Structure prediction of honey bee vitellogenin: a multi‐domain protein important for insect immunity

Abstract

Vitellogenin (Vg) has been implicated as a central protein in the immunity of egg‐laying animals. Studies on a diverse set of species suggest that Vg supports health and longevity through binding to pathogens. Specific studies of honey bees ( Apis mellifera ) further indicate that the vitellogenin ( vg ) gene undergoes selection driven by local pathogen pressures. Determining the complete 3D structure of full‐length Vg (flVg) protein will provide insights regarding the structure–function relationships underlying allelic variation. Honey bee Vg has been described in terms of function, and two subdomains have been structurally described, while information about the other domains is lacking. Here, we present a structure prediction, restrained by experimental data, of flVg from honey bees. To achieve this, we performed homology modeling and used AlphaFold before using a negative‐stain electron microscopy map to restrict, orient, and validate our 3D model. Our approach identified a highly conserved Ca 2+ ‐ion‐binding site in a von Willebrand factor domain that might be central to Vg function. Thereafter, we used rigid‐body fitting to predict the relative position of high‐resolution domains in a flVg model. This mapping represents the first experimentally validated full‐length protein model of a Vg protein and is thus relevant for understanding Vg in numerous species. Our results are also specifically relevant to honey bee health, which is a topic of global concern due to rapidly declining pollinator numbers.

Identification of 121 variants of honey bee Vitellogenin protein sequences with structural differences at functional sites

“…Our geographically broad sampling strategy resulted in over 100 full length Vg protein sequence variants, which is the largest collection of Vg protein variants in any species. Our data confirm the conserved nature of the ND: no changes were observed at aa positions in functional sites for DNA interaction 37 in the β-barrel subdomain, nor at positions suggested for homodimerization at either ND subdomain 28,33 (see Figures 4a and 5a). The oligomerization state in native honey bee Vg is uncertain, 28 but a requirement to protect surface properties involved in homodimerization is supported by our data.…”

How Honey Bee Vitellogenin Holds Lipid Cargo: A Role for the C-Terminal

“…We propose that the C-terminal region provides this shielding. As illustrated in Figure 3 , the “closed” position resembles the contour of the EM map ( Leipart et al, 2022 ), while the AlphaFold prediction would represent the “open” (flanking) position. Similar conformational shifts, including an “open” and “closed” state, have previously been reported for LLTPs ( Wang et al, 2006 ; Smolenaars et al, 2007 ).…”

“…To date, these have been most studied in honey bees ( Apis mellifera ) ( Pan et al, 1969 ), in which the protein influences social behavior, oxidative stress resilience, and cell-based and trans-generational immunity, in addition to its traditional role in yolk formation ( Salmela et al, 2015 ; Havukainen et al, 2013 ; Seehuus et al, 2006 ; Amdam et al, 2004 ). Recent progress made possible by DeepMind’s AlphaFold, a neural network for structure prediction ( Jumper et al, 2021 ), allowed us to generate a full-length structure prediction of honey bee Vg with high confidence ( Leipart et al, 2022 ). This structure prediction reveals the N-terminal domain folding around the lipid binding cavity, as expected for a Vg protein.…”

“…The C-terminal constitutes a small structural fold connected to the vWF domain through a presumably flexible linker. The folded C-terminal region does not appear to be in direct contact with the lipid binding site but instead appears at the flank of the large Vg structure [( Leipart et al, 2022 ) for more details].…”

“…Our previous study fitted the AlphaFold prediction into a low-resolution EM map ( Leipart et al, 2022 ). However, the C-terminal position was not compatible with EM density barriers.…”