Lecture 29 - Mitochondrial Diseases 2

Question 1

Where are the complexes in the inner mitochondrial membrane encoded?

Answer 1

Both nuclear DNA and mitochondrial DNA

Question 2

How prevalent are OXPHOS disorders?

Answer 2

1:5000 live births

(also see the figure 1:6500, but this is probably too conservative)

However, there are some ethnicities which have much greater rates: e.g. NSM Lebanese

Question 3

Describe the features of the mtDNA

Answer 3

• Double stranded
• DNA
• Circular
• 37 genes

Question 4

List the unique features of mtDNA

Answer 4

• Maternally inherited
• Multiple copies (within each cell, as opposed to one from mum, one from dad with nuclear DNA)
• High mutation rate
• Heteroplasmy
• Threshold effect
• mtDNA bottleneck
• Tissue-specific segregation / selection

Question 5

Describe mtDNA heteroplasmy, the bottleneck effect and the threshold effect

What is the clinical implication of this?

Answer 5

1. Heteroplasmy: different versions of the DNA within the same cell (due to high mutation rate)
2. Bottleneck effect: small number of mother’s mitochondria go into each oocyte; randomly selected
3. Threshold effect: for disease to be seen, the mutant load of mtDNA must be above a given threshold

Example:
Mother:
• 20% mutant mtDNA
• 80% normal mtDNA

In egg development, there will be different eggs with different amounts of the mutant mtDNA

Certain oocytes may have a greater or lesser proportion of the mutant mtDNA

Possible to get zygotes with:
 • 80% mutant load → severe disease
 • 60% mutant load → mild disease
 • 20% mutant load → no disease
etc

In summary:
Due to heteroplasmy, the bottleneck effect and the threshold effect, it is possible to have very different outcomes

Question 6

What is the most common OXPHOS disease in childhood?
Give some features of the disease
What is the main pathological feature?

Answer 6

Leigh Disease

• Age of onset: 6 months
• Progressive neurodegeneration
• Motor and/or intellectual regression
• Focal Spongiform lesions = demyelination, gliosis

Question 7

How many people carry pathogenic mtDNA mutation?
How many people are diagnosed with mtDNA disease?
Why is there this discrepancy?

Answer 7

mtDNA mutations: 1/200

mtDNA disease: 1/10000

This is probably because of mutant load: the mutation rate is below threshold

NB Some communities have higher rates of the mutation in the population, such as the NSW Lebanese

Question 8

What proportion of the mtDNA genes can have pathogenic mutations?

Answer 8

35 of the 37 mtDNA genes

Question 9

What are the different genes that can be mutated resulting in OXPHOS disease?

How are these genes inherited?

What proportion of OXPHOS diseases do these mutations account for?

Answer 9

1. Mutation in mtDNA genes
2. Mutation in subunits encoded by nuclear genes

3. Mutation in OXPHOS biogenesis genes
Involved in:
 • Insertion of Iron / Sodium
 • Assembly of complexes
 • Insertion of protein into membrane

4. mtDNA replication genes
5. mtDNA expression genes
6. Genes involved in membrane dynamics
7. RNA transport, nucleotide transport and synthesis genes

Inheritance:
These genes can be inherited in all different ways:
 • Autosomal recessive (most)
 • Autosomal dominant
 • X-linked

These mutations only account for 50% of OXPHOS diseases
For 50% of cases, the molecular basis is not known

Question 10

In what ways have OXPHOS disease genes been identified?

Answer 10

Mitochondrial genome discovered in 1988, whereafter it was sequenced:

• mtDNA sequencing
• Candidate studies
• Linkage studies
• Targeted exome sequencing
• Whole exome sequencing

About 15 genes were discovered each year

Question 11

Mutations in which genes can cause to Leigh syndrome?

Answer 11

Up to 50 genes that can lead to Leigh disease:
• 30 Autosomal genes
• 12 mtDNA genes
• 2 X-linked genes

Even though there is this heterogeneity, the disease presents very similar in all the cases

Question 12

What are the challenges of OXPHOS molecular diagnosis?

How can this be (partially) overcome?
What are the drawbacks?

Answer 12

1. Large number of candidate genes
2. Mostly private mutations; i.e. no hotspots like in other disorders
3. Common mutations only in a few genes
4. Genotype / Phenotype correlation often poor
   • Phenotype can not guide the genotype investigations
5. If performing sequential testing, there are so many genes to look at, and it requires expert guidance

→ NextGen sequencing is a good option
• Allows much greater volume of DNA to be sequenced
• However, doesn’t have same sensitivity or specificity of Sanger sequencing

Question 13

Describe the changes to the cost of sequencing of the genome

Answer 13

2001: $100M
2013: $8000

Question 14

Compare Sanger and NextGen sequencing

What can NextGen technology allow us to sequence?

Answer 14

Sanger:
• 1 target DNA
• Average of all DNA molecules
• around 800 bp per run

NextGen:
• Thousands of DNAs at a time (MitoExome: all of the known mitochondrial proteins)
• Single molecule DNA sequence
• 800 million bp per run
• Candidate gene approach, but at a much larger scale

Can sequence:
• 10, 40, up to 1000 genes (i.e. whole MitoExome)
• Whole exome (20,000 known genes)
• Whole Genome

Question 15

What is massively parallel sequencing?

Answer 15

aka NextGen sequencing

Question 16

How big is the human genome?

How many genes?

Answer 16

3 x 10^9

20,000 genes

Question 17

List some NextGen sequencing technologies

Answer 17

• Illumina
• Ion torrent
• MinION (3rd generation)

Question 18

Describe Illumina sequencing

Answer 18

“Sequencing of fluorescently labelled DNA fragments”

Method:

– Target preparation –

1. DNA sheared into small fragments
2. Addition of adapters to each end of the DNA fragment
3. Adapters bind DNA to a flow cell covered with oligonucleotides through hybridisation

– Cluster generation –

4. Amplification of different pieces of DNA that have bound to flow cell
   • Through bridge formation
5. Repeat 35 times
   • End up with many copies of the DNA fragment tethered to the flow cell (single stranded)

– Sequencing –

6. Sequential addition of fluorescently labelled nucleotides
7. When the correct nucleotide is incorporated, it becomes excited, and allows an emission specific for the nucleotide that was added
8. Emission read by the machine

Question 19

Describe Ion torrent sequencing

Answer 19

Chip with half a billion wells
Beads contained in the wells

1. DNA sheared into small fragments captured on a bead
2. DNA fragments on bead amplified by PCR in an emulsion
3. Bead in well

– Sequencing –

4. Addition of complimentary nucleotides
   • A given nucleotide will give off a proton when incorporated
5. Subtle chemical change detected on a machine
   • Based on which nucleotide (A, C, G, or T) is incorporated, there will be a different signal

Question 20

Compare whole genome and whole exome sequencing

Answer 20

Whole exome:
• All 20,000 known genes sequenced

Whole genome:
• 3 x 10^9 bp sequenced

Question 21

How much is 3Gb?

Why is it such a challenge to find mutations in the whole genome?

Answer 21

(3 x 10^9)

This is similar to the number of letter in the Harry Potter series if it had 5000 volumes instead of 7

Mutations are like:
 • Change in letter
 • Change in word
 • Change in a page
within these 5000 volumes

Question 22

Describe the MitoExome Project
 • The cohort
 • Method
 • Stats of results
 • General overview of results

Answer 22

Over 1000 genes sequenced:
including all the known mitochondrial encoding genes

Within 44 OXPHOS patients
• Had severe diseases, determined to by OXPHOS disorders
• Diverse group
• Defects in the various complexes (Complex I, Complex IV or a combination)

Method:
• Hybrid in-solution selection of the candidate genes
• Illumina sequencing

Results:
• Average 1 change per patient (i.e. gene not ‘normal’)
• However only 8 of the mutations in the 42 patients were relevant to the disease
• The rest are probably ‘normal variation’
• 10 of the patients identified to have known disease genes
• 10 more patients identified with 8 novel disease genes
• 43% of patients had no prioritised genes, i.e. the mutated genes causing their disease are still unknown → where could they be?

Statistics:
Nuclear DNA:
• Over 1000 genes sequenced
• More than 1 million bases sequenced
• Good depth (i.e. fidelity of sequencing?)
• 29 rare mutations identified (i.e. classified as rare from the ‘1000 Genomes Project’)
• 17 rare, protein modifying mutations

MtDNA:
 • All 37 genes sequenced
 • Good coverage
 • Only 3 rare mutations found
 • 0 rare, protein modifying mutations found

Question 23

What is Prioritisation, in the context of the MitoExome Study?

Describe the ‘Prioritisation’ observed in the MitoExome study

Answer 23

Prioritisation: an indication that one is looking at the correct genes in the disease group
Ie Looking at mutations in genes causing OXPHOS diseases

MitoExome study:
In the 31 subjects study, there was prioritisation observed compared to controls in:
 • Known or candidate genes
 • Known disease genes
 • Candidate disease genes

This means that in the subjects of the study, there were many more known disease genes etc observed than in the control group

Conclusion:
The study was unmasking the correct genes

Question 24

What is good about NextGen sequencing?

Answer 24

1. Price
   • Has come down markedly since 2007
2. Quantity
   • Allows sequencing of a much greater volume of genetic material

Question 25

What was observed in the various candidate genes in the MitoExome cohort?

Answer 25

1. Two of the candidate genes were mutated in two different patients from different families. These two individuals had the same clinical phenotype • AGK • MTFMT (2) ``` 2. Six mutated genes in patients, that are associated with a specific biochemical phenotype. In these six patients, the biochemical phenotype matched this • C1orf31 • C6orf10 • LYRM4 • MTFMT (2) • NDUFB3 • UQCR10 ``` 3. Four of these genes were then proved to be the cause of the disease • Through lentiviral correction studies 4. Four of the candidate genes have since had additional patients reported in the literature • Confirmation of these genes as causing OXPHOS disease 5. Two of the genes have been eliminated, because other disease causing mutations in these patients were found

Question 26

What did lentiviral correction studies show in the context of the candidate gene in the MitoExome study? Describe this process

Answer 26

These studies showed Causation of the mutated gene in the disease process 1. WT version of gene put into lentiviral vector 2. Viral particle put into fibroblast cells in patient 3. See whether the enzyme defect is corrected a. Correction → the gene plays a causative role in the disease b. No correction → no causation displayed Example: • Two patients, both with a defective Complex I, but with different mutations • Two lentiviral vectors, each with either the WT of the gene mutated in either Patient #1 or Patient #2 • Lentiviral vector was able to restore Complex I function only in the patient that had the corresponding mutation → Confirmation of causation

Question 27

What about the causal genes in those 43% of patient in the MitoExome study whose mutations are still unidentified?

Answer 27

These mutations could be: 1. In genes that were missed with the MitoExome sequencing (since only 1000 genes that were covered) For example: • Poor coverage • Exon deletions 2. Detected but not prioritised ``` 3. Non-targeted region For example: • Intron • Regulatory region • Unidentified genes in the mitochondrial genome ``` 4. Complex inheritance • Epigenetics etc

Question 28

What are the advantages and disadvantages of the various sequencing approaches: • MitoExome • Whole exome • Whole genome?

Answer 28

Mitoexome: • mtDNA + 1000 nuclear genes • ? still missing genes Whole exome: • 20,000 genes • Excludes mtDNA Whole genome: • Sequencing of everything • However, too much data

Question 29

Describe the Gartner Hype cycle in the context of NextGen sequencing

Answer 29

1. Technology trigger 2. Exponential rise 3. Peak of inflated expectations • Everyone was very excited 4. Trough of disillusionment • Patients weren't being diagnosed as well and as quickly as was thought possible • Expensive • Bio-informatic issues 5. Slope of enlightenment • Less expensive • Technology sufficiently utilised to put together good protocols in terms of analysing the data 6. Plateau of productivity At the moment with NextGen, we are on the 'Slope of enlightenment'

Question 30

What proportion of childhood-onset OXPHOS disorders have had their causation identified?

Answer 30

Only about 50%