The ϕPA3 phage nucleus is enclosed by a self-assembling 2D crystalline lattice

Abstract

To protect themselves from host attack, numerous jumbo bacteriophages establish a phage nucleus-a micron-scale, proteinaceous structure encompassing the replicating phage DNA. Bacteriophage and host proteins associated with replication and transcription are concentrated inside the phage nucleus while other phage and host proteins are excluded, including CRISPR-Cas and restriction endonuclease host defense systems. Here, we show that nucleus fragments isolated from ϕPA3 infected Pseudomonas aeruginosa form a 2-dimensional lattice, having p2 or p4 symmetry. We further demonstrate that recombinantly purified primary Phage Nuclear Enclosure (PhuN) protein spontaneously assembles into similar 2D sheets with p2 and p4 symmetry. We resolve the dominant p2 symmetric state to 3.9 Å by cryo-EM. Our structure reveals a two-domain core, organized into quasi-symmetric tetramers. Flexible loops and termini mediate adaptable inter-tetramer contacts that drive subunit assembly into a lattice and enable the adoption of different symmetric states. While the interfaces between subunits are mostly well packed, two are open, forming channels that likely have functional implications for the transport of proteins, mRNA, and small molecules.

Fig. 1. PhuN forms a broad range of assemblies.

a Micrograph showing an unlabeled phage nuclear shell fragment isolated from ΦPA3 infected wild-type Pseudomonas aeruginosa alongside the subsequent p2 (120 Å, 120° unit cell length, angle) and p4 (110 Å, 90° unit cell length, angle) symmetries observed upon 2D classification. Folds and differences in lattice orientation are visible. (n = 1 unlabeled shell isolation experiment). b Anion exchange trace showing extended elution profile and corresponding negative stain EM micrographs. Class I (blue) shows concentrated monomeric species. Class II (turquoise) shows medium-sized assemblies in the 25 nm range. Class III (green) reveals assemblies with elongated shapes and striated textures. c In vitro assembled 2D crystalline arrays at 0° tilt showing three main patches with distinct lattice orientations and gentle curvature. Borders between different crystal patch orientations appear smooth. (n = 6 grids imaged for this publication).

Fig. 2. PhuN assembles into a p2 symmetric lattice.

a Example 0° tilt 2D class average highlighting the p2 symmetry and resulting in four unique interfaces. The lattice is skewed to an ~100° angle. The positioning of the flexible loops that are difficult to resolve in 3D is visible. b Highest resolution map of the core tetramer centered on the Diamond channel as viewed from different angles. c The final PhuN-C subunit model (blue) overlaid with the AlphaFold prediction used as our starting model (gray). Residues 1–18, 276–287, and 556–602 had no corresponding density in our map and were deleted from the model. d The tetrameric model of PhuN-C and PhuN-O fit into the map in b. e The map from b is segmented and positioned to recreate the channels observed in 2D classes (left) paired alongside the corresponding 16-mer-model (right). In both the segmented maps and models, PhuN-O is represented in green while PhuN-C subunits are represented in blue. Unresolved and unmodeled loops are represented with dashed lines to reflect the 2D class while the C-terminal tails are excluded.

Fig. 3. A closer look at PhuN interactions and electrostatic potential.

a Surfaces of PhuN-C (top) and PhuN-O (bottom) colored by electrostatic potential with their corresponding models on the left. b Electrostatic potential coloration of the interior and exterior phage nucleus surfaces created by PhuN assemblies. Surface charge calculation was done at pH 6.5 and includes the C-terminal tail tips. Red corresponds to negative and blue corresponds to positive charge. c Direct comparison of PhuN-C and PhuN-O. The asymmetric subunits in the tetramer are overlaid showing minor differences (light and dark gray) while the interaction with the neighboring β-hairpins differs dramatically, shifting by ~20 Å from near the C-terminus (PhuN-O) towards a flexible loop (PhuN-C).

Fig. 4. ϕPA3 PhuN deletions show defects in shell integration in vivo and self-assembly in vitro.

Live fluorescence microscopy and negative stain EM micrographs of a full-length PhuN as well as the following PhuN deletions: b N-terminal tail (residues 1–37), c Loop (residues 272–291), d C-terminal tail (residues 556–602), and e β-Hairpin (residues 111–126). Fluorescence microscopy images of PhuN or PhuN deletions were collected at 40 min post-infection, deconvolved, and are displayed in green while the DAPI-stained DNA is shown in blue. The 1 µm and 50 nm scale bars apply to all fluorescence and EM panels, respectively. (Independent sample preparation and imaging: fluorescence microscopy n = 3, negative stain EM n = 1). f Rescaled anion exchange traces for the deletions shown in panels a–e. The decreased propensity for PhuN assembly on the column is evidenced by the sharp increase in the primary peak corresponding to the monomeric species (blue box) relative to the assembly peaks (green box). Samples from monomeric peaks were used for negative stain in panels a–e.

Fig. 5. PhuN proteins from ϕPA3 and 201ϕ2-1 share similar structures.

a Top and side views of an overlay of the ϕPA3 (blue) and 201ϕ2-1¹⁷ (green) monomeric subunits. The overlay was determined by fitting the monomers into the blue ϕPA3 EM density shown in panels c–e. The greatest backbone differences between the models reside in the extended positioning of the N-terminal residues, β-Hairpin, and a resolved flexible loop. b The binding pockets utilized by both ϕPA3 and 201ϕ2-1 proteins are the same, as highlighted in pink. This was determined by fitting the dark green 201ϕ2-1 subunit into the ϕPA3 map while retaining the light green subunit to preserve the N-terminal tail interaction. The N-terminal helix is positioned in the same pocket as that of the blue ϕPA3 subunit. c p2 ϕPA3 model with the corresponding ϕPA3 density (blue) and compared to the 201ϕ2-1 tomography density (gray, EMD-25221¹⁷). d Rigid body fitting of the published p4 201ϕ2-1 tetrameric model (PDB 7SQR¹⁷) into the ϕPA3 density (blue) and 201ϕ2-1 tomography volume (gray, EMD-25221¹⁷). The model protrudes outside of the tomography volume. e 201ϕ2-1 model (PDB 7SQR¹⁷) after independent rigid body fitting of each monomer into the ϕPA3 density alongside the improved fit of the resulting model into the 201ϕ2-1 tomography density (EMD-25221¹⁷).

Fig. 6. PhuN assembles through a series of complex and likely adaptive C-terminal tail exchanges exhibiting both p2 and p4 symmetries.

a Micrograph of in vitro assembled PhuN lattices with the p2 (blue) and p4 (green) particles displayed. (n = 8). b Cartoon showing core tetramer with N- and C-terminal tail interactions as well as the loops visible clearly in 2D classes. c Cartoon model tracing the N- and C-terminal tail interactions at all four unique interfaces compared to the p4 equivalent.