ANTI-VIRAL COMPOUND

Abstract

Phlorotannins with anti-viral properties, in particular those with a molecular mass of from about 1000 g/mol to about 3000 g/mol, are described. The phlorotannins may be obtained or obtainable as an extract from seaweed, and may be used in compositions or extracts as anti-viral agents. Methods for producing the extracts and their use for treating or preventing viral infections and for reducing or controlling a virus on a surface using the phlorotannins are also described.

Claims

1. A phlorotannin or a mixture of phlorotannins having a molecular mass of from about 1000 g/mol to about 3000 g/mol for use as an anti-viral agent.

2. An anti-viral agent comprising one or more phlorotannins according to claim 1.

3. (canceled)

4. A composition comprising one or more phlorotannins having a molecular mass of from about 1000 g/mol to about 3000 g/mol.

5. The composition according to claim 4, wherein the composition has anti-viral properties.

6. (canceled)

7. The phlorotannin or mixture of phlorotannins according to claim 1, wherein the phlorotannin or mixture of phlorotannins is obtained from brown seaweed.

8. The phlorotannin or mixture of phlorotannins according to claim 1, wherein the phlorotannin or mixture of phlorotannins is obtained from Ascophyllum nodosum.

9. The composition according to claim 5, wherein the virus is enveloped or non-enveloped.

10. The composition according to claim 5, wherein the virus is from one or more of group I, II, III or V of the Baltimore classification of viruses.

11. (canceled)

12. An extract obtained from brown seaweed, wherein the extract comprises one or more phlorotannins, and wherein at least about 70% by weight based on the dry weight of the phlorotannins in the dry extract have a molecular mass of from about 1000 g/mol to about 3000 g/mol based on the dry weight of the extract.

13. The extract according to claim 12, wherein the extract comprises from about 50% to about 100% phlorotannins by weight based on the dry weight of the extract.

14. The extract according to claim 12, wherein the extract is obtained from Ascophyllum nodosum.

15. A method of obtaining an extract according to claim 12, the method comprising using a hydro-alcoholic extraction solvent.

16. The method according to claim 15, wherein the method further comprises extraction with an organic solvent and/or purifying the extract.

17. (canceled)

18. (canceled)

19. An anti-viral composition comprising the extract according to claim 12.

20. A composition comprising the extract according to claim 12.

21. (canceled)

22. (canceled)

23. A method of treating or preventing a viral infection comprising administering the phlorotannin or mixture of phlorotannins according to claim 1 to a subject in need thereof.

24. A method of treating or preventing a viral infection comprising administering the composition according to claim 4 to a subject in need thereof.

25. A method of treating or preventing a viral infection comprising administering the anti-viral agent according to claim 2 to a subject in need thereof.

26. A method of reducing or controlling a virus on or at a surface comprising applying the composition according to claim 4 to the surface.

27. A method of reducing or controlling a virus on or at a surface comprising applying the anti-viral agent according to claim 2 to the surface.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0170] FIG. 1a shows the peak areas of phlorotannin oligomers having an odd number of repeating units in the extract of Example 3.

[0171] FIG. 1b shows the peak areas of phlorotannin oligomers having an even number of repeating units in the extract of Example 3.

[0172] FIG. 2 shows the size distribution of the phlorotannin oligomers in the extract of Example 3.

[0173] FIG. 3 shows the recovery of phenolics from LH20 fractionation in Example 5.

[0174] FIG. 4 shows the composition of the ORIGINAL, FUB and FUB fractions in Example 5.

[0175] FIG. 5 shows the possible structure of m/z 1117 component in Example 5.

[0176] FIG. 6 shows the characteristic phlorotannin m/z species are enriched in the BOUND fraction in Example 5.

[0177] FIG. 7 shows Fractionation of phlorotannins on Sephadex LH20 in Example 5.

[0178] FIG. 8 shows the MS spectra of phlorotannin peaks in LH20 fractions in Example 5.

[0179] FIG. 9 shows the Orbitrap MS evidence for doubly charged phlorotannin components in Example 5.

[0180] FIG. 10 shows the Positive mode LC-MS traces of original and SPE fractions in Example 5.

[0181] FIG. 11 shows the MS spectra of the 3 groups across their specific retention times in Example 5.

[0182] FIG. 12 shows the MS spectra of m/z 831 in the Bound Wash sample in Example 5.

[0183] FIG. 13 shows the HPLC-SEC RI detection of >1 kDa (left) and <1 kDa (right) TFF fractions of the extract of the invention in Example 6.

[0184] FIG. 14 shows the Left: HPLC-SEC RI detection of 80% MeOH insoluble fraction (higher MW material [ie earlier retention times] was observed). Right: HPLC-SEC RI detection of 80% soluble TFF fraction Example 6.

[0185] FIG. 15 shows the anti-viral activity of the Bound and Unbound fractions in Example 7.

[0186] FIG. 16 shows the anti-viral activity of the <1000 gmol−1; and (ii) >1000 gmol−1 fractions Example 7.

[0187] FIG. 17 shows a comparison of the anti-viral activity of seaweed extract and phlorotannin enriched fraction with Epigallocatechin, a polyphenol extracted from Green Tea in Example 8.

[0188] The invention is illustrated by the following non-limiting Examples.

Example 1—Small Scale Preparation of an Extract of the Invention

[0189] Fresh Ascophyllum nodosum (brown seaweed) which was sourced from Ireland having been harvested in January 2016 was used.

[0190] The water content of the fresh seaweed was determined as follows:

[0191] The moisture content of the seaweed was determined by drying a known weight of the seaweed in an oven to constant weight for approximately 24 hours at 103° C.

[0192] The seaweed was weighed, and the dry mass of the seaweed determined. The corresponding mass of water contained in the fresh seaweed was determined by subtracting the dry mass of the seaweed from the mass of the fresh seaweed, the percentage water is the fresh seaweed was then determined. For this seaweed, it was determined that the fresh seaweed was 70% water (ie 1000 g of seaweed contained 700 g of water).

[0193] An extract of the invention was prepared as follows:

[0194] 1000 g of the seaweed was used. This was pre-washed with seawater to remove any debris such as sand and stones. The seaweed was then washed with freshwater.

[0195] The seaweed was cut into pieces approximately 2-3 mm in diameter. This was done by first cutting with scissors and then using a Waring blender at low speed.

[0196] An Ethanol/Water blend was added to the seaweed. The ethanol used was 96% pure ethanol and 4% water. The amount of ethanol and water used in this ethanol/water blend satisfied the following two criteria: [0197] The ratio of the masses of ethanol in the blend to the total water present (i.e. water contained in the seaweed+the water in the blend+any additional water added in the blend) is 60:40 [0198] The total mass of the water and ethanol was three times the mass of the fresh seaweed to which it is added

[0199] The mass of water added as part of the ethanol/water blend to meet the above criteria can be determined through simple mathematics.

[0200] In this example, 1000 g of fresh seaweed is used and 700 g of this was determined to be water and 300 g the dry seaweed.

[0201] The total mass of the Ethanol and Water is 3000 g (i.e. 3×1000 g). And this is in an Ethanol: water ratio of 60:40. So 1875 g of Ethanol (considering the ethanol used is 96% ethanol and 4% water), and 1125 g of water. 700 g of this water is already contained in the fresh seaweed, so an additional 425 g of water is to be added as the Ethanol/water blend.

[0202] The Ethanol/Water blend was added to the washed, fresh seaweed in a large opaque beaker.

[0203] The mixture was agitated using an overhead stirrer for approximately 5 hours at room temperature.

[0204] After this time the sample was left to settle for approximately 30 minutes.

[0205] The mixture was then sieved using a Buchner filter.

[0206] The alcohol was then evaporated at low temperature (34-45° C.) using a rotary evaporator.

[0207] The extract was then frozen at −36° C. in shallow trays to maximize the surface area.

[0208] The extract was then freeze dried with Condenser temperature of −87° C. and Pressure 200-400 mTorr for approximately 3 days.

[0209] The extract was then ground to create a powder.

[0210] The powder extract was then frozen and protected from moisture, heat and light to avoid deterioration.

[0211] It was thawed as needed prior to testing.

Example 2—Antiviral Testing

[0212] Seaweed extracts obtained in Example 1 were used.

[0213] The thawed seaweed extract was dissolved in demineralised water to the required concentration for testing.

[0214] The extent to which a compound or composition renders the virus non-viable or inhibits the replication of the virus was measured using the test method BS EN 14476:2013+A1:2015 Chemical disinfectants and antiseptics—Quantitative suspension test for the evaluation of viricidal activity in the medical area. Test method and requirements (Phase 2/Step 1) have been defined as a standard European Test Method within the Biocidal Products Regulation (BPR). Full details of this method can be found in the regulation.

[0215] Using this method, the log reduction value for the particular virus the extract solution was being tested against was determined.

[0216] The contact time is the time that the test solution was in contact with the test virus during the test. The results are shown in Table 1 below.

TABLE-US-00001 TABLE 1 Example Concentration of (CC3687; Extract tested in Contact BYO2) water Virus Time Log.sub.10 R* 2-A 2% Norovirus 60 min ≥4.38 2-B 2% H1N1 60 min ≥2.63 2-C 2% Rotavirus 60 min ≥3.75 2-D 2% H1N1 60 min ≥4.00 2-E 2% H1N1 2 min ≥4.00 2-F 2% Norovirus 60 min ≥4.88 2-G 2% Norovirus 2 min ≥4.63 2-H 2% Rotavirus 60 min ≥5.44 2-I 2% Rotavirus 2 min ≥5.44 2-J 2% H1N1 60 min ≥4.00 2-K 2% H1N1 2 min ≥4.00 2-L 2% Poliovirus 60 min 4.67 2-M 2% Poliovirus 2 min 4.50 2-N 2% Adenovirus 60 min 4.83 2-O 2% Adenovirus 2 min 4.66 2-P 2% H1N1 60 min ≥5.00 2-Q 2% H1N1 2 min ≥5.00 2-R 2% Norovirus 60 min ≥4.75 2-S 2% Norovirus 2 min ≥4.50 2-T 2% H1N1 60 min ≥5.00 2-U 2% H1N1 2 min ≥5.00 2-V 2% Rotavirus 60 min ≥5.44 2-W 2% Rotavirus 2 min ≥5.44 2-X 2% Norovirus 60 min ≥4.85 2-Y 2% Norovirus 2 min ≥4.60 *where the Log.sub.10R value shows a ≥ sign it implies that complete elimination of the virus has occurred. H1N1 indicates influenza A virus subtype H1N1.

Example 3—Larger Scale Preparation of an Extract of the Invention

[0217] Fresh Ascophyllum nodosum (brown seaweed) which was sourced from Brittany in France having been harvested in November 2017.

[0218] The water content of the fresh seaweed was determined as follows:

[0219] The moisture content of the seaweed was determined by drying a known weight of the seaweed in an oven to constant weight for approximately 24 hours at 103° C.

[0220] The seaweed was weighed, and the dry mass of the seaweed determined. The corresponding mass of water contained in the fresh seaweed was determined by subtracting the dry mass of the seaweed from the mass of the fresh seaweed, the percentage water is the fresh seaweed was then determined.

[0221] 69 Kg of fresh Ascophyllum nodosum (brown seaweed) was used. The harvested seaweed was soaked overnight in seawater. The next day it was hand-sorted to remove epiphytic seaweed attached to the Ascophyllum biomass and drained for 30 minutes. It was then rinsed with freshwater for 3 minutes and drained again.

[0222] The seaweed was then milled in an Urschel grinder in two steps (Coarse grinding with 66708 grid, followed by fine grinding with 66896 grid). This produced 2-3 mm particles.

[0223] An Ethanol/Water blend was added to the seaweed. The ethanol used was 96% pure ethanol and 4% water. The amount of ethanol and water used in this Ethanol/water blend satisfied the following two criteria: [0224] The ratio of the masses of ethanol in the blend to the total water present (i.e. water contained in the seaweed+the water in the blend+any additional water added in the blend) is 60:40 The total mass of the water and ethanol was three times the mass of the fresh seaweed to which it is added

[0225] The mass of water added to the Ethanol/water blend to meet the above criteria is determined through simple mathematics.

[0226] In this example, 69 Kg of fresh seaweed is used and 52 Kg of this was determined to be water and 17 Kg the dry seaweed.

[0227] The total mass of the Ethanol and Water is 207 Kg (i.e. 3×69 Kg). And this is in an Ethanol: water ratio of 60:40. So 129.4 Kg of Ethanol (considering the ethanol used is 96% ethanol and 4% water), and 77.6 Kg of water. 52 Kg of this water is already contained in the fresh seaweed, so an additional 25.6 Kg of water is to be added as the Ethanol/water blend.

[0228] The extraction was performed in an agitated milk tank, for 17 hours at room temperature and pressure.

[0229] The spent biomass was separated from the extract on a 300 μm sieve (SWECO).

[0230] The alcohol was evaporated at low temperature in a Brouillon Process evaporator (initial temperature 34° C., final temperature 45° C.), taking approximately 8 hours.

[0231] The extract was further clarified using filtration aid FW12 (diatomaceous earth) and KDS12 filters (cellulose-based depth filter sheets) mounted on a Millipore frame.

[0232] The resulting filtered extract was transferred to polypropylene trays and freeze dried using a slow-cycle (25-30 Kg.prg) to avoid inhomogeneous freezing. This took approximately 3 days.

[0233] The extract was then grinded using a lab scale blender. 2.92 Kg of powdered extract resulted.

[0234] The powder extract was then frozen and protected from moisture, heat and light to avoid deterioration.

[0235] It was thawed as needed prior to testing.

[0236] The composition of the extract obtained was analysed.

[0237] A small sample of the extract was added to Folin-Ciocalteu's phenol reagent and the blue colour that developed after addition of sodium carbonate solution was assessed against a standard curve of gallic acid, as a standard phenolic compound.

[0238] This positive reaction with Folin-Ciocalteu's phenol reagent indicated the present of phlorotannins.

[0239] The nature of the phlorotannins was then investigated using liquid chromatography mass spectrometry techniques using an LC system with an LTQ Orbitrap XL mass spectrometer (Thermo Fisher Scientific). A neat sample of the extract was separated on a reversed phase C18 column and analysed in negative ESI mode (the method used is described in Austin et al. (2017) Extracts from the edible seaweed, Ascophyllum nodosum, inhibit lipase activity in vitro: Contributions of phenolic and polysaccharide components. Food & Function, 9, 502-510—DOI: 10.1039/C7FO01690E)

[0240] A series of phlorotannin oligomers were detected with apparent degree of polymerisation (DP) of phloroglucinol units from 10 to 23. The detection of phlorotannins with masses indicative of oligomers of odd numbered DP is shown in FIG. 1a (e.g. 11-23 DP) and even numbered DPs in FIG. 1b. Generally, more than one isomer was present for each oligomer.

[0241] The relative amounts of the PT oligomers were assessed by their MS peak areas (FIG. 2). There was a size distribution from DP 10 to 23 with the most abundant forms from DP11 to 18 and a sharp drop off in abundance from DP19 onwards.

Example 4—Antiviral Testing

[0242] Seaweed extracts obtained in Example 3 were used.

[0243] The thawed seaweed extract was dissolved in demineralised water to the required concentration for testing. The method for anti-viral testing described was used to determine the log reduction value for the particular virus the extract solution was being tested against. The contact time is the time that the test solution was in contact with the test virus during the test. The results are shown in Table 2 below.

TABLE-US-00002 TABLE 2 Concentration of Extract tested in Contact Example water Virus Time Log.sub.10 R 4-A 2% Norovirus 60 min ≥4.47 4-B 2% Norovirus 2 min ≥4.35 4-C 2% Norovirus 60 min ≥4.47 4-D 2% Norovirus 2 min ≥4.35 4-E 2% Norovirus 2 min ≥4.63 4-F 1% Norovirus 2 min 3.13 4-G 2% Norovirus 0.5 min ≥4.50 4-H 2% Poliovirus 60 min ≥5.00 4-I 2% Poliovirus 2 min ≥5.50 4-J 2% Adenovirus 60 min ≥4.25 4-K 2% Adenovirus 2 min ≥4.25 *where the Log.sub.10R value shows a ≥ sign it implies that complete elimination of the virus has occurred.

Example 5—Purification of Crude (Hydroethanolic Extract) Using Solid Phase Extraction (SPE) to Separate into Two Fractions (i) More Hydrophilic Fraction: (ii) More Hydrophobic Fraction

[0244] The extract (as obtained from Example 3) was freely soluble in ultra-pure water (UPW) containing 0.1% formic acid (FA) at 5% (w/w) and a 250 mL solution was prepared for use in the first test run (SPE-1). The SPE UNIT (Strata C18-E GIGA tube, 50 g capacity & 150 mL volume; Phenomenex Ltd) tube was prepared with 2 volumes of acetonitrile (ACN) containing 0.1% FA then equilibrated with 3 volumes of UPW containing 0.1% FA (Nwosu et al., 2011). The extract was applied in 60 mL batches and elution expedited using a side-arm flask and vacuum line.

[0245] After all of the extract sample solution was applied, the UNBOUND (hydrophilic) fraction was removed and the SPE unit washed with 2× volumes of UPW+FA, collected as the WASH fraction. Then the BOUND (hydrophobic) fraction was obtained by eluting the unit with 2 volumes of acetonitrile+FA. The unit could then be re-equilibrated for further use by washing with excess UPW+FA and stored in UPW+FA in the fridge until required.

[0246] For a second SPE (SPE-2) run, 400 mL of 5% (w/v) extract was prepared. The SPE procedure was the same apart from two changes. As the capacity of the SPE unit would be exceeded using this amount of material, the SPE procedure was repeated to re-capture any phenolic material that had passed into the UNBOUND fraction. Therefore, the BOUND fraction was composed of two combined fractions eluted from the unit. Also, the material that was eluted from the unit after removal of the bound fraction during the re-washing of the SPE unit was collected.

[0247] TOTAL CHOn and Phenolic Contents of Fractions

[0248] The total carbohydrate (CHOn) content of the BOUND and UNBOUND fractions were measured using the phenol-sulphuric acid method and the total phenol content (TPC) using the Folin-Ciocalteu method (Austin et al., 2017).

[0249] In SPE-1, the overall recovery of CHOn was 46.3% with 44.3% in the UNBOUND fraction and 2.0% in the BOUND fraction. The overall recovery of Total Phenol Content (TPC) was higher at 70.3% with the majority recovered in the BOUND fraction (56.8%) and less in the UNBOUND fraction (13.7%).

[0250] IN SPE-2, the overall recovery of CHOn was higher at 59.1% with 54.6% in the UNBOUND fraction and 4.5% in the BOUND fraction. However, the BOUND-WASH recovery fraction contained significant amounts of CHOn (i.e. 10.7% total) so the overall recovery of CHOn reached to 69.8%. That this “back-wash” fraction accounted for a fair amount of the “lost” CHOn in SPE-1 was unexpected given the chemistry of the SPE units.

[0251] For SPE-2, the TPC recovery figure reached 82.1% with the majority (70.0%) in the BOUND fraction and 12.1% in the UNBOUND fraction. The Bound Wash fraction accounted for a further 3.4% so the overall recovery of TPC reached 85.5%. It must be noted that neither of the two quantification methods is absolutely specific for CHOn or phenolics and therefore other components may interfere in their quantification.

[0252] Overall, the UNBOUND fraction was CHOn-rich and the TPC was poor whereas the BOUND fraction contained the majority of the TPC.

[0253] Expressed as a ratio of CHOn/TPC, the original extract had a value of −3.71 whereas the UNBOUND 1 ratio was 11.86 and the UNBOUND 2 was 17.24, which illustrates considerable enrichment in CHOn. The BOUND fraction 1 had a ratio of 0.136 and the BOUND fraction 2 had 0.255 which shows the enrichment of phenolics and removal of CHOn. The Bound Wash fraction was 12.04 also showing enrichment in CHOn.

[0254] Fractionation by Sephadex LH-20

[0255] A portion of the extract was fractionated using Sephadex LH20 using a technique (Pantidos et al., 2014) well-known to select for phlorotannin-like components (https://users.miamioh.edu/hagagermae/). A 25 mg/ml solution of the extract in UPW was produced and then 5 mL was added to 5 mL ethanol and mixed well. The extract was soluble in 50% ethanol and, in other tests, was fully soluble at up to 80% ethanol.

[0256] The extract solution was added to 5 mL of a slurry of Sephadex LH20 in 50% ethanol and mixed well for 10 mins at room temperature. After centrifugation at 2500 g for 5 mins at 5° C., the unbound fraction was removed and 5 mL of 50% ethanol added. The centrifugation procedure was repeated to give the WASH fraction then similarly with 50% acetone and then two washes with 80% acetone. The total CHOn and phenol contents were measured as before and 2×1 ml aliquots of each fraction were dried in a Speed-Vac for LC-MS analysis.

[0257] As shown in FIG. 3, the majority of the phenolic material was recovered in the first 80% acetone fraction with over 80% in the fractions released by acetone, which should be enriched in phlorotannin components.

[0258] Liquid Chromatography Mass Spectrometric (LC-MS) Analysis: Evidence for phlorotannins

[0259] Samples of the BOUND and UNBOUD fractions were analysed using a previously described method (Pantidos et al., 2014; Nwosu et al., 2011) using electrospray ionisation MS in both positive and negative mode to examine the composition of the fractions. The components were separated on a C18 reverse phase column and the elution of phenolic material monitored at 280 nm. Initial LC-MS data was obtained using a Fleet MS then selected samples were re-analysed using an Orbitrap MS which gives higher sensitivity combined with facilities to give exact mass data and discrimination of multiply charged ions.

[0260] As shown in FIG. 4, the BOUND sample was enriched in later eluting UV-absorbing peaks whereas the UNBOUND fraction was essentially devoid of these peaks.

[0261] In the BOUND fraction, the set of peaks from 12-21 mins gave a set of m/z signals in negative mode that are characteristic of phlorotannins. They present as two series of m/z values that differ by 124 amu, which is the extension unit mass equivalent to phloroglucinol minus 2 H atoms, which have been noted previously in our research (Nwosu et al., 2010; Pantidos et al., 2014; Austin et al., 2018). The series from m/z 745, 869, 993, 1117, 1241, 1365 and 1489 could arise from successive phloroglucinol additions to a trimer with m/z value of 373 (e.g. triphlorethol) and signals at m/z 621, 745 etc could be pentaphloroethol, hexaphloroethol etc (Martinez et al., 2013). Therefore, the major peak at m/z 1117 could be a nonaphloroethol (see FIG. 5). It is notable that the tannin MS signals are approx. 4 fold enriched in the BOUND fraction (FIG. 6).

[0262] These series are apparent if MS spectra are averaged across the separation zone between 12 and 21 mins. Not all the phlorotannin components could be readily separated by elution on C18 columns and most elute as mixed peaks. There were also multiple isomers possible for each phlorotannin species. However certain peaks appeared to represent discrete phlorotannin components. In particular, the major peak at 17.4 mins yielded a discrete m/z signal at 1117, the peak at 16.82 yields m/z 1055, 17.62 yields m/z 1179 and the peak at 16.22 yields m/z 933. However, further work is required to define the structures of these components.

[0263] The late-eluting phlorotannin peaks were only present in the bound samples released from Sephadex LH20 by acetone. The fractionation could be improved by more stringent washing with 50% ethanol then moving directly to elution with 80% acetone but the results support the expected fractionation of phlorotannins by acetone (Pantidos et al., 2014), see FIG. 7.

[0264] Therefore, we can be confident that the BOUND fraction is enriched in phlorotannins of a similar kind as noted previously and with apparent masses in the DP 4-13 range, with a molecular mass of from about 1000 g/mol to about 3000 g/mol.

[0265] Further studies using Orbitrap MS provided further information.

[0266] On the Orbitrap, the F-BOUND sample showed the same late eluting UV peaks as before and gave similar MS spectra across the separation zone of 12-21 mins (FIG. 9).

[0267] As before, the major UV peak gave a MS signal at m/z 1117 but on closer examination this signal had 0.5 amu variants indicative of a [M−H].sup.2− ion. As a doubly charged ion, the true mass would be twice as much and this suggests that this component was an oligomer of 18 phloroglucinol units rather than the nonaphloroethol described before. In fact all of the m/z values in the series from m/z 621 upwards were doubly charged which suggests that they are all larger than suggested by the original MS data. This range of DP fits with previous reports of phlorotannin structures from Ascophyllum nodosum (Steevensz et al., 2012).

[0268] LC-MS Evidence for Nature of Carbohydrate Components

[0269] In positive mode, there are three sets of ionizing species in the Original sample (FIG. 10). There are the phlorotannin components from RT 12-21, earlier eluting components from RT 2-3.7 mins and another set of components that elute between 4-10 mins. The phlorotannins accumulate in the BOUND sample but the early eluting components are enriched in the UNBOUND whereas the other components are present in the BOUND fraction but enriched in the BOUND WASH fraction (see FIG. 10).

[0270] The phlorotannins give characteristic MS spectra (see trace 3, FIG. 11) and, as seen earlier, were enriched in the Bound fraction. The early eluting peaks also give specific MS signals (see trace 1, FIG. 11) but these are not similar to any reported seaweed components. Initial work suggests sulphated components may be present as these signals show neutral losses of 80 amu but identification of their nature will require further work.

[0271] The mid-eluting components, which are also enriched in the Bound Wash fraction, give an interesting series of MS signals (see trace 2, FIG. 9). Given that the Bound Wash fraction was enriched in CHOn compared to the original, it is possible that these components are carbohydrate in nature. There is a series starting at m/z [M+H].sup.+=831 that differ by 162 amu.

[0272] A similar pattern can also be seen in negative mode but positive mode gives more intense signals. Addition of 162 is characteristic of addition of hexose sugar groups (e.g. glucose) to an existing structure.

[0273] The signal at m/z 831 fits with the molecular weight of laminaripentaitol (available from https://secure.megazyme.com/1˜3-Beta-D-Laminaripentaitol) and signals at m/z 993, 1155, 1317 and 1479 could be due to the addition of hexoses to the core structure of this laminarin oligosaccharide with a terminal mannitol group. Mannitol terminated laminarins (M-laminarins) are known to occur in Ascophyllum (Kadam et al., 2014) and these m/z signals suggest that such oligosaccharides are present in the crude extract material. These m/z signals gave very poor fragmentation, so the runs were repeated on the Obritrap MS. The same series of signals were apparent in the Orbitrap MS data (FIG. 12).

[0274] The putative penta-oligosaccharide gave MS.sup.2 fragments characterised by losses of 162 (loss of hexosyl group=glucose) or 182 (loss of mannitol). The signals at m/z 325 and 345 respectively could be protonated forms of a beta-linked glucose dimer (MW 324) and glucosyl-mannitol (MW 344). In addition, the m/z signal at 831 had an exact mass of 831.29499 which yielded a predicted formula of C.sub.30H.sub.55O.sub.26 (error of <2 ppm) which fits with the formula for laminaripentaitol.

[0275] Similar MS spectra and fragmentation data and exact mass formulae could be obtained for other members of the series at m/z 669 (DP4), 933 (DP6) and 1155 (DP7).

[0276] In summary, the above results confirm that the crude extract is enriched with phlorotannins having a degree of polymerisation of from about 10 to about 30, corresponding to a molecular weight from about 1000 g/mol to about 3000 g/mol.

[0277] The above results also confirm that the crude extract can be purified into phlorotannin enriched and polysaccharide enriched fractions using Solid Phase Extraction SPE (Chromatographic method), with the phlorotannins present in the bound fraction.

Example 6—Separation of Crude (Hydroethanolic Extract) Using Tangential Flow Filtration TFF to Separate into Two Fractions (i) <1000 gmol−1; and (ii) >1000 gmol−1

[0278] a) Initial Handling/Observations

[0279] The extract was stored at 2-8° C., covered in foil, and the work was carried out protecting the samples from light as far as possible.

[0280] The solubility of the extract in water was investigated to inform subsequent mass balances. The extract was made up at 2% in water, thoroughly mixed, centrifuged, and the insoluble and soluble fractions were freeze-dried.

[0281] b) Tangential Flow Filtration (TFF) Experiments to Remove Low MW Material/Size-Based Separation

[0282] An initial scoping run was conducted using an existing 1 kDa MW cut off hollow fibre cartridge on a Kros Flow Iii system (Spectrum), with subsequent runs carried out using a new cartridge with 2× surface area.

[0283] The extract samples were prepared by resuspension in water at 0.5%, followed by filtration down to GFC (1.2 um) (apart from first scoping experiment which was 0.4%).

[0284] The TFF concentration and diafiltration steps were standardised as far as possible between runs, with separation monitored by permeate conductivity.

[0285] All of the extract samples were freeze-dried and analysed by Folin-Ciocalteau assay and HPLC-size exclusion chromatography (SEC) with RI and PDA detectors.

[0286] Following the analysis and evaluation of samples, an additional fractionation run was also carried out on >1 kDa material using Vivaspin centrifugation units (Sartorius) at 30 kDa or 10 kDa MW cut off, to evaluate if further separation could be achieved using this approach.

[0287] An additional fractionation run was also carried out using a 5 kDa and 100 kDa MW cut off T-series membrane on a PALL centramate TFF system, to evaluate if this could improve separation. This has a larger surface area, allowing for higher loading of samples, although hold up volume is also larger.

[0288] c) Extraction Approaches Based on Solubility/Charge

[0289] All initial extraction experiments were carried out on >1 kDa material. Ethyl acetate was used to extract >1 kDa samples and generate an ethyl acetate soluble and insoluble fraction. Ethyl acetate was removed from samples by rotary evaporation and samples resuspended and freeze-dried. This was repeated on several batches of samples and the fractions were analysed by Folin-Ciocalteau assay and HPLC-SEC.

[0290] An alcohol extraction was carried out using 80% methanol on >1 kDa material. The 80% insoluble pellet was air-dried, resuspended and freeze-dried. Methanol was removed from the soluble fraction by rotary evaporation, and the resulting material freeze-dried. Fractions were analysed by Folin-Ciocalteau assay and HPLC-SEC.

[0291] Ion exchange chromatography was carried out on an ethyl acetate insoluble fraction to scope this method. The sample was loaded onto a Q-sepharose column in a Tris-NaCl buffer and eluted using a NaCl gradient. The two largest peaks were collected, dialysed and analysed.

[0292] d) Development of a Folin-Ciocalteau Assay to Estimate Polyphenol Content

[0293] A Folin-Ciocalteau assay was developed based on methods described in the literature and published theses. Phloroglucinol was used to generate a standard curve, and method optimised for timing and incubation temperatures. Results were generated in phloroglucinol equivalents. A PVPP blanking step was also introduced to allow estimation of how much of the signal can be attributed to phlorotannins, and how much may be other polyphenols or non-specific assay interference.

[0294] e) HPLC Size Exclusion Chromatography, HPLC C18 Analysis & GC-FID Monosaccharide Analysis

[0295] GlycoMar's existing HPLC-SEC method was adapted to evaluate Byotrol sample fractions. A Shodex SB806M column calibrated with dextran standards (1-1400 kDa), with RI and PDA detection and aqueous mobile phase (Tris-HCl/NaCl) was used, with sample prepared in mobile phase.

[0296] Development of an HPLC method using a Kinetex XB C18 column (Phenomenex) has been explored to improve polyphenol evaluation. The mobile phase is 2% acetic acid 0-100% methanol gradient with samples prepared in methanol. Phloroglucinol was used as a control.

[0297] Monosaccharide analysis of selected samples was carried out using an in-house methanolysis and TMS derivatisation approach, followed by GC-FID analysis.

[0298] Sulphate analysis was carried out using an in-house modified Terho method based on barium chloride.

[0299] Results Overview

[0300] a) Initial Observations

[0301] Dry weight yields indicated only a minimal amount of the original sample (˜0.4% w/w) was water insoluble debris, such as cellular material.

[0302] b) Tangential Flow Filtration Experiments to Remove Low MW Material.

TABLE-US-00003 TABLE 3 Scoping run using existing 1 kDa MW cut off hollow fibre unit, 2 g sample loading: TFF run Fraction Yield % (w/w) Folin PG.sup.1 equivalents μg/ml 23 Apr. 2018* >1 kDa 9.4 60.4 <1 kDa 57.8 2.6 *diafiltration permeate yield not quantified for this run .sup.1phloroglucinol equivalents

[0303] There was lower than expected recovery of >1 kDa material from this unit (based on previous CEVA data) and evidence of polyphenols sticking to the fibres. All subsequent 1 kDa runs were carried out using a new hollow fibre unit with higher surface area (Table 4 below).

TABLE-US-00004 TABLE 4 Subsequent runs using a new 1 kDa MW cut off hollow fibre unit with 2X surface area, 5 g sample loading: TFF Run Fraction Yield % (w/w) Folin PG equivalents μg/ml 7 May 2018 >1 kda 38.5 20.0 <1 kDa 46.5 2.6 15 May 2018* >1 kda 31.4 21.8 <1 kda 42.5 2.6 28 May 2018 >1 kda 21.1 34.9 <1 kda 43.2 2.6 *> and < 1 kDa fractions from this run sent to Byotrol. (Additional run from 12 Jun. 2018 still to be analysed)

[0304] The >1 kDa fraction % yield on the new unit was much higher than the scoping run, under the same handling conditions (volumes, concentration and diafiltration). Coloured fouling of the unit was again observed, but performance checks indicated it recovered well after use. However, >1 kDa yield reduced over time, probably suggesting a stabilisation of the unit.

[0305] Folin assay data: All the TFF fractions were tested in the Folin-Ciocalteau assay and the results suggested that polyphenol was retained in the >1 kDa fraction (see Table 1 & 2 above). If any polyphenol was present in the <1 kDa fraction it was below the limit of quantification of the Folin assay. Samples were also very different by appearance, with the >1 kDa fraction being dark brown and flakey, and the <1 kDa fraction being cream/white, crumbly and slightly tacky.

[0306] HPLC analysis: Preliminary HPLC-SEC analysis confirmed the low MW profile of both fractions, but with clear differences between the >1 and <1 kDa fractions (FIG. 13 below: >1 kDa left, <1 kDa right). The current method is not optimised to resolve this low MW material, but it is clear from the chromatography that the <1 kDa fraction is comprised of lower MW aqueous soluble material (ie later retention times) than the >1 kDa fraction. This is potentially due to the large amount of mannitol present in the <1 kDa fraction (mannitol standard elutes at Rt of 21.3 mins). The RI chromatography profile of these samples was consistent across all the TFF batches.

[0307] A dialysis trial was also carried out to assess yield using other separation methods (Table 5 below).

TABLE-US-00005 TABLE 5 Dialysis trial: >1 yield % Folin PG 1 kDa MWCO Sample/g (w/w) starting equivalents μg/ml Dialysis 7 May 18 0.5 12.6 44.5

[0308] Conclusions from fractionation using 1 kDa MW cut off TFF were that it successfully separated <1 kDa material from the sample. >1 kDa material may contain varying amounts of smaller components, dependent on success of the TFF step. Polyphenols were retained in the >1 kDa fraction. Fouling of the membrane with polyphenol like material was a problem.

[0309] c) Extraction Approaches Based on Solubility & Charge.

[0310] Attempts to further sub-fractionate >1 kDa fractions were carried out by extraction in either ethyl acetate or 80% methanol.

TABLE-US-00006 TABLE 6 Solubility based fractionation of >1 kDa fractions of 17/11/21 ASCOX Folin PG TFF Run Yield % (w/w) equivalents Date Fraction of >1 kDa fraction μg/ml 23 Apr. EtAc soluble 47.8  52* 2018 EtAc insoluble 56.4   53.4* 7 May EtAc soluble 2.6 21 2018 EtAc insoluble 86.8 22 Yield % (w/w) of EtAc soluble fraction 7 May 80% MeOH soluble 87.1   21.5 2018 80% MeOH insoluble 1.7 12 (200 mg) *prior of final Folin assay method development. A repeat extraction with 80% MeOH and higher loading is still to be analysed

[0311] Ethyl acetate extraction: A small amount of >1 kDa material was soluble in ethyl acetate, but yield was low, and the extraction appeared to be non-specific. In later batches much of the material remained in the insoluble fraction with polyphenol being detected in both soluble and insoluble fractions (Table 6 above). In the second trial only a small % of the >1 kDa fraction was soluble in the ethyl acetate. These samples also had very similar chromatography and MW profile as far as could be distinguished by the HPLC-SEC analysis.

[0312] 80% methanol extraction: Due to lack of success with ethyl acetate approach, a Sephadex LH 20 approach was considered. However, due to problems with polyphenols sticking on resins and on the membrane filters, it was decided this approach was unlikely to result in an improved separation. Therefore, differential solubility in 80% methanol was tested instead of ethyl acetate. Literature suggests that polyphenols are most soluble in 80% methanol (rather than 100% methanol) with a lower % methanol resulting in solubilisation of carbohydrates as well, ie. carbohydrates should be insoluble in 80% methanol, allowing separation of these species.

[0313] 87.1% of a >1 kDa fraction was soluble in 80% methanol (Table 6 above). There was only a low yield of the methanol insoluble material (˜1-2%), although this appeared to have a reduced polyphenol content (Table 6 above). Preliminary HPLC-SEC of these fractions indicated that the insoluble material comprised higher MW components, which did not appear to be polyphenol (by MW, and absorbance spectra) (FIG. 14, left). The 80% soluble material comprised all low MW material (FIG. 14, right). Analysis of monosaccharide and sulphate content of these samples suggested that fucoidan was present in the 80% insoluble component, but it was mixed with other compounds, including some polyphenols (Table 7 below). Glucose was also detected in the 80% MeOH soluble fraction and the origin of this sugar was not clear. Laminarin is unlikely to be soluble in 80% MeOH, although this should be further evaluated.

TABLE-US-00007 TABLE 5 Monosaccharide analysis and sulphate data for 80% methanol extraction fractions. 80% MeOH insoluble material contains fucose and galactose, indicative of fucoidan. Sulphate is present, but at lower amounts than would be expected for pure fucoidan. Glucose is also present in both soluble and insoluble fractions. Monosaccharide 80% MeOH insoluble 80% MeOH soluble Arabinose 6.21% 5.70% Rhamnose 2.23% 0.82% Fucose 11.73% 0.00% Xylose 4.71% 0.00% GalA 0.00% 0.00% Mannose 1.54% 0.00% Galactose 4.37% 1.22% Glucose 65.19% 89.13% Glucuronic acid 4.02% 3.12% N-acetyl galactosamine 0.00% 0.00% N-acetyl glucosamine 0.00% 0.00% Recovery % 20.60% 33.70% % sulphate 4.2 LOD

[0314] Conclusions from solubility based fraction approaches were that ethyl acetate was not effective in partitioning components in the >1 kDa fractions. Whilst the fractions from the 80% methanol extraction were different in composition, the yield of the insoluble fraction was only 1% of starting material (w/w), suggesting that if this is fucoidan rich, then it is only present at very low levels. Glucose was also detected in the 80% MeOH soluble fraction and the origin of this sugar is not clear. Laminarin is unlikely to be soluble in 80% MeOH, although this should be further evaluated. Further monosaccharide analysis is being carried out on other samples, incorporating a mannitol and laminarin control.

[0315] Ion exchange chromatography: Despite the lack of fractionation using ethyl acetate, ion exchange chromatography of an ethyl acetate insoluble fraction was carried out to evaluate this approach. Sodium chloride gradient elution resulted in multiple small peaks over the whole gradient, with evidence of polyphenols stuck to the top of the column. There was minimal baseline resolution and due to the material stuck on the column, it was not clear how much of the sample was being resolved. However, the two largest peaks were collected, dialysed and freeze-dried. Analysis of these peaks suggested they did not contain polyphenols (Table 8 below), although the low recovery may skew quantification. Preliminary HPLCSEC also suggested some higher MW components were present, but monosaccharide analysis suggested only minimal carbohydrate was present (recovery in this analysis was <5%). It may be possible to separate other components from polyphenols using this approach, but due to the polyphenols sticking to the column, and multiple peaks eluting, the method is very difficult to control and scale up.

TABLE-US-00008 TABLE 8 Ion exchange chromatography of an EtAc insoluble, >1 kDa TFF fraction of 17/11/21 ASCOX Folin PG Yield % equivalents TFF Run Fraction (w/w) μg/ml 23 Apr. 2018 >1 EtAc soluble 47.8 52*   kDa fraction EtAc insoluble 56.4 53.4* 23 Apr. 18 EtAc QSepharose peak 28-44 mins 6.9 1.4 insoluble (70 mg) QSepharose peak 58-72 mins 14.3 1.2 *prior to final Folin assay method development

[0316] d) Additional Fractionation Approaches to Target Carbohydrate and Polyphenol Components.

[0317] Attempts to fractionate >1 kDa samples to yield fractions enriched for polyphenol or carbohydrate had been largely unsuccessful based on solubility (EtAc/80% MeOH), and handling (such as ion-exchange) was limited by the nature of polyphenols and their non-specific adhesion to resins.

[0318] Whilst preliminary in nature, all the HPLC data indicated that polyphenols are low MW (largely <5 kDa, with literature indicating maybe <1 kDa). Therefore, based on the predicted low MW of polyphenols, further trials were carried out using MW cut off separations, to attempt to further fractionate the sample.

[0319] Vivaspin centrifugal separators: Further fractionation by MW using Vivaspin centrifugal units was carried out using two different MW cut offs with two 300 mg samples of >1 kDa 17/11/21 ASCOX. Whilst the resulting fractions were different in appearance (FIG. 4 below), polyphenol was detected by Folin assay in all fractions other than the <10 kDa fraction (Table 7 below). These and other samples confirm that the colour of samples does appear to correlate with the amount of polyphenol present.

TABLE-US-00009 TABLE 9 Vivaspin separation of >1 kDa material of 17/11/21 ASCOX Vivaspin Yield % Folin PG equivalents MWCO/kPa starting w/w Description μg/ml >10 45.2 Coloured 46.4 <10 51.8 Clear 2.6 >30 35 Coloured 43.9 <30 57.5 Slightly coloured 11

[0320] TFF using higher MW cut off membranes: Due to the difficulty in separating polyphenols from other components in 17/11/21 ASCOX, a trial was also carried out using a PALL centramate TFF system and 100 and 5 kDa T-series membranes. This system would also allow for a much high loading of sample (20 g), in order to generate larger amounts of fractions. 17/11/21 ASCOX was solubilised and filtered, as described above, prior to processing on the TFF. Membranes were used sequentially ie. sample passed through the 100 kDa and then all permeate (including from the diafiltration step) being pooled and applied to the 5 kDa membrane. See Table 10 for yield data.

TABLE-US-00010 TABLE 8 Higher MW cut off trial using centramate 100 and 5 kDa T-series membranes Sam- >100 kDa yield % 5-100 kDa yield % <5 kDa yield % ple/g (w/w) starting (w/w) starting (w/w) starting* 4 Jun. 20 g 1.4 9.8 44.4 2018 *Dialfiltration permeate yield not included

[0321] Preliminary HPLC-SEC analysis suggested that higher MW material from 17/11/21 ASCOX is enriched in the >100 kDa fraction. A defined peak is present in the 5-100 kDa fraction, with most low MW material present in the <5 kDa fraction. This approach, with a molecular sieve much higher than polyphenol MW, along with the <30 kDa vivapsin samples, suggested polyphenols will pass through filtration membranes if the MW cut off is much larger than the predicted polyphenol MW. However, some polyphenol is still retained, probably due to a membrane interaction or formation of large aggregates. It may be possible to improve this separation by changing conditions on the membrane—eg. less ionic, or non-polar conditions, low % alcohol, to prevent aggregation, and get a more accurate size-based separation of polyphenols.

[0322] Conclusions from additional centrifugal filtration and TFF approaches are that size-based membrane fractionation is limited due to membrane interaction by polyphenols. However, the use of a higher MW cut off allows for improved separation, with prediction of fucoidan being retained in the >100 kDa fraction, and the majority of polyphenols passing through the membrane.

Example 7—Anti-Viral Activity of the Crude Extract and the Fractions Obtained in Examples 5 and 6 Above

[0323] The hydrophilic, hydrophobic, (i)<1000 gmol.sup.−1; and (ii) >1000 gmol.sup.−1 fractions obtained in Examples 5 and 6 respectively, were then subjected to anti-viral testing against Murine norovirus using a method as described in Example 4.

[0324] The results are shown in FIGS. 15 and 16 and confirm that it is the BOUND (hydrophobic) and >1000 gmol−1 fraction that are active.

Example 8—Comparison of the Anti-Viral Activity of Seaweed Extract and Phlorotannin Enriched Fraction with Epigallocatechin, a Polyphenol Extracted from Green Tea

[0325] Seaweed extracts obtained from Example 3 (crude extract) and Example 5 (bound extract) were used, together with Epigallocatechin gallate (≥95%) obtained from Sigma Aldrich. The test conditions used were as described in Example 7.

[0326] The results show that the anti-viral activity for both the crude and bound extracts display significant anti-viral activity (see FIG. 17).