Novel reference transcriptomes for the sponges Carteriospongia foliascens and Cliona orientalis and associated algal symbiont Gerakladium endoclionum

Sponge transcriptomes are important resources for studying the stress responses of these ecologically important filter feeders, the interactions between sponges and their symbionts, and the evolutionary history of metazoans. Here, we generated reference transcriptomes for two common and cosmopolitan Indo-Pacific sponge species: Carteriospongia foliascens and Cliona orientalis. We also created a reference transcriptome for the primary symbiont of C. orientalis—Gerakladium endoclionum. Assemblies for C. foliascens, C. orientalis, and G. endoclionum contained 67,304, 82,895, and 28,670 contigs, respectively. Contigs represented 15,248–37,344 isogroups (~ genes) per assembly, and N50s ranged from 1672–4355 bp. Sponge transcriptomes were high in completeness and quality, with an average of 93% of core EuKaryotic Orthologous Groups (KOGs) and 98% of single-copy metazoan core gene orthologs identified. The G. endoclionum assembly was partial with 56% of core KOGs and 32% of single-copy eukaryotic core gene orthologs identified. These reference transcriptomes provide a valuable resource for future research assessing sponge stress responses.


Introduction
Sponges, phylum Porifera, are one of the oldest lineages of multicellular animals (Feuda et al. 2017); hence, investigating the transcriptomes of different sponge species can provide insight into the evolution of metazoans and their gene expression profiles. Sponges have an uncertain future in the face of global climate change (see Bell et al. 2013;IPCC 2014) as well as local stressors including coastal development and increased runoff of nutrients, pesticides and sediments (Kroon et al. 2012;Stender et al. 2014). Transcriptomic analysis of sponges that have been exposed to different environmental conditions would improve our understanding of molecular stress response pathways and enhance our ability to effectively manage these ecologically important filter feeders (e.g., Koutsouveli et al. 2020). Although there are approximately 9,000 described sponge species (Van Soest et al. 2020), relatively few species have published genomes or transcriptomes (e.g., Riesgo et al. 2014a, b).
Here, we assembled the transcriptomes of two common and widely distributed Indo-Pacific sponge species-Carteriospongia foliascens and Cliona orientalis. Both are emerging model species that have been used extensively to study the physiological and ecological effects of environmental stressors on sponges (e.g., Pineda et al. 2016Pineda et al. , 2017a. While both C. foliascens and C. orientalis host diverse populations of bacterial symbionts (see Pineda et al. 2016), C. orientalis additionally hosts an abundant population of eukaryotic Symbiodiniaceae: Gerakladium endoclionum (LaJeunesse et al. 2018), which comprises up to 96% of its algal symbiont community (Ramsby et al. 2018). We used sequences generated from the C. orientalis holobiont, i.e., host and symbiont, to construct a partial reference transcriptome for Gerakladium endoclionum. Matching host and symbiont transcriptomes provide a valuable tool to understand the holobiont response to changing environmental conditions and determine the cause-effect pathways for declining host health with environmental change.

Samples and sequencing
Samples of C. foliascens and C. orientalis were collected in May 2015 from Fantome Is. (S 18°41.028 0 E 146°30.706) and Pelorus Is. (S 18°32.903 0 E 146°29.172 0 ), respectively, in the Great Barrier Reef under permits G12/ 35,236.1 and G13/35,758.1. Since C. orientalis is a bioeroding sponge, ten C. orientalis drill cores (* 5 cm in diameter) were collected from a single individual growing on a dead colony of Porites sp. An individual of C. foliascens was cut into ten pieces (see Pineda et al. 2016). Sponges were healed and acclimated under natural light and flow-through seawater for 4 weeks before experiments were performed.
To capture the full range of gene expression responses, sponges were subjected to five different treatments at the Australian Institute of Marine Science (AIMS) National Sea Simulator: (i) decreased salinity, (ii) elevated temperature, (iii) elevated suspended sediment concentrations (SSCs) and sediment deposition, (iv) light attenuation, and (v) no stress control (see Supplemental for details). One genotype was used per species across all treatments to control for genetic variation. Two clones of each species were used for each treatment. Immediately after each treatment, * 1 cm 3 of tissue from each clone was frozen in liquid nitrogen and stored at -80C (Riesgo et al. 2012).
Approximately 50 mg of each clone was ground using a mortar and pestle under a thin layer of liquid nitrogen to limit RNA degradation. All tools were rinsed in ethanol followed by RNase Zap (Sigma-Aldrich, USA) to remove contamination and deactivate RNA degrading enzymes. Total RNA was isolated using the Zymo ZR RNA miniprep kit, with in-column DNase digestion, and cleaned using the Zymo RNA Clean and Concentrator kit (Zymo Research, USA). Total RNA quality was checked using gel electrophoresis and quantified using a Quant-iT RiboGreen Assay (Thermo Fisher Scientific, USA). For each sponge species, RNA from individual treatments was combined in equal amounts from all sponge clones to a final total RNA concentration of 93 ng ll -1 for C. foliascens and 160 ng ll -1 for C. orientalis. To isolate eukaryotic messenger RNA (mRNA), a TruSeq Stranded mRNA-seq sample prep was performed. The mRNA was then sequenced across two Illumina HiSeq2500 lanes at the Ramaciotti Centre for Genomics (University of New South Wales, Australia), generating 2 9 100 base pair (bp) paired-end (PE) rapid runs.

Transcriptome assembly and annotation
Reads were trimmed using publicly available scripts (Matz 2015;Meyer 2016) and assembled with Trinity v 2.8.5 (Grabherr et al. 2011) following established protocols (Kenkel and Bay 2017, see Supplemental Material and the github repository below for a full description). After assembly, additional quality control was performed to ensure that only target transcripts, i.e., derived from C. foliascens, C. orientalis or G. endoclionum, were included in their respective reference transcriptomes using protocols outlined by Kitchen et al. (2015) and Kenkel and Bay (2017). These quality controls also removed short contigs, ribosomal and mitochondrial RNA contamination (see Supplemental Material). Within each of the three transcriptomes, contigs were assigned to isogroups (* genes) and assigned gene names, gene ontologies (
The transcriptomes for C. foliascens and C. orientalis were largely complete based on BUSCO analysis (92.8% and 94.2% complete, respectively) and the representation of nearly all core eukaryotic Orthologous Groups (KOGs) 97.9% and 98.7%, respectively (Table 1). BUSCO analysis of the transcriptome of G. endoclionum was 32.3% complete, and 56% of core KOGs were identified (Table 1). A reduced completeness (33-42%) in genomes isolated from intracellular Symbiodiniaceae also occurs in corals (Liu et al. 2018), perhaps due to poor representation of related phyla within the BUSCO eukaryotic gene set. The G. endoclionum transcriptome contained 86% more isogroups than the Symbiodiniaceae transcripts identified within the transcriptome assembly of the closely related sponge holobiont, Cliona varians (Riesgo et al. 2014b). Therefore, the current transcriptome for G. endoclionum was considered useful for future studies in hospite.
C. foliascens and C. orientalis are widely distributed throughout the Indo-Pacific and are ecologically important components of the reef ecosystem, particularly in the GBR where C. foliascens is a dominant component of the benthos and C. orientalis bioerodes calcium carbonate substrate, including the skeletons of live coral. These data represent a substantial contribution to publicly available poriferan genetic resources and will provide the framework needed to develop these two sponge species into models for field and laboratory studies, particularly research examining the molecular mechanisms underpinning how reef sponges respond to environmental perturbation.
Acknowledgements This research was funded by the Western Australian Marine Science Institution (WAMSI) as part of the WAMSI Dredging Science Node and made possible through investment from Chevron Australia, Woodside Energy Limited, BHP Billiton as environmental offsets and by co-investment from the WAMSI Joint Venture partners. The views expressed herein are those of the authors and not necessarily those of WAMSI. Brian W. Strehlow was supported by a University of Western Australia (UWA) Scholarship for International Research Fees, University International Stipend, and UWA Safety-Net Top-Up Scholarships. Brian W. Strehlow was further supported by the Villum Investigator Grant awarded to Don Canfield (No. 16518). The funders had no role in study design, data collection, and analysis, decision to publish, or preparation of the manuscript. The authors are grateful to E. Botté and G. Millar for their assistance with computing issues. They also thank the staff at the AIMS National Sea Simulator for their expertise and assistance in the tank-based experiments. The authors acknowledge the Manburra People, as the Traditional Owners of the sea country where this work took place. We pay our respects to their elders' past, present and emerging, and we acknowledge their continuing spiritual connection to their sea country.
Availability of supporting data All data can be accessed here: https://www.dropbox.com/sh/82ue5l16n4xzxww/AABENUi-Cdbm_ z-6x4Gj3qICa?dl=0. Raw data have also been deposited on NCBI's SRA under accession numbers PRJNA639714 and PRJNA639798 for the C. orientalis holobiont and C. foliascens, respectively. The final, annotated assemblies as well as tabular files containing Kegg and GO information are also included in Supplemental Material. Scripts used throughout the assembly process can be found here: https://github. com/bstrehlow/Transcriptome_assembly/tree/master

Compliance with ethical standards
Conflict of interest On behalf of all authors, the corresponding author states that there is no conflict of interest.