Complementary mechanisms for neurotoxin resistance in a copepod

Toxin resistance is a recurring evolutionary response by predators feeding on toxic prey. These adaptations impact physiological interaction and community ecology. Mechanisms for resistance vary depending on the predator and the nature of the toxin. Potent neurotoxins like tetrodotoxin (TTX) and saxitoxin (STX) that are highly toxic to humans and other vertebrates, target conserved voltage-gated sodium channels (NaV) of nerve and muscle, causing paralysis. The copepod Calanus finmarchicus consumes the STX-producing dinoflagellate, Alexandrium fundyense with no effect on survival. Using transcriptomic approaches to search for the mechanism that confers resistance in C. finmarchicus, we identified splice variants of NaVs that were predicted to be toxin resistant. These were co-expressed with putatively non-resistant form in all developmental stages. However its expression was unresponsive to toxin challenge nor was there any up-regulation of genes involved in multi-xenobiotic resistance (MXR) or detoxification (phases I or II). Instead, adults consistently regulated genes encoding digestive enzymes, possibly to complement channel resistance by limiting toxin assimilation via the digestive process. The nauplii, which were more susceptible to STX, did not regulate these enzymes. This study demonstrates how deep-sequencing technology can elucidate multiple mechanisms of toxin resistance concurrently, revealing the linkages between molecular/cellular adaptations and the ecology of an organism.


Complementary mechanisms for neurotoxin resistance in a copepod
Vittoria Roncalli*, Petra H. Lenz, Matthew C. Cieslak and Daniel K. Hartline

Naupliar response to A. fundyense
Field collection and cultivation of C. finmarchicus Copepod development consists of an embryonic stage ("eggs") followed by six naupliar (NI-NVI) and six copepodite (CI-CVI) stages. Here, the transcriptomic response to A. fundyense was investigated in the "late naupliar stage" consisting of a mix of NV and NVI individuals.
These developmental stages were obtained by raising nauplii from eggs produced in the laboratory. C. finmarchicus were collected using a vertical net tow (75 cm diameter, 560 µm mesh) on July 1, 2012 in the Gulf of Maine near Mount Desert Rock (Lat: 44° 2'N; Long: 68°3'W). Adults were sorted from the plankton collection and transferred into baskets with bottoms covered with 560 µm mesh and suspended in 3.5 L culture jars of seawater containing live Rhodomonas sp. Eggs were collected and reared in a batch that differed in age by less than 48 hrs. After the nauplii reached the feeding stage (NIII) Rhodomonas sp. was added to the container. Nauplii were checked every 2-3 days, and after they reached the targeted developmental stages (NV-NVI), they were removed from culture and incubated with the experimental food as described below. All copepod cultures and experiments were maintained in a Percival Model I-36VL Incubator System (Percival Scientific, Inc., Perry, IA, USA) with temperature set at 10 ºC and the light:dark cycles set at 14L:10D.

Experimental design
For the experiment, 70-86 C. finmarchicus late nauplii (NV-NVI) were transferred into 100 mL crystallizing dishes with filtered seawater and fed for two days on one of two In the nauplius experiment, the non-toxic flagellate Rhodomonas sp. was added daily at 8,000 cells mL -1 d -1 to each control replicate. Nauplii in the HD group were fed a diet of 100% A.
fundyense at daily rations of 200 cells mL -1 d -1 per experimental replicate. Estimated carbon content for the two treatments was similar with 304 and 358 µgC L -1 respectively for the control and HD treatments 1-3 . Nauplii were checked under a dissecting microscope to assess mortality, algal ingestion (colored/filled guts), possible malformations (none were found) and behavior (active swimming, escape swims) after 1 and 2 days. On day 2, nauplii were harvested from each treatment and biological replicate and immediately processed for RNA extraction.

Mapping of short reads and identification of differentially expressed genes (DEGs)
Illumina sequencing for the six RNA-Seq libraries generated more than 121 million reads with 16 to 24 million reads per library with an average of 20 million across all samples (100 bp, bases. This was followed by the elimination of low quality reads (cutoff "Phred" score = 20) as well as Illumina adapters. An average of 24% of reads were removed, leaving from 12 to 18 million reads per sample for relative gene expression analysis. Each quality filtered RNA-Seq library was then mapped to an existing C. finmarchicus reference transcriptome 4,5 (96,090 contigs) using the software Bowtie 6 (v. 2.0.6). The reference transcriptome was generated through the de novo assembly of over 400 million reads from six developmental stages (embryo, early nauplius, late nauplius, early copepodite, late copepodite, adult female) as described previously and available online 5 . The reference transcriptome was designed to minimize ambiguous mapping (≤ 1% of mapped reads mapped > 1 time), thus, it did not include splice variants, and the Na V 1. CONTROL vs HD. Transcripts were identified as differentially expressed using the Exact test (p<0.05) followed by a multiple comparison correction using the Benjamini-Hochberg method (false discovery rate <5%) as implemented by edgeR 7 . Relative expression was quantified as a ratio in units of Log 2 (experimental/control) where a value of 0 represents equal expression between the experimental condition and control.

Functional annotation of differentially expressed genes (DEGs)
Functional annotation for genes identified as differentially expressed was undertaken using a local blast webserver. Blastx algorithm was used to search against the NCBI SwissProt protein database (downloaded on 18 th January, 2016) onto a local Beowulf Linux computer cluster; a maximum E-value for annotation of 10 -3 was employed. The Blast search yielded annotation for 66% of the DEGs. The resulting Blast annotations were then used to retrieve Gene Ontology (GO) terms with UniProt (http://www.uniprot.org/uploadlists/) under three categories: biological processes, molecular function and cellular component, which are hierarchically organized into levels. The maximum E-value used for this analysis was of 10 -6 . Enrichment analysis was performed separately for up-and down-regulated genes with GO terms (426 and 96 respectively) against the 10,344 genes with assigned GO terms in the C. finmarchicus reference transcriptome 4,5 . This analysis was comparable to an earlier study on the effects of A. fundyense on gene expression of adult females 5 . The analysis was implemented using the software BLAST2GO (v. 2.6.4) performing the Fisher's Exact Test followed by Multiple Testing correction of False Discovery rate (FDR <5%) 8 . It is important to note that in many cases multiple functions (GO terms) are assigned to individual genes. Table S1. Calanus finmarchicus Na V channel sequences identified by in silico searches of the reference transcriptome for individuals from Gulf of Maine (GOM) 5 I  GAXK01036301  GBFB01017903  i  II  GAXK01012592  GBFB01192791  e  III  GAXK01114023  not found  j  III  GAXK01009404  not found  k  III GAXK01063206 not found f IV GAXK01022998 GBFB01042925 * The "NaV1.X" category is a catchall of short sequences containing P-loops, but not necessarily all from the same gene. Sequences from GOM and NOR overlap extensively and share the same P-loops, but in general differ in length and in a few corresponding residues.  r1=74, r2=72,r3=77) and ALEX (r1=70, r2=84,r3=86). ** Behavior observation: A=active swimming, B= inactive on the bottom of container Ω Average of 3 biological replicates: CONTROL (r1=74, r2=72,r3=77) and ALEX (r1=66, r2=80,r3=81). Table S3. Summary of sequencing and mapping results for C. finmarchicus late nauplii (NV-NVI) feeding on CONTROL (100% Rhodomonas sp.) and ALEX (100% A. fundyense) for 2 days. Each treatment consists of 3 biological replicates (r1, r2, r3). For each replicate number of Illumina sequenced reads (100bp), number of high-quality filtered reads used for the mapping, mapping overall alignment rate (%) and number of reads that mapped 1 time.    Table S6. List of selected genes involved in protein turnover differentially expressed in C. finmarchicus late nauplii and adult females feeding on A. fundyense for 2 days. Genes are classified in different protein families based on their blast annotation. For each gene Accession No. (NCBI) and relative fold change in expression (absolute) are listed. The direction of expression (up-or down-regulated) and the magnitude are indicated by arrows ("red = up" and "green = down" regulated genes). Relative expression for all the serine proteases was similar in the two stages and ranged between 1 and 52 RPKM. Figure S1. Magnitude of response. Fold change of differentially expressed genes (DEGs) in C. finmarchicus late nauplii feeding on A. fundyense (HD) for 2 days. A) up-regulated genes and B) down-regulated genes. Fold change is absolute. Pie chart of the annotated DEGs regulated in C. finmarchicus late nauplii feeding on A. fundyense HD diet for 2 days. The pie chart includes Gene Ontology (GO) terms belonging to the biological process (BP) category.