Classifications

• • C—CHEMISTRY; METALLURGY
  • C07—ORGANIC CHEMISTRY
  • C07C—ACYCLIC OR CARBOCYCLIC COMPOUNDS
  • C07C59/00—Compounds having carboxyl groups bound to acyclic carbon atoms and containing any of the groups OH, O—metal, —CHO, keto, ether, groups, groups, or groups
  • C07C59/40—Unsaturated compounds
  • C07C59/42—Unsaturated compounds containing hydroxy or O-metal groups
• • C—CHEMISTRY; METALLURGY
  • C07—ORGANIC CHEMISTRY
  • C07C—ACYCLIC OR CARBOCYCLIC COMPOUNDS
  • C07C51/00—Preparation of carboxylic acids or their salts, halides or anhydrides
  • C07C51/347—Preparation of carboxylic acids or their salts, halides or anhydrides by reactions not involving formation of carboxyl groups
  • C07C51/367—Preparation of carboxylic acids or their salts, halides or anhydrides by reactions not involving formation of carboxyl groups by introduction of functional groups containing oxygen only in singly bound form

Definitions

• the present disclosure provides a two-step process for the production of 15-hydroxy fatty acid derivatives, such as 15-(S)-hydroxyeicosatrienoic acid (HETrE or 15-(S)-HETrE) or
• the first step involves the enzymatic oxidation of the fatty acid to a 15(S)-hydroperoxide fatty acid intermediate (e.g., using a liquid enzyme formulation), followed by reduction to the 15(S)-hydroxy fatty acid derivative.
• the enzymatic oxidation step includes contacting the fatty acid with lipoxygenase enzyme, optionally obtained from soy flour.
• the step of reducing the 15(S)-hydroperoxy fatty acid intermediate comprises an in-situ reduction with cysteine. In some embodiments, at least a portion of the process is conducted in the exclusion (e.g., partial exclusion) of air. In some embodiments, the process further comprises isolating and/or purifying the 15(S)-hydroxy fatty acid derivative to form a crude active pharmaceutical ingredient (API) grade product. In some embodiments, the isolating and/or purifying step comprises chromatography purification and/or crystallization. In some embodiments, the process is scaled to produce multi-kilogram quantities of the API, optionally conforming to cGMP.
• API active pharmaceutical ingredient
• the fatty acid is DGLA
• the 15(S)-hydroperoxy fatty acid intermediate is 15(S)-HPETrE
• the 15(S)-hydroxy fatty acid derivative is 15(S)-HETrE.
• the fatty acid is EPA
• the 15(S)-hydroperoxy fatty acid intermediate is 15(S)-HPEPE
• the 15(S)-hydroxy fatty acid derivative is 15(S)-HEPE.
• Figure 1 shows decrease in absorbance of methylene blue solution (680 nm) over time as a function of increasing lipoxygenase enzyme concentration in accordance with one embodiment of the present disclosure.
• Figure 2 shows enzyme cost contribution reduction due to reduced enzyme loadings and cheaper enzyme source in accordance with one embodiment of the present disclosure.
• Figure 3 shows the material cost comparison of commercial freeze-dried enzyme and soyflour extract.
• Figure 4 illustrates the time taken to bleach methylene blue solution using soyflour extract (0.1 M sodium acetate, pH 4.5) compared to commercial enzyme (2.06 mg/mL, 0.2 Munits/mL) in accordance with one embodiment of the present disclosure.
• Figure 5 is an overlay of H NMR spectra for 1891-029 (yellow: after 1 hour under O2, purple: after 2 hours under O2, green: after 3 hours under O2, + further 0.5 eq cysteine), red: stirred under atmospheric conditions over weekend, with further 0.5 eq of cysteine, blue: isolated crude HETrE)
• Figure 6 is an HPLC trace and peak table for 1822-155-4D1 .
• Figure 7 is an HPLC trace and peak table for 1891 -051 -7A.
• Figure 8 is an overlay of H NMR spectra of batch 1822-155-4D1 (red, original method) and batch 1891 -051 -7 A (blue, method according to the present disclosure).
• Figure 9 shows heat rate data for HETrE reaction according to one embodiment of the present disclosure.
• Figure 10 is a plot of pressure vs. temperature for an HETrE reaction according to one embodiment of the present disclosure.
• Figure 1 1 shows heat rate data for an HPETrE reaction according to one embodiment of the present disclosure.
• Figure 12 is a plot of pressure vs. temperature for an HPETrE reaction according to one embodiment of the present disclosure.
• Figure 13 shows an LC-MS analysis of over-oxidized material 1822-063-3 produced in an embodiment of the present disclosure.
• Figure 14 is an overlay of H NMR spectra comparing over-oxidized samples (1822-163-3, blue, 1822193-3, red) in reactions according to one embodiment of the present disclosure
• Figure 15 shows UPLC purity data for HETrE over 4.5 days of refluxing in a 40%
• Figure 16 is a representative spectrum showing overlap between DGLA and HETrE proton
• Figure 17 is an annotated UV chromatogram of front fractions (1891 -051 -6D) from the purification of 15-(S)-HETrE according to one embodiment of the present disclosure.
• Figure 18 is an annotated UV chromatogram of purified 15-(S)-HETrE (1891 -051-7A) produced according to one embodiment of the present disclosure.
• Figure 19 is an annotated UV chromatogram of tail fractions (1891-051-6E) from the purification of 15(S)-HETrE according to one embodiment of the present disclosure.
• Figure 20 shows chemical structures and other characteristic data for DGLA and major impurities.
• Figure 21 shows mass spectral data from fragmentation analysis of a dimeric impurity in purified HETrE72.
• Figure 22 proposes one possible pathway for the production of ester fragments observed by MS. Detailed Description
• the present disclosure provides methods of making 15-(S)-hydroxyeicosatrienoic
• DGLA dihomo-Y-linolenic acid
• a reducing agent e.g., sodium borohydride and/or cysteine
• Scheme 1 A representative process is shown in Scheme 1.
• the 15(S)-HETrE is formed in a single step (e.g., without isolating or purifying the 15(S)-HPETrE intermediate).
• a cost effective enzymatic process for the delivery of kilogram and greater amounts of GMP 15-(S)-HETrE and 15(S)-HEPE is needed (e.g., for use in clinical study programs).
• production of 100, 250 and 500 kg batches of GMP 15(S)-HETrE was carried out.
• the outcome of this program of work was the preparation of 15(S)-HETrE from DGLA, via bio-oxidation with soybean lipoxygenase P1 enzyme to give 15(S)-HPETrE, followed by reduction with 1 .1 equivalents of sodium borohydride.
• the isolated crude 15-(S)-HETrE was purified by column chromatography on silica gel giving a yield of 50% with a purity of >95% by HPLC area percent (235 nm).
• bio-oxidation reaction could also be performed using soy flour as a cheaper source of the enzyme; however, downstream processing was not convenient.
• a 500+ g demonstration batch of non-GMP 15(S)-HETrE (e.g., for toxicology studies) was prepared.
• 500+ g of 15(S)-HETrE (>95% purity by HPLC area%, 235 nm) was prepared from DGLA.
• LC-MS and MS-MS analysis of the isolated impurities suggested that there were di-and tri-oxygenated compounds amongst the ⁇ 5% impurities present in the final product.
• the method comprises oxidizing DGLA with an oxidant to form
• the oxidant is an enzymatic oxidant, such as a lipoxygenase.
• the oxidant is soybean lipoxygenase enzyme (e.g., lipoxygenase P1 enzyme).
• the soybean lipoxygenase enzyme is purified.
• the soybean lipoxygenase enzyme is used as a component of a mixture, for example, soy four.
• the oxidation step may occur in aqueous media, and at a pH suitable to enable enzymatic activity.
• the oxidant is soybean lipoxygenase enzyme
• the oxidation step may occur in buffered aqueous solvent at basic pH (e.g., pH of about 9, or about 9.6).
• the oxidant and/or enzyme may be present in a stoichiometric excess compared to the amount of DGLA.
• the oxidant and/or enzyme may be present in about 1 equivalents, about 1.1 equivalents, about 1 .2 equivalents, about 1 .3 equivalents, about 1 .4 equivalents, about 1.5 equivalents, about 1.6 equivalents, about 1 .7 equivalents, about 1 .8 equivalents, about 1 .9 equivalents, about 2 equivalents, about 2.1 equivalents, about 2.2 equivalents, about 2.3 equivalents, about 2.4 equivalents, about 2.5 equivalents, about 2.6 equivalents, about 2.7 equivalents, about 2.8 equivalents, about 2.9 equivalents, about 3 equivalents, about 3.1 equivalents, about 3.2 equivalents, about 3.3 equivalents, about 3.4 equivalents, about 3.5 equivalents, about 3.6 equivalents, about 3.7 equivalents, about 3.8 equivalents, about 3.9 equivalents, about 4 equivalents, or greater than about 4 equivalents.
• the oxidation step requires addition of an oxygen source, for example when the oxidation step includes contacting the DGLA with an enzyme.
• the oxidation step may occur in the presence of an oxygen source, such as atmospheric oxygen or purified (e.g., at least partially purified) gaseous oxygen.
• the oxidation step occurs under a pressurized oxygen atmosphere, for example at about 1.1 bar, about 1 .2 bar, about 1.3 bar, about 1 .4 bar, about 1 .5 bar, about 1 .6 bar, about 1.7 bar, about 1 .8 bar, about 1.9 bar, about 2 bar, about 2.1 bar, about 2.2 bar, about 2.3 bar, about 2.4 bar, about 2.5 bar, about 2.6 bar, about 2.7 bar, about 2.8 bar, about 2.9 bar, about 3 bar, about 3.1 bar, about 3.2 bar, about 3.3 bar, about 3.4 bar, about 3.5 bar, about 3.6 bar, about 3.7 bar, about 3.8 bar, about 3.9 bar, about 4 bar, or greater than about 4 bar.
• a pressurized oxygen atmosphere for example at about 1.1 bar, about 1 .2 bar, about 1.3 bar, about 1 .4 bar, about 1 .5 bar, about 1 .6 bar, about 1.7 bar, about 1 .8 bar, about 1.9 bar, about 2 bar, about
• the temperature of the oxidation step may be controlled to avoid excess heat generation.
• the oxidation step may occur at about 0-5°C.
• the oxidation step comprises contacting the DGLA with about 2 equivalents of soybean lipoxygenase P1 enzyme in the presence of aqueous buffer (e.g., 0.1 M sodium borate buffer) at pH about 9.6 under oxygen atmosphere at about 2.5 bar at 0-5°C.
• aqueous buffer e.g., 0.1 M sodium borate buffer
• the oxidation step comprises contacting the DGLA with about 2 equivalents of soybean lipoxygenase enzyme as a crude soy flour extract in the presence of aqueous buffer (e.g., 0.1 M sodium borate buffer) at pH about 9.6 in a pressurized oxygen atmosphere at 0-5°C.
• 15(S)-HETrE comprises contacting the 15(S)-HPETrE intermediate with a reducing agent to form the 15(S)-HETrE.
• the reducing agent is sodium borohydride.
• the reducing agent is cysteine. Cysteine, as a milder reducing agent, offers additional advantages over borohydride-type reducing agents. For example, cysteine does not form hydrogen gas as a byproduct, thus enabling safer scale-up opportunities.
• cysteine is a stable reagent and does not require the special handling or storage techniques required for borohydride-type reagents.
• the oxidized form of cysteine (cystine) is a stable dipeptide that is only partially soluble in water, which offers convenient purification opportunities over some other reducing agents.
• the method comprises contacting the DGLA with an oxidant and/or an enzyme, as described above, in the presence of the reducing agent.
• the reducing agent may be present in a stoichiometric excess amount, for example, about 2 equivalents compared to the amount of DGLA.
• the reducing agent is cysteine.
• the method comprises contacting the DGLA with an oxidant and/or an enzyme, as described above, in the presence of the reducing agent.
• the reducing agent may be initially present in a stoichiometric excess amount, for example, about 2 equivalents compared to the amount of DGLA.
• the method may further comprise adding an additional amount of the reducing agent, for example about another 1 equivalent, to the reaction vessel after a period of time.
• the reducing agent is cysteine.
• the method comprises oxidizing EPA with an oxidant to form
• the oxidant is an enzymatic oxidant, such as a lipoxygenase.
• the oxidant is soybean lipoxygenase enzyme (e.g., lipoxygenase P1 enzyme).
• the soybean lipoxygenase enzyme is purified.
• the soybean lipoxygenase enzyme is used as a component of a mixture, for example, soy four.
• the step of oxidizing the EPA and the step of reducing the 15(S)-HPEPE intermediate to form 15(S)-HEPE occur in a single reaction vessel without a step of isolating or purifying the 15(S)-HPEPE intermediate.
• the method comprises contacting the EPA with an oxidant and/or an enzyme, as described above, in the presence of the reducing agent.
• the reducing agent may be present in a stoichiometric excess amount, for example, about 2 equivalents compared to the amount of EPA.
• the reducing agent is cysteine.
• the step of reducing the 15(S)-HPEPE intermediate to form 15(S)-HEPE comprises contacting the 15(S)-HPEPE intermediate with a reducing agent to form the
• the reducing agent is sodium borohydride.
• the reducing agent is cysteine.
• Cysteine as a milder reducing agent, offers additional advantages over borohydride-type reducing agents. For example, cysteine does not form hydrogen gas as a byproduct, thus enabling safer scale-up opportunities.
• cysteine is a stable reagent and does not require the special handling or storage techniques required for borohydride-type reagents.
• the oxidized form of cysteine (cystine) is a stable dipeptide that is only partially soluble in water, which offers convenient purification opportunities over some other reducing agents.
• the method comprises contacting the EPA with an oxidant and/or an enzyme, as described above, in the presence of the reducing agent.
• the reducing agent may be initially present in a stoichiometric excess amount, for example, about 2 equivalents compared to the amount of EPA.
• the method may further comprise adding an additional amount of the reducing agent, for example about another 1 equivalent, to the reaction vessel after a period of time.
• the reducing agent is cysteine.
• the 15(S)-HEPE is converted to an ester (e.g., an ethyl ester) by treating the 15(S)-HEPE with alkyl bromide (e.g., ethyl bromide) and potassium carbonate in dry acetone.
• alkyl bromide e.g., ethyl bromide
• potassium carbonate e.g., potassium carbonate
• the crude ester formed in such embodiments may be conveniently purified by treatment with charcoal and silica gel to form high purity 15(S)-HEPE.
• the method provides compositions comprising a 15-hydroxy fatty acid derivative and one or more impurities.
• the one or more impurities consist of, essentially consist of, or comprise an over-oxidation product, an over-reduced product, a dimer, and/or a positional isomer of the 15(S)-hydroxy fatty acid derivative.
• the dimer is an ester-dimer formed between a carboxylic acid moiety of a first molecule and a hydroxy moiety of a second molecule.
• compositions comprising a
• the composition further comprises one or more impurities.
• the 15(S)-hydroxy fatty acid derivative is present in an amount of at least about 90%, by weight of all fatty acids present in the composition, for example at least about 90%, at least about 91 %, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%, by weight of all fatty acids present in the composition.
• the 15(S)-hydroxy fatty acid derivative is in the form of an ester, such as an ethyl ester.
• the composition further comprises an impurity, wherein the impurity is present in an amount of no more than about 10%, by weight of all fatty acids present, for example no more than about 10%, no more than about 9%, no more than about 8%, no more than about 7%, no more than about 6%, no more than about 5%, no more than about 4%, no more than about 3%, no more than about 2%, no more than about 1 %, of all fatty acids present in the composition.
• the present disclosure provides compositions comprising 15(S)-HETrE.
• the composition further comprises one or more impurities.
• the 15(S)-HETrE is present in an amount of at least about 90%, by weight of all fatty acids present in the composition, for example at least about 90%, at least about 91 %, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%, by weight of all fatty acids present in the composition.
• the 15(S)-HETrE is in the form of an ester, such as an ethyl ester.
• the composition further comprises an impurity, wherein the impurity is present in an amount of no more than about 10%, by weight of all fatty acids present, for example no more than about 10%, no more than about 9%, no more than about 8%, no more than about 7%, no more than about 6%, no more than about 5%, no more than about 4%, no more than about 3%, no more than about 2%, no more than about 1 %, of all fatty acids present in the composition.
• the present disclosure provides compositions comprising 15(S)-HEPE.
• the composition further comprises one or more impurities.
• the 15(S)-HEPE is present in an amount of at least about 90%, by weight of all fatty acids present in the composition, for example at least about 90%, at least about 91 %, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%, by weight of all fatty acids present in the composition.
• the 15(S)-HEPE is in the form of an ester, such as an ethyl ester.
• the composition further comprises an impurity, wherein the impurity is present in an amount of no more than about 10%, by weight of all fatty acids present, for example no more than about 10%, no more than about 9%, no more than about 8%, no more than about 7%, no more than about 6%, no more than about 5%, no more than about 4%, no more than about 3%, no more than about 2%, no more than about 1 %, of all fatty acids present in the composition.
• Example 1 Colorimetric Assay for Assessing Lipoxygenase Activity of Crude Vegetable Homogenates
• Linoleic hydroperoxide, 3-methyl-2-benzothiazoline (MBTH), and 3-(dimethylamino)benzoic acid (DMAB) are known to react in the presence of haemoglobin to give a purple indamine dye, absorbing at 590 nm. The absorbance at this wavelength is linear with the concentration of linoleic hydroperoxide up to 35 uM.
• assay solution B (MBTH and haemoglobin) (1 mL) was added and further incubated for -5 minutes before recording the absorbance reading (Table 1 ).
• a blue colour formation was observed in all samples, except the sodium borate, pH 10.4 extract samples and blank (all reagents, except enzyme).
• a slight grey colour was observed in this case with an absorbance of -0.2.
• a blank was run with all reagents and enzyme except linoleic acid, giving an absorbance of 0.2. It was determined that the enzyme solution was interfering with the absorbance reading.
• DMAB-MBTH assay 10 ⁇ _ of the enzyme solution was withdrawn and added to assay solution A (containing linoleic acid and DMAB) (0.5 mL), in a 2 mL eppendorf. After incubating for ⁇ 5 minutes at room temperature, assay solution B (MBTH and haemoglobin) (0.5 mL) was added and further incubated for ⁇ 5 minutes. Sodium lauryl sulphate solution (1 % w/v) (0.5 mL) was added to quench the reaction and the solutions were centrifuged at 16.1 rcf, 4°C for 10 minutes. The solution was then decanted into a 4 mL cuvette and analysed as before. This was performed in duplicate.
• the activity levels are determined by the time taken for the colour to disappear (i.e. absorbance at 680 nm to reach 0). In order to distinguish between higher enzyme loadings, the dilution series with purified enzyme was repeated using a lower concentration of the enzyme solution. Only 30 ⁇ _ of enzyme solution was added to the methylene blue bleaching solution and the volume was made up to 3 ml. with 270 ⁇ _ of distilled water. The results obtained are recorded in Table 4.
• FIG. 1 A graphical representation of the dilution series is shown in Figure 1.
• An induction time, (0.5 - 2 mins) for the absorbance to begin to decrease, is observed for samples with a lower enzyme concentration.
• the freeze dried enzyme supplied had good stability. Although recommended to be stored at -20°C, and desiccated after use, the batch was stored at ambient, atmospheric conditions for two weeks before being utilised in a reaction using the optimised enzyme loadings (1.8 Munits/ g DGLA). No deterioration in performance was observed.
• the reaction was carried out in buffer solution to maintain pH and solubilise reaction components. Cysteine, a mild reducing agent, was used in place of sodium borohydride.
• Use of cysteine enabled the oxidation/reduction reaction to be carried out in one step. In addition no flammable hydrogen was generated during reaction/quench. Furthermore, fewer over-oxidation products were produced, likely due to cysteine's anti-oxidant properties. It is also believed that the addition of the reducing agent at beginning of the process helps to lessen irreversible enzyme inactivation by high concentrations of hydroperoxide, thus leading to lower enzyme requirement (e.g., cost savings) and lower potential for over-oxidation.
• 15-(S)-HPETrE also helps to solubilise the DGLA substrate.
• cysteine Two equivalents of cysteine appear to be required initially. Addition of 3 equivalents caused the reaction to stall, or at least progress very slowly, likely due to the rapid removal of the enzyme activating hydroperoxide. However, if two equivalents of cysteine are used initially, a further equivalent of cysteine is required to be added to reduce hydroperoxide remaining after one hour reaction time. This is probably due to the oxidation of cysteine in the reaction mixture by oxygen, rather than the hydroperoxide. It was also found that reduction of the enzyme loading to as low as 1.8 Munits activity per gram of DGLA was achievable under these conditions, giving virtually complete consumption of DGLA within a few hours and reduced impurity (e.g., over-oxidised products) formation.
• DGLA DGLA residual DGLA (-6.5 wt%) by 1 H NMR. After standing at atmospheric, filtered through Celite to remove precipitated cystine and adjusted pH to 3 with solid citric acid. Extracted with MTBE to yield 70% crude HETrE. No HPETrE, residual DGLA (-1 wt%) by 1 H NMR. HPLC area% purity (252 nm) 93.19%.
• a 1 L final volume reaction was carried out in a 3L fermentor, where the substrate (40 g) and enzyme (1 .8 Munits/ g DGLA) were added at a constant rate (1 .6 mL/min and 2 mL/min)
• the active enzyme was instead extracted from the flour into buffer solution at room temperature. After centrifugation, the resulting turbid solution was used to perform the bio-oxidation. This removes the insoluble carbohydrates from the reaction mixture, but emulsions are formed during the work-up procedure due to the presence of significant amounts of proteins. Isolation of semi-purified enzyme from soy flour was attempted, although this was found to be inactive.
• the next set of extractions was carried out in 50 ml. centrifuge tubes in temperature controlled orbital shakers. A number of reactions were set-up at various pH and temperatures, and analysed at 3 time points; 1 , 2 and 14 hours and 3 temperatures 5, 25 and 40°C .
• the assay results (determined by methylene blue bleaching at the less discriminating conditions (300 ⁇ _ enzyme solution)) are shown in Table 8. Table 8.
• Table 8 Absorbance readings at 680 nm for soyflour extracts after 1 hour
• the extract solutions (1 : 10 w/v) which had been left over weekend at 25°C were filtered after standing for 1 day.
• the sample at pH 4.5 had a clear yellow supernatant, which was decanted and filtered through a pad of Celite.
• the residual flour was also filtered to give a solid cake.
• 20 mL extract was recovered from an initial buffer volume of 33 mL.
• the samples at pH 6.5 and 9.0, which had cloudy supernatants, were treated as above. When decanting the pH 6.5 extract, the supernatant became mixed with the flour residues.
• the suspension was filtered to give 14 mL of slightly turbid yellow filtrate.
• the pH 9.0 extract gave 20 mL of an orange coloured turbid filtrate.
• the pH 4.5 extract at 30°C (1 :5 w/v) was also filte red to give only 4 mL of filtrate from an initial buffer volume of 13.5 mL.
• cysteine reducing agent allowed reduction of the enzyme activity loading for the reaction by a factor of 7.6 (13.7 Munits to 1.8 Munits / g DGLA), yet giving an acceptable impurity profile. Reaction completion could be obtained within a few hours.
• the reaction was acidified with solid citric acid, thus reducing the total volume by -30%.
• HETrE product precipitated from the reaction mixture, along with cystine residues, which could be collected by filtration. Slurrying the collected precipitate with MtBE and filtration, followed by removal of the solvent, allowed isolation of >100% weight crude HETrE.
• the previous work-up had involved extraction with a comparatively expensive 50/50
• a Biotage KP Sil column was used to purify -45 g of 15(S)-HETrE using a 1 :10 w/w crude HETrE:silica gel ratio and was eluted with (1 ) 2L of Hexane, (2) 2L of 10% MTBE:Hexane, (3) 2L of 20% MTBE:Hexane, (4) 2L of 30% MTBE:Hexane, (5) 4L of 40% MTBE:Hexane , (6) 50% MTBE:Hexane, and (7) 4L of MTBE (total eluent volume: 19L).
• the improved chromatography conditions represent a 40% reduction in solvent volume, decreased processing time (e.g., for solvent removal), and decreased risk of product degradation.
• cysteine as reducing agent that a solid precipitated from the reaction as it progressed, likely the oxidised form cystine.
• the precipitate visible at the end of the reaction was collected by filtration through Celite to give a clear filtrate, before adjusting the pH to 3 using solid citric acid. Further cystine/cysteine precipitates at this stage along with the product.
• the doughy precipitate could be collected by filtration through a sintered funnel and air dried on the funnel. The 'dough' was then slurried in MTBE (1 x 100 mL, 3x 50 mL) to remove the product, which was collected by filtration and dried over Na2S0 4 .
• the mixture was then filtered through a 4L sintered funnel, under a flow of nitrogen. During filtration, which took 1 .5 - 2hours, approximately two thirds of the added MtBE evaporated. The filtrate layers were returned to the vessel and allowed to settle then separated. The aqueous layer was re-extracted (x2). The three extract layers were concentrated separately on the rotary
• reaction solution was lower 1.4°C vs 4.2°C.
• H NMR analysis after 1 hour showed ⁇ 3% HPETrE, and 1 1 % residual DGLA.
• CAD analysis of the reaction sample showed the residual DGLA to be present at 15.5% w/w.
• the crude materials were combined and purified by column chromatography on numerous 75L Biotage KP-Sil columns.
• the purified fractions were combined based on TLC analysis and concentrated on the rotary evaporator at 40°C (protected from light), before venting with nitrogen and storing at 80°C.
• the isolated pure HETrE fractions were dissolved in MtBE, filtered through a clean sintered funnel and combined before concentrating on the rotary evaporator at 40°C (protected from light).
• the purified material was protected from light and dried on a high vacuum pump for three days at room temperature, before venting with nitrogen and storing at -80°C. A total of 469 g of pale yellow oil was obtained, 97.4% area purity by UPLC (252 nm) and
• the batches have similar profiles and purities. A major discrepancy between the two spectra is noted at -35-36 mins, but this is probably due to integration of the broad, ill defined peaks.
• the impurity at 48.1 mins is present at 0.92% in the test batch vs 0.55% in the tox batch, which may be due to the modified reaction conditions, or more likely due to the composition of the DGLA starting material (a tocopherol stabilized batch was used to prepare the test batch).
• the impurities between 63 and 66 minutes correspond, with variations in area%. No new impurities above 0.1 % are apparent from a comparison of the traces.
• Table 13 gives a brief comparison of the previously utilized conditions for preparation of the tox batch with the conditions used to prepare the test batch.
• cysteine was found to give ⁇ 1 % residual DGLA as judged by H NMR.
• a process description (PD) was generated based on the results of reactions performed during this body of work.
• a draft process description was followed in the processing of 2x 300 g of DGLA to provide sufficient HETrE for the stability study project.
• the PD comprises the following exemplary steps:
• LPX1 enzyme powder (5.30 g, 17.7 mg/ g DGLA; 1 .8 Munits/ g DGLA, 1.77 wt%) to the cold reaction solution.
• Chemist check Remove an aliquot; acidify the extract to pH 3, using solid citric acid and extract with MTBE. Dry over Na2S0 4 , filter and remove the solvent on the rotary evaporator and
• DGLA and HETrE residues are readily soluble in alcohols such as ethanol and methanol, or acetone. Cysteine is soluble in water (280 g/L @ 25°C) and ethanol
• cystine (1 .5 g/ 100 g of ethanol @ 19°C) and cystine is sol uble in 1 M HCI with heating (50 g/L) and basic solutions.
• the hydrogenation vessel was cleaned with water and methanol to remove precipitated cysteine/ cystine, etc.
• the reactor was cleaned with a caustic solution (-0.5M), heated to 80°C and stirred overni ght, followed by a rinse with acetone to leave the vessel visually clean.
• Example 16 Thermal hazard assessment
• the Advanced Reactive System Screening Tool (ARSST) is an effective calorimeter that can quickly and safely identify potential chemical hazards in the process industry.
• the synthesis reaction involves the peroxidation of DGLA using lipoxgenase enzyme in sodium borate buffer.
• the resulting peroxide (15-(S)-HPETrE) is then reduced, in situ, to the final product 15-(S)-HETrE in the presence of cysteine.
• the reaction is intended to be carried out as a 1 pot procedure at 0-5°C. The main concern was that elevated temperatures could bring about adverse thermal events particularly as peroxide is formed during the course of the reaction.
• the temp in the vessel was 21 °C before addition.
• the self heat rate was approximately 0.9°C/min. There were no significant exotherms observed during the course of the reaction (Figure 9). No thermal hazards were observed using Advanced Reactive System Screening Tool (ARSST) or Direct Scanning Calorimetry (DSC).
• ARSST Advanced Reactive System Screening Tool
• DSC Direct Scanning Calorimetry
• Reaction 1891 -029 was stirred under oxygen for 3 hours and stored over weekend under air.
• CAD chemical aerosol detection
• DGLA does not absorb above 210 nm
• UV analysis was discounted.
• Charged aerosol detection (CAD) was found to be a suitable method where the IPC sample would be compared with a 0.1 % w/w DGLA standard. The IPC sample would be withdrawn from the reaction and quenched with 50% methanol to denature the enzyme, before the analysis was conducted. This method will allow more accurate determination of residual DGLA levels in reaction samples.
• Example 21 LC-MS investigation of impurities Purified 15(S)-HETrE and impurity enriched front and tail fractions prepared during the course of PRD development were analysed by LC-MS and MS-MS. Based on the supplied specifications for DGLA and the data obtained, tentative structures were suggested for the observed impurities.
• the three major impurities present were 20:2 ⁇ 6 (Eicosadienoic acid (EDA), C20H36O2, RMM 308.48), 20:3 ⁇ 3 (Eicosatrienoic acid (ETE) C20H34O2, RMM 306.48) and 20:4 ⁇ 3 (Eicosatetraenoic acid (ETA), C20H32O2, RMM 304.48) with FAME area% by GC analysis of 1 .1 , 1 .6 and 0.4%, respectively.
• a further 2 unidentified impurities each at 0.2% area (FAME) area were also present giving total impurities of 3.5 area% by FAME GC analysis.
• Tentative structures for the identified masses from the purified product are included in Table 15.
• the major impurities identified in the purified material at m/z 323 correspond with the hydroxylated products of EDA and ETA (HEDA and HETE, respectively).
• No immediately identifiable impurity related to hydroxylation of ETE could be detected (expected m/z 321 ) in the purified material, although in the tail fractions, a peak with m/z 321 (RT 7.19 mins) was observed on the main HETrE product peak.
• the third major impurity had an m/z of 583 (RT 1 1.41 ) which is probably related to the dimeric product visible at 12.22 mins with an m/z of 625. Initially, this was assumed to be formed due to either a Diels Alder mechanism, or self-esterification.
• LC MS-MS Figure 21 was performed on a sample (sample 229 from stability study 1773A0030E) which had been stored at +25°C, in which this dimeric impurit y had increased.
• the impurity formula was C40H66O5.
• the fragmentation mode supports the ester-dimer nature of the impurity. Identification by MS/MS algorithm gave daughter ions with the same profile as HETrE, which is the expected pattern from an ester cleavage ( Figure 22).
• the reaction was carried out by dissolved borax (194 g, 0.508 mol) in Dl water (5 L) completely.
• L-Cysteine 40 g, 0.33 mol
• PPG-2000 anti-foaming agent, 2 mL
• EPA 50 g, 0.165 mol
• the reaction mass was cooled to 0 to 5 °C and LPX1 enzyme (1 .25 g, 2.49%) was added.
• the reaction mixture was purged with oxygen (commercial grade, 99.5%) (1 Kg * 2) and pressurized the reaction vessel with oxygen gas (2.2 Kg). The reaction mixture was stirred under oxygen pressure (2.2 Kg) at 0-5 °C for 1 hour. After complete conversion of EPA, the mass was degassed with nitrogen gas (grade-l, 1 Kg * 2). L-Cysteine (20 g, 0.165 mol) was added and the reaction vessel was pressurized with nitrogen gas (grade-l, 2.2 Kg). The mass was stirred for 1 hour at 0 to 5 °C. After complete conversion of peroxy intermediate, nitrogen pressure was released; the mass was unloaded from the reactor and rinsed with DM water (150 mL * 2).
• the pH of the crude was adjusted to 3.0-4.0 using 40% w/w citric acid solution, followed by the addition of MTBE (1 L).
• the biphasic layers were stirred for 30 minutes and filtered.
• the main filtrate was taken for layer separation and the residue was washed with MTBE (500 mL * 2).
• the residue washed MTBE was taken for extracting the separated aqueous layer.
• the organic layers were combined, dried over anhydrous sodium sulfate (15 g) and filtered.
• the filtrate was concentrated under vacuum at 30 to 35 °C to get stage-l as pale yellow liquid (48 g) with 95.00% HPLC purity.
• a novel one-pot bio-oxidation/reduction of DGLA gave 15(S)-HETrE material of similar purity in a significantly increased yield (76% vs 53% for two-step methods) while using only a small fraction of the enzyme loading and much less extraction and chromatography solvents.
• the added advantage of the use of cysteine is that the reducing agent can be added at the beginning of the reaction, i.e. before the bio-oxidation step.
• the generated hydroperoxide intermediate is reduced in situ as it is formed, avoiding enzyme inactivation caused by high levels of the hydroperoxide, as well as avoiding impurity formation due to over-oxidation and hydroperoxide intermediate degradation. This has resulted in the required lipoxygenase enzyme loading to be reduced by a factor of 7.6. Combined with an alternative, cheaper supply of freeze-dried enzyme, this resulted in overall reduction of the enzyme cost contribution by a factor of 15.
• lipoxygenase extracted from soy flour is now a valid enzyme source for the 15-(S)-HETrE process. It is seen as a long term option towards large scale manufacture, with a suitable soy flour removal technique yet to be developed. In the short term and for smaller, multi-kilogram manufactures it is recommended to use freeze-dried isolated enzyme, since its higher cost is offset by the non-necessity of additional soy flour processing operations.
• the new processing method has included handling the material under inert atmosphere at all suitable stages, in an effort to reduce oxygen exposure and thus the generation of peroxides and subsequent decomposition products.
• the peroxide values obtained are now in the region of 10-15 compared with values of 90+ obtained previously. Results from the stability trial suggest that the initial quality of material has had a positive effect on the observed stability (discounting formation of esters).

Landscapes

• Chemical & Material Sciences (AREA)
• Organic Chemistry (AREA)
• Chemical Kinetics & Catalysis (AREA)
• Engineering & Computer Science (AREA)
• Oil, Petroleum & Natural Gas (AREA)
• Preparation Of Compounds By Using Micro-Organisms (AREA)
• Organic Low-Molecular-Weight Compounds And Preparation Thereof (AREA)

Priority Applications (4)

Applications Claiming Priority (2)

Publications (2)

Family

ID=52669636

Family Applications (1)

Country Status (6)

Cited By (1)

Families Citing this family (1)

Citations (1)

Family Cites Families (4)

• 2014
  • 2014-10-29 EP EP14843235.4A patent/EP3063118B1/en active Active
  • 2014-10-29 CN CN201480071389.7A patent/CN106170470B/zh active Active
  • 2014-10-29 WO PCT/IB2014/003086 patent/WO2015063617A2/en not_active Ceased
  • 2014-10-29 US US14/527,601 patent/US9670126B2/en active Active
  • 2014-10-29 JP JP2016551088A patent/JP6659557B2/ja active Active
  • 2014-10-29 ES ES14843235.4T patent/ES2670776T3/es active Active

Patent Citations (1)

Non-Patent Citations (1)

Also Published As

Similar Documents

Legal Events