Optimizing human SIRT6 protein with deep learning of 3D structures based on maximum lifespan
DOI:
https://doi.org/10.54097/8b4msf08Keywords:
SIRT6, Maximum lifespan, protein design, Convolutional Neural Network, RoseTTAFold, Elastic net, Protein language model, ESM model, ProteinMPNN, Immunofluorescence.Abstract
Aging is the largest risk factor for many diseases, and it is thus of great social and medical importance to delay the appearance of aging traits or promote healthy aging. The protein SIRT6’s function promotes longevity, but it is still unclear whether the function of SIRT6 can be further optimized through protein sequence engineering. Here, we developed a comprehensive framework of sequence optimization methods for functional proteins utilizing both sequence and structure. We used RoseTTAFold to predict the structural variance of SIRT6 sequences across 142 mammalian species and then analyzed the association between the maximum lifespan (MLS), sequence, 3D structures, and amino acid selection of SIRT6. By fine-tuning the ESM model, we developed a model that is capable of accurately predicting the MLS of the species from SIRT6 sequence (Pearson’s r = 0.818, MAE = 8.608 years). We further generated mutant sequences of the human SIRT6 using ProteinMPNN and analyzed different sites’ importance in longevity based on their impact on predicted MLS. 37 sites that play a key role in sequence function were identified. Among them, 20 sites located in the NAD+ binding area and β1-sheet, followed by 9 sites on the α1-helix, were found to have the greatest impact on MLS. We then tested the enzymatic efficiency of 2 novel SIRT6 sequences (with predicted MLS increase of 16.5% and 20.1%) in vitro, and showed that their DNA double-strand break repair efficiency is indeed higher than the human ortholog SIRT6 sequence. Our study demonstrates that despite being highly conservative in evolution, there is still room for optimization in human SIRT6. We not only found a high correlation between MLS and protein sequence or structure, generated experimentally-validated optimized strains, but also identified key sites affecting protein function that have varying importance, thereby providing novel insights into potential anti-aging interventions.
Downloads
References
Chang AY, Skirbekk VF, Tyrovolas S, et al. Measuring population ageing: an analysis of the Global Burden of Disease Study 2017. Lancet Public Health, 2019, 4(3):e159–67.
Aronson L. Healthy Aging Across the Stages of Old Age. Clin Geriatr Med, 2020, 36(4):549–58.
MacRae SL, Croken MM, Calder RB, et al. DNA repair in species with extreme lifespan differences. Aging, 2015, 7(12):1171–82.
Ui A, Chiba N, Yasui A. Relationship among DNA double‐strand break (DSB), DSB repair, and transcription prevents genome instability and cancer. Cancer Sci, 2020, 111(5):1443–51.
Gorbunova V, Seluanov A. DNA double strand break repair, aging and the chromatin connection. Mutat Res Mol Mech Mutagen, 2016, 788:2–6.
López-Otín C, Blasco MA, Partridge L, et al. The Hallmarks of Aging. Cell, 2013, 153(6):1194–217.
Kowalczyk A, Partha R, Clark NL, et al. Pan-mammalian analysis of molecular constraints underlying extended lifespan. eLife, 2020, 9:e51089.
Guo Z, Li P, Ge J, et al. SIRT6 in Aging, Metabolism, Inflammation and Cardiovascular Diseases. Aging Dis, 2022, 13(6):1787–822.
Sharma A, Mahur P, Muthukumaran J, et al. Shedding light on structure, function and regulation of human sirtuins: a comprehensive review. 3 Biotech, 2023, 13(1):29.
Tian X, Firsanov D, Zhang Z, et al. SIRT6 Is Responsible for More Efficient DNA Double-Strand Break Repair in Long-Lived Species. Cell, 2019, 177(3):622-638.e22.
You Y, Liang W. SIRT1 and SIRT6: The role in aging-related diseases. Biochim Biophys Acta BBA - Mol Basis Dis, 2023, 1869(7):166815.
Tasselli L, Zheng W, Chua KF. SIRT6: Novel Mechanisms and Links to Aging and Disease. Trends Endocrinol Metab, 2017, 28(3):168–85.
Roichman A, Elhanati S, Aon MA, et al. Restoration of energy homeostasis by SIRT6 extends healthy lifespan. Nat Commun, 2021, 12(1):3208.
TenNapel MJ, Lynch CF, Burns TL, et al. SIRT6 Minor Allele Genotype Is Associated with >5-Year Decrease in Lifespan in an Aged Cohort. PLoS ONE, 2014, 9(12):e115616.
Tacutu R, Craig T, Budovsky A, et al. Human Ageing Genomic Resources: Integrated databases and tools for the biology and genetics of ageing. Nucleic Acids Res, 2012, 41(D1):D1027–33.
Callaway E. ‘It opens up a whole new universe’: Revolutionary microscopy technique sees individual atoms for first time. Nature, 2020, 582(7811):156–57.
Jumper J, Evans R, Pritzel A, et al. Highly accurate protein structure prediction with AlphaFold. Nature, 2021, 596(7873):583–89.
Baek M, DiMaio F, Anishchenko I, et al. Accurate prediction of protein structures and interactions using a three-track neural network. Science, 2021, 373(6557):871–76.
Mansoor S, Baek M, Juergens D, et al. Accurate Mutation Effect Prediction using RoseTTAFold. 2022[Online] Biochemistry 2022.
Mirdita M, Schütze K, Moriwaki Y, et al. ColabFold: making protein folding accessible to all. Nat Methods, 2022, 19(6):679–82.
Lin Z, Akin H, Rao R, et al. Evolutionary-scale prediction of atomic-level protein structure with a language model. Science, 2023, 379(6637):1123–30.
Rives A, Meier J, Sercu T, et al. Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences. Proc Natl Acad Sci, 2021, 118(15):e2016239118.
Pearce R, Huang X, Omenn GS, et al. De novo protein fold design through sequence-independent fragment assembly simulations. Proc Natl Acad Sci, 2023, 120(4):e2208275120.
Tong X, Liu X, Tan X, et al. Generative Models for De Novo Drug Design. J Med Chem, 2021, 64(19):14011–27.
Giaimo S. Medawar and Hamilton on the selective forces in the evolution of ageing. Hist Philos Life Sci, 2021, 43(4):124.
Dauparas J, Anishchenko I, Bennett N, et al. Robust deep learning–based protein sequence design using ProteinMPNN. Science, 2022, 378(6615):49–56.
Lu AT, Fei Z, Haghani A, et al. Universal DNA methylation age across mammalian tissues. Nat Aging, 2023, 3(9):1144–66.
Edgar RC. Muscle5: High-accuracy alignment ensembles enable unbiased assessments of sequence homology and phylogeny. Nat Commun, 2022, 13(1):6968.
Montesinos López OA, Montesinos López A, Crossa J. Convolutional Neural Networks. In: Multivariate Statistical Machine Learning Methods for Genomic Prediction Springer International Publishing: Cham 2022, pp. 533–77.
Zou H, Hastie T. Regularization and Variable Selection Via the Elastic Net. J R Stat Soc Ser B Stat Methodol, 2005, 67(2):301–20.
Bisong E. Google Colaboratory. In: Building Machine Learning and Deep Learning Models on Google Cloud Platform Apress: Berkeley, CA 2019, pp. 59–64.
Beauharnois JM, Bolívar BE, Welch JT. Sirtuin 6: a review of biological effects and potential therapeutic properties. Mol Biosyst, 2013, 9(7):1789.
Pan PW, Feldman JL, Devries MK, et al. Structure and Biochemical Functions of SIRT6. J Biol Chem, 2011, 286(16):14575–87.
Zhang Z, Tian X, Lu JY, et al. Increased hyaluronan by naked mole-rat Has2 improves healthspan in mice. Nature, 2023, 621(7977):196–205.
Downloads
Published
Issue
Section
License
Copyright (c) 2024 Highlights in Science, Engineering and Technology
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
