Process for the purification of a neutral human milk oligosaccharide (HMO) from microbial fermentation

Abstract

The invention relates to a process for the purification of a neutral human milk oligosaccharide (HMO) from a fermentation broth, the process comprising the steps of: (i) separating biomass from the fermentation broth to provide a crude solution; (ii) treating the crude solution with: (a) a cation-exchange material; (b) an anion-exchange material; and (c) a cation-exchange adsorbent resin; thereby obtaining a purified solution containing the neutral human milk oligosaccharide. Further, the invention relates to a process for fermentatively producing HMO in a fermentation broth and purifying the HMO from the broth.

Claims

1. A process for purifying a neutral human milk oligosaccharide (HMO) from a fermentation broth, the process comprising: (i) separating biomass from the fermentation broth to provide a crude solution; (ii) treating the crude solution with: (a) a cation-exchange material; (b) an anion-exchange material; and (c) a cation-exchange adsorbent resin for adsorbing components other than HMO that are neutral at the pH of the broth and/or to remove color, thereby obtaining a purified solution containing the neutral human milk oligosaccharide.

2. The process according to claim 1, wherein the neutral human milk oligosaccharide is selected from the group consisting of 2′-fucosyllactose (2′-FL), 3-fucosyllactose (3-FL), lacto-N-tetraose (LnT), 6′-galactosyllactose and 3′-galactosyllactose.

3. The process according to claim 1, wherein the cation exchange material is selected from the group consisting of gel-type styrene/divinylbenzene cation exchange resins.

4. The process according to claim 1, wherein the cation exchange material is an acid cation exchange resin having a styrene/divinylbenzene gel-type matrix and sulfonic acid functional groups.

5. The process according to claim 1, wherein the anion exchange material is selected from the group consisting of cross-linked acrylic based anion exchange resins.

6. The process according to claim 1, wherein the anion exchange material is a gel-type cross-linked acrylic based anion exchange resin, having tertiary amine functional groups.

7. The process according to claim 1, wherein the cation-exchange adsorbent resin has a styrene/divinylbenzene copolymer matrix and sulfonic acid functional groups, a porosity of about 0.8 to 1.2 ml/g and an average surface area of ≥600 m.sup.2/g.

8. The process according to claim 1, wherein the crude solution is treated first with the cation exchange material, thereafter with the anion exchange material and thereafter with the adsorbent resin.

9. The process according to claim 1, wherein the separating comprises: (a) microfiltration (MF) to obtain a MF permeate; and (b) subjecting the MF permeate to ultrafiltration with a membrane having a molecular weight cut-off of 5 kDa or less.

10. The process according to claim 9, wherein the ultrafiltration is performed with a membrane having a molecular weight cut-off of 3 kDa or less.

11. The process according to claim 9, wherein the microfiltration is performed at a temperature in the range of 20-75° C.

12. The process according to claim 9, wherein the microfiltration is performed at a temperature in the range of 30-70° C.

13. The process according to claim 9, wherein the MF permeate is subjected to a heat treatment before ultrafiltration.

14. The process according to claim 9, further comprising: (iii) subjecting the purified solution to nanofiltration or reverse osmosis.

15. The process according to claim 1, further comprising (iii) drying or crystallizing the purified solution.

16. The process according to claim 15, wherein drying is by spray drying.

17. The process according to claim 1, wherein the neutral human milk oligosaccharide is obtained in the form of a powder having a water content of less than 10 wt. % and having a purity of 90 wt. % or more.

18. The process according to claim 1, wherein the neutral human milk oligosaccharide is obtained in the form of a powder having a water content of less than 8 wt. % and having a purity of 92 wt. % or more.

19. The process according to claim 1, wherein the neutral human milk oligosaccharide is obtained in the form of a syrup.

20. The process according to claim 1, which does not comprise an electrodialysis step.

21. The process according to claim 1, wherein the purified solution is further subjected to a treatment with a polishing material.

22. A method for producing a neutral human milk oligosaccharide, comprising producing the neutral human milk oligosaccharide by microbial fermentation in a fermentation broth and purifying the produced neutral human milk oligosaccharide using a process according to claim 1.

23. The process according to claim 1, wherein the cation-exchange adsorbent resin adsorbs organic components.

Description

EXAMPLE 1

(1) 2′-FL was produced by microbiologial conversion of glucose to GDP-fucose followed by intracellular enzymatic transfer to lactose in an alpha-1,2-linkage. The reaction was catalyzed by fucosyltransferase present in an engineered host strain of E. coli K12 bacteria, at 30° C. in a fermentation broth, generally as described in WO 2012/112777 and WO 2014/018596, with the difference that glycerol was replaced by glucose and DF204 antifoam by Basildon 98/007K. The other components in the fermentation broth include ammonium, sodium, magnesium, phosphate, potassium, sulphate, antifoaming agents, nitrilotriacetic acid (NTA) and trace elements.

(2) Demineralized water is used for highest purity. Temperature, pH, and dissolved oxygen are controlled within defined limits during the fermentation process.

(3) The fermentation was started with 1000 liter sterilized medium and continued until 3.52% 2′-FL had formed.

(4) The resultant broth containing 2′-FL was subjected to (ia) microfiltration, to remove the cells, using a membrane having a membrane area of 10.8 m.sup.2 and a 0.1 μm cut-off.

(5) Operational Data MF:

(6) TABLE-US-00001 Mass fermentation broth 2028 kg Recirculation flow 6000-6500 L/hr TMP 1.8 Temperature 45° C. Diafiltration 80% of feed Start diafiltration at VCR 1.7 Final concentration VCR 2.5

(7) Details MF:

(8) TABLE-US-00002 Protein according Measurements Total kg to Bradford assay Kg 2′-FL Fermentation 2028 1.6 69.4 broth diawater added 1600 Final retentate 800 3.0 6.7 Final permeate 3080 <0.2 61.60 Deviation in 252 kg* 2% mass balance *(or less diawater is added)

(9) Based on the total amount of 2′-FL present after MF, the yield over the MF was calculated as: 91%.

(10) The chemical analysis was as follows:

(11) TABLE-US-00003 On dry matter (DM) liquid DM % 2.71 2′-FL/DM % 74% Ash/DM % 8.1 Protein according to ppm 12657 Bradford assay/DM HPLC intern/DM % 4.03 Minerals LQS (sum) % 4.30 Heavy metals ppm 2.77 Organic acids (sum) % 13.99

(12) Next, the permeate of the MF was subjected to

(13) (ib) ultrafiltration, to remove protein and minerals, using 2×7 m.sup.2 membrane area, the membrane having a cut-off of 5 kDa.

(14) Operational Data

(15) TABLE-US-00004 Mass MF permeate 3080 kg Recirculation flow 2000 L/hr TMP 1.6 Temperature 14-15° C. Diafiltration 500 kg Start diafiltration at VCR 250 kg IBC - 12 Final concentration VCR 150 kg IBC - 20

(16) Details UF:

(17) TABLE-US-00005 Protein assay Total Acc. to % Kg kg Bradford BRIX OD420 OD450 2′-FL 2′-FL MF 3080 <0.2 2.005 2.0 62.0 permeate MF-P- 3.0 3.14 EOC (measured 23/02) Diawater 500 added Final 150 0.3 1.345 0.48 0.7 retentate UF perm 2.6 2.63 ex diaw Final UF 3430 <0.2 1.049 1.79 61.4 permeate Deviation 0 0.1 in mass balance

(18) The 2′-FL yield of UF was 99%.

(19) For the OD450 a reduction of a factor 2 was reached. This was mainly achieved by the UF. The ultrafiltration blocked part of the (color) components reducing already a part of the color that needed to be removed otherwise in the subsequent steps.

(20) TABLE-US-00006 UF permeate on DM % DM % 2.44 2′-FL/DM % 78 Ash/DM % 7.8 Protein assay acc. to ppm 451 Bradford/DM Minerals (sum) % 4.34 Heavy metals ppm 2.1 Organic acids (sum) % 14.63

(21) The main reduction of components was seen in the protein level as quantified with the Bradford assay. The level of Bradford in the MF permeate was 12000 ppm. This was reduced in the UF permeate to 600 ppm (both on total dry matter).

(22) Thereafter, the crude solution of HMO (permeate of UF) was subjected subsequently to

(23) (iia) cation exchange in 280 l column of a strong acid cation exchange resin having a styrene/divinylbenzene gel-type matrix and sulfonic acid functional groups;

(24) (iib) anion exchange in 225 l column of a weakly basic anion exchange resin having a cross-linked acrylic gel-type matrix and tertiary amine functional groups;

(25) (iic)) adsorption on adsorbent resin in 280 l of a cation exchange adsorbent resin having a styrene/divinylbenzene copolymer matrix and sulfonic acid functional groups, and having a porosity of about 1.0 ml/g and an average surface area of ≥700 m.sup.2/g.

(26) Before passing the crude solution through the columns, the columns were first rinsed with water for about 60 min (at about 1100 l/hr).

(27) The crude solution was passed through the columns at a rate of 1100 l/hr. at the outlet of the last column (iic), the eluent was monitored for its 2′-FL content, using BRIX (a generally known measure for saccharide content of a solution). Once the BRIX value reached 0.2, collection of eluent (containing 2′-FL) was started. Collection continued till BRIX dropped. The collected eluent was a purified solution containing the neutral human milk oligosaccharide.

(28) data after step (iic):

(29) TABLE-US-00007 Brad- Cond % Kg Total kg pH ford mS/cm BRIX OD450 2′-FL 2′-FL feed 3350 6.7 <0.2 2.54 2.57 1.049 1.79 59.6 Collected 3630 4.4 <0.2 0.028 1.7 1.63 59.2 eluent +10% dilution

(30) Based on the analytical data of 2′-FL and the total quantity of UF permeate and collected eluent, the 2′-FL yield of steps (iia)-(iic) was 99%.

(31) Thereafter, the collected eluent was further treated by NF to remove water (concentration of 2′-FL).

(32) Process Data:

(33) TABLE-US-00008 Volume NF retentate 178 Recirculation flow 1400 Pressure 30 Temperature 6-10 Final concentration VCR 31 BRIX

(34) The 2′-FL was concentrated from 1.7 BRIX to 32 BRIX. 2′-FL content was increased from 1.63% to 28.6%. Conductivity increased from 28 to 123 uS/m. OD290 (organics) and OD420 (colour) after concentration remained low at 0.356 and 0.002 respectively.

(35) Chemical Analysis

(36) TABLE-US-00009 NF retentate on DM liquid % DM % 32 2′-FL/DM % 91 Ash/DM % 0.1 Bradford/DM ppm 59 Heavy metals ppm Organic acids (sum) %

(37) The concentrate obtained after NF, having a dry matter content of 32% was subsequently subjected to a pasteurization step and the pasteurized concentrate was spray dried to obtain a 2′-FL powder.

(38) The overall yield starting from the broth was 83%.

EXAMPLE 2

(39) A fermentation broth obtained in a manner essentially as described in Example 1 was subjected to pasteurization (80° C., 15 sec), and subjected to steps (ia), (ib), (iia), (iib) and (iic), essentially as described in Example 1. The eluent from step (iic) was collected as long as BRIX was larger than 0.1. The eluent was subjected to RO instead of NF

(40) TABLE-US-00010 Conduc- Amount mass tivity 2′-FL RO [kg] Brix pH [mS/cm] 2′-FL Bradford kg RO feed 583 2.2 4.68 0.006 1.8 12 10.5 RO 46 25.3 3.75 0.030 21.4 18 9.8 retentate after displace- ment RO 542 0.0 5.64 0.002 0.3 6 1.6 permeate as is

(41) There amount of retentate was too low to concentrate to >30 Brix. The 2′-FL yield was 93%.

EXAMPLE 3

(42) 3-fucosyllactose (3-FL) was produced by microbiologial conversion of glucose to GDP-fucose followed by intracellular enzymatic transfer to lactose in an alpha-1,3-linkage. The reaction was catalyzed by fucosyltransferase present in an engineered host strain of E. coli in a fermentation broth, generally as described in WO 2012/112777.

(43) The fermentation broth comprising 3-FL was subjected to steps (i), (iia), (iib) and (iic), essentially as described in Example 1 above. Measurements of Brix, conductivity and color of the product showed a much increased purity of the 3-FL solution.

(44) Data after Step (ii)(c)

(45) TABLE-US-00011 Conductivity Brix (mS/cm) OD420 Fermentation broth 8.4 27.9 4.64 Purified solution after step (ii)(c) 1.9 0.080 0.008

EXAMPLE 4

(46) Lacto-N-tetraose (LNT) was produced as described in WO 2014/153253. The fermentation broth comprising LNT was subjected to steps (i), (iia), (iib) and (iic), essentially as described in Example 1 above. Measurements of Brix, conductivity and color of the product showed a much increased purity of the LNT solution.

(47) Data after step (ii)(c)

(48) TABLE-US-00012 Conductivity Brix (mS/cm) OD420 Fermentation broth 4.6 11.48 1.66 Purified solution after step (ii)(c) 0.4 0.012 −0.004