Re-equilibration of imbalanced NAD metabolism ameliorates the impact of telomere dysfunction

Abstract

Short telomeres are a principal defining feature of telomere biology disorders, such as dyskeratosis congenita (DC), for which there are no effective treatments. Here, we report that primary fibroblasts from DC patients and late generation telomerase knockout mice display lower nicotinamide adenine dinucleotide (NAD) levels, and an imbalance in the NAD metabolome that includes elevated CD38 NADase and reduced poly(ADP-ribose) polymerase and SIRT1 activities, respectively, affecting many associated biological pathways. Supplementation with the NAD precursor, nicotinamide riboside, and CD38 inhibition improved NAD homeostasis, thereby alleviating telomere damage, defective mitochondrial biosynthesis and clearance, cell growth retardation, and cellular senescence of DC fibroblasts. These findings reveal a direct, underlying role of NAD dysregulation when telomeres are short and underscore its relevance to the pathophysiology and interventions of human telomere-driven diseases.

Keywords: CD38 NADase; NAD metabolism; mitochondrial impairment; replicative senescence; telomere biology disorders.

Conflict of interest statement

The authors declare that they have no conflict of interest.

© 2020 The Authors. Published under the terms of the CC BY NC ND 4.0 license.

Figures

Figure 1. DC patient fibroblasts exhibit defects in NAD metabolism and NAD‐dependent PARylation and SIRT1 deacetylation activity

1. A, B

   Intracellular NAD levels and NAD/NADH ratio in DC and age‐matched healthy control fibroblasts. All values are presented as mean ± SD of four and eight replicates in (A) and (B), respectively. Student's t‐test was performed on DC cells vs controls.

2. C

   An overview of NAD metabolism. The activity of NAD‐consuming enzymes, such as PARPs, SIRTs, or CD38 consumes NAD and results in NAM production. The NAD biosynthesizing enzymes NMNATs and NAMPT recycle NAM back to NMN and then NMN to NAD, respectively. NR is converted to NMN, followed by conversion of NMN to NAD.

3. D, E

   Immunoblots of the expression of NAD synthesis and consuming proteins, and their activities in DC and control cells. Protein levels are normalized to GAPDH. Quantification values are presented as mean ± SD of four controls vs five DC samples as shown in (D). Student's t‐test was performed on DC cells vs controls. Con: control fibroblasts. Irrelevant intervening lanes in (D) have been removed for clarity (full blots are available online as Source Data).

4. F

   Effects of NR and NAM supplementation (24 h) on the NAD levels in DC fibroblasts. All values are presented as mean ± SD of four replicates. One‐way ANOVA was performed on DC cells in indicated conditions.

Source data are available online for this figure.

Figure 2. CD38 depletion ameliorates dysregulated NAD metabolism in DC cells

1. A

   Quantitative RT‐PCR and immunoblots of the expression of CD38 and SARM1 in DC and age‐matched health control fibroblasts. All values are presented as mean ± SD of three replicates in quantitative RT‐PCR. Protein levels are normalized to GAPDH, and mean (± SD) quantification values of four controls vs five DC samples are shown. Student's t‐test was performed on DC cells vs controls. Irrelevant intervening lanes have been removed for clarity (full blots are available online as Source Data).

2. B

   Immunoblots and quantification of indicated proteins in DC fibroblasts treated with DMSO and with 2 μM ATM inhibitor, KU‐55933 for 48 h. Mean (± SD) quantification values of three DC lines with and without ATMi treatment are shown. P values on the basis of Student's t‐test.

3. C

   The efficacy of CD38 shRNA (sh‐1 and sh‐2) was verified by quantitative RT‐PCR in DC fibroblasts and by RT‐PCR and immunoblots in A549 cells. The relative CD38 mRNA values in the CD38 knockdown cells were normalized to the scrambled shRNA. All values are presented as mean ± SD of three replicates. Student's t‐test was performed on DC and A549 cells in indicated conditions.

4. D, E

   Intracellular NAD levels and immunoblots of the expression of indicated proteins in the scramble and CD38 knockdown DC fibroblasts. The CD38 knockdown DC fibroblasts were treated with the PARP1 inhibitor Olaparib (400 nM) and the SIRT1 inhibitor EX 527 (1 μM) for 24 h. All values are presented as mean ± SD of four replicates. One‐way ANOVA was performed on DC cells in indicated conditions.

Source data are available online for this figure.

Figure 3. G3 Tert −/− mice display defects in the NAD metabolism

1. A, B

   Intracellular NAD levels and immunoblots of the expression of NAD consuming‐related proteins and their activities in brain tissues from 6‐month‐old G1 and G3 Tert−/− female mice. The CD38 antibody was validated using the CD38−/− mouse brain tissues.

2. C

   Quantification of immunoblots from the indicated proteins in (B).

3. D

   NADase activity in G1 and G3 Tert−/− mouse brain lysate was determined at various time points during the reaction. NADase activity was not detectable in the extracts treated with the CD38 inhibitor, 78c, at a concentration of 50 nM.

Data information: All values are presented as mean ± SD of G1 Tert−/− mouse brain tissues vs G3 Tert−/− mouse brain tissues. n = 4 in each group. P values on the basis of Student's t‐test.

Figure 4. NAD intervention impacts mitochondrial parameters of DC fibroblasts

1. A

   Representative immunoblots of the expression of PGC1‐1α in DC cells treated with vehicle or 3 mM NR. Quantification of the indicated proteins is from three immunoblots.

2. B–D

   Cellular and mitochondrial ROS in DC and age‐matched healthy control cells supplemented with vehicle and NR (B, C) and in the scramble and CD38 knockdown DC cells (D) were measured by flow cytometry from three replicates.

3. E

   Representative images captured by transmission electron microscopy of DC1 cells treated with vehicle or NR. White arrows: mitochondria. Yellow arrow: autophagosome‐like structures with engulfed mitochondria. Enlarged image within white frame is shown.

4. F–H

   Quantifications of percentage of damaged mitochondria per cell (F), mitochondrial length (G), and mitochondrial diameter (H). A minimum of 200 mitochondria counted per group.

Data information: All data are represented as mean ± SD. Student's t‐test was performed for individual pairs in indicated conditions. Con: control.

Figure 5. Impaired mitophagy process in DC fibroblasts is improved by NR supplementation

1. Representative images of mitophagy showing co‐localization between the mitophagy dye (red) and the lysosome dye (green) in DC and control fibroblasts treated with vehicle or 3 mM NR. White arrows: mitophagy.

2. The mean value of fluorescence intensity/cells in each image was scored. At least 15 images (˜ 200 cells) were counted per group. The relative values in each group were normalized to Con1.

3. Representative immunoblots of the expression of PINK1 and PARKIN in DC cells treated with vehicle or 3 mM NR. Quantification of immunoblots from the indicated proteins is from three replicates.

Data information: All values are represented as mean ± SD. One‐way ANOVA (B) and Student's t‐test (C) were performed for each individual pair. Con: age‐matched healthy controls.

Figure 6. NR supplementation reduces telomeric oxidative DNA lesions and TIF formation in DC cells

1. A

   PCR amplification efficiency at the telomere in the mock‐ and FPG‐treated DC and age‐matched healthy control fibroblasts supplemented with vehicle or 3 mM NR. All values are presented as mean ± SD of three replicates. The relative telomere PCR amplification in each sample was normalized to the 36B4 reference gene.

2. B

   Telomere restriction fragment analysis in DC and control fibroblasts treated with vehicle or 3 mM NR. Genomic DNA was isolated from the indicated cell lines at 0 or 14 days, with or without 3 mM NR treatment. At left, DNA molecular mass markers, in kilobase pairs (Kb).

3. C, D

   Representative images of γH2AX (green), telomere immuno‐FISH (red) in DC cells treated with vehicle or 3 mM NR. White arrows: co‐localization of γH2AX foci and telomeric DNA (or TIF). Enlarged views of co‐localizing foci are shown at the right panel and in Fig EV1. Percentage of DC and control cells with indicated TIFs per nuclei. Approximately 100 cells per condition were scored.

Data information: All values are represented as mean ± SD. P values were calculated using One‐way ANOVA (A, D). Con: age‐matched healthy controls.

Figure EV1. TIF formation in DC cells

Enlarged view of the upper right panel of Fig 6C, showing a representative image with γH2AX (green) and telomere immuno‐FISH (red) in a DC cell. White arrows: co‐localization of γH2AX foci and telomeric DNA (or TIF). Enlarged view of co‐localizing foci is shown at the right panel.

Figure 7. NAD intervention improves the proliferative capacity and suppresses cellular senescence and SASP in DC fibroblasts

1. A

   Cumulative population doubling analysis of the proliferation of representative scramble and CD38 knockdown DC fibroblasts or DC fibroblasts treated with vehicle or 3 mM NR. Each data point is represented as mean ± SD of three replicates.

2. B, C

   Representative images of BrdU (red) and DAPI (blue) staining of DC and age‐matched healthy control fibroblasts treated with vehicle or NR. Percentage of BrdU‐positive cells per condition. Each dot represents the percentage of cells with BrdU staining per image. Approximately 400 cells were counted per condition. All values are presented as mean ± SD.

3. D, E

   Representative images of SPiDER‐β‐gal (green) and DAPI (blue) staining of DC and age‐matched healthy control cells treated with vehicle or 3 mM NR. Percentage of SA‐β‐gal-positive cells per condition. Each dot represents the percentage of cells with SA‐β‐gal staining per image. Approximately 400 cells were counted per condition. All values are presented as mean ± SD.

4. F

   The levels of IL‐6, IL‐8, and MCP‐1 in both supernatant and cell lysate of indicated DC fibroblasts treated with vehicle or 3 mM NR. All values are represented as mean ± SD of three replicates.

5. G, H

   Representative immunoblots of p21 and p16 in DC fibroblasts treated with vehicle or 3 mM NR. Quantification of the indicated proteins is presented as mean ± SD from three immunoblots.

Data information: P values were calculated using one‐way ANOVA for multiple comparisons (C) and Student's t‐test between two groups (C, E, F, and H). Con: control.