GB1581824A

GB1581824A - Preparation of insulin

Info

Publication number: GB1581824A
Application number: GB1740/76A
Authority: GB
Original assignee: Laboratorios Leo SA
Current assignee: Laboratorios Leo SA
Priority date: 1976-01-16
Filing date: 1976-01-16
Publication date: 1980-12-31
Also published as: AT364717B; NO149875B; ATA7977A; NO770119L; EG13113A; SE7700322L; FR2338250A1; FR2338250B1; IE45013L; AU2131777A; NL7700455A; CA1112639A; NO822985L; FI770112A; FI64508B; FI64508C; ES455065A1; IE45013B1; SE445972B; BE850387A

Description

(54) IMPROVEMENTS IN OR RELATING TO THE PREPARATION OF INSULIN (71) We, LABORATORIOS LEO S.A., a company organized under the laws of Spain, residing at Avda. de Pio X[I No. 99, Madrid, Spain, do hereby declare the invention for which we pray that a patent may be granted to us, and the method by which it is to be performed, to be particularly described in and by the following statement: The invention relates to insulin having very little or no antigenicity and a process for its production. It, furthermore, relates to preparations containing such insulin.

It is now generally assumed that the antigenicity in insulin preparations is mostly due to impurities in the preparations, whereas it was previously believed that the insulin antibodies were produced by the insulin as such. These impurities may be accompanying proteins from pancreas distinct from insulin, proinsulin which is a precursor of insulin, intermediate insulin, the dimer, arginine insulin, ethylester insulin, desamido insulin desamidised to various extents, and other insulin modifications.

Efforts have therefore been made in order to produce insulin preparations consisting of pure insulin, the so called monocomponent insulin, free of impurities and accompanying substances of any kind.

For that purpose it has been proposed to subject amorphous or crystalline insulin prepared in a conventional manner to an extensive further purification, e.g. by gel filtration and/or ion exchange treatment.

The resulting highly purified insulins show a strong decrease in antigenicity, but not a complete removal thereof. This is no doubt due to the fact that in spite of the purification steps the purified preparations still contain substances different from insulin.

The invention is based on the recognition that some of the impurities present in insulin are formed during the recovery itself of insulin, the usually used processes resulting inter alia in a decomposition of the insulin and formation of aggregates. The insulin as it is accumulated in the pancreas glands, from which the insulin is recovered; is the pure monomer (having a molecular weight of about 6000), and it is extracted as such by conventional extraction with an acid i.e. aqueous 60-80 per cent alcohol. In the conventional further processing, which inter alia comprises a separation of fat by vacuum distillation of the extract at 25-300C, whereby the alcohol is evaporated and an aqueous extract with a strongly reduced alcohol content is obtained, the monomer is, however, subjected to conditions resulting in decomposition - inter alia due to the detrimental action of the enzymes which can display their activity in aqueous solution (solutions having an alcohol content less than 50 per cent) by causing formation of aggregates, etc., whereas they are not active in alcoholic solution (over 50 per cent alcohol).

We therefore propose to carry out the preparation of insulin in such a manner that the insulin, from the extraction from the pancreas glands until the obtaining of the final product, is only subjected to conditions that do not cause the formation of decomposition products, aggregates etc.

According to one aspect of this invention we provide a process of purifying insulin wherein an insulin-containing extract is prepared from pancreas glands, which extract is worked up in a plurality of steps to pure insulin, characterised in that the working up including concentration of the extract is carried out in such a way that until the final recovery of the insulin, the insulin is maintained in a solvent environment during the whole process, the undesired substances being from the beginning of the process until the end removed from the solvent or solvents used.

By "in a solvent environment" we mean that the insulin is in solution or in a like state.

It may remain at all times in the liquid phase in solution or if for example it is retained on an ion-exchange material will be protected by the solvent and hence still in the desired environment.

Insulin is very easily soluble in an acidic alcoholic extractant medium of from about 55 to about 80 per cent by volume of alcohol, especially ethanol, and in an extract the insulin will, if the extract is treated expediently, not decompose and form aggregates with itself or with impurities in the alcoholic extract. For this reason it is desirable to keep the extractant medium at an alcohol level which is 75% v/v and preferably 60-80to v/v throughout the working-up.

Use can of course be made of other extractants than acidic aqueous alcohol, extractants in which insulin is easily soluble and in which it will not form aggregates with itself or with impurities in the extract, e.g. aqueous acetone.

In this process purification of the obtained insulin extract takes place to remove dissolved impurities and fat and undesirable proteins and derivatives of insulin, until the pure insulin remains alone in the extract (or in another liquid phase).

The process may for instance be carried out in the following manner; Comminuted frozen pancreas glands are extracted with ethyl alcohol (60-80 per cent by volume) acidified with hydrochloric acid to about pH 3. After separation of the pancreas mass the pH of the extract is adjusted to about 8, whereby certain proteins different from insulin are precipitated and separated. The pH is again reduced to about 3. Instead of using this so-called "pH-8 precipitation" for separating undesirable proteins one can send the acid extract with a pH-value about 3 through an ultrafiltration/hyperfiltration plant, see below, using an appropriate membrane to retaining the said proteins, i.a. enzymes.

The resulting clear extract is cooled down in a special fat freezing plant to a temperature of about -30 to -450C, preferably about -35"C, whereby the fat is crystallized in compact fat crystals of small surface area. These crystals are separated, e.g. by centrifugation, at the same low temperature. At this low temperature of -30 to -450C and with the alcohol content remaining in the range of 60-80 per cent by volume, no harmful effect on the insulin takes place as is the case in the conventional fat removal by vacuum distillation at 25-300C with decreasing alcohol content in the extract.

Further details regarding this fat separation by freezing out are disclosed in the Specification No. 1 503 919.

After this gentle separation of the fat a comparatively large volume of fat-free extract is available, and this large volume has to be reduced, also in a manner which is harmless to the insulin.

Gel filtration and/or ion exchange treatment may be used, but the invention the following method is preferred: and indeed forms in itself a further aspect of the present invention.

Use is made of an ultrafiltration/hyperfiltration plant working according to the membrane method, the so-called reverse osmosis system (cf. for instance US Patent 3.623.610).

In this aspect and using reverse osmosis on a fat-free extract of insulin one can, by an appropriate choice of membrane types, obtain, together with a concentration of the volume of liquid, a separation of insulin from impurities present in the extract, both substances having larger molecules than the insulin and substances having smaller molecules.

One achieves the removal not only of undesirable substances of pancreatic origin which are dissolved in the extract, but also of any other molecules of non-pancreatic origin having a size different from that of the insulin, which for example the slaughterhouses might by error or inadvertence have included in the pancreas supplies.

The separation at that moment of by far the major part of these molecules, both larger and smaller than the insulin molecule, by ultraand hyperfiltration is of decisive importance to the obtaining of the monocomponent insulin, since these impurities are here separated at a moment when they are in solution and when they have had no possiblity of displaying a detrimental effect on the dissolved insulin, inter alia on account of the low temperature during the preceding process steps.

The fat-free clear extract is pumped through the reverse osmosis apparatus, which is provided with cooling means for the extract, for instance at a temperature between about +100C and -100C, preferably between 0 C and -10"C, and at a pressure of 2-20 atmospheres preferably 6 atmospheres, the type of membrane being chosen so that the large undesirable molecules are separated, down to a size as close to the size of the insulin molecule as possible.

As membranes e.g. GRX-6 or GRX-7 (polysulphone membranes manufactured by the De Danske Sukkerfabriker A/S) can be used. The large molecules remain in the concentrate, whereas the insulin and the impurities of smaller molecule size than that of the insulin form part of the permeate. This permeate is thereafter passed through a reverse osmosis apparatus wherein the type of membrane is chosen so that the major part of the undesirable molecules smaller than the insulin molecule are separated, up to a size as close to the size of the insulin molecule as possible. As membranes e.g. GR8-P or 930 (cellulose acetate) both manufactured by De Danske Sukkerfabriker A/S can be used. The insulin remains in the concentrate. The extract can be concentrated as strongly as desired, e.g. to 1/5 to 1/40, preferably 1/5 to 1/10, of the original volume.

This is a process in which the concentration ratio alcohol: water remains the same as in the starting extract.

The membranes used in the ultra- and hyperfiltration plant must be resistant to concentrated alcoholic solutions of for instance 62 per cent alcohol. Use can be made of members of Polymers other than the above-mentioned polysulphones and of cellulose acetate as such as certain polyamides.

With regard to the construction of the ultraand hyperfiltration plant, reference can, for instance be made to the following patent specifications: US patent No. 3,872,015, British patent No. 1,390,671 and Swiss patent No. 542,639, wherein embodiments of the plant and/or parts thereof are described. The invention is, of course, not limited to the use of plants of any definite constructions.

The resulting concentrate, which contains practically all the insulin from the raw extract, is transparent but more strongly coloured than the raw extract on account of the concentration. The colouring substances are washed out with, for instance 62 per cent alcohol, over membranes which retain the insulin but allow the colouring substances to pass through. The whole volume of washing liquid passes together with the said substances in the permeate, while the volume of the concentrate remains unaltered.

Simultaneously with the washing off of the colouring substances, the salts present in the extract and originating from the pancreas extraction are washed out.

The rather few dissolved impurities, which still remain in the insulin-containing extract, can suitably be separated by means of ion exchangers, cation exchangers as well as anion exchangers. The separation can also be effected by means of molecular sieve, e.g. Sephadex G-50 manufactured by Pharmacia Fine Chemicals, Uppsala, Sweden, (Sephader is a Registered trade mark) and other suitable separation methods known per se, but the use of ion exchangers is preferred.

Ion exchange processes for the cleaning of insulin by the column chromatography method as well as by the batch method are known per se, especially the use of column chromatography.

It is however hitherto unknown to apply these methods to a concentrated alcoholic raw insulin extract free from fat. The conventional alcoholic raw insulin extracts having all the fat dissolved therein will, especially by the column chromatography method, contaminate the ion exchangers with the impurities - the colouring substances and the fat - to such an extent that the ion exchangers after adsorption and elution once or twice will be so contaminated that it will be extremely difficult to purify and regenerate the exchangers for re-use.

When using the ion exchange or molecular sieve processes it is a condition that the extract being treated has been freed from fat and partially also from colouring substances, undesirable proteins, and as other impurities to such an extent that the ion exchangers and the molecular sieve substances can be regenerated and purified so that they are fully applicable for a substantial period of time, as their use otherwise will be economically prohibitive.

Such a sufficient and necessary purification is obtained by the method described aboye for prepurification of the alcoholic raw insulin extract comprising crystallization of the fat at a temperature as low as about -400C and removal by ultrafiltration through a reverse osmosis system of part of the colouring substances as well as a substantial part of the molecules which are larger or than and smaller than the insulin molecule, including molecules quite close to the size of the insulin molecule.

Most of the available types of ion exchangers can be used, including ion exchangers on a resin or cellulose bases as well as the Sephadex ion exchangers SP and QAE known per se, which are cation exchangers and anion exchan gers, respectively, and are modified crosslinked dextran chains having a tridimensional network of polysaccharide (manufactured by Pharmacia Fine Chemicals AB,Uppsala, Sweden). The ion exchangers SP-Sephadex C-25 (cation exchanger) and QAE-Sephadex A-25 (anion exchanger) are preferred.

One can choose between carrying out column chromatography or batch separation on the ion exchangers, or possibly both methods can be used.

The batch method has been found to present enormous advantages when used in larger industrial operation as a medium to fine purification process inserted between the separation and concentration by reverse osmosis and a chromatographic purification using an ion exchanger column. Batch separation is a very quick technique and the method causes no technical trouble as a consequence of swelling or shrinking of the ion exchanger. Separation on an ion exchanger can be carried out by binding the impurities and by allowing only the insulin to remain unbonded in the solution.

After the ion exchanger has been equilibrated in a suitable buffer chosen at the isoelectric point of the insulin, pH 5.2, the concentrated prepurified raw insulin extract is adjusted to pH 5.2 and stirred in contact with the ion exchanger and the whole is stirred for instance for 2 hours, whereafter the mixture is filtered.

All the impurities are bound to the ion exchanger, whereas the pure insulin is present in a dissolved state in the alcoholic filtrate, since the insulin at its isoelectric point is not bound to the ion exchanger.

The advantage of this process is that the insulin has not to be eluted from the ion exchanger and that the insulin extract immediately after the filtration can pass to the next process step.

The ion exchangers are regenerated in a conventional manner.

Alternatively, the insulin and the other proteins can be bound to the ion exchanger and subsequently be eluted in fractions by resuspending the mixture in a buffer of a higher ionic strength or a different pH value. By the latter method a smaller loss of insulin can be obtained than by the former: method, for which reason the latter method is preferred.

The method may for instance be carried out substantially in the following manner: 300 g of dry SP-Sephadex C-25 were swelled during 48 hours in 1000 ml of a buffer, 0.1 molar acetic acid (HAc) in 62% ethanol, pH 3.0. The buffer was changed several times. The weight of the swelled SP Sephadex was 720 g.

1000 ml of extract, concentrated 10 times by osmosis, having a content of 18000 i. units of insulin, was adjusted to pH 3.0 and stirred in a rotary container (12 revolutions per minute) for 1 hours together with half the swelled ion exchanger for adsorption. After filtration in vacuo the adsorption of the filtrate was continued on the second half of the ion exchanger (360 g swelled) likewise for 1 hours. The supernatant showed an insulin content of 288 i. units corresponding to a loss of 1.6% during the adsorption. Such a double adsorption has proved to reduce the loss to one third of the loss occurring in a single adsorption.

A change now was made to buffer: 1000 ml, 0.1 molar HAc in 62% ethanol, pIl 3.0, containing 0.075 mol NaCl. Impurities were eluted by stirring for 2 hours, whereas the insulin was not eluted.

After filtration a change was made to buffer: 1000 ml of 0.1 molar tris (hydroxymethyl) aminomethane (Tris) in 62% ethanol, pH 8.0, whereupon the insulin was eluted from the ion exchanger SP; insulin content 17000 i.

units, the loss during the eluating process being 4t/o; total loss 5.6%.

The ion exchanger was washed twice for removal of residual insulin attached to the ion exchanger.

In polyacrylamide gel electrophoresis (with 15% polyacrylamide) 4 bands (some of them weaker than the insulin band) were found above the insulin band, whereas no band was to be found below the insulin band. This means that the partially purified insulin has not during the production process been deamidized.

It is free from monodesamido insulin, which is an entirely new feature in the production of insulin.

Subsequently to eluting and washing, the ion exhanger was treated under slow stirring for 2-3 hours with 62% ethanol having added thereto 0.2-0.3 mol NaCl. Subsequently, the ion exchanger was regenerated with equilibration buffer pH 3, and it was now perfectly clean, chalk white and ready for the nextbatch.

The batch process can be carried out advantageously in a known apparatus of special design, having a being Y-shaped hollow body which is rotatable at an adjustable speed. Thereby a very efficient and mild treatment of the ion exchanger during the adsorption and eluating is provided while at the same time the loss during the adsorption as well as during the eluting has been minimized. The entire process can take place without removing the ion exchanger from the apparatus, which is provided with a filtering device and a vacuum tank for rapid filtration of the liquid from the ion exchanger.

The batch process is composed of the following steps: 1) Adsorption on the ion exchanger, pH 3, of the insulin and possibly of other proteins from the concentrated raw insulin extract, in 2 steps with intermediate vacuum filtration.

2) Vacuum filtration and removal of the protein-free extract.

3) Washing of the ion exchanger twice with 62% ethanol buffer, pH 3, and removal by filtration of the washing liquids.

4) Eluting impurities with buffer pH 3, ionic strength 0.075 mol NaCl, 2 hours, 12 r.p.m.

5) Filtration of eluting liquid with its content of impurities.

6) Washing twice of the ion.exchanger with 62% ethanol buffer, pH 3; filtration.

7) Eluting the insulin with buffer 0.1 molar Tris in 62% ethanol, pH 8; the whole mixture, eluting liquid and ion exchanger, is to be adjusted to pH 8 with Tris; the pH of the ion exchanger was 3.

8) Filtration of the eluting liquid containing the insulin.

9) Washing twice of the ion exchanger for removal of residual insulin attached to the ion exchanger.

10) Purification and regeneration of the ion exchanger.

The total production cycle 1-9 requires 12 hours, and the purification and regeneration of the ion exchanger require 4 hours, making a total of 16 hours.

The eluting extract from phase 7, after adjustment of the pH and the ion strength, can be applied directly to cation or anion exchangers according to the column methods known per se. This process is preferably carried out on programmed columns using automatic fractioning, wherein the central part of the insulin curve is removed which part now contains the pure insulin showing only one band in polyacrylamide electrophoresis with 15% polyacrylamide.

From the concentrated insulin-containing eluate from the last column the insulin can be recovered in a conventional manner, i.e. by dilution of the solution and subsequent precipitation for instance by the addition of zinc ions.

According to another aspect of the invention, it has however been found that it is not necessary to carry out a dilution of the highly concentrated aqueous alcoholic extract prior to precipitation. In this method aspect, insulin can be precipitated direct from a solution which has been concentrated to 1/5 to 1/10 of the volume of the original pancreas extract (for example by concentration in reverse osmosis as described above) by adding to it metal ions, for instance zinc ions, is effected at a temperature for instance at -300C to -45 C sufficiently low to yield quantitative precipitation of the insulin.

Thus, a zinc acetate solution may be added to the filtrate, subsequently leaving the solution to stand for 24 hours in a cold store at -350C.

Thereby the insulin is precipitated quantitatively from the for instance 62% alcoholic solution. The precipitated insulin is removed by centrifugation, washed e.g. with acetone, and dried. If desired, it may subsequently be crystallized.

As mentioned, the insulin produced by the present process and which is per se a further aspect of this invention is a pure monocomponent insulin showing merely a single component when analyzed by polyacrylamide gel electrophoresis (DISC PAGE) by the use of a 15% concentration of the polyacrylamide.

(Concerning the general principle of this method, reference is made to Ann. N.Y. Acad.

Sci., 121, pp. 321-349 and 407r427 (1964)).

Insulin of the purity described herein has to the best of our knowledge not been produced previously.

In the British Patent Specification Nos.

1,285,023 and 1,285,024 production of a high purity insulin is described. It is stated that the purified insulin in a polyacrylamide gel electrophoresis shows substantially a single component.

This insulin however still contains a smaller amount of desarnido insulin, and in polyacrylamide gel electrophoresis using a gel with 15% polyacrylamide, it will show two bands, i.e. besides the monoinsulin band a desamido insulin band, This result was confirmation of that reported by Yue and Tutle, (The Lancet, 26th October 1974) who used 20% polyacrylamide gel.

A smaller content of desamido insulin will, on the other hand, not be ascertainable in polyacrylamide gel electrophoresis using 71/2% polyacrylamide. At this content, complete separation of the individual components is not obtained, but inter alia an integration of the insulin and the monodesamido insulin into one band occurs where-as two bands will appear with 15% polyacrylamide.

The insulin produced by the process of the invention has proved to be extremely stable contrary to the hitherto known highly purified insulins, in which for instance during storing additional small amounts of desamido insulin are often produced; why this happens is unknown, but the cause may perhaps be that the insulins previously produced, despite extensive purification, still contain traces of substances having a decomposing effect on the insulin.

From the pure insulin described above insulin preparations can be prepared in any manner known per se, for instance by dissolving and/or suspending of the insulin, in amorphous and/or crystalline form, in an aqueous medium suitable for injection, infusion or implantation techniques.

The hitherto unknown steps of procedure described above, viz, the separation of impurities from a fat-free insulin-containing extract by means of reverse osmosis and precipitation of insulin from a highly concentrated alcoholic solution under cooled conditions, may of course be used not only in the described combination of process steps. The principle: separation of impurities from a fatfree insulin-containing extract by reverse osmosis can be used in any combination, and the solvent does not of course need to be a highly concentrated aqueous alcohol, but may be any suitable solvent for insulin. The precipitation under cooled conditions can be applied to any highly concentrated insulin-containing organic solvent extract, regardless of the manner in which it has been obtained.

The invention therefore resides not only in the combination of process steps described herein, but it also resides in said two hitherto unknown steps as considered separately.

WHAT WE CLAIM IS: 1. A process of purifying insulin wherein an insulin containing extract is prepared from pancreas glands, which extract is worked up in a plurality of steps to pure insulin, characterised in that the working up including concentration of the extract is carried out in such a way that until the final recovering of the insulin, the insulin is maintained in a solvent environment during the whole process, the undesired substances being from the beginning of the process until the end removed from the solvent or solvents used.

2. A process according to Claim 1 wherein the solvent or solvents used is or are such as to suppress detrimental action of enzymes in the impure insulin extract.

3. A process according to claim 2 wherein the solvent is an aqueous alcoholic medium whose content of watermiscible alcohol is always > 50% v/v.

4. A process according to Claim 3 wherein said alcohol content is always within the range 60-80% v/v.

5. A process according to Claim 3 or Claim 4 wherein the alcohol is ethanol.

6. A process according to any one of the preceding claims wherein the extract is substantially freed from fat prior to any dialysis and/or chromatographic steps.

7. A process according to any one of the preceding claims wherein the insulin-containing extract is subjected to a treatment for removal of fat comprising cooling of the extract to a temperature below -25"C followed by separation of the crystallized fat.

8. A process according to Claim 6 or Claim 7 wherein the extract after separation of the fat is subjected to reverse osmosis whereby the extract is concentrated and separated from impurities of greater and of lesser molecular size than insulin.

9. A process according to Claim 8, wherein the insulin containing concentrate from the reverse osmosis is washed in a reverse osmosis plant for removal of colouring substances and salts from the concentrate.

10. A process according to Claim 9, wherein the purified concentrate is subjected to further purification by means of ion exchange.

11. A process according to Claim 10, wherein a first treatment by ion exchange is carried out

**WARNING** end of DESC field may overlap start of CLMS **.

Claims

**WARNING** start of CLMS field may overlap end of DESC **. from the for instance 62% alcoholic solution. The precipitated insulin is removed by centrifugation, washed e.g. with acetone, and dried. If desired, it may subsequently be crystallized. As mentioned, the insulin produced by the present process and which is per se a further aspect of this invention is a pure monocomponent insulin showing merely a single component when analyzed by polyacrylamide gel electrophoresis (DISC PAGE) by the use of a 15% concentration of the polyacrylamide. (Concerning the general principle of this method, reference is made to Ann. N.Y. Acad. Sci., 121, pp. 321-349 and 407r427 (1964)). Insulin of the purity described herein has to the best of our knowledge not been produced previously. In the British Patent Specification Nos. 1,285,023 and 1,285,024 production of a high purity insulin is described. It is stated that the purified insulin in a polyacrylamide gel electrophoresis shows substantially a single component. This insulin however still contains a smaller amount of desarnido insulin, and in polyacrylamide gel electrophoresis using a gel with 15% polyacrylamide, it will show two bands, i.e. besides the monoinsulin band a desamido insulin band, This result was confirmation of that reported by Yue and Tutle, (The Lancet, 26th October 1974) who used 20% polyacrylamide gel. A smaller content of desamido insulin will, on the other hand, not be ascertainable in polyacrylamide gel electrophoresis using 71/2% polyacrylamide. At this content, complete separation of the individual components is not obtained, but inter alia an integration of the insulin and the monodesamido insulin into one band occurs where-as two bands will appear with 15% polyacrylamide. The insulin produced by the process of the invention has proved to be extremely stable contrary to the hitherto known highly purified insulins, in which for instance during storing additional small amounts of desamido insulin are often produced; why this happens is unknown, but the cause may perhaps be that the insulins previously produced, despite extensive purification, still contain traces of substances having a decomposing effect on the insulin. From the pure insulin described above insulin preparations can be prepared in any manner known per se, for instance by dissolving and/or suspending of the insulin, in amorphous and/or crystalline form, in an aqueous medium suitable for injection, infusion or implantation techniques. The hitherto unknown steps of procedure described above, viz, the separation of impurities from a fat-free insulin-containing extract by means of reverse osmosis and precipitation of insulin from a highly concentrated alcoholic solution under cooled conditions, may of course be used not only in the described combination of process steps. The principle: separation of impurities from a fatfree insulin-containing extract by reverse osmosis can be used in any combination, and the solvent does not of course need to be a highly concentrated aqueous alcohol, but may be any suitable solvent for insulin. The precipitation under cooled conditions can be applied to any highly concentrated insulin-containing organic solvent extract, regardless of the manner in which it has been obtained. The invention therefore resides not only in the combination of process steps described herein, but it also resides in said two hitherto unknown steps as considered separately. WHAT WE CLAIM IS:

1. A process of purifying insulin wherein an insulin containing extract is prepared from pancreas glands, which extract is worked up in a plurality of steps to pure insulin, characterised in that the working up including concentration of the extract is carried out in such a way that until the final recovering of the insulin, the insulin is maintained in a solvent environment during the whole process, the undesired substances being from the beginning of the process until the end removed from the solvent or solvents used.

5. A process according to Claim 3 or Claim 4 wherein the alcohol is ethanol.

as a batch process.

12. A process according to Claim 11, wherein the insulin-containing liquid phase from the batch-method ion-exchange is subjected to one or more further purifications by means of ion exchange column chromatography.

13. A process according to Claim 12 wherein the insulin remains in the liquid phase in solution.

14. A process according to any one of the preceding claims wherein the insulin is recovered from a concentrated purified solution by precipitation with metal ions at a temperature sufficiently low to ensure a quantitative precipitation of the insulin.

15. A process according to Claim 14 in which the metal ions are Zn++.

16. A process according to Claim 14 or Claim 15 in which the precipitation is effected in the temperature range -300C to -450C.

17. A process for preparation of purified insulin including the step of preparing a fat free organic solvent extract of insulin as a crude extract from pancreas gleands while maintaining the insulin in a solvent environment, and subjecting said extract to reverse osmosis to concentrate the extract and to remove substances having a larger or smaller molecular size than insulin.

18. A process of recovering insulin from a solution thereof in an organic solvent prepared by concentration of an insulin containing pancreas extract to 1/5 to 1/10 of its volume at the original extraction, wherein the insulin is precipitated from the said concentrated solution by adding metal ions at a temperature of the solution sufficiently low to ensure quantitative precipitation of the insulin.

19. A process according to Claim 18 wherein the metal ions are Zn++.

20. A process according to Claim 18 or Claim 19 wherein the precipitation is effected in the temperature range -30"C to -450C.

21. Insulin prepared by a process according to any one of Claims 1 to 16.

22. Insulin according to Claim 21 which shows only a single component when analysed by polyacrylamide gel electrophoresis using a gel containing 15% polyacrylamide.

23. Monocomponent insulin being insulin purified to such a degree that it shows only a single component when analysed by polyacrylamide gel electrophoresis using a gel containing 15% polyacrylamide.

24. Insulin prepared by a process according to any one of Claims 17 to 20.

25. An insulin preparation characterised in that it contains monocomponent insulin according to any one of Claims 21 to 24 in amorphous and/or crystalline form, dissolved and/or suspended in an aqueous isotonic medium suited for injection, infusion or implantation purposes.

\ t Tt TT\T rTs w T r T T T n n A