**2. Results**

### *2.1. Stress-Induced Mucus Secretion and Autolysis of A. coerulea*

Jellyfish are able to respond quickly to external environmental stimuli, although they have limited movement ability. We have previously noted the active secretion of jellyfish mucus induced by external stimulation, in that the surrounding seawater turns sticky when disturbed. In this study, we first checked the quantity-time relationship of stress-induced mucus secretion as well as the autolysis that rapidly occurs in dying jellyfish.

External stress was performed by removing *A. coerulea* from the environmental seawater and, as expected, the sticky liquid samples were largely secreted [13] and collected every 10 min for a total of 1 h (Figure 1). Two obvious phases in volume collection were displayed, whereby the volume decreases to a minimum at 30 min, followed by a gradual increase within 60 min (Figure 1A). However, protein concentration of each sample is positively associated with the time (Figure 1B), which is further confirmed by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) (Figure 1C). Proteins in mucus are mainly distributed in three concentrated molecular weight ranges—100–250 kDa, 50–100 kDa and 37–50 kDa—while proteins in tissue are more dispersed. A gentle trough of the curve for protein quantity (mg/kg) of each sample is shown at 30 min (Figure 1D). Meanwhile, obvious crevices in the umbrella part indicate that jellyfish autolysis starts at or even earlier than 30 min. Therefore, the decreases of mucus volume (Figure 1A) and protein amount (Figure 1D) in the first 30 min imply an adaption to the stress while the increase of mucus volume (Figure 1A) and protein amount (Figure 1D) in the latter 30 min is probably due to jellyfish autolysis. Interestingly, straight line correlations (R<sup>2</sup> > 0.99) for both mucus volume (mL/kg, Figure 1E) and protein quantity (mg/kg, Figure 1F) with time indicate a continuous release of proteins through two different mechanisms—i.e., stress-induced mucus secretion followed by jellyfish autolysis—without clear boundaries. The 20 min sample is less influenced by both residual seawater and jellyfish autolysis, and is therefore the sample selected for proteomics and metabolomics.

**Figure 1.** Stress-induced mucus secretion and autolysis of *A. coerulea*. ( **A**) Mucus volume of each sample/10 min; (**B**) Protein concentration of each sample/10 min was determined by the Bradford method; ( **C**) SDS-PAGE of collected mucus and tissue homogenate of *A. coerulea*. M: protein molecular size marker; Lanes 1–6: mucus samples of *A. coerulea* collected in 60 min; Lane 7: jellyfish tissue homogenate; ( **D**) Protein quantity (mg/kg)/10 min; (**E**) Accumulative volume of mucus (mL/kg)/60 min; (**F**) Accumulative protein quantity of mucus (mg/kg)/60 min. Mean ± SD (*n* = 4) is shown. \* *p* < 0.05 and \*\* *p* < 0.01 indicate a significance difference as compared to the control.

### *2.2. Proteomic Comparison of Secreted Mucus and Tissue Homogenate*

All proteomics raw MS data were aligned to obtain peptide sequence information and matched to proteins from our previously constructed transcriptomic database for *A. coerulea*. A total of 2729 proteins from 10,560 peptides were identified by LC-MS/MS, where 2421 proteins from 8866 peptides were matched in tissue and 1208 proteins from 4148 peptides were matched in mucus. A Venn diagram shows that 1438 and 225 proteins are separately located in tissue and mucus, respectively. Proteins numbering 183, 523, and 267 are elevated, lowered or unchanged, respectively, among the 983 overlapped proteins in mucus when compared to those in tissue (Figure 2A). This profile is further supported by the quantitative ratio histogram of the overlapped proteins between the two groups where the log2(FC) (fold change) value from mucus (numerator) vs. tissue (denominator) values were distributed from −6 to +6, with a peak located at around −2, rather than 0 (Figure 2B). Although protein number of secreted mucus is far less than that of tissue homogenate, two proteomic indexes including amino acid (AA) (Figure 2E) and molecular weight (MW) distributions (Figure 2F) are the same for the two groups. The distribution curves of four other indexes, including peptide count (Figure 2C), protein sequence coverage (Figure 2D), electric point (Figure 2G) and exponentially modified protein abundance index (emPAI) (Figure 2H), are slightly shifted to the right in mucus when compared to those in tissue. These proteomic analysis results indicate that proteins identified in secreted mucus were successfully separated to a high level of purity and that they were independent of those identified in tissue homogenate.

**Figure 2.** Proteomic comparison of secreted mucus and tissue homogenate. ( **A**) Venn diagram of protein composition in mucus and tissue. There were 2421 proteins identified in tissue and 1208 identified in mucus. 1438 and 225 proteins were only found in tissue and mucus, respectively. Of these 983 overlapping proteins in both groups, 183 were found at elevated levels in mucus, while 523 were at lower levels and 267 were at consistent levels when compared to those in tissue. (**B**) Histogram of quantitative ratio of the overlapped proteins between the two groups. The log2(FC) (fold change) value from mucus (numerator) vs. tissue (denominator) is distributed mainly between −6 to +6, with a peak located at around −2 instead of 0. Six indexes, including peptide count distribution ( **C**), protein sequence coverage ( **D**), AA distribution (**E**), MW distribution (**F**), electric point distribution (**G**), and emPAI distribution ( **H**), are compared for tissue and mucus.

### *2.3. Gene Ontology Analysis*

The Gene Ontology (GO) project provides an ontology of defined terms describing the characteristics of genes and their products in any organism [24]. It covers three domains: biological process (BP), cellular component (CC), and molecular function (MF). We categorized all identified proteins according to the levels of protein expression in secreted mucus and tissue homogenate. The distributing tendencies of tissue-enriched proteins (FC < 0.5), mucus-enriched proteins (FC > 2) and proteins with no change (0.5 < FC < 2) are similar, although obvious differences are seen in specific terms. Since the quantity of proteins in mucus is much less than that in tissue, we used percentage of the total identified proteins in each group as the horizontal axis, whereas the exact amount of proteins is labeled on the right side of the transverse column (Figure 3A–C).

Among the top 10 terms of BP, the ratios of proteins fall from near 50% in 'cellular process' to less than 5% in 'biological adhesion' in tissue-enriched proteins. The top three terms, 'cellular process', 'metabolic process' and 'biological regulation' in mucus-enriched proteins have much lower ratios than those in tissue-enriched proteins (Figure 3A). An intermediate distribution is seen in the proteins with no change. The largest difference between the two groups comes from the CC subcategories. Tissue-enriched proteins are mainly distributed in three intracellular locations: 'cell', 'cell part' and 'organelle', then followed by the membrane-related terms 'membrane', 'membrane part' and 'macromolecular complex'. Comparatively, the most abundant locations in mucus-enriched proteins are 'membrane' and 'membrane part'. The ratios of protein levels in the intracellular locations 'cell', 'cell part' and 'organelle' are much lower than those in tissue-enriched proteins, while the extracellular terms 'extracellular region' and 'extracellular part' show elevated percentages of proteins and larger ratios in mucus-enriched proteins when compared to those in tissue-enriched proteins (Figure 3B). The distribution profiles of MF subcategories are similar across all three groups. Interestingly, the ratios of the top two terms, 'binding' and 'catalytic activity', are close to 50%, which is significantly higher than that of other terms, with ratios of less than 10%. Although the number of 'molecular function regulator' proteins is similar between 'mucus-enriched' and 'tissue-enriched' samples, the ratio of 'molecular function regulator' proteins in mucus-enriched proteins is much higher than those in the other two groups, which may potentially be used as the molecular indicators of jellyfish stress (Figure 3C).

We turned our attention to the mucus-enriched proteins, of which the GO enrichment diagram (Figure 3D) is built with the parameters: protein number > 15, rich factor 0.2–0.8 and −log10(*p* value) > 7. The most significant feature is that the 'extracellular region' shows the highest −log10(*p* value) (bright red), although its 'rich factor' and 'protein quality' are not the largest. Moreover, a Venn diagram was constructed to further divide the extracellular proteins in mucus-enriched proteins into three subclasses—'extracellular region', 'extracellular matrix' and 'extracellular space'—with 23, 32 and 28 proteins, respectively, in each subclass. The 'extracellular region' and 'extracellular matrix', 'extracellular region and extracellular space', 'extracellular space and extracellular matrix' share five, eight and two proteins, respectively. Only two proteins overlap across all three subgroups (Figure 3E).

**Figure 3.** *Cont*.

**Figure 3.** Comparative Gene Ontology (GO) analysis of identified proteins in tissue and mucus of *A. coerulea*. Three groups, namely 'tissue-enriched proteins', 'mucus-enriched proteins', and 'proteins with no change' are displayed. The tissue-enriched proteins or mucus-enriched proteins represent proteins exclusively and highly expressed in tissue homogenate (FC < 0.5) or secreted mucus (FC > 2). The group 'proteins with no change' implies that the proteins expressed in both mucus and tissue show no obvious difference (0.5 < FC < 2). (**A**) Biological process (BP). The horizontal axis is the ratio of proteins in the total identified proteins, whereas the vertical axis provides description of the matched GO terms. The protein numbers are labeled on the right side of each transverse column. (**B**) Cellular component (CC). (**C**) Molecular function (MF). (**D**) Diagram of GO enrichment in mucus-enriched proteins. The horizontal axis indicates the rich factor, i.e., the proportion of the number of differentially expressed proteins vs. the total number of proteins in the same GO term. The vertical ordinates represent the matched GO terms. The bubble shows the number of proteins matched in each GO term. The color represents −log10(*p* value): Logarithmic conversion of Fisher exact test *p* value. (**E**) Venn diagram of the extracellular proteins in mucus-enriched proteins. Three subclasses 'extracellular matrix', 'extracellular region' and 'extracellular space' are colored by blue, yellow and green, respectively.

### *2.4. KEGG Pathway Analysis*

KEGG (Kyoto Encyclopedia of Genes and Genomes) is a common bioinformatics tool and was utilized in this study to provide pathway mapping of identified proteins. The top 20 matched pathways among the 226 successfully mapped pathways are mainly associated with the intracellular synthesis of metabolites, as well as intracellular functions (Figure 4A) in tissue-enriched proteins. The number of proteins gradually falls from 74 in the 'Ribosome' and 72 in 'Carbon metabolism' to only one protein in each pathway. By comparison, only 81 pathways were matched in mucus-enriched proteins. There are 25 proteins matched to the 'ECM (extracellular matrix)—receptor interaction' pathway, significantly more than in other pathways (Figure 4B). The largest spot (shown in red) has a rich factor of 0.52 and represents an 'ECM-receptor interaction'. This is shown in the KEGG enrichment diagram for protein-enriched mucus (Figure 4C). Three subclasses, namely collagen, laminin and thrombospondin (THBS) are further divided from the 25 matched proteins identified in the 'ECM-receptor interaction' that are known to function in the matrix-receptor interactions (Figure 4D).

**Figure 4.** Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway annotation of identified proteins in tissue and mucus of *A. coerulea*. (**A**) KEGG pathway annotation of tissue-enriched proteins. The horizontal axis is the number of proteins, whereas the vertical ordinates are the terms of the KEGG pathways. (**B**) KEGG pathway annotation of mucus-enriched proteins. (**C**) KEGG pathway enrichment of mucus-enriched proteins. The horizontal axis indicates rich factor and vertical ordinates are the terms of the KEGG pathways. Rich factor is the proportion of the number of differentially expressed proteins vs. the total number of proteins in the same KEGG pathway. The bubble shows the number of proteins matched in the KEGG pathway. The color represents −log10(*p* value): Logarithmic conversion of Fisher exact test *p* value. (**D**) The 'ECM-receptor interaction' pathway matched from the KEGG PATHWAY database where the elevated proteins from mucus-enriched proteins are colored by red.
