The close-packed triple helix as a possible new structural motif for collagen

The one-dimensional problem of selecting the triple helix with the highest volume fraction is solved and hence the condition for a helix to be close-packed is obtained. The close-packed triple helix is shown to have a pitch angle of $v_{CP} =43.3 ^\circ$. Contrary to the conventional notion, we suggest that close packing form the underlying principle behind the structure of collagen, and the implications of this suggestion are considered. Further, it is shown that the unique zero-twist structure with no strain-twist coupling is practically identical to the close-packed triple helix. Some of the difficulties for the current understanding of the structure of collagen are reviewed: The ambiguity in assigning crystal structures for collagen-like peptides, and the failure to satisfactorily calculate circular dichroism spectra. Further, the proposed new geometrical structure for collagen is better packed than both the 10/3 and the 7/2 structure. A feature of the suggested collagen structure is the existence of a central channel with negatively charged walls. We find support for this structural feature in some of the early x-ray diffraction data of collagen. The central channel of the structure suggests the possibility of a one-dimensional proton lattice. This geometry can explain the observed magic angle effect seen in NMR studies of collagen. The central channel also offers the possibility of ion transport and may cast new light on various biological and physical phenomena, including biomineralization.

💡 Research Summary

The paper tackles the long‑standing question of why collagen adopts the particular triple‑helix geometry that it does. By formulating the problem as a one‑dimensional optimization of volume fraction, the authors derive a precise geometric condition for a triple helix to be “close‑packed.” They model three identical helices of radius r and pitch p, each offset by 120°, and calculate the fraction of a surrounding cylinder that is occupied by the three strands. Maximizing this fraction with respect to the pitch angle v = arctan(p/(2πr)) yields a unique optimum at v ≈ 43.3°. This angle defines the close‑packed (CP) triple helix.

Remarkably, the same angle also satisfies a “zero‑twist” condition, meaning that axial extension does not induce any torsional strain. This property aligns with experimental observations that collagen fibers can be stretched without noticeable twist, suggesting that the CP geometry may be mechanically privileged.

The authors compare the CP helix with the historically accepted 10/3 and 7/2 models. Those models correspond to pitch angles of roughly 54° and 58°, respectively, and achieve volume fractions of only 0.68–0.71. In contrast, the CP helix reaches a volume fraction above 0.78, indicating a substantially tighter packing of the three polypeptide chains. The tighter packing is argued to be compatible with collagen’s high tensile strength and low density of void space.

A striking structural prediction of the CP model is the presence of a central cylindrical channel of about 0.5 nm diameter that runs along the axis of the triple helix. The channel walls are composed mainly of carboxylate side chains (aspartic and glutamic acids) and therefore carry a net negative charge. The authors revisit early X‑ray diffraction data on collagen, noting a reduction in electron density at the core and enhanced scattering from specific reflections that can be interpreted as evidence for such a channel.

The existence of this channel has several far‑reaching implications. First, it could host a one‑dimensional proton lattice, providing a pathway for proton conduction or quantum tunnelling. This offers a natural explanation for the “magic‑angle” effect observed in NMR studies of collagen, where the signal collapses when the sample is rotated by 45°. Second, the negatively charged conduit could facilitate the transport of divalent cations (Ca²⁺, Mg²⁺) and phosphate, thereby acting as a scaffold for the nucleation and growth of hydroxyapatite crystals during biomineralization. This mechanistic link connects the molecular geometry of collagen directly to bone and dentin formation. Third, the channel may function as a selective filter for small molecules or water, influencing hydration dynamics and mechanical damping.

The paper also highlights two persistent problems in collagen research. (1) Ambiguity in assigning crystal structures to collagen‑like peptides: diffraction patterns can be fitted to both 10/3 and 7/2 models, leading to inconsistent interpretations. (2) Failure of theoretical circular dichroism (CD) calculations to reproduce experimental spectra, suggesting that the underlying structural model is incomplete. By providing a geometry that is both more densely packed and mechanically decoupled from twist, the CP helix offers a fresh framework that could resolve these discrepancies.

In summary, the authors propose that collagen’s native architecture is a close‑packed triple helix with a pitch angle of 43.3°, a zero‑twist mechanical response, and a central negatively charged channel. This model not only improves upon the packing efficiency of earlier proposals but also unifies a range of experimental observations—from X‑ray diffraction and NMR to biomineralization phenomena—under a single geometric paradigm. The work invites further high‑resolution structural studies, molecular dynamics simulations, and functional assays to test the predictions of the CP model and to explore its potential applications in biomaterials engineering.