JPS5946939B2 - Manufacturing method of insulin - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a process for producing insulin drugs with sustained action and low antigenicity.

Insulin drugs extracted from pig or cow pancreas are commonly used to treat diabetes.

That is, about 30% of the world's insulin consumption depends on insulin from pigs, and the remaining about 70% depends on insulin from cows. Insulin from other animals, such as sheep, has also been proposed, but it is currently of little commercial significance. Conventional insulin therapy has many problems, particularly allergies and fat malnutrition.

Conventional insulin therapy also produces insulin antibodies in most patients, which necessitates increased insulin doses; some of the insulin binds to the antibodies produced, and insulin It has long been known that while bound in this manner, the drug loses its effectiveness as a blood sugar regulator. The primary cause of this antibody production and subsequent increase in insulin dosage is generally believed to be due to the presence of various impurities in the insulin sold. Examples of such impurities are insulin dimer, proinsulin, intermediate insulin (an intermediate between pyroinsulin and insulin), arginine insulin, ethyl ester insulin, monodesamide insulin, and didesamide insulin. .

The first three of these compounds are known to be highly antigenic. Furthermore, the fourth and fifth positions are also highly antigenic, while the sixth and seventh positions and the insulin molecule itself have no or only slight antigenicity (Reference, [MOnOcOmpOnentIns.
ullnanditsCllnicallmplica
Schllchtcrull, 197
It is known that it is possible to remove these impurities by gel filtration or ion exchange, etc., and obtain highly pure insulin consisting of essentially a single compound. There is. Recently, insulin drugs with these impurities removed have become commercially available, and in many cases have been able to reduce the amount of antibodies produced by diabetic patients. Also known from the above-mentioned documents are methods for producing highly purified fast-acting preparations of insulin from pigs and cows that hardly cause the production of insulin antibodies.

Furthermore, there is also a known method for producing long-lasting insulin drugs with low antigenicity based on highly purified porcine insulin (Reference, "DiabetOlOgia").
JT, Deckertetal, VOllO, No. 7
03-8, 1974). However, as described above, it is possible to produce a fast-acting preparation from bovine insulin by purifying it to the same degree as porcine insulin and producing almost no antibody production. Bovine insulin with highly antigenic purity is not yet known. This has become an important issue because there is a worldwide demand for long-acting drugs made from bovine insulin. Pig or bovine insulin, which is made of a single component without any impurities, has lower antigenicity than insulin produced by recrystallization from rice. However, considerable antigenicity remains. In other words, the problem of antigenicity cannot be solved simply by increasing the purity. The formation of antibodies against compounds such as insulin in living cells is based on chemical differences in their primary structure;
ndbOOkOfPhyslOlOgy''Chapter 9, Chapter 1
p. 60, 1972 6 Immunochemistry of Insulin 7)
, the three-dimensional structure of the compound is incompatible with cells and is rejected by the cells ("Diabetes" VOll8,
194, 1969), or the overall size (mass) of the compound is incompatible with the cell and is rejected by the cell (6Handb00k0fPhys1
010gy1) Chapter 9, p. 160, 1972). In addition, it is also possible to form antibodies by administering a component capable of activating the immune system together with insulin.

Several tests have shown that the physical state of insulin tends to be fundamentally involved in antibody formation (6H0
rnMetabRes, 6, (1974) 185-17
7, and "Neth J, Med, 17 (1974) pp. 234-238). The reason why fast-acting, high-purity insulin drugs cause no or only limited antibody formation is probably because the insulin in these cases is highly purified and also in a monomeric state or in the form of loose clumps. On the other hand, the long-acting, highly purified insulin drugs that have been manufactured and sold in the past have the property of forming antibodies due to the insulin being aggregated or polymerized. .

This problem is most evident for bovine insulin, which is more bulky or polymeric (e.g. dimer or
This seems to be due to the fact that it is easy to form It is known that bovine insulin, which does not contain additives, has a wider precipitation range due to its isoelectric point than pig insulin.

Furthermore, it is known that the sedimentation range due to the isoelectric point of cattle can be narrowed to the same level as insulin of pigs when a substance such as phenol or m-cresol is added (British Patent No. 1,222,100). The wide isoelectric precipitation range of bovine insulin is thought to be based on the fact that bovine insulin aggregates have a pH value close to neutrality. The present invention is based on the discovery that highly purified insulin can be stabilized so that it does not form clumps that can lead to antibody formation during manufacture, in the case of long-acting drugs, during storage, or during use. Based on.

That is, the present invention is characterized in that the reaction is carried out using monomers or aggregates in which highly purified insulin is stabilized.

According to the method of the invention, highly purified insulin is reacted with a stable monomer or a loose bulk organic compound having a basic group, such as an amino group or an amino substituent.

Preferred compounds having basic groups include polypeptides, such as polyarginine, somatostatin (SOma
tOstatine), protamine or glopin, or degradation products of such basic polypeptides.

Other suitable basic compounds are 1,3-bis(4-amino-
2-methyl-6-quinolyl) monourea. According to the invention, the reaction can be easily carried out in the presence of a stabilizer such as urea. A method for producing a long-lasting insulin complex by reacting insulin with a basic substance such as protamine is widely known. In this case, highly purified insulin is used, but such insulin is obtained by recrystallization and remelting, and as a result, it has aggregated or polymerized somewhat irreversibly. Before insulin was fixed by reaction with protamine or the like, some or most of it was carried out under conditions such that some or most of it irreversibly aggregated or polymerized. However, according to the present invention, insulin is treated in advance to avoid such irreversible aggregation or dipolymerization, and the reaction with protamine etc. is also prevented from such irreversible aggregation or dipolymerization. It is carried out under conditions such that polymerization does not occur.

In short, these conditions can be overcome by the presence of stabilizers that break up insulin aggregates or depolymerize them. This invention can be used with insulin from many different animals such as pigs, cows, and sheep.

However, it has particularly important implications for bovine insulin. A preferred embodiment of the present invention is to perform insulin exchange or gel filtration on an insulin solution containing a stabilizer using an eluent containing a stabilizer9.
In addition, the above-mentioned stabilizer held insulin in the form of a stabilized monomer or a loose lump.

Then, the insulin fraction from which impurities have been removed is added to a solution containing the basic polypeptide or basic polypeptide decomposition product, and the precipitated insulin complex is separated. After reaction with protamine, insulin is fixed in a stable monomeric form and does not form antigenic aggregates or polymers. This state also exists when this complex (or complex) is placed in a known carrier for injection after separation from the stabilizer.
It is maintained even after being dispersed. This method effectively prevents the aggregation of insulin and the highly purified insulin is not initially separated with the risk of consequent aggregation. Additionally, this method is extremely simple in that long-acting insulin drugs are directly precipitated in a stable form. Example 1 250 mf1 of recrystallized bovine insulin was mixed with 7 mol of deionized urea and 0.02 mol of Tris.
PHL8. 1 of stabilized buffer in 52 ml of buffer.

This solution was mixed with 5.2 ml of 7 mol urea. The pH of this liquid mixture was adjusted to 8.1. On the other hand, a high DEAE cellulose (WhatmannDE52) with a width of 2.1C77
! Filled diameter 5? column was equilibrated with a buffer of the above composition.

Then, the above insulin solution was injected into the column, and elution was carried out at a rate of 75 ml/hour according to the schedule below. (Schedule) (a) 2.5 hours with a buffer solution of the above composition.

(b) 3 hours with a buffer solution of the above composition to which 90.0045 mol of sodium chloride per liter was added.

(c) 12 hours with a buffer solution of the above composition to which 0.011 mol of sodium chloride was added per liter.

This eluate was fractionated.

High purity fractions were collected and the insulin content was determined. In a 7 molar urea solution equivalent to the mixture of the highly purified fractions, protamine was dissolved in a proportionate amount to the highly purified insulin to form isophane insulin. This insulin solution was gradually added dropwise to the protamine solution while stirring.

As a result, an amorphous protamine-insulin complex precipitated out, probably after the urea was diluted to a 1 molar concentration. This precipitate was separated by centrifugation. 300 μ9 of this precipitate was mixed in a polyacrylamide gel with 2% amphorin (Anlph) per 6 moles of deionized urea.
As a result of analysis using an isoelectric focusing method using a solution of Ollne (PH 3 to 10), substantially only a single band appeared. Example 2 The method of Example 1 above was repeated.

Zinc chloride was added to the promemine solution at a ratio of 0.5% zinc 4 ion based on the amount of insulin, and m-cresol was added at a rate of 0.3% based on the amount of the mixture. As a result, a protamine-insulin complex was precipitated as crystals. This precipitate was centrifuged and the 3001tf! was analyzed by the isoelectric focusing method in the same manner as in Example 1 above, and as a result, a substantially single band appeared. Example 3 The same method as in Example 1 above was repeated.

However, in this case, recrystallized bovine insulin was used as the starting material in Sephadex.
) Gel filtration with G-50 was used. The resulting compound was similar to Example 1.

Example 4 The method of Examples 1 and 2 above was repeated.

However, in this case, porcine insulin was used as a starting material, obtained by salting out a 3.5 molar aqueous extract formed by adding salt at pH 8.5 in the production of insulin. Further, the obtained cake-like salt was desalted according to the conventional method before being subjected to ion exchange. The compound obtained in this example exhibited the same purity as in Example 1.

Example 5 The same method as in Example 4 above was repeated.

Further, the obtained cake-like salt was subjected to gel filtration using Sephadex G-50. The resulting product had the same purity as Example 1 above.

Example 6 The method of Examples 1 to 3 above was repeated nine times.

However, in this case, instead of bovine insulin, porcine insulin was used instead of bovine insulin. The resulting protamine monopork insulin complex had a purity similar to that of Example 1. Example 7 The procedure of Example 1 above was repeated, but in this case poly-L-arginine (degree of polymerization, 295) was used instead of protamine.

The precipitate was separated in the same manner as in Example 1. Example 8 The method of Example 1 above was repeated.

However, in this case, bis-(4-amino-2-methyl-quinolyl 6) was added to the highly purified insulin fraction.
A 0.2% aqueous solution of urea monohydrogen chloride was added in an amount equivalent to 10% by weight of insulin in the insulin fraction. This presumably precipitated the insulin complex after diluting the urea concentration to 1 molar with 0.025 molar sodium phosphate buffer, pH 7.3. This precipitate was allowed to stand for a while and then separated by centrifugation. This product was of high purity as in Example 1. Example 9 The method of Example 8 was repeated nine times.

However, in this case, Cephadex G-5 was used as the starting material.
Recrystallized bovine insulin gel-filtered at 0 was used. The resulting product exhibited a purity similar to that of Example 1. Example 10 The method of Examples 8 and 9 was repeated.

However, in this case, porcine insulin was used instead of bovine insulin. The resulting porcine insulin was of high purity as in Example 1.

Example 11 The method of Example 8 was repeated.

However, in this case the starting material used was porcine insulin obtained by salting a 3.5 molar aqueous extract formed by addition of salt at pH 8.5 in the preparation of 4-insulin. Further, the obtained cake-like salt was desalted according to the conventional method before being subjected to ion exchange. The product obtained in this example had a purity similar to that of Example 1. Example 12 A method similar to Example 11 was repeated.

In this case, the cake-like salt obtained is Sephadex G-5
Gel filtration was performed at 0. The product obtained in this example was of high purity as in Example 1.

Example 13 In order to confirm the immunity of the insulin complex obtained according to the present invention, a comparative experiment with conventional insulin was conducted using Rapid.

Insulin complex sample (1) Impurity-containing N, P, H (Neutral→ROt
aminHagedOrn) Insulin (bovine) Contains antigenic impurities as described above.

Bovine insulin obtained by recrystallization. (2) High purity N
, P, H insulin (cow) Known high-purity isosulin crystals (cow) from which antigenic impurities have been removed by column chromatography (3) N, P, H insulin (cow) according to the present invention 1C Above examples Obtained by 3.

Test method Insulin of each sample (1), (2), (3) was
Rapid was injected subcutaneously every 10 days at a dose of 0.01 u (International Units).

However, no antibody expression could be observed when no adjuvant was used. Therefore, when expressing antibodies,
It was necessary to employ commonly used immunization methods.
That is, an insulin agent emulsified in Freund's complete adjuvant was used as the first injection, and for subsequent injections, an emulsion in Freund's incomplete adjuvant was used.

Insulin antibody formation is 0rtvedAndersen
eta1 method (ActaEndOcr(Kbh)69,1
95-208, 1972) every 20 days.

As a result, in the case of N, P, H insulin in sample (1), clear expression of antibodies was observed, and in sample (2)
In the case of N, P, H insulin, slightly less antibody expression was observed.

However, in the case of sample (3) N, P, H insulin, no antibody expression was observed. Note that the N, P, and H insulins of Sample (2) and Sample (3) were highly purified by chromatography under the same conditions, and the difference between them is that the sample (
In sample 2), insulin was first separated and then reacted with protamine using a known method. The reaction with protamine was carried out in an eluent containing urea without separation. Another test method used was a more gentle immunization method.

In this case, the initial immune stimulation by bacterial kill was stopped. In all cases, injections were used in which insulin samples were emulsified in Freund's incomplete adjuvant. This was rapidly injected every 10 days at a constant dose of 201,U. Insulin antibody formation was examined every 10 days by the immunity determination method described below.

As a result, similar to the above test results, antibody formation was observed with the insulin preparations made by Jumei (Sunfur 1, Sample 2), whereas antibody formation was observed with the insulin preparations of the present invention (Sample 3).
), no antibody formation was observed. Method for determining insulin antibodies using PEG Rapid serum 100μ11
1251-insulin 100 μl (approximately 2 μ units/ml
), 100 μl of 4-nsulin solution (250 μ units/ml
) and 0.04M phosphate buffer at pH 7.4 (0.04M).
700 μl of 15 M NaCl and 0.5% human albumin was incubated at 4° C. for 48 hours.

To this was added 500 μl of 360 g/l PEG6OOO and mixed with a rotary mixer. The sample obtained in this way was
I left it for 1 hour. The above sample was rotated at 3000 revolutions per minute for 10
After centrifugation for 1 minute, the upper layer was removed by decantation and discarded, and 1251 in the precipitate was counted.

In each test, samples containing excess anti-insulin serum from guinea pigs (maximum binding) and samples containing serum from non-immunized rapids (0-sample) were also used.

The results were calculated using the following formula. (Here, T is the count number of unknown sample, TO is the count number of O-sample, TM
is the maximum join count).

Claims (1)

[Claims] 1. In a method for producing a long-lasting, low-antigenicity insulin complex by reacting insulin with an organic compound having an amine group, first, insulin from which antigenic impurities have been removed is converted into monomer or loose insulin. A method for producing an insulin complex, which comprises reacting it with an organic compound having an amine group in a state in which it is dissolved in a buffer containing a stabilizer to maintain it in the form of an aggregate, and then separating the resulting insulin complex. 2. The manufacturing method according to claim 1, wherein the insulin is bovine insulin. 3. The production method according to claim 1 or 2, wherein the organic compound is a basic polypeptide or a decomposition product thereof. 4. The manufacturing method according to claim 3, wherein the basic polypeptide is selected from polyarginine, somatostatin, protamine, and globin. 5. The manufacturing method according to any one of claims 1 to 4, wherein the stabilizer is urea.