*2.2. Phylogenetic Analysis of OvoA Transcripts from Marine Metazoans*

Phylogenetic analysis of OvoA sequences deduced in this study for marine arthropods was used to support their annotation and to investigate their relationship to each other and to those from other marine metazoans. The analysis generated a consensus tree with several clades representing the different phyla. Consistent with what has been previously reported for marine metazoans, OvoA sequences from Mollusca, Echinodermata, Cnidaria, and Chordata phylum cluster separately and in individual clades (Figure 3). The OvoA sequences identified in this study for marine arthropods separated into two clades (Figure 3). The first clade, highly separated from all other metazoans, is closely related to OvoA from Placozoa and from two Porifera (Figure 3). This clade included 15 OvoA sequences, including 12 sequences from copepods and three from decapods (Figure 3). The second clade, with a total of 13 OvoA sequences, included seven copepods, four decapods and two amphipods. The seven copepods included five OvoA sequences from Calanoida and two sequences from Cyclopoida (Figure 3).

**Figure 3.** Phylogenetic tree for OvoA gene in marine metazoans. The unrooted tree shows the relationships between OvoA identified in this study for marine arthropods and OvoA from selected marine metazoans. Amino acid sequences were aligned using ClustalW, and FASTTREE was used to build a maximum-likelihood phylogenetic tree (Galaxy v. 2.1.10+ galaxy1) using the protein evolution model JTT+ CAT. Color coding refers to the different phylum: light blue = Arthropoda (sequences identified in this study), green = Mollusca, red = Chordata, orange = Echinodermata, pink = Cnidaria, purple = Porifera, yellow = Placozoa. Scale bar indicates the number of amino acid substitution per site.

## *2.3. Ovo A Expression in Calanus finmarchicus and C. helgolandicus*

The expression of OvoA was examined in the copepods *C. finmarchicus* and *C. helgolandicus* using previous RNA-Seq data for these copepods feeding on toxic algal species [28,29].

In *C. finmarchicus*, exposure to the saxitoxin producing dinoflagellate *Alexandrium fundyense* induced changes in the expression of OvoA (Figure 4A). The increase in expression of OvoA did not depend on the dose of the toxic algae but was time affected (Figure 4A). Specifically, at two days, OvoA was found up-regulated in females fed with toxic algae at low (LD) and high (HD) doses (LD = 5.1 ± 0.11 Log2 [RPKM + 1], HD = 5.03 ± 0.5 Log2 [RPKM + 1]) compared with females on the control diet *Rhodomonas* sp. (Figure 4A). In contrast, a longer exposure to the dinoflagellate had no significant effect; at five days, the expression of OvoA was similar between diets and not significantly different compared with females on the control diet (Figure 4A).

In *C. helgolandicus* exposure to a toxic diet did not affect the expression of OvoA. In females exposed for 5 days to the toxic diatom *Skeletonema marinoi*, OvoA expression was not significantly different from the expression of females feeding on the control diet *P. minimum* (TOXIC = 9.97 ± 1.35, Log2 [RPKM + 1]; CONTROL = 9.80 ± 0.36, Log2 [RPKM + 1]). The reported lack of OvoA differential expression after 5 days on the toxic algae is similar to that reported for *C. finmarchicus* (Figure 4A).

Changes in the expression of OvoA were also examined in *C. finmarchicus* through development. Using previously published RNA-Seq expression data, we examined the expression of OvoA in different stages including embryos (E), early nauplii (EN), early copepodites (CI), late copepodites (CIV and CV), and adult females (F). Compared with all other stages, the lowest and significantly different expression of OvoA was found in embryos (0.2 Log2 [RPKM + 1]) (Figure 4B). The OvoA expression significantly increased across development reaching its peak in the first copepodite stage (CI) (6.2 ± 0.19 Log2 [RPKM + 1]). A high expression was also maintained through the pre-adult stage (CV) (5.7 ± 0.19 Log2 [RPKM + 1]) but it decreased in the adult stage (Figure 4B).

**Figure 4.** OvoA expression in *Calanus finmarchicus*. In both graphs, relative expression was normalized by length RPKM (Log2) adding pseudocounts of 1. (**A**) Expression of OvoA in adult females exposed for 2 days (2d) and 5 days (5d) to a control diet *Rhodomonas* sp. (C), and low (LD) and high doses (HD) of the toxic dinoflagellate *Alexandrium fundyense* [29]. Bar graphs indicate mean with standard deviation (SD) of the three replicates in each diet; (**B**) Expression of OvoA across development: embryos [E], early nauplii (NII-NIII) [EN], early copepodites [CI], late copepodites (preadult CIV and CV) [CIV] [CV], females [F]. Bar graphs indicate 2-way ANOVA of the three replicates in each sample (exception CI and CIV with two replicates). Significant differences (*p* < 0.05; 2-way ANOVA followed by post-hoc Tukey's test) among stages are indicated by small letters over the bars.

#### **3. Discussion**

Ovothiols are small sulfur-containing natural metabolites playing a key role in protecting the organisms against oxidative stress. For its chemical properties and low molecular weight, Ovothiol A is one of the strongest antioxidants reacting with ROS and radicals significantly faster than other natural thiols. Ovothiols are biosynthesized in a two-way step with OvoA, 5-histidylcysteine sulfoxide synthase, being the key regulator. The OvoA gene is highly conserved and found in almost all marine metazoans including Porifera, Cnidaria, Hemicordata, and Echinoderamata [15]. Interestingly, some organisms such as insects and fish, are unable to produce those secondary compounds. This seems to be related to the two gene loss event that occurred through evolution for the ancestral Ecdysozoa (nematodes and arthropods) and the ancestor of Osteichthyes fish. Although fish lack the OvoA to biosynthesize the compounds, ovothiols have been found in different tissues of freshwater fish suggesting that those organisms acquire the metabolite through their nutrition [15]. For arthropods, there was no evidence of OvoA gene and no reports on the presence of ovothiols. In our study, we present the results for the mining of the publicly available transcriptomic resources on the National Center for Biotechnology Information (NCBI) database (TSA) limiting the results to the marine arthropods, mostly represented by the Crustacea subphylum. However, during our searches, OvoA hits resulted also for terrestrial species. For the completeness of our mining (data are not shown) we further examined those sequences with reciprocal blast and structural domain analysis as part of our workflow. This resulted in the identification of OvoA transcripts from eight insects (6 Hemiptera and 2 Diptera). Considering that the focus of the study is on marine organisms, we did not expand the investigation on the terrestrial arthropods. However, our preliminary results might suggest that more investigations are needed in order to clarify also the presence of this gene in terrestrial species.

Marine arthropods are among the most distributed living organism in aquatic environment; copepods, the insects of the sea, dominate the zooplankton. Zooplanktonic organisms have a key role in the energy transfer to higher trophic levels. Through their lifecycle, copepods are commonly exposed to abiotic and biotic stressors that disturb their cellular homeostasis with negative effects on their fitness and in extreme conditions their survival. To cope with the stress, organisms typically activate a cellular stress response (CSR) [30]. This response has the goal to repair and prevent macromolecular damage, to activate cell cycle checkpoints, to reallocate energy resources, and in extreme cases to activate a programmed cell death [30]. The critical part of the CSR is the antioxidant system, a set of enzymes acting against the oxidative protein damage induced by elevated concentration of reactive oxygen species (ROS). Antioxidant proteins, commonly used as biomarkers of oxidative stress, include thioredoxin/thioredoxin reductase, glutaredoxin/glutathione/glutathione reductases, metallothioneins, and cytochrome P450 proteins.

The increase of new transcriptomic resources for marine arthropods, mostly for organism from the copepoda order, has provided the opportunity to expand targeted gene discovery in these organisms. A better understanding of the complexity of gene families of interest, in particular those associated with response to stress, opens new opportunities to discover new biomarkers that could be used for functional studies. For its antioxidant properties, OvoA could be a new biomarker to evaluate antioxidant stress responses in marine organisms. Here we focused on the target identification in marine arthropods of the transcripts encoding for the OvoA enzyme. Our study describes the ovothiol gene distribution in many marine arthropods providing the first report of OvoA gene in the Arthropoda phylum. For mining transcriptomic resources for marine organisms, which included 19 copepods and nine malacrostacans, we found in each organism a single OvoA transcript encoding protein. Seventy percent of the proteins deduced from the predicted OvoA transcript appear to be full-length and showed the expected structural hallmark. Phylogenetic analysis was used to support the annotation and also to evaluate the relationship of OvoA from marine arthropods with other marine metazoans. Based on our results we found that the OvoA sequences from marine arthropods are significantly different from the other metazoans by clustering in independent clades. However, to our surprise, the OvoA sequences separated in two clades that included both copepods and malacrostacans with no distinction between copepod orders. The separation of two clades might suggest that through evolution the OvoA has diverged, however more studies are needed.

In many marine organisms, ovothiols are known for their role in the oxidative response against environmental stressors. In sea urchins, ovothiols are significantly over-expressed in response to metals and toxic algae [2]. In the starlet sea anemone *Nematostella vectensis* OvoA has been suggested as protector against environmental pollutants [31]. A significant high expression of OvoA was found in organisms exposed to dispersant and/or sweet crude oil exposure alone or combined with ultraviolet radiation (UV) [31]. Here, to support a possible role of OvoA as antioxidant in copepods, we used previously generated RNA-Seq data for two calanoid copepods exposed to harmful microalgae to examine the expression of OvoA. In the previously published study, the physiological response of *C. finmarchicus* females exposed to the neurotoxin-producing dinoflagellate *Alexandrium fundyense* was investigated after two- and five-days exposure [28]. The authors reported that at 2 days, *C. finmarchicus* activates a cellular stress response that involves differential expression of many genes associated with molecular chaperoning, apoptosis, cell cycle checkpoint, intracellular signaling, and protein turnover [28]. Although detoxification was not the major component of the response, up-regulated with the diet there were also some antioxidant biomarkers such as glutathione S-transferase (GST), sulfotransferase, and thioredoxin. Here, we found that the expression of OvoA was significantly high in *C. finmarchicus* exposed to the toxic diet compared with individuals on a control diet. The differential regulation of OvoA at 2 days reported here, is consistent with a role of OvoA as antioxidant as part of the CSR previously reported [28]. Furthermore, the lack of significant difference in the expression between the two toxic diets (low and high doses) agrees with the findings

that the *C. finmarchicus* response to the toxic algae is not dose dependent. At 5 days, we did not report differences in the expression of OvoA between individuals on the toxic diets (both doses) and the control. This agrees with the fact that at 5 days the *C. finmarchicus* CSR became a cellular homeostatic response, characterized by fewer differentially expressed genes [28]. A homeostatic response is activated when the organism physiologically adapts to new conditions and starts to re-establish homeostasis by counteracting the stress in a specific way [30].

Consistent with *C. finmarchicus*, we also observed no significant differences in OvoA expression when the congener *C. helgolandicus* was fed for 5 days on the oxylipin-producing diatom *S. marinoi*, with respect to the control diet. Oxylipins are lipid-derived info chemicals that regulate the structure and functioning of natural phytoplankton communities [25], and also act as defensive compounds against consumer copepods, by inducing offspring abnormalities, thereby reducing population recruitment at sea [25,32]. Interestingly, the same *C. helgolandicus* females for which we examined the expression of OvoA, showed strong up regulation of genes involved in stress response and xenobiotic detoxification, such as GST and cytochrome P450 [29]. Hence, it is possible that these *C. helgolandicus* individuals were activating a detoxification system based on other antioxidants than ovothiols, that protected the adult copepod from the direct ingestion of the harmful diet. Although we did not have information on OvoA expression in *C. helgolandicus* females feeding for 2 days on *S. marinoi*, a previous study showed that these copepods activated a CSR by over-expressing genes encoding for cellular chaperons, as well as for proteins involved in signal transduction pathways and cell cycle [33]. Given the similar cellular response of the two copepod species when exposed for short feeding times to harmful algal diets, we could speculate that ovothiols may play a role in the antioxidant defensive system of *C. helgolandicus*, as well. However, further studies are needed to confirm this hypothesis.

In addition to their role in response to stressor, ovothiols play a role also during development [2]. In the sea urchin *P. lividus*, a significant high expression of the OvoA gene was found not only in the post-fertilized eggs but also in the pluteus stage during larval development. At the pluteus stage, the organism undergoes a series of morphological changes leading to the final metamorphosis. Alteration of the redox homeostasis is common through development as a consequence of oxygen radicals or reactive oxygen species (ROS) that act as primary or secondary messengers to promote cell growth [34]. The periods associated with morphological changes are characterized by an increase of metabolic activity and apoptosis that leads to elevated concentration of ROS. Thus, it is not strange that OvoA transcript is highly expressed in those stages of development due to its role in ROS scavenging. Copepods have a molt cycle with six naupliar and five copepodite stages [35]. A significant high expression of OvoA was found here in *C. finmarchicus* in the first and the last copepodite stages. Those two stages are key through development. Between the six naupliar and the first copepodite stage the organism undergoes morphological rearrangements while the 5th copepodite stage is when the organism sexually differentiates [35]. Although copepods have a molt cycle that does not involve a pupal stage as in sea urchin (with final metamorphosis), the high expression at those two key stages is like the sea urchin findings, with OvoA playing a key role during late development. Taken together, the pattern of expression reported for OvoA transcript in *C. finmarchicus* fed on the toxic diet suggests that the production of ovothiols is a possible adaptive strategy to cope with environmental stressors. A differential expression through development could be associated either with an antioxidant response to the byproducts that are generated during metamorphosis or to a specific signaling role that these metabolites can have through development.

In addition to its role as antioxidant, ovothiols have also been reported to have antiproliferative and anti-inflammatory activities [2,10]. Our results provide the first evidence that OvoA is present in copepods, decapods, and amphipods, opening new questions on its distribution also among other zooplanktonic species. The high expression of OvoA in *C. finmarchicus* in response to a toxic diet as well as during the key transition stages of the

molt cycle, might suggest that ovothiol A has an antioxidant role also in copepods. In the recent years, drug discovery has focused on marine planktonic organisms, rather than macroorganisms because of their advantage of easy culturing in closed controlled systems and to obtain huge biomass [36,37]. Both phytoplanktonic and zooplanktonic species, have been shown to have the capability of producing several antioxidant, anti-inflammatory, anti-diabetes, anticancer, and other bioactivities compounds useful for the prevention and treatment of human pathologies [38–42]. Thus, the findings that ovothiol A might be produced by zooplankters can have interesting possible future biotechnological applications and can stimulate further studies on planktonic metabolites for pharmaceutical, nutraceutical, and cosmeceutical applications.

#### **4. Materials and Methods**
