A joint Congress by ERA-EDTA and ISN

FINDING THE GENES OF HYPERTENSION: LESSONS FROM THE HYPERGENES PROJECT

Daniele Cusi, Milan, Italy

Chair: Xavier Jeunemaitre, Paris, France

Gerjan Navis, Groningen, Netherlands

Prof Daniele Cusi Division of Nephrology and Dialysis San Carlo Borromeo Hospital Milano, Italy

Slide 1

Good afternoon to everybody. I think I will try to go fast and skip some of the previous assumptions in order to save time.

Slide 2

Slide 3

Slide 4

Slide 5

Slide 6

Slide 7

Slide 8

Since we are in a very renal environment, I will go from the beginning and I will start from a systematic study that is undergoing in our laboratory in collaboration with a group in San Raffaele hospital with my colleague Paolo Manunta and co-workers.

Slide 9

On one side we are performing an association study with standard techniques on the sodium transport system at the kidney level and this is just to show a cartoon, a basic cartoon on what we’re doing on some phenotypes on what is some genes on the chromosome 1, this is just to show the chloride channel on the top part of the slide and just to show what is seen; we have already discussed part of this in other previous communications, how the sodium moves on the thick ascending limb of Henle’s loop. The sodium enters through the sodium potassium chloride and then sodium goes out through the sodium potassium pump, whereas the potassium goes out through the ROM K and the chloride channel, provided that Barttin is functioning. Then a gradient occurs, so that also other cations move through a paracellular way.

Slide 10

In this way we tried several phenotypes and for example for the salt sensitivity phenotype (measured in an acute way) we tested different SNPs on chloride channel, type A and type B and we saw that chloride channel type A is significantly associated to sodium sensitivity phenotype, whereas it is not associated with sodium chloride channel type B and it is not at all associated with Barttin. Moreover, we found that there is a change of aminoacid also if we have no idea if these aminoacids are responsible for the association.

Slide 11

When we go and see the intermediate phenotype associated to these mutations and I plotted the three SNPs more associated, SNP 3, 4, and 6 of the association and we see the increase of blood pressure according to the genotype for the phenotype salt sensitivity, we see that plasma renin activity, measured the day before Na-sensityvity test, was decreased in the individuals that had the largest blood pressure increase according to volume expansion with the 3 different genotypes or the 3 different SNPs. Basal heart rate was decreased as well, according to what was expected for salt sensitive individuals.

Slide 12

Then we considered also Barttin for a different phenotype that was basal blood pressure.

Slide 13

We tested a first sample of about 300 cases and 300 controls and remember that we wanted to be sure of the definition of the cases and controls so that hypertensives have to have developed hypertension before 50 years of age and normotensives had to have remained with normal blood pressure until at least the 55th year of age and had to be followed up for at least 5 years and have their blood pressure be controlled for at least 5 years. We saw a statistically significant difference for one SNP.

Slide 14

Since the problem of replication in these studies is very, very crucial, we tested in a second sample

Slide 15

and we had the confirmation so we could pool the two samples and we had a very, very high significance, this study is going to be published very soon.

Slide 16

Then why did we enter this sticky business of these association studies with genome scan?

Slide 17

Because genome scan is in our hand a tool and also if we knew as we saw with Xavier a few minutes ago that hypertension is really not very promising we feel that the largest majority of the studies have problems and we agree completely with what Doctor Dominiczak said that if we are not very selective in selecting controls, we risk overinflating the controls of individuals that have a genetic inflation of genes for the cases.

Slide 18

This was recognised also in the WTCCC paper because hypertension is a late onset disease with very high prevalence. So we, as doctor Dominiczak decided to remain with the hypercontrol selection, so to select controls that were sure to remain with normal blood pressure for a long time.

Slide 19

So two consortiums started InGenious HyperCare that is led by Anna Dominiczak

Slide 20

and Hypergenes that is coordinated by myself

Slide 21

that want to perform a genome scan with very much care for the definition of controls.

Slide 22

Our consortium is based on the genome scan of a large number of individuals based on the study already available of course otherwise you will never end such kind of study and with the technical assistance of large enterprises and for the bioinformatics we based on the IBM research laboratory in Haifa that IBM devoted just for this kind of genetic study and for the technical part STMicroelectronics that will build up a lab-on-chip for the final custom array that will perform the diagnostic chip.

Slide 23

Within the cases we will have also important endo-ßphenotypes for a large number of cases that will never have received any kind of pharmacological treatment before have a good phenotyping so that we have between 600 and 700 echocardiographies. 650 patients underwent a formal pharmacogenetic study of hydrochlorathiazaide treatment and 550 had a pharmacogenomic study of losartan whereas all of them had biochemistry, anthropometry and renal function study measurements.

Slide 24

Well I’ll skip the genomic analyses

Slide 25

but just to give you an idea when you have such large number of data they cannot be seen individually and you need a bioinformatic infrastructure that helps you to analyse this data in terms of quality control and analysis. This was all set up during the first year of the study, actually we started doing well before otherwise we never would have been ready.

Slide 26

Where are we now? We finished the genotyping of the first 4000 individuals, 1 million SNPs per individual.

Slide 27

As soon as we will have finished the first genetic analysis, so that we are here now we will define a large number of SNPs for the intermediate phenotypes including the pharmacogenetic studies,

Slide 28

we will put them on a new custom made chip where we will also put other interesting SNPs that were candidate genes coming from the literature

Slide 29

and then we will retest them on the remaining cohort

Slide 30

Slide 31

Slide 32

Slide 33

and then we will put the most interesting chip on a lab-on-chip that is this gadget that already exists which can hold now 200 SNPs.

Slide 34

As a prototype it has a reasonable cost, it is around 150 dollars and can perform a genetic analysis in 2 hours with a machine that again already exists so I expect that by the end of the study the cost will be much less and we’ll be able to perform individual genetic analyses for the questions that we will be able to answer by the end of the study.

Slide 35

Maybe the response to treatment and just to give you a sip of what we have already been able to do as a preliminary analysis we performed a pharmacogenetic study of the blood pressure response to diuretic treatment in 201 hypertensives.

Slide 36

And this is the result. This was done with a minichip of the lumina and this was the result of the genetic study.

Slide 37

I think that this is all for now. Thank you very much for your attention.

Slide 38

Chairman: Thank you very much Professor Cusi for this beautiful presentation and these interesting results. This paper is now open for discussion.

Question: Daniele if I may ask the first question? We have probably all of us in the field seen the effort of Lifton’s laboratory trying to re-sequence genes involved in Bartter’s syndrome or Gitelman’s syndrome and so that there were some rare variants that could at least explain some mm of Hg of difference. I was very interested in your results on the polymorphism on CLC, MKB and -- and the fact that you might have also a mutation in MKB causing Bartter’s syndrome. Do you think it would be worth the effort to completely re-sequence these particular genes in your population to see in addition to the polymorphisms themselves if there are not some rare variants that might explain some other things?

Prof. Cusi: I think it is worth it and for sure when you see some difference or a signal, it is indeed a good idea now that the technology helps us and if you have good samples, good cases it is worth to go directly on the region and do re-sequencing. It was not wise 2 years ago because the costs were extreme, now if you have the good samples it should be kept in mind also in Europe where we don’t have enough money probably. Indeed we have the technology in our laboratory and it is in the programs. For example, we are going to re-sequence the chloride channel A for the best results on our patients. It’s not our number one priority because of time and manpower but it will one of the things in the program.

Question: May I ask you one question on the data presented here? Could you relate this difference in blood pressure to a difference in sodium balance?

Prof. Cusi: We don’t know because while the first sample was done trying to keep the patients in sodium balance, the second sample we don’t have the guarantee that the patients were keeping a reasonably stable salt intake which is very difficult in the clinic.

Question: But it would be interesting to dissect.

Prof. Cusi: Yes but I still want to see somebody who can give such a big sample of never treated before essential hypertension patients in sodium balance with these big numbers.

Question: Big challenge yes good.

Prof. Cusi: This is thanks to Paolo Manunta’s group that provided the cases.

Question: Yes wonderful.

Chairman: Are there any other questions in the audience? So if not thank you again Daniele and we’ll move to the next talk given by Professor Luft from Berlin.