Use of Fructosyl Peptide Oxidase for HbA1c Assay

Abstract

ARKRAY, Inc developed the world’s first automatic glycohemoglobin analyzer based on HPLC (1981). After that, ARKRAY developed enzymatic HbA1c assay “CinQ HbA1c” with the spread and diversification of HbA1c measurement (2007). CinQ HbA1c is the kit of Clinical Chemistry Analyzer, which uses fructosyl peptide oxidase (FPOX) for a measurement reaction. This report mainly indicates the developmental background, measurement principle, and future of the enzymatic method HbA1c reagent.

In 1981, ARKRAY, Inc developed the first commercial application of fully automated diagnostic equipment using high-performance liquid chromatography (HPLC) measurement principles.¹ Subsequently, in 2007, ARKRAY, Inc. developed the enzymatic method HbA1c reagent CinQ HbA1c in response to the broad adoption and diversification of HbA1c tests.^2,3 CinQ HbA1c is a clinical chemistry analyzer reagent kit that uses a measurement response to fructosyl peptide oxidase (FPOX). This technology report provides an overview of this reagent kit focusing on its developmental background and measurement principles.

Developmental Background

Increase in the Number of Diabetes Patients Worldwide

According to a survey by the International Diabetes Federation (IDF), the number of diabetes patients in the world has reached 382 million, which has a morbidity rate of 8.3%.⁴ Each year, 5.1 million people die due to diabetes, and it is estimated that the annual economic loss is $548 billion. If an effective policy is not implemented, the number of diabetes patients is projected to rise to 592 million by 2030.⁴

Diabetes policy is a common problem for all of mankind. At the May 2013 General Assembly of the World Health Organization (WHO) in Geneva, a new action plan (2013-2020) for the prevention and management of noncommunicable diseases, such as diabetes, was unanimously approved.

Diabetes and HbA1c Testing

HbA1c is a glycated protein in which glucose is attached to the N-terminal of the β-chain of hemoglobin, which reflects the average blood glucose levels for the past 1-2 months. As a result, HbA1c is widely used as an indicator of blood glucose levels in diabetes testing. The previous methods for measuring HbA1c were the HPLC assay method and the immunoassay method. Because the HPLC assay method is characterized by high precision and high accuracy, it is very effective in testing for a diagnosis of diabetes and in the treatment of diabetes. In contrast, since the immunoassay method is included in clinical chemistry analyzers, high throughput is easier to achieve with this method, a feature that can be easily applied in the measurement of large volumes of samples at the same time, for example in diabetes screening tests.

Problems With Previous Methods (HbA1c Immunoassay Method) and the Need for an H1A1c Enzymatic Assay Method

Recent progress in the treatment of diabetes increasingly requires improved reliability of HbA1c values. However, because the immunoassay method currently in use measures the change in the volume of absorption of amounts of aggregations produced by antibody-antigen reactions (immunonephelometry), measurement precision and accuracy are lower than those of the HPLC assay method. Using the immunoassay method in diabetes treatment unavoidably reduces treatment quality. The emergence of the enzymatic assay method was therefore anticipated as a measurement method to solve the drawbacks of the immunoassay method.

Measurement Principles of the HbA1c Enzymatic Assay Method

Technical Challenges in Realizing HbA1c Enzymatic Assay Method Reagents (FPOX Reactions and Hb1Ac Specificity)

HbA1c specificity in the enzymatic assay method depends greatly on the substrate specificity of FPOX.⁵ FPOX has substrate specificity for both fructosyl valyl-histidine (fructosyl-Val-His) derived from the β-chain N-terminal of hemoglobin and fructosyl amino acid. Therefore, the following possibilities exist for the realization of an enzymatic assay reagent with a high specificity for HbA1c.

a. Selectivity of glycation sites

The glycation sites of hemoglobin are the β-chain N-terminal amino acid group, the α-chain N-terminal amino acid group, and the lysine side chain amino acid group.

An HbA1c measurement reagent must measure only the glycation site of the β-chain N-terminal amino acid group and not that of the lysine side chain amino acid group.⁶ Here we have denatured the 3-dimensional structure of hemoglobin so that protease can easily act on the β-chain N-terminal amino acid group. As a result, it is possible to produce fructosyl-Val-His specifically from the β-chain N-terminal amino acid group without producing fructosyl amino acid derived from glycation of the lysine side chain.

b. Selectivity of stable HbA1c

It is necessary to measure stable HbA1c only without measuring labile HbA1c. Stable HbA1c and labile HbA1c have very similar structures. Because the FPOX used in the enzymatic assay method has a substrate specificity of not reacting to the glycated peptides derived from labile HbA1c, it is possible to measure stable HbA1c only with an enzymatic assay reagent.

c. Elimination of the influence of glycated amino acids and glycated peptides in blood and of total parenteral nutrition

Blood contains glycated amino acids and glycated peptides. Because high concentrations of amino acids and glucose are contained especially in total parenteral nutrition, total parenteral nutrition contains high concentrations of nonenzymatically produced glycated amino acids.⁷

Because preexisting glycated amino acids and glycated peptides contained in the reagent react with FPOX, these can be a source of false positive measurements. Here we have digested the preexisting glycated amino acids and glycated peptides contained in the reagent by use of FPOX as the first reagent. H₂O₂ produced in the digestion is digested by the catalase in erythrocytes and has no effect on the measurement value.

Measurement Principles of HbA1c Enzymatic Assay Reagent CinQ HbA1c

The enzymatic assay reagent CinQ HbA1c (CinQ HbA1c) measures HbA1c according to the following reaction principles. Erythrocytes are hemolyzed by mixing with water and whole blood. Glycated amino acids and glycated peptides contained in whole blood are digested by the action of the FPOX contained in the first reagent. The hemoglobin released is denatured into a special exposed hemoglobin structure by the action of a hemoglobin denaturing reagent on the hemoglobin β-chain N-terminal. Because the β-chain N-terminal is externally exposed in this special structure of hemoglobin, protease acts strongly on the β-chain N-terminal of the special structure of hemoglobin. As a result, protease has a narrow substrate selectivity and produces only fructosyl-Val-His, fructosyl-lysin-peptides, and nonglycated peptides without producing fructosyl-lysin.

FPOX has substrate selectivity for fructosyl-Val-His and fructosyl-lysin, but has no substrate selectivity for fructosyl-lysin peptides. As a result, FPOX produces hydrogen peroxide reacting only to fructosyl-Val-His derived from the β-chain N-terminal of hemoglobin. Hydrogen peroxide produces pigment in the presence of a color-forming coupler and peroxidase. The concentration of HbA1c can be measured by the absorption level of this pigment concentration. For hemoglobin, it is possible to obtain a measurement of the absorption level of the digested hemoglobin produced by the hemoglobin digestion reagent. HbA1c% is calculated from the HbA1c concentration and the Hb concentration.⁸

$$\mathrm{HbA1c}\%=\mathrm{HbA1c}\mathrm{concentration}\left(\mathrm{mmol}/\mathrm{L}\right)/\mathrm{Hb}\mathrm{concentration}\left(\mathrm{mmol}/\mathrm{L}\right)\times 1000\times 0.0915+2.15$$

Reagents for HbA1c Enzymatic Assay Reagent CinQ HbA1c

Major reagent composition of enzymatic method HbA1c “CinQ HbA1c” is as the following description.

Reagent 1 was composed of 1300 U/L FPOX, 10 KU/L Peroxidase, 4 mmol/L potassium nitrite (Hb denatured reagent), buffers.

Reagent 2 was composed of 1800 KU/L Protease, 0.030 mmol/L DA-67 (10-(carboxymethylaminocarbonyl)-3,7-bis(dimethylamino)phenothiazine), buffers

Figure 1 is an example of applying enzymatic methods HbA1c “CinQ HbA1c” to the automatic analyzer.

Figure 1.

Schematic of the enzymatic assay reagent CinQ HbA1c performed on the JCA-BM-6010 analyzer.

Whole blood are hemolyzed by purified water. Glycated amino acids and glycated peptides contained in whole blood are digested by the action of the FPOX contained in R1. The hemoglobin released is denatured into a special exposed hemoglobin structure by the action of a hemoglobin denaturing reagent in R1. The Hb concentration is measured by absorbance at optional wavelength 571 nm (main wavelength) / 694 nm (sub wavelength) at 4.5 minutes. The first measurement of HbA1c concentration is measured from DA-67 at an optional wavelength 694 nm (main wavelength) / 751 nm (sub wavelength) at 5 minutes. The Hb is resolved by the protease in R2, and fructosyl-Val-His is separated from Hb in about 3 minutes. FPOX produces hydrogen peroxide from the fructosyl-Val-His. hydrogen peroxide reacts with DA-67 via Peroxidase. The second measurement of HbA1c concentration is measured from DA-67 at an optional wavelength 694 nm (main wavelength) / 751 nm (sub wavelength) at 10 minutes. The HbA1c concentration is determined from the difference between the first measurement value and the second measured value.

HbA1c% is calculated from the HbA1c concentration and the Hb concentration.

Reagent Performance

HbA1c Selectivity

Selectivity for Glycation Sites (see Figure 2)

Figure 2.

Selectivity of HbA1c enzymatic assay method for GHbLys. Three samples with 65 g/L glucose were 0, 6, 12 hours incubation at 37°C, respectively. Three samples were measured by the enzymatic methods HbA1c “CinQ HbA1c” (line with dots), GHbLys amount (line with squares). The HbA1c enzymatic assay does not react with GHbLys, which is a glycation product of lysine side chains of hemoglobin.

Selectivity for glycation sites was determined by measurement of a sample incubated for 0, 6 and 12 hours at 37°C after the addition of 65 g/L glucose to whole blood. The amount of glycated lysine side chains of amino acids of hemoglobin (GHbLys) was determined after measurement of the amount of hydrogen peroxide produced by the action of fructosyl amino acid oxidase on the reaction of fructosyl-Lys and fructosyl-lysine peptides after sufficient protease digestion of the above mentioned sample. The amount of GHbLys increased in the incubation period of the initial stage from 5.4% to 8.8%, but the measured values of the HbA1c enzymatic assay method did not increase after incubation from 5.35% in the initial period. As a result, the HbA1c enzymatic assay method did not measure the glycation of lysine side chains of amino acids.

Influence of Labile HbA1c (See Figure 3)

Figure 3.

Reactivity of HbA1c enzymatic assay method to labile HbA1c. Glucose concentration in the 3 samples was 0, 7.5, and 15 g/L, respectively. Three samples were measured by the enzymatic methods HbA1c “CinQ HbA1c” (line with dots); the amount of labile HbA1c was measured with ADAMS A1c HA-8181 (line with triangles). The HbA1c enzymatic assay method did not react to labile HbA1c, which is an intermediate in the synthesis of HbA1c and is characterized by the reversible binding of glucose to hemoglobin β-chain N-terminal as a Schiff base.

The influence of labile HbA1c was determined by measuring HbA1c in whole blood samples from healthy persons with the addition of 0, 7.5, and 15 g/L glucose and incubated for 4 hours at 37°C. The amount of labile HbA1c was measured with HPLC assay method using an ADAMS A1c HA-8181. The amount of labile HbA1c increased with the glucose concentration from 1.8% in the initial period to 5.4%, while the values measured with the HbA1c enzymatic assay method did not increase beyond 4.7% in the initial period, regardless of the glucose concentration. As a result, the HbA1c enzymatic assay method did not measure labile HbA1c. This result is consistent with previously reported results.³

Influence of Glycated Amino Acids (See Figure 4)

Figure 4.

Influence of total parenteral nutrition on HbA1c enzymatic assay method Samples were whole blood with the addition of 0%, 1%, 3%, 5%, and 15% of AMICALC (diamonds), UNICALIC N (squares), UNICALIC L (triangles). The samples were measured in the enzymatic methods HbA1c “CinQ HbA1c.” The HbA1c enzymatic assay was not influenced by preexisting glycated amino acids contained in whole blood.

The influence of glycated amino acids was determined by measurement of samples of whole blood with the addition of 0%, 1%, 3%, 5%, and 15% total parenteral nutrition (AMICALC, UNICALIC N, UNICALIC L, all products of the TERUMO Corporation, Tokyo, Japan and Mitsubishi Tanabe Pharma Corporation, Osaka, Japan) in which glucose and amino acids were dissolved in the same solution.

Even with the addition of 15% total parenteral nutrition, which included glycated amino acids, glycated amino acids had no influence. This result is consistent with previously reported results.²

Basic Performance

The basic performance of CinQ HbA1c is shown below. The measurement equipment used was the JEOL BioMajesty BM-6010 (JEOL Ltd., Tokyo, Japan).

Relation to Measured Values of Previous Methods (See Figure 5)

Figure 5.

Correlation between measured values using the enzymatic assay method and immunoassay method. HbA1c values obtained with the HbA1c enzymatic method “CinQ HbA1c” (y-axis) were significantly correlated with those obtained with the Rapidia-auto HbA1c immunoassay (x-axis). r = 0.9975, y = 1.010× – 0.23, n = 80.

The correlation between the HbA1c enzymatic assay method and the Rapidia-auto HbA1c immunoassay (Fujirebio Inc., Tokyo, Japan) are shown. Eighty hemocyte samples were used as the sample from diabetes patients. Results of the comparative regression analysis of the enzymatic assay method (y-axis) and immunological assay method (x-axis) are as follows:

$$\mathrm{r}=0.9975,\mathrm{y}=1.010\mathrm{x}-0.23$$

It has been reported that the HbA1c enzymatic assay method shows a positive correlation with the HPLC assay method, and that both whole blood samples and hemocyte samples may be used.²

Reproducibility (See Table 1)

Table 1.

Within-Run Precision and Between-Day Precision.

Within run	Low	Middle	High	Between day	Low	Middle	High
Average	4.61	6.42	10.36	Average	4.70	6.93	10.81
SD	0.01	0.01	0.01	SD	0.05	0.04	0.04
CV%	0.15	0.20	0.09	CV%	0.96	0.52	0.34

Within-run precision was calculated from 20 measurements of whole blood samples from 3 persons. Between-day precision was calculated by measurement of 3 freeze-dried controls at a 20-day interval. The HbA1c enzymatic assay method had excellent within-run precision and between-day precision, with both below CV 1%.

Within-run precision was calculated from 20 measurements of samples hemolyzed from whole blood samples from 3 persons. CV% was below 0.20%.

It has been reported that reproducibility was the same even when whole blood was hemolyzed for each measurement.² Between-day precision was calculated by measurement of 3 freeze-dried controls at a 20-day interval. CV% was below 1.0%. In addition, when using a Hitachi 7180 as the measurement device, both within-run precision CV% and between-day precision CV% were stable, below 1.1% and 1.5%, respectively.³

Linearity (See Figure 6)

Figure 6.

Linearity of the HbA1c enzymatic assay method. Assay linearity was assessed by comparing the observed analytical values obtained for the proportional mixtures of the 2.7% and 15.6% samples. Linear regression analysis gave a relationship of y = 1.000 × 0.06% (r = .9999).

Linearity was determined by measurement of whole blood with a low (2.7%) HbA1c concentration and whole blood with a high (15.6%) HbA1c concentration. The relational equation is as follows:

$$\mathrm{y}=1.000\mathrm{x}-0.06,\mathrm{r}=0.9999.$$

The HbA1c enzymatic assay method had excellent linearity within the range of 3% to 15% concentration. This result is consistent with results previously reported.^2,3

Conclusion

Results of Development of HbA1c Enzymatic Assay Reagent

The calibration line of the previous immunoassay method is a curve based on 4-5 calibration measurement values. The curve of the calibration line varies arbitrarily based on calibration error. The immunoassay method measures the process of antigen-antibody reactions. Measured values can be easily influenced by disturbances external to the measurement such as measurement temperature or sample condition. As a result, between-run precision of the previous immunoassay method had a CV of 0.9%-2.6% and exceeded CV2%.^9,10

However, the calibration line of the enzymatic assay method is a direct line based on 2 calibration measurement values. The calibration line has a stabilized shape for all measurement regions. Because measurement is done after the reaction is completed, the enzymatic assay method is resistant to the influence of factors external to the measurement. The HbA1c enzymatic assay method is able to increase measurement precision in comparison with the previous immunoassay method and has realized CV below 1.0% (see Table 1). As a result of this, the HbA1c enzymatic assay method has made it possible to measure a high volume of samples with high accuracy using a clinical chemistry analyzer.

Future HbA1c Enzymatic Assay Method Reagents

High accuracy is required for the diagnosis of diabetes by HbA1c measurement. For this reason, the HPLC assay method, which has this accuracy, is being used in the diagnosis of diabetes.

The current enzymatic assay method uses FPOX after digestion of hemoglobin by protease. This is because FPOX does not act on glycated peptides consisting of 3 or more amino acids. Recently, a modified FPOX has been discovered that acts on glycated peptides consisting of 6 or more amino acids.¹¹

In the future, FPOX could be modified to act directly on the glycated β-chain N-terminal of hemoglobin. This kind of FPOX would be able to eliminate the measurement error in the protease reaction step. Development of this kind of technology will bring us closer to realizing the level of accuracy of the enzymatic assay method required in the diagnosis of diabetes.

We speculate that FPOX to react directly to the glycated β-chain N-terminal of hemoglobin has a wide substrate specificity.

Therefore, the reagent using the new FPOX has high specificity to HbA1c; developers must be careful by glycosylation site selectivity, influence of labile HbA1c, and influence of glycated amino acids and glycated peptides (see Figures 1, 2, and 3).

Future Possibilities for Enzymatic Assay Method Technology

The enzymatic method described above is a method that allows highly accurate measurement of HbA1c as described above. To date, POCT HbA1c measurement systems have been low accuracy systems because they used the immunoassay method. As a result, POCT HbA1c measurement systems have been used only within a limited scope.¹² The enzymatic method will improve the accuracy of POCT systems. In addition, an FPOX acting directly on the glycated β-chain N-terminal of hemoglobin is likely to have high accuracy. We believe that the enzymatic method will also promote the use of POCT HbA1c measurement systems.