19.03.13 One gene, many diseases and one disease, many genes

Question 1

What is pleiotropy?

Answer 1

When one gene influences two or more seemingly unrelated phenotypic traits

Question 2

One gene multiple disorders example 1 - RET (10q11.21)

Answer 2

• GoF variants cause multiple endocrine neoplasia type 2 (MEN2). AD disorder and three subtypes.
• MEN2A - phaeochromocytoma (adrenal medulla tumour), occurs in early adulthood. (moderate activating variants)
• MEN2B - Ganglioneuromas, occurs in early childhood (Aggressive variants - most severe MEN2 disorder)
• Familial medullary thyroid carcinoma (FMTC), occurs in middle age (mild activating variants)
• LoF variants cause Hirschsprung disease. AD disorder. Get complete absence of neuronal ganglion cells from intestinal tract. Leads to enlarged bowel and constipation in neonates.
• RET variants account for 40% of Hirschsprung disease (therefore this Hirschsprung also shows locus heterogeneity, as variants in other genes, e.g. EDNRB, also linked to it)

Question 3

One gene multiple disorders example 2 - PMP22 (17p12)

Answer 3

• PMP22 is a glycoprotein expressed by myelinating Schwann cells in the PNS
• PMP22 has role in initiation of myeline spirals, regulation of growth of schwann cells and controls thickness of myeline sheaths
• Recurrent 1.5Mb PMP22 duplication - increased dosage - causes CMT1A
• Recurrent 1.5Mb PMP22 deletion - haploinsufficiency - causes HNPP
• PMP22 point mutations - LoF (causes HNPP) and GoF (causes CMT1A)
• CMT1A - Peripheral demyelinating neuropathy - loss of myelin sheath of the PNS and subsequent degeneration of nerve fibres - leads to progressive muscle weakness.
• HNPP - Hereditary neuropathy with liability to pressure palsies. Peripheral demyelinating neuropathy - get segmental demyelination with ‘sausage-like’ myelin thickening. Results in numbness, muscle weakness and atrophy.

Question 4

What is the gene dosage model associated with PMP22?

Answer 4

• Overexpression of PMP22 leads to CMT1A, and haploinsufficiency leads to HNPP.
• CMT1A - increased levels of PMP22 - over production of PMP22 protein (meaning protein can’t be processed correctly) - Leads to reduced amounts of functional PMP22 protein - Impairs formation of myelin - Affects schwann cell growth and differentiation and increases myeline thickness. Model is supported by findings of increased PMP22 protein and mRNA in nerve biopsies from CMT1A patients.
• HNPP - Het del - reduced levels of PMP22 protein - so less myelin produced - increased susceptibility of nerves to mechanical forces (pressures).

Question 5

One gene multiple disorders example 3 - Androgen receptor gene (Xq12)

Answer 5

• Mutations in the AR gene can cause two disorders: Androgen insensitivity syndrome (AIS) and spinal and bulbar muscular atrophy (SBMA/Kennedy’s disease)
• AIS - AR essential for normal primary and secondary male development. Can get complete (CAIS), partial (PAIS) and mild (MAIS).
• AR variants that severely impair the amount, structure or function of AR protein cause CAIS. Males (XY) are born appearing female due to lack of masculinization of external genitalia. Normally not detected until puberty, where a lack of uterus, cervix is detected. Patients are sterile.
• PAIS - various degrees, and MAIS can have male genitalia but small
• Single base variants are most common (>90%) - most are missense. Most confined to DNA and steroid binding domains and splice sites
• SBMA - expansion of trinucleotide repeat (CAG)n in the first exon of the AR gene causes rare motor neuron disease
• onset at 30-50 years, proximal muscle weakness, muscle cramps, atrophy and twitching, progressive involvement of bulbar muscles. Also get signs of androgen insensitivity - reduced fertility, testicular atrophy and gynecomastia.
• HET females mostly normal
• 10-36 repeats = normal, 38-72 repeats = affected. The higher the number of repeats, the more severe the symptoms.
• SBMA molecular mechanism - get nuclear accumulation of polyglutamine-expanded AR in motor neurons - leads to neuronal dysfunction and cell death - also causes formation of toxic oligomers which leads to further neurodegeneration.

Question 6

What is genetic heterogeneity?

Answer 6

When a single disorder can be caused by mutations in different chromosome loci.

• Could be because disruption of a pathway causes phenotype, and multiple genes in pathway could be affected.
• Lots of examples: HCM/DCM, HNPCC, CAH, Noonans and CMT
• NGS is very helpful for these referrals, as multiple genes can be investigated in one panel at the same time.

Question 7

One disorder multiple genes example 1 - Hypertrophic cardiomyopathy (HCM)

Answer 7

• Unexplained left ventricular hypertrophy (and absence of other cardiac disease)
• Can range from asymptomatic to shortness of breath, chest pain, palpitations, progressive heart failure and sudden cardiac death.
• Caused by genes encoding the cardiac sarcomere components (50-60% of cases with FHx, and 20-30% with no FHx).
• Most common genes are MYH7, MYBPC3 and TNNT2
• Mechanism - dominant negative - mutant protein incorporated into sarcomere, but can’t interact properly with WT and inhibits function
• Mechanism - GoF - some variants cause increased contractile forces, causing impaired relaxation
• Mechanism - haploinsufficiency - NMD causes absence of certain proteins.

Question 8

One disorder multiple genes example 2 - Hereditary spastic paraplegia (HSP)

Answer 8

• Causes some muscles to be continuously contracted - get stiffness, tightness which interferes with gait, movement and speech
• Can be AD, AR, X-linked - >52 genes involved
• Most common are AD spastin (SPG4; 40% of cases), second is AD atlastin (ALT1; 20% of cases) - often early onset.
• Most genes have roles in intracellular membrane and protein trafficking

Question 9

Polygenic risk scores

Answer 9

• The likelihood of a single variant causing a specific condition can be expressed by a percentage (i.e. the penetrance of a gene).
• PRSs aim to quantify the cumulative effects of a number of genes, which may individually only have a small effect on susceptibility. It is a score reflecting the sum of all known risk alleles, weighted by how risky each variant was known to be.
• PRS can be used to predict a persons likelihood of displaying any trait with a genetic component.
• Recent research has focused on 5 common conditions – coronary artery disease, atrial fibrillation, type 2 diabetes, inflammatory bowel disease and breast cancer
• Genome-wide data allows access to millions of common genetic variations associated with the above conditions. A risk score for each person is then calculated with the aim to inform clinical management

Question 10

Benefits of PRSs

Answer 10

• Help to personalise preventative measures
• Accessing PRSs could be helpful in targeting people at higher-risk of conditions where increased surveillance and preventative treatment or surgery is available. E.g, inherited breast cancer where extra screening and risk reducing surgery is available this could improve outcomes.
• Good where there is no info regarding family history (i.e. due to adoption or donor-conceived).
• The PRSs may help to match drugs in clinical trials to individuals who are most likely to benefit from them
• Even for conditions which lack proven preventative measures, there may still be some benefit as there will be more awareness and more time to prepare.
• In time, the PRSs could be combined with other risk factors (e.g. environmental and lifestyle factors) to improve risk prediction and help define clinical action threshold.

Question 11

Challenges of PRS

Answer 11

• They are not diagnostic. A high risk score doesn’t mean that a person will definitely develop a condition, and a low score does not mean that they won’t.
• The results will need to be carefully communicated to patients do minimise the risk of confusion. We need to better understand how people react to receiving such scores, and then monitor any changes in health-related behaviours afterwards.
• Testing requires consent for genetic testing from patient (which may be difficult if the individual is currently not unwell)
• How accurate are the scores? To date, the majority are calculated from European DNA sequences and are therefore less accurate for other populations. There is a current drive to analyse multi-ethnic data to resolve this issue.