New β-Propellers Are Continuously Amplified From Single Blades in all Major Lineages of the β-Propeller Superfamily

Abstract

β-Propellers are toroidal folds, in which consecutive supersecondary structure units of four anti-parallel β-strands-called blades-are arranged radially around a central axis. Uniquely among toroidal folds, blades span the full range of sequence symmetry, from near identity to complete divergence, indicating an ongoing process of amplification and differentiation. We have proposed that the major lineages of β-propellers arose through this mechanism and that therefore their last common ancestor was a single blade, not a fully formed β-propeller. Here we show that this process of amplification and differentiation is also widespread within individual lineages, yielding β-propellers with blades of more than 60% pairwise sequence identity in most major β-propeller families. In some cases, the blades are nearly identical, indicating a very recent amplification event, but even in cases where such recently amplified β-propellers have more than 80% overall sequence identity to each other, comparison of their DNA sequence shows that the amplification occurred independently.

Keywords: amplification; differentiation; protein evolution; repetition; β-propeller.

FIGURE 1

The ECOD families matched. (A) Frequency of the different ECOD F-names (in their simplified form) matched. (B) Boxplots of the HHsearch probability of the match, per family. The most abundant families belonging to one of the three main groups of β-propellers, as described in the text, are marked: stars mark WD40, WRAP-like propellers, circles mark VCBS and VCBS-like (including PQQ) propellers, and triangles the three main families making the Asp-Box group. (C) Proportion of ECOD F-names mapped, with a list of those without any match to β-propeller-like sequences in our dataset. The UNKNOWN family corresponds to the sequences without significant matches to any of the 65 ECOD F-names.

FIGURE 2

Synonymous and non-synonymous mutations in the nucleotide sequence of four closely related WRAP domains (≥80% pairwise sequence identity, Supplementary Figure S3). Synonymous mutations relative to the majority rule consensus of the respective β-propeller are highlighted in teal, and non-synonymous ones in yellow. Positions where a nucleotide is present in at least two thirds of the blades of one β-propeller, but different from the equivalent nucleotides in at least two of the other β-propellers are highlighted in orange. The nucleotide sequence encoding the first β-strand of the structure is different from that of the equivalent strand in the amplified region and highly conserved between homologs; it is highlighted in dark grey. The individual strands in each β-propeller blade (based on the experimental structure of the Npun_R6612 WRAP domain, Afanasieva et al., 2019) are highlighted in grey. (A) Npun_R6612 of Nostoc punctiforme PCC 73102 (ACC84870.1), (B) H6G94_01,155 of Nostoc punctiforme FACHB-252 (MBD2609895.1), (C) A6V25_01,470 of Nostoc sp. ATCC 53789 (RCJ36011.1), and (D) LC607_02,335 of Nostoc sp. CHAB 5824 (MCC5641816.1).

FIGURE 3

Taxonomic distribution of highly repetitive β-propeller sequences. (A) The frequency, in logarithmic scale, of highly repetitive β-propeller sequences from different phyla in the collected dataset. (B) Heatmap of the frequency of highly repetitive β-propeller sequences of different ECOD families in different phyla. Colors are in logarithmic scale.

FIGURE 4

Globally versus locally highly repetitive β-propellers. (A) The proportion of β-propellers in the dataset categorised into “globally” or “locally” based on the absence or presence of non-repetitive β-propeller matches in their immediate domain environment. (B) Histograms for the absolute number of consecutive highly identical blades, separated by highly repetitive β-propeller type.

FIGURE 5

The locally highly repetitive β-propeller of Bombyx mandarina inactive dipeptidyl peptidase 10 (DPPY) (XP_028029463.1). (A) Predicted domain composition of the full-length DPPY sequence. (B) Annotated β-propeller sequence, highlighting the two highly identical, consecutive blades. β-propeller blade sequences were aligned with Promals3D (Pei and Grishin, 2014) based on their predicted structure. The individual strands in each β-propeller blade are highlighted in grey. (C) Three-dimensional model of the β-propeller region, predicted with ColabFold MMseqs protocol as of August 2021 (Mirdita et al., 2021) (QMEANDisCo Global (Studer et al., 2020) score of 0.59 ± 0.05, average pLDDT (Jumper et al., 2021) of 82.9). (D) Three-dimensional model of the β-propeller region, predicted with AlphaFold v2.1.2 (Jumper et al., 2021) (QMEANDisCo Global (Studer et al., 2020) score of 0.62 ± 0.05, average pLDDT (Jumper et al., 2021) of 85.3).

FIGURE 6

The globally highly repetitive β-propeller of WD40 repeat-like protein KAE9403162.1 from Gymnopus androsaceus JB14. (A) Annotated β-propeller sequence, highlighting the 15 nearly identical, consecutive blades and the flanking non-propeller sequences. The individual strands in each β-propeller blade are highlighted in grey. (B) Three-dimensional model of the β-propeller and immediate flanking regions predicted with AlphaFold v2.1.2 (Jumper et al., 2021) (QMEANDisCo Global (Studer et al., 2020) score of 0.65 ± 0.05, average pLDDT (Jumper et al., 2021) of 88.9). Modelling was carried out with the full-length protein sequence but only the region whose sequence is shown in (A) is depicted.

FIGURE 7

The globally highly repetitive β-propeller of Menosporascus sp. CRB-9-2 hypothetical protein DL7770_005,219 (RYP84315.1). (A) Annotated β-propeller sequence, highlighting the two highly identical, consecutive blades and the flanking non-propeller sequences. The individual strands in each β-propeller blade are highlighted in grey. (B–D) Three-dimensional models of the β-propeller and immediate flanking regions predicted with AlphaFold v2.1.2 (Jumper et al., 2021). Modelling was carried out with the full-length protein sequence but only the region whose sequence is shown in (A) are depicted. (B) The top ranking model (QMEANDisCo Global (Studer et al., 2020) score of 0.57 ± 0.05, average pLDDT (Jumper et al., 2021) of 80.1), (C) the second best model (QMEANDisCo Global score of 0.58 ± 0.05, average pLDDT of 75.9), and (D) the third best model (QMEANDisCo Global score of 0.58 ± 0.05, average pLDDT of 71.0).

FIGURE 8

The proximity of globally highly repetitive β-propellers to the termini. (A) The proportion of β-propellers in the dataset localised close to one of the termini of the corresponding host protein. For each globally repetitive β-propeller in the dataset, the distance to both the N- and C-termini was computed and the closest chosen as the closest to the β-propeller position. (B) Histogram of the number of amino acids separating the β-propeller from its closest terminus.

FIGURE 9

The β-propeller fragments found in the genomic neighbourhood of globally highly repetitive β-propellers. (A) Histogram of the HHsearch probability of the identified putative fragments. Boxplots for the distance of the (B) highly repetitive β-propeller (in amino acids) and (C) corresponding putative β-propeller fragment (in nucleotides) to the closest host protein/gene terminus, separated by confidence level. (D) Nucleotide sequence and genomic context of the exemplary globally repetitive β-propeller from Nostoc sp. HK-1 hypothetical WD40 repeat-containing protein BBD61132.1. The starting codon is highlighted in blue and in frame, disruptive, stop codons in red. The regions coding for the strands making up the different blades are highlighted in grey. (E) Amino acid sequence of the 6 highly similar blades annotated at the C-terminus of BBD61132.1 plus the additional 6 found beyond the stop codon (and that make the β-propeller fragment).