The effect of therapeutic administration of selective 5HT2A and B antagonists on the development of liver fibrosis in a mouse carbon tetrachloride (CCl4) liver injury model.

KK Changani, M Fulleylove, JL Morrell, G Vitulli, M Bergeal, MG Lennon, MV Haase, DEC Shannon, VJ Barrett, H Hobbs, CA Lovatt, F Chowdhury, J Denyer. GlaxoSmithKline, Stevenage, UK

Introduction: Serotonin has been implicated as a key driver of liver fibrosis in rodents, acting via 5-HT2 activation in the hepatic stellate cell. In house data from recent in vitro studies of isolated mouse hepatic stellate cells suggested that 5-HT2 receptors activate phospho-ERK responses with a pharmacological profile consistent with the predominant involvement of 5-HT2A receptor (unpublished data). The current study was conducted to explore the effects of selectively antagonising 5-HT2A, using volinanserin, or 5-HT2B using 6-((4,4-difluoropiperidin-1-yl)sulfonyl)-1-piperidin-4-yl)-1H-indole hydrochloride, in a mouse model of liver fibrosis (carbon tetrachloride induced (CCl4)) to better understand the selectivity profile required for anti-fibrotic 5-HT2 drug development. The 1-piperidinyl-6-piperidinylsulfonylindoles are known 5-HT2b receptor antagonists with good selectivity over the other 5-HT2 isoforms1. Due to its favourable pharmacokinetics in the rat 6-((4,4-difluoropiperidin-1-yl)sulfonyl)-1-piperidin-4-yl)-1H-indole hydrochloride1 was selected from this series. The effect on fibrosis development in liver was determined using a histological marker of collagen deposition, percentage picrosirius red area stained (% PSR) in liver tissue following 8 weeks exposure to CCl4.

Methods: C57Bl/6 mice (n=63) were divided into Groups 1 to 6. Each group was subdivided into two in order to stagger the start date of each sub group by one week. All animals were dosed intraperitoneally twice weekly with CCl4 (2µl/gram of body weight, 1:3; CCl4;olive oil) for up to 8 weeks. Group 2 animals were culled at the end of week 3 to determine the level of fibrosis prior to compound administration. All other animals were surgically implanted with minipumps (ALZET 2006, flow rate 3.6 µL/day) that had been loaded with: Group 1: compound vehicle (DMSO : PEG200 : water 10:45:45 (v/v/v); Group 3 and 4: Volinanserin 0.06 and 0.32mg/kg/day, respectively, Group 5 and 6: 6-((4,4-difluoropiperidin-1-yl)sulfonyl)-1-piperidin-4-yl)-1H-indole hydrochloride 0.15 and 1.53mg/kg/day, respectively. The dose levels were selected in order to achieve between 90-99% receptor occupancy as determined from the functional pKi s determined in cells recombinantly expressing the mouse receptors and exposure (unbound) from a pilot PK study. Blood samples were taken from animals throughout the in life phase and as a terminal procedure to determine drug exposure levels. At the end of the study two lobes were removed to determine tissue PK and histological assessment of % PSR.

Results: Liver from animals dosed with volinanserin (0.06 and 0.32mg/kg/day) showed no significant difference in % PSR area staining compared with vehicle at week 8. However, livers from animals dosed with the 5-HT2B antagonist had significantly increased % PSR area at both doses compared to the 8 week vehicle (Dunnetts adjusted p-value=0.0445 for 0.15mg; Dunnetts adjusted p-value=0.0036 for 1.53mg).

Neither volinanserin or the 5-HT2B antagonist produced an anti-fibrotic effect as determined by %PSR area in liver sections at blood concentrations consistent with over 90% receptor occupancy based on in vitro potency and unbound drug levels.

Conclusions: The data from this study does not support 5-HT2 antagonism as an anti-fibrotic therapy.

1. Bruton G et al, PCT Int. Appl. WO 2009016225 A1 20090205, 2009

All animal studies were ethically reviewed and carried out in accordance with Animals (Scientific Procedures) Act 1986 and the GSK Policy on the Care, Welfare and Treatment of Animals.