ALBERT

Description: Abstract 4244 Introduction: Long-term storage iron (hemosiderin) is insoluble and not directly accessible to iron chelation. Ascorbate is a potent facilitator of redox cycling and facilitates the mobilization of cellular iron stores in patients chelated with deferoxamine. Ascorbate deficiency is quite common in thalassemia major and may create a phenotype of relative chelator refractoriness. However, the synergistic role of ascorbate supplementation has never been demonstrated for other chelators, including the oral chelator deferasirox. Ascorbate can be freely synthesized in many animals, making them unsuitable models to address this question. The osteodystrophic syndrome (ODS) rat is a spontaneous mutant lacking L-gulono-gamma-lactone oxidase, the initial enzyme in ascorbate synthesis. Iron loading and chelation have never been performed in ODS rats, so we report a pilot study aimed at optimizing organ iron loading and clearance. Methods: Twenty-six 8-week old male ODS rats were maintained vitamin C replete by supplying ascorbate in the chow or drinking water. Fourteen animals were iron loaded using a combination of 0.4% TMH ferrocene and 3% carbonyl iron in standard rodent chow. Twelve animals were loaded with iron dextran (200 mg/kg/week) by subcutaneous injection. To characterize loading kinetics, two animals were analyzed at five, ten, and twelve weeks of oral iron loading as well as five and ten weeks of iron dextran injections. Sixteen animals were iron loaded for ten weeks (half enterally, half parenterally) and transitioned into the chelation arm. Iron chelation was performed with oral deferasirox at 75 mg/kg/dose, once daily five times per week. Chelation was administered for six and twelve weeks in six animals, divided equally between parenteral and enteral iron loading; sham chelation was given to the remaining four animals. Following euthanasia, H&E and iron staining were performed and tissue iron was quantified by atomic absorption. Results: Parenteral iron loading was well tolerated and yielded liver iron concentration (LIC) values of approximately 3 and 4.5 mg/g wet weight (∼11-16.8 mg/g dry weight) at five and ten weeks, respectively. Iron loading was predominantly reticuloendothelial, with little parenchymal redistribution. Spontaneous iron loss after 12 weeks of iron chelation was modest at 1.3% per week. In contrast, oral iron loading with the combination of TMH-ferrocene and carbonyl iron was poorly tolerated. Animals developed severe diarrhea and required fluid replacement and frequent dose reductions. These GI disturbances gradually lessened over a period of four weeks and animals received approximately 80% of the targeted iron dose. LIC values were lower at five weeks (2.2 mg/g ww) but nearly equivalent by 10 weeks (4.2 mg/g ww). Iron was entirely parenchymal, with little reticuloendothelial deposition. Spontaneous iron losses after 12 weeks of iron chelation were strikingly higher than for parenteral iron loading, measuring 4.7% per week. Iron chelation with deferasirox was well tolerated in both groups. Iron chelation was less efficient in iron dextran loaded animals. LIC was 3.1 ± 0.7 mg/g ww compared with 3.8 ± 0.2 mg/g ww in sham chelated animals, p=0.25. In contrast, deferasirox treatment nearly completely cleared liver iron in TMH treated animals 0.4 ± 0.1 mg/g ww vs 1.5 ± 0.04 mg/g ww, p = 0.001, with most of the residual iron located primarily in reticuloendothelial store on histologic analysis. Discussion: Iron loading in the ODS rat can be performed with either iron dextran or TMH-ferrocene but the characteristics of iron staining, spontaneous iron loss, and chelator accessibility are completely different. Iron dextran loads the reticuloendothelial system. Iron redistribution to parenchymal tissues is sluggish, based upon the low spontaneous iron elimination rate and modest response to deferasirox therapy (1.5% per week). In contrast, TMH produced exclusively parenchymal loading, high spontaneous iron losses (4.7% per week) and more vigorous response to chelation (6.3% per week). Subsequent studies will determine whether the degree of ascorbate sufficiency modulates deferasirox efficacy in animals with primarily reticuloendothelial iron loading. Disclosures: Nick: Novartis: Employment. Wood:Novartis: Research Funding; Ferrokin Biosciences: Consultancy.

Description: Introduction: Deferasirox (ICL670) is a novel tridentate oral iron chelator currently being evaluated for the treatment of transfusional iron overload. Phase III clinical trials have demonstrated that once-daily ICL670 (20 mg/kg) is equally effective at controlling liver iron concentration as standard deferoxamine therapy (40 mg/kg/day, 5 days per week). While ICL670’s long serum half-life should offer good protection against cardiac iron accumulation, little is known regarding its ability to remove stored cardiac iron. Therefore, we compared the relative efficacy of ICL670, deferiprone (L1), and deferoxamine (DFO) in removing cardiac iron from iron-loaded gerbils. Methods: 37 8–10 week old female gerbils underwent ten weekly iron dextran injections of 200 mg/kg/D, followed by a 13 day equilibration period. Five animals were then sacrificed to determine pre-chelation iron burdens. Chelation was initiated in 3 groups of 8 animals (ICL670 100 mg/kg/D po QD, L1 375 mg/kg/D po divided TID, DFO 200 mg/kg/D sub Q divided BID) five days per week and maintained for 12 weeks. The remaining 8 animals received sham chelation. All animals underwent ECG and treadmill assessment at baseline, following iron loading, and after completing chelation therapy. Animals were sacrificed for liver and heart iron measurement (Mayo Medical Laboratory) and semiquantitative histology. Hearts were evaluated for iron loading/distribution, tissue fibrosis, and myocyte hypertrophy, while livers were scored for iron loading/distribution and fibrosis. Results: Chelator-independent iron excretion and redistribution was evident, unlike in humans. Cardiac and liver iron contents fell 30.4% and 23.2%, respectively, with sham chelation; all subsequent chelator comparisons are reported with respect to the sham-chelated animals. ICL670 reduced cardiac iron content 20.5%. There were no changes in cardiac weight, myocyte hypertrophy, fibrosis, or wet-to-dry weight ratio. ICL670 treatment reduced liver iron content 51%. Iron elimination was greatest in hepatocytes with no detectable Kupfer-cell iron clearance. L1 produced comparable reductions in cardiac iron content (18.6%). Wet weight cardiac iron concentration fell nearly 30% but this was offset by greater cardiac mass (16.5% increase). Histologic analysis demonstrated decreased iron staining but increased myocyte hypertrophy. L1 decreased liver iron content 24.9%. Wet weight liver iron concentration fell 43.8% but was offset by a 30% increase in liver weight and water content. Iron elimination was balanced between Kupfer cells and hepatocytes. DFO did not reduce biochemically-assayed cardiac or liver iron content, although it improved histologic iron scores in both organs. Hearts from DFO treated animals were enlarged and had greater fibrosis. Cardiac and liver iron contents were closely correlated (r = 0.66), but ICL670 animals had lower hepatic iron contents for any given cardiac iron content. Iron loading broadened QRS duration by 10.6%; this effect was antagonized by both L1 and ICL670 therapy. PR, QRS, and QTc interval were weakly correlated with cardiac and liver iron contents. Treadmill exercise time was independent of chelation therapy. Conclusion: ICL670 and L1 were equally effective in removing stored cardiac iron in a gerbil animal model but ICL670 removed more hepatic iron for a given cardiac iron burden.

Description: HIV-1 replication is induced by the excess of iron and iron chelation by desferrioxamine (DFO) inhibits viral replication in HIV-1 infected CEM T cells [1]. Treatment of cells with DFO or 2-hydroxy-1-naphthylaldehyde isonicotinoyl hydrazone inhibits expression of proteins that regulate cell-cycle progression, including cycle-dependent kinase 2 (CDK2) [2]. HIV-1 transcription is activated by Tat protein, which recruits transcriptional co-activators to the HIV-1 promoter. Elongation of HIV-1 transcription is mediated by the interaction of HIV Tat with host cell cycle-dependent kinase 9 (CDK9)/cyclin T1, which phosphorylates the C-terminal domain of RNA polymerase II. Our recent studies showed that CDK2 participates in HIV-1 transcription by phosphorylating Tat [3]. Thus inhibition of CDK2 by iron chelators might present a new approach to inhibit HIV-1 transcription. We evaluated the effect of a clinically approved orally effective iron chelator, 4-[3,5-bis-(hydroxyphenyl) -1,2,4-triazol-1-yl]-benzoic acid (ICL670 or deferasirox) on HIV-1 transcription. ICL670 inhibited Tat-induced HIV-1 transcription in CEM, 293T and HeLa cells at concentrations that did not induce cytotoxicity. The chelator decreased cellular activity of CDK2 but not its protein level and reduced HIV-1 Tat phosphorylation by CDK2. ICL670 did not decrease CDK9 protein level but significantly reduced association of CDK9 with cyclin T1 and reduced phosphorylation of Ser-2 residues of RNA polymerase II C-terminal domain. In conclusion, our findings add to the evidence that iron chelators may inhibit HIV-1 transcription by deregulating CDK2 and Cdk9. Further consideration should be given to the evaluation of ICL670 for future anti-retroviral therapeutics and to the development of iron chelators specifically as anti-retroviral agents.

Description: Introduction: Transfusional therapy for thalassemia major and sickle cell disease can lead to iron deposition and damage to the heart, liver, and endocrine organs. Iron causes the MRI parameters T1, T2, and T2* to shorten in these organs, creating a potential mechanism for iron quantitation. Validation of liver MRI has been achieved by studying patients undergoing clinically indicated liver biopsy. However, because of the danger and variability of cardiac biopsy, cardiac MRI studies have relied upon “clinical” validation, i.e., the association between low cardiac T2* and cardiac function. In this study, we demonstrate that iron produces similar T1, T2, and T2* changes in the heart and liver, using a gerbil iron overload model. Methods: Twelve gerbils underwent iron dextran loading (200 mg/kg/week) from 2–14 weeks; 5 age-matched controls were studied as well. Animals had in-vivo assessment of cardiac T2* as well as hepatic T2 and T2* using a General Electric 1.5 T CVi system with custom isofluorane anesthesia delivery system, imaging enclosure, coil and pulse sequences. Liver and heart were harvested following imaging, weighed, and portions collected for histology and quantitative iron (Mayo Metals Laboratory, Rochester, MN). Ex-vivo cardiac and liver T1 and T2 measurements were performed on fresh specimens (〈 30 minutes post-sacrifice) using a Bruker minispectrometer. Results: Control animals had minimal detectable iron at baseline and did not accumulate iron in the liver or the heart over the 14-week study interval. Chemically-assayed heart iron concentration increased 0.078 mg/g(wet wt)/wk (r2=0.98) and iron content 0.022 mg/wk (r2=0.92) by linear regression analysis. Similarly, assayed liver iron concentration increased 1.15 mg/g(wet wt)/week (r2=0.93) over a 10 week interval and liver iron content increased 3.82 mg/wk (r2=0.96). Liver iron deposition was prominent in both sinusoidal cells and hepatocytes. Interstitial fibrosis was mild and there was no necrosis. Cardiac iron deposition was predominantly endomysial, generally sparing the myocytes themselves. Interstitial fibrosis was prominent, originating from areas of high iron concentration. No myocyte necrosis was observed, however myocyte hypertrophy was evident at high iron concentrations. Cardiac and liver R2* (1/T2*), R2 (1/T2), and R1 (1/T1) rose linearly with tissue iron concentration (r2 averaged 0.94 [0.74 to 0.98]. The slope of these parameter with respect to iron was15–29% steeper in heart than in liver, although these differences reached statistical significance only for R2. Systematic differences in wet-to-dry weight ratio between heart and liver (5.07 vs 3.82) antagonized this effect, however, such that calibrations were similar on a dry-weight basis. Conclusion: Cardiac iron is the primary determinant of cardiac MRI relaxivity. Calibration curves were similar between heart and liver on a dry weight basis. Extrapolation of liver calibration curves to heart may be a rationale approximation in humans where direct tissue validation is difficult and dangerous. Regardless of systematic differences in absolute calibration, these data support prior claims that cardiac R2 and R2* measurements reflect cardiac iron concentration