Alternative oxidase encoded by sequence-optimized and chemically-modified RNA transfected into mammalian cells is catalytically active

Abstract

Plants and other organisms, but not insects or vertebrates, express the auxiliary respiratory enzyme alternative oxidase (AOX) that bypasses mitochondrial respiratory complexes III and/or IV when impaired. Persistent expression of AOX from Ciona intestinalis in mammalian models has previously been shown to be effective in alleviating some metabolic stresses produced by respiratory chain inhibition while exacerbating others. This implies that chronic AOX expression may modify or disrupt metabolic signaling processes necessary to orchestrate adaptive remodeling, suggesting that its potential therapeutic use may be confined to acute pathologies, where a single course of treatment would suffice. One possible route for administering AOX transiently is AOX-encoding nucleic acid constructs. Here we demonstrate that AOX-encoding chemically-modified RNA (cmRNA), sequence-optimized for expression in mammalian cells, was able to support AOX expression in immortalized mouse embryonic fibroblasts (iMEFs), human lung carcinoma cells (A549) and primary mouse pulmonary arterial smooth muscle cells (PASMCs). AOX protein was detectable as early as 3 h after transfection, had a half-life of ~4 days and was catalytically active, thus supporting respiration and protecting against respiratory inhibition. Our data demonstrate that AOX-encoding cmRNA optimized for use in mammalian cells represents a viable route to investigate and possibly treat mitochondrial respiratory disorders.

Fig. 1. Construct design and expression test in iMEFs, of different AOX cmRNA variants.

a Schematic of cmRNA design. AOX, codon-optimized Ciona intestinalis alternative oxidase; mutAOX, catalytically inactive form of codon-optimized AOX; UTR, untranslated region; MTS, human ATP5F1B-derived mitochondrial targeting sequence; poly(A), polyadenylate tail. b Schematic illustrating where AOX integrates into the mitochondrial respiratory chain of mammalian cells and how AOX branches off electrons from ubiquinol to reduce oxygen to water. I–V, respiratory chain complexes. A, alternative oxidase (AOX). c, cytochrome c. IM, inner mitochondrial membrane. OM, outer mitochondrial membrane. Q, quinone pool. c Representative Western blots of iMEF lysates 24 h after transfection with catalytically active or inactive cmRNA constructs as indicated. α-tubulin, loading control. “ATP5F1B”, “HBA1” and “Minimal” indicate the different 5′ UTR constructs used. WT (wild-type) and AOX^Rosa26 (AOX-transgenic) iMEFs serving as negative and positive control for AOX expression, respectively. d Immunocytochemistry of iMEFs transfected with AOX cmRNA constructs using human ATP51B 5’UTR. Hoechst, nuclear stain; ATP5A, mitochondrial stain. Scale bars, 30 µm. e Representative Western blots using iMEFs transfected with cmRNA constructs bearing ATP51B 5′ UTR and encoding catalytically active or inactive AOX as indicated. Time after transfection shown in hours. α-tubulin, loading control. f Representative Western blots of MEFs, pre-treated with mitomycin C, transfected with cmRNA constructs as indicated. Time after transfection shown in days. α-tubulin, loading control. g Densitometric analysis of AOX protein expression in cmRNA-transfected MEFs normalized to α-tubulin. All data are shown as mean ± SD in arbitrary units (a.u.) with one being the average expression seen in AOX^Rosa26 iMEFs or MEFs of n = 3 independent experiments.

Fig. 2. cmRNA-encoded AOX is catalytically active in iMEFs.

a Respirometry using permeabilized iMEFs transfected with AOX-encoding cmRNA variants as indicated. WT iMEFs treated with or without lipofectamine served as negative control. Date are shown as mean ± SD of n ≥ 3 experiments. cII, respiratory complex II activity upon addition of succinate in the presence of cI inhibitor rotenone; AOX, Ciona intestinalis alternative oxidase activity upon further addition of antimycin A normalized to AOX inhibitor n-propyl gallate (nPG). b Representative images showing cellular morphology of iMEFs transfected with AOX cmRNA variants as indicated and treated with antimycin A (5 µM) for 48 h in 10 mM galactose-containing medium. Scale bars indicate 30 µm.

Fig. 3. cmRNA-encoded AOX is catalytically active in A549 cells.

a Representative Western blots of A549 cells 24 h after transfection with AOX cmRNAs as indicated (nomenclature as in Fig. 1; a, active; i, inactive). b Respirometry using A549 cells transfected with AOX cmRNA variants. Nomenclature as in Fig. 2. Data are shown as mean ± SD of values from n > 3 experiments, shown individually as filled circles.

Fig. 4. AOX cmRNA transfection has minor effects on A549 cell respiration and expression of respiratory subunits.

a Respirometry of A549 cells transfected with AOX cmRNAs encoding active or inactive AOX as indicated. Endo, oxygen consumption of intact cells prior to permeabilization with digitonin (endogenous substrates); cI (−ADP), measurements made after permeabilization and supplementation with complex I-linked substrates (pyruvate 5 mM, glutamate 5 mM, malate 2 mM) in absence of ADP (non-phosphorylating respiration); cI (+ADP), measurements made after the further addition of ADP (phosphorylating respiration, state 3); cIV, measurements in the presence of antimycin A and after the addition of ascorbate/TMPD and the subtraction of residual oxygen consumption (resistant to azide). Data are shown as mean ± SD of n ≥ 3 experiments (indicated individually by filled circles), with horizontal bars representing significant differences (p ≤ 0.05) based on two-way ANOVA and Tukey’s multiple comparisons test. b Representative Western blots (n = 2) probed for selected subunits of the mitochondrial respiratory complexes as indicated, as well as AOX, and α-tubulin used for loading control.

Fig. 5. cmRNA-encoded AOX is expressed and catalytically active in primary mouse pulmonary artery smooth muscle cells (PASMCs).

a Representative Western blots (n = 2) and b respirometry, using the same nomenclature as Fig. 4, except that the loading control used in the Westerns was α-SMA, as indicated. Data are shown as mean ± SD of values from n = 4 experiments, indicated individually by filled circles. c Respirometry using the same nomenclature as Fig. 4. Data are shown as mean ± SD of values from n = 4 experiments, indicated individually by filled circles.