1.2.1 Work Package 1 – Objective 1
Since both objectives rely on the same sample preparation pipeline, they were initiated simultaneously. To date, three sample preparation protocols have been executed (WP1.1 and WP2.1) enabling multiple cryo-FIB milling and data acquisition sessions (WP1.2 and WP2.2). These efforts have yielded hundreds of high-quality tomograms. The datasets include regions enriched in keratin filaments (Objective 1 – WP1) and areas containing lamellar bodies and adjacent vesicles involved in lipid biogenesis (Objective 2 – WP2).
During the development of Objective 1 (WP1), I was advised to attempt an in-plane lift-out—a 90° tilt of the conventional lift-out orientation. This approach aims to produce lamellae aligned with the Z-axis of the skin layers, potentially allowing the imaging of an entire 200 nm slice of the same keratinocyte in the stratum corneum. These cells exhibit varying densities of keratin filament bundles, offering a unique opportunity to obtain multiple tomograms from the same cell or layer. Due to the technical complexity of this approach, only one data acquisition was completed, and WP1.2 remains in progress.
From the datasets already obtained, image processing began immediately (WP1.3). I compared tomogram reconstruction using different software tools: AreTomo, IMOD, and WARP, to determine which was best suited for the processing pipeline. As these programs produce different outcomes, AreTomo was selected for tomogram reconstruction and analysis, while WARP was used to obtain defocus values and perform subtomogram averaging.
Two denoising software packages were also evaluated: IsoNet and cryoCARE. Both were tested on the same dataset to ensure fair comparison. IsoNet’s inclusion of missing-wedge correction yielded the best results. For all software comparisons, the same dataset was used consistently, and the final outcomes were evaluated against the original data to verify that no processing artifacts were introduced. A complete workflow for tomogram reconstruction and denoising was established and updated as new software versions or tools became available. This pipeline is currently being applied to both WP1.3 and WP2.3.
As part of WP1.3 I conducted a comparative evaluation of Amira and Ais software for filament tracing. Both tools are suitable, but they yield different results. Tracings generated with Amira were used for subtomogram averaging and for extracting ultrastructural data that later could be used to measure different parameters such as filament number, length, curvature, orientation, and more. In contrast, Ais proved more suitable for segmenting datasets in both WP1.3 and WP2.3. This analysis is still ongoing.
Subtomogram averaging of keratin filaments was also initiated. For this, tomograms were reconstructed using WARP, filaments were traced in Amira, and coordinates were extracted. Particle extraction, classification, and refinement were performed using Linux WARP, Relion 5, and Dynamo. This pipeline has not yet been fully optimized, and the analysis is still in progress.
Explain the work carried out in WP1 during the reporting period giving details of the work carried out by each beneficiary/affiliated entity involved.
1.2.2 Work Package 2 – Objective 2
While working on datasets for Objective 1 (WP1.2) an intriguing lipid-based structure composed of multiple stacked lipid membranes was identified. These structures vary in length and thickness and are found exclusively in the stratum corneum layer of the epidermis. To our knowledge, such structures have not been previously described in the literature—likely due to the limitations of conventional room-temperature techniques commonly used in skin research. We have contacted collaborators who specialize in lipid biology to assist in identifying these structures and understanding their potential function.
Several datasets have also been acquired specifically for WP2. The processing pipeline developed and optimized during WP1.2 was applied in parallel to WP2.2. As such, a preliminary visual annotation of lamellar bodies and their distinct contents has been performed, along with the annotation of other notable structures potentially involved in lipid biogenesis. Some tomograms rich in lamellar bodies and other secondary vesicles were segmented using Ais software. Data analysis for Objective 2 is ongoing.
1.2.3 Work Package 3 - Professional Development and Training Activities
As part of my training activities, I successfully completed the courses “Electron Microscopy” and “Scientific Visualization with Adobe Illustrator: From Basics to Publication-Ready Figures”, both organized by ISTA. These courses significantly contributed to my technical skillset and my ability to effectively communicate scientific findings.
1.2.3 Work Package 4 - Communication and Dissemination
I presented this project at the Vienna Region Cryo-EM meeting, where I shared insights into the processing pipeline and software tools used, as well as preliminary results. The feedback received during this event was valuable for further refinement of the project. I also participated in the Vienna Cryo-EM Symposium in both 2023 and 2024. At the 2023 symposium, I presented a poster on this project, which was awarded the Best Poster prize.
This project was also featured in the “ISTA Open Campus” event in 2024 as part of the institute’s outreach activities. For this occasion, I designed a hands-on experiment using UV-sensitive beads aimed at educating young children about the importance of skin protection against ultraviolet (UV) radiation—a major risk factor for various types of skin cancer.
Children were invited to test the effect of different protective items such as T-shirts, sunglasses, sunscreen, and other gear by covering the beads and observing the change in color under sunlight. This simple but effective visual demonstration helped illustrate the varying levels of protection each item offers, raising awareness of the importance of sun safety in an engaging and age-appropriate way.
1.2.3 Work Package 5 - Project Management
I held regular meetings with my supervisor, Prof. Florian Schur, to address all aspects of the project, including experimental planning, data management, and milestone tracking. All experimental data, metadata, and lab notes were systematically organized and archived on our institute’s internal server.
We also discussed the financial aspects of the fellowship, including budgeting and administrative procedures, where I received valuable guidance on best practices for grant and project management.
