A simple high-performance liquid chromatography ...

Clinica Chimica Acta 433 (2014) 150–156

Contents lists available at ScienceDirect

Clinica Chimica Acta journal homepage: www.elsevier.com/locate/clinchim

A simple high-performance liquid chromatography (HPLC) method for the measurement of pyridoxal-5-phosphate and 4-pyridoxic acid in human plasma Rona Cabo a, Karolina Kozik b, Maciej Milanowski b, Sigrunn Hernes c, Audun Slettan a, Margaretha Haugen d, Shu Ye e, Rune Blomhoff f,g, M. Azam Mansoor a,f,⁎ a

Department of Natural Sciences, University of Agder, Kristiansand, Norway Department of Biochemistry, University of Technology and Life Sciences, Bydgoszcz, Poland c Department of Nutrition and Public Health, University of Agder, Kristiansand, Norway d National Institute of Public Health, Oslo, Norway e William Harvey Research Institute, Queen Marry University of London, London, UK f Department of Nutrition, University of Oslo, Olso, Norway g Division of Cancer Medicine, Surgery and Transplantation, Oslo University Hospital, Olso, Norway b

a r t i c l e

i n f o

Article history: Received 6 February 2014 Received in revised form 3 March 2014 Accepted 3 March 2014 Available online 19 March 2014 Keywords: HPLC Vitamin B6 PLP and 4-PA

a b s t r a c t Background: Low concentration of plasma pyridoxal-5-phosphate (PLP) is associated with hyperhomocysteinemia and inflammation. Most methods for the measurement of plasma PLP require large specimen volume and involve the use of toxic reagents. Methods: We have developed a HPLC method for the measurement of PLP and 4-pyridoxic acid (4-PA) in plasma, which requires small specimen volume. The samples are prepared without adding any toxic reagents. Furthermore, we have examined whether intake of vitamin B6 affects the concentration of plasma PLP and 4-PA. Results: The coefficient of variation of the method was 6% and the recovery of the added vitamin in plasma was about 100%. The concentrations of plasma PLP and 4-PA in 168 healthy subjects were 40.6 (8.4–165.0) nmol/L, median and (range) and 17.5 (3.7–114.79) nmol/L, median and (range) respectively. In the multiple regression analyses, the concentration of plasma PLP was associated with the concentration of plasma 4-PA (p b 0.0001), BMI, (p = 0.02) and sex, (p = 0.0008). The concentration of plasma 4-PA was associated with plasma PLP (p b 0.0001), serum folate (p = 0.004), smoking (p = 0.03) and vitamin B6 intake (p = 0.01). Conclusion: The present method is suitable for large clinical studies for the measurement of plasma PLP and 4-PA. Our findings demonstrate that plasma 4-PA, BMI and sex are the major determinants of plasma PLP in healthy individuals. © 2014 Elsevier B.V. All rights reserved.

1. Introduction It has been reported that about 4% of all intracellular enzymatic activity depends upon pyridoxal-5-phosphate (PLP), which is a cofactor form of vitamin B6. 4-Pyridoxic acid (4-PA) is a degradation product of PLP and can be measured in blood plasma [1–3]. PLP modulates the activity of the enzyme cystathionine β-synthase (CBS), which converts homocysteine to cystathionine and that is further catabolized to cysteine by the enzyme cystathionine lyase (CL). The activity of CL also depends upon PLP. A deficiency of PLP results in the elevation of plasma total homocysteine (p-tHcy), which is associated with cardiovascular disease (CVD) [4–6]. Most studies, but not all, have demonstrated that ⁎ Corresponding author at: Department of Natural Sciences, University of Agder, 4616 Kristiansand, Norway. E-mail address: [email protected] (M.A. Mansoor).

http://dx.doi.org/10.1016/j.cca.2014.03.003 0009-8981/© 2014 Elsevier B.V. All rights reserved.

the deficiency of vitamin B6 is associated with inflammation [7–11]. It has also been demonstrated that vitamin B6 has antioxidant properties and its deficiency may play a role in oncogenesis [12,13]. Several methods for the measurement of plasma PLP have been reported in the literature. However, most methods involve the use of toxic reagents and require large sample volume, which are probably the major drawbacks for their application in large clinical studies. Furthermore, in some assays, within-day coefficient of variation (CV) is very high, up to 37% and between the days CV% varies from 2% to 26%. Mean recoveries of added PLP in the plasma samples is as low as 53% and as high as 144% [14–17]. Findings based on precision parameters obtained in these analytical methods suggest that there is an urgent need for a new and an accurate method, which requires small specimen volume, for the measurement of plasma PLP and 4-PA. Here, we describe a HPLC method, which is simple and accurate and requires a small amount of a specimen. The method has a single protein

R. Cabo et al. / Clinica Chimica Acta 433 (2014) 150–156

precipitation step and does not require toxic reagents for the measurement of PLP and 4-PA in plasma or serum. To compare the newly developed method with other methods, we have measured the concentration of plasma PLP and 4-PA in 168 apparently healthy subjects. It is also interesting to investigate whether the concentration of plasma PLP has association with the concentration of plasma 4-PA, p-tHcy and vitamin B6 intake. 2. Materials and methods 2.1. Reagents and standard solutions PLP, 4-PA, trichloroacetic acid (TCA), potassium phosphate monobasic and sodium perchlorate were purchased from Sigma. Sodium bisulfite was purchased from ACROS Biochemicals. We prepared PLP standards in double distilled water. Solubility of 4-PA in water is low; therefore 4-PA was dissolved in 1% NaOH solution first, which was later diluted with double distilled water. 2.2. Inclusion criteria Healthy subjects, both male and female, age from 18 to 67 years. Exclusion criteria were; the subjects who were diagnosed for CVD, cancer, and diabetes, and individuals who were taking medication or using vitamin-B supplements. The participants were mostly employees or students at the University of Agder, Norway. All subjects fasted for 12 h before they provided the blood samples. We used heparin plasma for the measurement of PLP, 4-PA, homocysteine, cysteine and cysteinylglycine. Serum samples were used for the determination of folate and vitamin B12. All participants provided written informed consent to participate in the studies. The Ethical Committee Helse Sør-Øst, Norway approved the studies. 2.3. Blood samples from the study subjects Blood samples were collected in heparin vials and vials without any coagulant, for serum, from 168 apparently healthy subjects. In one study, we recruited 102 individuals for an intervention study, whereas in the second study, we recruited 68 subjects for the measurement of biochemical parameters to compare them between healthy smokers and non-smokers. Main findings of the two studies will be presented later in two separate communications. At least 6 subjects could not provide samples for all analyses.

151

for 10 min at 13,000 rpm at 4 °C. The plasma proteins were removed and 25 μL clear supernatant was injected in the auto-sampler for the measurement of PLP and 4-PA. 2.7. HPLC system The UHPLC Ultimate 3000-Dionex System was obtained from Thermo Scientific, USA. The fluorescence detector attached to HPLC was also from Dionex. The fluorescence detector had excitation at λ 300 nm and emission at λ 400 nm. We used Restek, Pinnacle II, 150 × 4.6 mm, C18 column with particle size 3 μm and a 5 μm Drop-In Cartridge as a guard column (Thermo Scientific). 2.8. Mobile phase The mobile phase consisted of buffer A: 0.1 M potassium phosphate monobasic, 0.1 M sodium perchlorate and 0.5 g/L sodium bisulfite [18]. Buffer B: contained 0.1 M potassium phosphate monobasic, 0.1 M sodium perchlorate, 0.5 g/L sodium bisulfite and 20% acetonitrile. The pH of the buffers A and B were adjusted to 3.0 by adding phosphoric acid. The flow rate of the buffer was 0.5 mL/min. The following gradient for the mobile phase was used: Buffer A, 100%, from 0 to 4 min and buffer B, 27%, from 4.1 to 13 min. The column was washed with 50% acetonitrile for 3 min and then it was primed for 4 min with buffer A, before analyzing the next sample. 2.9. Measurement of PLP and 4-PA in plasma We plotted standard concentrations of PLP or 4-PA on the x-axis and peak area on the y-axis. Using linear regression analysis, a straight line was drawn between the parameters at x-x and y-y-axis. Simple regression analysis resulted in an equation, which was used for the calculation of PLP, and 4-PA concentration in plasma. For the measurement of PLP and 4-PA in plasma samples, we prepared daily fresh PLP and 4-PA standards. 2.10. Measurement of homocysteine, related molecules in plasma and B-vitamins in serum The concentrations of p-tHcy, plasma total cysteine and plasma total cysteinylglycine were measured according to the HPLC method published by us previously [19]. The concentrations of serum folate and serum vitamin B12 were measured with Cobas 6000 (Roche Diagnostics) in the laboratory of medical biochemistry, SSHF Hospital, Arendal, Norway.

2.4. Blood samples for the analysis of coefficient of variation (CV) test 2.11. Analyses of dietary intakes Blood samples were collected from 10 apparently healthy subjects in heparin vials, EDTA vials and vials without any anticoagulant for serum for the analysis and calculation of CV. All blood samples were placed on the ice, in the dark, and were centrifuged within 30 min at 4 °C. Vials for serum were incubated at 4 °C for 45 min before the centrifugation. Heparin plasma, EDTA plasma and serum were isolated and were used to test the precision of the assay. 2.5. Blood samples for the recovery assay and the controls We collected blood samples from 6 subjects in heparin vials. Heparin plasma was used for the recovery assay and for the preparation of inhouse controls. One control was tested for each batch of 11 unknown plasma samples.

Dietary intakes of vitamin B6, proteins, carbohydrates and fat have been reported previously [20]. 2.12. Statistical methods The distribution of plasma PLP and 4-PA concentrations in the studied population did not match with the Gaussian distribution curve (not normal distribution). The Mann–Whitney U test was used to explore differences between two groups. We used Kruskal–Wallis test to investigate whether anticoagulants had any impact on the concentration of plasma PLP and 4-PA. The concentrations of plasma PLP and 4-PA were converted into log values for testing in multiple regression analysis. P b 0.05 was considered significant. All statistical analyses were performed by the StatView for MacOS (USA).

2.6. Preparation of plasma or serum samples for HPLC analyses 3. Results In a 50 μL plasma or in a 50 μL serum sample, proteins were precipitated by adding 50 μL 16% TCA; the final concentration of TCA was 8% in the samples. The samples were vortexed for 1 min and were centrifuged

Fig. 1 shows the chromatograms of a blank, a PLP standard, a 4-PA standard, and a spiked plasma sample.

152


Fluorescence units

A

Fluorescence units

B

Fluorescence units

C

Fluorescence units

D

Fig. 1. HPLC chromatograms showing peaks for PLP and 4-PA; Panel A; a standard, Panel B; a blank, Panel C; a plasma sample, and Panel D; a spiked plasma sample.

R. Cabo et al. / Clinica Chimica Acta 433 (2014) 150–156 Table 1 Demographic, biochemical and nutrient intake variables in the study participants. Number of subjects BMI (n = 160) Age, years (n = 164) Sex ratio, male/female Ratio, number of smokers/non-smokers Serum folate, nmol/L (n = 168) Serum vitamin B12, pmol/L (n = 168) Plasma total homocysteine, μmol/L (n = 167) Plasma total cysteinylglycine, μmol/L (n = 167) Plasma total cysteine, μmol/L (n = 167) Vitamin B6 intake, mg/day (n = 166) Protein intake, g/day (n = 166) Carbohydrate intake, g/day (n = 166) Fat intake, g/day (n = 166)

168 24.0 (17.0–37.0) 30 (18–68) 112/51 41/127 13.7 (6.0–37.1) 272 (112–887) 7.5 (3.3–33.0) 31.1 (12.2–56.0) 239 (124–357) 1.7 (0.6–6.8) 91.2 (43.1–262.1) 297 (62–902) 79.3 (22.3–270.1)

Data are presented in median and in range in the parentheses.

3.1. Demographic data The concentrations of plasma aminothiols, B-vitamins and nutrient intake data are provided in Table 1.

153

Table 2A CV test performed for PLP and 4-PA in plasma and in serum. PLP

4-PA

Concentrationa

CV%

Concentrationa

CV%

Heparin plasma Day 1 (n = 10) Day 2 (n = 10) Day 3 (n = 10) Total (n = 30)

26.2 (24.9–27.0) 23.9 (22.5–25.2) 23.0 (22.6–25.3) 24.1 (22.5–27.0)

2.3 2.9 3.6 5.5

10.2 (10.1–10.8) 10.0 (9.5–10.3) 9.5 (9.0–9.9) 10.0 (9.0–10.8)

2.2 2.9 2.9 4.4

EDTA plasma Day 1 (n = 10) Day 2 (n = 10) Day 3 (n = 10) Total (n = 30)

30.5 (29.1–31.4) 27.2 (25.8–28.2) 27.0 (26.4–29.4) 27.6 (25.8–31.4)

2.2 2.5 3.6 5.9

11.4 (11.1–11.6) 10.2 (9.9–10.8) 10.2 (9.9–10.5) 10.4 (9.9–11.6)

1.8 2.4 1.9 5.6

Serum Day 1 (n = 10) Day 2 (n = 10) Day 3 (n = 10) Total (n = 30)

25.8 (24.1–27.5) 23.9 (20.8–25.1) 23.0 (22.3–25.6) 24.1 (20.8–27.5)

4.2 5.2 3.9 6.0

12.5 (12.0–12.8) 11.5 (9.5–12.1) 11.7 (11.1–12.0) 11.9 (9.5–12.8)

1.9 7.2 2.5 6.0

Data are presented in median and in range in the parentheses. aThe concentration units are nmol/L. A significant effect of anticoagulants on PLP and 4-PA concentrations (p b 0.0001 and p b 0.0001).

3.2. Standard curves Fig. 2 shows the standard curves, the linear regression plots for PLP and 4-PA standards. We plotted from 0 to100 nmol/L PLP and from 0 to 100 nmol/L 4-PA standard concentrations on the x-axis and their respective peak areas on the y-axis in two separate graphs. The correlation coefficient (R) for PLP was 0.999 (p b 0.0001), Y = 0.02 + 0.15 × X and R = 0.999 (p b 0.0001), Y = 0.04 + 0.045 × X for 4-PA respectively. The 4-PA in the standard and in the plasma samples resulted in higher peak areas than PLP.

heparin plasma, EDTA plasma and serum (Table 2A, 2B and 2C). The day-to-day CV for PLP in heparin plasma, EDTA plasma and serum was in the range of 2.2% to 4.2%, whereas the total CV for all days ranged from 5.5% to 6.0%. The concentration of PLP in EDTA plasma, heparin plasma and serum was significantly different (p b 0.0001). Day-to-day CV for 4-PA in heparin plasma, EDTA plasma and serum was found in the range of 1.8% to 7.2%, whereas the total CV for all days ranged from 4.4% to 6.0%. The concentration of 4-PA in the serum, heparin plasma and in EDTA plasma was significantly different (p b 0.0001) (Table 2A).

3.3. Coefficient of variation 3.4. Recovery assay

Area mvx min -PLP

The day-to-day coefficient of variations, for PLP and 4-PA (n = 10) for 3 days were determined in three types of biological materials;

1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0

3.5. Concentration of PLP and 4-PA in healthy subjects

5.0

The median, and range concentrations of PLP and 4-PA in plasma are given in Table 3. The frequency distribution of PLP and 4-PA concentration in plasma is shown in Fig. 3. In the multiple regression analyses, the concentration of plasma PLP was significantly associated with plasma 4-PA (p b 0.0001), with sex (p = 0.0008) and BMI (p = 0.02). The concentration of plasma 4-PA was associated with the concentration of plasma PLP (p b 0.0001), concentration of serum folate (p = 0.004), smoking (p = 0.03) and vitamin B6 intake (p = 0.01) (Table 4).

4.0

4. Discussion

0

20

40 60 80 100 120 PLP concentration

Y = 0.02 + 0.015 x X; R = 0.999 Area mv x min -4-PA

Two different concentrations of PLP and 4-PA solutions were added in the heparin plasma samples for the recovery assay. The recovery of PLP was from 100% to 107% and the recovery of 4-PA in plasma was from 99% to 104% (Table 2B and 2C).

3.0 The HPLC method reported here has only one preparation step, which is the precipitation of proteins in the plasma samples. Our

2.0 1.0

Table 2B Recovery assay performed for PLP in heparin plasma.

0 0

20

40

60

80

100

120

4-PA concentration

Y = 0.04 + 0.045 x X; R = 0.999 Fig. 2. Linear regression plot between the PLP standards and the corresponding peak areas. A linear regression plot between 4-PA standards and the corresponding peak areas.

Sample 1 Sample 2 Sample 1 Sample 2

PLP in matrix

PLP added

PLP measured

PLP recovered

Recovery%

6.7 6.4 6.7 6.4

27.8 29.8 14.8 14.9

36.1 36.2 22.5 22.3

29.4 29.8 15.8 15.9

106 100 107 107

154


60

Table 2C Recovery assay performed for 4-PA in heparin plasma. 4-PA added

4-PA measured

4-PA recovered

Recovery%

5.4 5.3 5.4 5.3

36.0 35.8 16.0 16.9

41.2 42.2 22.1 22.9

35.8 29.8 16.7 17.6

99 103 104 104

Count

Sample 1 Sample 2 Sample 1 Sample 2

50

4-PA in matrix

40 30 20 10

The concentration units are nmol/L.

0

0

20

40

60

80 100 120 140 160 180

Concentration of plasma PLP, nmol/L 80 70 60

Count

method has high precision and requires small specimen volume, and the separation of the PLP and 4-PA peaks in the chromatogram is good; therefore, it is suggested that this method is suitable for large clinical studies. Many analytical methods reported in the literature are tedious, have several preparation steps, involve the use toxic reagents and require large specimen volume, the major drawbacks for their application in large clinical studies (Table 3). The variability of analytical methods may increase with an increased number of derivatization steps for the preparation of the samples for the HPLC. Some of the previously published reports on HPLC methods for the measurement of PLP and 4-PA have not provided the data on precision parameters (Table 3). In some HPLC methods, the fluorescence intensity of PLP and 4-PA in the samples is enhanced by derivatization the samples, pre or postcolumn, with bisulfite, semicarbazide, chlorite or cyanide [15–18,27]. Cyanide is a hazardous substance; therefore, its use requires special laboratory facilities. We used sodium bisulfite in the mobile phase, which is not a toxic substance [18]. It has been reported that the aldehyde group at position 4 of PLP molecule is partially responsible for the low fluorescence of the native molecule. To increase the fluorescence of PLP in the samples injected in the HPLC, bisulfite was added in the mobile phase, which changed the function of the aldehyde group of PLP [25]. Methods based on HPLC-MS/MS systems are also applied for the measurement of plasma PLP and 4-PA [29–31]. HPLC-MS/MS systems are usually used in specialized laboratories and require highly competent laboratory personal, which generally makes application of these systems and the methods difficult. Nonetheless, the precision parameters reported by these methods are not significantly better than our HPLC-fluorescence method (Table 3). At present, established and verified quality controls for PLP or 4-PA are not available. Similarly, a comparison method for the measurement of plasma vitamin B6 is lacking. Therefore, heparin plasma from 6 donors was collected to prepare a pool of in-house control. One control was tested for each batch of 11 unknown plasma samples.

50 40 30 20 10 0

0

20

40

60

80

100

120

Concentration of plasma 4-PA, nmol/L Fig. 3. The frequency distribution chart for PLP and 4-PA in plasma.

The plasma PLP and 4-PA concentrations determined in healthy subjects in our study are similar to PLP and 4-PA concentrations reported in the literature (Table 3). Previously, plasma PLP b 34.4 or b 20.0 nmol/L had been reported as cut-off values; however, these cut-off values had been questioned [27,32,33]. In the UK, only 0.5%, young people age 4–18 years, on the basis of a b 20 nmol/L cut-off value for PLP, had deficiency of vitamin B6 [33]. The findings based on these reports suggest that more research is required to establish a valid cut-off value of plasma PLP in healthy subjects. In the present study, plasma PLP concentration was not associated with p-tHcy. We detected in the elderly in the UK that the association between plasma PLP and p-tHcy was attenuated when age, sex, BMI, domicile, serum folate, serum vitamin B12 were included in the multiple regression analysis [34,35].

Table 3 Description of the methods reported in the literature for the measurement of plasma PLP and 4-PA. Sample

Prep.

Detection

CV%

Volume

Steps

System

PLP (w)

4-PA (w)

PLP (t)

4-PA (t)

PLP

Recovery % 4-PA

PLP

Plasma concentration 4-PA

Ref

– 1000 μL 1000 μL 600 μL 500 μL 500 μL 500 μL 200 μL 150 μL 100 μL 100 μL – 60 μL 60 μL 60 μL 50 μL

3 5 4–5 4–5 3–4 1 1 4 5 3–5 5–6 2 2 2 1 1

Fluor Fluor/bisulfite Fluor/semicarbazide Fluor/cyanide Fluor/semicarbazide Fluor/bisulfite Fluor/bisulfite Fluor/cyanide Fluor/chlorite Fluor/bisulfite Fluor/cyanide Fluor/bisulfite LC–MS/MS ESI–MS/MS LC–MS/MS Fluor/bisulfite

– 5.4 5.9 3.7 3.7–5.2 – 0.5–2.4 1.9–3.5 0.6–1.2 0.8–4.6 2.0 5.1–6.1 5.1–6.1 2.6–3.0 2.6–5.3 2.9

– 4.5 – 4.3 7.6–11.6 – – 3.3–5.6 0.9–1.8 2.5–4.0 – 8.1–8.8 3.1–5.1 2.7–4.1 4.3–5.3 2.7

– 8.1 11.8 – 4.6–6.9 – 2.9–5.3 – 3.6–6.7 – 5.2–5.7 2.6–6.6 5.5–7.4 5.9–7.1 5.1–6.3 5.5

– 9.8 – – 8.4–13.7 – – – 3.7–5.6 – – 7.1–12.3 9.4–11.0 6.7–11.1 4.0–4.2 4.4

87–98 92–99 98 ± 4 98 ± 3 97 ± 4–98 ± 3 96 91–103 98–103 97 ± 4–101 ± 3 90 ± 22 93 (90–97) 97 ± 12 98–102 96–97 94–101 101–104

91 116–118 – 97 ± 1 91 ± 4–93 ± 3 102 – 98–102 97 ± 3–99 ± 3 93 ± 6 ND 88 ± 5 95–102 103–108 104–116 101–104

77 (n = 1) 61.5 ± 33.6 46.3 ± 25.4 – 56 (21–138) 19 ± 9 (n = 3) 111–113 (n = 12) 37.2 42.4 (40.1) 80 ± 21 (n = 8) 43.5 – 34.4 37.2 46–321 45.7 ± 25.8 (40.6)

– 25.1 ± 8.7 (n = 10) – – 23 (9–60) (n = 126) 3.9 ± 3 (n = 3) – 15.2 (n = 136) 27.3 (21.8) (n = 303) 75 ± 20 (n = 8) – – 22.4 (n = 94) 24.3 (n = 94) 16.4–139 (n = 24) 20.1 ± 12.0 (17.5)

[21] [15] [22] [17] [23] [18] [24] [16] [25] [26] [27] [28] [29] [30] [31] our study

Data are presented as mean ± SD nmol/L. CV% for PLP (w), PLP within run, PLP (t) PLP total, 4-PA (w) 4-PA within-run, 4-PA (t), 4-PA total. Ng/mL are converted to nmol/L [15,22], 50 percentile males all ages [16], geometric mean and median in the parentheses [25], geometric mean and 95% RI in the parentheses Ref. [23], (Low–High) [24], ND—not detectable Ref. [27], median Ref. [29,30], age (4.3y–16y) [31]. (–) Data are not provided. Fluor, fluorescence. ESI, electrospray ionization.


155

Table 4 Data on multiple regression analyses. PLP dependent variable

4-PA dependent variable

Independent variables

Std. coeff

t-value

p-value

Std. coeff

t-value

p-value

Age BMI Sex Smoking P-tHcy P-total cysteine Serum folate Serum vitamin B12 Vitamin B6 intake Folate intake Vitamin B12 intake Protein intake Fat intake Carbohydrate intake Plasma 4-PA Plasma PLP

−0.166 −0.183 −0.320 −0.002 −0.068 0.079 0.049 0.010 0.097 0.081 0.186 −0.117 −0.081 −0.188 0.430

−1.965 −2.400 −3.424 −0.025 −0.802 0.879 0.499 0.139 0.956 0.509 1.382 −0.602 −0.540 −1.006 5.27

0.051 0.018 0.008 0.980 0.433 0.381 0.681 0.890 0.340 0.611 0.169 0.548 0.590 0.316 b0.0001

0.115 0.130 −0.002 0.173 0.105 −0.134 0.267 0.015 0.239 −0.015 −0.168 0.051 −0.092 0.109

1.403 1.758 −0.016 2.197 1.284 −1.51 2.952 0.212 2.484 −0.096 −1.296 0.269 −0.636 0.601

0.163 0.081 0.987 0.030 0.201 0.125 0.004 0.832 0.014 0.924 0.197 0.788 0.526 0.549

0.381

5.193

b0.0001

Log converted values of plasma PLP and 4-PA were used in the analyses (n = 152).

5. Conclusion Our HPLC method is suitable for large clinical studies for the measurement of PLP and 4-PA in plasma and serum. The method has very low CV%, it requires small sample volume and has a single sample preparation step. The present method is an important and a significant addition to the previously reported HPLC methods for the measurement of PLP and 4-PA.

Conflict of interest There is no conflict of interest between the authors.

Acknowledgments We are thankful to our students for their contribution in this work. A minor portion of the data is presented in student assignments. This research work was supported by the University of Agder, Norway.

References [1] Coburn SP. Vitamin B6. In: Dakshinamurti K, editor. Annals NY Acad Sci, 585; 1990. p. 76–86. [2] Spinneker A, Sola R, Lemmen V, Castillo MJ, Pietrzik K, Gonzalez-Gross M. Vitamin B6 status, deficiency and its consequences—an overview. Nutr Hosp 2007;22:7–24. [3] Leklem JE. Vitamin B-6: a status report. J Nutr 1990;120(Suppl. 11):1503–7. [4] Finkelstein JD. Methionine metabolism in mammals. J Nutr Biochem 1990;1:228–37. [5] Clarke R, Collins R, Lewington S, Donald A, Alfthan G, Tuomilehto J, et al. Homocysteine and risk of ischemic heart disease and stroke: a meta-analysis. Homocysteine Studies Collaboration. JAMA 2002;288:2015–22. [6] Welch GN, Loscalzo J. Homocysteine and atherothrombosis. N Engl J Med 1998;338:1042–50. [7] Friso S, Jacques PF, Wilson PW, Rosenberg IH, Selhub J. Low circulating vitamin B6 is associated with elevation of inflammation markers C-reactive protein independently of plasma homocysteine levels. Circulation 2001;103:2788–91. [8] Sakakeeny L, Roubenoff R, Obin M, Fontes JD, Benjamin EJ, Bujanover Y, et al. Plasma pyridoxal-5-phosphate is inversely associated with systematic markers of inflammation in a population of U.S. adults. J Nutr 2012;142:1280–5. [9] Lotto V, Choi SW, Friso S. Vitamin B6: a challenging link between nutrition and inflammation in CVD. Br J Nutr 2011;106:183–95. [10] Ulvik A, Midttun Ø, Pedersen ER, Nygård O, Ueland PM. Association of plasma B-6 vitamers with systemic markers of inflammation before and after pyridoxine treatment in patients with stable angina pectoris. Am J Clin Nutr 2012;95:1072–8. [11] Folsom AR, Desvarieux M, Nieto FJ, Boland LL, Ballantyne CM, Chambless LE. B vitamin status and inflammatory markers. Atherosclerosis 2003;169:169–74. [12] Wondrak GT, Jacobson EL. Vitamin B6: beyond coenzyme functions. In: Stanger O, editor. Water soluble vitamins. Sub-cellular biochemistry Springer Science; 2012. p. 291–300 [Chapter 15].

[13] Galluzzi L, Vacchelli E, Michels J, Garcia P, Keep O, Senovilla L, et al. Effect of vitamin B6 metabolism on oncogenesis, tumor progression and therapeutic responses. Oncogene 2013;32:4995–5004. [14] Rybak ME, Jain RB, Pfeiffer CM. Clinical vitamin B6 analysis: an interlaboratory comparison of pyridoxal 5-phosphate measurement in serum. Clin Chem 2005;51:1223–31. [15] Hollins B, Henderson JM. Analysis of B6 vitamers in plasma by reversed-phase column liquid chromatography. J Chromatogr 1986;380:67–75. [16] Bates CJ, Pentieva KD, Matthews N, Macdonald A. A simple, sensitive and reproducible assay for pyridoxal 5-phosphate and 4-pyridoxic acid in human plasma. Clin Chim Acta 1999;280:101–11. [17] Hess D, Vuilleumier JP. Assay of pyridoxal-5-phosphate, pyridoxal and pyridoxic acid in biological material. Int J Vitam Nutr Res 1989;59:338–43. [18] Kimura M, Kanehira K, Yokoi K. Highly sensitive and simple liquid chromatographic determination in plasma of B6 vitamers, especially pyridoxal 5-phosphate. J Chromatogr 1996;722:295–301. [19] Mansoor MA, Svardal AM, Ueland PM. Determination of the in vivo redox status of homocysteine, cysteine, cysteinylglycine, and glutathione in human plasma. Anal Biochem 1992;92:218–29. [20] Hernes S, Cabo RN, Mansoor MA, Haugen M. Eating patterns are associated with biomarkers in selected population of university students and employees. J Nutr Sci 2012;1:1–8. [21] Vanderslice JT, Marie CE, Beecher GR. B6 vitamer analysis in human plasma by high performance liquid chromatography: a preliminary report. Am J Clin Nutr 1981;34:947–50. [22] Ubbink JB, Serfontein WJ, de Villiers LS. Stability of pyridoxal 5-phsphate semicarbazone: applications in plasma vitamin B6 analysis and population surveys of vitamin B & nutritional status. J Chromatogr 1985;342:277–84. [23] Talwar D, Quasim T, McMillan DC, Kinsella J, Williamson C, O´Reilly DS. Optimisation and validation of a sensitive high-performance liquid chromatography assay for routine measurement of pyridoxal-5-phosphate in human plasma and red cells using pre-column semicarbazide derivatisation. J Chromatogr B Analyt Technol Biomed Life Sci 2003;792:333–43. [24] Deitrick CL, Katholi RE, Huddleston DJ, Hardiek K, Burrus L. Clinical adaptation of a high-performance liquid chromatographic method for the assay of pyridoxal 5-phosphate in human plasma. J Chromatogr B Biomed Sci Appl 2001;751:383–7. [25] Rybak ME, Pfeiffer CM. Clinical analysis of vitamin B6: determination of pyridoxal 5-phosphate and 4-pyridoxic acid in human serum by reversed-phase highperformance liquid-chromatography with chlorite post column derivatization. Anal Biochem 2004;333:336–44. [26] Sharma SK, Dakshinamurti K. Determination of vitamin B6 vitamers and pyridoxic acid in biological samples. J Chromatogr 1992;578:45–51. [27] Bailey AL, Wright AJA, Southon S. High performance liquid chromatography method for the determination of pyridoxal-5-phosphate in human plasma: how appropriate are cut-off values for vitamin B6-deficiency. Eur J Clin Nutr 1999;53:448–55. [28] Bisp MR, Bor MV, Heinsvig EM, Kall MA, Nexø E. Determination of vitamin B6 vitamers and pyridoxic acid in plasma: development and evaluation of highperformance liquid chromatography assay. Anal Biochem 2002;305:82–9. [29] Midttun Ø, Hustad S, Solheim E, Schneede J, Ueland PM. Multi-analyte quantification of vitamin B6 and B2 species in the nano-molar range in human plasma by liquid chromatography-tandem mass spectrometry. Clin Chem 2005;51:1206–16. [30] Midttun Ø, Hustad S, Ueland PM. Quantitative profiling of biomarkers related to B-vitamin status, tryptophan metabolism and inflammation in human plasma by liquid chromatography/tandem mass spectrometry. Rapid Commun Mass Spectrom 2009;23:1371–9.

156


[31] Footitt EJ, Clayton PT, Mills K, Heales SJ, Neergheen V, Oppenheim M, et al. Measurement of plasma B6 vitamer profiles in children with inborn errors of vitamin B6 metabolism using a LC-MS/MS method. J Inherit Metab Dis 2013;36:139–45. [32] Haller J, Lowik MR, Ferry M, Ferro-Luzzi A. Nutritional status: blood vitamins A, E, B6, B12, folic acid and carotene. Euronet SENECA investigators. Eur J Nutr 1991;45(Suppl. 3):63–82. [33] Bates CJ, Pentieva KD, Prentice A. An appraisal of vitamin B6 status indices and associated confounders in young people aged 4-18 years and people aged 65 years and over, in two national British surveys. Public Health Nutr 1999;2:529–35.

[34] Bates CJ, Pentieva KD, Prentice A, Mansoor MA, Finch S. Plasma pyridoxal phosphate and pyridoxic acid and their relationship to plasma homocysteine in a representative sample of British men and women aged 65 years and over. Br J Nutr 1999;81:191–201. [35] Bates CJ, Mansoor MA, Pentieva KD, Hamer M, Mishra GD. Biochemical risk indices, including plasma homocysteine, that prospectively predicts mortality in older British people: the National Diet and Nutrition Survey of People Aged 65 Years and Over. Br J Nutr 2010;104:893–9.