Validation of Analytical Method for Determination of Adenine and

Hypoxanthine Purine Bases in Melinjo Chips by HPLC-UV

Hanifah Nuryani Lioe, Dahrul Syah, Mutiara Pratiwi and Annisa Defriana

Department of Food Science and Technology, Faculty of Agricultural Engineering and Technology,

IPB University (Bogor Agricultural), IPB Darmaga, Bogor 16680, Indonesia

Keywords: Validation, Analytical Method, Purine, Melinjo Chips, HPLC-UV.

Abstract: Melinjo chips, which is commonly consumed by Indonesian people, are considered as one of the causes which

triggers gout disease due to its purine content. The method to analyze purine in food is limitedly known by

food laboratories in Indonesia. The objective of this research was to validate the analytical method for purine

bases determination in melinjo chips by HPLC-UV. Adenine and hypoxanthine were of the known purine

bases and chosen to be analyzed due to their characteristics which cause more uric acid accumulation in the

body rather than other purine bases, guanine and xanthine. Guanine and xanthine were insoluble in the mobile

phase used in this study, so that they might not be able simultaneously analyzed with adenine and

hypoxanthine. Adenine and hypoxanthine standards were used in the instrumental performance experiment,

method linearity and recovery test. The results showed that the HPLC-UV instrument with RP-C18 column

and UV 257 nm detection had a good linearity in the concentration range of 7.81–125.00 µg/mL. The

coefficients of determination (R2) were more than 0.999 for both adenine and hypoxanthine. Adenine and

hypoxanthine were detected by HPLC-UV at retention time of 5.9–6.8 and 4.8–5.5 min respectively, and both

retention times had an acceptable precision, less than 2.0 %. Detection limit (LOD) and quantification limit

(LOQ) of the instrument were found at 0.72 and 2.39 µg/mL for analysis of adenine, while for analysis of

hypoxanthine were at 0.69 and 2.30 µg/mL, respectively. The analytical method showed a good linearity at a

concentration range of 50–800 µg/g sample with R2 more than 0.990 for both adenine and hypoxanthine

analysis. Method detection limit (MDL) of adenine and hypoxanthine analysis was 19.44 and 14.42 µg/g

respectively. Accuracy of the method was determined by a recovery test at spiking concentrations of 100,

500, and 1000 µg/g. In the analysis of adenine, the respective recovery results were 79.33%, 89.39%, and

90.37% with respective precisions were 5.19%, 4.50%, and 3.46%. While in the analysis of hypoxanthine,

the recovery results were 66.75%, 92.29%, and 100.15%, and the precisions were 2.98%, 3.15%, and 2.22%,

respectively. Based on these results, the analytical method for determination of purine bases in melinjo chips

has been validated and was found to be accurate at concentration more than 100 µg/g wet weight of sample.

1 INTRODUCTION

Purines consist of adenine and guanine found in

deoxyribonucleic acid (DNA) and ribonucleic acid

(RNA), whereas hypoxanthine and xanthine are

purine-derived natural compounds that are rarely

found as bases in DNA and RNA, but often act as

important intermediate compounds in the process of

formation and breakdown of nucleotides (Garret

2005). Since purines can be synthesized and reused

by human body, the need for purines from food is

very small. Purines from food that are absorbed by the

body but are not needed will be catabolized to

produce the final product of uric acid (Zöllner 1982).

In the research of Kaneko et al. (2014), total purines

in food were described as follows: in cereals 157 ̶ 759

µg/g, beans 188 ̶.776 µg/g, soybean products 200 ̶

2931 µg/g, dried seaweeds 154 ̶ 5917 µg/g, eggs not

detected (nd = <2 µg/g), dairy products nd ̶ 129 µg/g,

mushrooms 69 ̶ 3795 µg/g, fruits 24 ̶ 35 µg/g, beef

meat 774 ̶ 1064 µg/g, chicken meat 700 ̶ 1539 µg/g,

pork meat 814 ̶ 1197 µg/g, and fish meat 669 ̶ 2114

µg/g.

Gout or also known as gouty is a condition of the

accumulation of uric acid crystals in the joints. The

accumulation occurs due to the excess of uric acid

production or suboptimal excretion of uric acid as a

product of purine catabolism. According to the

118

Lioe, H., Syah, D., Pratiwi, M. and Defriana, A.

Validation of Analytical Method for Determination of Adenine and Hypoxanthine Purine Bases in Melinjo Chips by HPLC-UV.

DOI: 10.5220/0009978300002833

In Proceedings of the 2nd SEAFAST International Seminar (2nd SIS 2019) - Facing Future Challenges: Sustainable Food Safety, Quality and Nutrition, pages 118-126

ISBN: 978-989-758-466-4

research results of Clifford et al. (1976a) and Clifford

and Story (1976b), adenine and hypoxanthine show a

greater impact on gouty or uric acid levels, compared

to guanine and xanthine.

The analytical method for purine base analysis in

food has rarely developed by food laboratories in

Indonesia. Melinjo (Gnetum gnemon L.) is

stigmatized to cause uric acid due to its purine

content, however a study on serum uric acid, after

consuming the product of melinjo (fried melinjo

chips), mentioned it didn’t raise the uric acid level

(Saifudin et al 2018). Considering that melinjo chips

are generally consumed by Indonesian people, an

analysis of purine content in melinjo chips is

important to give an information for consumers of the

product. Thus, there is a need for a validated

analytical method to analyze the purine content in

melinjo chips.

Purines are known able to be analyzed by high

performance liquid chromatography (HPLC)

instrument with UV detection in ppm levels (mg/kg

or µg/g). The aromatic ring functional group in purine

molecules can absorb strongly light at ultraviolet

(UV) wavelengths. This can be used for both

quantitative and qualitative analyses of purines

(Garret 2005). Reversed phase HPLC (RP-HPLC) has

been proven to be very efficient for nucleic acid

analysis (Titkova et al. 1983), and has been

commonly used to separate and quantify purine bases

(Kaneko et al. 2014).

RP-HPLC-UV method requires the hydrolysis of

nucleic acids to become nucleotides and free purine

bases using strong acids. Brulé et al. (1989)

developed a sample preparation method for purine

base analysis with RP-HPLC using acid hydrolysis.

In the research of Brulé et al. (1989), samples were

hydrolyzed using 11.6 N perchloric acid for 1 hour at

100 °C, pH was adjusted using NH4OH to obtain pH

4.0, and to mark with distilled water in a 50 mL

volumetric flask, filtered, and finally analyzed by RP-

HPLC. The purine bases were isocratically separated

with a RP-HPLC column C18 and a mobile phase of

0.1 M potassium phosphate buffer mixture and

phosphoric acid at pH 4.0.

In a research conducted by Sotelo et al. (2002)

regarding the determination of purine bases in sea

urchin gonads, samples were hydrolyzed with a

mixture of trifluoroacetic acid/formic acid (1/1, v / v)

at 90 °C for 15 min, and to mark in 250 mL

volumetric flask, and dried with a rotary vacuum

evaporator at 75 °C. The purine base was dissolved

with 10 mL buffer KH2PO4 0.3 M (pH 4.0) and

filtered using a filter membrane before being

analyzed with RP-HPLC equipped with a UV-VIS

detector at a wavelength of 255 nm. Gradient analysis

was carried out using a mobile phase buffer solution

KH2PO4 0.3 M with a pH of 4.0.

The main objective of this research was to validate

the purine base analysis method with a HPLC

instrument. In this study a method validation of the

purine base analysis was conducted on melinjo chip

sample with a HPLC instrument equipped with RP-

HPLC column and a UV-Vis detector which was set

for UV detection, called as RP-HPLC-UV method.

The purine bases chosen were adenine and

hypoxanthine which are known to have a greater

impact on the increase in uric acid in the body than

guanine and xanthine (Clifford et al 1976a). Adenine

and hypoxanthine were analyzed by RP-HPLC with

operating conditions referring to the adenosine

analysis method in royal jelly in the study of Xue et

al. (2009). The sample preparation method was

adapted from the qualitative analysis procedure for

xanthine purine base (AOAC 2012a). The melinjo

chips were hydrolyzed with 6 N HCl at 100 °C for 1

hour, neutralized with 25% NH4OH, and then treated

with aquabidest in a 10 mL volumetric flask. Finally,

the sample was passed through a SPE (Solid Phase

Extraction) column containing silica, and is injected

into a HPLC equipped with a UV-Vis detector.

2 MATERIAL AND METHOD

2.1 Materials

The materials used in this research were melinjo chips

(Sriti, Sriti Food Co., Jakarta, Indonesia). The

chemicals used for analysis include adenine and

hypoxanthine standards (98-99%, Sigma, Sigma-

Aldrich, USA), phosphoric acid 0.4% (Merck,

Germany), methanol (pa, Merck, Germany), ethanol

80% (pa, Merck, Germany), aquabidest, NH4OH

25% (Merck, Germany), HCl 37% (Merck,

Germany), and silica 60 (Merck, Germany).

Analytical balance, oven, food processor, hot

plate, magnetic stirrer, vacuum filter, spatula, vortex,

Millipore nylon filter membrane 0.45 µm, column

SPE (Solid Phase Extraction), and glasswares were

used. The analytical instruments used were pH-meter

and high performance liquid chromatography

(HPLC) LC 6A model (Shimadzu, Shimadzu Corp.,

Kyoto, Japan), equipped with a SPD-10AV model

UV-Vis detector (Shimadzu, Shimadzu Corp., Kyoto,

Japan), Chromatopac semi-automatic data recorder

(Shimadzu, Shimadzu Corp., Kyoto, Japan), and C18

column (Zorbax, Agilent Technologies, USA).

Validation of Analytical Method for Determination of Adenine and Hypoxanthine Purine Bases in Melinjo Chips by HPLC-UV

119

2.2 Method Validation

In general, this study consisted of four parts, namely

a preliminary test for the determination of retention

time precision and resolution of separation, HPLC-

UV instrument performance testing, development of

purine base analysis procedure, and validation of

purine base analysis method on melinjo chip. The

validation of the analytical method included the

specificity of the method, the linearity of the method,

the accuracy and precision of the method by recovery

test, the limit of the method detection, and the intralab

reproducibility. Adenine and hypoxanthine were the

purine bases chosen in this study.

2.3 Preliminary Retention Time

Precision and Peak Resolution Test

The preliminary test was carried out by separately

injecting the standards adenine and hypoxanthine

which had been dissolved in 0.4% (90%) phosphoric

acid and methanol (10%) at several concentrations to

determine the chromatogram profile and the retention

time of each compound. Once adenine and

hypoxanthine appeared at different retention times,

both were then re-injected in the form of a mixture to

determine the peak resolution of the two compounds.

The resolution shows the ability of the column to

separate the two peaks and is declared good if it has a

value greater than 1.50 (Zhang 2007). This test was

done in duplicate. Resolution was determined using

the equation (1).

Rs = 2 (tR-B - tR-A) / wb-A + wb-B (1)

tR-A and tR-B is the retention times of the two

peaks (compound A is the compound that was eluted

first), while wb-A and wb-B are the width of the

baseline (the bottom) of the two peaks.

The operating conditions of HPLC for analysis of

purine bases in melinjo chips refer to the adenosine

analysis method in royal jelly in Xue et al. (2009)

with modifications of the isocratic elution method and

the flow rate of mobile phase. The analysis of purine

adenine and hypoxanthine bases using HPLC with

UV-Vis detector was performed with the condition as

follows: Zorbax C18 (octadecyl silane or ODS)

column, particle size 5 µm, L 250 mm, inner

diammeter 4.6 mm, isocratic mobile phase of

phosphoric acid 0.4% in water mixed with methanol

(pro analysis) at ratio 90:10 and the pH adjusted to

4.0 by NH4OH 1 M, flow rate at 0.5 mL/min, ambient

temperature, 20 µL injection volume, and detection at

UV 257 nm.

2.4 HPLC-UV Instrument

Performance Test for the Analysis

of Adenine and Hypoxanthine

Purine Bases

Standard stock solution. Standard stock solutions of

adenine and hypoxanthine were made with a

concentration of 500 µg/mL by dissolving 0.025 g of

each standard into 50 mL of the HPLC mobile phase.

The mobile phase consists of a mixture of 0.4%

(90%) phosphoric acid and methanol pro analysis

(10%). The standards adenine and hypoxanthine were

mixed to obtain a standard mixed concentration with

a concentration of 250 µg/mL.

Instrument linearity. Linearity was tested by

injecting a serial solution of a mixture of adenine and

hypoxanthine at concentrations of 1.95, 3.91, 7.81,

15.62, 31.25, 62.50, 125.00, and 250.00 µg/mL into

HPLC with the condition above, so that the peak area

of the serial concentrations were known. Testing on

standard mixed solutions with eight different

concentrations was carried out in triplicate from three

different serial standard solutions. The HPLC results

were then used for making a calibration curve which

plotted between the concentrations (µg/mL) and the

averaged peak area, then the coefficient of

determination (R2) was calculated. Linearity is

considered good if it has R2 greater than 0.990

(AOAC 2012b).

Precision of peak area and retention time. The

precision determination of the area and retention time

was done by injecting seven times a standard mix

solution of adenine and hypoxanthine at the same

concentration into the HPLC, in this case a

concentration of 7.81 μg/mL was used. This

concentration was chosen because it gave an

acceptable recovery result (greater than 80%) at

relatively low concentration. The peak area and

retention times from the seven repeatations of

injection were calculated for their average, standard

deviation (SD), and relative standard deviation

(RSD). RSD acceptance in this test was less than

2.0% (JECFA 2006).

Instrument detection limit (limit of detection or

LOD) and quantification limit (limit of quantification

or LOQ). The LOD of instrument was determined

from the above seven injections of a standard mix

solution. Each concentration of each standard from

each injection was calculated using a calibration

curve obtained from the above test. The SD of the

concentrations obtained from the seven repeatations

of injection was calculated, then the LOD was

determined as three times of the SD, meanwhile the

LOQ was determined as ten times of the SD.

2nd SIS 2019 - SEAFAST International Seminar

120

2.5 Analytical Procedure Orientation

The sample preparation procedure for purine base

analysis in melinjo chip was using the acid hydrolysis

method adapted from AOAC (2012a), AOAC

Official Method 960.56 Microchemical Tests for

Xanthine Alkaloids, with some modifications. The

modifications were the HCl volume used to hydrolyze

the sample as well as the length of the hydrolysis

process. The operating conditions of HPLC-UV for

purine base analysis in emping melinjo samples

followed the adenosine analysis method in royal jelly

(Xue et al. 2009) with some modifications. The

modifications were made in terms of the elution

method with the isocratic mobile phase and its flow

rate. In the preparation stage, the melinjo chips that

have been mashed with a food processor were

hydrolyzed with 6 N HCl at different HCl volumes,

0.5, 1.0, and 2.0 mL at 100 °C for 1 hour, neutralized

with 25% NH4OH, and to mark with aquabidest in a

10 mL volumetric flask. Then the sample solution

was passed through a solid phase extraction (SPE)

column containing silica (approximately 1 gram,

weighed after activated in an oven 105 °C for at least

2 hours) and then it was injected into the HPLC

equipped with a UV-Vis detector with detection at

UV 257 nm as above condition.

Samples to be analyzed by HPLC-UV consist of

unspiked samples (without the addition of standards)

and spiked samples (with the addition of standards).

The standard mixture of adenine and hypoxanthine

with a concentration of 400 µg/g sample was applied.

Thus, the recoveries obtained by using procedures

with different acid hydrolysis time lengths and

volumes were determined and the procedure with the

best recovery was chosen for the method validation

below. This orientation test was done in duplicate.

Analysis of the adenine and hypoxanthine purine

bases content in the sample was compared to the

results of the standard injection only from the

instrumental performance test above. If the peak of

each purine base in the sample could be detected

proportional to its concentration, then the developed

procedure was used for the method validation stage.

2.6 Method Validation

The validation of the purine adenine and

hypoxanthine analytical method consisted of: method

specificity, linearity, accuracy and precision,

detection limit, and intralab reproducibility. Method

validation was carried out following EURACHEM

(1998).

Analytical Procedure. Sample preparation. In the

sample preparation stage, a total of 0.5 grams of

melinjo chips that have been mashed with a food

processor were weighed using an analytical balance.

Then, the sample was hydrolyzed with 6 N HCl at 100

°C for 1 hour. The 6 N HCl volume used for acid

hydrolysis was 0.5 mL (the selected volume resulting

from the above development). The hydrolyzed

sample was then neutralized with NH4OH 25%. The

solution was transferred into a 10 mL volumetric

flask and was fixed to the mark with aquabidest. The

analytical solution was passed through an SPE

column containing about 1 gram of silica before being

injected into HPLC.

Determination by HPLC. The analysis of purine

adenine and hypoxanthine bases using HPLC with

UV-Vis detector was performed isocratically

following the conditions described above. The purine

base in sample was calculated by multiplying the

concentration from calibration curve (from

instrumental performance test) with the final sample

volume (10 mL), then divided by sample weight.

Method specificity. Specificity test was done by

injecting a standard mixture of adenine and

hypoxanthine, samples without standard addition,

and samples that have been added with the standard

mixture of adenine and hypoxanthine. In this study a

standard mixed concentration of 400 μg/g of sample

was used. Thus, at least three chromatograms were

obtained. If the chromatogram shows well-separated

peaks and these peaks were not having interference

by other peaks of sample components, then the

specificity of the analytical method was considered

good.

Method linearity. This test was carried out using

samples spiked with adenine and hypoxanthine

standards at concentrations of 50, 100, 200, 400, and

800 µg/g, then sample treatment according to the

analytical procedure for the sample above was

applied to these mixtures, and then injected into the

HPLC. The linearity test of the method was carried

out in triplicate by making three series of samples

which were spiked at the specified concentrations.

After that, the method callibration curve was made,

that is a plot between the peak area and the purine

concentrations (µg/g). The linearity requirement for

the method is R2 greater than 0.990 or r greater than

0.995 (AOAC 2012b).

Accuracy by recovery test. This test was carried

out using samples spiked with standards at three

different concentrations, low (100 µg/g), medium

(500 µg/g), and high concentrations (1000 µg/g).

Each analysis was carried out at seven replications.

The percentage of recovery was using formula (2).

Validation of Analytical Method for Determination of Adenine and Hypoxanthine Purine Bases in Melinjo Chips by HPLC-UV

121

Acceptance of the recovery percentages is according

to AOAC (2012b).

Recovery











x100% (2)

Method precision. The data obtained in the

recovery test for accuracy at three different

concentrations were used for the determination of

method precision by calculating the RSD in each

purine analysis at low, medium, and high

concentrations. The value of RSD analysis (RSDa)

was compared to RSD Horwitz (RSDh). Good

precision was for a RSDa value smaller than RSDh.

RSDh was calculated using formula (3).

RSDh = [2 exp (1-0.5 log C)] (3)

C = Analyte concentration (in fraction of sample)

Method detection limit. The method detection

limit was determined from a plot between the

standard deviations and the concentrations of adenine

and hypoxanthine from the recovery test results at

concentrations of 100, 500, and 1000 µg/g. Through

a linear equation of the curve, the standard deviation

SD0 was determined when the concentration was

equal to zero. MDL value is three times the SD0 value

obtained. Determination of the MDL was referred to

EURACHEM (1998).

Intralab reproducibility. Intralab reproducibility

test was carried out using the same melinjo chip

sample and the same operator and laboratory, but

carried out on different weeks. The results of the

analysis were then calculated for the mean and the

RSD values. The RSD value obtained was then

compared with its RSDh. Good intralab

reproducibility was that with a smaller RSDa value

than its RSDh. In addition, the results obtained were

processed by one-way ANOVA followed by Duncan

posthoc test if there was a significant difference at 5%

level, using the IBM Statistic SPSS 20 program

between the results of analysis from different weeks.

3 RESULTS AND DISCUSSION

3.1 Retention Time and Resolution of

Adenine and Hypoxanthine

The chromatogram obtained showed that adenine

appeared at 5.9-6.8 min, meanwhile hypoxanthine

appeared at 4.8-5.5 min. The quite different retention

times of adenine and hypoxanthine gave the fact that

the two compounds could be analyzed

simultaneously. Figure 1 shows the chromatograms

of standard adenine, standard hypoxanthine, and the

mixture of adenine and hypoxanthine at 62.50 µg/mL.

The analysis of the standard mixture have a resolution

of 4.51. The resolution shows the ability of the

column to separate the two peaks and is declared good

if it has a value greater than 1.50 (Zhang 2007). Thus,

the resolution of the peak adenine and hypoxanthine

was acceptable, so that the two compounds can be

analyzed simultaneously further.

Figure 1: Chromatograms of standards adenine (A),

hypoxanthine (B), a mixture of adenine and hypoxanthine

solution. HPLC column was C18, and the isocratic mobile

phase was phosphoric acid 0.4% in water mixed with

methanol (pro analysis) at ratio 90:10 and the pH adjusted

to 4.0 by NH4OH 1 M. Peaks: (1) adenine, (2)

Hypoxantine.

3.2 Instrument Performance Test

Instrument linearity test results are presented in Table

1. Instrument linearity test results in the table shows

good slope precision, with RSD values less than 5%.

Meanwhile, the value of the intercept obtained

showed a lack of precision both in the analysis of

adenine and hypoxanthine with RSD values greater

than 20%. However, the linearity of the HPLC-UV

instrument is considered as good due to the R2 more

than 0.990, which means the instrument is able to

produce a linear response to the concentration of the

analyte at various levels of concentration.

Retention time (min)

bso

2nd SIS 2019 - SEAFAST International Seminar

122

Table 1: The results of HPLC-UV instrument linearity test

for the analysis of adenine and hypoxanthine standards at

serial concentrations of 1.95, 3.91, 7.81, 15.62, 31.25,

62.50, 125.00, and 250.00 µg/mL.

Slope Intercept R R

Adenine

1 65840 92344 0.9998 0.9997

2 66281 139408 0.9999 0.9998

3 66773 44532 0.9999 0.9999

Average 66298 92095 0.9999 0.9998

SD 467 47438 5.77. 10

-5

1.00.10

-4

RSD (%) 0.70 51.51 0.01 0.01

Hypoxanthine

1 45434 103801 0.9995 0.9991

2 43748 60649 0.9998 0.9997

3 43076 92983 0.9995 0.9990

Average 44086 85811 0.9996 0.9993

SD 1215 22452 1.73.10

-4

3.79.10

-4

RSD (%) 2.76 26.16 0.02 0.04

The chromatogram precision determined in this study

covers the precision of peak area and the precision of

retention time. Each precision was shown by its RSD

value obtained from seven injections. The precision

of retention time in analysis of adenine by the HPLC-

UV was 0.67%, meanwhile the precision of peak area

was 3.18%. The similar result was obtained for

hypoxanthine, the precision of retention time was

0.96% and the precision of peak area was 2.97%. The

acceptable precision is 2.0% or less according to

JECFA (2006). The precisions of all retention times

were acceptable, however the precisions of peak area

were not. The poor peak area precision was caused by

the use of semi-automatic data recorders and printers,

so that the peak area could be affected by the feed

speed. Besides this, according to Barwick (1999), the

precision of peak area was probably influenced by the

flow rate of the mobile phase. A constant mobile

phase flow rate can only be produced by HPLC

pumps that in good condition. However, the HPLC

flow rate has been callibrated by an external

calibration service.

LOD and LOQ values obtained in the analysis of

adenine and hypoxanthine were 0.72 and 2.39 µg/mL,

respectively. Meanwhile, the LOD and LOQ values

in the hypoxanthine analysis were 0.69 and 2.30

µg/mL, respectively. As a comparison, in the research

of Sotelo et al. (2002) regarding the determination of

purine base levels in gonads of sea urchins with

HPLC instruments, the detection limits of adenine

and hypoxanthine were 0.076 and 0.060 µg/mL,

respectively, which are ten times lower than those

obtained in this current study.

The precision of the analysis results in

determining the detection limit and the quantification

limit of the instrument was determined by calculating

RSD of analysis (RSDa) and RSD Horwitz (RSDh).

The RSDa of adenine analysis (3.90%) was smaller

than 2/3 RSDh (8.12%). Similar result obtained for

hypoxanthine analysis, the RSDa value (3.97%) was

smaller than 2/3 RSDh (8.19).

3.3 Analytical Procedure Orientation

Filtering the final sample solution with silica using a

SPE column was aimed to remove impurity

components contained in the sample. In developing

this analytical procedure, samples prepared with the

addition of 6 N HCl at various volumes were injected

into the HPLC, where the samples consisted of

unspiked samples (without the addition of standards)

and spiked samples (with the addition of standards).

Each standard was added at a concentration of 400

µg/g sample which was ten times of the instrument

LOQ.

The results of sample analysis, without the

addition of spikes, using HCl volumes of 0.5, 1.0, and

2.0 mL showed that the average melinjo chips

contained adenine in the concentration range of

70.37-171.88 µg/g sample and hypoxanthine in the

concentration range of 48.37-155.58 µg/g sample.

The total of purine bases in emping melinjo was

predicted between 500-1500 µg/g. The recoveries

obtained at the use of HCl volumes of 0.5, 1.0, and

2.0 mL were presented in Tables 2. The results show

that HCl volume of 0.5 mL provides the best recovery

result of 92.99% and 113.84% for the analysis of

adenine and hypoxanthine, respectively. This was

acceptable according to AOAC (2002b), which

mentioned the acceptable recovery between 85-

110%.

Validation of Analytical Method for Determination of Adenine and Hypoxanthine Purine Bases in Melinjo Chips by HPLC-UV

123

Table 2: The results of the analytical procedure orientation

of adenine and hypoxanthine analysis in melinjo chip using

different volumes of HCl for acid hydrolysis in the sample

preparation stage*.

Volume of

HCl 6 N

(mL)

Concentra

tion of

unspiked

sample

(µg/g)

Concentra

tion of

spiking

(µg/g)

Concentra

tion of

spiked

sample

(µg/g)

Recov

ery

(%)

Adenine

0.5

79.43 ±

2.61 400

451.37 ±

12.82 92.99

1.0

171.88 ±

8.85 400

495.54 ±

11.35 80.91

2.0

70.37 ±

3.48 400

559.03 ±

17.25 122.16

Hypoxant

hine

0.5

103.29 ±

12.12 400

558.64 ±

39.18 113.84

1.0

48.37 ±

4.42 400

650.55 ±

18.93 150.54

2.0

155.58 ±

0.76 400

703.58 ±

5.60 137.00

*in duplicate

3.4 Method Specificity

The chromatogram in Figure 2 shows that the peaks

of adenine and hypoxanthine could be separated to

each other, either analyzed in standard mix solution

or in melinjo chips. Both adenine and hypoxanthine

peaks were not interferred by other peaks of sample

component. Guanin and xanthine which could be

extracted during the sample preparation process,

were not be able analyzed by HPLC because they

cannot be eluted with the mobile phase used in this

study. This mentioned that the method had a good

specificity. Adenine and hypoxanthine in the melinjo

sample were detected at 5.9 ̶ 6.8 min and 4.8 ̶ 5.5 min,

respectively. In Sotelo et al. (2002) study regarding

the determination of purine bases in sea urchin

gonads by HPLC, adenine and hypoxanthine

respectively detected at about 18 and 11 min. A

considerable difference between the retention times

obtained in this study and Sotelo et al. (2002) due to

the different mobile phase used. The mobile phase

used in the Sotelo et al. (2002) study was a 0.3 M

KH2PO4 buffer solution, whereas in this study a

mixture of 0.4% phosphoric acid (90%) mixed with

methanol p.a. (10%) was as the mobile phase. In

general, hypoxanthine was detected earlier than

adenine.

Figure 2: Chromatograms of a standard mixture of adenine

and hypoxanthine (a), unspiked sample (b), and spiked

sample at spiking concentration of 400 µg/g sample (c).

Peaks: (1) adenine, (2) Hypoxanthine.

3.5 Method Linearity

The callibration curve of the method in the analysis

of adenine and hypoxanthine had a linear equation

obtained in the linearity test method for adenine

analysis is y = 3637.7x + 305030 with R and R2

values of 0.9998 and 0.9996, respectively.

Meanwhile testing conducted on hypoxanthine

yielded a curve with the equation of y = 2535.8x +

157650, while R and R2 values were 0.9978 and

0.9956, respectively. R and R2 values obtained in the

adenine and hypoxanthine tests met the requirements

set by AOAC (2012b), that is, R was greater than

0.995 or R2 was greater than 0.990. Thus, the method

used had a good linearity, which could provide a

linear response to the concentration of analytes in the

sample.

3.6 Method Accuracy and Precision

According to AOAC (2002b), the acceptable

percentages of recovery for concentrations of 100 and

500 µg/g is 85-110%, while for concentration of 1000

µg/g is 90-108%. The recovery test results were

presented in Table 3. The results at the spike of 500

and 1000 µg/g in adenine analysis met the AOAC

(2002) requirement, namely by recovery of 89.39%

Retention time (min)

Absorbance at 257 nm

2nd SIS 2019 - SEAFAST International Seminar

124

and 90.37%, respectively. While at 100 µg/g spiking,

the recovery did not meet the standard, which was

79.33% (less than 85%). The recovery results in

hypoxanthine analysis, were only acceptable at the

spiking concentration of 500 and 1000 µg/g, with

recoveries of 92.29% and 100.15%, respectively.

While at the spiking concentration of 100 µg/g, its

recovery was far below the AOAC standard, which

was 66.75%. This might be due to the relatively low

concentration of spikes and the analytes content

naturally in the sample (without the standards

addition). The loss of analytes during sample

preparation seems more significant in the results of

the analysis.

The precision of the analytical method was

determined through the recovery test. The precision

of the method is considered good if the RSDa value

is smaller than the RSDh. If the RSDa value is smaller

than 2/3 RSDh, then the precision of the method is

even better. Determination of the 2/3 RSDh value for

a more stringent standard is to ensure the results of

the analysis. The method precision is presented in

Table 4. In the adenine analysis with the addition of

standard at 100, 500, and 1000 µg/g, RSDa value was

smaller than RSDh, but only at concentrations of

1000 µg/g RSDa value was smaller than 2/3 RSDh.

This shows that in the analysis of adenine, the spiking

at 100 and 500 µg/g was less precise than the spiking

at 1000 µg/g. On the other hand, in the hypoxanthine

analysis, the RSDa values were all smaller than 2/3

RSDh. Thus, hypoxanthine analysis had a good

precision at all spiking studied.

Table 3: Recovery of analytical method for the

determination of adenine and hypoxanthne in melinjo chip

by HPLC-UV.

Concentration

of spiking

(µg/g)

Concentration

found in spiked

sample (µg/g)

Concentration

found in

unspiked

sample (µg/g)

Averaged

recovery

(%)

Adenine

100

222.04 ±

11.53* 142.71 ± 1.35 79.33

500

591.97 ±

26.65* 145.00 ± 4.70 89.39

1000

1044.26 ±

36.19** 140.53 ± 4.51 90.37

Hypoxanthine

100

285.30 ±

8.49* 218.55 ± 7.44 66.75

500

592.29 ±

18.67* 130.84 ± 2.74 92.29

1000

1261.64 ±

28.07**

260.12 ±

11.83 100.15

*Obtained from 7 replications

**Obtained from 5 replications

Table 4: Precision of analytical method for the

determination of adenine and hypoxanthne in melinjo chip

by HPLC-UV.

Concentration of

spiking (µg/g)

(µg/g)

RSDa*

(%)

RSDh**

(%)

2/3 RSDh

(%)

Adenine

100 11.53 5.19 7.09 4.73

500 26.65 4.50 6.12 4.08

1000 36.19 3.46 5.62 3.75

Hypoxanthine

100 8.49 2.98 6.83 4.55

500 18.67 3.15 6.12 4.08

1000 28.07 2.22 5.46 3.64

*RSDa is relative standard deviation of the analysis

**RSDh is RSD Horwitz 2(1-0.5 log c) with c is a fraction

in sample

3.7 Method Detection Limit

The method detection limit was determined by

plotting the standard deviations obtained from

recovery test and the concentrations of adenine and

hypoxanthine found. Through the linear equation, the

standard deviation at zero concentration was

determined (SD0). The method detection limit is

three times the value of SD0. The linear curve is

presented in Figure 3.

Figure 3: A plot between analyte concentration (µg/g) and

its standard deviation from recovery test of adenine and

hypoxanthine analyses by HPLC-UV to determine SD0 for

calculating method detection limit (3SD0).

Based on the curve in Figure 3, the SD0 value for

adenine was 6.4794. Thus, the method detection limit

value for adenine analysis was 19.44 µg/g. While the

SD0 value for hypoxanthine was 4.8082, therefore the

method detection limit for hypoxanthine analysis was

14.42 µg/g. The curve had R2 values greater than

0.900 which was acceptable.

y = 0.0295x + 6.4794

R² = 0.9642

y = 0.0191x + 4.8082

R² = 0.9462

0 500 1000 1500

Standard deviation (µg/g)

Concentration (µg/g)

Adenin Hiposantin

Validation of Analytical Method for Determination of Adenine and Hypoxanthine Purine Bases in Melinjo Chips by HPLC-UV

125

3.8 Intralab Reproducibility

The values of intralab reproducibility in the analysis

of adenine and hypoxanthine were greater than the

RSDh value. Intralab reproducibility values for

adenine and hypoxanthine analyses were 34.53% and

39.89%, respectively. The results of one-way

ANOVA followed with Duncan test gave that they

were significantly different, the result from week 1

was different from the results from weeks 2 and 3.

Meanwhile, the results of hypoxanthine analysis at

weeks 1, 2, and 3 were significantly different to each

other. This might be caused by a poor area precision

observed in the instrument performance test results.

4 CONCLUSIONS

The analysis procedure for alkaloids in AOAC

Official Method 960.56 can be applied to the analysis

of the purine adenine and hypoxanthine bases in

melinjo chips with HPLC-UV. Development of the

analytical procedures carried out showed that the

analytical method using acid hydrolysis provided the

best recovery in the use of 6 N HCl volumes of 0.5

mL. The developed procedure can then be used at the

method validation stage.

In general, the method had a good performance

on all validation parameters except the intralab

reproducibility. The analytical method for the

determination of adenine and hypoxanthine in

melinjo chip by HPLC-UV was validated and gave

an accurate result at concentrations more than 100

µg/g sample.

REFERENCES

[AOAC] Association of Official Analytical Chemists,

2012a. Chapter 18, AOAC Official Method 960.56

Microchemical Tests for Xanthine Alkaloids. AOAC

International. Maryland, USA, 19th edition, page 53.

[AOAC] Association of Official Analytical Chemists,

2012b. AOAC Official Methods of Analysis, Appendix

K: Guidelines for Single Laboratory Validation of

Chemical Methods for Dietary Supplements and

Botanical. AOAC International. Maryland, USA, 19th

edition.

Barwick, V. J, 1999. Sources of uncertainty in gas

chromatography and high-performance liquid

chromatography. Journal of Chromatography 849: 13 ̶

33.

Brulé, D., Sarwar, G., Savoie, L., 1989. Effect of methods

of cooking on free and total purine bases in meat and

fish. Canadian Institute of Food Science and

Technology Journal 22: 248 ̶ 251.

Clifford, A. J., Riumallo, J. A., Young, V. R., Scrimshaw,

N. S., 1976a. Effect of oral purines on serum and

urinary uric acid of normal, hyperuricemic and gouty

humans. The Journal of Nutrition 106: 428 ̶ 434.

Clifford, A. J., Story, D. L, 1976b. Levels of purines in

foods and their metabolic effects in rats. The Journal of

Nutrition 106: 435 ̶ 442.

EURACHEM. 1998. The Fitness for Purpose of Analytical

Methods: a Laboratory Guide to Method Validation and