Commentary Nearly 400 types of skeletal malformations, including those that involve the spine, have been characterized by molecular and cellular analyses1. Identification of genetic components that control the integrity of intervertebral discs—and that predispose to spinal degeneration in aging patients when mutated—represents a challenging goal. Spinal conditions that occur later in life may emerge from faulty musculoskeletal tissue repair after accidental injury and the influences of comorbidities. The collective stochastic and often undocumented events in the life of a patient can easily complicate attempts at defining genetic predispositions. To manage the statistical noise that comes from random mutations within the human population (“nature”) and the randomness of life’s events (“nurture”), it is necessary to examine massive genetic data sets with many thousands of patients. Technological advances that allow cost-effective DNA sequencing have made it feasible to compare large patient populations matched on the basis of specific codes for medical conditions. Consequently, several studies have made impressive attempts at defining genetic variations linked to a range of surgically relevant musculoskeletal complications, including degenerative rotator cuff disease2, arthroplasty3, adhesive capsulitis of the shoulder4, and end-stage knee osteoarthritis5. The current paper by Bovonratwet et al. at the Hospital for Special Surgery in New York provides the latest installment in a growing set of orthopaedic studies in this journal that investigate the relationship between genetic variations and the risk of surgery for spinal conditions. Similar to prior published efforts by Yanik et al.2, Brüggemann et al.3, and Kulm et al.4,5, the study by Bovonratwet et al. leverages the power of the UK Biobank, which has genetic data for about 400,000 patients, of whom a total of about 20,000 (∼5%) had 1 of 4 spinal conditions (i.e., lumbar spondylolisthesis, spinal stenosis, degenerative disc disease, and pseudarthrosis after spinal fusion). Results from this large data set (“training population”) were then tested using the FinnGen database (“test population”) for validation. The analysis yielded multiple different genetic variants across 7 chromosomes. On average, the authors discovered 2 distinct loci per spinal condition. These findings are consistent with the genetic complexities of spinal conditions and suggest that the diseases are polygenic, as expected. Several nucleotide variants associated with 4 common spinal conditions were located in loci encoding anonymous genes that have not been experimentally explored. Degenerative disc disease was associated with a locus containing 2 genes that affect chondrogenesis: CHST3 (for carbohydrate sulfotransferase 3) promotes sulfation of chondroitin, and SMAD3 is the inducible target of chondrogenic transforming growth factor (TGF)-β signaling. Remarkably, 1 chromosomal region significantly associated with both lumbar spondylolisthesis and spinal stenosis contained overlapping sets of genes (i.e., GFPT1 and NFU1) that encode proteins involved in cell metabolism. These genes represent attractive targets for studies on genetic causality for these 2 spinal conditions. While it is tempting to speculate about potential mechanistic implications, the genetic biomarkers for spinal conditions mapped to large regions containing several genes. Whether these genes are even expressed in cell types and tissues relevant to homeostasis and repair of spinal tissues remains to be determined. Lack of similarity among studies indicates that endogenous tissue-specific repair processes may be predominant in the genetic landscape. One major strength of this study is the control for demographics and comorbidities. Yet, even with access to nearly 400,000 people who form a fairly homogeneous study cohort (i.e., White elderly people in Britain), this population size is not nearly large enough to identify the many genes that undoubtedly contribute to spinal conditions. The observation that only a few genetic variants were replicated in the Finnish cohort may further reflect the polygenicity of disease factors and genetic differences of Finnish patients. In closing, just as our emotions can run up or down our spine depending on whether we are in a positive or negative mood, genetic variation appears to affect spinal diseases. Some DNA mutations may literally be running down our spine, by destroying either disc integrity or our capacity for repair after injury. The positive news is that recognition of genetic predispositions may enhance consideration of behavioral modifications or guide surgical decisions. The paper by Bovonratwet et al. will inspire future studies, while bringing us 1 step closer to that goal.
Read full abstract