Prolonged metformin treatment diminishes the multidimensional biological age and alleviates neuronal aging through the Nrf2 pathway in male primates. 长期二甲双胍治疗通过 Nrf2 通路降低雄性灵长类动物的多维生物年龄并缓解神经元衰老。
Resource 资源
Metformin decelerates aging clock in male monkeys
Yuanhan Yang, ^(1,10,23){ }^{1,10,23} Xiaoyong Lu, ^(2,3,10,23){ }^{2,3,10,23} Ning Liu, ^(7,8,23){ }^{7,8,23} Shuai Ma, ^(1,4,5,10,23){ }^{1,4,5,10,23} Hui Zhang, ^(1,10,13,23){ }^{1,10,13,23} Zhiyi Zhang, ^(7,23){ }^{7,23}Kuan Yang, ^(2,3,10){ }^{2,3,10} Mengmeng Jiang, ^(1,4,5){ }^{1,4,5} Zikai Zheng, ^(2,3,10){ }^{2,3,10} Yicheng Qiao, ^(7,8){ }^{7,8} Qinchao Hu, ^(1,11){ }^{1,11} Ying Huang, ^(12){ }^{12}Yiyuan Zhang, ^(1,4,5){ }^{1,4,5} Muzhao Xiong, ^(2,3,10){ }^{2,3,10} Lixiao Liu, ^(2,3,10){ }^{2,3,10} Xiaoyu Jiang, ^(1,10){ }^{1,10} Pradeep Reddy, ^(15){ }^{15} Xueda Dong, ^(7,9){ }^{7,9} Fanshu Xu, ^(7,8){ }^{7,8} 张艺源, ^(1,4,5){ }^{1,4,5} 木照雄, ^(2,3,10){ }^{2,3,10} 刘立晓, ^(2,3,10){ }^{2,3,10} 蒋晓宇, ^(1,10){ }^{1,10} 普拉迪普·雷迪, ^(15){ }^{15} 董学达, ^(7,9){ }^{7,9} 徐凡舒, ^(7,8){ }^{7,8}Qiaoran Wang, ^(2,3,10){ }^{2,3,10} Qian Zhao, ^(6){ }^{6} Jinghui Lei, ^(6){ }^{6} Shuhui Sun, ^(13){ }^{13} Ying Jing, ^(6){ }^{6} Jingyi Li, ^(1,4,5,22){ }^{1,4,5,22} Yusheng Cai, ^(1,4,5){ }^{1,4,5} Yanling Fan, ^(2,3){ }^{2,3}Kaowen Yan, ^(1,4,5){ }^{1,4,5} Yaobin Jing, ^(1,4,5,14){ }^{1,4,5,14} Amin Haghani, ^(15){ }^{15} Mengen Xing, ^(19){ }^{19} Xuan Zhang, ^(20,24){ }^{20,24} Guodong Zhu, ^(21,24){ }^{21,24}Weihong Song, ^(19,24){ }^{19,24} Steve Horvath, ^(15,24){ }^{15,24} Concepcion Rodriguez Esteban, ^(15,24){ }^{15,24} Moshi Song, ^(1,5,10,24){ }^{1,5,10,24} Si Wang, ^(6,22,24){ }^{6,22,24}Guoguang Zhao, ^(16,17,18,24){ }^{16,17,18,24} Wei Li, ^(1,4,5,10,24){ }^{1,4,5,10,24} Juan Carlos Izpisua Belmonte, ^(15,24){ }^{15,24} Jing Qu, ^(1,4,5,10,13,22,**){ }^{1,4,5,10,13,22, *} 郭光赵, ^(16,17,18,24){ }^{16,17,18,24} 李伟, ^(1,4,5,10,24){ }^{1,4,5,10,24} 胡安·卡洛斯·伊兹皮苏亚·贝尔蒙特, ^(15,24){ }^{15,24} 曲静, ^(1,4,5,10,13,22,**){ }^{1,4,5,10,13,22, *}Weiqi Zhang, ^(2,3,4,5,10,22,**){ }^{2,3,4,5,10,22, *} and Guang-Hui Liu ^(1,4,5,6,10,22,25,**){ }^{1,4,5,6,10,22,25, *}^(1){ }^{1} Key Laboratory of Organ Regeneration and Reconstruction, State Key Laboratory of Membrane Biology, State Key Laboratory of Stem Cell and Reproductive Biology, Institute of Zoology, Chinese Academy of Sciences, Beijing 100101, China ^(2){ }^{2} China National Center for Bioinformation, Beijing, China ^(3){ }^{3} Beijing Institute of Genomics, Chinese Academy of Sciences, Beijing, China ^(4){ }^{4} Institute for Stem Cell and Regeneration, CAS, Beijing 100101, China ^(5){ }^{5} Beijing Institute for Stem Cell and Regenerative Medicine, Beijing 100101, China ^(6){ }^{6} National Clinical Research Center for Geriatric Disorders, Aging Translational Medicine Center, International Center for Aging and Cancer, Xuanwu Hospital Capital Medical University, Beijing 100053, China ^(7){ }^{7} State Key Laboratory of Brain and Cognitive Science, Institute of Biophysics, Chinese Academy of Sciences, Beijing 100101, China ^(8){ }^{8} College of Life Sciences, University of Chinese Academy of Sciences, Beijing 100049, China ^(9){ }^{9} Sino-Danish College, University of Chinese Academy of Sciences, Beijing 100049, China ^(10){ }^{10} University of Chinese Academy of Sciences, Beijing 100049, China ^(11){ }^{11} Hospital of Stomatology, Guanghua School of Stomatology, Guangdong Provincial Key Laboratory of Stomatology, Sun Yat-Sen University, Guangzhou 510060, China ^(12){ }^{12} Chongqing Fifth People's Hospital, Chongqing 400060, China ^(13){ }^{13} Beijing Institute of Heart, Lung and Blood Vessel Diseases, Beijing Anzhen Hospital, Capital Medical University, Beijing 100029, China ^(14){ }^{14} International Center for Aging and Cancer, Hainan Medical University, Haikou 571199, China ^(15){ }^{15} Altos Labs San Diego Institute of Science, San Diego, CA, USA ^(16){ }^{16} Department of Neurosurgery, Xuanwu Hospital Capital Medical University, Beijing 100053, China ^(17){ }^{17} National Medical Center for Neurological Diseases, Beijing 100053, China ^(18){ }^{18} Beijing Municipal Geriatric Medical Research Center, Beijing 100053, China ^(19){ }^{19} Oujiang Laboratory, Center for Geriatric Medicine and Institute of Aging, Key Laboratory of Alzheimer's Disease of Zhejiang Province, Zhejiang Provincial Clinical Research for Mental Disorders, The First-Affiliated Hospital, Wenzhou Medical University, Wenzhou, Zhejiang 325035, China ^(20){ }^{20} Department of Rheumatology, Beijing Hospital, National Center of Gerontology, Institute of Geriatric Medicine, Clinical Immunology Center, Chinese Academy of Medical Sciences, Beijing 100730, China ^(21){ }^{21} Institute of Gerontology, Guangzhou Geriatric Hospital, Guangzhou Medical University, Guangzhou, China ^(22){ }^{22} Aging Biomarker Consortium (ABC), Beijing 100101, China ^(23){ }^{23} These authors contributed equally ^(24){ }^{24} Senior author ^(25){ }^{25} Lead contact *Correspondence: qujing@ioz.ac.cn (J.Q.), zhangwq@big.ac.cn (W.Z.), ghliu@ioz.ac.cn (G.-H.L.) https://doi.org/10.1016/j.cell.2024.08.021
Abstract
SUMMARY In a rigorous 40-month study, we evaluated the geroprotective effects of metformin on adult male cynomolgus monkeys, addressing a gap in primate aging research. The study encompassed a comprehensive suite of physiological, imaging, histological, and molecular evaluations, substantiating metformin’s influence on delaying age-related phenotypes at the organismal level. Specifically, we leveraged pan-tissue transcriptomics, DNA methylomics, plasma proteomics, and metabolomics to develop innovative monkey aging clocks and applied these to gauge metformin’s effects on aging. The results highlighted a significant slowing of aging indicators, notably a roughly 6-year regression in brain aging. Metformin exerts a substantial neuroprotective effect, preserving brain structure and enhancing cognitive ability. The geroprotective effects on primate neurons were partially mediated by the activation of Nrf2, a transcription factor with anti-oxidative capabilities. Our research pioneers the systemic reduction of multi-dimensional biological age in primates through metformin, paving the way for advancing pharmaceutical strategies against human aging.
INTRODUCTION
Aging, a progressive process, induces tissue dysfunction and physiological deterioration, culminating in the emergence of age-related conditions, including neurodegenerative, cardiovascular, and diabetic disorders. ^(1-5){ }^{1-5} Notably, accumulating evidence suggests that aspects of aging are malleable in rodents through interventions including small-molecule drugs, genetic manipulations, exercise, and diet. ^(6-11){ }^{6-11} Metformin, a first-line treatment for type 2 diabetes, developed from a guanidine derivative in Galega officinalis, shows promise in slowing physiological aging across a range of models, including nematodes, fruit flies, and rodents. ^(12-21){ }^{12-21} Previous studies, including our own, also demonstrated metformin’s potential to alleviate senescence in human diploid cells. ^(21-25){ }^{21-25} Moreover, retrospective studies indicate that metformin appears to reduce the mortality rate in diabetic patients. ^(26-28){ }^{26-28} However, whether metformin can delay aging and ameliorate aging-related tissue degeneration in primates remains unclear.
Leveraging high-throughput omics technologies, a cuttingedge toolkit for gauging biological aging has emerged, affording us the ability to precisely quantify aging rates at a molecular level. ^(29){ }^{29} Machine learning integration of epigenomics, transcriptomics, proteomics, and metabolomics data paves the way for “aging clocks,” offering a means to evaluate the effectiveness of interventions against aging. ^(30-36){ }^{30-36} Additionally, the evolution of single-cell sequencing technologies enhances our comprehension of the intricate cellular and molecular underpinnings of the aging process and its interventions. ^(37){ }^{37} Yet the potential of metformin to catalyze systemic rejuvenation across various biological dimensions at the pan-tissue level remains to be fully understood.
To explore whether metformin alleviates age-related declines, we conducted a comprehensive, 40-month study assessing adult-onset metformin supplementation in aged primates. Specifically, our investigation encompassed a broad spectrum of analytical techniques, including physiological examinations, medical imaging, pan-tissue histological analysis, organismwide transcriptomics, and single-nucleus RNA sequencing (snRNA-seq). By quantifying these parameters and integrating them into a comprehensive “primate aging clock,” we provided evidence that metformin decelerates the aging process in male cynomolgus monkeys across various tissues. Importantly, our findings highlighted a notable neuroprotective effect of metformin, which was further validated using a human stem cellderived neuronal senescence model. Our results affirm that extended metformin administration mitigates aging in primates, indicating its clinical potential for aging management and disease prevention.
RESULTS
Long-term metformin treatment exhibits geroprotective effects in primates
To assess whether long-term metformin therapy delays aging in healthy primates, we conducted a proof-of-concept study involving male cynomolgus monkeys (Macaca fascicularis) aged between 13 and 16 years, roughly equivalent to approxi-
mately 40-50 years in humans. At the start of the study, monkeys were evenly divided by age and randomly assigned to either the metformin or vehicle treatment groups (hereafter referred to as O-Met and O-Ctrl). The monkeys in the O-Met group were administered a daily dose of 20mg//kg20 \mathrm{mg} / \mathrm{kg} metformin, a standard dosage used in diabetes management for humans, ^(38,39){ }^{38,39} while being maintained under the same environmental and care conditions as the O-Ctrl group (Figure 1A). One participant in the O-Ctrl group succumbed to kidney failure on the 1,126th day of the study, as confirmed by a veterinarian. The remaining monkeys adhered to this regimen for a period of 1,200 days, approximately 3.3 years, which corresponds to about 10 years in humans.
Both groups underwent routine physical examinations every 3 months (Figure 1A). We observed that prolonged administration of metformin was not associated with compromised blood glucose homeostasis, nor did it lead to a reduction in body weight (Table S1). Similarly, we did not detect significant changes in the blood cell composition or the physiological characteristics of urine (Table S1).
We also included two additional control groups: young (3-5 years old) and middle-aged (10-12 years old) male adult cynomolgus monkeys, referred to as the Y -Ctrl group and M-Ctrl group, respectively. Once the O-Met group had completed 1,200 days of metformin treatment, we analyzed all four groups of monkeys for 68 biological parameters. These included morphometric indicators (BMI and organ indices), blood tests (routine blood tests, blood biochemical tests, and hormones), and imaging indicators (computed tomography [CT] scans and magnetic resonance imaging [MRI]) (Figure 1B). Taken together, these results implied the high safety profile of long-term metformin treatment (Table S1). Additionally, we noted that agingassociated periodontal bone loss was mitigated in the O-Met group relative to the O-Ctrl group (Figure 1C).
To evaluate memory, learning, and cognitive flexibility, we employed the Wisconsin General Test Apparatus (WGTA) method. ^(40,41){ }^{40,41} In the delay task, which evaluates memory retention, the O-Met group demonstrated higher accuracy in retrieving food after a delay compared with the O-Ctrl group, suggesting that metformin may enhance memory in aged animals (Figure 1D). Additionally, in the object discrimination task, the O-Met group showed superior learning abilities, indicating metformin’s potential to improve learning in older subjects (Figure 1D). Likewise, in the object reversal learning, the O-Met group displayed enhanced cognitive resilience relative to the O-Ctrl group (Figure 1D).
When we investigated brain morphology using MRI, general linear mixed models (GLMMs) revealed reduced cortical thickness in aged monkeys compared with young ones, particularly in the frontal and temporal lobes (Figures 1E and S1A). In metfor-min-treated aged monkeys, frontal lobe cortical thickness was preserved, with a trend toward increased thickness in the parietal lobe, compared with the O-Ctrl group (Figures 1E and S1A). Consistently, histological examination revealed that metformin treatment enhanced the thickness of the frontal cortex, an area that typically thinned with age in monkeys (Figure 1F). By subdividing the brain into 88 regions using the CHARM5 atlas, ^(42){ }^{42} we identified 9 regions, predominantly in the frontal lobe, with a
