CME Slides Forum

A joint Congress by ERA-EDTA and ISN

LESSONS FROM AUTOSOMAL DOMINANT HYPERTENSION AND BRACHYDACTYLIE

Friedrich Luft, Berlin, Germany

Chair: Xavier Jeunemaitre, Paris, France

Gerjan Navis, Groningen, Netherlands

luft

Prof Friedrich C. Luft
Charité Campus Buch
Experimental and Clinical Research Center (ECRC)
Berlin, Germany

Ladies and Gentlemen I’m happy to be here. What we see here is a map of the Middle East. You may not recognise the names because it’s the map of the Middle East the way it was 5000 years ago. What we see here is the travels of Jason and his adventurers going to Colchis to capture the Golden Fleece. They came to unhappy ends and their voyage took them 20 years.

Colchis is located at the same place where the modern Turkish town of Trabzon is located and we had occasion to go to Trabzon in 1994 to work up a family with a Mendelian form of hypertension and short fingers that was first described in the English language by Nihat Bilginturan, that was15 years ago. As you see here, in affected individuals they invariably have a type E brachydactyly compared to non affected individuals. This is mean arterial blood pressure plotted against age and as you can see here in affected individuals there is a dramatic increase in mean arterial blood pressure so that when they’re my age they have a mean arterial blood pressure, not systolic but mean of 180 mmHg and the other remarkable thing is that when they’re my age they’re all dead with the exception of these two individuals that are living in Germany whose hypertension was treated.

Now we did a linkage analysis and got a LOD score of 9 on the short arm of chromosome 12 and so in 1995 when I submitted this for publication I thought that we would be getting somewhere and my lesson is that life is not always easy. Now we didn’t have Grant Liddle to phenotype these individuals so we had to do it ourselves. As you see here their responses in terms of renin, aldosterone, norepinephrine, in terms of normal salt diet, high salt diet, low salt diet or low salt diet plus furosemide are perfectly normal. So this separates them from all other Mendelian forms of hypertension in that this is not a salt and water syndrome.

Here’s an example of the brachydactyly. It’s a type E brachydactyly the affected individuals have cone shaped epiphyses that are directed forward or anteriorly. Here’s an x-ray from a 5-year old Turkish child with her family.

Some investigators from Japan published a deletion syndrome on the short arm of chromosome 12 in 1997 with exactly the same kind of brachydactyly. So we asked these investigators to let us have some of their DNA and we mapped this deletion syndrome and there’s an overlap with our linkage interval and the deletion syndrome so we reasoned that the gene that’s responsible for our autosomal dominant brachydactyly and hypertension has to lie within this overlap.

Since then we’ve discovered 4 other families that also map to this locus but they have not helped us narrow down the linkage interval any further and here are some examples of two other families, this is an American family and this is a family from Canada. They’re not of Turkish origin and they also map to this locus.

Now the next thing that we did being clinicians is to see if we could outfox Grant Liddle and do some clinical pharmacology. So we did a randomised controlled crossover trial of 6 regimens and the placebo over the course of one and a half years to see if some medication would work better than another medication to help us figure out what’s wrong with these people. As you see here, this is factored for placebo all these different medicines lowered blood pressure to a similar degree as is the case in patients that have essential hypertension.

So this was an interesting study but it did not help us further. We did observe in our 24-hour blood pressure measurements that there were some aberrancies in baroreflex sensitivity and so my impetuous young associates looked at how this is regulated, how does the baroreflex work. Here’s the nucleus tracti solitarii and here’s an area called the rostral ventrolateral medulla that’s important for baroreflex regulation and we performed magnetic resonance imaging on these persons affected and non-affected individuals in this Turkish family and we found out that all the affected individuals have an aberrant vascular structure that’s immediately adjacent to the rostral ventrolateral medulla.

This is called a neurovascular contact syndrome and it’s well known for conditions such as trigeminal neuralgia. So we decided to bring the family back to Berlin to do some detailed measurements of autonomic function and this is microneurography looking at sympathetic nerve signals and there is no difference between affected persons and a person with essential hypertension or a normal individual. So these patients do not have increased nerve traffic.

Now we tested their baroreflex in terms of their capacity of being able to buffer a dose of phenylephrine and if I give you an injection of phenylephrine, you will easily put away 100 µg of phenylephrine in order to raise your blood pressure 12.5 mmHg. So we did that with these patients and what these data show is they are not able to buffer changes in blood pressure at all. Although their heart rate regulation in terms of baroreflex function is ok, they’re not able to buffer changes in blood pressure. So we’re not quite certain what this means but it might support the notion that this is a problem with autonomic regulation. We’ve also done other phenotyping including buttocks biopsies and examining their blood vessels and molvani miographs and this kind of testing.

This is a blood vessel section of a 20-year old that’s affected compared to a 20-year old that’s not affected.

The difference in blood pressure between these two individuals is not yet profound. We tested genes in the interval pharmacologically with various techniques. This is a potassium channel that we tested with diazoxide. It turned out to be innocent.

This is PDE3 that was also within our linkage interval, it was also found to be innocent. We sequenced every gene that’s in the linkage interval and found no mutations in any.

Now this is rather frustrating work but here’s something that sort of cheered us up a little bit. This is a linkage study in isolated Chinese families with essential hypertension, they don’t have brachydactyly but in this linkage study from these Chinese patients with essential hypertension there was linkage to the short arm of chromosome 12 with a LOD score of 3.5 and there’s pretty good overlap between this linkage interval and ours.

Since we found no mutations in any genes the next hypothesis that was tested was, is the stuff really ok or is it merely in the wrong order?

So we obtained BACs that we labelled and did interphase FISH analysis in these linkage intervals to see if it’s possible that this could be a rearrangement syndrome and indeed it is a rearrangement syndrome. As you can see here in the Turkish family there’s this BAC that’s deleted here and reinserted at this site and then these two BACs were inverted compared to the normal arrangement that’s shown up here. Here’s an individual from South Africa, here’s the American family, here’s a Canadian family, here’s a French family and they all have rearrangements but the frustrating thing is that the rearrangements are all different but they do have this one area that all of these families have in common.

So we thought there might be something going on here. There’s a shared minimal inverted region and it lies between two known genes that we sequence, SOX5 and this gene. This is a gene poor region and according to human genome information there were no known genes in it. But we did have 3 expressed sequence tags that nobody knew exactly what to do with. So with 3 prime and 5 prime race PCR we’ve come up with this construct that has Exons a total of 32 Exons. But this construct has no open reading frames and no cosar sequences suggesting that it does not code for protein.

What could it be? We did expression studies on every one of these Exons and cells that we had from affected and non-affected persons. What we found is that in affected individuals this Exon 22 is never expressed. Now we have some pretty good bioinformatics people and people that are interested in non-coding sequences at the Max Delbrook Centre and my friend Nicholas Raievsky and his colleagues tell me that the structure of what this thing could code for, what it could eventually be looks like this.

This looks like it could be a microRNA. Now we next looked to see if we could find this in the mouse and we could not find it in the mouse but we found it in the chimpanzee and we found it in the orang-utan and we found it in various other human tissues, so we think that it’s real.

Now you’ll recall that microRNAs that are all in vogue now are processed by a complex situation into pieces. These small pieces then tend to suppress other genes and have an important regulatory function. Our next challenge is to see if we can find out is this indeed a microRNA? And can we find out what this microRNA does? One approach that we’re using is to try to confirm all the stuff that we learned from 3 and 5 prime race PCR. We have some associates in -- that can cut out a portion of chromosome 12 and we’re going to subject that to deep sequencing.

Then what we’re also doing is expressing this microRNA. I think, there it is, this is the sequence and we found it in human placenta and a condrocyte line and a gliomal cell line and HeLa cells and blood cells and primary fibroblasts of these patients.

What could it do? Well with bioinformatics we can look at genes that it might address and we have hits in terms of genes that have to do with angiogenesis. Genes that are involved with the nervous system and genes that are involved with skeletal genesis which are related to the phenotypes that our patients have.

What we’re also doing with collaborators at our institute is to express this putative microRNA in HeLa cells and HeLa cells make about 5000 proteins, that’s not all the proteins that we have but it is a lot. We want to see with the same technique that was used in this study for another microRNA. I’m not going to go into details here, this paper was recently published but we’re going to use the same technology looking at the expression of proteins that could be influenced by this microRNA with a technique called pSILAC.

So what’s the novelty here? We published the possibility that this was a microRNA last year and recently, very recently there was a paper in Nature Genetics describing another Mendelian syndrome that is involved with hearing loss. This also related to a problem in a microRNA. We’re going to use a clinical application of a recently described method to determine this microRNA’s function and we hope that we’ll learn some information that maybe some other day I can tell you something about.

Now it takes a lot of people to do work like this and I can’t mention them all but Yvette, Sylvia and Atakan have spent a lot of time on this project. Thank you very much for listening to me.

Chairman: A great story that has not ended yet. Questions in the audience yes please?

Question: Well, I always lecture about Arthur Guyton and that he was right and that all the genes identified for chronic regulation of hypertension so far have to do with salt and water handling by the kidney and I don’t want to modify that lecture yet. So how sure are you that your renal sodium phenotyping is definitive first? And second what do you know about the anatomy or structure of the kidneys in these patients who have genes involved in angiogenesis, skeletal genesis etc could it be that you still have a renal sodium disorder through an unexpected mechanism?

Prof. Luft: Well, I’m also a student of Doctor Guyton and Doctor Guyton also noticed that the renal function curve in most people with essential hypertension is pretty darn steep. It looks pretty darn steep in these people also. Their renal function is normal, they don’t have proteinuria, their GFR is ok, all the imaging studies that we’ve done of their kidneys and their blood vessels are ok. But I agree renal function curves can be steep and this one is steep but I don’t think that that’s a contradiction to Guytonian logic.

Question: Fred the story of this possible microRNA is quite interesting. How do you understand from the chromosomal rearrangement that you have seen in these families that this particular gene will be non expressed because it doesn’t seem that the arrangements themselves are just cutting down the gene but just the interval is comprising the gene right?

Prof. Luft: I don’t think the rearrangements that we’ve found can explain the phenomenon that we’ve noticed that this one Exon is not expressed in affected individuals and I assume that there must be some microdeletion or something else going on that might explain that because using these BACs to find these rearrangements is pretty crude.

Question: Do you envision to use for example other cheap NimbleGen chips with dedicated CGHR in there just to go further?

Prof. Luft: We used chips and looked at all the SNPs in the region and thus far it has not helped us.