CME Slides Forum - T. Miyata

A joint Congress by ERA-EDTA and ISN

ADVANCED GLYCATION, HYPOXIA AND OXIDATIVE STRESS

Toshio Miyata, Sendai, Japan

Chair: Paola Fioretto, Padua, Italy

Carol Pollock, St. Leonards, Australia

miyata

Prof. T. Miyata
Center for Translational and Advanced Research
Tohoku University Graduate School of Medicine
Sendai, Japan

Thank you very much Professor Fioretto. First of all, I would like to appreciate the organising committee for kindly providing me with this opportunity.

The prevention of diabetic nephropathy has made significant progress as classical factors contributing to this pathology, for example hypertension, hyperglycemia, hyperinsulinemia and hyperlipidemia are now amenable to treatment.

Yet current therapies do not fully prevent renal complications. Recent studies of animal experiments have unravelled new culprits stemming from abnormal oxygen metabolisms such as hypoxia, oxidative stress and advanced glycation. I’m very much happy to share with you some of our data to delineate their contribution and also to discuss potential therapeutic approaches.

The first factor concerns renal hypoxia and relies on the inhibition of oxygen sensor.

The peritubular capillary plexus fed by glomerular afferent arteries supplies oxygen and nutrients to the tubulointerstitial tissues and diabetic renal lesions decrease the number of peritubular capillaries and thus, oxygen diffusion to the tubulointerstitial species. This chronic hypoxia engenders tubular dysfunction and eventually fibrosis.

To verify this concept we relied on pimonidazole that specifically detects the hypoxic cells. As you see, the pimonidazole staining was very positive in diabetic but not in control normal kidneys. 33 weeks after birth confirming therefore, chronic hypoxia in the tubulointerstial species.

How do we defend ourselves against hypoxia? Our defence system hinges upon the hypoxia inducible factor: HIF. Under hypoxic status HIF translocates from the cytoplasm into the nucleus where it activates a master gene switch which results in broad and coordinated downstream reactions involved EPO, VEGF and GLUT all of which are involved critically in the defence against hypoxia.

In the presence of the sufficient oxygen HIF is constantly hydroxylated by prolylhydroxylase, PHD and then degraded within the proteasomes.

So increasing HIF stability by inhibiting the PHD activity may thus offer protection against hypoxia in general.

In order to evaluate the role of HIF/PHD system in diabetic nephropathy we took advantage of the ability of cobalt that inhibits HIF degradation by PHD. Cobalt inhibits a PHD activity by replacing iron an essential element for the activity of PHD, so that HIF is not degraded and activates a master gene switch.

So we gave cobalt for 20 weeks to hypertensive type 2 diabetic rats, here SHR/ND. Cobalt did not affect hypertension and metabolic abnormalities in these models such as obesity, hyperglycemia, and hyperlipidemia are not affected.

As anticipated, expressions of HIF regulated genes, for example EPO, VEGF and GLUT are significantly increased by the cobalt treatment. Most impressively the proteinuria, as well as the histological abnormalities, are significantly improved by the cobalt treatment.

Unfortunately, the toxicity of cobalt hampers its clinical use in humans, so the availability of less cumbersome, non-toxic small molecular activators of HIF should prove very useful for the therapeutic interventions.

To this end we have tried to obtain such a novel compound. We have developed a very unique in vitro assay system relying cells expressing luciferase under the control of 5 tonne repeat of the hypoxia responsible elements and eventually identified the two very unique compounds here TM6008 and TM6010 able to stimulate the HIF ability like cobalt in vitro.

This is the 3 dimensional structure of the human PHD2 structure. As you see, our computer simulation docking pictures disclosed that our compounds, I put several compounds, our compound is preferentially binding to the PHD pocket where HIF binds. So our compounds are HIF inhibitor.

Initial studies with these inhibitors of oxygen sensor erudite very provocative results. Local administration into a sponge implanted under the mouse’s skin enhanced angiogenesis and increased the haemoglobin content.

In addition, they protected neurons from ischemic injuries in a model of acute – mouse model. In gerbil mice transient ischemia followed by reperfusions resulted in neuronal death but damage was prevented significantly by our inhibitors of oxygen sensor.

Unfortunately, recent demonstration that both erythropoietin and VEGF independently accelerate the diabetic retinopathy should call for caution but we are very fortunate that there are three isoforms in PHD. The respective role of each PHD isoform has been successfully elucidated.

The angiogenesis and erythropoiesis are primarily regulated by the PHD2 isoform and it was just recently shown that the specific disruption of the PHD1 isoform induces hypoxia tolerance by reprogramming basal oxygen metabolism. This later adaptation to oxygen conservation provides protection against hypoxia.

A variable of PHD inhibitors, the inhibitors of the oxygen sensors are listed in the latest Nature Review, an article provided by Peter Carmeliet. Our compounds are also kindly introduced as novel inhibitors. However, current available PHD inhibitors are not specific to PHD1 isoform.

So, given the specific PHD1 inhibitor it may be a very interesting candidate for future therapy in diabetic nephropathy.

It might provide a novel therapeutic concept, i.e. hibernation therapy. To prevent ischemic cells from apoptosis by preserving them under the hibernation status until the sufficient oxygen supply to occur just like preserving the cells in liquid nitrogen.

The second factor, aetiological factor that relates to oxidative stress and advanced glycation another derangement during oxygen metabolism.

To document their consequences we have quantified in human diabetic renal tissues the level of oxidatively modified proteins.

Here advanced glycation end products. As you see, oxidative stress-dependent AGE structures such as pentosidine here appearing the so-called glycosylation products are detected remarkably in human diabetic kidneys where as oxidation-independent glycation products here appearing were virtually undetectable. These findings suggest an enhanced oxidative stress and its resultant, advanced glycation in human diabetic nephropathy.

Can we modify this process? We previously demonstrated that angiotensin receptor inhibitors inhibit indeed the in vitro formation of AGEs and various other types of anti-hypertension agents such as calcium channel blockers and beta blockers lack this possibility to inhibit AGEs in vitro.

Very interestingly the renal AGE content correlates significantly with proteinuria in our diabetic rat whatever the type of given anti-hypertensive agent.

So angiotensin receptor blockers protect the kidneys by lowering blood pressures and inhibiting the renin-angiotensin system but its renal benefit may be partly mediated by its inhibitory effect on advanced glycation and oxidative stress.

We would like to know the respective role of the benefit of anti-AGE formation and the anti-oxidative stress properties of ARB.

So we have tried to dissociate these properties. We designed to develop novel ARB derivatives with a marked inhibitory effect on AGE formation and oxidative stress but with a minimal affinity to angiotensin II receptor and thus little anti-hypertensive effect. More than 100 olmesartan derivatives were assayed for AGE inhibition and affinity to angiotensin II receptors. After synthesis of the novel derivatives the toxicity and the pharmacological parameters we eventually identified one hit compound named here R147176.

This is a collaboration of the Daiichi Sankyo pharmaceutical companies founded by the Japanese government. Our compound shares a common core structure of known sartans. The upper panel demonstrates the structure of our compound, each shares a common core structure with olmesartan, losartan, valsartan and candersatan.

This slide summarises the angiotensin II receptor affinity, as well as the AGE inhibitory effect of two sartans; olmesartan and losartan and our compound. Losartan is the first sartan that appeared in clinical medicine. Olmesartan is the latest one. As you see, in this panel olmesartan has an angiotensin affinity but not a magnitude higher than losartan. Two other magnitudes higher than our compound. On the other hand, our compound is the most effective in vitro inhibitor of AGE formations.

So, now can we prevent with these two diabetic nephropathies independently of blood pressure? The answer is apparently yes, given to hypertensive type 2 diabetic rats SHR/ND it reduces proteinuria and improved the morphological abnormalities with a modification of the diabetic nephropathy.

Let me summarise my talk. The prevention of diabetic nephropathy will rely on a multi-pronged approach. Current therapies interfering with the classical pathways deriving from hemodynamics and metabolics may be not insufficient to fully prevent diabetic renal complications underlying the need for novel agents able to interfere with biochemical alterations. Only time will tell us if the renewed approach postulated from the study of animal experiments can be extended to humans. Thank you very much for your attention.

Chairman: Thank you very much for sharing with us this very interesting and provocative data. The lecture is open for discussion. Are there any questions?

Question: I was just interested, you obviously developed novel PHD inhibitors to maintain expression of HIF, yet there is some recent literature to suggest that HIF has dual effects some of which may be positive and some of which may be detrimental particularly to the kidney. Would you mind commenting on that?

Prof. Miyata: So far we have administered cobalt. We did not observe any detrimental effect but we do not know the long-term toxicity, as well as the long term adverse effects. So you are completely right. We do not know the long-term toxicity and adverse effects at this moment.

Question: I take the opportunity to ask you whether you had a chance to test your PHD1 specific inhibitor in animal models to see whether it’s able to prevent the classic lesions in the glomeruli in this animal.

Prof. Miyata: As I mentioned, so far available PHD inhibitors are not PHD1 specific and our compound could protect ischemic injury in the apoptotic model but we have never tried in a diabetic model due to the tissue penetrations. So once we have PHD1 specific inhibitors maybe we could test that very important possibility. Thank you very much.

Question: I might just ask another follow on question then. With regards to your inhibition of AGE formation, I think some of the problems have come because they’re used too late in the development of nephropathy. Have you got a model that actually has established diabetic nephropathy and does it actually show any benefit by preventing AGE formation as opposed to cross link inhibitors that might actually be disrupt pre-formed AGEs?

Prof. Miyata: Yes in animal models there is a lot of evidence. Indeed we have generated triple transgenic mice of AGE ions and – it exaggerates diabetic renal lesions and this glomerular damage has been protected by AGE inhibitors; aminoguanadine, pyridoxamine and currently the human clinical studies for pyridoxamine are in progress so there is some evidence.