Unveiling the stochastic nature of human heteropolymer ferritin self-assembly mechanism

Abstract

Despite ferritin's critical role in regulating cellular and systemic iron levels, our understanding of the structure and assembly mechanism of isoferritins, discovered over eight decades ago, remains limited. Unveiling how the composition and molecular architecture of hetero-oligomeric ferritins confer distinct functionality to isoferritins is essential to understanding how the structural intricacies of H and L subunits influence their interactions with cellular machinery. In this study, ferritin heteropolymers with specific H to L subunit ratios were synthesized using a uniquely engineered plasmid design, followed by high-resolution cryo-electron microscopy analysis and deep learning-based amino acid modeling. Our structural examination revealed unique architectural features during the self-assembly mechanism of heteropolymer ferritins and demonstrated a significant preference for H-L heterodimer formation over H-H or L-L homodimers. Unexpectedly, while dimers seem essential building blocks in the protein self-assembly process, the overall mechanism of ferritin self-assembly is observed to proceed randomly through diverse pathways. The physiological significance of these findings is discussed including how ferritin microheterogeneity could represent a tissue-specific adaptation process that imparts distinctive tissue-specific functions to isoferritins.

Keywords: cryo‐EM; ferritin microheterogeneity; ferritin subunits; human heteropolymer ferritin; isoferritins; self‐assembly mechanism.

FIGURE 1

(a) Structure of a ferritin molecule showing (a) Fe²⁺ ions entrance through the threefold channel, oxidation at di‐iron centers, and inorganic Fe³⁺ iron core deposition inside the ferritin inner cavity, (b) the five helices (A‐E) of one bundle monomer, and the inter‐subunit interactions of each subunit at the C2 (dimer), C3 (trimer), and C4 (tetramer) symmetry interfaces. The figure is generated using UCSF ChimeraX software (version 1.7.1) (Meng et al., 2023) and ferritin PDB:7jgk.

FIGURE 2

Different proposed mechanisms of ferritin self‐assembly. The left schematic illustrates three distinct assembly pathways (1, 2, and 3) involving various configurations of ferritin subunits (i.e., dimers, trimers tetramers, hexamers, octamers, 12‐mer), culminating in the formation of a 24‐mer assembly. On the right, one configuration highlights the favored interaction between H and L heterodimers, resulting in the assembly of a 24‐mer human heteropolymer ferritin structure. Due to the preference for H/L heterodimer formation and the specific number of threefold and fourfold channels formed (as shown in Table 1), only certain structures are permissible, despite the potential for different spatial arrangements of H and L subunits.

FIGURE 3

Electron micrographs of H‐rich (70%H:30%L, left) and L‐rich (38%H:64%L, right) ferritins.

FIGURE 4

Top 40 out of 100 classes are shown as an example for the H‐rich ferritin sample (70%H:30%L). Particles with no dense iron density were selected based on their 2D class averages (red box).

FIGURE 5

Cryo‐EM reconstruction of (a) H‐rich ferritin (70%H:30%L) with symmetry imposed, and of (c) L‐rich ferritin (38%H:62%L) without symmetry for modeling. (b) Overlay of published monomeric structures of ferritin heavy H‐chain (blue, PDB 7jgk) and light L‐chain (red, PDB 6tsj) with the main location of structural divergence at the (D–E) loop circled in green. (d) Cryo‐EM map of one subunit, fitted with H‐chain model (blue) or L‐chain model (red). (e) Example of side‐chain densities at 2.1 Å resolution of the H‐rich ferritin sample showing hybrid densities of both heavy chain (magenta) and light chain (green). (f, g) Gold standard fourier shell correlation (GSFSC) curves for H‐rich (f) and L‐rich (g) samples.

FIGURE 6

Comparative analysis of raw DAQ scores for various ferritin EM‐maps with different H and L compositions. DAQ scores for each residue are colored red when DAQ score < −1.0 and blue for DAQ score >1.0. Blue indicates high DAQ scores reflecting model‐map alignment, while red denotes low DAQ scores indicating disagreement. The DAQ value below each EM map represents the average DAQ score.

FIGURE 7

Distribution curves of ferritin species with their average H and L subunits composition displayed at the peak of the curve, extending to +/− 20% variation on each side of the curve (i.e., +/− up to 2H subunits/protein). (Left) H‐rich ferritin (70%H:30%L or 17H:7L), (Right) L‐rich ferritin (38%H:62%L or 9H:15L). The smoothed curves were generated using the kernel density estimation technique. The H‐monomers are depicted in cyan while the L‐monomers in red.

FIGURE 8

Depiction of the predicted H‐ and L‐chain distribution in two heteropolymer ferritin samples, an H‐rich ferritin (70%H:30%L or 17H:7L) and an L‐rich ferritin (38%H:62%L or 9H:15L) using the DAQ score and the Kolmogorov–Smirnov two‐sample test.

FIGURE 9

Probability tree illustrating every possible outcome originating from the specified H:L ratio representing +/− 20% variation from the parent H‐rich (17L:7H) and the L‐rich (9H:15L) heteropolymer ferritin samples. One valid path for the L‐rich Sample (11H:13L) is highlighted in red until 1H:1L is reached (See text for more details).

FIGURE 10

Example illustration of an H‐rich heteropolymer ferritin (15H:9L) self‐assembly mechanism. This ferritin type is similar to that found in the heart or the brain. Alternative mechanisms and conformations exist, all leading to the same final 24‐mer structures, in accordance with the data in Table 1. In this diagram, H‐monomers are depicted in blue, while L‐monomers are in red. Each progression in this self‐assembly process builds upon the previous step by incorporating new H‐L dimers, maintaining the same number of threefold or fourfold channels identified in Table 1. Dotted arrows in steps four and five represent multiple consecutive additions of HH, LL, or HL dimers to achieve the depicted structures.