Warm temperatures, cool sponges: The effect of increased temperatures on the Antarctic sponge [i]Isodictya[i] sp

Although the cellular and molecular responses to exposure to relatively high temperatures (acute thermal stress or heat shock) have been studied previously, only sparse empirical evidence of how it affects cold-water species is available. As climate change becomes more pronounced in areas such as the Western Antarctic Peninsula, both long-term and occasional acute temperature rises will impact species found there, and it has become crucial to understand the capacity of these species to respond to such thermal stress.

While 5 individual samples are noted here (a single control, and two replicates for both 239 the 3 °C and 5 °C treatments) another control sample was also initially taken and sequencing 240 performed as described above. Despite being morphologically identical, our initial assays of the 241 sequences from that sample, and particularly the sequence of common molecular markers used 242 in phylogenetic comparison, including 28S rRNA and cytochrome c oxidase 1 (CO1) markers, 243 revealed that it is in fact a cryptic related species, which will be the subject of description in the 244 future (see Fig 1C). We have not included this sample in any of the analyses in this work. The 245 lack of a second control sample meant that comparisons of the treatment conditions limited 246 some conclusions, as discussed later in this manuscript.

247
Basic sequencing metrics can be seen in Table 1, alongside those after read cleaning.
248 Our initial FastQC analysis revealed the presence of low quality nucleotide sequence in the 249 second file of many pairs. Cleaning was therefore stringent, and resulted in markedly fewer, but 250 much better average quality, reads for all samples, which were then used for assembly of the

340
In general, we found a large number of genes whose expression was perturbed by heat 341 treatment at both 3 and 5 °C (although please note this is in comparison to a single control 342 sample). Between our control and 3 °C treatments, the transcriptional landscape changed 343 considerably. Stringent criteria were applied for significance given the single control: cut off for 344 FDR of 0.0001, a min abs(log2(a/b)) change of 2, but these results should be taken as indicative 345 rather than quantitative.

457
The full-length sequence of these proteins showed the presence of three classical 458 signature motifs of the HSP70 family (IDLGTTYS; IIDLGGGTFDVSIL; IVLVGGSTRIPKI/VQK) 459 and the ATP/GTP binding site (Fig 4B). These four forms of HSP70 in Isodictya possess 460 relatively low sequence similarity (between 41.4% and 68.1% amino acid identity), and 461 individual isoforms are more closely related to some previously described forms, which may 462 indicate that HSP70A and B are ancestrally shared paralogues within the Porifera (Fig 4A). The 463 HSP70A1 and 2 genes are most closely related to the non-inducible or HSP70 cognate (HSC) 464 described for others invertebrates and vertebrates (Fig 4, Supplementary File 7). Their C-  (Fig 4B).

471
In contrast, the isoform HSP70B showed high sequence similarity to the HSP70 472 sequence of Xestospongia testudinaria (75.4%) and this sequence does not possess 473 tetrapeptide repeat sequences at the C-terminal (Fig 4B). HSP-B is therefore likely the 474 "inducible" HSP gene in this species. It was up-regulated during thermal stress in this 475 experiment, listed among the most strongly-changed in expression (Table 3)

546
In the most similar study to this work previously performed, in the warm water sponge H.
547 tubifera from Philippines 34 , multiple G protein coupled receptors (GPCRs) were noted as 548 differentially expressed. In that experiment, sponges with a normal growth temperature of 29 °C 549 were exposed to 32°C and 34°C for between 4 and 12 hours. 583 We also see a contig upregulated with similarity to oxidative stress-induced growth inhibitor 2-584 like (TRINITY_DN30012_c0_g1_i1), which could play a role in downstream organisation after 585 the detection of an oxidative stress state. It seems possible that these sponges are under 586 severe oxidative stress, and the expression of these genes reflects this, although the 587 experiment was stopped before any outward signs of damage (e.g. necrosis) were observed.

588
The presence of allograft inflammatory factor (TRINITY DN_12683_c0_g1_i1) in our 589 datasets was noted as these have been previously studied in Antarctic species such as the sea 591 high sequence similarity with that known from the sponge Suberites domuncula, and has two 592 conserved calcium binding motifs known as EF hands (Fig 6). AIF-1 sequences from several 593 invertebrate and vertebrate species were aligned and showed relatively high levels of 594 conservation of the protein with both groups. Phylogenetic analyses generated a congruent tree 595 positioning the AIF from sponges in a cluster with corals, distinct from vertebrate sequences. In 596 general, allograft inflammatory factor is known for its role in recovery from injury, rather than any 597 specific role in protecting against heat shock 85 . It has been studied in sponges previously 86 , 598 where it has been shown to act to activate immunocyte-like activity, as well as in protection 599 immediately after trauma. In the Japanese oyster Crassostrea gigas AIF-1 stimulates hemocyte 600 immune activation by enhancing phagocytosis and expression of inflammatory cytokines 87 . Its 601 role in the thermal stress response might therefore be in protection against infection, rather than 602 directly in tissue repair, but both these processes are undoubtedly necessary under prolonged 603 periods of acute thermal stress.

604
That Antarctic species of sponge utilize many of the same pathways to respond to 605 thermal stress as warmer water species is a useful if unsurprising finding. Their efficacy at 606 moderating response to different levels of stress does however seem to be diminished, with 607 many of these genes expressed at both 3 and 5 °C, with no significant difference in their 608 expression levels between these temperatures, as noted elsewhere in this work.  , which was excluded due to its being identity as a cryptic, but separate, species.