WO1983003476A1

WO1983003476A1 - Ion-exchange system and method for isolation and determination of glycosylated hemoglobin in human blood

Info

Publication number: WO1983003476A1
Application number: PCT/US1982/000288
Authority: WO
Inventors: James Lynn Sanders
Original assignee: James Lynn Sanders
Priority date: 1982-03-25
Filing date: 1982-03-25
Publication date: 1983-10-13
Also published as: EP0104165A1

Abstract

An ion-exchange system and method is provided for isolating glycosylated hemoglobin (HB Al) from other hemoglobins in human blood together with a quantitative determination of glycosylated hemoglobin. Separation of the glycosylated hemoglobin is accomplished by the ion-exchange system with little temperature dependence in the range of from about 15<o>-37<o>C and without rigid control of pH and ionic strength. The ion-exchange system lowers the pH of the human blood to about pH = 6.9 through the use of an organic buffer preferably a zwitterionic buffer. The ion-exchange system also contains a cation-exchange resin. The preferred composition of the ion-exchange system contains about 0.05 molar 3-(N-morpholino) propanesulfonic acid as the buffer and carboxy-methyl dextran as the ion-exchange resin present in an amount of from about 30 milliequivalents to about 50 milliequivalents of binding capacity per liter thereof.

Description

I. ION-EXCHANGE SYSTEM AND METHOD FOR ISOLATION AND DETERMINATION OF GLYCOSYLATED HEMOGLOBIN IN HUMAN BLOOD

II. DESCRIPTION OF INVENTION A. BACKGROUND OF THE INVENTION

This invention relates to the selective separation of glycosylated hemoglobin (Hb Al ) from non-glycosylated hemoglobin in human blood. Another aspect of this invention relates to the selective separation of non-glycosylated hemoglobin from human blood using an ion-exchange system which does not require rigid control of pH and ionic strength and which shows little temperature dependence in the range of from about 15°-37°C. Still another aspect of this invention relates to a method for separating glycosylated hemoglobin from non-glycosylated hemoglobin and quantitatively determining the fractional amount of glycosylated hemoglobin present in human blood through use of a reference material prepared from human blood.

Throughout the circulatory life of the human red cell, glycosylated hemoglobin is formed continuously by the adduction of glucose to the N-terminal of the hemoglobin beta chain. This process, which is non-enzymatic, reflects the average exposure of hemoglobin to glucose over an extended period. Several classical studies have shown that glycosylated hemoglobin in diabetic subjects can be elevated 2-3 fold over the levels found in normal individuals. (Trivelli, L.A., et al., 1971, New Eng. J. Med. 284: 353; Gonen, B. , and Rubenstein, A.H., 1978, Diabetolcgia 15: 1; and Gabbay , K.H., et al., 1977, J. Clin. Endocrinol. Metab. 44: 859). These cited investigators have recommended that glycosylated hemoglobin serve as an indicator of diabetic control since the glycosylated hemoglobin levels approach normal values for diabetics responding to treatment. Historically, fasting plasma glucose and urinary glucose tests have been employed as measures of diabetic control. The use of glycosylated hemoglobin determinations offers several advantages over these methods: 1) the glycosylated hemoglobin level is unaffected by the ingestion of a recent meal; 2) it appears quite stable in the bleed; and 3) it reflects the average blood glucose level over an extended period (3-4 weeks) rather than at a single time point, thus providing a better criterion of diabetic control. Thus, an accurate, reproducible and dependable in vitro quantitative test for glycosylated hemoglobin is important in medicine for the clinical management of diabetic patients.

Glycosylated hemoglobin has been defined operationally as the fast fraction hemoglobins (Hb Ala, Alb, Ale) which elute first during column chromatography with cation-exchange resin. The nonglycosylated hemoglobin, which consists of the bulk of the hemoglobin, remains attached to the resin and can be removed by lowering the pH or raising the ionic strength of the εluting buffer. In the past, a carboxy derivative of cellulose or polystyrene has been ccmmonly employed as the ion-exchange resin. Elution of the glycosylated hemoglobin was accomplished by use of a phosphate buffer containing cyanide. These methods are disclosed in Trivelli, L.A., et al., 1971, New Eng . J. Med. 284: 353; and Gabbay, K.H., et al., 1977, J. Clin. Endocrinol. Metab. 44: 859. Other methods used to separate glycosylated hemoglobin include high-performance cation-exchange chromatography and electrcphoresis. These latter procedures require expensive equipment and usually prove tec slew and cumbersome for practical uses.

Difficulties can occur in practicing these methods for separating glycosylated hemoglobin from non-glycosylated hemoglobin. For proper quantitative determinations, the composition, pH and ionic strength of the eluting buffer must be maintained within narrow limits. More importantly, temperature control is critical, usually being confined to a range of 21º-24ºC. Such rigid limitations on a procedure can cause operating difficulties beyond the capacibilities of many clinical laboratories. Furthermore, substantial amounts of cyanide are used in the eluting buffer and represent a hazard to the user. These problems are described by Simon, M., and Eissler, J. , 1980, Diabetes 29: 467; Rand, P.B., and Nelson, C. 1980, Clin. Chem. 26 : 1209; and Schellekens, A. P.M., et al., 1981, Clin. Chem. 27: 94.

Thus, a need exists for an ion-exchange system and method for determining glycosylated hemoglobin in human blood which offers ease of handling, has little temperature dependence, does not require rigid control of pH and ionic strength, and uses a minimal amount of cyanide.

B. SUMMARY OF THE INVENTION

According to the invention, an ion-exchange system and method for separating glycosylated hemoglobin from non-glycosylated hemoglobin and a method for quantitative determination of the glycosylated hemoglobin is provided. The ion-exchange system for selectively binding non-glycosylated hemoglobin in human blood contains a cation-exchange resin an a zwitteriαnic buffer having a pH of from about 6.4 to about 7.2 and a concentration of from about 0.02 molar to about 0.1 molar. Preferably, the ion-exchange system contains about 0.05 molar 3-( N-morpholino) propanesulfonic acid and carboxymethyl dextran as the ion-exchange resin present in an amount of from about 30 milliequi-valents to about 50 milliequivalents of binding capacity per liter thereof. According to the invention, a lysed preparation of human blood is added to the ion-exchange system and mixed, causing the non-glycosylated hemoglobin to bind to the ion-exchange resin. The glycosylated hemoglobin remains free in solution. The preferred pH of the ion-exchange system is about pH = 6.9. The ion-exchange system lowers the pH of the blood to about pH = 6.9 and causes the binding of non-glycosylated hemoglobin without binding glycosylated hemoglobin. The solution containing glycosylated hemoglobin is separated from the resin containing non-glycosylated hemoglobin by filtration. The fractional amount of glycosylated hemoglobin present in human blood is determined by comparing the absorbance of the glycosylated hemoglobin fraction at a particular wavelength with the absorbance of a diluted sample of the lysed human blood. The use of a reference material prepared from human blood and containing a known amount of glycosylated hemoglobin facilitates determination of the unknown concentrations in human blood.

The ion-exchange system and method is dependable, accurate and reproducible. Furthermore, rigid control of pH and ionic strenght is not required, and there is little temperature dependence in the range αf from about 15º-37ºC.

C. DETAILED DESCRIPTION OF THE INVENTION

The ion-exchange system of the subject invention contains a cation-exchange resin and a zwitterionic buffer having a pH of from about 6.4-7.2 and a concentration of from about 0.02-0.1 molar.

The preferred buffer of the subject invention is 3-(N-morpholino) propanesulfonic acid (MOPS) at a concentration of about 0.05 molar. MOPS is a zwitterionic buffer havin a pKa of about 7.20 at 20°C and a useful buffering range from about pH 6.4 to 7.9. Although the preferred buffer is MOPS, the ion- exchange system of the subject invention works effectively with several zwitterionic buffers having a pKa in the range of from about 6.6-7.5 at 20°C and having a concentration of from about 0.02-0.1 molar. These buffers include N-2-acetamidoiminαdiacetic acid; N-2-acetamido-2-aminoethanesulfonic acid; piperazine-N,N'- bis-2-ethanesulfonic acid; N,N'-bis-(2-hydroxyethyl-2-aminoetha nesulfonic acid; and 2-[tris-(hydroxymethyl)methyl]aminoethane sulfonic acid. For separation of glycosylated hemoglobin from non-glycosylated hemoglobin, the preferred system pH should be about pH = 6.9. The buffer is used to lower the blood pH to about pH = 6.9. Use of a zwitterionic buffer offers several advantages over conventional ionic buffers because buffer interaction with proteins is small, ionic strength is easily controlled, and pH shifts with ermperature changes are minimized.

The preferred cation-exchange resin of the subject invention is carboxymethyl dextran with a binding capacity of from about 4.0-5.0 milliequivalents per gram of resin. The dextran is cross-linked and beaded to form a particle of from about 40-120 microns in diameter. For purposes of the method, the preferred amount of the resin in the ion-exchange system is about 40 milliequivalents of binding capacity per liter thereof. Although the preferred cation-exchange resin is carboxymethyl dextran, the ion-exchange system of the subject invention works effectively with several other cation-exchange resins having similar binding properties, including sulfopropyl dextran, carboxymethyl cellulose, carboxy cellulose, carboxymethyl agarose and carboxy polystyrene.

The combination of zwitterionic buffer and cation-exchange resin allows a rapid and effective separation of glycosylated hemoglobin from non-glycosylated hemoglobin. Because of the use of the zwitterionic buffer, the ion-exchange system of the subject invention does not require rigid control of pH and ionic strength and has little dependence on temperature in the range αf from about 15º-37ºC.

Preservatives can be employed to assist in stabilizing the ion-exchange system at room temperature. The preferred preservative is boric acid present in a concentration of about 0.01 molar. Boric acid acts to inhibit microbial growth.

The preffered method of the present invention for the determination of glycosylated hemoglobin includes the following steps and the total test time requires about 15 minutes. About 0.1 milliliters of well-mixed, whole blood is added to about 0.5 milliliters of a lysing agent comprised of about 0.25% polyoxyethylene octyl phenol in water. The polyoxyethylene octyl phenol is a surfactant which acts to disrupt the cell membrane and causes the release of hemoglobin, thus forming a lysate. The preferred molecular weight of the polyoxyethylene octyl phenol is about 650 daltons. Although other volumes can be used, the ratio of lysing agent to whale blood should be approximately constant at 5. Potassium cyanide should also be included in the lysing agent if the whale blood contains signifioant amounts αf methemoglobin. Glycosylated methemoglobin has ion-exchange binding properties which differ from those of the usual glycosylated oxyhemoglobin. The glycosylated methemoglobin binds to the ion-exchange resin causing a falsely low result for the glycosylated hemoglobin determination. Cyanide complexes with methemoglobin to form cyanmethemoglobin. Glycosylated cyanmethemoglobin has ion-exchange properties essen¬tially the same as the binding properties of glycosylated oxyhemoglobin. The inclusion of cyanide in the lysing agent then converts glycosylated methemoglobin to glycosylated cyan- methemoglobin and the correct result is obtained for the glycosylated hemoglobin determination. The preferred concentration of potassium cyanide in the lysing agent is about 0.01 molar.

For separation of glycosylated hemoglobin from non-glycosylated hemoglobin, about 0.1 milliliters of the lysate is added to about 3.0 milliliters of the ion-exchange system and the combined system is mixed for about 5 minutes. Upon addition to the ion-exchange system, the lysate pH is lowered to about pH = 6.9. At about pH = 6.9, an electrostatic charge difference exists between glycosylated hemoglobin and non-glycosylated hemoglobin, which allows the cation-exchange resin to bind non- glycosylated hemoglobin while leaving glycosylated hemoglobin free in solution. The use of MOPS as the buffer in the preferred method allows pH control at different temperatures. Use of the preferred ion-exchange system assures a fast and effective separation of glycosylated hemoglobin from non-glycosylate hemoglobin in the temperature range of from about 15º-37°C. The resin is separated from the surrounding solution by filtration. Most preferably, the ion-exchange system is filtered with a porous-plastic serum filter capable of retaining the resin. (Such filters are available from Glasrock Products, Inc., Fair-burn, Georgia). The filtered solution contains glycosylated hemoglobin while the ion-exchange resin retains the non-glycosylated hemoglobin.

The method described herein is preferred for assaying the amount of glycosylated hemoglobin, although any established method for determining hemoglobin may be used. Hemoglobin, both glycosylated and non-glycosylated, absorbs quite str ongly in the Soret Band wavelength region of from about 400 nm to about 440 nm. For purposes of the method, absorbance measurements for hemoglobin are made at the preferred wavelength of 415 nm.

The preferred means for expressing the analytical results for glycosylated hemoglobin is as the percent of total hemoglobin - i.e., glycosylated plus non-glycosylated. The absorbance at 415 nm for the glycosylated hemoglobin is made directly on the filtered solution containing glycosylated hemoglobin. The absorbance at 415 nm for total hemoglobin is made on a diluted sample of the blood lysate, prepared by adding about 0.02 milliliters of the blood lysate to about 5.0 milliliters of deionized water and mixing well. The glycosylated hemoglobin as percent of total hemoglobin is then determined by calculating the ratio of absorbances at 415 nm for the glycosylated hemoglobin to the total hemoglobin and comparing the ratio to that of a reference material which is also carried through the separation procedure. The reference material is a stable preparation of human blood and contains a known amount of glycosylated hemoglobin . (Such reference material is available from Sandare Chemical Company, Dallas, Texas). The following equation is used:

The use of a reference material in the method adjusts for any inaccuracies in volume dispensing and compensates for whatever temperature dependence that might be present.

The method of the subject invention shows linearity in the range of 5% to 20% glycosylated hemoglobin. Bloods having a total hemoglobin concentration exceeding 180 grams per liter should be diluted two-fold with deionized water before assay.

Sensitivity of the method indicates a change of about 0.02% glycosylated hemoglobin for every change of 0.001 absorbance units. The final separation fractions appear quite stable, but absorbance measurements should be made within 1 hour of separation before evaporation of the samples becomes significant.

EXAMPLE 1 The ion-exchange system and method of the invention was used to determine the expected values for glycosylated hemoglobin in a non-diabetic population. One hundred subjects were used in the study. These individuals had normal blood glucose values and no history of diabetes. The ion-exchange system contained about 0.05 molar 3-(N-morpholino)propanesulfonic acid and carboxymethyl dextran present in an amount of about 40 milliequivalents of binding capacity per liter thereof. The ion-exchange system also contained boric acid as a preservative present in an amount of about 0.01 molar. About 0.1 milliliters of well-mixed, whole blood was added to about 0.5 milliliters of a lysing agent comprised of about 0.25% (v/v) polyoxyethylene octyl phenol in water to prepare a lysate. The lysing agent also contained about 0.01 molar potassium cyanide to convert any methemoglobin to cyanmethemoglobin. About 0.1 milliliters of the lysate was added to about 3.0 milliliters of the ion-exchange system and the system was mixed for about 5 minutes. The ion-exchange resin was then separated from the surrounding solution by filtering through a porous-plastic serum filter. A spectrophotometer calibrated to read absorbance at 415 nm was zeroed using deionized water as the blank, and the absorbance of the filtered solution was then determined. The absorbance of the total hemoglobin fraction was made by diluting about 0.02 milliliters of the lysate with about 5.0 milliliters of deionized water and measuring the diluted sample against water as the blank. The glycosylated hemoglobin concentration, expressed as percent of total hemoglobin, was determined by calculating the ratio of the glycosylated hemoglobin absorbance to the total hemoglobin absorbance and comparing the ratio to that of a reference material containing a known amount of glycosylated hemoglobin and carried through the separation. The glycosylated hemoglobin values cf the normal subjects ranged from 6.4% to 8.7%.

EXAMPLE 2 The preferred ion-exchange system and method were used to establish the expected values range for a diabetic population. The bloods of 42 individuals diagnosed as diabetic and who were receiving medication for this condition were analyzed using the ion-exchange system and method set forth in Example 1. The glycosylated hemoglobin values for the diabetic subjects ranged from 7.5% to 14.8%, with a mean value of 10.7%. Five of the 42 diabetic subjects had glycosylated hemoglobin values which were within the observed normal range - i.e., 8.7% or below. These five individuals had bleed glucose levels close to normal, indicating well-managed treatment. The correlation coefficient was 0.71 between fasting glucose levels and the glycosylated hemoglobin values. Thus, the clear separation of glycosylated hemoglobin expected values for the diabetic and non-diabetic populations, and the general agreement between fasting glucose levels and glycosylated hemoglobin values in the diabetic population, indicate the usefulness and the efficacy of the invention.

EXAMPLE 3 Within run reproducibility of the preferred ion-exchange system and method was determined by conducting separation of glycosylated hemoglobin twenty times each for normal and diabetic bloods using the ion-exchange system and method set forth in Example 1. The following results were obtained:

TYPE MEAN STD DEV %CV

NORMAL 7.8 0.21 2.7

DIABETIC 13.4 0.23 1.7

EXAMPLE 4 Run to run reproducibility of the preferred ion-exchange system and method was determined by conducting separation of glycosylated hemoglobin for ten successive runs for both normal and diabetic bloods using the ion-exchange system and method set forth in Example 1. The following results were obtained: TYPE MEAN STD DEV %CV

NORMAL 7.6 0.31 4.1

DIABETIC 13.0 0.60 4.6

EXAMPLE 5 Temperature dependence of the preferred ion-exchange system and method was determined by conducting separation of glycosylated hemoglobin from normal and diabetic bloods at temperatures of 15°, 24°, 30° and 37°C using the ion-exchange system and method set forth in Example 1. The results showed an average difference of 0.4% glycosylated hemoglobin for separations carried out over the temperature range of 15°-37°C, thus establishing the little temperature dependence of the ion-exchange system.

This invention has been described in detail with reference to its preferred embodiments, and many modifications will now be apparent to those skilled in the art and those modifications are intended to be within the scope of the appended claims

Claims

CLAIMSWHAT IS CLAIMED IS:

1. An ion-exchange system for selectively separating glycosylated hemoglobin from non-glyccsylated hemoglobin in human blood which is comprised of: a) an organic buffer having a pH cf from about 6.4-7.2 and a concentration of from about 0.02-0.1 molar; and b) a cation-exchange resin.

2. The ion-exchange system as recited in Claim 1 wherein the organic buffer is selected from the group of zwitterionic buffers derived from ethanesulfenic acid, propanesulfonic acid, and imincdiacetic acid.

3. The ion-exchange system recited in Claim 2 wherein the cation-exchange resin is selected from the group consisting of carboxymethyl, carboxy and sulfopropyl derivatives of dextran, cellulose, agarose and polystyrene.

4. An ion-exchange system for selectively binding non-glycosylated hemoglobin inhuman blood which comprises: a) a zwitterionic buffer having a useful pH buffering range of from about pH = 6.4 to about pH = 7.2, and a pKa of from about 6.6 to about 7.5 at 20°C and present in a concentration of from about 0.02 molar to about 0.1 molar; and b) a cation-exchange resin present in an amount of from about 30 milliequivalents to about 50 milliequivalents of binding capacity per liter thereof.

5. The ion-exchange system as recited in Claim 4 wherein: a) the zwitterionic buffer is 3-(N-morpholinc)propanesulfcnic acid; and b) the cation-exchange resin is carboxymethyl dextran.

6. The ion-exchange system in Claim 5 wherein: a) 3-(N-morpholino)propanesulfonic acid is present at a concentration of about 0.05 molar; and b) the carboxymethyl dextran is present in an amount of about 40 milliequivalents of binding capacity per liter thereof.

7. A method of determining the amount of glycosylated hemoglobin in human blood without rigid ccntrol of pH and ionic strength and with little temperature dependence which comprises: a) preparing a lysate of human bleed; b) adjusting the pH of said lysate to from about pH = 6.4 to about pH = 7.2 with an organic buffer; and c) introducing a cation-exchange resin into said lysate and thus binding the non-glycosylated hemcglobin.

8. The method as recited in Claim 7 wherein: a) the concentration of the organic buffer in the lysate is from about 0.02 molar to about 0.1 molar; and b) the cation-exchange resin is present in the lysate in an amount of from about 30 milliequivalents to about 50 milliequivalents of binding capacity per liter thereof.

9. The method as recited in Claim 8 further comprising: a) mixing the said cation-exchange resin, said organic buffer and said lysate; b) separating the said resin from the surrounding solution; c) determining the absorbance at 415 nm of said separated solution; d) preparing a dilution of said lysate in water and determining the absorbance at 415 nm; e) calculating the ratio of the absorbance cf the said separated solution to the absorbance of the said diluted lysate; f) performing the ion-exchange separation of glycosylated hemoglobin according to Claim 9 on a reference material prepared from human blood and containing a known amount of glycosylated hemoglobin; g) calculating the ratio of the absorbance of the separated solution to the absorbance of the diluted lysate for the said reference material; h) calculating the glycosylated hemoglobin concentration in said blood by dividing the ratio of absorbances cf said blood lysate by the ratio of absorbances of said reference material and multiplying by the glycosylated hemoglobin concentration of the reference material.

10. The method as recited in Claim 8 wherein the pH of the blood lysate is adjusted and a cation-exchange resin is introduced by the addition of a single ion-exchange system comprising: a) a zwitterionic buffer having a useful pH buffering range of from about 6.4 to about 7.2 and present in a concentration of from about 0.02 molar to about 0.1 molar; and b) a cation-exchange resin present in an amount of from about 30 milliequivalents to about 50 milliequivalents of binding capacity per liter thereof .

11. The method as recited in Claim 9 wherein: a) the zwitterionic buffer is 3-(N-morpholino)propanesulfonic acid; and b) the cation-exchange resin is carboxymethyl dextran present in an amount of about 40 milliequivalents of binding capacity per liter thereof.

12. The method as recited in Claim 10 wherein a lysate of said human bleed is utilized.

13. The method as recited in Claim 10 wherein the said bleed lysate is mixed with the said ion-exchange system for about 5 minutes.

14. The method as recited in Claim 10 wherein the separation of said resin from said surrounding solution includes filtration.

15. The method as recited in Claim 10 wherein said blood lysate is diluted with water for absorbance measurement.

16. The method as recited in Claim 10 wherein the glycosylated hemoglobin concentration in human blood is determined by comparison to a reference material prepared from human blood and containing a known amount of glycosylated hemoglobin.

17. The method as recited in Claim 16 wherein the glycosylated hemoglobin of said reference material is separated according to Claim 10.