Computational Structural Genomics Meets Cryo-EM for the Analyses of Mitochondria-Encoded Enzyme Complexes: A Transformational Approach for Deep Variant Phenotyping in Myelodysplastic Neoplasms

Background: While the central role of genomic alterations in myelodysplastic neoplasms (MDS) is well recognized, most studies have focused on nuclear genome mutations, overlooking the potentially critical importance of the mitochondrial genome. Mitochondria are dynamic organelles essential for maintaining hematopoietic cell homeostasis and differentiation. They produce cellular energy and reactive oxygen species via oxidative phosphorylation, trigger apoptosis, and control heme biosynthesis and iron metabolism. Dysfunction in any of these processes can be pathogenic for MDS. Recently, we conducted whole-genome sequencing to characterize the prognostic landscape of mitochondrial genome in MDS patients receiving allo-HCT. Our findings, using conventional genomic annotation tools, reveal that genetic variants in frequently mutated mitochondrial DNA (mtDNA) genes, such as those encoding mitochondrial complex I (MTC-1) (e.g., MT-ND4L and MT-ND5), are highly prognostic for allo-HCT outcomes in MDS patients. However, the underlying biological mechanisms remain unclear.

Methods: This study applies computational structural genomics, a new translational discipline that integrates conventional genomics, computational biophysics, computational biochemistry, small molecule chemistry, and enhanced multi-omics annotation, to reach a deep characterization of variants. This allows us to gain a better insight into their pathogenicity as well as infer molecular mechanism for their dysfunction. We focus on MTC-1, as a case example. We selected one potentially pathogenic missense variant for each MTC-1 gene from patient, prioritizing the highest predicted damaging variants scored using sequence-based (2D) annotation methods (i.e., MT-ND1 m.3470T>C, p.L55P; MT-ND2 m.5016T>C, p.S183P; MT-ND3 m.10176G>A, p.G40S; MT-ND4 m.11423G>A, p.E222K; MT-ND4L m.10726G>A, p.G86D; MT-ND5 m.12770A>G, p.E145G; and MT-ND6 m.14601G>A, p.P25S). We compared these scores with those derived using 3D assessment of the structure and dynamics of the encoded gene products (i.e., proteins). For this purpose, we used a multi-tier approach that integrates well-established biophysically based metrics (i.e., amino acid local packing density, solvation energy, hydrogen bonds, and salt bridges) and more advanced metrics (i.e., multi-body statistical contact potentials, energetic frustration, and thermodynamic cooperativity).

Results: The precision and scale achieved by our methodology exhibit similar reliability to structural analyses generated by Cryo-EM. By examining the environments around the selected mtDNA variants, we found that MT-ND4:E222 and MT-ND5:E145 likely change the dynamics of the complex. Mutation to a different charge (e.g., E222K) is predicted to alter the molecular dynamic effects of the complex or the electron transport process. We also performed evolutionary coupling analysis with tools from the Marks lab, allosteric path analysis with tools from the Bahar lab, and calculated structural stability using the same tools plus those of Rousseau. This approach identified all seven variants as deleterious. Moreover, allosteric path analysis identified six out of seven variants as deleterious with one neutral call (MT-ND4:E222K) but within 1% of the threshold required for assignment to this category. Structural stability analysis, which is based on molecular mechanic calculations of folding energy identified one variant with neutral (MT-ND5:E145G), one stabilizing (MT-ND4:E222K) and the remaining five variants with highly destabilizing effect.

Conclusions: The comparison of 2D methods for variant classification and functional inferences with our computational structural genomics demonstrates the superior performance of the latter. Our method excels in both predicting potential pathogenicity and providing insights into mechanisms of dysfunction. The primary advantage of our methods is its reliance on computation rather than expensive structural biology equipment like Cryo-EM. Additionally, it surpasses Cryo-EM in its ability to perform extensive molecular dynamic simulations, which are impossible with the former. Finally, the application of this approach suggests that the prognostic effects of these mtDNA variants on MDS patients undergoing allo-HCT are likely due to their deleterious impacts on mitochondrial function, such as respiration.