Mitochondria are essential constituents of every eukaryotic cell. Like many other cellular compartments the molecular structure of the secluded interior of mitochondria is difficult to study by conventional structural biology approaches. Recent revolutionary technological and methodological developments now enable detailed structural studies of macromolecular complexes in their native settings using cryo-electron tomography (CET). This approach allows studying processes that are difficult to analyze in isolation, in particular those involving membrane-associated complexes, and it can reveal the native supramolecular organization of macromolecular machines. Mitochondria with their limited proteome of ~1,000 proteins and their restricted spatial dimensions are uniquely suited to establish a groundbreaking ‘holistic’ approach to structural biology.
Key to mining biological information from CET images (tomograms), which have a very low signal-to-noise ratio, is computational analysis, which is my specialty. I propose to develop a computational workflow for in situ structural biology, involving localization of macromolecules in tomograms with high specificity and determining their high-resolution structure by subtomogram averaging. This approach will be applied to elucidate the biogenesis of proteins of the inner mitochondrial membrane (IMM) in yeast and humans. Firstly, I will investigate the translation of mitochondrially-coded IMM proteins, including the conformational cycle of the mitoribosome during translation, the identification and structural characterization of transiently binding mitoribosomal cofactors and the mitoribosome-associated IMM-insertion machinery for nascent peptides. Furthermore, I will analyze the supramolecular organization of mitochondrial DNA, the mRNA processing machinery, mitoribosomes and the translation products in isolated mitochondria and in whole cells. Lastly, I will address how the majority of IMM proteins translocate across the outer mitochondrial membrane (OMM) during their synthesis in the cytosol. This project will be the blueprint for a new approach to structural biology of membrane-associated
processes. It will contribute to our mechanistic understanding of age-related pathologies, including neurodegenerative diseases, diabetes mellitus and cancer.

The Endoplasmic Reticulum (ER) membrane in all eukaryotic cells has an intricate protein network that facilitates protein biogenesis and homeostasis. The molecular complexity and sophisticated regulation of this machinery favours studying it in its native microenvironment by novel approaches. Cryo-electron tomography (CET) allows 3D imaging of membrane-associated complexes in their native surrounding. Computational analysis of many subtomograms depicting the same type of macromolecule, a technology I pioneered, provides subnanometer resolution insights into different conformations of native complexes.
I propose to leverage CET of cellular and cell-free systems to reveal the molecular details of ER protein biogenesis and homeostasis. In detail, I will study: (a) The structure of the ER translocon, the dynamic gateway for import of nascent proteins into the ER and their maturation. The largest component is the oligosaccharyl transferase complex. (b) Cotranslational ER import, N-glycosylation, chaperone-mediated stabilization and folding as well as oligomerization of established model substrates such as major histocompatibility complex (MHC) class I and II complexes. (c) The degradation of misfolded ER-residing proteins by the cytosolic 26S
proteasome using cytomegalovirus-induced depletion of MHC class I as a model system. (d) The structural changes of the ER-bound translation machinery upon ER stress through IRE1-mediated degradation of mRNA that is specific for ER-targeted proteins. (e) The improved ‘in silico purification’ of different states of
native macromolecules by maximum likelihood subtomogram classification and its application to a-d.
This project will be the blueprint for a new approach to structural biology of membrane-associated processes. It will contribute to our mechanistic understanding of viral immune evasion and glycosylation disorders as well as numerous diseases involving chronic ER stress including diabetes and neurodegenerative diseases.

The aim of this proposal is to develop new computational methodology to enable high-resolution insights into macromolecular complexes from cryo-electron tomographic data. In particular, the proposed algorithms will be key to molecular structural biology in situ, i.e. to study macromolecular complexes in their native settings. The specific aims are:
1. Determination and correction of local beam-induced motion for subtomogram analysis
2. A constrained, regularized Fourier-based iterative tomographic reconstruction algorithm
3. Integration of advanced tomographic reconstruction into subtomogram analysis
4. Documentation of algorithms and appropriate tutorials