NAD+ dependent UPRmt activation underlies intestinal aging caused by mitochondrial DNA mutations

Abstract

Aging in mammals is accompanied by an imbalance of intestinal homeostasis and accumulation of mitochondrial DNA (mtDNA) mutations. However, little is known about how accumulated mtDNA mutations modulate intestinal homeostasis. We observe the accumulation of mtDNA mutations in the small intestine of aged male mice, suggesting an association with physiological intestinal aging. Using polymerase gamma (POLG) mutator mice and wild-type mice, we generate male mice with progressive mtDNA mutation burdens. Investigation utilizing organoid technology and in vivo intestinal stem cell labeling reveals decreased colony formation efficiency of intestinal crypts and LGR5-expressing intestinal stem cells in response to a threshold mtDNA mutation burden. Mechanistically, increased mtDNA mutation burden exacerbates the aging phenotype of the small intestine through ATF5 dependent mitochondrial unfolded protein response (UPR^mt) activation. This aging phenotype is reversed by supplementation with the NAD⁺ precursor, NMN. Thus, we uncover a NAD⁺ dependent UPR^mt triggered by mtDNA mutations that regulates the intestinal aging.

Fig. 1. Increased mtDNA mutation burden induces an aging-like phenotype in the small intestine.

a Diagram depicting the experimental methods for detecting the changes of small intestine and organoids derived from isolated crypts in the mice with four categories of mtDNA mutation burden at 3, 8, and 12 month. The figures of small intestine, intestinal crypt, and organoids were generated with the help of Servier Medical Art, provided by Servier, licensed under a Creative Commons Attribution 4.0 unported license (https://creativecommons.org/licenses/by/4.0/deed.en). PolgA^WT/Mut (WT/Mut**) mice were crossed to generate three types of offspring with a theoretically maternally transmitted mtDNA mutations, i.e. PolgA^WT/WT (WT/WT*), PolgA^WT/Mut (WT/Mut**) and PolgA^Mut/Mut (Mut/Mut***). PolgA^WT/WT mice with negligible level of mtDNA mutations (WT/WT) were used as an additional control. b Representative images of small intestine stained with H&E in WT/WT, WT/WT*, WT/Mut**, and Mut/Mut*** mice at 3, 8, and 12 months of age (Scale bar, 100 μm). The villus length and number of crypts per millimeter of small intestine are quantified below (Data are presented as the mean ± S.D of 9 intestinal segments from 3 mice per group; two-way ANOVA test). c Representative images of intestinal crypts stained with TUNEL in WT/WT*, WT/Mut**, and Mut/Mut*** mice at 3, 8, and 12 months of age (Scale bar, 20 μm). Since no difference was observed during intestinal aging between WT/WT and WT/WT*, WT/WT* mice was used a control. The number of TUNEL-positive cells per crypt is quantified on right (Data are presented as the mean ± S.D and n = 3 mice per group; two-way ANOVA test). d Representative crypt images at Day 0 and organoids at Day 8 in the intestine of WT/WT*, WT/Mut**, and Mut/Mut*** mice at 3, 8, and 12 months of age (Scale bar, 100 μm). Crypt size, organoid size and relative organoid number per well were quantified. Average organoid number per well was normalized to WT/WT* at 3 months. Data are presented as the mean ± S.D of 6 wells from 3 mice per group; two-way ANOVA test. Source data are provided with this paper.

Fig. 2. Increased mtDNA mutation burden results in NAD⁺ depletion during the intestinal aging.

a Gene Ontology (GO) enrichment of differentially upregulated genes in Mut/Mut*** mice at 8 months of age versus WT/WT* mice was performed using clusterProfiler package, and significance was determined by Fisher exact test of adjusted p-value by Benjamini–Hochberg (BH) method (n = 3 mice per group). b Detection of NADH/NAD⁺ ratio in the intestinal crypts of Mut/Mut*** mice and WT/WT* mice at 8 months of age using SoNar and cpYFP imaging (Scale bar, 10 μm). c Quantification of NADH/NAD⁺ ratios as in (b) (90 intestinal crypts from 3 mice per group; unpaired comparison using nonparametric Mann–Whitney test). Central line: median, box: interquartile ranges (IQR), whisker: ranges except extreme outliers (>1.5*IQR), circles: individual data points; d Detection of NAD⁺ content in the intestinal crypts of Mut/Mut*** mice and WT/WT* mice using mCherry-FiNad and mCherry-cpYFP imaging (Scale bar, 10 μm). e Quantification of mCherry-FiNad fluorescence corrected by mCherry-cpYFP as in (d) (90 intestinal crypts from 3 mice per group; unpaired comparison using nonparametric Mann–Whitney test). Central line: median, box: IQR, whisker: ranges except extreme outliers (>1.5*IQR), circles: individual data points. Source data are provided with this paper.

Fig. 3. NAD⁺ repletion alleviates small intestine aging phenotypes caused by mtDNA mutation burden.

a Representative images of small intestine stained with H&E in Mut/Mut*** mice at 8 months of age with NMN or control (water) (Scale bar, 100 μm). Villus length and number of crypts per millimeter are quantified on right; WT/WT* is used as an additional control (Data are presented as the mean ± S.D of 9 intestinal segments from 3 mice per group; unpaired two-tailed Student’s t-test). b Representative images of intestinal crypts stained with TUNEL in Mut/Mut*** mice at 8 months of age with NMN or control (Scale bar, 20 μm). The number of TUNEL-positive cells per crypt is quantified on right (Data are presented as the mean ± S.D and n = 3 mice per group; unpaired two-tailed Student’s t-test). c Representative images of crypts at Day 0 and organoids at Day 8 from the intestine of Mut/Mut*** mice at 8 months of age with 200 μM NMN or water control (Scale bar, 100 μm). Organoid size and relative organoid number per well are quantified at bottom. Average organoid number per well is normalized to the control. Data are presented as the mean ± S.D of 6 wells from 3 mice per group; paired two-tailed Student’s t-test. Source data are provided with this paper.

Fig. 4. NAD⁺ depletion impairs Wnt/β-catenin signaling during the intestinal aging caused by mtDNA mutation burden.

a Heatmap of differential genes associated with Wnt/β-catenin signaling in WT/WT*, WT/Mut** and Mut/Mut*** mice at 8 months of age (n = 3 mice per group). b Schematic illustration of LGR5-expressing ISCs, proliferated epithelial cells and Wnt/β-catenin signaling in the intestinal crypts. The figure of intestinal crypt was generated with the help of Servier Medical Art, provided by Servier, licensed under a Creative Commons Attribution 4.0 unported license (https://creativecommons.org/licenses/by/4.0/deed.en). c, d Anti-β-catenin (c) and anti-CD1 (d) IF in the intestinal crypts of WT/WT*, WT/Mut** and Mut/Mut*** mice at 3, 8 and 12 months of age. Relative FIs of β-catenin and CD1⁺ cell number per crypt are quantified on right (Scale bar, 10 μm; data are presented as the mean ± S.D and n = 3 mice per group; two-way ANOVA test). e Anti-β-catenin and anti-CD1 IF in the intestinal crypts of Mut/Mut*** mice at 8 months of age treated with NMN or water. Relative FI of β-catenin and CD1⁺ cell number per crypt are quantified on right (Scale bar, 10 μm; data are presented as the mean ± S.D and n = 3 mice per group; unpaired two-tailed Student’s t-test). Source data are provided with this paper.

Fig. 5. Increased mtDNA mutation burden induces ISC decrease through NAD⁺ depletion.

a Generation of LGR5-GFP reporter mice with four categories of mtDNA mutation burden. LGR5^WT/GFP female mice were first crossed to WT/Mut** male mice, and the offspring of LGR5^WT/GFP& WT/Mut** mice were further used to generate WT/WT*&LGR5^WT/GFP, WT/Mut**&LGR5^WT/GFP and Mut/Mut***&LGR5^WT/GFP. WT/WT&LGR5^WT/GFP mice generated by crossing WT/WT male mice to female LGR5^WT/GFP female mice were used as an additional control. The figure was partly generated with the help of Servier Medical Art. b Representative images of LGR5-expressing ISCs in WT/WT&LGR5^WT/GFP, WT/WT*&LGR5^WT/GFP, WT/Mut**&LGR5^WT/GFP and Mut/Mut***&LGR5^WT/GFP male mice at 3, 8 and 12 months of age (Scale bar, 20 μm). ISC number is quantified at bottom (Data are presented as the mean ± S.D and n = 3 mice per group; two-way ANOVA test). c Representative images of LGR5-expressing ISCs in Mut/Mut***&LGR5^WT/GFP mice at 8 months of age treated with NMN or water control (Scale bar, 10 μm). ISC number is quantified on right (Data are presented as the mean ± S.D and n = 3 mice per group; unpaired two-tailed Student’s t-test). d Anti-WNT2, anti-WNT4 and anti-WNT5A IF in the intestine of WT/WT*, WT/Mut** and Mut/Mut*** mice at 8 months of age treated with NMN or water control. Relative FI is quantified on right (Scale bar, 100 μm; data are presented as the mean ± S.D and n = 3 mice per group; one-way ANOVA test). Source data are provided with this paper.

Fig. 6. NAD⁺ depletion activates ISR to regulate mtDNA mutation-induced aging phenotypes through impaired Wnt/β-catenin signaling.

a Kyoto Encyclopedia of Genes and Genomes pathway enrichment of differentially upregulated genes in Mut/Mut*** mice at 8 months of age versus WT/WT* mice was performed using clusterProfiler package, and significance was determined by Fisher exact test of adjusted p-value by BH method (n = 3 mice per group). b Protein expression of eIF2α-P and ATF5 by western blot analysis in the intestinal crypts of WT/WT*, WT/Mut** and Mut/Mut*** mice at 8 months of age. Relative band densities quantified using ImageJ are shown at bottom (Data are presented as the mean ± S.D and n = 3 mice per group; one-way ANOVA test). c Protein expression of eIF2α-P and ATF5 by western blot analysis in the intestinal crypts of Mut/Mut*** mice treated with NMN or water. Relative band densities quantified using ImageJ are shown at bottom (Data are presented as the mean ± S.D and n = 3 mice per group; one-way ANOVA test). d Protein expression of ATF5 by western blot analysis in Mut/Mut*** intestinal crypt cells treated with Control or ISRIB. Relative band densities quantified using ImageJ are shown on right (Data are presented as the mean ± S.D and n = 3 mice per group; paired two-tailed Student’s t-test). e Protein expression of β-catenin by western blot analysis in Mut/Mut*** intestinal crypt cells treated with ISRIB or control. Relative band densities quantified using ImageJ are shown on right (Data are presented as the mean ± S.D and n = 3 mice per group; paired two-tailed Student’s t-test). f Representative organoid images at Day 8 from the intestinal crypts of Mut/Mut*** mice at 8 months of age with 500 nM ISRIB or water control (Scale bar, 100 μm). Organoid size and relative organoid number per well are quantified on right, and the average organoid number per well is normalized to the control. (Data are presented as the mean ± S.D and n = 3 mice per group; unpaired two-tailed Student’s t-test) g Representative organoid images at Day 8 from the intestinal crypts of Mut/Mut*** mice at 8 months of age with siAtf5 or control siNC (Scale bar, 100 μm); organoid size and relative organoid number per well are quantified at bottom. Average organoid number per well is normalized to the control. Data are presented as the mean ± S.D and n = 3 mice per group; unpaired two-tailed Student’s t-test. Source data are provided with this paper.

Fig. 7. The ISR caused by mtDNA mutation burden regulates UPRmt activation.

a Protein expression of HSP60, LONP1, and ClpP by western blot analysis in the intestinal crypts in WT/WT*, WT/Mut**, and Mut/Mut*** mice at 8 months of age. Relative band densities quantified using ImageJ are shown at bottom (Data are presented as the mean ± S.D and n = 3 mice per group; one-way ANOVA test). b Protein expression of LONP1 by western blot analysis in the intestinal crypts of Mut/Mut*** mice treated with ISRIB or control. Relative band densities quantified using ImageJ are shown on right (Data are presented as the mean ± S.D and n = 3 mice per group; paired two-tailed Student’s t-test). c Protein expression of LONP1 by western blot analysis in the intestinal crypts of Mut/Mut*** mice treated with siAtf5 or siNC. Relative band densities quantified using ImageJ are shown on right (Data are presented as the mean ± S.D and n = 2 mice per group; paired two-tailed Student’s t-test). d Summary of mechanism regulating intestinal aging caused by mtDNA mutation burden. The figures of small intestine and mitochondria were generated with the help of Servier Medical Art, provided by Servier, licensed under a Creative Commons Attribution 4.0 unported license (https://creativecommons.org/licenses/by/4.0/deed.en).