{"paper_id":"188a4e6c-ab3e-4f73-be0b-58b525cafa37","body_text":"1 \nFully genetically encoded low-molecular-weight 1 \nprotein tags with defined shapes for direct molecular 2 \nidentification by cryo-electron tomography 3 \nFeng Luo1,2,3, Rong Sun1,2,3, Oliver Chalkley1,2,3, Pingping Li2,4, and Qiangjun Zhou1,2,3* 4 \n1Department of Cell and Developmental Biology, Vanderbilt University, Nashville, USA 5 \n2Vanderbilt Brain Institute, Vanderbilt University, Nashville, USA 6 \n3Center for Structural Biology, Vanderbilt University, Nashville, USA 7 \n4Department of Pharmacology, Vanderbilt University, Nashville, USA 8 \n*Correspondence: qiangjun.zhou@vanderbilt.edu  9 \n 10 \nAbstract 11 \nCryo-electron tomography (cryo-ET) enables three-dimensional visualization of cells in near-12 \nnative states, but direct identification of specific proteins in situ remains challenging due to 13 \ncrowded cellular environments and the low intrinsic contrast of most proteins smaller than ~500 14 \nkDa. Consequently, molecular identification often relies on indirect labeling strategies or bulky 15 \nprobes that can perturb native structures. Here we present a “shape-as-signal” strategy that 16 \nuses fully genetically encoded protein tags with defined shapes as a molecular signal for direct 17 \nidentification by cryo-ET. We designed two single-chain, monomeric, low-molecular-weight tags: 18 \nan extended V-shaped tag (62 kDa) and a compact triangular tag (85 kDa). Both adopt rigid 19 \ngeometries validated by cryo-electron microscopy and remain compatible with fluorescence 20 \nmicroscopy when fused to fluorescent proteins. Their characteristic shapes are readily 21 \nrecognized and computationally detected in vitro. In cells, the V-shaped tag yields clear, non-22 \ndisruptive signals at native locations. These results demonstrate that low-molecular-weight 23 \nprotein tags can be unambiguously detected and assigned in situ within crowded cellular 24 \nenvironments. This single-step genetic tagging strategy enables seamless dual fluorescence 25 \nand electron microscopy without exogenous probes, challenging the assumption that small 26 \nprotein tags are unsuitable for direct cryo-ET identification. More broadly, this approach 27 \nestablishes a scalable and minimally perturbative framework for visual proteomics and paves 28 \nthe way for multiplexed, shape-encoded molecular mapping in intact cells.  29 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 2 \nMain 30 \nLight microscopy (LM) and electron microscopy (EM) reveal how proteins are organized and 31 \nmove in cells. Fluorescence microscopy (FM)—including modern super-resolution methods—32 \ncan localize specific targets with 10-100 nm precision1–7, owing to protein and small-molecule 33 \nfluorophores that enable selective labeling. EM provides a complementary view at higher spatial 34 \nresolution. In particular, cryo-electron tomography (cryo-ET) visualizes cellular ultrastructure in 35 \nthree dimensions (3D) under near-native, vitrified conditions, resolving membranes, cytoskeletal 36 \nelements, and large protein assemblies in intact cells8–14. However, cryo-ET is fundamentally 37 \nlimited by molecular identification. Only large, structurally distinctive complexes—such as 38 \nribosomes (>2.5 MDa), the 26S proteasome (~2 MDa), mitochondrial respiratory 39 \nsupercomplexes (>1 MDa), and cytoskeletal polymers—are recognized directly in tomograms15–40 \n20. These complexes can be determined in situ at sub-nanometer resolution by subtomogram 41 \naveraging (STA) only when they are abundant in cells. However, most proteins are <70 kDa and 42 \npresent at low abundance, making them difficult to identify effectively and precisely. As a result, 43 \ncryo-ET often reveals where a structure resides in the cell, but not what it is. Precisely and 44 \nunambiguously identifying these smaller, low-abundance proteins remains a central limitation of 45 \ncryo-ET.  46 \nMultiple strategies have attempted to overcome this barrier by attaching high-contrast or 47 \nphysically large (>10 nm) markers to proteins of interest through affinity targeting or chemically 48 \ninduced coupling. Recent examples include nanogold particles21, iron-loaded ferritin cages22,23, 49 \nDNA origami “signpost” scaffolds24, and multimeric protein tags such as genetically encoded 50 \nmultimeric particles (GEMs)25. These methods generate visible landmarks principally but 51 \naccompanied with practical constraints: they often require post hoc labeling, have limited 52 \nefficiency in cells, and can generate false positives through off-target binding, therefore limiting 53 \ntheir general use. Cryogenic super-resolution fluorescence imaging improves localization of 54 \ntagged proteins beyond the diffraction limit26–34, but correlation with cryo-ET is still only precise 55 \nto tens of nanometers due to high fluorescence background and alignment error at cryogenic 56 \ntemperatures—typically not sufficient to assign identity to individual molecules26,27,35,36.  57 \nHere, we addressed this molecular identification limitation with a “shape-as-signal” strategy: 58 \ndeveloping a new class of genetically encoded, shape-defined protein tags that are directly 59 \nvisible by cryo-ET. Rather than attaching heavy metal particles or bulky scaffolds, we 60 \nengineered low-molecular-weight, single-chain proteins that fold into rigid, geometrically 61 \ndistinctive 3D shapes intended to be recognizable by morphology (size and shape) alone. We 62 \nengineered two single-chain, monomeric, shape-defined tags—a V-shaped protein (62 kDa) and 63 \na triangular protein (85 kDa)—whose rigid architectures were verified by single-particle cryo-EM. 64 \nUsing ferritin as a visibility benchmark in vitro, both tags produced clear densities in 3D cryo-65 \ntomograms, and STA resolved ferritin cages and tags. Inside the Escherichia coli (E. coli), 66 \nextended V-shaped protein architectures are inherently easier to distinguish around target 67 \nassemblies, whereas more compact triangular designs, although still detectable, are more likely 68 \nto be confused with surrounding punctate densities. In HeLa cells, fusion of both tags 69 \nrespectively to TOM70NTD targeted them to the mitochondrial outer membrane without 70 \ndetectable trafficking or morphological defects; cryo-ET revealed a characteristic V-shaped 71 \ndensity for enabling unambiguous, molecular-resolution (<2 nm) localization using standard 200 72 \nkeV cryo-EM, while triangular tag signals were subtler but size-consistent. Fusion of GFP to the 73 \nV-shaped tag provided dual optical and ultrastructural visibility, enabling broadly applicable 74 \ncorrelative light and electron microscopy (CLEM) without the need for additional physical 75 \ncorrelation steps. Together, these results establish a fully genetically encoded strategy for direct 76 \nprotein identification in cryo-ET and point toward a modular toolbox of shape-defined tags.  77 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 3 \nThis strategy establishes a fundamentally new route for efficient, direct in situ protein labeling 78 \nand opens opportunities to tackle an expanding set of scientific questions that demand both 79 \nprecise molecular mechanisms and intact ultrastructural context—critical areas that have long 80 \nlacked practical solutions. It enables, for example, mapping the spatial arrangement of adaptor 81 \nproteins or even individual protein isoforms at nascent membrane structures such as vesicles, 82 \ndistinguishing closely spaced paralogs within a membrane complex, and tracking assembly 83 \nintermediates inside cells. Because the tags are fully genetically encoded, they can be 84 \nseamlessly combined with standard molecular perturbations (mutants, truncations, rescue 85 \nconstructs), enabling coordinated functional and structural analyses and pushing visual 86 \nproteomics toward single-molecule-level imaging in the native cellular context. 87 \nDesign and in vitro validation of V- and Δ -shaped tags  88 \nTo generate EM-visible protein tags with special geometries, we engineered an extended V-89 \nshaped tag (Fig. 1a-d) and a compact triangular tag (Fig. 1e-h). 90 \nFor the V-shape, we built on a three-helix bundle scaffold37 and introduced rigid turn inspired by 91 \nsterile α  motif (SAM)38 to stabilize the angular junction between two bundles. Two bundles were 92 \nconnected by a rigid α -helical linker to form an extended V structure with a defined angle 93 \n(Extended Data Fig. 1a). AlphaFold239,40 predicted four V-shaped designs with ~12 nm arms and 94 \ninter-arm angles of approximately 60°, 72°, 90°, and 140° (Extended Data Fig. 1b). To preserve 95 \nstructural rigidity while preventing undesired interactions, we neutralized SAM oligomerization 96 \nresidues and turned surface hydrophilicity (Extended Data Fig. 2a). For the triangular tag, we 97 \nadopted a C3-symmetric oligomeric motif41 as a structural template to generate an equilateral 98 \ntriangle with ~6 nm sides (Fig. 1e). Surface residues were optimized for hydrophilicity to 99 \nmaintain solubility (Extended Data Fig. 2b). Thus, the V-shaped protein (~62 kDa) forms 100 \nextended 12 nm structures, whereas the triangular protein (~85 kDa) adopts a compact ~6 nm 101 \ntriangle (Figs. 1a, e and Extended Data Fig. 1b). 102 \nWe expressed and purified all five designed constructs for single-particle cryo-EM analysis 103 \n(SPA) to assess whether they folded as intended (Extended Data Fig. 3). Synthetic genes were 104 \ncloned and expressed in Escherichia coli (E. coli); proteins were purified by Ni-NTA immobilized 105 \nmetal affinity chromatography and analyzed by size-exclusion chromatography (SEC) to 106 \ndetermine oligomeric state. Raw 200-keV cryo-EM micrographs showed well-defined particles 107 \nfor the 72° V-variant and the triangular construct; individual V- and triangular shapes were 108 \ndirectly visible despite their low mass (Fig. 1b, f). As observed by EM and consistent with SEC, 109 \nboth proteins behaved as monomer with no detectable oligomerization (Fig. 1b, f). The other 110 \nthree V variants (60°, 90°, 140°) did not fold into the intended architectures. We designate the 111 \n12-nm 72° V-variant as V12 and the 6-nm triangular construct as Delta6 (Δ 6). SPA 112 \nreconstructions closely matched the designed models (Fig. 1c, d, g, h and Extended Data Fig. 113 \n4), confirming that both tags fold as intended and demonstrating our design strategy in which 114 \nlow-molecular-weight proteins are engineered to adopt defined geometries that enhance EM 115 \nvisibility. As is common for purified proteins, the samples exhibited preferred orientation on EM 116 \ngrids, with both V12 and Δ 6 appearing predominantly in a “top” view (Fig. 1b, c, f, g). 117 \nTagging apoferritin cages in vitro 118 \nAs a visibility benchmark, we fused V12 or Δ 6 tag to E. coli ferritin (FtnA), which naturally 119 \nassembles into a ~12-nm nanocage composed of 24 subunits42. To minimize potential steric 120 \nstress, we designed constructs containing two FtnA copies fused to either V12 (Extended Data 121 \nFig. 5a) or Δ 6 (Extended Data Fig. 5c), such that a fully assembled cage could carry up to 12 122 \ntags. We expected peripheral tag densities surrounding the apoferritin cage both in vitro and in 123 \nsitu (Extended Data Fig. 5b, d). This design also enabled a direct test of whether tagging 124 \nperturbs apoferritin cage assembly. 125 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 4 \nTomograms of purified V12-ferritin revealed spherical cages with additional peripheral densities 126 \nattributable to V12 (Fig. 2a). To further validate tagging, we evaluated several automated 127 \nparticle-picking pipelines43–48 and performed STA46,47,49. Notably, current cryo-ET picking and 128 \nSTA workflows are largely developed and optimized for large complexes (typically >500 kDa), 129 \nwhich limits performance on low-molecular-weight features. Even so, using the deep-learning 130 \nbased program crYOLO43,44 and the template matching and correlation-based package 131 \nPyTom45,50 (Extended Data Fig. 6a, b), particle picking followed by STA yielded independent 132 \naverages for the apoferritin cage and V12 (Fig. 2b, c). The apoferritin cage was readily 133 \nreconstructed to high resolution (5 Å), consistent with its size (~12-nm outer diameter; ~8-nm 134 \ncavity; ~465 kDa) and high symmetry (octahedral symmetry, 432 point group). By contrast, only 135 \n16.3% of V12 picks contributed to a low-resolution average (Fig. 2c, d), underscoring a known 136 \nlimitation: existing cryo-ET picking/STA pipelines, tuned for larger assemblies, struggle with low-137 \nmolecular-weight targets like V12 due to low signal to noise ratio (SNR) and orientation 138 \nambiguity, as well as the inherent missing-wedge in cryo-ET43–45,49,51. 139 \nIn 2D tomographic slices, only views aligned near the V apex display a clear V (Fig. 2a, c); 140 \nwhereas most other orientations appear as two dots or a short line (Fig. 2a, c). In 3D, however, 141 \nthe V shape is evident: the averaged apoferritin cage and V12 volumes fit unambiguously into 142 \nthe tomographic densities, producing a coherent structural model (Fig. 2e and Supplementary 143 \nVideo 1), that confirms intact cage assembly and direct detectability of V12. Slice-wise densities 144 \nagree with 2D projections of the fitted model (Fig. 2e, f), demonstrating that nearly the entire 145 \ntags are visualized across orientations—further clarifying why existing particle picking and STA 146 \nalgorithms struggle with V12 despite its clear visibility in tomograms. 147 \nFor Δ 6-ferritin, tomograms likewise showed peripheral tag densities (Fig. 2g). Automated 148 \npicking (crYOLO, PyTom; Extended Data Fig. 6c, d) and STA yielded independent averages for 149 \nthe cage and Δ 6 (Fig. 2h, l), with a usable-particle fraction of 71.5% for Δ 6 and 43.0% for the 150 \ncage (Fig. 2j), underscoring the compact tag’s strong in vitro performance. Relative to V12-151 \nferritin, the lower cage fraction of the cage in Δ 6–ferritin datasets suggests that the compact Δ 6 152 \ndensity may influence apoferritin cage picking. The averages recapitulated the expected 153 \ngeometries and fit perfectly into tomographic densities (Fig. 2k, l and Supplementary Video 2). 154 \nSlice views revealed triangular densities in top views and one or two discrete spots in side views, 155 \nconsistent with Δ 6 orientation (Fig. 2i, l).  156 \nTagging apoferritin cages in E. coli  157 \nWe next examined whether the V12 and Δ 6 tags were detectable in situ. V12- and Δ 6-tagged 158 \nferritin were expressed in E. coli, and 80–250-nm thick lamellae were prepared by a cryogenic 159 \nfocused ion-beam scanning electron microscope (cryo-FIB-SEM) (Extended Data Fig. 7a, b).  160 \nFor V12-tagged ferritin, tomograms reconstructed with missing-wedge compensation and 161 \ndenoising using IsoNet52 revealed membranes, ribosomes, and numerous ~12-nm nanocages 162 \n(Fig. 3a, b). Template-based particle picking using PyTom (Extended Data Fig. 7c) followed by 163 \nSTA identified apoferritin cages in situ (Fig. 3d). Only a small fraction of particles contributed to 164 \nthe final average (Fig. 3e), underscoring the difficulty of detecting small features in crowded 165 \ntomograms. Notably, close inspection of individual cages revealed extended densities 166 \nconsistent with the expected V-shaped geometry despite the tag’s modest mass (62 kDa). As 167 \nanticipated, existing algorithms did not reliably detect or reconstruct the low-molecular-weight 168 \nV12 tag in this context. Nevertheless, manual inspection consistently revealed V-shaped 169 \ndensities adjacent to nanocages—matching the in vitro structures and demonstrating direct 170 \nrecognition of V12 in situ (Fig. 3c, f, g and Supplementary Video 3).  171 \nFor Δ 6-tagged ferritin, high quality of tomograms of cryo-FIB-milled E. coli likewise revealed 172 \nnanocages (Fig. 3h, i). Around the cages, ~5–6 nm dot-like densities were frequently observed 173 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 5 \n(Fig. 3j-l and Supplementary Video 4), consistent with the compact triangular geometry of Δ 6 174 \nand matching the in vitro structures (Fig. 2g-l). However, because similar punctate features are 175 \nabundant throughout the cytoplasm, individual Δ 6 tags, while detectable, were more prone to 176 \nmisidentification with surrounding densities.  177 \nTogether, these results indicate that both tags could be detected in the crowded bacterial 178 \ncytoplasm, with the extended V12 tag providing a more distinctive and recognizable shape cue 179 \nthan the compact Δ 6 tag.  180 \nDisplay on the mitochondrial surface in HeLa cells 181 \nHaving confirmed the visibility of both tags in bacteria, we next tested their labeling performance 182 \non the mitochondrial surface in mammalian HeLa cells. To target a native membrane, V12 or Δ 6 183 \nwas fused to the N-terminal targeting fragment of (TOM70NTD)25,53 and appended GFP for 184 \nfluorescence readout (Fig. 4a, b). Western blotting with anti-GFP confirmed robust expression of 185 \ntagged constructs (Fig. 4c), and GFP fluorescence colocalized with the mitochondrial marker 186 \nHsp60 (Fig. 4d, m). Consistent results from anti-HA antibody and Mito-Tracker Red staining in 187 \nHeLa cells, together with western blotting in HEK293T cells, confirmed proper expression and 188 \nmitochondrial localization for both tags without detectable interference (Extended Data Fig. 8).  189 \nHeLa cells were transiently transfected; and GFP-positive cells were isolated by fluorescence-190 \nactivated cell sorting (FACS), allowed to attach onto EM grids, and plunge-frozen for cryo-FIB 191 \nmilling (Extended Data Fig. 9). Cryo-fluorescence imaging of the resulting lamellae guided cryo-192 \nET data acquisition and tracking of the tag (Fig. 4e, Extended Data Fig. 9). We reconstructed 193 \nhigh-quality tomograms; after missing-wedge compensation and denoising with IsoNet52, V12-194 \nexpressing cells showed well-resolved mitochondria, ribosomes, and vesicles (Fig. 4f, g). Cryo-195 \nfluorescence correlated with the 3D tomograms, revealing the signal on the mitochondrial 196 \nsurface (Fig. 4e). On the mitochondrial outer membrane, extended densities with the 197 \ncharacteristic V-shaped geometry were clearly visible and annotatable, enabling precise 3D 198 \nmapping of tag distribution (Fig. 4g, h-l and Supplementary Video 5).  199 \nIn mito-Δ 6-GFP expressing cells, ~6 nm dot-like densities were observed on the mitochondrial 200 \nsurface and colocalized with fluorescence, with orientation-dependent appearances consistent 201 \nwith a compact triangular tag (Fig. 4n-u). However, these features were less distinct than those 202 \nof V12 and difficult to assign unambiguously without reference. No V-shaped densities were 203 \ndetected in mito-Δ 6 tomograms, further underscoring the uniquely identifiable morphology of the 204 \nV12 tag. 205 \nTogether, these results demonstrate that the V12 tags produce clear, detectable densities on 206 \nmitochondrial surface in mammalian cells, correlates well with the GFP fluorescence signal. In 207 \ncontrast, the smaller and more compact Δ 6 tag is challenging to resolve in situ without 208 \nsupporting experiments or subtomogram averaging results, consistent with the relative 209 \ndetectability observed in bacterial cells. 210 \nDiscussion 211 \nWe introduce a fully genetically encoded, shape-defined tagging strategy based on a “shape-as-212 \nsignal” principle, enabling direct identification of specific proteins in cryo-electron tomograms 213 \nwithout post hoc labeling or chemical targeting. By encoding geometry rather than contrast, 214 \nthese low-molecular-weight, single-chain, monomeric tags form rigid, distinctive densities that 215 \nare recognizable by eye at the electron microscope and amenable to computational validation. 216 \nThis creates a direct link between molecular identity and ultrastructural context—an essential 217 \nstep toward molecular-resolution maps of macromolecular organization in intact cells and 218 \ntoward routine in situ counting and positioning of individual proteins. 219 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 6 \nCompared with existing approaches such as nanogold labeling21, DNA-origami scaffolds24, or 220 \nmultimeric particles like GEMs25, our tags are fully genetically encoded, small enough to 221 \nminimize perturbation of trafficking or localization, and engineered to fold into unambiguous 3D 222 \ngeometries. Their visibility arises not from contrast enhancement but from distinctive shape, 223 \nanalogous to how cytoskeletal filaments or membrane structures can be recognized in 224 \ntomograms by morphology (size and shape) alone. Tagging ferritin or the mitochondrial outer 225 \nmembrane with either tag did not introduce detectable defects in protein assembly, trafficking, or 226 \nmorphology (Figs. 2-4). Notably, V12 is clearly visible on a standard 200-keV cryo-TEM 227 \n(Glacios) in purified samples, and mammalian cells (Figs. 1, 4)—particularly in 3D tomograms—228 \nbroadening accessibility and underscoring its potential for widespread application.  229 \nThe two prototype designs illustrate a tunable design space. The extended V12 tag produces a 230 \ncharacteristic V-shaped density that is readily detectable in situ on the mitochondrial outer 231 \nmembrane and in the cytoplasm of bacteria. The more compact Δ 6 tag, although less visually 232 \nstriking in cells, is robustly identifiable in vitro. Together, these results suggest that tag geometry 233 \ncan be tailored to experimental needs—for example, maximizing detectability in crowded 234 \ncytoplasm, minimizing footprint on a sensitive target protein, or introducing asymmetry so that 235 \nthe tagged terminus (N- or C-terminal) can be unambiguously assigned. 236 \nBeyond manual annotation, these tags have the potential to support automated analysis. In 237 \nvitro, tagged complexes could be detected by both template matching45,50 and deep-learning 238 \nbased particle picking43, demonstrating feasibility for computational identification (Figs. 2, 3, and 239 \nExtended Data Figs. 6, 7). Extending these approaches in situ should enable automated 240 \nrecognition of specific tagged molecules directly in cells. In particular, developing 3D (not merely 241 \n2D) detection algorithms specialized for V-shaped densities would improve recall and precision 242 \nfor low-molecular-weight features and accelerate both particle picking and subtomogram 243 \naveraging, enabling automated detection and statistical analysis without requiring subtomogram 244 \naveraging.  245 \nIn cells, V12 could be directly recognized in tomograms and correlated with fluorescence signals 246 \nfrom fusion to a fluorescence protein (e.g., GFP), allowing precise 3D mapping of its distribution 247 \non mitochondria. This ability to annotate the tagged protein’s position within its native 248 \nultrastructural environment creates a route to follow how localization changes across conditions 249 \nsuch as signaling states, metabolic stress, or disease-associated mutations. More broadly, this 250 \nbridges the LM-EM resolution gap: light microscopy provides temporal context and molecular 251 \nspecificity, while cryo-ET supplies molecular-resolution ultrastructure in the same cell with the 252 \nexact same V12-FP fusion tag, without relying on CLEM post hoc physical correlation. 253 \nLooking forward, the protein-origami design framework is inherently extensible. Engineering 254 \nadditional tags with distinct, non-overlapping geometries would enable multiplexed labeling of 255 \ndifferent proteins in the same cell, allowing simultaneous mapping of multiple targets in 3D. In 256 \nparallel, incorporation of heavy-atom clusters or tailored mass distributions could further 257 \nimprove detectability and support automated in situ identification. 258 \nThis work is primarily a proof-of-concept demonstration of shape-defined, genetically encoded 259 \nEM tags, and we do not yet use the approach to derive new biological insights. Our experiments 260 \nestablish feasibility in selected test systems, but each future application will require empirical 261 \noptimization of tag placement, linker design, and expression levels, as well as functional 262 \ncontrols to verify that the fusion does not perturb the behavior of the protein of interest—263 \nanalogous to the validation routinely performed for fluorescent protein fusions. In addition, V-264 \ntags are currently identified mainly by visual inspection in cells and by simple template matching 265 \nor deep-learning based particle picking in vitro. Nonetheless, our data show that V-shaped 266 \ndensities are readily identifiable in 3D volumes in situ, suggesting that robust automated 267 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 7 \ndetection is likely achievable. To fully realize large-scale, quantitative “visual proteomics,” 268 \ndedicated 3D detection algorithms tailored to V-shaped densities will need to be developed and 269 \nintegrated into tomogram analysis pipelines. 270 \nIn summary, these shape-specific, genetically encoded EM tags provide proof-of-principle for a 271 \ngeneral strategy to assign molecular identity directly in cryo-electron tomograms, practically 272 \nbridging fluorescence imaging (temporal control, live-cell specificity, etc.) and cryo-ET 273 \n(molecular-resolution ultrastructure), enabling direct tracking of protein localization under 274 \nphysiological and disease-relevant conditions. This approach opens a route to quantitative, 275 \ncontext-aware maps of protein localization, organization, and interaction networks inside intact 276 \ncells, laying the groundwork for truly integrative, in situ structural and functional proteomics. 277 \n 278 \nMaterials & Correspondence 279 \nSupplementary Information is available for this paper. 280 \nCorrespondence and requests for materials should be addressed to Qiangjun Zhou 281 \n(qiangjun.zhou@vanderbilt.edu). 282 \nPeer review information includes the names of reviewers who agree to be cited and is 283 \ncompleted by Nature staff during proofing. 284 \nReprints and permissions information is available at www.nature.com/reprints. 285 \n 286 \nData and code availability 287 \nAll cryo-EM/cryo-ET data will be deposited in EMPIAR (accession to be provided upon 288 \nacceptance). The density maps and structure coordinates have been deposited in the EMDB 289 \nand PDB under accession numbers EMD-73933 and 9Z9D (V12 tag) and EMD-73947 and 9Z9I 290 \n(Δ 6 tag). The original and/or analyzed data sets generated during the current study are 291 \navailable from the corresponding author upon reasonable request.  292 \nThis paper does not report original code. 293 \nAny additional information required to reanalyze the data reported in this paper is available from 294 \nthe lead contact upon request. 295 \n 296 \nAcknowledgements 297 \nWe are grateful to Drs. David Miller, Ege Kavalali, Lisa Monteggia, Borden Lacy, Hassane 298 \nMchaourab, Ian Macara (VU), Eric Skaar (VUMC), Z. Hong Zhou (UCLA) and Stella Sun (Pitt) 299 \nfor insightful discussions. We also thank Drs. Yun-Tao Liu and Hongcheng Fan (UCLA) for their 300 \nsupport with IsoNet processing. EM data collection was performed at the Center for Structural 301 \nBiology Cryo-EM Facility at Vanderbilt University. We acknowledge use of the Glacios cryo-302 \nTEM, which was acquired under NIH award S10 OD030292. Flow cytometry experiments were 303 \ncarried out in the VMC Flow Cytometry Shared Resource, which is supported by the Vanderbilt 304 \nIngram Cancer Center (P30 CA68485) and the Vanderbilt Digestive Disease Research Center 305 \n(DK058404). Cryo-FIB milling was conducted at the Vanderbilt Institute of Nanoscale Science 306 \nand Engineering with technical support from Dr. James McBride. Cryo-CLEM and cell imaging 307 \nstudies were performed in part through the Vanderbilt Cell Imaging Shared Resource, supported 308 \nby NIH grants CA68485, DK20593, DK58404, DK59637, and EY08126. This work was 309 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 8 \nsupported by CDB Destination Postdoc Award to F.L., and grant from the National Institute of 310 \nHealth (R01MH132918 to Q.Z.). 311 \nAuthor contributions  312 \nConceptualization: F.L., Q.Z.; Methodology: F.L., R.S., O.C., P.L., Q.Z.; Investigation: F.L., R.S., 313 \nP.L., Q.Z.; Visualization: F.L., Q.Z.; Funding acquisition: Q.Z.; Project administration: Q.Z.; 314 \nSupervision: Q.Z.; Writing – original draft: F.L., Q.Z.; Writing – review & editing: F.L., R.S., O.C., 315 \nP.L., Q.Z. 316 \n 317 \nDECLARATION OF INTERESTS 318 \nAuthors declare that they have no competing interests. 319 \n 320 \nMethods 321 \nProtein Design and Computational Modeling of V- and Δ -shaped Tags 322 \nAll shaped tags were designed as single-chain proteins with rigid, predefined geometries. We 323 \nused an iterative, AlphaFold2-guided protein-engineering workflow (“protein nanoblocks”/Lego 324 \nstrategy): initial designs were modeled in AlphaFold240,54, inspected for geometry and 325 \nconfidence, and refined through successive design–prediction cycles (Extended Data Fig. 3). All 326 \nsurface residues were tuned for hydrophilicity. Electrostatic surface potentials were calculated in 327 \nPyMOL (APBS plugin) 55 to verify balanced charge distribution across exposed surfaces and to 328 \nreduce the risk of nonspecific interactions or oligomerization. 329 \nFor the V-shaped protein, the V scaffold was derived from a three-helix-bundle (PDB: 4TQL) 330 \nwith the two bundles connected by a rigid turn inspired by sterile α -motif (SAM) domains38,56,57 331 \nand a de novo-designed mini-protein motif58. AlphaFold2 predicted four candidates with inter-332 \narm angles of ~60°, 72°, 90°, and 140° (Extended Data Fig. 1). To maintain solubility and 333 \nprevent oligomerization or undesired interactions, SAM-interface residues were neutralized.  334 \nFor the Δ -shaped protein, we used the same design strategy, Δ 6 was built from a C3-symmetric 335 \ntrimeric scaffold (C3triangle120_C3_A) to form an equilateral triangular assembly (~6 nm per 336 \nside)41. Two short linkers were engineered to concatenate three repeats into a single chain, 337 \npreserving the C3 geometry.  338 \nProtein Expression and Purification 339 \nFor V12 and Δ 6 proteins, codon-optimized genes encoding V12 and Δ 6 were cloned into 340 \npET27b vectors with N-terminal His6 tags for expression in E. coli BL21(DE3) (NEB). Cultures 341 \nwere grown in LB at 37°C to OD600 ≈  0.6, induced with 0.1 mM isopropyl-β -D-thiogalactoside 342 \n(IPTG), and incubated for 12 h at 20°C. Cells were pelleted and resuspended in lysis buffer (20 343 \nmM Tris-HCl pH 8.0, 300 mM NaCl, 10 mM imidazole) supplemented with a protease inhibitor 344 \ncocktail tablet (Roche). After sonication and centrifugation (18,000 × g, 60 min) at 4°C, 345 \nsupernatants were purified by Ni–NTA affinity chromatography (Ni-NTA Agarose, Qiagen), 346 \nanion-exchange chromatography (Resource Q, Cytiva), and size-exclusion chromatography 347 \n(Superdex 200 Increase 10/300 GL, Cytiva) in 20 mM Tris-HCl pH 8.0, 300 mM NaCl. Protein 348 \nfractions were verified by SDS-PAGE and concentrated to ~0.5 mg/mL for cryo-EM. 349 \nFor V12-ferritin and Δ 6-ferritin nanocages, the E. coli ferritin (ftnA) gene was fused at its N 350 \nterminus to either V12 or Δ 6 via a flexible linker and were cloned into pJ414 vectors with N-351 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 9 \nterminal His6 tags. Cultures were grown and induced with as above but harvested after 4 h at 352 \n20°C. A portion of each culture (1 mL) was used directly for plunging freezing and cryo-FIB 353 \nmilling. The remaining cells were pelleted, resuspended in lysis buffer (20 mM Tris-HCl pH 7.4, 354 \n300 mM NaCl, 10 mM imidazole), supplemented with a protease inhibitor cocktail tablet (Roche) 355 \nat 4°C. The cells were lysed by sonication, and clarified by centrifugation (18,000 × g, 60 min) at 356 \n4°C. Purification followed the same chromatography workflow as above with the buffer at pH 357 \n7.4. Purified samples were verified by SDS-PAGE and concentrated to ~0.5 mg/mL for cryo-EM. 358 \nSingle particle cryo-electron microscopy 359 \nPurified V12 and Δ 6 proteins were applied to glow-discharged Quantifoil R1.2/1.3 Cu 300-mesh 360 \ngrids and vitrified using a Vitrobot Mark III (FEI) (95% humidity, 4°C, blot time 3 s). Data were 361 \nacquired on a 200-keV Thermo Fisher Glacios TEM equipped with a Falcon 4 direct detector at 362 \n120,000× magnification (pixel size 0.73 Å) with a total dose of 60 e⁻ /Å2 in EER format. Beam 363 \ninduced motion-correction and dose-weighting to compensate for radiation damage over spatial 364 \nfrequencies were performed using Patch Motion correction and Contrast Transfer Function 365 \n(CTF) estimation were performed in cryoSPARC59. Particle picking, two-dimensional (2D) 366 \nclassification, and 3D refinement produced final reconstructions, reached overall resolutions of 367 \n5.7 Å for V12 and 6.8 Å for Δ 6 by gold-standard Fourier shell correlation (FSC) at the 0.143 368 \ncriterion. Both datasets were processed without applying symmetry (C1), allowing unbiased 369 \nreconstruction of the full asymmetric architectures of the tags. 370 \nMammalian cell culture, transfection, and FASC 371 \nHeLa (ATCC, no. CCL-2) and HEK293T (ATCC, no. CRL-3216) were cultured in DMEM (Gibco, 372 \nno. 31053028) supplemented with 10% (v/v) fetal bovine serum (FBS, Gibco, no. A5669701), 373 \nand 1% MEM nonessential amino acids (Gibco, no. 11140-050) at 37°C with 5% CO2.  374 \nFor mitochondrial targeting, TOM70NTD-V12 and TOM70NTD-Δ 6 constructs tagged with GFP or 375 \nHA were cloned into pFUGW backbone under the UBC promoter. TOM70NTD corresponds to 376 \nresidues 1-59 of human TOM70 protein, which mediates outer mitochondrial membrane 377 \nlocalization. 378 \nCells were seeded into 10 cm dishes one day before transfection. At ~70% confluency, 379 \ntransfections were performed using FuGENE 6 (Promega, no. F6-1000) with 5µg of plasmids 380 \nDNA and Opti-MEM (Gibco, no. 31985062) following the manufacturer’s protocol. Two days 381 \npost-transfection, GFP-positive cells were sorted by flow cytometry using a BD FACS Aria III. 382 \nParallel transfections were carried out in 6-well or 24-well plates for immunoblotting and 383 \nimmunofluorescence assays. 384 \nCryo-ET sample preparation 385 \nFor E. coli expressing V12-ferritin and Δ 6-ferritin, E. coli cultures (1mL) expressing V12-ferritin 386 \nor Δ 6-ferritin (described above) were centrifuged at 2500 × g for 5min, washed once with PBS 387 \n(pH 7.4) and resuspended into ~60 µL PBS. Cell suspensions were applied to glow-discharged 388 \nQuantifoil R2/2 Cu 200-mesh grids and plunge-frozen using a Vitrobot Mark III (FEI) at 95% 389 \nhumidity and 24°C with a 3s blot time. 390 \nFor HeLa cell preparation, Gold Quantifoil R2/2 SiO2 film grids were UV-sterilized for 30min per 391 \nside and coated with sterilized 0.05 mg/mL poly-L-lysine (PLL, Sigma-Aldrich, no. P2636-392 \n100MG) in 0.1M borate buffer (pH 8.5; Boric Acid, Sigma-Aldrich, no. B-0252; Borax, Sigma-393 \nAldrich, no. B-9876) overnight at room temperature. Grids were rinsed 3 times with ddH2O and 394 \nequilibrated in culture medium. 395 \nAfter cell sorting, GFP-positive cells were pelleted with 200 × g for 5 min and resuspended in 396 \nmedium containing 4 µM AraC (to prevent division) and HEPES and seeded onto 3-well dishes 397 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 10\n(Culture-Insert 3 Well in 35 mm µ-Dish, ibidi, no. 80366 ) with the amount of ~1× 104 cells per 70 398 \nµL with 2 grids each well. Six hours after attaching, grids cultured with GFP-positive HeLa cells 399 \nwere plunge-frozen in pre-warmed PBS using Leica EM GP2 with one side blotting at 37°C, 95% 400 \nhumidity, 3 s blotting time. 401 \nImmunoblotting and immunofluorescence 402 \nCells were lysed in RIPA buffer (25 mM Tris pH 7.6, 150 mM NaCl, 1% NP-40; Sigma, no. 403 \nR0278) supplemented with protease inhibitors. Lysates were separated by SDS-PAGE using 404 \n4%-20% Mini-PROTEIN TGX Precast Protein Gels (Bio-RAD, no. 4561094) and transferred to 405 \nPVDF membranes. Immunoblotted was performed with anti-GFP (Roche, no. 11814460001, 406 \n1:1,000) or anti-HA (Invitrogen, no. 26183, 1:5,000) primary antibodies, and GAPDH (Cell 407 \nsignaling, no. 2118S, 1: 1,000) served as a loading control. IRDye secondary antibodies (LI-408 \nDOR) were used for detection, and signals were imaged with an Odyssey DLx system (LI-COR). 409 \nFor immunofluorescence, cells were fixed with 4% paraformaldehyde (PFA), permeabilized with 410 \n0.1% Triton X-100, and stained with anti-HA (Invitrogen, no. 26183, 1:500; magenta), anti-411 \nHsp60 (Cell signaling, no. 12165S, 1:200), MitoTracker Red CMXRos (Invitrogen, no. M46752), 412 \nand DAPI (blue). Images were acquired using a Nikon CSU-W1 SoRa confocal microscope and 413 \nNikon SIM system. Colocalization with mitochondria was quantified in FIJI60 using Pearson’s 414 \ncorrelation coefficient. 415 \nCryo-FIB lamella preparation 416 \nCryo-focused ion beam (cryo-FIB) milling was performed using an FEI Helios NanoLab G3 CX 417 \nwith a Quorum PP3010T cryo-SEM system at liquid nitrogen temperature. Prior to milling, 418 \nmetallic platinum was deposited by sputter coating (10 mA, 20 s), followed by a protective layer 419 \nof organometallic platinum applied via the gas injection system (6 mm working distance, 25° 420 \nstage tilting angle and 8s injection).  421 \nTwo notches were first created ~1 μ m away from the lamella to relieve mechanical stress and 422 \nprevent warping or bending during subsequent thinning and transfer. Cells were then milled to 423 \n~1 μ m thickness at a 20° stage tilt using ion beam currents of 0.43 nA and 0.23 nA at 30 keV. 424 \nThe stage was then tilted to 16º, and lamellae were thinned to a target thickness of 400–500 nm 425 \nusing beam currents of 80 pA and 40 pA. Finally polishing was performed at 16° with cross-426 \ncleaning at 23 pA to achieve a final thickness of 100-250 nm. Before unloading, SEM overview 427 \nimage of all lamellae and the corresponding grid was acquired to provide localization references 428 \nfor subsequent cryo-CLEM. Finally, lamellae were sputter-coated with platinum (3 mA, 2 s) to 429 \nminimize charging and beam-induced drift during cryo-ET imaging. 430 \nCryo-correlative light and electron microscopy (Cryo-CLEM)  431 \nCryo-FIB-milled lamellae of HeLa cells expressing TOM70NTD-V12-GFP or TOM70NTD-Δ 6-GFP 432 \nwere imaged using Leica STELLARIS Cryo-confocal microscope. FIB-milled grids were 433 \ntransferred with a Leica EM VCM under fresh liquid nitrogen to limit ice containment.  434 \nLamellae were first located in widefield mode based on overview SEM reference images. 435 \nSubsequently, z-stacks encompassing the entire lamellae and adjacent notches were acquired 436 \nin Lighting mode using 491 nm and 587 nm lasers to capture GFP fluorescence and 437 \nautofluorescence, respectively, for later correlation with TEM search maps. Z-stacks were 438 \nprocessed to generate sum-intensity projections. Correlation between cryo-fluorescence images 439 \nand low-magnification TEM search maps (lamella overviews) was performed using IMOD61,62 440 \nand FIJI60. 441 \nCryo-ET image acquisition 442 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 11 \nFor purified V12-ferritin and Δ 6-ferritin nanocages, the purified samples were applied to glow-443 \ndischarged Quantifoil R2/2 Cu 200 mesh grids and plunge-frozen as described above. Tilt 444 \nseries were collected from -55° to +55° in 5° increments with dose-symmetric tilt scheme (8e-/Å2 445 \nper tilt; total accumulated dose ~184 e-/Å2) on a 300 kV Titan Krios G4 microscope equipped 446 \nwith a Gatan K3 detector and a BioQuantum energy filter. Data were acquired at a nominal 447 \ndefocus of 3-4 µm, using Thermo Fisher Tomography software.  448 \nFor bacterial and mammalian lamellae, the stage was tilted by ±9º to compensate for the final 449 \nmilling angle. Tilt series were collected from -60° to +60° using a dose-symmetric tilt scheme 450 \nwith 2° increments (total dose ~183 e⁻ /Å2). The E. coli lamellae were imaged on a 300 kV Titan 451 \nKrios G4 microscope equipped with a Gatan K3 detector and energy filter, using a defocus of 3-452 \n5 µm and a calibrated pixel size of 1.6 Å. HeLa cell lamellae were first screened by collecting 453 \nlow-magnification search maps for all existing lamellae. Cryo-fluorescence correlation with 454 \nCLEM data was performed as described above to identify regions containing both GFP signal 455 \nand mitochondria for targeted cryo-ET data acquisition. Tilt series were collected on a 200 kV 456 \nThermo Fisher Glacios TEM equipped with a Falcon 4 direct detector, using 4-5 µm defocus, a 457 \n70 µm objective aperture, and a pixel size of 1.5 Å. 458 \nCryo-ET data processing 459 \nFor purified V12-ferritin and Δ 6-ferritin nanocages, tilt series were aligned and reconstructed in 460 \nRELION547,63 with integrated motion correction and CTF correction. Reconstructed tomograms 461 \nwere binned fourfold and processed with IsoNet for missing-wedge compensation and denoising, 462 \nenabling improved model fitting and visualization.  463 \nSubtomogram averaging (STA) was performed using crYOLO43,44 for automatic ferritin cage 464 \npicking and PyTom45 for localization of smaller tag particles. Amond tested approaches, 465 \ncrYOLO43,44 was most effective for large in vitro particles, whereas PyTom45 performed better for 466 \nsmall tag features in vitro and in situ cage detection. Independent refinements of cage and tag 467 \nsubtomograms were carried out in RLION547,64,65, yielding final resolution of 5 Å and 22 Å for 468 \nferritin cage and the V12 tag, respectively, and 6.7 Å and 7.3 Å for ferritin cage and Δ 6 tag, 469 \nrespectively. Averaged densities were fitted into corresponding tomograms using UCSF 470 \nChimeraX66 for visualization, tags detection and structural interpretation. 471 \nFor Bacterial and mammalian cell tomograms, tilt series of E. coli and HeLa cell lamellae were 472 \nmotion corrected with Motioncor367 and reconstructed using IMOD (weighted back-projection 473 \nmode)61,62 and binned fourfold, yielding final pixel size of 6.4 Å (E. coli) and 6 Å (HeLa). The 474 \ntomograms were subsequently processed with IsoNet52 for missing-wedge compensation and 475 \ndenoising, using custom masks generated to focus on regions enriched in ferritin cages or 476 \nmitochondrial membranes and associated tags. Ribosomes, membranes, and ferritin nanocages 477 \nwere segmented using AI-assisted tools in Amira (Thermo Fisher Scientific).  478 \nTag-like densities were identified through manual inspection and validated by docking averaged 479 \ntag models obtained from purified samples into tomographic volumes using ChimeraX66. While 480 \nPyTom45 enabled efficient in situ cage picking, existing algorithms failed to reliably detect the 481 \nsmaller tag densities due to the combination of the missing wedge and the crowded cellular 482 \nenvironment. STA of in situ ferritin cages, performed using Warp46 and RELION563,64, achieved a 483 \nfinal resolution of ~12 Å. 484 \nCurrent algorithmic limitations hinder robust automated identification of small, shape-defined 485 \ntags in situ. Ongoing efforts aim to develop new computational approaches tailored for these 486 \ngeometrically defined tags to enhance their detection and verification within cellular tomograms. 487 \nAlthough technically challenging, such advancements are expected to substantially broaden the 488 \napplicability and usability of both tags in future studies. 489 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 12\nData analysis and visualization 490 \nAll density maps were visualized in UCSF ChimeraX66 and segmented in Amira (Thermo Fisher 491 \nScientific). Electrostatic potential surfaces were rendered in PyMOL with APBS55. Fourier shell 492 \ncorrelation (FSC) was used to estimate resolution68. For 3D modeling, structures were fitted into 493 \ntomograms using ChimeraX66. Figures were prepared in ChimeraX, PyMOL, BioRender 494 \n(Extended Data Figs. 3, 9A), and Adobe Illustrator. 495 \n 496 \nREFERENCES 497 \n1. Gustafsson, M. G. L. Surpassing the lateral resolution limit by a factor of two 498 \nusing structured illumination microscopy. J Microsc 198, 82–87 (2000). 499 \n2. Wichmann, J. & Hell, S. W. Breaking the diffraction resolution limit by stimulated 500 \nemission: stimulated-emission-depletion fluorescence microscopy. Optics Letters, 501 \nVol. 19, Issue 11, pp. 780-782 19, 780–782 (1994). 502 \n3. Klar, T. A. & Hell, S. W. Subdiffraction resolution in far-field fluorescence 503 \nmicroscopy. Optics Letters, Vol. 24, Issue 14, pp. 954-956 24, 954–956 (1999). 504 \n4. Betzig, E. et al. Imaging intracellular fluorescent proteins at nanometer resolution. 505 \nScience 313, 1642–5 (2006). 506 \n5. Rust, M. J., Bates, M. & Zhuang, X. Sub-diffraction-limit imaging by stochastic 507 \noptical reconstruction microscopy (STORM). Nature Methods 2006 3:10 3, 793–508 \n796 (2006). 509 \n6. Schermelleh, L. et al. Super-resolution microscopy demystified. Nature Cell 510 \nBiology 2019 21:1 21, 72–84 (2019). 511 \n7. Sahl, S. J., Hell, S. W. & Jakobs, S. Fluorescence nanoscopy in cell biology. Nat 512 \nRev Mol Cell Biol 18, 685–701 (2017). 513 \n8. Lu č ić , V., Leis, A. & Baumeister, W. Cryo-electron tomography of cells: connecting 514 \nstructure and function. Histochemistry and Cell Biology 2008 130:2 130, 185–196 515 \n(2008). 516 \n9. Turk, M. & Baumeister, W. The promise and the challenges of cryo-electron 517 \ntomography. FEBS Lett 594, 3243–3261 (2020). 518 \n10. Wan, W. & Briggs, J. A. G. Cryo-Electron Tomography and Subtomogram 519 \nAveraging. Methods Enzymol 579, 329–67 (2016). 520 \n11. Schur, F. K. Toward high-resolution in situ structural biology with cryo-electron 521 \ntomography and subtomogram averaging. Curr Opin Struct Biol 58, 1–9 (2019). 522 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 13\n12. McCafferty, C. L. et al.  Integrating cellular electron microscopy with multimodal 523 \ndata to explore biology across space and time. Cell 187, 563–584 (2024). 524 \n13. Medalia, O. et al. Macromolecular Architecture in Eukaryotic Cells Visualized by 525 \nCryoelectron Tomography. Science (1979) 298, 1209–1213 (2002). 526 \n14. Nicastro, D. et al. The molecular architecture of axonemes revealed by 527 \ncryoelectron tomography. Science 313, 944–8 (2006). 528 \n15. Medalia, O. et al. Macromolecular architecture in eukaryotic cells visualized by 529 \ncryoelectron tomography. Science 298, 1209–13 (2002). 530 \n16. Nicastro, D. et al. The molecular architecture of axonemes revealed by 531 \ncryoelectron tomography. Science 313, 944–8 (2006). 532 \n17. Waltz, F. et al. In-cell architecture of the mitochondrial respiratory chain. Science 533 \n387, 1296–1301 (2025). 534 \n18. Asano, S. et al. Proteasomes. A molecular census of 26S proteasomes in intact 535 \nneurons. Science 347, 439–42 (2015). 536 \n19. Xue, L. et al. Visualizing translation dynamics at atomic detail inside a bacterial 537 \ncell. Nature 610, 205–211 (2022). 538 \n20. Pfeffer, S. et al. Dissecting the molecular organization of the translocon-539 \nassociated protein complex. Nat Commun 8, 14516 (2017). 540 \n21. Young, L. N. et al. ExoSloNano: Multi-Modal Nanogold Tags for identification of 541 \nMacromolecules in Live Cells & Cryo-Electron Tomograms. bioRxiv 542 \nhttps://doi.org/10.1101/2024.10.12.617288 (2024) doi:10.1101/2024.10.12.617288. 543 \n22. Wang, Q., Mercogliano, C. P. & Löwe, J. A ferritin-based label for cellular electron 544 \ncryotomography. Structure 19, 147–154 (2011). 545 \n23. Clarke, N. I. & Royle, S. J. FerriTag is a new genetically-encoded inducible tag for 546 \ncorrelative light-electron microscopy. Nat Commun 9, 1–10 (2018). 547 \n24. Silvester, E. et al. DNA origami signposts for identifying proteins on cell 548 \nmembranes by electron cryotomography. Cell 184,  1110--1121.e16 (2021). 549 \n25. Fung, H. K. H. et al. Genetically encoded multimeric tags for subcellular protein 550 \nlocalization in cryo-EM. Nat Methods 20, 1900–1908 (2023). 551 \n26. Last, M. G. F., Voortman, L. M. & Sharp, T. H. Imaging intracellular components in 552 \nsitu using super-resolution cryo-correlative light and electron microscopy. in 553 \nMethods in Cell Biology vol. 187 223–248 (Academic Press Inc., 2024). 554 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 14\n27. Last, M. G. F., Voortman, L. M. & Sharp, T. H. Building a super-resolution 555 \nfluorescence cryomicroscope. in Methods in Cell Biology vol. 187 205–222 556 \n(Academic Press, 2024). 557 \n28. Chang, Y.-W. et al. Correlated cryogenic photoactivated localization microscopy 558 \nand cryo-electron tomography. Nat Methods 11, 737–739 (2014). 559 \n29. Liu, B. et al. Three-dimensional super-resolution protein localization correlated 560 \nwith vitrified cellular context. Sci Rep 5, 13017 (2015). 561 \n30. Tuijtel, M. W., Koster, A. J., Jakobs, S., Faas, F. G. A. & Sharp, T. H. Correlative 562 \ncryo super-resolution light and electron microscopy on mammalian cells using 563 \nfluorescent proteins. Sci Rep 9, 1369 (2019). 564 \n31. Dahlberg, P. D. et al.  Cryogenic single-molecule fluorescence annotations for 565 \nelectron tomography reveal in situ organization of key proteins in Caulobacter. 566 \nProc Natl Acad Sci U S A 117, 13937–13944 (2020). 567 \n32. Wolff, G., Hagen, C., Grünewald, K. & Kaufmann, R. Towards correlative 568 \nsuper‐ resolution fluorescence and electron cryo‐ microscopy. Biol Cell 108, 245–569 \n258 (2016). 570 \n33. Dahlberg, P. D. & Moerner, W. E. Cryogenic Super-Resolution Fluorescence and 571 \nElectron Microscopy Correlated at the Nanoscale. Annu Rev Phys Chem 72, 253–572 \n278 (2021). 573 \n34. Kounatidis, I. et al. 3D Correlative Cryo-Structured Illumination Fluorescence and 574 \nSoft X-ray Microscopy Elucidates Reovirus Intracellular Release Pathway. Cell 575 \n182, 515-530.e17 (2020). 576 \n35. DeRosier, D. J. Where in the cell is my protein? Q Rev Biophys  54, e9 (2021). 577 \n36. Klumpe, S. et al. A modular platform for automated cryo-FIB workflows. Elife 10, 578 \n(2021). 579 \n37. Huang, P. S. et al.  High thermodynamic stability of parametrically designed helical 580 \nbundles. Science (1979) 346, 481–485 (2014). 581 \n38. Di Pietro, S. M., Cascio, D., Feliciano, D., Bowie, J. U. & Payne, G. S. Regulation 582 \nof clathrin adaptor function in endocytosis: novel role for the SAM domain. EMBO 583 \nJ 29, 1033–44 (2010). 584 \n39. Jumper, J. et al. Highly accurate protein structure prediction with AlphaFold. 585 \nNature 596, 583–589 (2021). 586 \n40. Mirdita, M. et al. ColabFold: making protein folding accessible to all. Nat Methods 587 \n19, 679–682 (2022). 588 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 15\n41. Huddy, T. F. et al. Blueprinting extendable nanomaterials with standardized protein 589 \nblocks. Nature 627, 898–904 (2024). 590 \n42. Stillman, T. J. et al. The high-resolution X-ray crystallographic structure of the 591 \nferritin (EcFtnA) of Escherichia coli; comparison with human H ferritin (HuHF) and 592 \nthe structures of the Fe3+ and Zn2+ derivatives11Edited by R. Huber. J Mol Biol 593 \n307, 587–603 (2001). 594 \n43. Wagner, T. et al. SPHIRE-crYOLO is a fast and accurate fully automated particle 595 \npicker for cryo-EM. Commun Biol 2, 218 (2019). 596 \n44. Wagner, T. & Raunser, S. The evolution of SPHIRE-crYOLO particle picking and 597 \nits application in automated cryo-EM processing workflows. Communications 598 \nBiology 2020 3:1 3, 1–5 (2020). 599 \n45. Hrabe, T. et al. PyTom: A python-based toolbox for localization of macromolecules 600 \nin cryo-electron tomograms and subtomogram analysis. J Struct Biol 178, 177–601 \n188 (2012). 602 \n46. Tegunov, D. & Cramer, P . Real-time cryo-electron microscopy data preprocessing 603 \nwith Warp. Nat Methods 16, 1146–1152 (2019). 604 \n47. Bharat, T. A. M. & Scheres, S. H. W. Resolving macromolecular structures from 605 \nelectron cryo-Tomography data using subtomogram averaging in RELION. Nat 606 \nProtoc 11, 2054–2065 (2016). 607 \n48. Rice, G. et al. TomoTwin: generalized 3D localization of macromolecules in cryo-608 \nelectron tomograms with structural data mining. Nat Methods 20, 871–880 (2023). 609 \n49. Castaño-Díez, D., Kudryashev, M., Arheit, M. & Stahlberg, H. Dynamo: A flexible, 610 \nuser-friendly development tool for subtomogram averaging of cryo-EM data in 611 \nhigh-performance computing environments. J Struct Biol 178, 139–151 (2012). 612 \n50. Chaillet, M. L. et al. Extensive Angular Sampling Enables the Sensitive 613 \nLocalization of Macromolecules in Electron Tomograms. Int J Mol Sci 24, 13375 614 \n(2023). 615 \n51. Liu, G. et al. DeepETPicker: Fast and accurate 3D particle picking for cryo-616 \nelectron tomography using weakly supervised deep learning. Nat Commun 15, 1–617 \n15 (2024). 618 \n52. Liu, Y . T. et al. Isotropic reconstruction for electron tomography with deep learning. 619 \nNat\n Commun 13, 1–17 (2022). 620 \n53. Nguyen, T. T. & Voeltz, G. K. An ER phospholipid hydrolase drives ER-associated 621 \nmitochondrial constriction for fission and fusion. Elife 11, (2022). 622 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 16\n54. Jumper, J. et al. Highly accurate protein structure prediction with AlphaFold. 623 \nNature 596, 583–589 (2021). 624 \n55. Baker, N. A., Sept, D., Joseph, S., Holst, M. J. & McCammon, J. A. Electrostatics 625 \nof nanosystems: application to microtubules and the ribosome. Proc Natl Acad Sci 626 \nU S A 98, 10037–41 (2001). 627 \n56. Sayou, C. et al. A SAM oligomerization domain shapes the genomic binding 628 \nlandscape of the LEAFY transcription factor. Nat Commun 7, 11222 (2016). 629 \n57. Trevelyan, S. J. et al. Structure-based mechanism of preferential complex 630 \nformation by apoptosis signal-regulating kinases. Sci Signal 13, (2020). 631 \n58. Rocklin, G. J. et al. Global analysis of protein folding using massively parallel 632 \ndesign, synthesis, and testing. Science 357, 168–175 (2017). 633 \n59. Punjani, A., Rubinstein, J. L., Fleet, D. J. & Brubaker, M. A. cryoSPARC: 634 \nalgorithms for rapid unsupervised cryo-EM structure determination. Nat Methods 635 \n14, 290–296 (2017). 636 \n60. Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis. Nat 637 \nMethods 9, 676–82 (2012). 638 \n61. Mastronarde, D. N. Accurate, automatic determination of astigmatism and phase 639 \nwith Ctfplotter in IMOD. J Struct Biol 216, 108057 (2024). 640 \n62. Mastronarde, D. N. & Held, S. R. Automated tilt series alignment and tomographic 641 \nreconstruction in IMOD. J Struct Biol 197, 102–113 (2017). 642 \n63. Burt, A. et al. An image processing pipeline for electron cryo-tomography in 643 \nRELION-5. FEBS Open Bio 14, 1788–1804 (2024). 644 \n64. Bharat, T. A. M. & Scheres, S. H. W. Sub-Tomogram Averaging in RELION . 645 \nNature protocols http://biorxiv.org/lookup/doi/10.1101/030544 (2015) 646 \ndoi:10.1101/030544. 647 \n65. Scheres, S. H. W. RELION: Implementation of a Bayesian approach to cryo-EM 648 \nstructure determination. J Struct Biol 180, 519–530 (2012). 649 \n66. Meng, E. C. et al. UCSF ChimeraX: Tools for structure building and analysis. 650 \nProtein Science 32, (2023). 651 \n67. Zheng, S. Q. et al. MotionCor2: anisotropic correction of beam-induced motion for 652 \ni\nmproved cryo-electron microscopy. Nat Methods 14, 331–332 (2017). 653 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 17\n68. Rosenthal, P. B. & Henderson, R. Optimal Determination of Particle Orientation, 654 \nAbsolute Hand, and Contrast Loss in Single-particle Electron Cryomicroscopy. J 655 \nMol Biol 333, 721–745 (2003). 656 \n  657 \n 658 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 1 \n 1 \nFig. 1. Design and structural characterization of V- and Δ -shaped protein tags.  2 \n(a) Design of V-shaped tag (V12) predicted by AlphaFold2. Two three-helix arms connected by 3 \na rigid SAM-turn motif, forming an angle of ~72° with an arm length of ~12 nm.  4 \n(b) Cryo-EM micrograph of purified V12 collected on a 200-keV Glacios cryo-TEM. Orange 5 \nboxes mark representative V-shaped particles; enlarged views are shown at right.  6 \n(c) Representative 2D class averages showing the characteristic V-shaped architecture, mostly 7 \nin top view. Scale bar, 10 nm.  8 \n(d) Cryo-EM density map of extended V12 (62 kDa) with the predicted model fitted into the 9 \ndensity.  10 \n(e) Design of the compact Δ 6 tag predicted by AlphaFold2, consisting of three copies of C3-11 \nsymmetric trimeric motif assembly ~6 nm in diameter.  12 \n(f) Cryo-EM micrograph of purified Δ 6 collected on a 200-keV Glacios cryo-TEM. Blue boxes 13 \nmark individual triangular particles; enlarged views are shown at right.   14 \n(g) 2D class averages of Δ 6 showing compact triangular geometries. Predominant top views are 15 \nshown; side views are indicated by red boxes. Scale bar, 5 nm.  16 \n(h) Cryo-EM density map and fitted predicted model of the compact Δ 6 (85 kDa) reveal the 17 \nexpected triangular architecture. 18 \n 19 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 2 \nFig. 2. In vitro visualization and analysis of V12- and Δ 6-tagged ferritin nanocages. 20 \n(a) Representative cryo-electron tomogram slice of purified V12-tagged ferritin nanocages. 21 \nInsets show enlarged regions highlighting individual cages and associated V-shaped densities. 22 \nScale bar, 20 nm (left) and 10 nm (right). 23 \n(b, c) STA structures and representative orientated slice views of the ferritin cage (b) and the 24 \nV12 tag (c), each reconstructed independently from purified tomograms. 25 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 3 \n(d) Fraction of particles retained after classification for V12 (orange) and ferritin cages 26 \n(magenta), illustrating the challenge of identifying small V-shaped tags in crowded tomograms. 27 \n(e) Model of the ferritin cage and V12 tag obtained by STA and fitted into the 3D tomographic 28 \ndensity. 29 \n(f) Comparison of model and tomogram slices. Top, representative model slice corresponding to 30 \nthe tomogram slice; middle, tomogram slice; bottom, fitted model slices showing close 31 \nagreement between model and density. Scale bar, 10 nm. 32 \n(g) Representative cryo-electron tomogram slice of purified Δ 6-tagged ferritin nanocages. Insets 33 \nshow enlarged regions highlighting individual cage and associated compact, triangular densities 34 \nsurrounding the cages corresponding to the Δ 6 tag. Scale bar, 20 nm (left) and 10 nm (right). 35 \n(h, i) STA structures and representative orientated slice views of the ferritin cage (h) and the Δ 6 36 \ntags (i) reconstructed independently from purified tomograms. 37 \n(j) Fractions of particles retained after classification for Δ 6 (blue) and ferritin cages (magenta) 38 \nshowing that the compact triangular tags are more readily identified in vitro but may influence 39 \nthe structural analysis of target protein.  40 \n(k)Model of the ferritin cage and Δ 6 tag obtained by STA and fitted into the 3D tomographic 41 \ndensity. 42 \n(l) Comparison of model and tomogram slices. Top, representative model slice corresponding to 43 \nthe tomogram slice; middle, tomogram slice; bottom, fitted model slices showing close 44 \nagreement between model and density. Scale bar, 10 nm. 45 \n 46 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 4 \n 47 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 5 \nFig. 3. In situ visualization of V12- and Δ 6-tagged ferritin cages in E. coli by cryo-ET. 48 \n(a) Representative cryo-tomographic slice of a FIB-milled E. coli cell expressing V12-tagged 49 \nferritin nanocages. Orange boxes mark examples of nanocages. Scale bar, 20 nm. 50 \n(b) Automated segmentation with Amira showing apo-ferritin cages (pink), ribosomes (green) 51 \nwithin the cytoplasm and cell membranes (grey).  52 \n(c) Enlarged views of boxed regions in panel (a) showing peripheral extended densities 53 \ncorresponding to V12 tags with annotated views at right. Scale bar, 10nm. 54 \n(d) STA structure of the apo-ferritin cage from in situ particle picking. 55 \n(e) Fraction of ferritin cage particles retained after in situ classification, illustrating the low yield 56 \nof usable particles in crowded cellular environments. 57 \n(f) Comparison of model and in situ tomogram slices. Top, representative model slice 58 \ncorresponding to the tomogram slice; middle, tomogram slice; bottom, fitted model slices 59 \nshowing close agreement between model and density. Scale bar, 10 nm. 60 \n(g) Model of the ferritin cage and V12 tag obtained by in vitro STA and fitted into the in situ 3D 61 \ntomographic density. 62 \n(h) Representative cryo-tomographic slice of a FIB-milled E. coli cell expressing Δ 6-tagged 63 \nferritin nanocages. Blue boxes mark examples of nanocages. 64 \n(i) Segmentation highlighting apo-ferritin cages (pink). 65 \n(j) Enlarged views of boxed regions in (h) showing compact peripheral densities corresponding 66 \nto Δ 6 tags with the annotation at right. Scale bar, 10nm. 67 \n(k) Comparison of model and in situ tomogram slices. Top, representative model slice 68 \ncorresponding to the tomogram slice; middle, tomogram slice; bottom, fitted model slices 69 \nshowing close agreement between model and density. Scale bar, 10 nm. 70 \n(l) Model of the ferritin cage and Δ 6 tag obtained by in vitro STA and fitted into the in situ 3D 71 \ntomographic density. 72 \n 73 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 6 \n74 \nFig. 4. Mitochondrial surface display of V12- and Δ 6-tagged TOM70NTD fusion proteins in 75 \nHeLa cells. 76 \n(a, b) Schematic of TOM70NTD-V12-GFP and TOM70NTD-Δ 6-GFP constructs. The TOM70 N-77 \nterminal domain (TOM70NTD) anchors to the mitochondrial outer membrane (OM), positioning 78 \nthe V12 or Δ 6 tags on the cytosolic face. 79 \n(c) Immunoblot of HeLa cell lysates expressing TOM70NTD-V12-GFP or TOM70NTD-Δ 6-GFP 80 \nprobed with anti-GFP antibody. GAPDH served as a loading control. 81 \n(d and m) Confocal fluorescence images showing mitochondrial localization of TOM70NTD-V12-82 \nGFP (d) and TOM70NTD-Δ 6-GFP (m). GFP signal colocalizes with the mitochondrial marker 83 \nHsp60 (Pearson’s R = 0.61 and 0.89, respectively). Scale bars, 10 μ m. 84 \n(e) Cryo-correlative light and electron microscopy (cryo-CLEM) of TOM70NTD-V12-GFP cell. 85 \nFluorescence overlay shows GFP colocalized mitochondria on a FIB-milled lamella.  86 \n(f) Tomographic slice of the corresponding region showing mitochondria, ribosomes, and 87 \ncytosolic features; orange arrowheads indicate V-shaped densities.  88 \n(g) Segmented tomogram showing mitochondria (green), ribosomes (red), and V12-tag 89 \ndensities (yellow). In panels e-g, purple asterisks (*) mark the same mitochondrial cristae, and 90 \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint \n\n 7 \npurple hash symbols (#) mark the same vesicle. 91 \n(h-l) Enlarged tomographic slices showing surface V12-tag densities (orange arrowheads) along 92 \nthe mitochondrial outer membrane (OM, green lines) and annotated V12 (yellow). IM, inner 93 \nmembrane. Scale bar, 10 nm. 94 \n(n) Cryo-CLEM of TOM70NTD-Δ 6-GFP cell. Fluorescence overlay shows GFP colocalized 95 \nmitochondria on a FIB-milled lamella. 96 \n(o) Tomographic slice of the corresponding region showing mitochondrion and cytosolic 97 \nfeatures; blue arrowheads indicate triangular-shaped densities. In panels N and O, purple 98 \nasterisks (*) mark the same mitochondrion. 99 \n(p-u) Enlarged tomographic slices showing compact Δ 6-tag densities (blue arrowheads) on the 100 \nmitochondrial outer membrane (green lines) of TOM70NTD-Δ 6-GFP cells. OM, outer membrane; 101 \nIM, inner membrane. Scale bar, 10 nm. 102 \n 103 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted January 20, 2026. ; https://doi.org/10.64898/2026.01.16.700029doi: bioRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}