Results of the English Q&A with Dr. Bohn held on July 10


Q&A with Dr. Markus Bohn – Topic: Genetics of Connective Tissue Disorders (main interest in Tenascin X)

Part 2 – Results from the English Q&A held on July 10 2016

1

Q

I haven’t looked into TNXB and Tenascin X very much at all and a lot of this is over my head, so this might be a silly question. How do TNXB and Tenascin X relate specifically to EDS? It makes sense that they go hand in hand, but do all people with EDS most likely have a TNXB variation? Or can you have EDS without the TNXB factor?

A

There are a few different connections between EDS and TNXB. It looks like ~10% of patients with the hypermobile EDS type have a TNXB variant. 9% of patients with CAH have a defunct TNXB allel and suffer from CAH-X, which is a separate disease type. Patients with Tenascin-X deficiency (no serum level of the protein encoded by TNXB) resemble the classical type of EDS. This does not mean that everyone who has EDS necessarily has a TNXB variant or a deficiency on the protein level.

2

Q

What is the raw data from the lab that did the genetic testing? Is it something I need to get? Is it something a geneticist uses or reads?

A

The kind of data depends highly on the type of testing that’s being done. Let’s assume we use some kind of sequencing for now. The real “raw” data are intensities for color values that correspond to individual nucleotide positions (because of the technology used to sequence). Those color values are then translated to the actual nucleotide sequence. Each of those read sequences is only a few hundred bases long, a longer gene can be >70.000 bases long (like TNXB), so that needs lots of overlapping reads to be covered in full.

Now sequencing is not perfect, it’s in fact very easy to get a nucleotide wrong or to assemble multiple reads in the wrong way, so it needs a lot of repeats as well. Usually somewhere between 50x to 100x redundancy is what’s considered ok nowadays (as far as I saw).

Genetic testing can also mean something that’s not sequencing. Allel-specific hybridization or SNP microarrays are forms of genotyping that rely on specific probes for known variants.

If you’re going for genotyping the data you could potentially take home could be a list of 500.000 genetic variants and which of those you do or do not have (that’s if they use the Affymetrix Human SNP 5.0 GeneChip for example). If it’s gene/exome/whole genome sequencing that you’re getting you might get a list of all sites that differ from a reference sequence. The benefit of sequencing is that if you have something that has never seen before, it can be detected.

Sorry this was a bit longer, but I was thinking this was a something that might be of interest to a lot of people. I just had to look up myself how many variants are picked up by an Affymetrix chip. When going for genetic testing I’d always request a list of all variants found or – if possible – the sequences themselves. Those will never change for the rest of your life, so you’ll be able to take them to any other specialist or new type of analysis that may exist in the future.

3

Q

I have have a TNXB single variation. I am being told it has never been recorded before. I am hypermobile, but I am mostly showing symptoms of vascular connective tissue disease. I have small and medium artery ruptures, several large artery aneurysms in aorta and internal carotid artery, tortuosity in iliac and internal carotid arteries and aorta, and severe stenosis due to tortuosity in cervical artery. I have no other reason for this to be happening except connective tissue disease. Vascular issues started in my 40s. There is a history in my family with two related sudden deaths. There is no hypermobility in my family. Is it possible that my TNXB variant could be causing this or is it impossible to be caused by TNXB? I had Next Generation testing which rules out Marfans, Loeys-Dietz or VEDS. Could there be some other genetic issue causing my vascular problems?

A

That depends a lot of the mutation itself, some can be very disruptive (introducing/deleting cysteines or bulky amino acids), some can be neutral (e.g. exchanging a valine with a leucine in a flexible part that makes no interactions). Otherwise, Tenascin-X is localized around blood vessels and we do know it’s an “organizer” of connective tissue properties so it could be very well one of the keys to what you’re describing.

But one big caveat: TNXB is one of the few genes for which we are able to draw this kind of connection (and I worked with it, so it’s the first that comes to my mind). There will certainly be other candidate genes and maybe even unknown types of connective tissue disorders that might explain what you are describing.

4

Q

Also–do people with EDS and neuropathy always have small fiber neuropathy? And should a biopsy of the skin be positive for SFN?

A

Hmmm, I’m not aware why it should be limited to SFN. Interestingly, Tenascin-X can be found in peripheral nerves and there was some speculation about Tenascin-X involvement in muscle-nerve junctions … So I wouldn’t be surprised if there’s some causal connection between EDS (at least with Tenascin-X defects) and peripheral neuropathies.

5

Q

Hi! I’m wondering if it’s possible that CRISPR technology could one day be used to correct the TNXB defect.

A

Oooh, I do actually have a CRISPR experiment on another (non-EDS) project 🙂 and I love the technology. The hope is (I think) to one day be able to edit out any genetic cause for disease. But it would probably have to happen way before birth, since only then you could make sure to hit the target gene in every single cell where it could cause problems.

6

Q

What is the easiest way to detect the TNXB gene mutation? Blood test, skin biopsy..?

A

The “flavor of the day” for finding those is exon sequencing. There are lots of commercial offerings for whole exome sequencing (that’s sequencing of all protein-coding regions in the genome usually with their regulatory elements). Caveat here is, that if 2 or more exons share high sequence identity (e.g. because of non-functional pseudogenes) that can lead to errors (as it has happened to some people who wanted to get their TNXB sequence that way).

Q

Thank you for your answer.

Ok, so the full exome sequencing might lead to errors? Are there better/other options?

A

The errors other were getting were along the lines of “exon 40 of TNXB couldn’t be sequenced because of pseudogenes”. So at least you wouldn’t be misled by a false-positive mutation or something. If you know you only care about the TNXB, then single gene sequencing. If you want to have all your bases covered: whole genome sequencing.

7

Q

How does Tenascin X play a role if any in CSF leak or Chairi?

A

Since Tenascin-X mutations can affect connective tissue properties and Cerebrospinal fluid leak and Chiari malformations have a definite connective tissue component, there could indeed be a connection. There is no data on that (just like on any other connection between connective tissue disorders and their comorbidities), but if there were any 🙂 it could look similar to what’s observed in collagen defects.

8

Q

One more question. Are there any ideas whether TNXB and other collagen mutations affect cell-to-cell communication?

A

Extracellular matrix proteins often have signaling functions and collagen (as an example) is involved in inducing epithelial-to-mesenchymal transitions, Tenascin-X is involved in TGF-beta signaling. I’m not aware of specific mutations that have been studied in the context of cell signaling, but it’s definitely something that will happen further down the line.

9

Q

What are the benefits of biomedical research?

A

We don’t just see association between a special genotype with a phenotype but we are in a position to find causative relationships and develop new options for diagnostics and treatments.

10

Q

Can you give us a few insights into your new TNXB paper?

A

CAH-X patients suffer from TNX haploinsufficiency because one of the two allels is non-functional. Despite this, the EDS phenotype exhibited high variability. We hypothesized that point mutations on the remaining TNXB allel lead to the different EDS phenotypes.

11

Q

What is the difference between a scientist and a medical doctor?

A

The job of medical doctors is to apply recent knowledge and insights. Scientists have to generate broadly applicable knowledge and insights which then are again applicable by practicing medical doctors.

12

Q

What is a protein and how does a mutation affect the protein?

A

Genes encode for proteins that have a function. A gene is only information whereas the protein is the functional entity. Mutations can lead to different amounts of protein if they are in a non-coding region or they can affect the function of the protein.

13

Q

What are the different TNX EDS types?

A

1. TNX deficiency: No TNX detectable in serum (homozygous)

2. TNX haploinsufficiency: Half the amount of TNX in serum compared to unaffected people

3. CAH-X: TNX haploinsufficiency; EDS + CAH

4. TNXB mutations we don’t know what they are doing at the moment (functional impairment?): Full amount TNX in serum. Produce a protein with altered characteristics.

14

Q

How does a genetic predisposition lead to a phenotype?

A

The Cysteine is replaced by a Tryptophan. The replacement of Cysteine with a Tryptophan leads to  breakage of a disulfide bond disrupting the folding of secondary structure. This again leads to failure in organizing fibrillin fibers in tissue and to the EDS tissue phenotype.

15

Q

How meaningful / precise are genetests and which method is the best?

A

Exome Sequencing has the potential to detect defects in the protein-encoding region but not in the regulatory region for example promotor or intron splicing sites.

Whole Genome Sequencing examines the whole DNA with all regulatory elements.

Both are limited by a high error rate if one is looking for a single nucleotide differences. It is easier to find larger deletions. Many mutations are silent mutations that have no impact.

Panel diagnostics and single gene tests are more precise but rely on a predetermined group of markers.

16

Q

If I remember right, some genetic diseases can be diagnosed by metabolic markers in blood and urine (for example hypophosphatasia, in which some values are too low and others too high because there is a metabolic blockage). Would it be possible to conclude from abnormal blood or urine tests in which area there might be a metabolic blockage, even if the location of a potential mutation isn’t known yet? Would it be feasible to systematically look into the different metabolic regulatory circuits which might be responsible for abnormal lab results, and conclude in which genetic areas there might be a mutation? Is this a path research is taking?  Is there any way in which low alkaline phosphatasia might be related to EDS (hypermobile, mutation unknown)?

A

There is a lot of exciting research going on in the area of linking metabolic markers with the genetic underpinnings of a disease. As an example, there is a spectacular technology in which during surgery tissue is vaporized right on the surgeons “blade” and analyzed for metabolic markers of a tumor type in real time while the surgeon is removing it (so he knows where the healthy tissue starts and the tumor ends). This data (specific metabolic tumor markers) is combined with deep sequencing data (mRNA levels for every gene) and across patients to look for new connections or entire networks of interdependent factors that might be involved in disease outcome. So yes, there is strong research into that 🙂

Regarding the alkaline phosphatase in EDS: As far as I know there is no clear connection between low enzyme levels and EDS. But the individual role enzymes that are broadly involved in metabolic functions are hard to be pinned down in a multifaceted syndrome.

17

Q

My mother, both my daughter’s, and I are all diagnosed EDS Hypermobile Type. However, I recently learned that one of my daughters and I have the genetic variation associated with Stickler Syndrome (COL11A2) and are heterozygous for AMPD1 associated with myopathy (we were the only two family members to have genetic testing done). The geneticist said these findings were of “unknown clinical significance”. Since the specific gene responsible for EDS-HT has not yet been identified, I wonder if it’s possible multiple, fairly common variations such as these may be responsible in combination or with environmental factors. Are you aware of any research being done in this area?

A

I am not working in diagnostics, but as far as I understand “clinical significance” can only be granted if that precise mutation has been reported to be correlated with a disease phenotype in the literature. It is very hard to impossible to predict what the phenotype of individual mutations is going to be. Several groups are working hard to find genetic variations that are linked to EDS so the number of mutations with “clinical significance” should keep going up.

Regarding combinations of common/insignificant variants: small changes to different parts of complex networks can indeed add up or have synergistic effects (leading to a “perfect storm”) and cell/tissue functions form an extraordinary difficult network. So research mostly starts with the low hanging fruit looks for clear 1 genetic variant – 1 defect relations and if that doesn’t work we move up. At the very end people stop looking for specific variants and start comparing all variants. A current example would be search for the genetic basis of schizophrenia, which turns out to be very difficult and where groups started looking genome-wide at correlations between single nucleotide variants. That kind of approach would indeed discover if combinations of common variants correlate with a clinical phenotype. I don’t think this is being done with EDS (yet).

Environmental factors: Without a clean cut cause-effect relationship it’s hard to get applicable insight, but it does appear that in some cases of EDS there are hormonal or immunological circumstances that might influence disease onset or progression. So I would count that as environmental factors, but I’m not aware of anything actionable that could be used for diagnostics (yet).

18

Q

How accurate/reliable is the gene test for vascular EDS?

A

Currently >95% of pathogenic variants for vEDS can be detected by sequence analysis (and there are other genetic tests for some rare variants). But that’s not 100% so there is always a chance for vEDS symptoms and no identifiable genetic cause. That’s why clinical consequences > tissue phenotype > genetic variants.

Also, there are overlapping EDS types that exhibit vascular-like symptoms on other genes than COL3A1. For example, a COL1A1 mutation leading to an overlap between classical and vEDS.

De Paepe A, Malfait F. The Ehlers-Danlos Syndrome, a disorder with many faces. Clin Genet 2012; 82; 1-11.

19

Q

Do you think all cases of HEDS are genetic mutations vs. autoimmune?

A

So far the working hypothesis is that there are direct genetic causes for the different EDS types. Also, there is no clear vs. in what you’re asking. Quite a few autoimmune diseases can be linked to genetic mutations (e.g. there are epidemiological studies that give a 65% chance for inheritability of osteoarthritic hands), but those often imply polygenic inheritance.

20

Q

Do you feel like a cure or better treatment options can come from these findings?

A

Assuming you are talking about point mutations in TNXB that affect severity of an EDS phenotype, the answer is no. Finding the genetic basis for a disease does sadly not translate into curative approaches. Otherwise it would be very easy to deal with negative consequences of aneuploidies (e.g. trisomies). In the case of structural extracellular matrix proteins like Tenascin-X it’s very hard to think about therapeutic/curative approaches since it’s a protein that fulfills its function by being at the right time at the right place in the right amount. There is no enzyme that can be blocked by a small molecule or a receptor that needs to be activated, so it doesn’t fall within what’s currently possible with pharmaceutical intervention. Fixing a “broken” or supplementing “missing” Tenascin-X in the entire connective tissue is something for which no technology exists.