Investigation of Best Practices for Venom Toxin Purification in Jellyfish towards Functional Characterisation

The relative lack of marine venom pharmaceuticals can be anecdotally attributed to difficulties in working with venomous marine animals, including how to maintain venom bioactivity during extraction and purification. The primary aim of this systematic literature review was to examine the key factors for consideration when extracting and purifying jellyfish venom toxins to maximise their effectiveness in bioassays towards the characterisation of a single toxin.An up-to-date database of 119 peer-reviewed research articles was established for all purified and semi-purified venoms across all jellyfish, including their level of purification, LD50, and the types of experimental toxicity bioassay used (e.g., whole animal and cell lines). We report that, of the toxins successfully purified across all jellyfish, the class Cubozoa (i.e., Chironex fleckeri and Carybdea rastoni) was most highly represented, followed by Scyphozoa and Hydrozoa. We outline the best practices for maintaining jellyfish venom bioactivity, including strict thermal management, using the “autolysis” extraction method and two-step liquid chromatography purification involving size exclusion chromatography. To date, the box jellyfish C. fleckeri has been the most effective jellyfish venom model with the most referenced extraction methods and the most isolated toxins, including CfTX-A/B. In summary, this review can be used as a resource for the efficient extraction, purification, and identification of jellyfish venom toxins.


Introduction
Animal venom is a proven source of potent bioactive compounds, many of which have been adapted into pharmaceuticals. Specifically, 11 venom-derived drugs are commercially available and approved by both the United States Food and Drug Administration and the European Medicines Agency [1]. The wide range of venom bioactivity can be seen in these different drugs being utilised in the treatment of coronary syndrome, chronic pain treatment, diabetes, hypertension, and sciatica [1]. Only one of these drugs, Prialt ® , comes from a marine organism (Conus magus), which is symptomatic of the widening gap between terrestrial and marine venom research. This deficiency in marine venom research can be anecdotally attributed to the difficulties of working with marine venom in contrast to terrestrial venom. These difficulties include the availability of target organisms in the wild (often seasonal) and poor knowledge of aquaculture of the organisms and extraction procedures from tissue(s) that maintain bioactivity. Regarding the extraction of biomolecules from tissues, a significant diversity of methodologies exists in the literature for jellyfish venom. later in this review. This is best highlighted by the vast majority of the captured articles including a recommendation for the continuation of the research, yet the venom and toxin fractions they identified did not receive any subsequent publications detailing further refinement. While speculative, this demonstrates the difficulty of jellyfish purification in the initial stages of research, regularly causes the delivery of results that do not justify further research into the venom when more favourable sources of bioactive molecules may be present. Overall, the class Scyphozoa represented the most common jellyfish class for venom purifications (90 publications), followed by Cubozoa (39 publications) and Hydrozoa (12 publications). No venom had been reported purified from any Staurozoa. gradually increased from 2001, peaking in 2012 (13 publications), and then trending downwards up to now. This decrease may be attributed to the inherent difficulties associated with jellyfish venom extraction and purifications relative to terrestrial venoms which is further discussed later in this review. This is best highlighted by the vast majority of the captured articles including a recommendation for the continuation of the research, yet the venom and toxin fractions they identified did not receive any subsequent publications detailing further refinement. While speculative, this demonstrates the difficulty of jellyfish purification in the initial stages of research, regularly causes the delivery of results that do not justify further research into the venom when more favourable sources of bioactive molecules may be present. Overall, the class Scyphozoa represented the most common jellyfish class for venom purifications (90 publications), followed by Cubozoa (39 publications) and Hydrozoa (12 publications). No venom had been reported purified from any Staurozoa. year . In the inset, the pie chart shows the relative distribution (number and percentage) represented by jellyfish class.

Jellyfish Venom Identification
A total of 202 venom proteins have been reported from jellyfish, with varying degrees of purification (Table 1), which could be classified into (1) crude venom, (2) a venom fraction, and (3) isolated venom toxins. Crude venom was defined as venom that had been removed from nematocysts, filtered and/or sonicated and/or centrifuged, and intended to be representative of a jellyfish's entire venom toxin repertoire (i.e., the entire venome). A venom fraction was defined as crude venom that had had its components separated (e.g., liquid chromatography) to refine the potential number of venom toxin products. Finally, Figure 1. The bar graph shows the number of jellyfish venom purification papers identified per year . In the inset, the pie chart shows the relative distribution (number and percentage) represented by jellyfish class.

Jellyfish Venom Identification
A total of 202 venom proteins have been reported from jellyfish, with varying degrees of purification (Table 1), which could be classified into (1) crude venom, (2) a venom fraction, and (3) isolated venom toxins. Crude venom was defined as venom that had been removed from nematocysts, filtered and/or sonicated and/or centrifuged, and intended to be representative of a jellyfish's entire venom toxin repertoire (i.e., the entire venome). A venom fraction was defined as crude venom that had had its components separated (e.g., liquid chromatography) to refine the potential number of venom toxin products. Finally, isolated toxins were defined as a single homogenous product, pure enough to obtain a single kilodalton (kDa) measurement, which varied from 1 to 600 kDa. For each classification of purification, the venom had additionally been tested in toxicity assays to obtain an LD50 (or LD50 equivalent for cellular assays). Of the 202 reported venoms, 99 (49%) were crude venom, 62 (31%) were venom fractions, and 38 (20%) were isolated venom toxins.

Purification Levels of Jellyfish Venom
A total of 31 jellyfish species had at least one venom purification level reported (Table 1 and Figure 2), with Chironex fleckeri, Chrysaora quinquecirrha (Atlantic Sea nettle), and Cyanea capillata (Lion's mane jellyfish) most extensively investigated. C. quinquecirrha shares C. fleckeri's status as a 'medically relevant' jellyfish [31], due to its rapidly acting cardiotoxic venom [40] and powerful haemolytic properties [36,37]. In addition, it inhabits the East Coast of the United States and the Gulf of Mexico, thereby serving as a convenient model for venom researchers in North America. The observed abundance of C. capillata venom research follows a similar logic, with a powerful cardiotoxic venom [40,52,54,57] and a niche distribution within the northern oceans, extending to the Artic, which helps to explain their disproportionately large representation in the literature.

Purification Levels of Jellyfish Venom
A total of 31 jellyfish species had at least one venom purification level reported (Table  1 and Figure 2), with Chironex fleckeri, Chrysaora quinquecirrha (Atlantic Sea nettle), and Cyanea capillata (Lion's mane jellyfish) most extensively investigated. C. quinquecirrha shares C. fleckeri's status as a 'medically relevant' jellyfish [31], due to its rapidly acting cardiotoxic venom [40] and powerful haemolytic properties [36,37]. In addition, it inhabits the East Coast of the United States and the Gulf of Mexico, thereby serving as a convenient model for venom researchers in North America. The observed abundance of C. capillata venom research follows a similar logic, with a powerful cardiotoxic venom [40,52,54,57] and a niche distribution within the northern oceans, extending to the Artic, which helps to explain their disproportionately large representation in the literature. The data highlight that Aurelia aurita, Cassiopea xamachana, Catostylus mosaicus, Cyanea nozakii, Nemopilema nomurai, and Pelagia noctiluca have had toxic fractions identified, yet not a single toxin purified. These jellyfish represent an excellent opportunity for research as there exists a strong foundation of effective venom purification and fractionation. Moreover, these jellyfish only represent 32 species of over 3500 known jellyfish species from which no, or very little, research has been performed on their venom. Although hydrozoans represent around 77% of all medusozoan species [2,3], they only occupied 12% of the venom purifications reported.

Jellyfish Venom Extraction
Jellyfish venom for subsequent purification of single toxins and toxicity bioassay con- The data highlight that Aurelia aurita, Cassiopea xamachana, Catostylus mosaicus, Cyanea nozakii, Nemopilema nomurai, and Pelagia noctiluca have had toxic fractions identified, yet not a single toxin purified. These jellyfish represent an excellent opportunity for research as there exists a strong foundation of effective venom purification and fractionation. Moreover, these jellyfish only represent 32 species of over 3500 known jellyfish species from which no, or very little, research has been performed on their venom. Although hydrozoans represent around 77% of all medusozoan species [2,3], they only occupied 12% of the venom purifications reported.

Jellyfish Venom Extraction
Jellyfish venom for subsequent purification of single toxins and toxicity bioassay confirmation have been successfully obtained using a variety of different extraction and purification procedures ( Table 2). As categorised by Carrette and Seymour (2004) [33] venom extraction methods can be separated into, natural discharge of intact nematocysts, mechanical disruption of nematocysts, chemically induced discharge of intact nematocysts and removal of liquid from individual intact nematocysts [33]. Only natural and mechanic methods of extraction returned individual toxins.
The most commonly referenced isolation methods [22,23,37,89,94,113,114] were developed by Burnett et al. (1992) [43] and Bloom et al. (1998) [26], the latter of which had been refined from Burnett's original method. Briefly, jellyfish tentacles were freshly removed from beachside individuals, then immediately refrigerated within two volumes of seawater for 1-4 days. For venom extraction, the tentacle samples were shaken, and an aliquot of water was taken daily for filtration with filter paper prior to being viewed under a microscope to confirm the release of intact cnidocytes from tissue. The cnidocytes were then lyophilised and stored (−70 • C) prior to extraction. The venom was extracted from the cnidocytes via sonication (3 × 20 s with 1 min intervals) in ice-cold water (4 • C) before centrifugation at 20,000× g for 1 h at 4 • C and then immediately used in toxicity assays. This approach was regularly referred to as the 'autolysis' venom extraction technique since it relied on autolysis of tentacle tissue (but not the cnidocytes themselves). Variations in autolysis venom extraction included tentacle incubation times of one day [88] to up to 6 weeks [82]. However, the longest time of autolysis that yielded a single proteinaceous toxin was 4 days. Therefore, it is recommended that autolysis be no longer than 4 days.
The efficiency of the venom extraction autolysis technique may be attributed to a few factors. First, it has been speculated that the intracellular environment of cnidocytes, including high pressure, contributes to venom stability. In addition, venom is known to act rapidly to facilitate processes that disable prey; therefore, they do not require a high degree of structural stability that could maintain a longer half-life outside of the cellular environment [26].
A variation of the autolysis technique involved the agitation of jellyfish tentacles in a buffered solution for up to 6 weeks to allow for maximal separation of the cnidocytes. Cnidocytes were then chilled in a pressure cell at 12,000 psi for 15 min, followed by a freeze-thaw cycle to help liberate the venom [82]. While this method is a conventional approach for liberating the intracellular components of a cell for small molecule analysis, it should not be used when isolating proteinaceous toxins from jellyfish venom, as multiple studies have reported that freeze-thaw cycles cause a significant decrease in bioactivity [65,83] particularly for proteinaceous compounds. This is relatively well known, as the phase change of an aqueous environment can cause alterations at the tertiary and quaternary levels of proteins by mechanical disruption of the structure during freezing.
A strict thermal regulation regime appears to play a critical role in the preservation of jellyfish venom bioactivity and, therefore, its potency. This is irrespective of cnidocyte separation since a large number of jellyfish toxins have been successfully isolated using whole tentacle homogenisation [84,91,93,95,98,102] and whole tentacle lysis [83]. Thermal denaturation of jellyfish toxins has been studied to investigate their sensitivity. For example, the activity of the CnPH toxin was significantly reduced when exposed to temperatures above 45 • C [116] and the venom activity of the flame jellyfish (Rhopilema esculentum) was found to be reduced at 40 • C [8]. It was additionally inferred during the purification of ClGp1 from lysed cnidocytes that it should be performed as fast as possible to preserve bioactivity [98].  Whole tentacle homogenisation and lysis use mechanical, electrical, or chemical means to break open cells, and in doing so, cause nematocysts to fire, releasing the venom. Thus, whole tentacle extraction undoubtedly adds ambiguity as to the true source of the toxin and introduces background biomolecules to the subsequent purification process. For instance, the enzyme hydralysin was identified from a Chlorohydra viridissima tentacle homogenate [91], yet it is a cytolytic/neurotoxic proteinaceous toxin that originates from a non-cnidocyte source. This, as well as other examples, indicates the possibility that not all jellyfish "venom" is delivered from nematocysts [15,91]. Whole tentacle extraction may be a better choice if the sample size is low as it will provide a more concentrated venom extract. However, if the jellyfish venom is of low toxicity and if the venom readily loses bioactivity under extraction conditions, it may be significantly impacted during the hours or days that the tentacles are suspended in the separation solution.

Jellyfish Venom Purification
A total of 38 individual jellyfish venom toxins have been identified ( Table 2). In all of the isolated toxins within the literature-captured jellyfish venom purification studies, liquid chromatography was the primary method of choice, while most of them utilised a combination of at least two chromatography techniques, including HPLC [93], RP-HPLC [6], SEC [7], CEX [94], and AEX [102]. Only two studies required additional chromatography to ensure the purity of a single toxin [7,59]. Non-chromatography purification processes have also been used in tandem with liquid chromatography, using acidic precipitation [102] and a "salting out" protein precipitation method [103]. While these methods were shown to have produced single toxins from crude venom, they should only be used as a final purification step for identification, when the activity and purity of the toxin have been confirmed. This is due to protein precipitation causing the denaturation of proteins, thereby removing any chance to confirm its native function in bioassays.
Size exclusion chromatography was used in more than 70% of the jellyfish toxin purification studies, thereby representing the most common liquid chromatography method. SEC is particularly useful when purifying unknown toxins, as it does not require organic solvents. Organic solvents, such as acetonitrile, methanol, and ethanol, commonly used in RP-HPLC and chloroform in HPLC, are known to modify protein structure, thereby potentially affecting a toxin's function. Unlike CEX and AEX, SEC does not require prior knowledge of the proteins' solubility, charge, or isoelectric point and will provide better resolution across a sample with many compounds. SEC can be performed as high-throughput [23] or used as a last step to "polish" toxins [98].

Toxicity Assays of Venom towards Proposed Function
The combined data obtained, from jellyfish purification to identification, indicate that overall direct comparisons should not be performed due to the lack of commonality between studies regarding the toxicity assay ( Figure 3). For jellyfish venoms, mice have been the most used animal model system for toxicity characterisation (to establish lethal dosage; Figure 3A). This is consistent with studies on other types of venom (derived from marine or terrestrial organisms), whereby functional toxicity assays using mice are considered relatively cheap with accessible ethical requirements, as well thiersimilar biological, physiological, and symptomatic effects to humans. However, as animal ethics for vertebrate experimentation become stricter, invertebrate animal models (e.g., crayfish, flies, moths, and spiders) may be more widely used. In addition, jellyfish predate most commonly on small fish and crustaceans making them the most biologically relevant models for whole animal toxicity testing. marine or terrestrial organisms), whereby functional toxicity assays using mice are considered relatively cheap with accessible ethical requirements, as well thiersimilar biological, physiological, and symptomatic effects to humans. However, as animal ethics for vertebrate experimentation become stricter, invertebrate animal models (e.g., crayfish, flies, moths, and spiders) may be more widely used. In addition, jellyfish predate most commonly on small fish and crustaceans making them the most biologically relevant models for whole animal toxicity testing.  Erythrocytes are excellent for cell-based toxicity assays to establish haemolytic activity. To assess jellyfish venom toxicity, sheep erythrocytes have been the most used, likely due to their ethical sourcing, relatively low cost, and high availability ( Figure 3B). Chicken erythrocytes have also been commonly used, which although non-mammalian, are still nucleated cells that have demonstrated clear comparative results against different types of jellyfish venom [8,40,45,78,88]. Human erythrocytes have also been well used in cell-based toxicity assays, primarily with relevance to investigating cardiovascular and haemolytic impacts on humans; they are also more sensitive to venom compared to sheep erythrocytes [92]. The haemolytic effects have been studied in human erythrocytes to reveal venom protein interactions with membrane lipids, which were proposed to be directly and most likely involved in pore formation [34,71,92,109]. Research investigating these impacts has used various venom purification methods as this model gives autonomy in method designation, although the HU50s (one haemolytic unit, the amount of protein sample required to induce 50% haemolysis, referred to as LD50 in this review) were quite different between methods conducted on the same species [34]. However, the most dependable method was presented by Yanagihara and Shohet (2012) [34], where human erythrocytes were recognised as being a reproducible source for experiments. They showed specific membrane breakdown when interacting with venom toxins; although considering that different jellyfish venoms contain different repertoires of toxins, they consequently have different haemolytic activity that might not be possible to compare; in addition, reportedly, careful consideration should be taken with the enzymes used in bioassays to avoid inhibitory effects on venom proteins [91].
Human cancer cell lines are often used to elucidate the harmful effects of venoms on humans and to understand the molecular mechanisms of venom toxicity, and they have been used on numerous occasions for jellyfish venom toxins [57,65,72,118,123,124,129]. They provide a reliable, high-throughput, convenient, and relatively cheap approach [124], which is primarily relevant for medical and pharmaceutical applications, specifically when it comes to first aid treatments regarding understudied jellyfish venom stings [123,124]. This model, however, has its own limitation, with a bias for the most accessible cell lines available, thus causing variation in the obtained results [57,65,72,118,123,124,129]. The most common mechanistic function reported using these cell lines involved reactive oxygen species (ROS) and cell apoptosis.
However, it was revealed that comparisons using the same animal model and the crude venom of the same species may not aid in elucidating the most effective purification or extraction technique. Separate investigations of Chironex fleckeri "crude venom", described in the articles as crude tentacle venom [28], stock venom [9], nematocyst venom [26], and Chironex venom [32], all used near identical methodologies while producing significantly different LD50s in mice of 11, 12, 23.4, and 150 µg/mL, respectively. This variation in LD50 is most likely due to individual variations in venom potency between jellyfish or venom degradation that may have occurred prior to individual capture and isolation. As jellyfish are difficult to capture while in the ocean, especially smaller species, specimens are most often collected after they have washed up on the beach, where animal death and thermal desiccation quickly degrade the proteinaceous compounds within the venom. Furthermore, as very few jellyfish species are able to be maintained in captivity, repeated bioassays using the same individual animal's venom is not possible, thus compounding the source of error due to individual variability of venom potency. Although not captured in the scope of this review, it is important to highlight a 2019 study of Nemopilema nomurai which showed that individual nomurai venom showed significantly different enzymatic metalloproteinase activity, with up to a 77-fold difference in haemolytic activity [130]. Moreover, the geographical location and the jellyfish's age are correlated with significant variations in jellyfish venom of the same species [131].
Published studies have noted difficulties associated with the attempted analysis of reported jellyfish venom due to accidental misreporting of the specific activity of venom and the vastly different methodologies involved in the preparation and bioassays [82]. Complicating analysis further, jellyfish venom can have a unique haemolytic activity profile that can be highly variable between jellyfish species [65,116] and that very rarely aligns with general cytotoxicity assay results. Furthermore, the objective of many research studies was not to classify the general function or toxicity of the venom towards the discovery of a toxin. Instead, their objective was focused on the biodiscovery of a compound with a predetermined, specific function: for example, studies of jellyfish venom targeting angiotensin I, converting enzyme (ACE) inhibitors [76,93,112]. As such, jellyfish venom studies currently lack a standard methodology for toxicity measurements in cells and animal models using LD50 measurements. Function-specific bioassays of toxins are too specific and lack any type of standardisation, meaning that the comparison of bioassay data between articles for the purpose of elucidating the most effective purification techniques may not be possible. This highlights that jellyfish venom LD50s should not be used as a means to compare the effectiveness of venom purification or extraction methods. The limitations of venom potency can be best mitigated by obtaining the largest possible sample size when collecting venom for extraction and excluding juvenile or desiccated jellyfish from extractions.

Chironex fleckeri, a Model for Jellyfish Venom Purification
The box jellyfish, Chironex fleckeri, is often referred to as the most medically relevant jellyfish [17,31,111], given that its powerful haemolytic and cardiotoxic toxins have, to date, been prime sources for biodiscovery. In fact, the increase in jellyfish venom purification studies can generally be attributed to the attention brought to the field by extensive research performed on cardiotoxic venoms derived from C. fleckeri (Figure 2). The box jellyfish is well known for an extremely lethal and fast-acting venom that can be lethal to humans and prey alike. Thus, C. fleckeri represents the most researched jellyfish, with the most venomous LD50 score, and from which the most jellyfish toxins have been identified. Although there are outcomes that have reported on the toxin properties and lethality of the whole venom [9,10,23,24,[27][28][29][30][32][33][34]99] leading to many bioactive fractions [9,[24][25][26][27], the primary lethal toxin responsible for human fatalities has not been reported. Nonetheless, C. fleckeri represents a well-established model for the purification of jellyfish toxins and could be further used by researchers for methods of venom characterisation of Medusozoan.

Conclusions
This review has systematically compiled all kinds of Medusozoan purified venoms (crude venom, toxic fractions, and single toxins), which may serve as a valuable resource for venom researchers. Despite initial intensive foundational research since 1974, with research peaking in 2011, very few toxins have been successfully purified across the Medusozoans, with the exceptions being C. fleckeri and C. rastoni. Importantly, in order to characterise a single toxin when guided by bioassays, a strict thermal management plan is required to avoid loss of potency, thereby improving the success rate of purifications. Although a clear universal methodology for venom purification could not be elucidated, there is a clear trend for two-step liquid chromatography involving SEC to isolate a single toxin. This variation in the methodologies used, as well as a lack of duplication and the natural variation in individual species' venom, currently precludes our ability to compare bioassay data. This points towards jellyfish venom extraction and purification methods being entirely dependent upon the observed chemical and functional behaviour of the toxins of interest. Therefore, they should be experimentally determined per species via functional bioassays of the toxin, to confirm the conservation of the bioactivity at each step. Performing comparisons of venom LD50, even when using the same model organism and same jellyfish species, has been shown to be of questionable value given the discussed limitations. To date, C. fleckeri has been the most effective jellyfish venom model, from which researchers have developed widely used extraction and purification methods, leading to a large number of venoms being categorised. While jellyfish venom research continues to trail behind terrestrial venom research, this does provide ample opportunity for novel discovery.

Method
Only literature captured using the methodology described herein is referenced in this review to maintain the integrity of the analysis, unless specifically stated. The inclusion criteria were as follows: Scopus, Web of Science, and Pub Med databases were searched using the keywords "purification or purified" and "venom" in "all fields" using the access provided by the University of the Sunshine Coast. Additionally, five secondary search terms, including "Jellyfish", "Cubozoa" or "Cubozoan", "Hydrozoa" or "Hydrozoan", "Scyphozoa" or "Scyphozoan", and "Staurozoa" or "Staurozoan", were added and combined to ensure the capture of all jellyfish purification publications for analysis (Figure 4), with duplicate articles removed. The exclusion criteria were applied as follows: (1) if no venom purification was detailed in the publication, (2) if the venom detailed in the publication is not a single Medusozoan venom, and (3) if the venom purification detailed in the publication was non-guided with no bioassays. Using the above methodology, the literature from between 1 January 1975-15 December 2022 was captured as the earliest and latest articles. This methodology followed the Transparent Reporting of Systematic Reviews and Meta-Analysis (PRISMA) [20]; the database of all of the literature information is provided in File S1. cation is not a single Medusozoan venom, and (3) if the venom purification detailed in the publication was non-guided with no bioassays. Using the above methodology, the literature from between 1 January 1975-15 December 2022 was captured as the earliest and latest articles. This methodology followed the Transparent Reporting of Systematic Reviews and Meta-Analysis (PRISMA) [20]; the database of all of the literature information is provided in File S1.