Precision design of single and multi-heme de novo proteins

Abstract

The de novo design of simplified porphyrin-binding helical bundles is a versatile approach for the construction of valuable biomolecular tools to both understand and enhance protein functions such as electron transfer, oxygen binding and catalysis. However, the methods utilised to design such proteins by packing hydrophobic side chains into a buried binding pocket for ligands such as heme have typically created highly flexible, molten globule-like structures, which are not amenable to structural determination, hindering precise engineering of subsequent designs. Here we report the crystal structure of a de novo two-heme binding “maquette” protein, 4D2, derived from the previously designed D2 peptide, offering new opportunities for computational design and re-engineering. The 4D2 structure was used as a basis to create a range of heme binding proteins which retain the architecture and stability of the initial crystal structure. A well-structured single-heme binding variant was constructed by computational sequence redesign of the hydrophobic protein core, assessed by NMR, and utilised for experimental validation of computational redox prediction and design. The structure was also extended into a four-heme binding helical bundle resembling a molecular wire. Despite a molecular weight of only 24kDa, imaging by CryoEM illustrated a remarkable level of detail in this structure, indicating the positioning of both the secondary structure and the heme cofactors. The design and determination of atomic-level resolution in such de novo proteins is an invaluable resource for the continued development of novel and functional protein tools.

Rational design of photosynthetic reaction center protein maquettes

“…Addition of heme B decreases the measured ellipticity at 222 nm at 25°C from –27,900 deg cm 2 dmol −1 res −1 in the apo-state to –21,100 deg cm 2 dmol −1 res −1 when two hemes per protein are present, a 24% change ( Figure 6C ). This loss of ellipticity upon heme binding is similar to that observed in the RC maquette, but it stands in contrast to many other designed porphyrin-binding four-helix bundle proteins in which cofactor binding intensifies the ellipticity at 222 nm, presumably by promoting folding ( Ghirlanda et al, 2004 ; Rau and Haehnel, 1998 ; Polizzi et al, 2017 ; Bender et al, 2007 ; Hutchins et al, 2020 ). One exception is the highly stable 4 PA maquette, in which a small decrease in ellipticity at 222 nm upon porphyrin binding was attributed to a porphyrin transition ( McAllister et al, 2008 ).…”

“…Antiparallel four-helix bundles that bind porphyrins are common in nature ( Hunte et al, 2000 ; Malone et al, 2019 ; Mathews et al, 1979 ; Weber et al, 1980 ; Yankovskaya et al, 2003 ; Dautant et al, 1998 ), and experimental structures of de novo -designed versions have been published recently ( Polizzi et al, 2017 ; Pirro et al, 2020 ; Mann et al, 2021 ; Hutchins et al, 2020 ). One common trend in all of the examples cited here is that the plane of the porphyrin ring cuts through the e - e interface of the helical bundle, causing the e - e interface to expand while the g - g interface can remain narrow.…”

De novo protein design of photochemical reaction centers

“…Inter-cofactor distances are subtly influenced by core packing, the helical register, and the orientations of porphyrin rings, but larger distance changes are more easily accomplished by moving cofactor-ligating residues along the helices in increments of a full helical turn (~5.2 Å). Previous multi-cofactor de novo protein designs 10,13,14,38 have edge-to-edge distances between porphyrins that are too short or too long for high-yield charge separation in DPA triads as shown in the theoretical contour plot of Fig. 1c (calculated from Eq.…”

Tailorable Tetrahelical Bundles as a Toolkit for Redox Studies

“…These de novo designed proteins are practical systems for studying biological redox reactions. 19 , 36 …”

“…De novo proteins can function as redox tools that do not require significant genetic modifications, stumble over evolutionary artifacts, or poison the system with harsh chemicals. − To date, there have been an impressive number of these systems and more are being developed every day. ,− However, we are missing a set of fundamental engineering guidelines that allow for the reliable tuning of E m values impeding their utility. , Currently, designed proteins are characterized as they are made and used where appropriate . More broadly, our knowledge of how natural proteins modify E m values comes from a limited set of data, which may be complicated by evolutionary artifacts like misfolding or compensating amino acids we have not identified.…”