The Tetragnatha kauaiensis Genome Sheds Light on the Origins of Genomic Novelty in Spiders

The Tetragnatha kauaiensis Genome Sheds Light on the Origins of Genomic Novelty in Spiders Jos eCerca 1,2,3, *, Ellie E. Armstrong 14 ,JoelVizueta 5,6 ,RosaFern  andez 7 ,DimitarDimitrov 8 , Bent Petersen 9,10 , Stefan Prost 11,12,13 , Julio Rozas 5 , Dmitri Petrov 4 , and Rosemary G. Gillespie 1 1 Berkeley Evolab, Department of Environmental Science, Policy, and Management, UC Berkeley, California, USA 2 Frontiers in Evolutionary Zoology, Natural History Museum, University of Oslo, Norway 3 Department of Natural History, NTNU University Museum, Norwegian University of Science and Technology, Trondheim, Norway 4 Department of Biology, Stanford University, California, USA 5 Departament de Gene`tica, Microbiologia i Estad ıstica & Institut de Recerca de la Biodiversitat (IRBio), Universitat de Barcelona, Spain 6 Villum Centre for Biodiversity Genomics, Section for Ecology and Evolution, Department of Biology, University of Copenhagen, Denmark 7 Institute of Evolutionary Biology (CSIC—Universitat Pompeu Fabra), Barcelona, Spain 8 Department of Natural History, University Museum of Bergen, University of Bergen, Norway 9 Section for Evolutionary Genomics, The GLOBE Institute, Faculty of Health and Medical Sciences, University of Copenhagen, Denmark 10 Centre of Excellence for Omics-Driven Computational Biodiscovery, Faculty of Applied Sciences, AIMST University, Kedah, Malaysia 11 Central Research Laboratories, Natural History Museum Vienna, Vienna, Austria 12 University of Veterinary Medicine, Konrad Lorenz Institute of Ethology, Vienna, Austria 13 South African National Biodiversity Institute, National Zoological Garden, Pretoria, South Africa *Corresponding author: E-mail: jos[email protected]om. Accepted: 22 November 2021 Abstract Spiders (Araneae) have a diverse spectrum of morphologies, behaviors, and physiologies. Attempts to understand the genomic-basis of this diversity are often hindered by their large, heterozygous, and AT-rich genomes with high repeat content resulting in highly fragmented, poor-quality assemblies. As a result, the key attributes of spider genomes, including gene family evolution, repeat content, and gene function, remain poorly understood. Here, we used Illumina and Dovetail Chicago technologies to sequence the genome of the long-jawed spider Tetragnatha kauaiensis, producing an assembly distributed along 3,925 scaffolds with an N50 of 2 Mb. Using comparative genomics tools, we explore genome evolution across available spider assemblies. Our findings suggest that the previously reported and vast genome size variation in spiders is linked to the different representation and number of transposable elements. Using statistical tools to uncover gene-family level evolution, we find expansions associated with the sensory perception of taste, immunity, and metabolism. In addition, we report strikingly different histories of chemosensory, venom, and silk gene families, with the first two evolving much earlier, affected by the ancestral whole genome duplication in Arachnopulmonata (450 Ma) and exhibiting higher numbers. Together, our findings reveal that spider genomes are highly variable and that genomic novelty may have been driven by the burst of an ancient whole genome duplication, followed by gene family and transposable element expansion. Key words: gene family, Araneae, arthropod, repeatome, hawai’i, transposable element. Introduction With nearly 50,000 described species (World Spider Catalog 2021), and dating back 350 Myr (Fern andez et al. 2018), spiders (Chelicerata, Araneae) have conquered most terrestrial ecosystems, from the cold Arctic to arid deserts (Jackson and Cross 2011;Dimitrov et al. 2012;Garrison et al. 2016; ßThe Author(s) 2021. Published by Oxford University Press on behalf of the Society for Molecular Biology and Evolution. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://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. 13(12) https://doi.org/10.1093/gbe/evab262 Advance Access publication 26 November 2021 1 GBE Downloaded from https://academic.oup.com/gbe/article/13/12/evab262/6443144 by Consejo Superior de Investigaciones Cientificas (CSIC) user on 14 March 2022 Fern andez et al. 2018). Spiders play a key role in terrestrial ecosystems regulating community dynamics as major arthropod predators (Herberstein and Wignall 2011;Wilder 2011), having evolved a diverse array of adaptive solutions, which include, a rich cocktail of venoms to neutralize prey (Binford 2001;King and Hardy 2013), a color palette essential for camouflaging, mimicking, and signaling (Oxford and Gillespie 1998;Croucher et al. 2013;Cotoras et al. 2016), and the ability to produce silk for spinning webs and subduing prey (Vollrath 1999;Garb et al. 2010;Sanggaard et al. 2014). Despite the advances in spider ecology, evolution, and systematics, knowledge of spider genomes still lags relative to other taxa. Most of the available spider genomes are of poor quality, being highly fragmented (Garb et al. 2018)andlacka substantial part of the genome, with only three recent exceptions involving chromosome-resolved genomes (Escuer et al. 2021;Fan et al. 2021;Sheffer et al. 2021). Several factors contribute to the sparse availability of high-quality spider genome assemblies, including the lack of a model organism among spiders (sensu Drosophila melanogaster in flies and Tribolium castaneum in beetles) (Brewer et al. 2014), and the challenges associated with sequencing spider genomes, which are characterized by high AT-content, repeats, heterozygosity, and often large genome sizes (Garb et al. 2018). Focus on non-model organism genomes shows that increased taxon-sampling leads to an improved understanding of the diversity and function of molecular mechanisms across the tree of life (McGregor et al. 2008), as it overcomes the biases from the limited number of model taxa, and highlights the idiosyncrasies throughout the tree of life. Consequently, a better representation of spider genomes will certainly help understanding spider diversity and evolution (McGregor et al. 2008). A systematic analysis of spider genomes has the potential to unveil the genomic foundation of spider evolution. For example, the detection of duplicate Hox clusters suggested an ancestral whole genome duplication in the common ancestor of modern spiders and scorpions (Arachnopulmonata; Schwager et al. 2007), and this evidence was later on confirmed by the first spider genomes (Clarke et al. 2015; Schwager et al. 2017;Leite et al. 2018). The implications of whole genome duplications may, however, be multifarious and complex (Ohno 1970). On one hand, genome duplication may act as a catalyst for molecular novelty. Under this framework, the retention of duplicated genes and other genetic components may act as “reservoirs of genetic variation,” through processes of gene neoand sub-functionalization (Lynch and Force 2000), and be of use when organisms encounter novel selective pressures (Li et al. 2018;Nieto Feliner et al. 2020;Schmickl and Yant 2021). Considering the evidence for gene duplicates in spider genomes, including spidroins (silk genes) (Sanggaard et al. 2014;Clarke et al. 2015; Babb et al. 2017;Garb et al. 2018;Sheffer et al. 2021), venoms (Sanggaard et al. 2014;Gendreau et al. 2017; Haney et al. 2019), chemosensory (Vizueta et al. 2018, 2019;Vizueta, Escuer, et al. 2020) gene families may yield insights on phenotypic innovation and the adaptation to novel environments. On the other hand, because genome duplication leads to a significant re-organization of the genome, it may cause deregulation of gene-expression networks or unlock the epigenetic suppression of transposable elements, which may proliferate across the genome and result in decreased fitness for the organism—“the genomic shock hypothesis” (McClintock 1984;Choi et al. 2020). In such a scenario, one expects to find variation in transposable element proliferation across genomes, and ultimately a substantial variation of genome size. The proliferation of transposable elements may thereby underlie genome size variation in spiders, which ranges between 0.74 and 5.73 C values (0.7– 5.6 Gb) (Gregory and Shorthouse 2003)(http://www. genomesize.com/ checked in April 15, 2021; values for: Habronattus borealis,Tetragnatha elongata, respectively). Comparisons between different genome assemblies may yield important insights on the prevalence of gene duplications, neofunctionalization, and transposable element dynamics across different lineages. Here, we report a genome assembly of the Hawaiian spider Tetragnatha kauaiensis and place it in the context of currently available spider genomes to assess signatures of genome evolution across spider lineages (supplementary table 1, Supplementary Material online). To do so, we first explore the completeness and duplication rates across the spider assemblies. Considering the role of transposable elements in driving genome size variation, we also assess transposable Significance Despite being one of the most charismatic animal lineages, progress on spider genome evolution lags due to the challenges in sequencing and assembling their genomes, which involve genome size and repeat content. Here, we sequence the genome of Tetragnatha kauaiensis, a spider endemic to Hawai’i, and compare it with other available spider genomes. We find variation in terms of repeats and transposable elements; expansions in gene-content associated with metabolism, sensory perception, and immunity; and wide variation of chemosensory genes and venom genes. Cerca et al. GBE 2Genome Biol. Evol. 13(12) https://doi.org/10.1093/gbe/evab262 Advance Access publication 26 November 2021 Downloaded from https://academic.oup.com/gbe/article/13/12/evab262/6443144 by Consejo Superior de Investigaciones Cientificas (CSIC) user on 14 March 2022 element load in each genome. Third, we quantify the expansion and contraction of gene families (based on gene similarity metrics), and classify the function of these families using Gene Ontology (GO). Finally, we delve deeper into the identification and comparison of chemosensory, venom, and spidroin (silk) genes, studying duplicates in a phylogenetic context. Focus on these three categories is grounded on their central role to the survival and fitness of spiders, and benefits from extensive research, including hand curated genes and databases. Results The Tetragnatha kauaiensis Genome The T. kauaiensis genome assembly has a size of 1.08 Gb, distributed along a total of 132,391 contigs, comprising 3,925 scaffolds. The largest scaffold is ca. 10.5 megabases (Mb), whereas the estimated scaffold-N50 for the assembly is 2Mb (supplementary table 2,Supplementary Material online). The assembly has a GC content of 33.3%, in line with the remaining spider genomes (lowest GC content Latrodectus hesperus with 28.59% and highest content is Stegodyphus mimosarum with a GC content of 33.62; supplementary table 2,Supplementary Material online). The assembly has no obvious contaminants or associated symbionts, as determined by Blobtools (supplementary fig. 1, Supplementary Material online). The majority of scaffolds have a similar GC composition, despite variations in coverage. From all 3,925 scaffolds, 2,774 were labeled as no-hits (comprising only a total of 32.46 Mb of the assembly), and 889 labeled as Arthropods (886 Mb). Annotation of the T. kauaiensis genome yielded 38,907 genes, comprising 213,695 exons and 171,423 introns (supplementary table 3,Supplementary Material online). Together, all genes cover 290,369,064 bp (290 Mb) representing 26.7% of the genome with 41,209,078 bp (41 Mb, 3.8% of the genome) being coding sequences (cds). The mean gene length is 7,463bp (supplementary table 3, Supplementary Material online), the longest gene is 208,580 bp long (208 kb), and 89.7% of BUSCOs are retrieved as complete. Genome Characterization and Evolution The analyzed assemblies vary widely in size. Araneus ventricosus has the largest assembly with 3.6 Gb (supplementary table 2,Supplementary Material online), whereas T. kauaiensis has the smallest assembly with 1,085,571,486 bp (1.1 Gb). In between these extremes, we find the genomes of S. mimosarum (2.7 Gb), Trichonephila clavipes (2.4 Gb), Argiope bruennichi (1.7 Gb), Dysdera silvatica (1.4 Gb), Parasteatoda tepidariorum (1.5 Gb) and L. hesperus (1.1 Gb). Considering the 3-fold variation in genome size and the evidence for ancient whole genome duplications in Chelicerata (Shingate et al. 2020) and Arachnida (Schwager et al. 2017;Harper et al. 2021), and the suggestion that there has been a large-scale (whole genome or chromosomal) duplication event within spiders (Clarke et al. 2015), we explored the possibility of whole genome duplication private to spider genomes by interrogating the number of homologs in the Hox genes clusters. Using Hox genes 1–5, and based on a threshold of 95% identity, we find no evidence for an additional ancestral whole genome duplication in the studied spider genomes. We found zero, one, or two homologs for Hox 1(supplementary table 4,Supplementary Material online). For Hox 2, we found two homologs in all genomes, with the exception of A.ventricosus, where we only find a single homolog (supplementary table 4,Supplementary Material online). For Hox 3, there was only one homolog in all genomes, with the exception of P. tepidariorum (two candidates) and T. clavipes (no candidate). For Hox 4, we found two homologous genes in T. kauaiensis,P. tepidariorum,L. hesperus,and S. mimosarum, one in T.clavipes and another in D. silvatica. A.ventricosus, however, had four homologs for the Hox4 gene.Finally,forHox 5, we identified one homolog in all genomes, with the exception of A.ventricosus and P. tepidariorum where we found two homologous genes. This suggests that, with the exception of the outlier with four copies (Araneus Hox4), Hox genes are present in one or two copies. Transposable Element Variation We find variation in repeat content and tempo of repeat accumulation across the spider assemblies (fig. 1;supplementary table 5,Supplementary Material online). For example, 10.3% of the D. silvatica genome is composed of Long Interspersed Nuclear Elements (LINEs), whereas all other studied spiders had at most 3% LINEs (fig. 1A). Stegodyphus mimosarum had 5.40% of its genome covered by long terminal repeat (LTR) elements, whereas A.ventricosus,whichis the second LTR-element-most rich genome, had only 1.60% (fig. 1). Interspersed repeats varied between 52.84% in D. silvatica and 16.53% in L. hesperus (supplementary table 5,Supplementary Material online). Unclassified repeats ranged between 32.64% (A.ventricosus), and 4.71% (L. hesperus)(supplementary table 5,Supplementary Material online). Overall, Repeatmasker identified between 16.71% and 52.84% of total repeat content (fig. 1A;supplementary table 5,Supplementary Material online). The correlation coefficient (R) between genome size and the percent of masked genome is R¼0.65, and the correlation coefficient (R) between total length of the masked genome and genome size is R¼0.962. Finally, we find variability in the accumulation of transposable elements through time, as represented by the shape of the transposable element/repeat landscape plot curves (fig. 1B). For instance, the A. bruennichi and P. tepidariorum assemblies show two peaks in transposable element accumulation, whereas all the others display a single Tetragnatha kauaiensis and Origins of Genomic Novelty in Spiders GBE Genome Biol. Evol. 13(12) https://doi.org/10.1093/gbe/evab262 Advance Access publication 26 November 2021 3 Downloaded from https://academic.oup.com/gbe/article/13/12/evab262/6443144 by Consejo Superior de Investigaciones Cientificas (CSIC) user on 14 March 2022 peak. Stegodyphus mimosarum, however, has a recent burst in Tc1/mariner (DNA/TcMar) transposable elements (fig. 1B). Despite the differences in the accumulation of transposable element/repeats through time, we note that the Tc1/mariner group (DNA/TcMar) is present as one of the top three most represented transposable elements in all the assemblies, and hAT transposons (DNA/hAT) are also among the threedominant categories in six assemblies. There is, however, variation across assemblies, as shown by the high numbers of Helitrons (RC/Helitron) in two of the Araneidae assemblies (A. bruennichi and A.ventricosus), Gypsy (LTRGypsy) in S. mimosarum, and Jockey (LINE/Jockey-l) in L. hesperus. The analysis of genome completeness, as assessed by BUSCO scores, suggests that spider assemblies are considerably fragmented and missing substantial parts of the genome (supplementary table 6,Supplementary Material online). For 0 5 10 15 20 25 30 35 40 45 50 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 Kimura substitution level (CpG adjusted) % of genome Centruroides sculpturatus DNA/hAT DNA/TcMar Unknown 0 5 10 15 20 25 30 35 40 45 50 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Argiope bruennichi Unknown RC/Helitron DNA/TcMar 0 5 10 15 20 25 30 35 40 45 50 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Araneus ventricosus Unknown RC/Helitron DNA/TcMar 0 5 10 15 20 25 30 35 40 45 50 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 Trichonephila clavipes Unknown DNA/hAT DNA/TcMar 0 5 10 15 20 25 30 35 40 45 50 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Parasteatoda tepidariorum Unknown DNA/hAT DNA/TcMar 0 5 10 15 20 25 30 35 40 45 50 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 Tetragnatha kauaiensis Unknown DNA/hAT DNA/TcMar 0 5 10 15 20 25 30 35 40 45 50 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Stegodyphus mimosarum Unknown DNA/TcMar LTR/Gypsy 0 5 10 15 20 25 30 35 40 45 50 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Dysdera silvatica DNA/hAT DNA/TcMar Unknown 0 5 10 15 20 25 30 35 40 45 50 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Latrodectus hesperus LINE/Jockey-I Unknown DNA/TcMar A B 0 15 30 45 60 Argiope bruennichi (1.7 Gb) Araneus ventricosus (3.7 Gb) Centruroides sculpturatus Dysdera silvatica (1.6 Gb) Latrodectus hesperus (1.1 Gb) Trichonephila clavipes (2.4 Gb) Parasteatoda tepiorarium (1.4 Gb) Tetragnatha kauaiensis (1.1 Gb) Stegodyphus mimosarum (2.7 Gb) Centruroides sculpturatus (925 Mb) SINEs LINEs LTR elements DNA elements Unclassified Total masked % of genome FIG.1.—Transposable element and repeat characterization (A) Web diagram showing the representation of TE and repeats in the assemblies. Assemblies and correspondent assembly sizes are represented on the edges of the web diagram. Different transposable element families or repeats are presented in different colors on the plot, and the total content masked by RepeatMasker is shown in blue. The numbers for each web-line indicate the percent of the genome occupied by each transposable element, or the percent masked. (B) Repeat/transposable element landscape plots for the various assemblies. The three most represented transposable element categories are present for every genome (e.g. DNA/TcMar, DNA/hAT, and unknown for T. kauaiensis). Each plot shows the Kimura substitution level (xaxis) and percent of genome covered by repeats (yaxis). Cerca et al. GBE 4Genome Biol. Evol. 13(12) https://doi.org/10.1093/gbe/evab262 Advance Access publication 26 November 2021 Downloaded from https://academic.oup.com/gbe/article/13/12/evab262/6443144 by Consejo Superior de Investigaciones Cientificas (CSIC) user on 14 March 2022 instance, the D. silvatica,L. hesperus and T.clavipes genomes have only, respectively, 66%, 38.6%, and 52% complete BUSCOs (Arachnid odb10). Completeness in the remaining genomes ranged between 80% and 99%. Duplicated BUSCOs ranged between 30.5% (P. tepidariorum)and 3.2% (S. mimosarum). Notably, the two biggest genomes, A.ventricosus (3.6 Gb) and S. mimosarum (2.7 Gb) have 18.4% and 3.2% duplicated BUSCOs (supplementary table 6,Supplementary Material online, Arachnid data set odb10). The percentage of complete single-copy, duplicated, fragmented, and missing BUSCOs is concordant between the Arthropod and Arachnid sets (supplementary table 6, Supplementary Material online). Gene-Family Evolution Because studying gene family evolution requires a phylogenetic backbone, we used the tree obtained from OrthoFinder based on 286 single-copy orthologs (orthologs are genes in different species that evolved from a common ancestral gene; fig. 2A). The tree topology has T. kauaiensis (Tetragnathidae) as sister lineage to the clade comprising the two members of Araneidae (A. bruennichi and T.clavipes). The clade encompassing all the aforementioned is sister to the Theridiidae (L. hesperus and P. tepidariorum). In turn, S. mimosarum (Eresidae) is the sister to Araneoidea (represented here by Tetragnathidae, Araneidae, and Theridiidae). Dysdera silvatica (Dysderidae) is the sister to the clade comprising all the 0.1 Trichonephila clavipes Latrodectus hesperus Tetragnatha kauaiensi s Stegodyphus mimosarum Dysdera silvatica Parasteatoda tepidariorum Centruroides sculpturatus Argiope bruennichi Araneidae 20 122 12 93 27 12 9 108 80 110 15 Centruroides sculpturatus protein glycosylation cell division defense response Dysdera silvatica sensory perception of taste biosynthesis carbohydrate metabolism DNA integration immune response regulation of cell adhesion viral DNA genome packaging mannose metabolism Stegodyphus mimosarum oxidation− -reduction process DNA integration intracellular signal transduction DNA integration protein import Theridiidae Parasteatoda tepidariorum calcium ion transmembrane transport ethanol oxidation immune response RNA processing microtubule −based movement protein ufmylation microtubule− -based movement nuclear−transcribed mRNA catabolic process, nonsense- -mediated decay biosynthe- -sis viral DNA genome packaging Latrodectus hesperus DNA integration sensory perception of taste intracellular signal transduction Node 1 Tetragnatha kauaiensis nuclear−transcribed mRNA catabolic process, nonsense- -mediated decay protein lipoylation regulation of apoptotic process sodium ion transport carbohydrate metabolism chitin metabolism sensory perception of taste cation transport glyoxylate catabolic process regulation of nitrogen utilization DNA integration phosphorylation Araneidae A B Trichonephila clavipesArgiope bruennichi immune response protein peptidylprolyl isomerization sodium ion transport protein folding regulation of neurotransmitter levels sensory perception of taste DNA methylation on adenine response to heat zinc II ion transport protein unfolding regulation of nitrogen utilization ubiquitin−dependent protein catabolism via the N−end rule pathway pilus assembly protein refolding Node 1 Theridiidae FIG.2.—Gene family expansion (A) Tree topology obtained for single-copy orthologs. Numbers in blue indicate significantly expanded gene families as determined by CAFE. (B) Treemap representation of Gene Ontology Biological Function Annotation of the significantly expanded gene families as retrieved by REVIGO. Branches/Nodes with significant expansions, including Araneidae, Theridiidae, and Node 1 are represented together with the different genomes. Tetragnatha kauaiensis and Origins of Genomic Novelty in Spiders GBE Genome Biol. Evol. 13(12) https://doi.org/10.1093/gbe/evab262 Advance Access publication 26 November 2021 5 Downloaded from https://academic.oup.com/gbe/article/13/12/evab262/6443144 by Consejo Superior de Investigaciones Cientificas (CSIC) user on 14 March 2022 aforementioned spiders (fig. 2A). This topology is in agreement with recent and comprehensive phylogenomic analyses of spiders (Fern andez et al. 2018). From a total of 608 significant gene family expansions in all branches, 572 occurred in terminal branches (fig. 2B). There were 451 significant expansions, and 157 significant contractions, of which 124 occurred in terminal branches (supplementary figs. 1–4,Supplementary Material online). GO annotations of the significantly expanded gene families which were characterized under “biological process” were organized by REVIGO and are represented in fig. 2B. Broadly, we find expansions associated with feeding metabolism and sensory perception, mannose metabolism in the genome of D. silvatica and chitin metabolism in T. kauaiensis (fig. 2B). Expansions in carbohydrate metabolism are found in D. silvatica and T. kauaiensis, whereas Araneidae has glyoxylate catabolic process expanded (fig. 2B). Expansions in sensory perception of taste are found in D. silvatica,T. kauaiensis, A. bruennichi, and in Node 1 (fig. 2B). Immune response is found in the genomes of D. silvatica,P. tepidariorum,and A. bruennichi, whereas sodium ion transport is found in T. kauaiensis and A. bruennichi (fig. 2B). When considering significant expansions in all GO categories (i.e. biological process, molecular function, and cellular component), we find expansions associated with taste (including sensory perception of taste in Node 1, A. bruennichi,andD. silvatica; detection of chemical stimulus involved in sensory perception of taste in A. bruennichi and Node 1; molecular function taste receptor activity is found in A. bruennichi and T. kauaiensis; supplementary table 7, Supplementary Material online). We also find evidence for expansions related to various metabolic processes, including carbohydrate metabolic process, and mannose metabolic process in D. silvatica, whereas protein catabolic process, 3,4dihydroxybenzoate catabolic process, fatty acid catabolic process, pyruvate metabolic process, glucose metabolic process, protein metabolic process, lipid catabolic process, lipid metabolic process, and fatty acid metabolic process are found in T. clavipes.TheP. tepidariorum genome includes expansions in peptidoglycan catabolic process and lipid metabolic process, whereas that of T. kauaiensis includes expansions in chitin metabolic process and carbohydrate metabolic process. Theridiidae includes expansions in lipid metabolic process, whereas Araneidae includes changes in taurine catabolic process. Finally, catalytic activity is expanded in the genomes of D. silvatica,L. hesperus,T.clavipes,T. kauaiensis. Other notable expansions include the regulation of neurotransmitter levels, structural constituent of eye lens in A. bruennichi, defense response and toxin activity in C. sculpturatus, and response to heat in T.clavipes. The biological process for “sodium channel activity” is found expanded in A. bruennichi,T.clavipes,and P. tepidariorum, whereas the molecular function for “sodium channel activity” is found in A. bruennichi and T. kauaiensis. Proteolysis (i.e. breakdown of proteins), the breakdown of process is expanded in A. bruennichi,C. sculpturatus, D. silvatica,L. hesperus,P. tepidariorum,S. mimosarum, T. kauaiensis, and Theridiidae. Venom Gene-Family Variation The combination of BLAST and TOXIFY identified a total of 559 toxins in the studied genomes (supplementary table 8, Supplementary Material online), included as part of 189 orthogroups. The orthogroups with most genes are displayed in figure 3 and include OG0000175 (135 genes, Astacin-like metalloproteases as determined by NCBI-BLAST), OG0000314 (105 genes, Neprilysins or endothelinconverting proteins), OG0000346 (99 genes, uncharacterized proteins), OG0000432 (86 genes, Techylectin), OG0000639 (68 genes, various toxin-types), OG0000761 (61 genes, Zonadhesins, various toxin-types), OG0000803 (59 genes, Astacin-like metalloproteases), OG0000916 (54 genes, Papilins, Kunitz-type serine protease inhibitor) OG0000930 (54 genes, Astacin-like metalloproteases), OG0001436 (41 genes, uncharacterized proteins). The two most toxin-rich assemblies were the A. bruennichi and P. tepidariorum where 154 and 200 toxins were identified, respectively. The scorpion genome, C. sculpturatus, yielded 31 toxins, whereas D. silvatica and L. hesperus yielded 13 and 16 toxins, respectively (supplementary table 8,Supplementary Material online). Phylogenetic analyses of the orthogroups show that most venom families were present before the split between scorpions and spiders (fig. 3). Different spider genomes include species-specific expansions (i.e. groups of five or more genes from a single genome that cluster as a monophyletic clade), and many of these have relatively large branch lengths. Specifically, we find evidence for various expansions in P. tepidariorum (4 expansions, one with 7 genes, another with 12, one with 7 and one with 9 genes), one expansion in A.ventricosus (oneexpansionwith11closelyrelated genes), one in D. silvatica (one expansion in 6 genes) and one in C. sculpturatus (5 genes expanded) in OG0000175 (fig. 3). In OG0000314, we found an expansion private to the three Araneidae genomes, including A. bruennichi,A. ventricosus,andT.clavipes), various expansions exclusive to the A.ventricosus genome, and one expansion specific to the scorpion genome (nine genes). In OG000346, we found various expansions on the S. mimosarum (nine genes), P. tepidariorum (five genes), A.ventricosus (eight genes) genomes. In OG000432 we found genome-specific expansions in D. silvatica (eight genes; fig. 3). In OG0000639, we found an expansion in C. sculpturatus (five genes), and in OG0000803 there are two five-gene expansions, one in C. sculpturatus, another in A.ventricosus. OG0000930 is only present in T. kauaiensis (1expansionwith20genes), A.ventricosus,A. bruennichi,T.clavipes,and P. tepidariorum. OG0001436 is expanded in C. sculpturatus (five genes). Cerca et al. GBE 6Genome Biol. Evol. 13(12) https://doi.org/10.1093/gbe/evab262 Advance Access publication 26 November 2021 Downloaded from https://academic.oup.com/gbe/article/13/12/evab262/6443144 by Consejo Superior de Investigaciones Cientificas (CSIC) user on 14 March 2022 Chemosensory Gene-Family Variation We identified a total of 5,595 candidate gustatory receptors (GRs), 1,934 candidate ionotropic receptors (IRs), 25 candidate Odorant binding proteins (OBP-like), 147 candidate Niemann-Pick type C2 (NPC2), 137 candidate carrier protein (CCP), and 998 candidate cluster of differentiation 36 and neuron membrane proteins (CD36-SNMP; supplementary table 9,Supplementary Material online;figs. 4 and 5). GRs exhibited a large interspecific variation (fig. 4), ranging between1,436GRsinA.ventricosus and84inL. hesperus. Centruroides sculpturatus, the outgroup, had 1,648 GRs (supplementary table 9,Supplementary Material online). The D. silvatica genome has the most IR/iGluR genes with 443 genes (supplementary table 9,Supplementary Material online;fig. 4). We detected a total of 25 OBP-like genes, with 5 being present in T. kauaiensis,4inD. silvatica and in S. mimosarum,3inP. tepidariorum and all remaining genomes having only 1 or 2 OBP-like genes (supplementary table 9,Supplementary Material online;fig. 5). From the 147 OG0000346 Tree scale OG0000639 Tree scale OG0000803 Tree scale OG0000930 Tree scale OG0001436 Tree scale OG0000175 Tree scale OG0000314 Tree scale OG0000432 Tree scale OG0000761 Tree scale OG0000906 Trees cale Argiope bruennichi Araneus ventricosus Centruroides sculpturatus Dysdera silvatica Latrodectus hesperus Parasteatoda tepidariorum Stegodyphus mimosarum Tetragnatha kauaiensis Trichonephila clavipes Scorpion genes indicated by arrows FIG.3.—Venom gene phylogenies. Phylogenies for the ten largest orthogroups of identified venom genes. For each tree, we indicate the Orthogroup ID and tree scale. Different colors correspond to different species, as displayed in the legend. Arrows highlight scorpion toxin genes and show that most orthogroups were already present in before the split between scorpions and spiders. Tetragnatha kauaiensis and Origins of Genomic Novelty in Spiders GBE Genome Biol. Evol. 13(12) https://doi.org/10.1093/gbe/evab262 Advance Access publication 26 November 2021 7 Downloaded from https://academic.oup.com/gbe/article/13/12/evab262/6443144 by Consejo Superior de Investigaciones Cientificas (CSIC) user on 14 March 2022 identified NPC2, D. silvatica had the least NPC2-genes (7 genes) and A.ventricosus the most (23). Argiope bruennichi had the most CCP, with 41 genes, whereas C. sculpturatus and T.clavipes had only 1 CCP (supplementary table 9, Supplementary Material online;fig. 5). Finally, we identified at least 8 and at most 16 CD36-SNMP genes. T.clavipes and C. sculpturatus had the most CD36-SNMP genes with 16 and 14, respectively, whereas P. tepidariorum and A. bruennichi hadtheleastwith8(supplementary table 9,Supplementary Material online). An analysis of phylogenetic patterns suggests that the chemosensory portfolio is driven by a highly dynamic diversification process. For instance, within GRs there are two genome-specific expansions of genes in the scorpion, one including 1,237 genes and another 235 genes (fig. 4). A similar pattern is observed in the IRs where we find two genome-specific expansions private to the scorpion genome (88 genes, and 382 genes; fig. 4), a large genome-specific gene group with 392 genes in D. silvatica, and another in the Tetragnatha genome including 139 genes. In CCPs, we found expansions in A. bruennichi (5 genes and 13 genes), P. tepidariorum (21 genes), A.ventricosus (8 genes), and D. silvatica (6 genes; fig. 5A). In CD36-SNMP we found expansions in the scorpion (9 genes) and in T. kauaiensis (5 genes; fig. 5B). In NPC2, we found expansions in L. hesperus (14 genes), P. tepidariorum (6 genes), and C. sculpturatus (14 genes; fig. 5C), whereas in CD36-SNMP (fig. 5D) we found expansions in the T. kauaiensis (5 genes) and C. sculpturatus (9 genes) genomes. Silk Gene-Family We identified a total of 24 putative spidroins in the genome of T. kauaiensis (supplementary table 9,Supplementary Material online). After querying these to the NCBI protein database, we identified one Flagelliform spidroin (Flag), four Aggregate spidroins (AgSp), eight Major Ampullate spidroins (MaSp), three Minor Ampullate spidroins (MiSp), one Tubuliform spidroins (TuSp), one Pyriform spidroin (PySp), and one Aciniform spidroin (AcSp). There was one spidroin for which NCBI did not yield any results, and four where the database retrieved more than a single gland as a top-hit (supplementary table 9, Supplementary Material online). Alignments are provided in the Supplementary Material online. Phylogenetic patterns of spidroin shows several genomespecific expansions of the Ma/Mi spidroins, including two separate expansions in the P. tepidariorum genome (25 genes and 10 genes; supplementary table 10,Supplementary Material online;fig. 6), a single expansion in S. mimosarum including 7 genes, another in A.ventricosus including 8 genes, and another in T. kauaiensis including 7 genes. In the remaining spidroins, we find genome-specific expansions in AgSp andPySpinP. tepidariorum, with nine and six genes, respectively. In AcSp there are two smaller lineage-specific clades in A. bruennichi and A.ventricosus. There is a genome-specific expansion in A. bruennichi for the TuSp gland, with seven genes (supplementary table 10,Supplementary Material online;fig. 6). Discussion In this study, we report the sequence assembly of the T. kauaiensis genome, and explore genome evolution across the available spider assemblies. To do so, we controlled for the quality of the assemblies, by focusing on contiguity and completeness (i.e. how complete a genome is from a gene Tree scale A B Tree scale duplications signaled by arrows conserved clusters signaled by squared arrows Argiope bruennichi Araneus ventricosus Centruroides sculpturatus Dysdera silvatica Latrodectus hesperus Parasteatoda tepidariorum Stegodyphus mimosarum Tetragnatha kauaiensis Trichonephila clavipes FIG.4.—Gustatory and ionotropic reception phylogenies (A) Gustatory receptor phylogeny. The phylogeny has 5,595 genes and includes every GR identified in the assemblies herein studied. (B) Ionotropic receptor phylogeny. The phylogeny has 1,932 genes and includes every IR identified in the assemblies herein studied. Arrows indicate major duplications private to specific genomes, whereas squared arrows highlight potentially conserved IR genes (small branch length and small duplicates). Cerca et al. GBE 8Genome Biol. Evol. 13(12) https://doi.org/10.1093/gbe/evab262 Advance Access publication 26 November 2021 Downloaded from https://academic.oup.com/gbe/article/13/12/evab262/6443144 by Consejo Superior de Investigaciones Cientificas (CSIC) user on 14 March 2022 content perspective based on the presence of universal single copy genes), finding that many of these assemblies are highly fragmented and incomplete. We find a wide variation in gene content, repeat content, and genome size in the surveyed spider genomes, which indicates a highly dynamic pattern of genome evolution. Although the low quality of some assemblies did not hamper comparative analyses of the surveyed spider genomes, results should be interpreted with caution. By surveying all repeats and transposable elements (hereafter “the repeatome”) and studying Hox gene duplications, we find that the observed genome size differences are likely driven by the expansion of the repeatome. We also find significant gene-family expansions associated with sensory perception of taste, immunity, and metabolism, which may underlie the diverse biology of spiders. We confirm previous work showing that venoms and chemosensory genes are present in high numbers across the assemblies, and discuss the role of putative ancient whole genome duplication in generating the diversity we observe in spiders. Repeat Content Underlie Genome Size Variation in Spiders Previous evidence from flow cytometry, Feulgen image analysis densitometry, and genome assembly sizes have found wide variation in genome size in spiders (Gregory and Shorthouse 2003;Sanggaard et al. 2014;Kr al et al. 2019). For instance, Gregory and Shorthouse (2003) assembled a large data set comprising 115 species from 19 different families of spiders, finding that spider genomes vary between 5.73 and 0.79 C (7 Gb for the jumping spider H. borealis—724 Mb for the long-jawed orbweaver T. elongata). They also reported a wide variation within relatively closely related species. For instance, genome size in the Salticidae family ranged between 1.73 and 5.73 C (between H. borealis and the peppered jumping spider Pelegrina galathea). Our results are in line with this evidence, because we found variation in genome size among spider assemblies (in our data set the largest genome was A.ventricosus with 3.6 Gb, and the smallest was T. kauaiensis with 1.08 Gb). We also report Candidate carrier protein (CCP) Tree scale Cluster of Differentiation 36 and Neuron Membrane Proteins (CD36-SNMP) Tree scale Niemann-Pick Type C2 (NPC2) Tree scale Odorant binding proteins OBPi Treescale Argiope bruennichi Araneus ventricosus Centruroides sculpturatus Dysdera silvatica Latrodectus hesperus Parasteatoda tepidariorum Stegodyphus mimosarum Tetragnatha kauaiensis Trichonephila clavipes AB CD FIG.5.—Phylogeny of other chemosensory genes. (A) CCP phylogeny; (B) CD36-SNMP phylogeny; (C) NPC2, phylogeny; (D) OBP-like phylogeny. Tetragnatha kauaiensis and Origins of Genomic Novelty in Spiders GBE Genome Biol. Evol. 13(12) https://doi.org/10.1093/gbe/evab262 Advance Access publication 26 November 2021 9 Downloaded from https://academic.oup.com/gbe/article/13/12/evab262/6443144 by Consejo Superior de Investigaciones Cientificas (CSIC) user on 14 March 2022 Gnerre S, et al. 2011. High-quality draft assemblies of mammalian genomes from massively parallel sequence data. Proc Natl Acad Sci U S A. 108(4):1513–1518. Gregory TR, Shorthouse DP. 2003. Genome sizes of spiders. J Hered. 94(4):285–290. Hakes L, Pinney JW, Lovell SC, Oliver SG, Robertson DL. 2007. All duplicates are not equal: the difference between small-scale and genome duplication. Genome Biol. 8(10):R209. Haney RA, Matte T, Forsyth FS, Garb JE. 2019. Alternative transcription at venom genes and its role as a complementary mechanism for the generation of venom complexity in the common house spider. Front Ecol Evol. 7:85. Harper A, et al. 2021. Widespread retention of ohnologs in key developmental gene families following whole-genome duplication in arachnopulmonates. G3 (Bethesda). https://doi.org/10.1093/g3journal/ jkab299 Herberstein ME, Wignall A. 2011. Introduction: spider behavior. In: Herberstein, ME, editor. Spider behaviour: flexibility and versatility. Cambridge: Cambridge University Press. p. 1–30. Hoang DT, Chernomor O, von Haeseler A, Minh BQ, Vinh LS. 2018. UFBoot2: improving the ultrafast bootstrap approximation. Mol Biol Evol. 35(2):518–522. Hoff KJ, Lomsadze A, Borodovsky M, Stanke M. 2019. Whole-genome annotation with BRAKER. In: Kollmar M, editor. Gene prediction: methods and protocols. New York: Springer New York. p. 65–95. Huson DH, Scornavacca C. 2012. Dendroscope 3: an interactive tool for rooted phylogenetic trees and networks. Syst Biol. 61(6):1061–1067. Jackson RR, Cross FR. 2011. Spider cognition. In: Casas J, editor. Advances in insect physiology. Vol. 41. New York: Academic Press. p. 115–174. Kalyaanamoorthy S, Minh BQ, Wong TKF, von Haeseler A, Jermiin LS. 2017. ModelFinder: fast model selection for accurate phylogenetic estimates. Nat Methods. 14(6):587–589. Katoh K, Standley DM. 2013. MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Mol Biol Evol. 30(4):772–780. Keilwagen J, Hartung F, Grau J. 2019. GeMoMa: homology-based gene prediction utilizing intron position conservation and RNA-seq data. Methods Mol Biol. 1962:161–177. King GF, Hardy MC. 2013. Spider-venom peptides: structure, pharmacology, and potential for control of insect pests. Annu Rev Entomol. 58:475–496. Koch L. 2016. Chicago HighRise for genome scaffolding. Nat Rev Genet. 17(4):194. Kono N, et al. 2019. Orb-weaving spider Araneus ventricosus genome elucidates the spidroin gene catalogue. Sci Rep. 9(1):8380. Koutsovoulos G, et al. 2016. No evidence for extensive horizontal gene transfer in the genome of the tardigrade Hypsibius dujardini. Proc Natl Acad Sci U S A. 113(18):5053–5058. Kr al J, et al. 2019. Insights into the karyotype and genome evolution of haplogyne spiders indicate a polyploid origin of lineage with holokinetic chromosomes. Sci Rep. 9(1):3001. Laetsch DR, Blaxter ML. 2017. KinFin: software for taxon-aware analysis of clustered protein sequences. G3 (Bethesda) 7(10):3349–3357. Leite DJ, et al. 2018. Homeobox gene duplication and divergence in arachnids. Mol Biol Evol. 35(9):2240–2253. Letunic I, Bork P. 2019. Interactive Tree Of Life (iTOL) v4: recent updates and new developments. Nucleic Acids Res. 47(W1):W256–W259. Li Z, et al. 2018. Multiple large-scale gene and genome duplications during the evolution of hexapods. Proc Natl Acad Sci U S A. 115(18):4713–4718. Lynch M, Force A. 2000. The probability of duplicate gene preservation by subfunctionalization. Genetics 154(1):459–473. McClintock B. 1984. The significance of responses of the genome to challenge. Science 226(4676):792–801. McGregor AP, et al. 2008. Cupiennius salei and Achaearanea tepidariorum: spider models for investigating evolution and development. Bioessays 30(5):487–498. Nguyen L-T, Schmidt HA, von Haeseler A, Minh BQ. 2015. IQ-TREE: a fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies. Mol Biol Evol. 32(1):268–274. Nieto Feliner G, Casacuberta J, Wendel JF. 2020. Genomics of evolutionary novelty in hybrids and polyploids. Front Genet. 11:792. Ohno S. 1970. The enormous diversity in genome sizes of fish as a reflection of nature’s extensive experiments with gene duplication. Trans Am Fish Soc. 99(1):120–130. Oxford GS, Gillespie RG. 1998. Evolution and ecology of spider coloration. Annu Rev Entomol. 43:619–643. Peona V, Weissensteiner MH, Suh A. 2018. How complete are “complete” genome assemblies?—an avian perspective. Mol Ecol Resour. 18(6):1188–1195. Pineda SS, et al. 2018. ArachnoServer 3.0: an online resource for automated discovery, analysis and annotation of spider toxins. Bioinformatics. 34(6):1074–1076. Putnam NH, et al. 2016. Chromosome-scale shotgun assembly using an in vitro method for long-range linkage. Genome Res. 26(3):342–350. R Core Team 2013. R: A language and environment for statistical computing. [cited 2021 January 01]. Available from: https://cran.microsoft. com/snapshot/2014-09-08/web/packages/dplR/vignettes/xdate-dplR. pdf S anchez-Herrero JF, et al. 2019. The draft genome sequence of the spider Dysdera silvatica (Araneae, Dysderidae): a valuable resource for functional and evolutionary genomic studies in chelicerates. Gigascience 8(8). Available from: http://dx.doi.org/10.1093/gigascience/giz099 Sanderson MJ. 2003. r8s: inferring absolute rates of molecular evolution and divergence times in the absence of a molecular clock. Bioinformatics 19(2):301–302. Sanggaard KW, et al. 2014. Spider genomes provide insight into composition and evolution of venom and silk. Nat Commun. 5:3765. Schmickl R, Yant L. 2021. Adaptive introgression: how polyploidy reshapes gene flow landscapes. New Phytol. 230(2):457–461. Schwager EE, Schoppmeier M, Pechmann M, Damen WGM. 2007. Duplicated Hox genes in the spider Cupiennius salei. Front Zool. 4:10. Schwager EE, et al. 2017. The house spider genome reveals an ancient whole-genome duplication during arachnid evolution. BMC Biol. 15(1):62. Seppey M, Manni M, Zdobnov EM. 2019. BUSCO: assessing genome assembly and annotation completeness. Methods Mol Biol. 1962:227–245. Sheffer MM, et al. 2021. Chromosome-level reference genome of the European wasp spider Argiope bruennichi: a resource for studies on range expansion and evolutionary adaptation. Gigascience 10(1). Available from: http://dx.doi.org/10.1093/gigascience/giaa148 Shingate P, et al. 2020. Chromosome-level assembly of the horseshoe crab genome provides insights into its genome evolution. Nat Commun. 11(1):2322. Sim~ ao FA, Waterhouse RM, Ioannidis P, Kriventseva EV, Zdobnov EM. 2015. BUSCO: assessing genome assembly and annotation completeness with single-copy orthologs. Bioinformatics 31(19):3210–3212. Supek F, Bo snjak M,  Skunca N,  Smuc T. 2011. REVIGO summarizes and visualizes long lists of gene ontology terms. PLoS One. 6(7):e21800. Tarailo-Graovac M, Chen N. 2009. Using RepeatMasker to identify repetitive elements in genomic sequences. Curr Protoc Bioinformatics. Chapter 4:Unit 4.10. Thomas GWC, et al. 2020. Gene content evolution in the arthropods. Genome Biol. 21(1):15. Cerca et al. GBE 16 Genome Biol. Evol. 13(12) https://doi.org/10.1093/gbe/evab262 Advance Access publication 26 November 2021 Downloaded from https://academic.oup.com/gbe/article/13/12/evab262/6443144 by Consejo Superior de Investigaciones Cientificas (CSIC) user on 14 March 2022 Vienneau-Hathaway JM, et al. 2017. Duplication and concerted evolution of MiSp-encoding genes underlie the material properties of minor ampullate silks of cobweb weaving spiders. BMC Evol Biol. 17(1):78. Vizueta J, Escuer P, S anchez-Gracia A, Rozas J. 2020. Genome mining and sequence analysis of chemosensory soluble proteins in arthropods. Methods Enzymol. 642:1–20. Vizueta J, et al. 2017. Evolution of chemosensory gene families in arthropods: insight from the first inclusive comparative transcriptome analysis across spider appendages. Genome Biol Evol. 9(1):178–196. Vizueta J, Mac ıas-Hern andez N, Arnedo MA, Rozas J, S anchez-Gracia A. 2019. Chance and predictability in evolution: the genomic basis of convergent dietary specializations in an adaptive radiation. Mol Ecol. 28(17):4028–4045. Vizueta J, Rozas J, S anchez-Gracia A. 2018. Comparative genomics reveals thousands of novel chemosensory genes and massive changes in chemoreceptor repertories across chelicerates. Genome Biol Evol. 10(5):1221–1236. Vizueta J, S anchez-Gracia A, Rozas J. 2020. bitacora: a comprehensive tool for the identification and annotation of gene families in genome assemblies. Mol Ecol Resour. 20(5):1445–1452. Vollrath F. 1999. Biology of spider silk. Int J Biol Macromol. 24(2–3):81–88. Wilder SM. 2011. Spider nutrition: an integrative perspective. In: Casas J, editor. Advances in insect physiology. Vol. 40. New York: Academic Press. p. 87–136. World Spider Catalog. 2021. World Spider Catalog. Version 22.5. Switzerland: Natural History Museum Bern. Available from: http:// wsc.nmbe.ch. Wu C, Lu J. 2019. Diversification of transposable elements in arthropods and its impact on genome evolution. Genes. 10(5):338. Yim KM, Brewer MS, Miller CT, Gillespie RG. 2014. Comparative transcriptomics of maturity-associated color change in Hawaiian spiders. J Hered. 105 Suppl 1:771–781. Associate editor: Toni Gossmann Tetragnatha kauaiensis and Origins of Genomic Novelty in Spiders GBE Genome Biol. Evol. 13(12) https://doi.org/10.1093/gbe/evab262 Advance Access publication 26 November 2021 17 Downloaded from https://academic.oup.com/gbe/article/13/12/evab262/6443144 by Consejo Superior de Investigaciones Cientificas (CSIC) user on 14 March 2022