Dual mRNA therapy restores metabolic function in long-term studies in mice with propionic acidemia

Abstract

Propionic acidemia/aciduria (PA) is an ultra-rare, life-threatening, inherited metabolic disorder caused by deficiency of the mitochondrial enzyme, propionyl-CoA carboxylase (PCC) composed of six alpha (PCCA) and six beta (PCCB) subunits. We herein report an enzyme replacement approach to treat PA using a combination of two messenger RNAs (mRNAs) (dual mRNAs) encoding both human PCCA (hPCCA) and PCCB (hPCCB) encapsulated in biodegradable lipid nanoparticles (LNPs) to produce functional PCC enzyme in liver. In patient fibroblasts, dual mRNAs encoded proteins localize in mitochondria and produce higher PCC enzyme activity vs. single (PCCA or PCCB) mRNA alone. In a hypomorphic murine model of PA, dual mRNAs normalize ammonia similarly to carglumic acid, a drug approved in Europe for the treatment of hyperammonemia due to PA. Dual mRNAs additionally restore functional PCC enzyme in liver and thus reduce primary disease-associated toxins in a dose-dependent manner in long-term 3- and 6-month repeat-dose studies in PA mice. Dual mRNAs are well-tolerated in these studies with no adverse findings. These studies demonstrate the potential of mRNA technology to chronically administer multiple mRNAs to produce large complex enzymes, with applicability to other genetic disorders.

Fig. 1. Dual mRNAs encode functional PCC with proper mitochondrial localization in patient-derived fibroblasts and human Hep3B cells.

a–c Comparison of co-transfection of hPCCA+hPCCB (dual) mRNAs with single transfection of hPCCA or hPCCB mRNA alone. Human fibroblasts isolated from a PCCA-deficient PA patient and a PCCB-deficient PA patient (n = 2 per condition) were transfected with 0.5 or 1 µg of the total dual mRNAs (1:1 molar ratio), hPCCA mRNA alone, or hPCCB mRNA alone, or 1 μg of eGFP control mRNA for 24 h. Cells were lysed, and mitochondrial matrix fractions were collected to assess PCC enzymatic activity by a radiometric assay (a), and protein levels of PCCA (b) and PCCB (c) by capillary electrophoresis. d Identification of the optimal molar ratio of hPCCA mRNA:hPCCB mRNA. Patient fibroblasts were transfected with 0.67 nM (1 µg) of total dual mRNAs across a range of molar ratios of hPCCA mRNA:hPCCB mRNA or with the same molar concentration of eGFP mRNA (n = 4 per condition) for 24 h. PCC activity in the cellular mitochondrial matrix fractions was measured. Data are shown as mean ± SEM. e, f Subcellular localization of dual mRNAs-encoded PCC subunits in PCCA-deficient patient fibroblasts and human liver-derived Hep3B cells. Cells were transfected with 25 ng of dual mRNAs or luciferase (Luc) control mRNA for 24 h, and were stained for PCCA, PCCB, and TOM20, a mitochondrial marker. Scale bars are 20 µm. Representative images are shown from n = 24 replicates.

Fig. 2. Kinetics of dual mRNAs-encoded hepatic PCC and comparison of dual mRNA therapy with carglumic acid in PA mice.

a, b Kinetics of dual mRNAs-encoded hepatic PCC subunits and enzyme activity. PA hypomorphic mice of mixed gender were administered a single IV bolus injection (1 mg/kg) of dual mRNAs (n = 4 mice/time point) or Luc control mRNA (n = 3 mice/time point) encapsulated in LNPs and were sacrificed at various time points. a Absolute quantification of hPCCA and hPCCB proteins in livers from dual mRNAs-injected PA mice by LC-MS/MS using signature peptides for wild-type (WT) hPCCA and hPCCB protein sequences. hPCCA and hPCCB were below the LLOQ in all livers from Luc control mRNA-injected mice. b PCC enzyme activity in hepatic mitochondria was measured. Shaded bars represent the range of endogenous PCC enzyme activity levels from all Luc mRNA-injected mice across all time points (n = 21). c–e Substantial reductions in plasma ammonia and primary disease biomarkers due to dual mRNAs in contrast to carglumic acid. PA hypomorphic female mice were administered a single IV bolus dose of Tris-sucrose buffer (n = 11) or 1 mg/kg of dual mRNAs or Luc control mRNA (n = 12) encapsulated in LNPs and were sacrificed 7 days post injection. Additional PA hypomorphic female mice were administered repeat doses of 1% carboxymethyl cellulose (CMC, n = 11) or carglumic acid (1200 mg/kg/day, twice daily, n = 12) via oral gavage for 7 consecutive days and then sacrificed. Unaffected Pcca^+/− mice (n = 9) were IV injected with Tris-sucrose buffer as control. Mice were bled at study days 0 (pre-treatment), 2, and 7. Plasma ammonia (c) and liver 2MC (e) at day 7 upon sacrifice, and plasma 2MC (d) were assessed. Data are shown as mean ± SEM. Plasma and liver 2MC levels in unaffected mice were <LLOQ. P values were obtained from Tukey’s post hoc pairwise multiple comparison test following a one-way or two-way ANOVA, and are provided in the source data. ***P < 0.001 comparing dual mRNA therapy with carglumic acid. ⁺⁺⁺P < 0.001 comparing dual mRNAs or carglumic acid group with control groups (1% CMC, Tris-sucrose, and Luc mRNA). Pre, pre-treatment on day 0.

Fig. 3. Long-term efficacy in 3- and 6-month repeat-dose studies in PA hypomorphic mice.

a–d 3-month study showed sustained and reproducible biomarker reductions due to dual mRNAs. PA hypomorphic mice of mixed gender were IV bolus administered 0.5 or 2 mg/kg dual mRNAs or 2 mg/kg Luc control mRNA or Tris-sucrose buffer (n = 12/group/sacrifice time point) every 3 weeks for 12 weeks. Mice were sacrificed 2 days after the first dose (day 2) and 2 days after the last dose (day 86). PCC enzyme activity in liver mitochondria (a), liver 2MC (b), and plasma ammonia (female mice only as male PA mice do not exhibit elevated ammonia levels) (c) were assessed on days 2 and 86. d Plasma primary disease biomarkers were assessed weekly throughout the 3-month study. Biomarker levels in unaffected heterozygote mice (n = 24) were <LLOQ for plasma 2MC and 3HP, and <0.13 for plasma C3/C2 ratio at all time points. e, f Amelioration of heart weight/body weight ratio and plasma CTNT levels in the 6-month study. PA hypomorphic female mice received a total of six IV bolus doses of 0.5 or 1 mg/kg dual mRNAs or 1 mg/kg Luc control mRNA (n = 6/group) at weeks 0, 7, 11, 15, 20, and 24. Mice were sacrificed 2 days after the last dose. Age-matched untreated female WT mice (n = 6) were added to the end of the study at sacrifice. Hearts were weighed and normalized to body weights (e), and plasma CTNT was assessed (f). The dotted lines indicate the mean values obtained from n = 6 untreated WT age-matched female mice. Levels of heart weight/body weight and plasma CTNT in these WT mice were 3.3 ± 0.13 mg/g and 181.1 ± 43.9 pg/mL, respectively. Data are presented as mean ± SEM. P values were obtained from Tukey’s or Dunnett’s post hoc pairwise multiple comparison test following a one-way or two-way ANOVA, and are provided in the source data. ***P < 0.001.

Fig. 4. Dual mRNAs are well-tolerated in 3- and 6-month repeat-dose studies in PA hypomorphic mice.

a–c Select clinical chemistry parameters (serum ALT, AST, and BUN) in PA hypomorphic and unaffected mice (n = 12/group/sacrifice time point) from the 3-month study. d, e Select clinical chemistry parameters (plasma ALT and AST) in PA hypomorphic female mice (n = 6/group) from the 6-month study. The dotted lines indicate the mean values from n = 5 untreated WT age-matched female mice. Levels of plasma ALT and AST in these untreated WT mice were 65.60 ± 9.49 and 103.60 ± 14.15 U/L, respectively (one outlier was excluded). Data are presented as mean ± SEM. P values were obtained from Tukey’s post hoc pairwise multiple comparison test following a one-way or two-way ANOVA, and are provided in the source data. ***P < 0.001. ALT alanine aminotransferase, AST aspartate aminotransferase.