Molecular Evolutionary Analysis of Nematode Zona Pellucida (ZP) Modules Reveals Disulfide-Bond Reshuffling and Standalone ZP-C Domains

GBE Molecular Evolutionary Analysis of Nematode Zona Pellucida (ZP) Modules Reveals Disulfide-Bond Reshuffling and Standalone ZP-C Domains Cameron J. Weadick*1 1 Department of Biosciences, University of Exeter, United Kingdom Accepted: 13 May 2020 Abstract Zona pellucida (ZP) modules mediate extracellular protein–protein interactions and contribute to important biological processes including syngamy and cellular morphogenesis. Although some biomedically relevant ZP modules are well studied, little is known about the protein family’s broad-scale diversity and evolution. The increasing availability of sequenced genomes from “nonmodel” systems provides a valuable opportunity to address this issue and to use comparative approaches to gain new insights into ZP module biology. Here, through phylogenetic and structural exploration of ZP module diversity across the nematode phylum, I report evidence that speaks to two important aspects of ZP module biology. First, I show that ZP-C domains—which in some modules act as regulators of ZP-N domain-mediated polymerization activity, and which have never before been found in isolation—can indeed be found as standalone domains. These standalone ZP-C domain proteins originated in independent (paralogous) lineages prior to the diversification of extant nematodes, after which they evolved under strong stabilizing selection, suggesting the presence of ZP-N domain-independent functionality. Second, I provide a much-needed phylogenetic perspective on disulfide bond variability, uncovering evidence for both convergent evolution and disulfide-bond reshuffling. This result has implications for our evolutionary understanding and classification of ZP module structural diversity and highlights the usefulness of phylogenetics and diverse sampling for protein structural biology. All told, these findings set the stage for broad-scale (cross-phyla) evolutionary analysis of ZP modules and position Caenorhabditis elegans and other nematodes as important experimental systems for exploring the evolution of ZP modules and their constituent domains. Key words: gene family evolution, supradomain, domain architecture, cysteine connectivity, nematode cuticle, cuticlin. Introduction Secreted proteins help cells withstand, react to, and shape external conditions (Agrawal et al. 2010; Naba et al. 2016; Cuesta-Astroz et al. 2017). The extracellular environment can be variable and stressful, and in order to properly function under such challenging conditions, secreted proteins often employ specialized domains that can be repurposed to different ends by being recombined into different protein architectures (Bork et al. 1996; Martin et al. 1998). Obtaining an appreciation of the structural diversity of secreted proteins is key to understanding the many biological processes that extend beyond the cellular membrane. In many cases, however, insights into the biology of secreted protein families derive from restricted and potentially nonrepresentative sets of model proteins (e.g., those linked to particular biomedical conditions, those expressed in already established model systems, and those that can be collected at high levels). Taking a broad, comparative view can uncover important but otherwise overlooked aspects of secreted protein structure and function. The zona pellucida (ZP) module is a key component of many secreted proteins (Bork and Sander 1992; Plaza et al. 2010; Litscher and Wassarman 2015; Bokhove and Jovine 2018). Named after the mammalian egg coat (from which the first family-members were found), ZP modules mediate extracellular protein–protein interactions. Through these actions, ZP-module-bearing proteins (hereafter referred to simply as “ZPD proteins,” following Litscher and Wassarman [2015]) contribute to a variety of critical cellular and developmental processes, including regulating C The Author(s) 2020. Published by Oxford University Press on behalf of the Society for Molecular Biology and Evolution. V This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Genome Biol. Evol. 0(0):pp. 1240–1255. doi:10.1093/gbe/evaa095 Advance Access publication 18 May 2020 1 *Corresponding author: E-mail: GBE Weadick 2 are capable of folding independently in vitro (Lin et al. 2011; Diestel et al. 2013; Bokhove et al. 2016) and they contribute to protein–protein binding interfaces in some ZPD proteins (Hanet al. 2010; Linet al. 2011; Diestelet al. 2013; Okumura et al. 2015). These points combine to suggest that standalone ZP-C domains could in theory prove functional and exist on their own in nature. ZP modules are characterized by the presence of multiple intradomain disulfide bonds (Bork and Sander 1992). However, the number of cysteine residues found per module varies and this has led to contrary views about how the cysteines connect and whether this variation has any functional effect (Jovine et al. 2005; Yonezawa 2014). ZP modules have often been classified as either Type I or Type II based on the number of cysteines found within the ZP-C domain; these two groups were alleged to have nonnested connectivity patterns, and to differ functionally, with Type II but not Type I modules able to homopolymerize (Boja et al. 2003; Darie et al. 2004; Kanai et al. 2008). However, in light of the solved structures of a few ZP modules and isolated ZP-C domains, it was subsequently argued that there is no reliable distinction between these groups, and that polymerization tendencies are unrelated to cysteine connectivity patterns (Bokhove and Jovine 2018). Rather, Bokhove et al proposed that ZP-C domains typically have a standard set of three disulfide bonds (Cys5– Cys7, Cys6–Cys8, and CysA–CysB), with cysteine variation among ZPD proteins resulting primarily from lineage-specific gains and losses of disulfide pairs. For example, the ZP module component of the BMP coreceptor endoglin lacks the Cys6–Cys8 and CysA–CysB disulfides found in uromodulin (Saito et al. 2017), whereas additional disulfides associated with lineage-specific insertions have been found in some vertebrate egg-coat proteins (e.g., trout VEa/b and chicken ZP3; Darie et al. 2004; Han et al. 2010). The case of ZP3 is an interesting example, as this family of egg-coat proteins possesses a ZP-C subdomain that introduces four additional cysteine residues that are closely situated both along the sequence and in 3D space. Through protein crystallography of chicken ZP3, Han et al. (2010) showed that disulfide bonds covalently link the ZP-C core to its subdomain. By contrast, the results of earlier mass spectrometric analysis of other vertebrate ZP3 proteins (but not including chicken ZP3) indicated several cases where the subdomain’s cysteines paired only among each other (Boja et al (...truncated)