Hepato-selective dihydroquinolizinones active against hepatitis A virus in vitro and in vivo

Abstract

Despite the considerable clinical and economic burden imposed by hepatitis A virus (HAV) infection, both globally and in U.S., there are currently no available antiviral therapies for the treatment of type A hepatitis. Here we describe novel third-generation hepato-selective dihydroquinolizinones (HS-DHQs) with cellular uptake mediated by transport via hepatocyte-specific solute organic anion transporter family members 1B1 and 1B3 (OATP1B1-B3). The lead HS-DHQ compound, HS83128, demonstrates robust inhibition of the host cell TENT4A/B terminal nucleotidyltransferases required for efficient HAV RNA synthesis (IC₅₀ 6-25nM), and potent antiviral activity against HAV in cell culture (EC₅₀ 0.6 nM). Pharmacokinetic studies in CD-1 mice receiving comparable oral doses of HS83128 and a first-generation dihydroquinolizinone, RG7834, revealed a 5-fold increase in intrahepatic drug concentration and more than 10-fold improvement in liver versus nervous system tissue selectivity. Twice-daily oral administration of HS83128 rapidly arrested viral replication in HAV-infected Ifnar1^-/- mice, reducing fecal virus shedding and cytokine markers of hepatic inflammation and reversing virus-induced liver injury. The hepato-selective nature of HS83128 may reduce the risk of neurologic and reproductive track toxicities observed with long-term administration of other dihydroquinolizinones, making it a candidate for the first antiviral therapy of hepatitis A.

Keywords: Antiviral drug; Hepatovirus; OATP1 transporter; PAPD poly(A) polymerase; Pharmacokinetics; Prophylaxis; TENT4 terminal nucleotidyltransferase; Therapy.

Fig. 1.

A. Structures of RG7834, HS65042, HS79186 and HS83128. B. upper: genome organization of HAV/18f-NLuc reporter virus; lower: antiviral activities of RG7834, HS65042, HS79186 and HS83128 against 18f-NLuc virus in cells. Error bars denote SEM (n = 3).

Fig. 2.

A. a polyadenylation reaction with recombinant TENT4A or TENT4B. ATP consumption during the reaction is measured by a luciferase assay as the surrogate of TENT4 enzymatic activity. B-D: effect of RG7834, HS79186 and HS83128 on the enzymatic activities of TENT4A and TENT4B measured as in A. Error bars denote SEM (n = 2).

Fig. 3.

Tissue distribution of (left) RG7834 and (right) HS83128 following p.o. administration of 10 mg/kg of each drug to CD-1 outbred mice. See also Table 2.

Fig. 4.. HS79186 blocks HAV replication and reverses hepatitis in infected Ifnar1^−/− mice.

(A) Experimental scheme to determine antiviral activity of HS79186 in mice. Cohorts of Ifnar1−/− mice (n = 4) were challenged with 2 × 10⁶ GE of HAV-mp7 inoculum and then treated with HS79186 at 25 μmol/kg given by gavage t.i.d. between days 4.5–8 post-infection (p.i.), or mice were given vehicle during this time. Viral RNA in fecal pellets and liver, as well as liver histology and blood ALT activity were assessed at day 8 p.i. (B) Serum ALT activities, (C) fecal HAV shedding, and (D) intrahepatic HAV RNA on day 8 p.i. (E) Representative images of H&E-stained sections of liver harvested from mice treated with (top) vehicle only or (bottom) HS79186 at necropsy on day 8 p.i. Arrows indicate apoptotic hepatocytes. (F) Liver pathology scores on a scale of 1–10. (G) Transcript levels determined by RT-PCR in livers on day 8, relative to levels expressed in naïve mice.

Fig. 5.. HS83128 treatment of acute hepatitis A in mice.

(A) Experimental scheme for the HS83128 experiment. Ifnar1−/− mice were challenged with 2 × 10⁶ GE HAV-mp7. From days 4–8 p.i., mice received either HS83128 at 25 μmol/kg per os (p.o.) q.d., or they received vehicle only. (B) Serum ALT activities and (C) fecal HAV RNA quantified by qRT-PCR at days 4 or 8 p.i. (D) Intrahepatic HAV RNA on day 8 p.i. (E) Interferon-β and proinflammatory chemokine transcript levels determined by RT-PCR in livers on day 8, relative to levels expressed in naïve mice. (F–K) Cohorts of mice (n = 5) were challenged with 2 × 10⁶ GE HAV-mp7 and treated with HS83128 or RG7834 (25 μmol/kg; p.o. b.i.d.), or vehicle during days 6–15 p.i. The virological and inflammatory responses in these mice were tracked during and after treatment. (F) (top) Serum ALT activities and (bottom) fecal HAV RNA quantified by qRT-PCR in Ifnar1^−/− mice treated with HS83128 or RG7834 or vehicle only between days 6 and 15 p.i. with HAV. (G) Intrahepatic HAV RNA at necropsy on day 33 p.i. (H) Representative images of H&E-stained liver sections at necropsy on day 33 p.i. Black arrows indicate apoptotic hepatocytes; white arrow, extramedullary hematopoiesis. (I) Pathology scores on a scale of 1–10. (J) Cytokine transcript levels determined by RT-PCR in livers on day 33, relative to levels expressed in naïve mice. (K) Body weight changes.(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 6.. Chemoprevention of hepatitis A in Ifnar1−/− mice.

(A) Cohorts of male mice (n = 5 each) were infected with a high-dose HAV inoculum (10⁷ genome equivalents, GE), one day after starting treatment with HS83128, 25 μmol/kg given by gavage once daily (q.d.), or vehicle control. Starting on day 7 post-infection (p.i), dosing was increased to twice a day (b.i.d.). Treatment was discontinued on day 15, and animals necropsied on day 33 p.i. (top) Serum ALT activities and (bottom) fecal HAV RNA quantified qRT-PCR between day 4 and day 33 p.i. (B) Liver pathology was scored on a scale of 1–10 according to the number of inflammatory foci and apoptotic hepatocytes present in random 20× fields of view.