Beyond the clock: How tissues reveal the true story of ageing

-

 

Ageing, a process we all experience, is often perceived as a uniform decline dictated by the passage of time. However, groundbreaking research is challenging this notion, revealing a much more complex picture where different parts of our bodies age at varying rates. This study delves into the intricate details of tissue ageing, utilizing a vast collection of over 25,000 histopathological images from 40 distinct tissue types, sourced from the Genotype-Tissue Expression Project (GTEx). The scientists harnessed the power of deep learning, a sophisticated form of artificial intelligence, to meticulously analyze these images and uncover age-related morphological changes. This approach offers a unique perspective, moving beyond traditional molecular and cellular views of ageing to focus on the structural and architectural changes in tissues.

Tissue clocks: Biological age predictors

One of the most significant outcomes of this research was the development of what the scientists call "tissue clocks". These are essentially predictive models based on the visual features extracted from tissue images, which estimate the biological age of a tissue. Unlike chronological age, which is simply how long we have lived, biological age reflects the physiological state of a tissue and can vary from an individual's actual age. For instance, certain tissues might appear "older" than expected, showing signs of accelerated ageing, while others might appear "younger", indicating slower ageing. The average prediction error of these tissue clocks was remarkably low, at just 4.9 years, highlighting their accuracy. Further, the study showed a strong correlation between the biological age predicted by these tissue clocks and other hallmarks of ageing, such as telomere shortening. Telomeres, the protective caps on the ends of our DNA, naturally shorten as we age, and the tissue clocks showed a more robust association with telomere length than chronological age alone, emphasizing the biological relevance of these clocks.

Variable ageing rates across the body

A pivotal discovery of this research is that the ageing process is not uniform across the body; different organs and tissues age at different rates. Some organs, particularly the lungs and glandular organs such as the ovaries and thyroid, showed signs of more rapid ageing in early adulthood, between the ages of 20 and 40. Conversely, organs like the colon and aorta displayed a bimodal pattern of ageing, with periods of accelerated change in the third decade of life and again between 45 and 55 years of age. The uterus presented a particularly unique trajectory, with a noticeable inflection point around the time of menopause, indicating a significant shift in the rate of ageing during this period. This heterogeneity underscores the complex nature of the ageing process and challenges the notion of a single, unified ageing clock for the entire organism.

The influence of lifestyle and health

The study also explored how various demographic, clinical, and lifestyle factors can influence tissue-specific ageing. The researchers found that several health conditions are linked to accelerated ageing in specific tissues. For example, renal failure was strongly associated with accelerated ageing across multiple tissues, not just the kidneys themselves. Similarly, chronic respiratory disease was found to be strongly linked to accelerated ageing in lung tissue. Furthermore, the study identified associations between type II diabetes and pancreas ageing, heart disease and adipose tissue ageing, and the effects of steroid use on the spleen. These findings highlight that our health, lifestyle, and past medical history play a significant role in how our tissues age.

Predicting tissue ageing from blood

Seeking a less invasive method to assess tissue-specific ageing, the researchers innovatively linked tissue ageing data derived from images with gene expression data from blood samples. This led to the development of a novel approach to predict tissue-specific ageing from a simple blood test. This blood-based method was tested on various cohorts, including individuals with Crohn’s disease and rheumatoid arthritis. The results demonstrated that the blood-based predictions could identify which tissues were ageing more rapidly in those individuals affected by these conditions. This has significant implications, potentially allowing for the monitoring of tissue-specific ageing rates through routine, minimally invasive blood tests.

Implications of the findings

This study offers a new lens through which to view ageing, emphasizing the importance of tissue-level analysis. It establishes that ageing is a complex, non-uniform process that is influenced by both internal biological processes and external factors. The study paves the way for new approaches to monitor and even intervene in the ageing process. By using sophisticated deep learning techniques, the researchers have been able to develop highly informative tissue clocks and explore non-invasive blood based tests for monitoring tissue-specific aging. The scientists do acknowledge limitations of their study, including the sex imbalance within the GTEx cohort and the use of post-mortem samples which may impact results. However, despite these limitations, this research offers a new and comprehensive understanding of the human ageing process and highlights the potential for new avenues of research in the future.

 

Additional information: Tissue clocks derived from histological signatures of biological aging enable tissue-specific aging predictions from blood. BioRxiv (2024) https://doi.org/10.1101/2024.11.14.618081

Journal information: https://www.biorxiv.org/


Location: Oxford, UK

Comments

Post a Comment

Popular posts from this blog

Unlocking insights: simplifying single-cell data with Strand’s scRNA portal

-
Image
Image credit:   https://scrna-curation.mystrand.org/ Simplifying single-cell data exploration with Strand’s scRNA portal In the rapidly advancing world of single-cell RNA sequencing (scRNA-seq), navigating complex datasets can be a major challenge. Strand’s scRNA Portal is designed to transform this challenge into an opportunity by harmonizing publicly available datasets, making disease-specific omics data accessible, actionable, and ready for discovery. Why strand’s scRNA portal was built 1) Focus on unmet needs in key diseases Starting with datasets with diseases with significant unmet clinical need: Inflammatory bowel disease :  Covering Ulcerative Colitis and Crohn’s Disease. Neurodegenerative disorders :  Including Alzheimer’s Disease, Parkinson’s Disease and Frontotemporal Dementia. 2)  The rise of scRNA-seq in aiding precision medicine:  Single-cell RNA sequencing provides unprecedented insights into cellular heterogeneity, which is pivotal for identifyi...
Read more

Understanding Your Immune System: A Lifespan Journey

-
Image
Image credit:   https://openai.com/index/dall-e/   Have you ever wondered how your immune system changes as you get older? It’s not just about catching colds, our immune system is always working hard to protect us and it’s constantly evolving throughout our lives. A new study has looked at how our immune system changes from birth right through to old age, providing some fascinating insights. What the study did : Scientists looked at blood samples from a large group of healthy people, ranging from newborns to those over 90 years old. They used advanced techniques to study the different types of immune cells in these samples. They looked at how these cells behave and change over time. This is called a "single-cell atlas," because they looked at each cell individually. Key findings Immune cells change over time: The study identified 25 different types of immune cells in the blood. The number of each type of cell changes throughout life. Some become more commo...
Read more

Cracking the code of itchy skin: A new way to diagnose and treat skin problems

-
Image
a : The ridgeline plots show the activity scores of different immune response types (Th1, Th2, Th17, type I interferon, neutrophil-related, macrophage-related, and eosinophil-related responses) in skin samples from different inflammatory skin diseases (Psoriasis, Atopic Dermatitis, Lichen Planus, Cutaneous Lupus Erythematosus, Neutrophilic Dermatosis, and Wells Syndrome). The solid lines mark the point where each response type is considered active. b : Box plots compare the immune response scores in skin samples across all these diseases for the "sentinel" (reference) biopsies. Each dot is one sample. The number of samples for each condition is listed: Psoriasis (25), Atopic Dermatitis (16), etc., along with healthy controls (8). c : Box plots show the immune response scores for "test" samples (a different set of biopsies). Sample numbers for each disease are also listed, like Psoriasis (8) and Atopic Dermatitis (15). b–c : In the box plots, the middle line shows th...
Read more