Mitochondrial bioenergetics deficiency in cisd-1 mutants is linked to AMPK-mediated lipid metabolism

Abstract

Background: CISD-1 is a mitochondrial iron-sulfate [2Fe-2S] protein known to be associated with various human diseases, including cancer and diabetes. Previously, we demonstrated that CISD-1 deficiency in worms lowers glucose and ATP levels. In this study, we further explored how worms compensate for lower ATP levels by analyzing changes in cytoplasmic and mitochondrial iron content, AMPK activities, and total lipid profiles.

Materials and methods: Expression levels of CISD-1 and CISD-1GFP fusion proteins in wild-type worms (N2), cisd-1-deletion mutants (tm4993 and syb923) and GFP insertion transgenic worms (PHX953 and SJL40) were examined by Western blot. Fluorescence microscopy analyzed CISD-1GFP pattern in PHX953 embryos and adults, and lipid droplet sizes in N2, cisd-1, aak-2 and aak-2;cisd-1 worms. Total and mitochondrial iron content, electron transport complex profiles, and AMPK activity were investigated in tm4993 and syb923 mutants. mRNA levels of mitochondrial β-oxidation genes, acs-2, cpt-5, and ech-1, were quantified by RT-qPCR in various genetic worm strains. Lipidomic analyses were performed in N2 and cisd-1(tm4993) worms.

Results: Defects in cisd-1 lead to an imbalance in iron transport and cause proton leak, resulting in lower ATP production by interrupting the mitochondrial electron transport chain. We identified a signaling pathway that links ATP deficiency-induced AMPK (AMP activated protein kinase) activation to the expression of genes that facilitate lipolysis via β-oxidation.

Conclusion: Our data provide a functional coordination between CISD-1 and AMPK constitutes a mitochondrial bioenergetics quality control mechanism that provides compensatory energy resources.

Keywords: AMPK; C. elegans; Iron-sulfate containing protein; Lipid droplet size; Lipidomic; Mitochondria outer membrane protein.

Graphical abstract

Fig. 1

Characterization of protein encoded by W02B12.15 (cisd-1 gene) (A) Genomic structure of cisd-1. The red boxes above the gene show the deletion region in cisd-1(tm4993) and cisd-1(syb923). The GFP translational reporter strain, PHX953, was created by using the CRISPR editing method. (B) Western blot analysis of five strains of worms - wild-type (N2), cisd-1(tm4993), cisd-1(syb923), PHX953 (Pcisd-1:CISD-1GFP), and SJL40 (Pfib-1:CISD-1GFP expressing transgenic worm) - using anti-GFP (left panel) and anti-mouse CISD-1 (right panel) antibodies. (C) Fluorescence images of the expression pattern of CISD-1GFP in the embryonic stages of development. Scale bar, 10 μm. (D) Fluorescence images of the expression pattern of CISD-1GFP in an adult worm. CISD-1GFP is expressed throughout the body, with higher expression levels in the pharynx and intestine. Scale bar, 100 μm. (E) Z-stack images at various depths (3 μm–24 μm) acquired for the head, midbody, and tail regions marked in (D), with specific tissues labeled at high magnification. Scale bar, 100 μm.

Fig. 2

Depletion of CISD-1 causes iron transport imbalance between mitochondria and cytoplasm (A) Total iron content in wild-type (N2), cisd-1(tm4993), and cisd-1(syb923) worms, measured and normalized to the total protein concentration (N = 9 at N2, N = 5 at tm4993, N = 4 at syb923). (B) & (C) The total iron, ferrous iron, and ferric iron in the cytosolic (B) or the mitochondrial (C) fractions of wild-type (N2), cisd-1(tm4993), and cisd-1(syb923) worms, measured and normalized with total protein concentration (N = 6 at N2, N = 3 at tm4993, N = 4 at syb923). (D) Aconitase activities in cytosolic and mitochondrial fractions of wild-type (N2), cisd-1(tm4993), and cisd-1(syb923) worms, normalized to protein concentration (N = 7 at N2, N = 3 at tm4993, N = 4 at syb923). N indicates the number of experiments conducted. ∗p < 0.05. ∗∗p < 0.01. ∗∗∗p < 0.001. Error bars represent SEMs.

Fig. 3

Loss of CISD-1 disrupts mitochondria electron transport chain function (A) Composition of mitochondrial electron transport chain proteins analyzed using Blue native PAGE. Representative gel of isolated mitochondria from young adults of wild-type (N2) and cisd-1(tm4993). MW markers (in kDa) are shown on the left, and ETC complexes with designations on the right. BSA was used as a protein loading control. (B) Quantification of electron transport chain complexes from three independent Blue native PAGE experiments as shown in (A). The dimer of complex V (CV2), complex I (CI), monomer of complex V (CV), complex III (CIII), and complex IV (CIV) are presented (N = 6). (C) Oxygen consumption rate (OCR) in wild-type (N2), cisd-1(tm4993) and cisd-1(syb923) animals. Around 100 young adult worms were measured in the presence of 20 μM DCCD, 25 μM FCCP, or 10 mM sodium azide using the Seahorse XF24 (N = 3). (D) ATP-linked respiration of in wild-type (N2), cisd-1(tm4993) and cisd-1(syb923) young adults. ATP-linked respiration indicates the basal OCR minus DCCD response OCR (N = 3). (E) Spare respiratory capacity in wild-type (N2), cisd-1(tm4993) and cisd-1(syb923). Spare respiration capacity represents the FCCP response OCR (maximum OCR) minus basal OCR (N = 3). (F) Proton leak in wild-type (N2), cisd-1(tm4993) and cisd-1(syb923) worms. Proton leak is demonstrated by the DCCD response OCR minus sodium azide-inhibited OCR (N = 3). (G) AMP to ATP ratio in wild-type (N2), cisd-1(tm4993) and cisd-1(syb923) animals (N = 6). (H) Level of AMPK activation determined by the ratio of phosphorylated AMPK to total AMPK. A representative Western blot gel shows the amount of p-AMPK (upper gel) and total AMPK (lower gel) in wild-type (N2), cisd-1(tm4993) and cisd-1(syb923) worms (N = 4). “N” indicates the number of experiments conducted, ∗p < 0.05. ∗∗p < 0.01. ∗∗∗p < 0.001. Error bars represent SEMs.

Fig. 4

Depletion of CISD1 decreases the amount and size of lipid droplets. (A) Representative images of the BoDIPY staining in wild-type (N2) and cisd-1 (tm4993) worms, with mutants additionally stained with WGA in red (indicated by arrowheads). Scale bar, 250 μm. (B) Quantitative data from the BoDIPY staining of N2, cisd-1(tm4993) mutant, aak-2 mutant and aak-2; cisd-1(tm4993) double mutant (n = 14 at N2, n = 18 at tm4993, n = 25 at aak-2, n = 24 at aak-2; cisd-1). “n” indicates the number of animals used. (C) Representative high-magnification images showing the size and amount of lipid droplets in intestine cells of each strain as indicated. Images were taken under a laser-scanning confocal microscope. Scale bar, 10 μm. (D) Lipid droplet size profiles of each strain quantified by the DropletFinder plugin in ImageJ. (E) Quantitative data of the averaged size of lipid droplets in each strain of worm. ∗p < 0.05. ∗∗p < 0.01. ∗∗∗p < 0.001. Error bars represent SEMs.

Fig. 5

Upregulation of genes in b-oxidation in cisd-1 mutant is AMPK and NHR-49 dependent (A), (B), and (C) mRNA levels of AMPK responsive β-oxidation genes: acs-2 (acetyl-CoA synthetase; ortholog of human ACSF2, acyl-CoA synthetase family member 2), cpt-5 (carnitine palmitoyl transferase; homolog of human Carnitine O-palmitoyltransferase 1) and ech-1 (enoyl-CoA hydratase; ortholog of human HADHA, hydroxyacyl-CoA dehydrogenase trifunctional multienzyme complex subunit) analyzed by qRT-PCR in four strains of worms, N2, cisd-1(tm4993), aak-2, and aak-2; cisd-1(tm4993) (N ≥ 3). ∗p < 0.05. ∗∗p < 0.01. ∗∗∗p < 0.001. Error bars represent SEMs.

Fig. 6

Metabolomic analysis of lipid metabolism in wild-type (N2) and cisd-1 mutant (tm4993). (A) Separation of wild-type (N2) and cisd-1 (tm4993) mutant worms in a PCA scores plot. The values of t [1] and t [2] represent the score of each sample in the principal component 1 and component 2. The two groups were separated with R2X and Q2 were 0.905 and 0.845, respectively. (B) Loading S-plot scatter was arising from the components of the PCA-X model. The high correlation and covariance variables were selected and highlighted in green color. Identification of each metabolite was marked in the relative position in the plot. (C) Heat map of significant metabolite changes in wild-type and cisd-1 mutant worms. The heat map was created using the MetaboAnalyst 3.6. The number at the top of the heat map is the biological replicates of each strain. The color ranges from dark red to dark blue meaning the metabolites abundance from high to low. Isoforms are indicated by alphabets.

Fig. 7

A model of CISD1 action mechanism. In wild-type worms (N2), CISD-1 regulates iron homeostasis between mitochondria and cytoplasm to maintain a proper structure of ETC complexes, which contain [Fe–S] clusters and produce a large amount of ATP for lipogenesis (lipid storage) (left panel). In cisd-1 mutants, the iron homeostasis between mitochondria and cytoplasm is imbalanced. Subsequently, the structure of the ETC complex, primarily complex I, is altered to reduce ATP biosynthesis. In the presence of AMPK activation, β-oxidation genes are elevated to facilitate lipolysis to produce a small amount of ATP (right panel) for maintaining normal motility and lifespan described previously [8].