape in female mammals is the next focus to complete the picture of aging in mammals and to potentially unmask sex-dependent aging mechanisms. Further finetuning on the mechanisms of IgG accumulation in cellular aging could lead to the development of immunoglobulin-targeted therapies to treat age-related diseases.

References:

Systematic comparison of sequencing-based spatial transcriptomic methods. You et al., Nature Methods, Sep; 21(9):1743-1754, (2024).

Spatial transcriptomic landscape unveils immunoglobin-associated senescence as a hallmark of aging. Ma et al., Cell, Nov 2:S0092-8674(24)01201-7, (2024).

Cellular senescence in ageing: from mechanisms to therapeutic opportunities. Di Micco et al., Nat Rev Mol Cell Biol, Feb;22(2):75-95, (2021).

Cellular Senescence: Defining a Path Forward. Gorgoulis et al., Cell, Oct 31;179(4):813-827, (2019).