An international project aimed at mapping the cells of the musculoskeletal system has been launched by Oxford University, and we asked Dr Sarah Snelling, an Associate Professor in the Nuffield Department of Orthopaedics Rheumatology and Musculoskeletal Science and one of the project’s coordinators, to walk us through it.
What was the impetus for the musculoskeletal HCA network?
The Human Cell Atlas (HCA) was launched in 2016 as a huge international collaborative project to create comprehensive reference maps of all human cells—the fundamental units of life—as a basis for both understanding human health and diagnosing, monitoring, and treating disease. A truly global initiative, there are now more than 2,000 HCA members, from 75 countries around the world, and the data is openly available for use.
Single cell RNA sequencing, which reveals which genes are expressed in individual cells and in what quantity is a key method. This gives each a cell a unique ‘signature’ – assigning it a particular type or state. Spatial methods are also used, to map the exact location of cells. There are an estimated 37 trillion cells in the human body, and mapping each cell type is an enormous endeavour.
The Human Cell Atlas has different Biological Networks, each focussed on a particular tissue, organ or system, to help with this huge task. The Musculoskeletal Biological Network was borne out of this. It aims to bring together clinicians and basic scientists working within the musculoskeletal field, to ensure that the atlases delivered are informed by the community, and address the challenges and needs of musculoskeletal research.
What are some of the ways a comprehensive atlas of musculoskeletal cells will help with research into relevant diseases and conditions?
One of the exciting things about knowing what every cell of our musculoskeletal system is making, which tissue it is in and whereabouts within the tissue it is found, is that it gives us a blueprint for what a tissue looks like in health. The HCA aims to map healthy tissues across development, childhood and adulthood, providing reference maps of cell types, locations, cellular communication and trajectories of development. These healthy reference maps will therefore be relevant across all musculoskeletal pathologies – of genetic, traumatic, inflammatory, autoimmune, infective or malignant origins that affect humans throughout their lifetime.
If for example we know the cells that should be present within the healthy synovium of an adult, we can compare this to cells taken from the synovium of a patient with psoriatic arthritis, and establish the cell types, or cell products, that could potentially be targeted therapeutically. This provides new avenues for not only identifying mechanisms of disease, and novel therapeutics through functional studies, but also repurposing of therapeutics from other diseases that share similar mechanistic origins. The great thing about having a reference dataset is that researchers working on any pathology affecting a tissue for which a reference dataset exists can rapidly compare health to disease.
Aside from the research applications, are there immediate clinical applications that such an atlas will provide?
The methods used to develop healthy reference maps can, as discussed above, be applied to pathological tissues. Within the cancer field, single cell sequencing can reveal tumour heterogeneity and pave the way for precision-based treatments based on the cells present within a tumour. The same argument can be applied to patients for example with rheumatoid arthritis or psoriatic arthritis – whereby the specific cells and pathways disrupted within an individual’s synovium can inform the best choice of therapeutic – for example whether a patient has significant TNFa-related profiles and will be a responder to anti-TNFa.
Can you describe a bit about the processes involved, in terms of tissue collection, sequencing, and so on?
Tissues from biopsies of healthy patients, donor tissues or cadaver tissues, or from surgeries for conditions that don’t affect the tissue of interest, are donated to HCA researchers for ethically approved studies. These generously donated tissues are vital to build healthy reference datasets. The tissues contain a heterogeneous mixture of cells embedded within their extracellular matrix. Tissue sample are enzymatically digested to yield a suspension of single cells. These individual cells can then be sequenced, with each individual cell (and its contained transcriptome) being assigned an ID or “barcode” allowing each RNA to be traced back to its original cell.
Very similar methods can be employed in a process called single nuclei RNAseq, with the difference lying at the start of the process – instead of isolating individual cells enzymatically from fresh tissue, you can isolate individual nuclei using detergents and mechanical disruption from samples that are generally snap frozen.
What are some of the challenges or obstacles you anticipate in this effort? Where might the difficulties be found?
Access to healthy donated tissues is one of the major challenges, and it is important to study and utilise cadaver, donor, surgical waste tissue and biopsy samples to understand the limitations and strengths of each tissue source.
A further challenge is the complex extracellular matrix (ECM) of musculoskeletal tissues – this means that prolonged enzymatic digestions are often necessary to release individual cells. However this can result in the cell loss due to cell death, or the cells may mount a stress response therefore altering their transcriptome. Single nuclei methods can overcome this, but may still result in cell loss due to freezing or failure to be released from the ECM. This makes it particularly important to benchmark the various methods of tissue disruption and use imaging validation methods to confirm cell signatures – allowing us to understand their biases and reveal ground truth. A consensus on optimal methods for revealing ground truth, and an understanding bias, will make comparisons and integration of tissues analysed by different groups far more valuable and helpful to all, especially if comparing a pathological tissue against a reference dataset generated for the HCA. This will take a real community effort especially due to the diverse nature of musculoskeletal tissues – but is really important in allowing efficient delivery of atlases and maximising patient benefit.
And finally, since the network is designed to bring together clinicians and researchers, how might people get involved to help the project along?
The HCA is an open community and we encourage clinicians and researchers interested in the musculoskeletal field to get involved. There’s information on the HCA Musculoskeletal Biological Network at https://www.humancellatlas.org/areas-of-impact/ and anyone wishing to join should email the network coordinators at [email protected].