Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Jan 31.
Published in final edited form as: Chem Commun (Camb). 2026 Jan 27;62(7):2304–2308. doi: 10.1039/d5cc06640a

β-barrels from short macrocyclic peptides

Viet Thuc Dang a, Michael W Martynowycz b, Dan McElheny a, Andy I Nguyen a,*
PMCID: PMC12857771  NIHMSID: NIHMS2134597  PMID: 41489626

Abstract

β-Barrels are ubiquitous motifs in protein structures, but the fundamental rules underlying their formation are unclear, and their de novo design remains highly challenging. Small peptides that form barrels are especially scarce. Here, we report barrels with the shortest staves (6 residues, ~60% of previous record) and smallest shear number (S = 4) so far, formed from 12-residue macrocyclic peptides. The miniature barrel has anomalous structural features, demonstrated by solution phase and crystallographic characterization; there is a pronounced and essential backbone kink imparted by an achiral residue, N-methylglycine, as well as four structural water molecules stitching the seams of the barrel. These results provide insights into how extremely short sequences could form barrel assemblies.


The β-barrel motif is composed of β-strands that assemble into a cylinder.1,2 β-barrels are found in many important classes of proteins, including transmembrane proteins, TIM barrels, and transport proteins. Additionally, some toxic amyloid oligomers are proposed to have β-barrel structures.35 However, de novo β-barrels are extraordinarily rare and have only been reported recently, underscoring the limited understanding of their sequence–structure relationship.610 There are very few structurally-characterized barrels made from small peptides (<25 amino acids)4,1113 since β-sheet-forming peptides tend to generate 1-D fibrillar structures.14,15 The factors that govern whether sheets become fibrils or oligomers are crucial for understanding protein folding, as well as amyloid-related diseases.16 β-barrels appear to contain irregular features like β-bulges and kinks,8,17 and thus, discovery of minimalistic β-barrel forming sequences is necessary for informing and validating the requirements for folding, as well as for better understanding and predicting amyloid oligomers. Here, we report the serendipitous discovery of 12-residue peptides that assembles into the smallest β-barrel known so far, ~60% the height of the previous record.18,19

The peptides discussed herein were designed with the initial intention to mimic the metal-binding active site of laccase, which is a β-sandwich (Fig. 1(a)).20 We chose to extract and stabilize one “bun” of the β-sandwich from laccase (PDB 1GYC),21 composed of residues 62–66 and 107–111 (Fig. 1(b)). In order to ensure that the excised peptides fold into a β-sheet, we employed the ornithine macrocyclization strategy developed by Nowick and coworkers.22 Ornithine residues are appended onto the N-terminus of each strand and connected to the C-terminus of the opposing antiparallel strand by the sidechain amino group, Nε, forming the macrocycle (Fig. 1(c)). Furthermore, N-methylation of a residue with an edge-facing backbone nitrogen is added to mitigate aggregation into infinite fibrils, and for this, we chose to mutate W65 into sarcosine (N-methyl glycine). Y108 and S62 appeared inconsequential for β-sheet stabilization, and so, they were mutated to Phe to increase the hydrophobicity of the peptide to favor crystallization. This 12-mer peptide design is named MC4H, and the residues are renumbered according to Fig. 1(c). Three other derivatives of MC4H were also synthesized. The first variant is a Sar9 to N-methylalanine (MeA) mutant, which is expected to increase the rigidity of the backbone and strengthen folding into a β-sheet (MC4H-MeA). The other two variations are H8 to πMe-His (MeH) mutants of MC4H and MC4H-MeA, corresponding to MC4H-MeH and MC4H-MeAMeH, respectively, which were originally intended to further direct metal binding to the correct N on H8.

Fig. 1.

Fig. 1

Design of MC4H peptides. (a) Laccase, from which the peptide sequence was extracted from. (b) Excised polypeptide from laccase. (c) Key design features to stabilize the β-sheet secondary structure, prevent aggregation, and increase propensity for crystallization.

Circular dichroism (CD) spectroscopy of the four peptides revealed information on the secondary and quaternary structure (Fig. 2(a)). While peptides containing N-methylalanine, MC4H-MeA and MC4H-MeAMeH, show features typical of β-sheet CD spectra with a minima at ~232 nm, the peptides with sarcosine, MC4H and MC4H-MeH, show a combination of a positive peak at 224 nm and minima at 210 nm that is characteristic of Trp–Trp exciton coupling.2326 Since all sequences only have one Trp residue, the observation of a Trp–Trp exciton coupling implies oligomerization that bring two or more Trp residues in close proximity. The strength of the 224 nm peak is weaker than that of previously reported trpzip peptides, indicating a monomer–oligomer equilibrium.25,26

Fig. 2.

Fig. 2

CD spectroscopy collected at 0.1 mg mL−1 (4.5 × 10−5 M) concentration and buffered with 10 mM HEPES pH 7.0. (a) Spectra of MC4H, MC4H-MeH, MC4H-MeA, and MC4H-MeAMeH at 298 K. (b) Molar ellipticity at 220 nm of MC4H (squares) and MC4H-MeH (triangles) from 5 °C to 90 °C. Data were fit to a two-state dimer-to-monomer model (R2 of 0.98 and 0.99 for MC4H and MC4H-MeH, respectively).27

The oligomerization seen by CD spectroscopy is further supported by diffusion NMR spectroscopy at 280 K, which shows that MC4H and MC4H-MeH are mostly dimeric in solution (average oligomeric state >1.5, or >50% dimer species), whereas MC4H-MeA and MC4H-MeAMeH are closer to being monomeric (Table S6).2730 Variable temperature CD (VT-CD) of both peptides from 278 to 363 K showed significant attenuation of the 220 nm peak, suggesting dissociation of the dimers that removes the Trp–Trp interaction (Fig. 2(b) and Fig. S26). Modelling the data to a two-state homodimer-to-monomer mechanism yielded excellent fits (eqn (S7)(S24)).31 Overall, the fitting results give exothermic reactions with negative enthalpies consistent with a dimerization reaction (Table 1). The melting temperatures, Tm, for the dimers are near room temperature, which is consistent with the oligomeric coefficients seen by DOSY at 280 K. Together, the CD and DOSY NMR spectroscopies suggest Trp–Trp π-stacking drives dimerization of MC4H and MC4H-MeH in the solution state (Fig. 3).

Table 1.

Thermodynamic parameters of dimerization derived from VT-CD spectroscopy

Tm (K) ΔH (kcal mol−1) ΔS (298 K) (kcal K−1) Kdimera (298 K)
MC4H 288 ± 18 −9.2 ± 4.5 −11 2.2 × 104 M−1
MC4H-MeH 306 ± 57 −10.4 ± 3.2 −15 2.2 × 104 M−1
a

See eqn (S17).

Fig. 3.

Fig. 3

(a) Lowest energy NMR structure of MC4H, (b) lowest energy NMR structure of MC4H-MeAMeH. All NMR measurements taken using a sample concentration of 10 mg mL−1 in 0.05 M potassium phosphate buffer pH 6.90 with 10% D2O and 90% H2O at 280 K.

Single crystal structures were obtained for the sarcosine-containing peptides, MC4H and MC4H-MeH, to resolutions of 1.55 Å and 1.25 Å, respectively (Fig. 4 and 5). The crystal structures of both these peptides reveals further assembly of the dimers into tetramers in the solid-state. Their structures are nearly identical, both being tetrameric 8-stranded β-barrels with an unusually small shear number, S, of 4. The samples were grown in different buffer and precipitant conditions (Table S14), which suggests their assembly is robust. The shear number is the sum of residue shifts relative to the starting point after completing a trace on the surface of barrel perpendicular to the strand direction, and it along with the number of strands, uniquely defines barrel topology. MC4H-MeH crystallizes in I222 with two chains in the asymmetric unit (ASU), whereas MC4H crystallizes in P21 space group and contains 8 chains in the ASU.32,33 However, the conformations of all chains in both MC4H and MC4H-MeH are highly similar, with the main differences due to sidechain rotations (Fig. 5). With staves of only ~18 Å that are composed of 6 residues, these are the smallest β-barrels to date, significantly smaller in length than the next two smallest examples, which are 30 Å (11-residue stave) and 30 Å (10-residue stave) (Table 2).4,11

Fig. 4.

Fig. 4

Crystal structure of MC4H (only chains A, B, C, D are shown since the other barrel composed of chains E, F, G, H is highly similar). (a) side view with red spheres being water molecules. (b) W12–W12 interactions and Sar9 highlighted with sphere representations, (c) the top-down view highlighting the packing of hydrophobic residues. (d) Hydrogen bonds involving peptide backbone (dashed lines). Regions folded int β-sheets are highlighted with arrows. (e) Ramachandran plot for chains A–H of MC4H. Residues that are not in the β-sheet region are color-coded by their residues: His4 (blue), Sar9 (pink), and His10 (orange). Orn residues are excluded. (f) Alignment of the 8 chains in the ASU. (g) Comparison of chain B in the crystal structure (blue) with the lowest energy NMR structure (orange).

Fig. 5.

Fig. 5

Crystal structure of MC4H-MeH. (a) the side view, (b) Overlay of MC4H-MeH with MC4H, (c) the top-down view of the overlay of MC4H-MeH with MC4H. (d) Ramachandran plot for chains A and B of MC4H-MeH. Points not in the β-sheet region are color-coded by their residues: His4 (blue), Sar9 (pink), and His10 (orange). Orn residues are excluded.

Table 2.

Structural comparisons of barrels made from small peptides

Ref. Residues in stave Stave length (Å) Strand number (n) Shear number (S)
4 11 30 6  6
4 11 30 6  6
11 10 30 6 12
This work  6 18 8  4

The barrel anatomy is a dimer of dimers, and it has approximately D2 point group symmetry (Fig. 4(d)). The pairs of dimers comprising the barrel (chains A/B and chains C/D in MCH4) each have chains arranged in an antiparallel fashion and W12–W12 intermolecular π-stacking (Fig. 4(b)), consistent with exciton coupling observed in CD spectra (Fig. 2). Furthermore, six intermolecular hydrogen bonds involving residues W12, F1, H2, S3, and H4 hold the pair together (Fig. 4(d)). The concave face of the dimers is hydrophobic, defined by F1, S3, and I7 (Fig. 4(c)). The dimer surface is pronouncedly curved due to the large twisting of each sheet (~52°) caused by Sar9. The curvature enables assembly into a finite barrel rather than an extended fibre. These dimers are brought together in an antiparallel fashion into the final tetrameric assembly by the hydrophobic effect and a different set of hydrogen bonds involving water molecules. Hydrophobic residues F1 and I7 comprise the interior of the barrels (Fig. 4(c)). Dimer–dimer hydrogen bonding from the strands defined by F6, I7, H8, Sar9, and H10 requires two additional bridging water molecules (Fig. 4). The strand bulge and lack of NH moiety imparted by Sar9 disallows canonical antiparallel β-sheet H-bonding, and in this geometry water molecules are needed to complete the H-bonding between staves.

Based on the different types of interactions in the barrel, we surmise that the dimer observed in solution by NMR and CD spectroscopy is the pair that contains the W12–W12 packing and six hydrogen bonds (i.e. chain A/B or chain C/D). The high concentration of salt or precipitants present in the crystallization conditions is proposed to drive the assembly of the dimers into tetramers.

Notably, the secondary structure of the MC4H macrocycles in the solid-state is not uniformly β-sheet throughout the strand regions, and its structure is highly similar to the solution phase NMR structure (Fig. 4(g)). Sar9 in MC4H display unusual (φ, ψ) values of (+126.8°, +137.0°), which is enabled by the achiral Cα center of N-methylglycine (Fig. 4(e)). This ability to adopt dihedrals in the sparsely populated positive quadrant of the Ramachandran plot appears to be critical for barrel formation, as mutation to the more rigid N-methylalanine in the MC4H-MeA series did not crystallize nor display W12–W12 exciton coupling in the CD spectra. “Glycine kinks”, where the Gly residue adopts positive φ angles, have been shown to enable sheet curvature that is conducive for barrel formation with proteins.10,17 While bulges are crucial for barrel geometry, they also disrupt H-bonding. Here, we have shown that water molecules can remedy disruptions caused by bulges by forming new H-bonding patterns between strands. Thus, it may be important to consider structural water molecules in the design or prediction of barrel-forming sequences.

In conclusion, we report crystal structures of β-barrels with the smallest dimensions known to date, with the stave length being 60% of the previously shortest example. The S value of 4 is also unprecedented, especially for an 8-stranded barrel where S is optimally ~12 for proteins.17 This finding adds to Eisenberg’s example of an 11-residue 6-stranded cylindrin barrel that has S of 6, providing evidence that short peptides can form barrel structures with different requirements than those for larger proteins.18 Notably, unlike the canonical barrel architecture, the barrels in this work leverage multiple water molecules to complete the hydrogen bonding between sheets. Interestingly, Eisenberg’s cylindrin barrel also contains one structural water bridging the strands, though those occur at the tips of the barrel, whereas MC4H and MC4H-MeH uses bridging water molecules in the middle of the strands.18 These observations reveal new roles for water in barrel assembly of very short peptides. These unusual features offer concrete exceptions to the currently known general principles of β-barrel formation. We also show that the oligomerization is sensitive to the twist and uniformity of the sheet, which can be triggered by only a single mutation. Though with these reported peptides, the stability of the dimer and tetramer is low, mutations could feasibly increase their stability. Nevertheless, these findings reveal further insights into the minimal principles underlying β-barrel formation and may guide the discovery of small barrel oligomers relevant to human health.

Supplementary Material

Supplementary Information

Footnotes

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information: experimental procedures, additional figures, and tables (PDF). See DOI: https://doi.org/10.1039/d5cc06640a.

Crystal and NMR structures have been deposited to the Protein Data Bank with codes 9OXQ, 9OXP, 9Q63, and 9Q6E.

References

  • 1.Cheng P-N, Pham JD and Nowick JS, J. Am. Chem. Soc, 2013, 135, 5477–5492. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Cowan SW, Schirmer T, Rummel G, Steierf M, Ghosh R, Pauptitt RA, Jansonius JN and Rosenbusch JP, Nature, 1992, 358(6389), 727–733. [DOI] [PubMed] [Google Scholar]
  • 3.Haass C and Selkoe DJ, Nat. Rev. Mol. Cell Biol, 2007, 8, 101–112. [DOI] [PubMed] [Google Scholar]
  • 4.Laganowsky A, Liu C, Sawaya MR, Whitelegge JP, Park J, Zhao M, Pensalfini A, Soriaga AB, Landau M, Teng PK, Cascio D, Glabe C and Eisenberg D, Science, 2012, 335, 1228–1231. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Liu C, Zhao M, Jiang L, Cheng P-N, Park J, Sawaya MR, Pensalfini A, Gou D, Berk AJ, Glabe CG, Nowick J and Eisenberg D, Proc. Natl. Acad. Sci. U. S. A, 2012, 109, 20913–20918. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Sakai N, Mareda J and Matile S, Acc. Chem. Res, 2008, 41, 1354–1365. [DOI] [PubMed] [Google Scholar]
  • 7.Berhanu S, Majumder S, Müntener T, Whitehouse J, Berner C, Bera AK, Kang A, Liang B, Khan N, Sankaran B, Tamm LK, Brockwell DJ, Hiller S., Radford SE, Baker D. and Vorobieva AA, Science, 2024, 385, 282–288. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Kim DE, Jensen DR, Feldman D., Tischer D, Saleem A, Chow CM, Li X, Carter L, Milles L, Nguyen H, Kang A, Bera AK, Peterson FC, Volkman BF, Ovchinnikov S and Baker D, Proc. Natl. Acad. Sci. U. S. A, 2023, 120, e2207974120. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Vorobieva AA, White P, Liang B, Horne JE, Bera AK, Chow CM, Gerben S, Marx S, Kang A, Stiving AQ, Harvey SR, Marx DC, Khan GN, Fleming KG, Wysocki VH, Brockwell DJ, Tamm LK, Radford SE and Baker D, Science, 2021, 371, eabc8182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Dou J, Vorobieva AA, Sheffler W, Doyle LA, Park H, Bick MJ, Mao B, Foight GW, Lee MY, Gagnon LA, Carter L, Sankaran B, Ovchinnikov S, Marcos E, Huang P-S, Vaughan JC, Stoddard BL and Baker D, Nature, 2018, 561, 485–491. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Yamagami M, Sawada T and Fujita M, J. Am. Chem. Soc, 2018, 140, 8644–8647. [DOI] [PubMed] [Google Scholar]
  • 12.Liu C, Zhao M, Jiang L, Cheng P-N, Park J, Sawaya MR, Pensalfini A, Gou D, Berk AJ, Glabe CG, Nowick J and Eisenberg D, Proc. Natl. Acad. Sci. U. S. A, 2012, 109, 20913–20918. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Samdin TD, Jones CR, Guaglianone G, Kreutzer AG, Freites JA, Wierzbicki M and Nowick JS, Chem. Sci, 2023, 15, 285–297. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Sawaya MR, Sambashivan S, Nelson R, Ivanova MI, Sievers SA, Apostol MI, Thompson MJ, Balbirnie M, Wiltzius JJW, McFarlane HT, Madsen AØ, Riekel C and Eisenberg D, Nature, 2007, 447, 453–457. [DOI] [PubMed] [Google Scholar]
  • 15.Cheng P-N, Pham JD and Nowick JS, J. Am. Chem. Soc, 2013, 135, 5477–5492. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Do TD, LaPointe NE, Nelson R, Krotee P, Hayden EY, Ulrich B, Quan S, Feinstein SC, Teplow DB, Eisenberg D, Shea J-E and Bowers MT, J. Am. Chem. Soc, 2016, 138, 549–557. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Murzin AG, Lesk AM and Chothia C, J. Mol. Biol, 1994, 236, 1369–1381. [DOI] [PubMed] [Google Scholar]
  • 18.Laganowsky A, Liu C, Sawaya MR, Whitelegge JP, Park J, Zhao M, Pensalfini A, Soriaga AB, Landau M, Teng PK, Cascio D, Glabe C and Eisenberg D, Science, 2012, 335, 1228–1231. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Yamagami M, Sawada T and Fujita M, J. Am. Chem. Soc, 2018, 140, 8644–8647. [DOI] [PubMed] [Google Scholar]
  • 20.Rodgers CJ, Blanford CF, Giddens SR, Skamnioti P, Armstrong FA and Gurr SJ, Trends Biotechnol., 2010, 28, 63–72. [DOI] [PubMed] [Google Scholar]
  • 21.Piontek K, Antorini M and Choinowski T, J. Biol. Chem, 2002, 277, 37663–37669. [DOI] [PubMed] [Google Scholar]
  • 22.Spencer R, Chen KH, Manuel G and Nowick JS, Eur. J. Org. Chem, 2013, 3523–3528. [Google Scholar]
  • 23.Grishina IB and Woody RW, Faraday Discuss., 1994, 99, 245–262. [DOI] [PubMed] [Google Scholar]
  • 24.Wu L, McElheny D, Huang R and Keiderling TA, Biochemistry, 2009, 48, 10362–10371. [DOI] [PubMed] [Google Scholar]
  • 25.Cochran AG, Skelton NJ and Starovasnik MA, Proc. Natl. Acad. Sci. U. S. A, 2001, 98, 5578–5583. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Thuc Dang V, Engineer A, McElheny D, Drena A, Telser J, Tomczak K and Nguyen AI, Chem. – Eur. J, 2024, 30, e202402101. [DOI] [PubMed] [Google Scholar]
  • 27.Wang Y, Truex NL, Vo NDP and Nowick JS, Bioorg. Med. Chem, 2018, 26, 1151–1156. [DOI] [PubMed] [Google Scholar]
  • 28.Hwang TL and Shaka AJ, J. Magn. Reson., Ser. A, 1995, 112, 275–279. [Google Scholar]
  • 29.Aug’e S, Schmit P-O, Crutchfield CA, Islam MT, Harris DJ, Durand E, Clemancey M, Quoineaud A-A, Lancelin J-M, Prigent Y, Taulelle F and Delsuc M-A, J. Phys. Chem. B, 2009, 113, 1914–1918. [DOI] [PubMed] [Google Scholar]
  • 30.Stejskal EO and Tanner JE, J. Chem. Phys, 1965, 42, 288–292. [Google Scholar]
  • 31.Greenfield NJ, Nat. Protoc, 2006, 1, 2527–2535. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Tsunekawa E, Otsubo Y, Yamada Y, Ikeda A, Adachi N, Kawasaki M, Takasu A, Aramaki S, Senda T, Sato S, Yoshida S, Fujita M and Sawada T, J. Am. Chem. Soc, 2023, 145, 16160–16165. [DOI] [PubMed] [Google Scholar]
  • 33.Afonine PV, Grosse-Kunstleve RW, Echols N, Headd JJ, Moriarty NW, Mustyakimov M, Terwilliger TC, Urzhumtsev A, Zwart PH and Adams PD, Acta Crystallogr., Sect. D: Biol. Crystallogr, 2012, 68, 352–367. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Information

Data Availability Statement

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information: experimental procedures, additional figures, and tables (PDF). See DOI: https://doi.org/10.1039/d5cc06640a.

Crystal and NMR structures have been deposited to the Protein Data Bank with codes 9OXQ, 9OXP, 9Q63, and 9Q6E.

RESOURCES