Our understanding of how genes influence disease is better than ever, and increasing at an incredible rate. While some diseases like Huntington’s disease are caused by mutations in a single gene, most diseases are influenced by a huge number of variants across the genome. Multiple sclerosis, diabetes, and cardiovascular disease are examples of conditions which are not caused by a single gene problem, but nonetheless are influenced by various genetic risk ‘hotspots’ (also called loci).
A recent study blew this paradigm to bits by suggesting that certain cases of MS might be attributable to mutations in a single gene. Wang et al found that a mutation in the NR1H3 gene was present in 7 people with primary progressive MS from two separate families with a significant family history of the disease. This variant was present in 1 out of a further 2053 people with MS who they studied. They concluded that this mutation might be a cause of primary progressive MS in certain people.
This finding was big news – it had massive implications for both our understanding of MS pathogenesis and for the care of people with MS. What role could this gene have? How could its dysfunction be sufficient to cause such a complex disease? Why would it only cause PPMS and not RRMS? Did this mean that these diseases were even more distinct than we thought? In terms of real-life implications, did this mean that people with lots of affected family members should be routinely offered genetic testing, and even assisted reproduction?
But while revolutionary discoveries do happen, they are rare. The big players in international MS genetics sought to replicate Wang et al’s findings in bigger cohorts to check that their incredible result was genuine. The exome aggregation consortium (ExAC) has a database of pooled genetic data from over 60,000 people. They found that the ostensibly disease-causing variant in NR1H3 detected by Wang et al was present in 21 of their healthy controls (0.031%). This frequency is very similar to the frequency of 0.049% reported by Wang et al in their cohort of people with MS. If this variant does cause MS, you would expect its frequency to be much higher in the MS cohort. The possible explanations for similar frequencies between people with MS and controls are:
- 1. That there is no association between this variant and disease risk, or
- 2. That there is an association between this variant and disease risk but…
o Some people in the control group have MS, or
o The variant has a low penetrance (i.e. it confers an increased risk of disease, but does not always cause disease), or
(IMSGC) has another huge database of genetic information. They looked at data from 32000 people with MS and 36500 controls. The ostensibly disease-causing variant in NR1H3 was present in 31 people in each cohort. There was no association with disease risk or with disease sub-type.
This disagreement highlights a few important issues surrounding the way MS genetics research is both done and reported. It emphasises that reporting of these kinds of discoveries should be circumspect, and should clearly distinguish between fact and hypothesis. Second, it highlights the merits and weakness of two very different approaches to discovering disease-associated gene variants: big data vs small data. Big, international consortia like ExAC and IMSGC have vast swathes of genetic data at their disposal and can work in a data-driven way to discover associations. Family-based case-control studies, on the other hand, try to work the other way up, in a hypothesis-driven way. The problem is that rare variants with a low penetrance are incredibly difficult to detect and validate using either of these approaches. Broadly speaking, small case-control studies are likely to yield false positive results, as people who are closely related will share rare genetic variants that may or may not be associated with disease risk. Using larger, international cohorts is more likely to yield false negatives, as it is impossible to have up-to-date clinical information on tens of thousands of people, and so lots of people in the apparently healthy control group may have various undiagnosed or undocumented medical problems. This makes it tricky to detect small associations between risk variants and disease.
So neither massive genome-wide association studies nor small case-control studies are perfect for elucidating the genetics of MS. In this case, the evidence seems to support a lack of association between NR1H3 mutations and MS. However, this does not mean that the Wang et al paper is dishonest in any way – it merely reflects how difficult it is to show associations between rare gene variants and disease. There is an outside chance that future studies with more multiply-affected families might put this issue to bed. For now though, I don’t think we can say that NR1H3 mutations are a monogenic cause of MS in anyone.
Multiple sclerosis (MS) is an inflammatory disease characterized by myelin loss and neuronal dysfunction. Despite the aggregation observed in some families, pathogenic mutations have remained elusive. In this study, we describe the identification of NR1H3 p.Arg415Gln in seven MS patients from two multi-incident families presenting severe and progressive disease, with an average age at onset of 34 years. Additionally, association analysis of common variants in NR1H3identified rs2279238 conferring a 1.35-fold increased risk of developing progressive MS. The p.Arg415Gln position is highly conserved in orthologs and paralogs, and disrupts NR1H3 heterodimerization and transcriptional activation of target genes. Protein expression analysis revealed that mutant NR1H3 (LXRA) alters gene expression profiles, suggesting a disruption in transcriptional regulation as one of the mechanisms underlying MS pathogenesis. Our study indicates that pharmacological activation of LXRA or its targets may lead to effective treatments for the highly debilitating and currently untreatable progressive phase of MS.