TECHNICAL FIELD

The invention relates to the semisynthetic preparation of neosaxitoxin (neoSTX) and the purification from extracts of cultures of selected isolates of Alexandrium pacificum of the intermediate GTX1,4 used in its preparation.

BACKGROUND ART

As stated in the publication of Garcia-Altares (2017), marine microalgal toxins constitute one of the most diverse and sophisticated groups of natural products. Examples are paralytic shellfish toxins (PSTs) such as saxitoxin (STX), its analogues and derivatives. Gonyautoxins (GTXs) are sulphated analogues of STX and marine bacteria can transform GTXs into STX through reductive eliminations. In marine environments the main producers of STX are eukaryotic dinoflagellates .

STX is a monoterpenoid indole alkaloid containing a tricyclic 3,4- propinoperhydropurine system with 2 guanidinium moieties formed by the N¾ groups in the positions C2 and C8 of the reduced purine:

STX blocks voltage-gated sodium channels (VGSCs), but also binds to calcium and potassium channels. The nature of the substituents greatly influences the overall toxicity of saxitoxin analogues. The hydroxylation of Nl, e.g. as in neosaxitoxin (neoSTX) does not play a major role in binding affinity, but seems to increase potency.

The prior art is replete with disclosures of the cosmetic and therapeutic applications of PSTs, including their use as local anaesthetics and

analgesics. The publication of Mezher (2018) discloses that the US Food and Drug Administration (FDA) plans to develop guidance documents to encourage the development of extended-release local anaesthetics which could replace the need for systemic oral opioids in certain situations. The expectations of the US FDA are for the development of new non-opioid drugs to treat chronic pain that could provide a safer alternative for patients who require long term use of analgesic drugs. A limitation on the exploitation and widespread adoption of these applications is the availability of the PSTs in sufficient quantity and of sufficient purity to render their use in the manufacture of pharmaceutical preparations commercially viable.

The following publications disclose the preparation of gonyautoxin 1 (GTX1), gonyautoxin 4 (GTX4) or neosaxitoxin (neoSTX) . Often the preparation is on an analytical scale, or does not provide the quantity and purity required for use of the preparation as an active pharmaceutical ingredient (API) .

The publication of Hall et al (1984) discloses the confirmation by x-ray crystallography of the position and identity of the three substituents which, with the parent compound, form the array of twelve saxitoxins found in protogonyaulax .

The publication of Daigo et al (1985) discloses the extraction and isolation of neosaxitoxin (neoSTX) from specimens of crab. The dose-death time curve obtained for the isolated neoSTX was clearly distinguishable from the curve for saxitoxin (STX) .

The publication of Laycock et al (1994) discloses methods for the

purification of some of the common paralytic shellfish poisoning (PSP) toxins in quantities sufficient for use as analytical standards. The PSP toxins were purified from the dinoflagellate Alexandrlum excavatum, the giant sea scallop (Plagopecten magellanlcus) hepatopancreas and the cyanobacterium

Aphanlzomenon flos-aquae .

The publication of Ravn et al (1995) discloses what are asserted to be optimal conditions for extraction of paralytic shellfish toxins from a clone of Alexandrlum tamarense . The paralytic shellfish toxins are extracted with acetic acid and hydrochloric acid in the concentration range 0.01 to 1.0 N. Concentrations of hydrochloric acid in the range 0.03 to 1.0 N were observed to cause the amount of Cl and C2 toxins to decrease sharply with a

concomitant increase in the amount of gonyautoxins 2 (GTX2) and 3 (GTX3) .

The publication of Tsai et al (1997) discloses the detection of paralytic toxicity by a tetrodotoxin bioassay in specimens of crab. Partial

purification and characterisation of the toxins demonstrated the main toxin to be tetrodotoxin with minor amounts of gonyautoxins (GTXs) and neosaxitoxin (neoSTX) .

The publication of Siu et al (1997) discloses the examination of the effects of environmental and nutritional factors on population dynamics and toxin production in Alexandrlum catenella. Optimum conditions for the growth of this species of dinoflagellate are disclosed along with the toxin profile for a species grown under these conditions. The toxin profile as detected by HPLC was found to include in descending order GTX4, GTX3, GTX1, B2, neosaxitoxin (neoSTX) and saxitoxin (STX) .

The publication of Sato et al (2000) discloses the transformation of the 0- sulfate group of GTX1 and GTX4 to methylene to form neosaxitoxin. The transformation was achieved using thiols such as glutathione and

intermediates of the conversion were isolated.

The publication of Parker et al (2002) discloses an investigation of the autotrophic growth of the toxic dinoflagellate, Alexandrium minutum, in three different high biomass culture systems, assessing growth, productivity and toxin production. The organism was grown in aerated and non-aerated two litre Erlenmeyer flasks, 0.5 litre glass aerated tubes, and a four litre lab scale alveolar panel photobioreactor. A marked increase in biomass and productivity in response to aeration was observed. A maximum cell concentration of 3.3 x 10 ⁵ cells/mL, a mean productivity of 0.4 x 10 ⁴ cells/mL/day and toxin production of approximately 20 pg/L/day with weekly harvesting was reported.

The publication of Baker et al (2003) discloses the production by bacterial strains isolated from saxitoxin-producing dinoflagellates of compounds that could easily be mistaken for gonyautoxin 4 (GTX4) .

The publication of Miao et al (2004) discloses the isolation of gonyautoxins (GTX1, GTX2, GTX3 and GTX4) from two strains of Alexandrium minutum Halim.

The strain of Alexandrium minutum Halim designated Amtk4 is asserted to be suitable for the preparation of gonyautoxins.

The publication of Jiang and Jiang (2008) discloses the establishment of optimal conditions for the extraction of paralytic shellfish poisoning toxins from the gonad of Chlamys nobilis. The extraction uses acetic acid and hydrochloric acid in the concentration range of 0.04 to 1.0 mol/L. The use of hydrochloric acid in the concentration range of 0.25 to 1.0 mol/L was shown to cause a significant decrease of the toxins Cl, C2 and GTX5 and the concomitant increase in the toxins GTX2,3. The amount of the three unstable toxins did not show any change when acetic acid was used in the extraction.

The publication of Liu et al (2010) discloses the culture of toxin producing Alexandrium catenella in the laboratory. A maximum cell density of 0.4 x 10 ⁴ cells/mL was obtained within eight days of culture. Analysis by high performance liquid chromatograph (HPLC) of a crude extraction showed the major toxic components to be Cl/2, GTX4, GTX5 and neoSTX at concentrations of about 0.04550, 0.2526, 0.3392, 0.8275 and 0.1266 pmol/L, respectively. The publication of Foss et al (2012) discloses a comparison of extraction methods for paralytic shellfish toxins (PSTs) from the filamentous

cyanobacterium Lyngbya wollei. In the absence of commercially available standards for the unique toxins produced by this cyanobacterium it was not possibly to quantify the toxins extracted.

The publication of Li et al (2013) discloses a method for the rapid screening and identification of paralytic shellfish poisoning (PSP) toxins in red tide algae. The method utilises hydrophilic interaction chromatography-high resolution mass spectrometry (HILIC-HR-MS) combined with an accurate mass database. Limits of detection (LOD) of ten common PSP compounds were in the range of 10 to 80 nmol/L. The developed method was asserted to be a useful tool for the rapid screening and qualitative identification of common PSP toxins in harmful algae.

The publication of Bernardi Bif et al (2013) discloses the sensitivity of sea urchins to toxic cell extracts containing saxitoxins.

The publication of Poyer et al (2015) discloses the development of an analytical method to characterise and differentiate saxitoxin analogs, including sulfated ( gonyautoxins ) and non-sulfated analogs. Hydrophilic interaction liquid chromatography (HILIC) was used to separate sulfated analogs. Ion mobility mass spectrometry (IM-MS) was used as a new dimension of separation based on ion gas phase confirmation to differentiate the saxitoxin analogs. Positive and negative ionisation modes were used for gonyautoxins. Positive ionisation mode was used for non-sulfated analogs. The coupling of three complementary techniques, HILIC-IM-MS, permitted the separation and identification of saxitoxin analogs, isomer differentiation being achieved in the HILIC dimension with non-sulfated analogs separated in the IM-MS dimension.

The publication of Rubio et al (2015) discloses a method to purify saxitoxin using a liquid chromatography methodology based on ionic pairs. The saxitoxin is extracted using hydrochloric acid and treated with ammonium sulfate following a treatment with trichloroacetic acid and hexane/diethyl ether (97/3) . Samples were analysed by a semi-preparative HPLC in order to collect pure fractions of saxitoxin and these fractions were eluted in solid-phase cationic interchange STX extraction columns. The purified saxitoxin was reported to be stable and homogenous and its identity confirmed by LC-MS-MS. Analogs such as neosaxitoxin of a decarbamoyl saxitoxin were reported to be absent from the purified saxitoxin.

The publication of Chen et al (2016) discloses the application of serial coupling of reverse-phase liquid chromatography (RPLC) and hydrophilic interaction chromatography (HILIC) combined with high resolution mass spectrometry (HR-MS) to the simultaneous screening and identification of known lipophilic and hydrophilic toxins in the algae of harmful algal blooms (HABs) . Lipophilic and hydrophilic toxins were extracted simultaneously by the use of ultrasound-assisted extraction (UAE) . The publication demonstrated that HPLC/HILIC-HR-MS combined with an accurate mass list of known marine algal toxins may be used as a powerful tool for screening of different classes of known toxins in marine harmful algae.

The publication of Cho et al (2016) discloses the analysis of crude extracts of toxin-producing dinoflagellates by column switching and two-step gradient elusion using hydrophilic-interaction chromatograph (HILIC) combined with mass spectrometry. The publication states that the data obtained supports the hypothesis that the early stages of the saxitoxin biosynthesis and shunt pathways are the same in dinoflagellates and cyanobacteria.

The publication of Beach et al (2018) discloses the sensitive multiclass analysis of paralytic shellfish toxins, tetrodotoxins and domoic acid in seafood using a capillary electrophoresis (CE) -tandem mass spectrometry (MS/MS) method. A novel, highly acidic background electrolyte comprising 5 M formic acid was used to maximise protonation of analytes and is asserted to be generally applicable to simultaneous analysis of other classes of small, polar molecules with differing pKa values.

The publication of Kellmann and Neilan (2007) discloses the fermentative production of neosaxitoxin and its analogs in recombinant Escherichia coli strains .

The publications of Lagos Gonzales (2010, 2015a, 2015b and 2016) disclose the purification of phycotoxins from cyanobacteria produced in a continuous culture. The phycotoxins are isolated primarily from the bacteria, but can also be isolated from the culture medium. In one embodiment of the process disclosed only neosaxitoxin (neoSTX) and saxitoxin (STX)are produced. In another embodiment of the process disclosed only gonyautoxin 2 (GTX2) and gonyautoxin 3 (GTX3) are produced.

The publication of Wang et al (2010) discloses the preparation of a paralytic shellfish poison (PSP) standard solution. The standard solution is prepared by removing impurities from shellfish material, collecting shellfish meat, adding distilled water and 0.1-0.3 mol/L hydrochloric acid solution, regulating pH to 1.5 to 5.0, and homogenising to obtain homogenate,

precooling at -20°C for 30 minutes to 24 hours, and lyophilising to obtain a core sample, grinding, and sieving, precooling at -20°C for 10 minutes to six hours and lyophilising to obtain the standard sample. The method of

preparation is asserted to have the advantages of low raw material cost and a simple preparation process. The publication of Xiong and Qiu (2009) discloses the application of

biguanido purine derivatives and their salts and esters for improving the therapeutic effect and reducing the side effects of antitumor agents. The biguanido purine derivates are saxitoxin analogs.

It is an object of the invention to provide a semisynthetic method of producing neosaxitoxin in sufficient quantity and of sufficient purity to enable its use in the manufacture of pharmaceutical preparations. It is a further object of the invention to provide a method of producing the

intermediate for use in the semisynthetic method. These objects are to be read in the alternative with the object to at least provide a useful choice.

SUMMARY OF INVENTION

In a first aspect the invention provides a method of preparing a volume of concentrated aqueous extract for use in the preparation of a quantity of GTX1,4 comprising the steps:

1. Culturing a selected isolate of a dinoflagellate in a vertical column of aerated amended seawater for a period of time and at a temperature sufficient to provide a culture having a predetermined cell density;

2. Harvesting the cells from the culture having the predetermined cell

density to provide a quantity of cellular biomass;

3. Resuspending the quantity of cellular biomass in an aqueous solution of a weak organic acid for a period of time and at a temperature sufficient to provide a mixture of residual biomass and an extract in solution;

4. Separating the residual biomass from the extract in solution; and then

5. Reducing the volume of the extract in solution to provide the volume of concentrated aqueous extract, where the selected isolate has been selected to produce a ratio of GTX2,3 to GTX1,4 of less than 0.1, the amended seawater is seawater amended with a nutrient medium, and the predetermined cell density is in the range 7 x 10 ⁴ to 10 ⁵ cells/mL.

Preferably, the selected isolate is an isolate of the dinoflagellate

Alexandrium pacificum. More preferably, the isolate is an isolate of the dinoflagellate Alexandrium pacificum that produces a ratio of GTX2,3 to GTX1,4 of less than 0.01. Most preferably, the isolate is the isolate of the dinoflagellate Alexandrium pacificum designated CAWD234.

Preferably, the nutrient medium comprises nitrates, phosphates, trace metals and vitamins . Preferably, the aqueous solution of a weak organic acid is 0.25 to 0.75% acetic acid.

Preferably, the volume of concentrated aqueous extract a density between 1.06 to 1.14 g/mL .

In a second aspect the invention provides a method of fractionating a volume of concentrated aqueous extract to provide a solution of partially purified GXT1,4 comprising the steps:

1. Reducing the volume of the aqueous extract by ultrafiltration to provide a reduced volume;

2. Loading the reduced volume on a column of activated carbon sorbent to provide a loaded column; and

3. Eluting the loaded column with a stepwise gradient of water followed by aqueous acetic acid/acetonitrile to provide the solution of partially purified GTX1,4, where the volume of concentrated aqueous extract is an extract of a culture of a dinoflagellate .

Preferably, the volume of concentrated aqueous extract is an extract of a culture of a dinoflagellate prepared according to the method of the first aspect of the invention.

In a third aspect the invention provides a method of preparing a quantity of neoSTX comprising the steps of:

1. contacting in solution in a reaction solvent a quantity of purified

GTX1,4 and a quantity of dithiol for a period of time and at a

temperature sufficient to provide a reaction product in which greater than 97.5% (w/w) of the GTX1,4 has been converted to neoSTX; and then

2. applying the reaction product to a silica based weak cation exchange sorbent and eluting with an aqueous weak organic acid to separate the neoSTX from the dithiol and provide the quantity of neoSTX, where the pH of the solution is in the range 7.4 to 7.6.

Preferably, the reaction solvent is buffered aqueous acetic acid.

Preferably, the dithiol is selected from the group consisting of

dithiothreitol (DTT) and dithiobutylamine (DTBA) . More preferably, the dithiol is dithiothreitol (DTT) . Preferably, the quantity of purified GTX1,4 is obtained from a solution of purified GXT1,4 prepared according to the method of the second aspect of the invention .

Preferably, the quantity of neoSTX is greater than 100 mg with a purity greater than 99.5% (w/w) .

The methods of the invention provide for the batch preparation of GTX1,4 and neoSTX in quantities and of purities not previously obtainable {cf. Lagos Gonzales (2010, 2015a, 2015b and 2016)).

In the description and claims of this specification the following

abbreviations, acronyms, phrases and terms have the meaning provided: "batch preparation" means prepared discontinuously; "biosynthetic" means prepared within living organisms or cells; "CAS RN" means Chemical Abstracts Service (CAS, Columbus, Ohio) Registry Number; "comprising" means "including", "containing" or "characterized by" and does not exclude any additional element, ingredient or step; "consisting of" means excluding any element, ingredient or step not specified except for impurities and other incidentals; "consisting essentially of" means excluding any element, ingredient or step that is a material limitation; "GTX" means gonyautoxin; "GTX1" mean

gonyautoxin 1 [CAS RN 60748-39-2]; "GTX4" means gonyautoxin 4 [CAS RN 64296- 26-0]; "GTX1,4" means an unresolved mixture (as solid or in solution) comprising gonyautoxin 1 and gonyautoxin 4; "GTX2,3" means an unresolved mixture (as solid or in solution) comprising gonyautoxin 2 and gonyautoxin 3; "neoSTX" means (3aS, 4R, lOaS) -2-amino-4- [ [ ( aminocarbonyl ) oxy] methyl ] - 3a, 4 , 5 , 6, 8 , 9-hexahydro-5-hydroxy-6-imino-lH, 10H-pyrrolo [ 1 , 2-c] purine-10, 10- diol [CAS RN 64296-20-4]; "nutrient medium" means a medium comprising trace metals and vitamins; "preparative scale" means prepared in batches of greater than 100 mg; and "semi-synthetic" means prepared by chemical conversion of an at least partially purified biosynthetic precursor. A paronym of any of the defined terms has a corresponding meaning.

The terms "first", "second", "third", etc. used with reference to elements, features or integers of the matter defined in the Summary of Invention and Claims, or with reference to alternative embodiments of the invention, are not intended to imply an order of preference. Where concentrations or ratios of reagents are specified the concentration or ratio specified is the initial concentration or ratio of the reagents. Where values are expressed to one or more decimal places standard rounding applies. For example, 1.7 encompasses the range 1.650 recurring to 1.749 recurring. Purity of the isolated neoSTX is determined according to Method 3 [F. Analysis] .

The invention will now be described with reference to embodiments or examples and the figures of the accompanying drawings pages.

BRIEF DESCRIPTION OF FIGURES

Figure 1. A plan view of a hanging bag (1) formed from a length of tubular plastic for use in the bulk culture of isolates. Traversing single (3, 4, 5 and 6) and double solid lines (9 and 10) identify where two or more layers of the tubular plastic are heat welded together. Traversing broken lines identify where four (7 and 8) or two (12 and 13) layers of the tubular plastic are cut to provide a hanging loop (2) and cone, respectively.

Figure 2. The profiles of toxins produced by individual isolates (Table 1) when cultured in vertical columns of aerated amended sea water. The isolates designated as CAWD12, CAWD20 and CAWD121 are identified as non-producers of toxins under these culture conditions.

Figure 3. Amounts of toxins produced by individual isolates (Table 1) when cultured in vertical columns of aerated amended sea water.

Figure 4. Plot of concentration of gonyautoxins (GTX 1 (¨) and GTX 4 (■) ) versus time.

DESCRIPTION

According to the present invention a quantity of GTX1,4 is purified on a preparative scale, i.e. in batch quantities of greater than 100 mg, from a concentrated extract of a culture of a dinoflagellate and converted in a solution buffered to a pH around 7.5 to neoSTX by reductive desulfonation using a dithiol as the preferred reducing agent.

Extracts of cultures of isolates of Alexandrium pacificum that produce relatively low amounts of gonyautoxin 2 (GTX2) and gonyautoxin 3 (GTX3) have been determined to be most suitable as a source for the purification of GTX1,4 on a preparative scale. Extracts of cultures of dinoflagellates that may be used for the purification of GTX1,4 in accordance with the method described here are available from the Cawthron Institute, Nelson, New

Zealand .

The introduction of an ultrafiltration step prior to desalting has been found to be particularly advantageous when purifying GTX1,4 from these aqueous extracts on a preparative scale. Without wishing to be bound by theory it is believed that ultrafiltration removes a substantial portion of the solutes that might otherwise interfere with the desalting step on sorbents such as activated carbon. The introduction of the ultrafiltration step thereby reduces the quantity of sorbent that would otherwise be required. In solution, gonyautoxin 1 (GTX1) and gonyautoxin 4 (GTX4) exist as a pair of epimers of which GTX1 is the thermodynamically most favoured. Epimerisation is believed to occur under most conditions via keto-enol equilibration at C- 12. In the first step of the proposed 2-step reaction mechanism according to SCHEME I a thiol group of the dithiol (R-SH) attacks the electrophilic C-12 of the keto form (I) to form a thiohemiketal (II) . Conversion to a thioether (IV) occurs via an episulfonium ion intermediate (III) when the leaving group (O-sulfate) is oriented anti to the sulphur atom (as in the reactive epimer GTX1) . In the second step of the proposed reaction mechanism the thiol group of the dithiol reacts with the sulphur of the thioether (IV) to form a disulfide thereby yielding an enolate that readily hydrates to neoSTX (V) .

The optimal pH for the conversion of GTX1,4 to neoSTX has been determined to be around 7.5 and without wishing to be bound by theory is believed to ensure both (i) an optimal rate of epimerisation between the gonyautoxin epimers and (ii) optimal degrees of electrophilicity at C-12 and deprotonation of the dithiol used as the reducing agent. The use of dithiols such as

dithiothreitol (DTT) and dithiobutylamine (DTBA) are preferred over the use of monothiols such as glutathione (GSH) and mercaptoethanol (ME) ( cf.

Sakamoto et al (2000) and Sato et al (2000)) . Higher rates of conversion are obtained when using the dithiols, rendering them more suitable for use in the production of neoSTX on a preparative scale

The excess dithiol, sodium phosphate buffer and unreacted GTX1,4 has been found to be most conveniently removed from the neoSTX containing conversion product by the use of cation exchange chromatography. The silica based weak cation exchange sorbent Sepra™ WCX has been determined to be a suitable sorbent as it has been determined not to retain DTT . Trials of the polymeric based weak cation exchange sorbent Strata-X™ CW (Phenomenex) determined this sorbent to be unsuitable for purification of neoSTX from the conversion product on a preparative scale. The excess dithiol is retained by both an ion exchange and a reverse phase mechanism when using this sorbent. Although a portion of the excess DTT is eluted with organic solvents such as

acetonitrile/water a further portion is eluted with 1 M acetic acid

frustrating the purification of the neoSTX when using this sorbent.

o s

c oo o EXAMPLE

A. Materials

10 mM acetic acid (0.6 g/L in deionised water) ; amended seawater (7 mL/L LI nutrient medium in seawater) ; deionised water (Milli-Q™, Merck-Millipore ) ; LI nutrient medium (1.25 g/L EDTA, 0.91 g/L FeCl ₃ .6H ₂ 0, 0.29 mL/L trace metal stock solution, 21.6 g/L NaN0 ₃ , 1.44 g/L NaH ₂ P0 ₄ and 14.4 mL/L vitamin stock solution) ; Mobile Phase A (2.2 g/L sodium heptane sulphonic acid (sodium salt) and 0.31 g/L 85% phosphoric acid adjusted to pH 7.1 with 25% ammonium hydroxide) ; Mobile Phase B (0.1% acid in acetonitrile) ; Mobile Phase C (0.1% acetic acid in deionised water) ; roll of tubular (230 mm x 200 m x 250 pm) low density poly ( ethylene ) (LDPE) plastic (Amcor Limited); seawater (30 to 37 ppt salinity); trace metal stock solution (2.5 g/L CUSO ₄ .5H ₂ O, 20 g/L

Na ₂ Mo0 ₄ .2H ₂ 0, 23 g/L ZnS0 ₄ .7H ₂ 0, 11.9 g/L COC1 ₂ .6H ₂ 0, 178 g/L MnCl ₂ .4H ₂ 0, 1.3 g/L ¾Se0 ₃ , 2.6 g/L N1SO ₄ .6H ₂ O, 1.8 g/L Na ₃ V0 ₄ and 1.9 g/L K ₂ Cr0 ₄₎ ; vitamin stock solution (0.01 g/L biotin, 2 g/L thiamine and 0.01 g/L vitamin B12) .

Vitamin stock solutions are filter sterilised and aseptically added through a 0.22 pm syringe filter during preparation of amended seawater following the autoclaving of other ingredients.

B . Inocula

Isolates are obtained from naturally occurring algal blooms in coastal waters. Species responsible for harmful algal blooms include Alexandrium minutum, Alexandrium pacificum (formerly referred to as Alexandrium

catenella) , Alexandrium tamarense and Gymnodium catenatum.

The dinoflagellate Alexandrium pacificum is an armoured, marine, planktonic dinoflagellate . The species was originally identified as Alexandrium

catenella using the morphological characteristics described in the

publication of Balech (1985) . A more recent description of the

characteristics of species of the genus is provided in the publication of MacKenzie et al (2004) .

A chain-forming species, Alexandrium pacificum typically occurs in

characteristic short chains of 2, 4 or 8 cells. Single cells are round, slightly wider than long, and are anterio-posteriorly compressed. A small to medium sized species, it has a rounded apex and a slightly concave antapex.

The thecal plates are thin and sparsely porulated. Cells range in size between 20 to 48 pm in length and 18 to 32 pm in width.

Table 1. Isolates evaluated for their production of toxins and isolated from New Zealand coastal waters prior to October 2015.

As stated in the publication of MacKenzie (2014), Alexandrium pacificum is the cause of most paralytic shellfish toxin contamination in New Zealand. Individual isolates (Table 1), including isolates of the species Alexandrium pacificum, have been evaluated for their production of toxins when cultured in bulk according to the following protocols. Representations of the

structures of the toxins referred to in Figure 2 and Figure 3 are provided below. The representations are of the structures of the toxins in their neutral (uncharged) form.

A. Culture

Individual isolates are cultured in aerated vertical columns of amended seawater in a plurality of hanging bags. The bags are hung in an incubation room maintained at a temperature of 16 to 20 °C. The bags are illuminated according to a 16-hour/8-hour light/dark cycle. Automated carbon dioxide (CO2) dosing is used to maintain the pH in the range 8 to 9 during the light phase of the light/dark cycle. Standard personal protection equipment is worn by operators to minimise the risk of exposure to toxins.

Reference is made to Figure 1 of the accompanying drawings pages where a plan view of a cut and welded hanging bag (1) for use in the bulk culture of isolates is provided. The bags are formed from a roll of tubular plastic by cutting and heat welding according the following protocol.

After discarding the first two metres of a new roll of tubular plastic a 20 cm section is cut and the open ends of the excised section each sealed by a heat weld. The inner surfaces of the sealed section of tubular plastic roll are subjected to microbiological evaluation before the remainder of the new roll of tubular plastic is used for the fabrication of hanging bags.

Gonyautoxin 1 (GTX1) Gonyautoxin 2 (GTX2 ) Gonyautoxin 3 (GTX3)

Gonyautoxin 4 (GTX4) Gonyautoxin 5 (GTX5 (Bl) ) Gonyautoxin 6 (GTX6(B2))

A two-metre length (A+G) is dispensed from the tubular plastic roll and a hanging loop (2) formed at the first end. The hanging loop (2) is formed by folding back a 10 cm section (A) of the tubular plastic and heat welding together the four layers of plastic to provide a first heat weld (3) proximal to the cut first end. A second heat weld (4) is formed within the folded back section (A) parallel and spaced apart from the first heat weld (3) by about 2 cm followed by diagonal heat welds (5,6) traversing the region between the horizontal first and second heat welds (3,4) . The corners of the folded back section distal from the welds are then cut (7,8) to provide a hanging loop (2) of a width (B) capable of supporting the length of tubular plastic when filled with a volume of amended seawater.

The second end of the length of tubular plastic is sealed by double diagonal heat welds (9,10) converging to a point (11) proximal to the centre of the second end. The integrity of each double diagonal heat weld (9,10) is inspected visually before each triangle of plastic outside the conical sealed end is cut (12,13) away. The conical portion of the sealed end has a depth (C) of around 20 cm. Each hanging bag (1) is capable of containing a culture volume of approximately 24 L

Prior to filling with inoculum and amended seawater a hanging bag (1) is hung in position and the top outside corner surface sterilised by wiping with isopropanol. A downward pointing first hole is formed in the surface sterilised region using a sterile pick. The tip of a sterilised air vent inserted into the hole and taped to the outside of the bag using PVC insulation tape (50 mm width) . A region of the outer surface of the conical sealed end is also surface sterilised by wiping with isopropanol. A second hole is formed in the surface sterilised region using a sterile pick. The downward pointing tip of a sterile inoculation line is inserted into the hole and the line taped to the outside of the bag using PVC insulation tape (50 mm width) .

Inoculum is fed into the hanging bag from a parent culture via the

inoculation line. The line from the parent culture and the inoculation line to the hanging bag are each connected to a manifold. Air is purged from the lines by pumping amended seawater into both. The parent culture is then allowed to flow into the hanging bag. Pressurised air is introduced into the hanging bag containing the parent culture via its sterilised air vent. Equal volumes of the parent culture are transferred to multiple hanging bags. The inoculum containing hanging bags are then filled with the amended seawater to a fill line (D) . Once the hanging bags are filled, the inoculation lines are disconnected from the manifold and connected to an air line via a sterilised air filter. The culture volume is aerated via the air line. To monitor the pH of the medium during the culture the surface of a pH probe, including the glass bulb, is washed with deionised water and 70% (w/w) ethanol. A region of the outer surface of the hanging bag around 4 cm above the fill line is surface sterilised by wiping with isopropanol and a hole made in the bag within this region using a sterilised 10 mL pipette tip. The pH probe is inserted via the hole and into the culture volume and held flat against the inner side wall of the bag.

Under these incubation conditions the isolates have a doubling time of about two days and are harvested when the cell density has reached 7 x 10 ⁴ to 10 ⁵ cells/mL as determined by microscopic image cytometry.

B . Harvest

With aeration maintained the pH probe is removed and a volume of about 9 mL glacial acetic acid introduced into the culture volume. After about 10 to 20 minutes a volume of about 100 mL of a suspension of hydrated bentonite clay is added to the culture to provide a dosage of about 4 mL/L of culture. After a further 5 to 10 minutes the inoculation line is clamped and disconnected from the air filter. Settling of cells occurs over a period of time of at least 2 hours. The settled cells are drained into a centrifuge bottle and the collected volume (300 to 400 mL/bag) centrifuged in balanced bottles at a force of 1,500 x g for a period of time of 5 minutes. The supernatant is discarded, and the harvested cells weighed before storing frozen at -20°C.

C . Extraction

Extracts for (i) determining toxin profiles and monitoring toxin production, or (ii) preparation of GTX1,4 are prepared according to the following methods .

For monitoring toxin production, 10 mL of the culture volume is transferred to a polypropylene centrifuge tube and the cells pelleted by centrifugation at a force of 1,500 x g for a period of time of 5 minutes. The supernatant is discarded, and the pellet resuspended in a volume of 250 pL ImM acetic acid. The volume is sonicated for a period of time of 2 minutes and heated to 80 °C for a period of time of 10 minutes. The cellular material is pelleted by centrifugation at a force of 3,220 x g for a period of time of 5 minutes and 10-fold and 20-fold dilutions of the supernatant then prepared in 80% (v/v) acetonitrile/ 0.25% (v/v) acetic acid.

For preparation of GTX1,4, the frozen pellet of harvested cells is defrosted before resuspending in an equal volume of 0.5% acetic acid and leaving at room temperature for a period of time of 30 to 60 minutes. The suspension in acetic acid is then heated and maintained at a temperature of 85+2 °C in a water bath for a period of time of 10 to 15 minutes. The heated suspension is then cooled in an ice slurry before centrifugation at a force of 3,990 x g for a period of time of 2 minutes and the first supernatant decanted to a collection vessel. An equal weight of 0.5% acetic acid is added to the pelleted cellular material, mixed well, and centrifuged at a force of 3,990 x g for a period of time of 5 minutes. The second supernatant is decanted to the collection vessel and the total volume of collected supernatants reduced by rotary evaporation under vacuum (less than 15 mBar) at a temperature of 30 °C. (For example, portions of a total volume of 10 litres of supernatant are transferred from the collection vessel to a weighed 5 litre round bottom flask and the volume reduced by rotary evaporation until an extract having a weight of 800+50 g and a density between 1.08 to 1.12 g/mL is obtained.) The concentrated extract is stored frozen at -20°C.

D . Analysis

A sample of extract is prepared for analysis using activated carbon solid phase extraction (SPE) . A Supelclean™ ENVI-Carb™ SPE tube (bed wt . 250 mg, volume 6 mL) is conditioned with a volume of 3 mL of 20% acetonitrile/ 1% acetic acid followed by a volume of 3 mL of 0.025% ammonia. Following elution to the top frit the tube is loaded with a total volume of 400 pL consisting of 10 to 400 pL of extract and quantum sufficit deionised water. The cartridge is eluted to the top frit using a vacuum of -15 to -20 kPa and washed with a volume of 3 mL of deionised water before elution with a volume of 5 mL of 20% acetonitrile/ 1% acetic acid. The eluate is collected in a polypropylene tube and a volume of 10 pL diluted 4-fold with the addition of acetonitrile in a polypropylene autosampler vial. The contents of the vial are mixed on a vortex mixture with further dilution as required.

Method 1

Aqueous extracts and fractions enriched for the presence of GTX1,4 are analysed by high pressure liquid chromatography (HPLC) (Shimadzu Prominence) with ultraviolet (UV) detection according to this Method 1.

Samples are diluted with 10 mM acetic acid to provide a nominal concentration of 200 pg/mL. A volume of 40 pL of diluted sample is injected onto a column (4.6 x 150 mm) of 3.5 pm Zorbax Bonus RP eluted with Mobile Phase A at a flow rate of 1 mL/min for a period of time of 30 minutes while being maintained at a temperature of 20°C. The absorbance of the eluate is monitored at

wavelengths of 210 nm (purity) and 245 nm (quantity) using a photodiode array detector (PAD) . The purity of samples is calculated based on the percentage area at 210 nm and retention time with reference to a standard at a

concentration of approximately 200 pg/mL GTX1,4. Samples are analysed in triplicate with 10 mM acetic acid used as a blank and the chromatogram obtained subtracted from that obtained for all samples.

Method 2

Extracts and fractions enriched for the presence of GTX1,4 are analysed and quantified by liquid chromatography-mass spectrometry (LC-MS) (Shimadzu 8050) according to this Method 2.

Samples are diluted to an appropriate concentration using 80%

acetonitrile/ 0.25% acetic acid. A mixed standard containing a number of reference paralytic shellfish toxins (PSTs) is used. A maximum volume of 2 pL of diluted sample is injected by means of an auto sampler maintained at 4°C onto a column (2.1 x 100 mm) of 1.7 mM Waters Acquity UPLC BEH amide eluted at a flow rate of 0.4 mL/min while being maintained at a temperature of 20°C. The column is eluted stepwise with 75% Mobile Phase B/25% Mobile Phase C for a period of time of 5 minutes following injection, followed by 55% Mobile Phase B/45% Mobile Phase C for a period of time of 0.50 min before reverting to 75% Mobile Phase B/25% Mobile Phase C. The eluate is monitored by mass spectrometry using selective ion monitoring in ESI- and ESI+ ionisation modes .

Method 3

Purified products of extraction, fractionation and conversion are analysed for quantity and purity according to this Method 3.

Isolated products (GTX1,4 or neoSTX) are diluted to a concentration of 200 pg/mL in 10 mM acetic acid. The diluted sample is then further diluted 100- fold in 8% acetonitrile/ 0.25% acetic acid to provide a solution of product at a concentration of 20 mg/mL for quantitative analysis. A mixed standard containing a number of reference paralytic shellfish toxins (PSTs) is also prepared in the same solvent. A solution of 2 pL of the diluted product (20 ng/mL) is injected by means of an autosampler maintained at a temperature of 40°C onto a column (2.1 x 100 mm) of 1.7 pm Waters Acquity UPLC BEH amide eluted at a flow rate of 0.6 mL/min while being maintained at a temperature of 60°C. The column is eluted stepwise with 80% Mobile Phase B/20% Mobile Phase C for a period of time of 6 minutes following injection, followed by 55% Mobile Phase B/45% Mobile Phase C for a period of time of 0.50 minutes before reverting to 80% Mobile Phase B/20% Mobile Phase C. The eluate is monitored by mass spectrometry monitoring in ESI- and ESI+ ionisation modes.

E. Purification

The concentrated extract is thawed at room temperature. The extract is divided between two balanced 500 mL conical centrifuge bottles and centrifuged at a force of 4000 x g for a period of time of 10 minutes. The supernatant is decanted and filtered under reduced pressure through a series of 110 mm diameter hardened ashless and glass microfiber filter papers (Whatman™ grade 540, Whatman™ grade 542 and Whatman™ grade GF/A) .

The filtered extract is then subjected to crossflow ultrafiltration by recirculation through two filters (VivaFlow 200) connected in parallel with an outlet pressure no greater than 2.5 bar until the volume of the extract has been reduced to a volume of 10 to 20 mL of retentate.

The retentate is transferred to a 100 mL bottle and made up to a total volume of 100 mL with deionised water. The diluted retentate is similarly subjected to rounds of crossflow ultrafiltration reducing the volume of retentate to 10 to 20 mL before making up to a total volume of 100 mL with deionised water.

A 32 mm 5pm syringe filter (Pall Corp.) is installed on the inlet of a 50 g Sepabeads™ SP-270 (Supelco) SPE cartridge conditioned with 500 mL of deionised water at a flow rate of 30 mL/minute. The washed retentate is passed through the conditioned SPE cartridge at a flow rate of 30 mL/minute and the effluent collected. The SPE cartridge is then eluted with 200 mL of deionised water at a flowrate of 30 mL/minute and the effluent collected.

The volume of the combined effluents is reduced by rotary evaporation under vacuum (less than 15 mBar) at a temperature of 30±2°C. The volume is reduced to provide a weight of 500+50 g, the density determined gravimetrically, and the total reduced volume calculated on this basis.

Prior to purification of the GTX1,4 from the reduced volume on a preparative scale a sample of the reduced volume is prepared and analysed as described (F. Analysis) . A 10,000-fold dilution of the sample should provide a concentration of GTX1,4 in the range 20 to 50 ng/mL. The calculated yield of the GTX1,4 should exceed 90% (w/w) .

The total reduced volume is desalted by loading on a 100 g 25 x 450 mm carbon column ( Enviro-Clean™ Graphitized Carbon Non-Porous, UCT) conditioned with 1L of deionised water. The reduced volume is loaded at a flow rate of 30 mL/min using a Mini-Flash Pump (Sorbent Technologies, Inc.) and eluted at a flow rate of 15 mL/min with a stepwise gradient of deionised water for a period of time of 40 minutes followed by 0.2% (v/v) acetic acid/30% (v/v) acetonitrile for a period of time of 40 minutes.

The eluate is monitored at 205 nm using an inline UV detector and sequential volumes of 10 mL of eluate collected as fractions. Volumes of 5 pL of selected fractions are diluted to a total volume of 10 mL with 0.25% (v/v) acetic acid/80% (v/v) acetonitrile and submitted to LC-MS analysis as described (F. Analysis) . The GTX1,4 containing fractions are combined. The pH of the desalted combined fractions collected from the carbon column is adjusted to 7.8 using concentrated ammonium hydroxide at a rate of

approximately 2 pL/mL. The pH adjusted volume is then loaded on a 35 x 460 mm column of 280 g Sepra™ WCX-NH4 ⁺ conditioned using a volume of 1 L of 50 mM ammonium bicarbonate at a flow rate of 30 mL/min. The column is eluted with a gradient of 10 to 80% 0.5 M ammonium bicarbonate over a period of time of 100 minutes. The eluate is monitored at wavelengths of 205 nm and 245 nm using an inline UV detector. Sequential volumes of 25 ml are collected as fractions and combined according to the UV monitoring and LC-MS analysis where required (F. Analysis) . Fractions containing greater than 2% of the total amount of GTX1,4 are combined and the pH of the combined fractions reduced to 6.5 by the dropwise addition of glacial acetic acid. A 5,000-fold dilution of the acidified volume is subjected to LC-MS analysis as described (F. Analysis) and the total quantity and yield of GTX1,4 and the ratio to GTX5 calculated. The yield should be greater than 90% and the ratio of GTX5 to GTX 1,4 should be less than 1%.

The combined fractions from the weak cation exchange chromatography are loaded onto a 35 x 480 mm column of 175 g Sepra™ ZT-WCX-H ⁺ form conditioned using 1 L of deionised water at a flowrate of 30 mL/min. The loaded column is eluted with a continuous gradient of 0 to 100% 1M acetic acid in water over a period of time of 100 minutes and sequential volumes of 25 mL of eluate collected as fractions while monitoring at a wavelength of 205 nm using an inline UV detector.

The GTX1,4 containing fractions are collected from baseline to baseline and the total volume reduced to a volume of 10 to 20 mL by rotary evaporation at 30°C under reduced pressure of less than 15 mBar. A 500,000-fold dilution of the reduced volume is subjected to LC-MS analysis as described (F. Analysis) and the total quantity and yield of GTX1,4 calculated on this basis.

The GTX1,4 containing reduced volume is loaded on a 50 x 500 mm column of P2 gel conditioned with 2 L of 100 mM acetic acid at a flowrate of 50 mL/min and eluted isocratically with the same conditioning mobile phase. The eluate is monitored at a wavelength of 205 nm for a period of time of 200 minutes using an inline UV detector and sequential volumes of 10 mL of eluate collected as fractions. The GTX1,4 containing fractions are collected from baseline to baseline and the total volume reduced by rotary evaporation at a temperature of 30°C under reduced pressure of less than 15 mBar.

The reduced volume of combined fractions obtained by gel filtration is transferred to a pre-weighed 100 mL round bottom flask and evaporated to dryness by rotary evaporation at a temperature of 30°C under a reduced pressure of less than 15 mbar followed by drying in a freeze dryer at a shelf temperature of 10°C and pressure of 0.05 mbar for 24 hours. The open mouth of the round bottom flask is securely covered with an air permeable, lint free tissue before placing in the freeze dryer. The yield of purified GTX1,4 is determined gravimetrically . The purified GTX1,4 is dissolved in a known volume of deionised water to provide a solution containing 70 to 100 mg/mL of GTX 1,4.

A 500-fold dilution of the solution in 10 mM acetic acid is prepared and analysed as described (F. Analysis) . The final volume required to provide a concentration of 40 to 45 mg/mL of GTX 1,4 is calculated and the solution transferred via a filter to a pre-weighed 10 mL amber glass vial, rinsing with deionised water to provide a transferred volume having the target concentration of 40 to 45 mg/mL. A 200-fold dilution of the transferred solution is prepared in 10 mM acetic acid and analysed as described (F.

Analysis) . The dilution is diluted a further 10-fold in 10 mM acetic acid and analysed for purity as described (F. Analysis) .

F . Conversion

A quantity of 183 mg (as the free base) of GTX1,4 purified according to the preceding steps from an extract of a culture of the isolate of Alexandrium pacificum designated CAWD234 is dissolved in a total volume of 5 mL of dilute acetic acid and mixed with a volume of 45 mL of 0.2 M phosphate buffer at a pH of 7.5 in a 100 mL round bottom flask. The mixture is placed on ice and the pH adjusted from 6.8 to 7.5 with the addition whilst stirring of solid sodium carbonate. A quantity of 1.5 g of dithiothreitol (DTT) is added to the pH adjusted mixture and its dissolution promoted by placing the reaction mixture containing round bottom flask in an ultrasonic bath before

transferring to a water bath maintained at a temperature of 50°C. Aliquots of a volume of 10 pL are removed from the reaction mixture and transferred to the water bath (T=0) and periodically (every 15 minutes) thereafter. Aliquots are diluted 50-fold by the addition of a volume of 490 pL 80% acetonitrile 0.25% acetic acid immediately following removal from the reaction mixture and analysed by LC-MS as described (F. Analysis) to monitor the progress of the reaction in near real time (Figure 4) . After incubation for 45 minutes at

50°C the reaction mixture is chilled by transferring the round bottom flask to an ice slurry. Under these conditions near quantitative conversion of GTX1,4 to neoSTX is observed with close to 100% yield.

G. Isolation

The conversion product is loaded onto a quantity of 39 g Sepra™ WCX packed in an empty flash cartridge (Grace) and preconditioned with a volume of 250 mL of 50% (w/w) acetonitrile followed by a volume of 250 mL of deionised water. The conversion product is loaded onto the packed cartridge with rinses of deionised water with collection of the effluent (about 200 mL) . The

dissolution of any crystals formed during storage of the conversion product at 4°C is achieved by the addition of a minimal amount of deionised water.

The loaded packed cartridge is then eluted at a rate of 50 mL/min with a total volume of 1.5 L, followed by elution with a continuous gradient to 1 M acetic acid over 20 minutes and the collection of sequential volumes of 10 mL of eluate as fractions while monitoring UV absorbance at 205 nm and 254 nm.

A volume of 5 pL of fractions demonstrating UV absorbance at 205 nm is diluted 100,000-fold in 80% acetonitrile 0.25% acetic acid and analysed by LC-MS . Fractions confirmed to comprise neoSTX are combined, frozen at -70°C and lyophilised. The dried neoSTX is dissolved in a small volume of 10 mM and transferred to a pre-weighed 10 mL glass vial and a volume of 10 pL analysed. The purity and quantity of a batch (CNC00063) of neoSTX prepared according to the foregoing method is provided in Table 2.

Table 2. Specification for a batch (CNC00063) of neoSTX prepared according to the semisynthetic method described.

Although the invention has been described with reference to embodiments or examples it should be appreciated that variations and modifications may be made to these embodiments or examples without departing from the scope of the invention. Where known equivalents exist to specific elements, features or integers, such equivalents are incorporated as if specifically referred to in this specification. Variations and modifications to the embodiments or examples that include elements, features or integers disclosed in and selected from the referenced publications are within the scope of the invention unless specifically disclaimed. The advantages provided by the invention and discussed in the description may be provided in the alternative or in combination in these different embodiments of the invention.

REFERENCED PUBLICATIONS

Baker et al (2003) GTX4 imposters: characterization of fluorescent compounds synthesized by Pseudomonas stutzeri SF/PS and Pseudomonas/Alteromonas PTB-1, symbionts of saxitoxin-producing Alexandrium spp . Toxicon, 41(3), 339-347.

Balech (1995) The genus Alexandrium Halim (Dinoflagellata) Sherkin Island Marine Station, Sherkin Island, Ireland 151 pp . Beach et al (2018) Capillary electrophoresis-tandem mass spectrometry for multiclass analysis of polar marine toxins Analytical and Bioanalytical Chemistry, 410(22), 5405-5420.

Bernardi Bif et al (2013) Evaluation of mysids and sea urchins exposed to saxitoxins Environmental Toxicology and Pharmacology, 36(3), 819-825.

Chen et al (2016) Simultaneous screening for lipophilic and hydrophilic toxins in marine harmful algae using a serially coupled reversed-phase and hydrophilic interaction liquid chromatography separation system with high- resolution mass spectrometry Analytica Chimica Acta, 914, 117-126.

Cho et al (2016) Column switching combined with hydrophilic interaction chromatography-tandem mass spectrometry for the analysis of saxitoxin analogues, and their biosynthetic intermediates in dinoflagellates Journal of Chromatography A, 1474, 109-120.

Daigo et al (1985) Isolation and some properties of neosaxitoxin from a xanthid crab Zosimus aeneus Nippon Suisan Gakkaishi, 51(2), 309-13.

Foss et al (2012) Investigation of extraction and analysis techniques for Lyngbya wollei derived Paralytic Shellfish Toxins Toxicon, 60(6), 1148-1158.

Garcia-Altares (2017) Structural diversity of microalgal marine toxins

Comprehensive Analytical Chemistry, 78, 35-75.

Hall et al (1984) Dinoflagellate neurotoxins related to saxitoxin: structures of toxins C3 and C4, and confirmation of the structure of neosaxitoxin

Tetrahedron Letters, 25(33), 3537-8.

Jiang and Jiang (2008) Investigation of extraction method for paralytic shellfish poisoning toxins in shellfish Fenxi Huaxue, 36(11), 1460-1464.

Kellmann and Neilan (2007) Fermentative production of neosaxitoxin and its analogs in recombinant Escherichia coll strains International application no. PCT/EP2017/ 053077 [publ. no. WO 2017/137606 Al].

Lagos Gonzales (2015a) Methods for producing phycotoxins United States patent no. 8,957,207 B2.

Lagos Gonzales (2015b) Methods for purifying phycotoxins, pharmaceutical compositions containing purified phycotoxins and methods of use thereof United States patent no. 8,952,152 B2.

Lagos Gonzales (2016) Methods for purifying phycotoxins, pharmaceutical compositions containing purified phycotoxins and methods of use thereof United States patent no. 9,249,150 B2.

Lagos Gonzales (2010) Methods for purifying phycotoxins, pharmaceutical compositions containing purified phycotoxins, and methods of use thereof International application nos. PCT/IB2010/051187 [publ. no. WO 2010/109386 Al] and PCT/IB2010/051188 [publ. no. WO 2010/109387 Al] .

Laycock et al (1994) Isolation and purification procedures for the

preparation of paralytic shellfish poisoning toxin standards Natural Toxins, 2(4), 175-83.

Li et al (2013) Rapid screening, identification of paralytic shellfish poisoning toxins in red tide algae using hydrophilic interaction

chromatography-high resolution mass spectrometry with an accurate-mass database Fenxi Huaxue, 41(7), 979-985.

Liu et al (2010) Cultivation of Alexandrium catenella and extraction and detection of paralytic shellfish poisoning toxins Shuichan Xuebao, 34(11), 1783-1788.

MacKenzie (2014) The risk to New Zealand shellfish aquaculture from paralytic shellfish poisoning (PSP) toxins: A review New Zealand Journal of Marine and Freshwater Research, 48(3), 430-465.

MacKenzie et al (2004) The dinoflagellate genus Alexandrium (Halim) in New Zealand coastal waters: comparative morphology, toxicity and molecular genetics Harmful Algae, 3, 71-92.

Mezher (2018) FDA to replace analgesic drug development guidance with new documents Regulatory Focus News Article (www.raps.org/news-and-articles/news- articles/2018/8/ fda-2-replace-analgesic-drug-development-guidance ) .

Miao et al (2004) Isolation and purification of gonyautoxins from two strain of Alexandrium minutum Halim Journal of Chinese Pharmaceutical Sciences, 13(2), 103-105.

Parker et al (2002) Growth of the toxic dinoflagellate Alexandrium minutum (Dinophyceae) using high biomass culture systems Journal of Applied

Phycology, 14(5), 313-324.

Poyer et al (2015) Identification and separation of saxitoxins using hydrophilic interaction liquid chromatography coupled to traveling wave ion mobility-mass spectrometry Journal of Mass Spectrometry, 50(1), 175-181.

Ravn et al (1995) Standardized extraction method for paralytic shellfish toxins in phytoplankton Journal of Applied Phycology, 7(6), 589-94.

Rubio et al (2015) Purification and characterization of saxitoxin from

Mytilus chilensis of southern Chile Toxicon, 108, 147-153.

Rutman et al (2011) Treatment of loss of sense of touch with saxitoxin derivatives International application no. PCT/EP2011/051992 [Publ. no. WO 2011/098539 Al] .

Rutman et al (2013) Paralytic shellfish poison International application no. PCT/EP2013/ 055448 [Publ. no. WO 2013/135884 Al] . Sakamoto et al (2000) Formation of intermediate conjugates in the reductive transformation of gonyautoxins to saxitoxins by thiol compounds Fisheries Science, 66, 136-141.

Sato et al (2000) Identification of thioether intermediates in the reductive transformation of gonyautoxins into saxitoxins by thiols Bioorganic and

Medicinal Chemistry Letters 10, 1787-1789.

Siu et al (1997) Environmental and nutritional factors which regulate population dynamics and toxin production in the dinoflagellate Alexandrium catenella Hydrobiologia, 352, 117-140.

Tsai et al (1997) Toxicity and toxic components of two xanthid crabs,

Atergatis floridus and Demania reynaudi, of Taiwan Toxicon, 35(8), 1327-1335.

Wang et al (2010) Paralytic shellfish poison or PSP standard substances and preparations for monitoring sea food industries Chinese patent application no. 2010-10179534 [publ. no. CN 101858833 A].

Xiong and Qui (2009) Application of biguanido purine derivatives for improving therapeutic effect and reducing side effect of antitumor agents Chinese patent application no. 2008-10007215 [publ. no CN101513408] .