Nrf2 activation by pyrroloquinoline quinone inhibits natural aging-related intervertebral disk degeneration in mice

Abstract

Age-related intervertebral disk degeneration (IVDD) involves increased oxidative damage, cellular senescence, and matrix degradation. Pyrroloquinoline quinone (PQQ) is a water-soluble vitamin-like compound with strong anti-oxidant capacity. The goal of this study was to determine whether PQQ can prevent aging-related IVDD, and the underlying mechanism. Here, we found that dietary PQQ supplementation for 12 months alleviated IVDD phenotypes in aged mice, including increased disk height index and reduced histological scores and cell loss, without toxicity. Mechanistically, PQQ inhibited oxidative stress, cellular senescence, and senescence-associated secretory phenotype (SASP) in the nucleus pulposus and annulus fibrosus of aged mice. Similarly, PQQ protected against interleukin-1β-induced matrix degradation, reactive oxygen species accumulation, and senescence in human nucleus pulposus cells (NPCs) in vitro. Molecular docking predicted and biochemical assays validated that PQQ interacts with specific residues to dissociate the Keap1-Nrf2 complex, thereby increasing nuclear Nrf2 translocation and activation of Nrf2-ARE signaling. RNA sequencing and luciferase assays revealed Nrf2 can transcriptionally upregulate Wnt5a by binding to its promoter, while Wnt5a knockdown prevented PQQ inhibition of matrix metalloproteinase-13 in NPCs. Notably, PQQ supplementation failed to alleviate aging-associated IVDD phenotypes and oxidative stress in aged Nrf2 knockout mice, indicating Nrf2 is indispensable for PQQ bioactivities. Collectively, this study demonstrates Nrf2 activation by PQQ inhibits aging-induced IVDD by attenuating cellular senescence and matrix degradation. This study clarifies Keap1-Nrf2-Wnt5a axis as the novel signaling underlying the protective effects of PQQ against aging-related IVDD, and provides evidence for PQQ as a potential agent for clinical prevention and treatment of natural aging-induced IVDD.

Keywords: IVDD; Keap1–Nrf2 signaling; Wnt5a; aging; pyrroloquinoline quinone.

FIGURE 1

PQQ supplementation alleviates the IVDD phenotype in aging wild‐type mice. (a) Experimental design for investigating the effects of PQQ on natural aging‐induced IVDD; wild‐type mice at the age of 12 months were given the normal diet and PQQ diet (4 mg/kg standard feed) for 12 months, respectively. Control mice were given the normal diet. Mice were sacrificed and IVD phenotypes analyzed at indicated ages. (b) Representative X‐ray scans and (c) the changes in the disk height index (DHI). (d) Safranin O and Fast Green (SO&FG) staining. (e) Histological scores of lumbar IVDs in indicated groups of mice. (f) Cell number in NP tissues. One‐way ANOVA with Tukey's post hoc test. **p < 0.01, ***p < 0.001.

FIGURE 2

PQQ rescued aging‐induced ECM degradation and cellular senescence during aging. Representative images of IVD sections immunostained for P16, MMP13, and Collagen II in (a) NP and (b) AF. (c–e) Quantification of the percentage of (c) p16⁺ NPCs and AFCs, (d) MMP13⁺ area, and (e) Collagen II⁺ area in NP and AF. (f) Western blot detection and (g) the quantitative analysis of Collagen II, MMP13, P16, P21, and Lamin B1 protein levels in indicated groups. One‐way ANOVA with Tukey's post hoc test. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 3

Effects of PQQ on ROS accumulation, cell proliferation, and senescence phenotype in IL‐1β‐treated human NPCs. (a) Human NPCs viability was determined using CCK8 assay following PQQ (0, 1, 5, 10, 20, 50, 100 μM) treatment for 24 and 48 h. (b) Crystal violet staining and (c) relative clonal expansion analysis. (d) EdU staining and (e) the quantitative analysis of EdU⁺ NPCs. (f) ROS levels determined using dihydroethidium (DHE) staining in vitro and (g) the quantitative analysis of ROS levels. (h) SA‐β‐gal staining and (i) the quantitative analysis of SA‐β‐gal⁺ NPCs. (j) Western blot detection and (k) qPCR detection of P16, Col2a1, and MMP13 in indicated groups. One‐way ANOVA with Tukey's post hoc test. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 4

PQQ binds to and disrupts Keap1–Nrf2 complex resulting in Nrf2–ARE activation. (a) The model of PQQ. (b) The ribbon model of the Keap1–Nrf2 complex. (c) The interaction between PQQ and Keap1–Nrf2 complex; ‐CDOCKER energy is 46.4490 kcal/mol. (d) Other potential targets of PQQ. (e) 2D binding model between PQQ and Keap1–Nrf2 complex, and the molecules of Keap1–Nrf2 complex that interact with PQQ through van der Waals, salt bridges, alkyl and hydrogen bonds. (f) The interaction between Keap1 and Nrf2 in PQQ‐treated human NPCs relative to vehicle‐treated control was determined using endogenous Co‐IP. (g) Western blot detection of nuclear‐Nrf2, total‐Nrf2, and HO1 in human NPCs following PQQ treatment for 24 h in the presence and absence of IL‐1β. (h) IF detection of Nrf2 in indicated groups of human NPCs. (i) Relative luciferase activity driven by ARE in human NPCs following PQQ treatment for 24 h in the presence and absence of IL‐1β were determined using dual luciferase assay. (j) qPCR detection of HO‐1 and Nqo1 in indicated groups. One‐way ANOVA with Tukey's post‐hoc test. *p < 0.05, **p < 0.001, ***p < 0.001. ns: not significant.

FIGURE 5

Nrf2 activation inhibits ECM degradation by upregulating Wnt5a in NPCs. (a) RNA‐seq heatmap showing the mentioned genes expressed in BM‐MSCs isolated from WT and Nrf2^−/− mice. (b) The mRNA levels of Wnt5a in NPCs from 18‐month‐old WT, Nrf2^−/− mice and PQQ‐treated Nrf2^−/− mice. (c) The mRNA levels of Wnt5a in mouse NPCs from young (3‐month‐old), old (18‐month‐old), and PQQ‐treated 18‐month‐old WT mice. (d) A predictive Nrf2‐binding element in mouse Wnt5a promoter region. (e) Chromatin immunoprecipitation (ChIP) with Nrf2 antibody or IgG antibody were performed in vehicle‐ or tBHQ‐treated mouse NPCs and relative enrichment of Wnt5a promoter was determined using qPCR assay. (f) Mouse Wnt5a promoter or Wnt5a promoter mutant Luc‐plasmids were transfected into mouse NPCs following lentivirus‐mediated Nrf2 overexpression and relative luciferase activity were analyzed after 48 h. (g) The mRNA levels of Wnt5a in mouse NPCs from 3‐month‐old WT, Nrf2^−/− mice, and NAC‐treated Nrf2^−/− mice. (h) The protein and mRNA levels of Wnt5a in NP tissues from young (3‐month‐old) and old (18‐month‐old) wild‐type mice. Western blot detection of P16, P21, and Lamin B1 was used as cell senescence markers. (i) Knock‐down efficiency of si‐Wnt5a determined using Western blot. (j) Western blot detection of MMP13 in vehicle‐ or PQQ‐treated human NPCs in the presence or absence of si‐Wnt5a. (k) Quantitative analysis of MMP13 in (j). One‐way ANOVA with Tukey's post hoc test. *p < 0.05, **p < 0.01, ***p < 0.001. ns: not significant.

FIGURE 6

Nrf2 knockout blunts the preventing effects of PQQ on aging‐related IVDD. (a) Experimental design for investigating the effects of PQQ on Nrf2 deficiency‐induced IVDD; Nrf2‐deficient mice at the age of 6 months were given the normal diet and PQQ‐containing diet (4 mg/kg standard feed) for 12 months, respectively. Six‐month‐old control mice were given the normal diet. (b) Representative X‐ray scans and (c) the quantitative analysis of DHI. (d) Safranin O and Fast Green (SO&FG) staining and (e) histological scores of lumbar IVD in indicated groups of mice. (f–h) Representative images of lumbar IVD sections immunostained for (f) P16, (g) MMP13, and (h) Collagen II in NP tissues. (i–k) Quantification of the percentage of (i) p16⁺ NPCs, (j) MMP13⁺ area, and (k) Collagen II⁺ area in NP tissues. One‐way ANOVA with Tukey's post hoc test. *p < 0.05, ***p < 0.001. ns: not significant.