Epigenetics in Musculoskeletal Disease

What epigenetics is. Epigenetics is heritable, potentially reversible change in gene expression that occurs without any change in the underlying DNA sequence. It is the interface through which environment, ageing and mechanical load reach the behaviour of musculoskeletal cells, and the reversibility is what makes it a therapeutic target.

The three mechanisms. Gene expression is regulated epigenetically in three ways, and all three are implicated in osteoarthritis, osteoporosis and tumours. DNA methylation at promoter CpG islands generally silences a gene, histone modification opens or closes chromatin, and non-coding RNAs (microRNAs and lncRNAs) act after transcription.

View larger

Osteoarthritis. Epigenetic dysregulation drives the abnormal chondrocyte phenotype and mesenchymal stem cell (MSC) senescence: aged MSCs differentiate poorly and release pro-inflammatory cytokines. MSC-derived exosomes, carrying DNA, RNA, proteins and lipids including regulatory microRNAs, are being explored as a therapy to promote cartilage repair.

Osteoporosis and tumours. Epigenetic dysregulation also contributes to osteoporosis through the osteoblast-osteoclast balance, and to bone and soft-tissue tumours through silencing of tumour-suppressor genes and characteristic methylation and chromatin changes. Across these and other degenerative conditions, epigenetics is a unifying theme of musculoskeletal disease.

View larger

Acetylation opens, deacetylation closes. Histone acetyltransferases (HATs) add acetyl groups to lysine tails, neutralising their positive charge and loosening the histone-DNA grip, so chromatin opens (euchromatin) and transcription is activated. Histone deacetylases (HDACs) remove them, the chromatin condenses and the gene is repressed, which is why HDAC inhibitors can re-express silenced genes.

Methylation depends on position. Unlike acetylation, histone methylation can activate or repress. H3K4me3, found at active promoters, is activating, whereas H3K9me3 and H3K27me3, the heterochromatin and Polycomb marks, are repressive.

Writers, erasers and readers. This machinery is what makes the code reversible and druggable.

• Writers add marks: HATs, histone methyltransferases, DNMTs
• Erasers remove them: HDACs, histone demethylases, TET enzymes
• Readers interpret them: bromodomain proteins read acetyl, chromodomain proteins read methyl

View larger

View larger

Laying down the mark. DNMTs add a methyl group to the 5-carbon of cytosine in CpG dinucleotides, forming 5-methylcytosine. DNMT1 is the maintenance enzyme, copying the pattern to the daughter strand at replication, and DNMT3A/3B are the de novo enzymes. TET enzymes oxidise 5-methylcytosine towards demethylation.

How methylation silences. Methyl-CpG directly blocks transcription-factor binding. It also recruits methyl-CpG-binding proteins (MeCP2/MBD), which pull in HDACs and co-repressors to condense the chromatin, and this is where the two mechanisms meet.

The licensed drugs. Because the marks are reversible, epi-drugs exist:

• DNMT inhibitors (hypomethylating agents): azacitidine and decitabine, for myelodysplastic syndrome and AML
• HDAC inhibitors: vorinostat and romidepsin, for cutaneous T-cell lymphoma
• EZH2 inhibitor: tazemetostat, for advanced INI1-deficient epithelioid sarcoma

The orthopaedic one. Over 90% of epithelioid sarcomas, a soft-tissue sarcoma seen in orthopaedic oncology, lose INI1/SMARCB1, a subunit of the SWI/SNF chromatin-remodelling complex. That loss creates an oncogenic dependence on EZH2, the histone methyltransferase that writes the repressive H3K27me3 mark, and tazemetostat exploits exactly that dependence. In the pivotal phase 2 cohort of 62 patients the objective response rate was 15%, with median progression-free survival 5.5 months and median overall survival 19.0 months, and little toxicity.

View larger

Epigenetic clocks. These estimate biological age from DNA-methylation patterns, and clocks such as Horvath's, GrimAge and DunedinPACE are associated with degenerative musculoskeletal disease. The concrete figures come from a systematic review:

• OA cartilage is 3.7 years older than chronological age on Horvath's clock
• DunedinPACE tracks chronic low-back-pain severity and impairment at r = 0.39-0.45
• GrimAge correlates most strongly with chronic pain (rho = 0.47)

Reading the direction. All fourteen underlying studies are observational, so none shows that epigenetic ageing causes degenerative disease rather than tracking its consequences. GrimAge was found to mediate a socioeconomic effect on pain, so part of what these clocks measure is the biology of disadvantage. An r of 0.4 explains about a sixth of the variance: real, reproducible, and nowhere near a test.

Biomarkers. Epigenetic marks and epigenetic age acceleration are potential diagnostic and prognostic tools. They are not ready: no clock has longitudinal validation, the correlations are moderate, and disease-specific algorithms are still needed.

Therapeutic potential. Reversibility makes epigenetic changes candidate drug targets (agents acting on methylation or HDACs) and underpins regenerative strategies (MSC and exosome therapy). Those regenerative therapies are preclinical, and no randomised trial has shown they change human osteoarthritis, which matters because patients are being sold them.

Where the field stands. The core concepts are secure, and the field is advancing rapidly but still emerging. For the orthopaedic surgeon the relevance is a conceptual understanding of how environment and ageing translate into musculoskeletal disease rather than current routine practice, with one exception.

Related pages: Cell and Molecular Biology in Orthopaedics and Genetics and Inheritance in Orthopaedics hold the layer beneath this one - the distinction that defines the topic is that epigenetics changes expression without changing the sequence, so read them together. Epithelioid Sarcoma is the single place where this science is licensed treatment: INI1/SMARCB1 loss creating EZH2 dependence, drugged by tazemetostat; Synovial Sarcoma is the neighbouring translocation sarcoma in which SS18-SSX also acts by hijacking chromatin remodelling, and Osteosarcoma with Chondrosarcoma are where tumour-suppressor silencing by promoter methylation is best described. On the degenerative side, Osteoarthritis Pathophysiology, Chondrocyte Metabolism and Articular Cartilage Structure are where the methylation and non-coding-RNA changes described here are expressed as a phenotype, and Intervertebral Disc Biology is the tissue in which the senescence data are strongest. Osteoporosis and Bone Remodeling carry the RANKL-OPG axis that epigenetic regulation acts on. Stem Cells and MSCs in Orthopaedics and Gene Therapy and Tissue Engineering cover the exosome and regenerative strategies discussed here - and the reason they remain investigational.