Acetylation level was normalized to H3 and H4 expression level. (C) and (D) Quantification of (A) and (B) respectively. **P,0.01, *P,0.05 versus vehicle (veh). n = four per treatment. Error bars indicate SEM.oral bioavailability and very minimal CNS exposure for compound 4b. Despite an unfavourable predicted in vivo pharmacokinetic profile, we pursued further in vivo studies in light of the published positive efficacy study [30]. Surprisingly, 4b did not dissolve in the previously described formulation [30]. Unlike SAHA, which dissolves in cyclodextrins upon heating and is stable for seven days at room temperature [39], investigation of both free-base and salt forms of 4b showed that neither was capable of full dissolution to 1 mg/mL as previously described. However, the physicochemical instability of the acidic salts of 4b, combined with the efficient and irreversible conversion of the parent compound to the inactive benzimidazole C1, might explain the disparity between the published studies and our data. The benzimidazole C1 is readily soluble in aqueous media and the conversion process is exacerbated under acidic conditions and at elevated temperature. Hence, it is likely that there was very significant contamination of C1 in the chronic drinking water preparations previously used. In agreement with our in vitro ADME predictions, 4b reaching the systemic circulation was rapidly cleared, which resulted in a plasma half-life of approximately 40 min after subcutaneous dosing.
Subcutaneous dosing resulted in much higher relative concentrations in plasma and brain tissue as the impact of first pass metabolism and efflux by MDR1 was reduced via this route. To put these results in the context of the previous study [30], the subcutaneous administration is easier to interpret, since both studies used 4b in soluble form for acute administration, which largely sets aside any salt versus free-base stability disparities. Indeed, a comparative stability analysis of 4b and 4bNTFA prepared at 1 mg/ml in DMSO showed both are soluble, but the TFA salt again showed signs of degradation (day 7 values being 59% of the value measured at day 1 after formulation). According to Thomas et al, verification of an acute pharmacodynamic response (the reversal of transcriptional dysregulation in the R6/2 model) was achieved by the repeated once daily injection of 150 mg/kg 4b in 50:50 DMSO: PBS for three consecutive days. Assuming linear kinetics and based on our values when 4b was dosed at 4.22 mg/kg (s.c), the brain Cmax achieved in that study would be in the region of 4? mM ( = (150 mg/kg/4.22 mg/kg)* Cmax of 134 nM)), sufficient to potentially fully inhibit Class I HDAC cellular activity, based on our cellular in vitro potency values (Fig. 1), and consistent with the positive results obtained. However, given the $12 fold lower exposures shown here from the oral administration versus the sc route, we would expect to achieve a brain Cmax of only ,72 nM ( = (150 mg/kg/50 mg/kg)* Cmax of 24 nM)) via equivalent oral bolus administration of 150 mg/kg. This estimated Cmax is approximately 25 fold lower than the most potent cellular HDAC in vitro IC50 value we measured (1.8 mM). With the slower continuous administration of 4b via a drinking water study (even with the best case assumption that the 4b was stable and fully dissolved in the drinking water), we would expect the Cmax achieved to be significantly lower. The lack of any pharmacodynamic response predictive of central Class I HDAC inhibition when 4b was dosed orally at 150 mg/kg twice daily for 5 days confirmed this prediction. In conclusion, our findings demonstrate that the physicochemical properties, metabolic and P-glycoprotein substrate liabilities of 4b render it unsuitable as a molecular tool to investigate central Class I HDAC inhibition in vivo in mouse by oral administration. In the pivotal proof of concept trial [30], 4b was given to R6/2 mice in drinking water, leading to improved behavioral phenotypes. We conclude that this is highly unlikely to be due to HDAC inhibition in the CNS or related to the finding of the reversal of transcriptional dysregulation detected in the acute pharmacodynamic trial [30], where brain concentration of 4b was most likely ,65 fold higher. Our results cast serious doubts on the validation of CNS HDAC3 as a target for the treatment of HD. Our findings are consistent with Moumne et al [40], who demonstrated that a ?genetic cross of Hdac(+/2) heterozygotes with R6/2 mice effectively reduced nuclear HDAC3 levels, but did not ameliorate physiological or behavioural phenotypes and had no effect on molecular changes including dysregulated transcripts. We cannot rule out that a metabolite of 4b or C1 was responsible for the therapeutic benefit seen in R6/2 mice per oral dosing of 4b in the previous study. To our knowledge, no data has been published on the ADME properties of 4b or related compounds used in the FRDA mouse models. Our results underscore the absolute necessity for appropriate ADME evaluation of compounds prior to in vivo target validation.
Materials and Methods Ethics Statement
All animal work was conducted according to relevant national and international guidelines. The pharmacokinetic analysis of HDACi 4b was conducted at Charles River Laboratories (CRL), which is subject to legislation under the Animal Welfare Act. At CRL all animal studies are governed by CRL’s Institutional Animal Care and Use Committee (IACUC). CRL is fully accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC), and is registered with the United States Department of Agriculture (USDA). All procedures involving animals were conducted humanely and were performed by or under the direction of trained and experienced personnel. The protocol was reviewed and approved by the IACUC of CRL prior to study initiation. The veterinarian was consulted in the overall study design for this study type. The pharmacodynamic evaluation of HDACi 4b was approved by the King’s College London Ethical Review Panel and experimental procedures were performed in accordance with the UK Home Office regulations.