Toshio OKANO,3 Naoko MIzuNo,3 Sumiko SHIDA,3 Norlko TAKAHASHI,3 Tadashl KOBAYASHI,3* Elzo KURODA,4 Solchl

Transcription

1 J. Nutr. Sci. Vitaminol., 27, 43-54, 1981 A Method for Simultaneous Determination of 25-Hydroxyvitamin D2 and 25-Hydroxyvitamin D3 in Human Plasma by Using Two Steps of High-Performance Liquid Chromatography1,2 Toshio OKANO,3 Naoko MIzuNo,3 Sumiko SHIDA,3 Norlko TAKAHASHI,3 Tadashl KOBAYASHI,3* Elzo KURODA,4 Solchl and Tamotsu KODAMA,4 MATSUO4 3 Department of Hygienic Chemistry, Kobe Women's College of Pharmacy, Higashinada-ku, Kobe 658, Japan 4Department of Pediatrics, Kobe University School of Medicine, Ikuta-ku, Kobe 650, Japan (Received September 22, 1980) Summary A method for simultaneous determination of 25-hy droxyvitamin D2 (25-OH-D2) and 25-hydroxyvitamin D3 (25-OH-D3) in human plasma has been developed by using two steps of high-performance liquid chromatography (HPLC). Lipids extracted from 0.5ml of human plasma were first subjected to the preparative HPLC using a Nucleosil SC18 column (reversed-phase type) and a 25-OH-D fraction containing 25-OH- D2 and 25-OH-D3 was separated. The separated fraction was subsequently subjected to the analytical HPLC using a Zorbax SIL column (straight phase type). Since the peaks corresponding to 25-OH-D2 and 25-OH-D3 were clearly separated from one another on the chromatogram of the analytical HPLC, the metabolites could be simultaneously determined by estimating the respective peak heights. When the fractions corresponding to the respective peaks were separately collected by repeatedly applying rather large quantities of human plasma and were subjected to gas chromatography-mass spectrometry (GC-MS), they were identified as * To whom correspondence should be addressed. 1 A preliminary communication in this paper was reported in J. Pharmaco-bio Dynamics, 3, s-15 (1980). 2 Following abbreviations are used: 25 -OH-D2/D3, 25-hydroxyvitamin D2/D3 (25- hydroxy-ergocalciferol/-cholecalciferol); 25-OH-pyro-D2/D3, 25-hydroxypyrovitamin D2/D3; 25-OH-isopyro-D2/D3, 25-hydroxyisopyrovitamin D2/D3; 1,25-(OH)2-D2/D3, 1ƒ,25- dihydroxyvitamin D2/D3; HPLC, high-performance liquid chromatography; GC-MS, gas chromatography-mass spectrometry; UV, ultraviolet; AUFS, absorbance unit full scale; TIC, total ion collector. 43

2 44 T. OKANO et al. containing 25-OH-D2 and 25-OH-D3, respectively. The proposed method was applied to plasma samples of human adults taking 400 I.U./day of vitamin D2 for 8 weeks and the values were 22.5 }8.1ng/ml for 25-OH-D3 and 11.5 }1.8ng/ml for 25-OH-D2 (mean }S. D.), respectively. Key Words gas chromatography-mass spectrometry, 25-hydroxyvitamin D2, 25-hydroxyvitamin D3, 25-hydroxyergocalciferol, 25-hydroxycholecalciferol, human plasma, high-performance liquid chromatography ultraviolet detector, vitamin D2, vitamin D3, It has been documented that the major circulating metabolite of vitamin D in blood (mainly existing in plasma) is 25-OH-D whose plasma concentration directly reflects the repletion status of vitamin D (1). Therefore, it is very important to establish a simplified method for determining this metabolite in human plasma from the viewpoints of clinical and nutritional sciences. Vitamin D2 and D3 are similarly metabolized to 25-OH-D2 and -D3 in the liver and subsequently to 1,25-(OH)2-D2 and -D3 in the kidney to show physiological effects (1). When normal humans receive commercial drugs or enriched foods with exposure to sunlight, the existence of exogenous 25-OH-D2 besides endogenous 25-OH-D3 derived from skin is usually observed in their plasma because vitamin D2 is predominantly used in such drugs and enriched foods for commercial reasons (The cost of vitamin D2 is much lower than that of vitamin D3 and both vitamins show practically the same biological activity to humans). Therefore, we tried to establish a simplified method for simultaneous determination of 25-OH-D2 and 25-OH-D3 in human plasma by HPLC. Existing HPLC assay methods for the metabolites usually require clean-up procedures to eliminate lipophilic concomitants. Chromatography on "open" columns using silica gel (2, 3), Sephadex LH-20 (4-8), hydroxyalkoxypropyl Sephadex (9) or others (10,11) was used for this purpose, but is rather time consuming and large quantities of organic solvents are required. On the other hand, Jones (12) reported a method using two different types of HPLC, preparative and analytical HPLC, for separative assay of vitamin D2, D3, 25-OH-D2 and 25-OH-D3 in human plasma. The first preparative HPLC on a straight-phase column (Zorbax SIL) was used to eliminate concomitants, whereas the second analytical HPLC on a reversed-phase column (Zorbax ODS) was used to measure the compounds. The use of preparative HPLC for clean-up procedure is an excellent idea, because the technique is time-saving and permits good separation and reproducibility. However, when we re-examined the method, sometimes we met with the problem that the disorder of an expensive reversed-phase column for the analytical HPLC soon occurred from repeated use. We thought that this might be due to the overload of lipids on the column. Saponification procedure was applied to diminish the overload in a previous report (13), but this was rather complicated and time-consuming. After trying to revise Jone's method, we found that good results were obtained when the J. Nutr. Sci. Vitaminol.

3 ASSAY OF 25-OH-D2 AND 25-OH-D3 IN PLASMA 45 column order was reversed from Jone's (A reversed-phase column was used for the first preparative HPLC and a straight-phase column for the second analytical HPLC). The detailed procedure on the revised method is described in this paper. EXPERIMENTAL 1. Materials and reagents Commercial grades (Philips-Duphar Co., The Netherlands) of 25-OH-D3 and the in vivo synthesized 25-OH-D2 purified from the blood of rabbits dosed with large amounts of vitamin D2 as reported in a previous paper (14) were used as the respective standard compounds. Organic solvents were purified according to the usual methods and distilled before use. Other guaranteed reagents were used. 2. Plasma samples Plasma samples of adults were collected from healthy volunteers of laboratory workers taking 400 I.U. (10ƒÊg)/day of vitamin D2 for 8 weeks. 3. Instrumentation 1) First preparative HPLC. The first preparative HPLC to isolate a 25-OH-D fraction including both 25-OH-D2 and 25-OH-D3 from the lipid extracts of plasma was performed on a Shimadzu-DuPont 841 high-performance liquid chromato graph equipped with a UVD-1 detector (fixed at 254nm, AUFS 0.01, Japan). Separation was carried out in a stainless tube (300 ~7.5 mm i.d.) packed with Nucleosil SC18 (reversed-phase type, Nishio Industrial Co., Japan) using 20% methanol in acetonitrile as the mobile phase. Elution was carried out at a constant flow rate of 1.5ml/min (column pressure: 55kg/cm2) and the eluate between 1,450 and 1,700 drops (The drops corresponded to the eluate between about 15 and 18min of retention time) was collected as a 25-OH-D fraction by a mini-fraction collector with a drop counter (Gilson Co., USA). 2) Second analytical HPLC. The second HPLC to determine 25-OH-D3 and 25-OH-D2 was performed on a Shimadzu LC-2F high-performance liquid chromatograph with a Shimadzu UVD-2 detector (fixed at 254nm, AUFS 0.001). Separation was carried out in a stainless tube (250 ~4.6 mm i.d.) packed with Zorbax SIL (straight-phase type, DuPont Co., USA) using 5.5% isopropanol in n hexane as the mobile phase. Flow rate was 0.56ml/min (column pressure: 20kg/cm2). 3) Gas chromatography-mass spectrometry (GC-MS). The GC-MS was performed on a Hitachi M-80 gas chromatograph-double focussing mass spectrometer (Japan) equipped with a unit of electron impact as an energy source. A glass column (100 ~0.3cm i.d.) packed with 2% OV-1 on Chromosorb W (60-80 mesh) was run at 243 Ž with a helium flow of 40ml/min. Separator temperature, injection port temperature and ionizing voltage were controlled at 300 Ž, 320 Ž and 20 ev, respectively. When the peak corresponding to 25-OH-pyro-D2 or 25-OHpyro-D3 was observed on a gas chromatogram, mass spectra were estimated. The Vol. 27, No. 1, 1981

4 46 T. OKANO et al. analysis of GC-MS was kindly performed by Hitachi Co. 4. Procedure for the determination of 25-OH-D2 and 25-OH-D3 in plasma 1) Extraction of lipids from plasma. Exactly 0.5ml of plasma was placed in a test tube with a stopper. Threeml of methanol and 1.5ml of methylene dichloride were added and mixed with a Vortex mixer for 2min. After allowing to stand for 20min, the same procedure for mixing and standing was repeated. The mixed solution was centrifuged at 3,000rpm for 5min. The centrifuged solution showed mono-layer with protein precipitation at the bottom. One ml of methylene dichloride and 1ml of water were added to it. Since the solution was separated into two layers by the addition, the lower methylene dichloride layer was taken with a Pasteur pipet and placed in a round bottomed flask. The upper water layer was extracted further three times with each 2ml of methylene dichloride and the extracted layers were combined in the round bottomed flask. After evaporating the solvent to dryness under reduced pressure, the resulting residue was dissolved in 10ml of methylene dichloride. 2) Isolation of the 25-OH-D fraction including both 25-OH-D2 and 25-OH-D3 by the first preparative HPLC. Exactly 8ml of the solution obtained above was placed in a test tube and evaporated to dryness under reduced pressure. The resulting residue was dissolved in 200ƒÊl of methanol. Exactly 180ƒÊl of the solution was subjected to the first preparative HPLC described in the above section. The eluate corresponding to the 25-OH-D fraction (1,450-1,700 drops) was collected and evaporated to dryness under reduced pressure. By use of authentic compounds, both 25-OH-D2 and 25-OH-D3 were confirmed to be included in the fraction. The resulting residue was dissolved in 200ƒÊl of 5.5% isopropanol in n-hexane. The solution was denoted as a sample solution for the second HPLC. 3) Simultaneous determination of 25-OH-D2 and 25-OH-D3 by the second analytical HPLC. Exactly 100ƒÊl of the sample solution obtained above was submitted to the second analytical HPLC described in the above section. In addition, 5ƒÊl of either 25-OH-D2 or 25-OH-D3 standard solution (1,000ng/ml of the respective authentic compounds in 5.5% isopropanol in n-hexane) was subjected to the HPLC in the same manner. The peak heights corresponding to 25-OH-D2 and 25-OH-D3 were estimated on the HPL-chromatograms of the sample or standard solutions, respectively. The concentration of 25-OH-D2 and 25-OH-D3 in plasma (ng/ml) can be calculated by the following formula: Concentration of 25-OH-D2 or 25-OH-D3 (ng/ml) = Psa/Pst ~S/w ~10/8 ~200/180 ~200/100 Psa: Peak height of 25-OH-D2 or 25-OH-D3 on the HPL-chromatogram obtained from a sample solution. Pst: Peak height of 25-OH-D2 or 25-OH-D3 on the HPL-chromatogram obtained from the respective standard solutions. S: Quantity of 25-OH-D2 or 25-OH-D3 (ng) applied to the HPLC of J. Nutr. Sci. Vitaminol.

5 ASSAY OF 25-OH-D2 AND 25-OH-D3 IN PLASMA 47 the respective standard solutions. This is 5 in the above case. W: Volume (ml) of plasma taken for the assay. This is 0.5 in the above case. 10/8: The ratio between the volume of methylene dichloride for dissolving the lipid extract and that taken for making the solution applied to the preparative HPLC. 200/180: The ratio between the volume of methanol for dissolving the residue from the above methylene dichloride solution and that applied to the preparative HPLC. 200/100: The ratio between the volume of 0.5% isopropanol in n-hexane for making the sample solution and that applied to the analytical HPLC. RESULTS AND DISCUSSIONS 1. Profiles of the preparative HPLC Profiles of the preparative HPLC on a mixture of authentic 25-OH-D2 and 25- OH-D3 and the lipid extract of human plasma are shown in Figs. 1 a and l b. The plasma sample was obtained from a healthy adult taking 400 I.U./day of vitamin D2. The retention times of authentic 25-OH-D3 and 25-OH-D2 were 16.7 and Fig. 1. Profiles of the preparative HPLC of authentic 25-OH-D2, 25-OH-D3 and the lipid extract obtained from a plasma sample of human adult. A mixture of authentic 25-OH-D2 and 25-OH-D3 or the lipid extract was applied to the preparative HPLC described in EXPERIMENTAL. (a) Authentic 25-OH-D2 and 25-OH-D3, (b) the lipid extract. Vol. 27, No. 1, 1981

6 48 T. OKANO et al. 17.4min, respectively, but separation between the two compounds was insufficient as shown in Fig. 1(a). When a new column of Nucleosil 5C18 was used, clear separation was initially observed but it gradually became insufficient after repeated use of the column. We thought this might be due to the deterioration of the column derived from an overload. If the column is used for analytical purposes, the insufficient separation must be undesirable. However, since we used the column for preparative purposes, the insufficient separation was convenient rather than undesirable because the eluate including both 25-OH-D2 and 25-OH-D3 could be simultaneously collected as a 25-OH-D fraction and both compounds could be simultaneously determined by applying the fraction to the second analytical HPLC using a straight-phase column (Zorbax SIL). The 25-OH-D fraction was confirmed to be the eluate between 1,450 and 1,700 drops (about 15 and 18 min of retention time) from the chromatogram of authentic 25-OH-D2 and 25-OH-D3 (Fig. la). Good reproducibility for collecting the fraction was obtained even after repeated use of the column. Many large irrelevant peaks which might be due to unknown concomitants were observed in the chromatogram of plasma (Fig. 1b), but these were eliminated from the fraction because most of them were eluted before the 25-OH-D fraction. From the results, we concluded that the preparative HPLC was useful for the purpose. Other reversed-phase columns, e.g., Zorbax ODS (DuPont Co.) were also useful for the same purpose. Peaks corresponding to 25-OH-D2 and 25-OH-D3 were not observed in the chromatogram of plasma (Fig. lb) but this was due to the low sensitivity of the Fig. 2. Profiles of the analytical HPLC of authentic 25-OH-D2, 25-OH-D3 and the 25- OH-D fraction isolated from the preparative HPLC of a plasma sample of human adult. A mixture of authentic 25-OH-D2 and 25-OH-D3 or the 25-OH-D fraction was applied to the analytical HPLC described in EXPERIMENTAL. (a) Authentic 25- OH-D2 and 25-OH-D3, (b) 25-OH-D fraction. J. Nutr. Sci. Vitaminol.

7 ASSAY OF 25-OH-D2 AND 25-OH-D3 IN PLASMA 49 detector UVD-l equipped with the chromatograph. When the detector UVD-2 was used, the peaks were detectable because it has a high sensitivity as shown in the chromatogram of analytical HPLC (Fig. 2b). 2. Profiles of the analytical HPLC Profiles of the analytical HPLC on a mixture of authentic 25-OH-D 2 and 25- OH-D3 and the 25-OH-D fraction isolated from the preparative HPLC are shown in Fig. 2. The retention times of authentic 25-OH-D2 and 25-OH-D3 were 20.4 and 23.6min, respectively, which were clearly separated from one another (Fig. 2a). The chromatogram of the sample solution also gave two clearly separated peaks tentatively denoted as peaks I and II whose retention times agreed with those of the respective authentic compounds (Fig. 2b). Since no peaks interfering with these two were observed in the chromatogram, we considered the clean-up procedure by the preparative HPLC successfully performed. 3. Identification of peaks I and II as 25-OH-D2 and 25-OH-D3 1) Co-chromatography. Co-chromatography was performed by mixing the sample solution used in the previous section with either authentic 25-OH-D2 or 25- OH-D3. As shown in Fig. 3, peak I or II gave an increase of peak height by addition of the respective authentic compounds. The results showed that peaks I and II Fig. 3. Profiles of the co-chromatography on the analytical HPLC of the 25-OH-D fraction with authentic 25-OH-D2 or 25-OH-D3. The 25-OH-D fraction isolated from the preparative HPLC of a plasma sample of human adult was applied to the analytical HPLC described in EXPERIMENTAL with addition of either authentic 25- OH-D2 or 25-OH-D3. The profiles of left side, middle and right side represent those of the 25-OH-D fraction without addition, with addition of authentic 25-OH-D2 and with addition of authentic 25-OH-D3, respectively. *, 25-OH-D2; **, 25-OH-D3. Vol. 27, No. 1, 1981

8 50 T. OKANO et al. were due to 25-OH-D2 and 25-OH-D3, respectively. 2) GC-MS. Each 0.5ml of plasma was taken from about 80 healthy humans including adults and neonates and each plasma sample was subjected to the whole procedure according to Section 4 of EXPERIMENTAL. The fractions corresponding to the peaks I and II on the analytical HPLC were respectively collected and combined. After evaporating the solvent under reduced pressure, the fractions were individually applied to the GC-MS described in Section 3 of EXPERIMENTAL. As shown in the upper side of Fig. 4, the gas chromatograms of total ions on both the peaks I and II showed two peaks each due to the respective pyro- and isopyroderivatives. Mass spectra were estimated on the respective eluate corresponding to their pyro-derivatives and the results are shown in the lower side of Fig. 4. The mass spectrum of peak I gave the molecular ion (M+) at m/e 412 (the molecular weight of 25-OH-pyro-D2) and the fragment ions at m/e 394 (M+-H2O), 379 (M+-H2O- CH3), 271(M+-side chain) and 253 (M+-side chain-h2o). On the other hand, the mass spectrum of peak II gave the molecular ion (M+) at m/e 400 (the molecular weight of 25-OH-pyro-D3) and the fragment ions at m/e 367 (M+-H2O-CH3), 341 (M+-C3H70, loss of C-25 carbon, 25-OH and C-26,27 dimethyl groups from the Fig. 4. GC-MS of the 25-OH-D2 and 25-OH-D3 fractions isolated from human plasma. Plasma collected from healthy humans was subjected to the whole procedure described in EXPERIMENTAL and the 25-OH-D2 and 25-OH-D3 fractions were isolated from the analytical HPLC. The isolated fractions were individually applied to the GC-MS described in EXPERIMENTAL. The profiles of upper side represent the gas chromatograms described by total ion collector (TIC) of the 25- OH-D3 and 25-OH-D2 fractions, respectively. The peaks of a, b, a L and b L were due to those of 25-OH-pyro-D3, 25-OH-isopyro-D3, 25-OH-pyro-D2 and 25-OHisopyro-D2, respectively. The profiles of lower side represent the mass spectra of the peaks a and a L (25-OH-pyro-D3 and 25-OH-pyro-D2 peaks), respectively. J. Nutr. Sci. Vitaminol.

9 ASSAY OF 25-OH-D2 AND 25-OH-D3 IN PLASMA 51 molecular ion) and 271(M+-side chain). The mass spectra agreed with those of the respective authentic compounds. From the results of GC-MS and cochromatography, peaks I and II were completely confirmed to be due to 25-OH-D2 and 25-OH-D3, respectively. 4. Calibration curves of 25-OH-D2 and 25-OH-D3 Either authentic 25-OH-D2 or 25-OH-D3 was dissolved in 5.5% isopropanol in n-hexane to concentrations of 200 to 2,000ng/ml. Exactly 5ƒÊl of the standard solutions was subjected to the analytical HPLC as described in EXPERIMENTAL in order to obtain calibration curves. As shown in Fig. 5, straight lines passing through the origin were observed between the peak heights and the amounts of the metabolites (0-10ng) in both the cases of 25-OH-D2 and 25-OH-D3. Fig. 5. Calibration curves of 25-OH-D2 and 25-OH-D3 on the analytical HPLC. 5. Extraction of lipids from plasma In order to extract lipids from plasma, the Bligh and Dyer method (15) using a solvent mixture of chloroform and methanol for the extraction has proved very popular. However, this method resulted in an emulsion being formed between the upper water phase and the lower chloroform phase, the technique for taking the lower layer being rather difficult. On the other hand, since in the proposed method we used rather small quantities of methylene dichloride instead of chloroform (1.5ml of methylene chloride and 3ml methanol) for the first extraction, the precipitation derived from protein was formed at the bottom of the test tube and the precipitation was not disturbed by the following extraction. Therefore, the procedure taking the lower methylene dichloride layer was easier than the Bligh and Dyer method. When a recovery test on the proposed extraction method was performed by using 3H-25-OH-D3, good results (Recovery was almost 100%) were Vol. 27, No. 1, 1981

10 52 T. OKANO et al. obtained. 6. Recovery of 25-OH-D2 and 25-OH-D3 for the whole procedure Half ml of pooled human plasma was treated according to the whole procedure described in EXPERIMENTAL with or without the addition of either l0ng of authentic 25-OH-D2 or 20ng of authentic 25-OH-D3 in order to perform recovery tests. As shown in Table 1, the overall recoveries of 25-OH-D2 and 25-OH-D3 were 96.5 }6.4 and 96.8 }5.4 (mean }S. D.), respectively, which were satisfactory. Table 1. Recovery test of 25-OH-D2 and 25-OH-D3 in human plasma. a n.d. (not detected) 1ng of 25-OH-D2/ml of plasma. The data are shown as mean }S. D. 7. Assayed values of 25-OH-D3 and 25-OH-D2 in human plasma of adults Plasma samples were collected from healthy adults in September who had been taking multivitamin tablets including vitamin D2 (400 I.U./day) and were then subjected to the procedure described in EXPERIMENTAL. As shown in Table 2, the values of 25-OH-D3 and 25-OH-D2 were 22.5 }8.1 and 11.5 }1.8ng/ml Table 2. Assayed values of 25-OH-D3 and 25-OH-D2 in human plasma of healthy dults. M=male, F=female. The subjects had been taking multivitamin tablets including vitamin D2 (400 I.U./day) every day for 8 weeks and the blood was collected. J. Nutr. Sci. Vitaminol.

11 ASSAY OF 25-OH-D2 AND 25-OH-D3 IN PLASMA 53 (mean }S. D.), respectively. The data of endogenous 25-OH-D3 were similar to those reported in a previous paper (13), while the values of exogenous 25-OH-D2 showed about half on those of 25-OH-D3. The results suggested that the plasma level of endogenous 25-OH-D3 might be unaffected by the administration of vitamin D2. The detailed data on the levels of the metabolites in human plasma will be reported in the future. The authors wish to thank the Analytical Center of Hitachi Co. for measuring the GC MS for the study. REFERENCES 1) DeLuca, H. F. (1978): Vitamin D and calcium transport. Ann. N. Y. Acad. Sci., 307, ) Koshy, K. T., and Vandersilk, A. K. (1976): High-performance liquid chromatographic method for the determination of 25-hydroxycholecalciferol in cow plasma. Anal. Biochem., 74, ) Gilbertson, T. J., and Stryd, R. P. (1977): High-performance liquid chromatographic assay for 25-hydroxyvitamin D3 in serum. Clin. Chem., 23, ) Eisman, J. A., Shepard, R. M., and DeLuca, H. F. (1977): Determination of 25- hydroxyvitamin D2 and 25-hydroxyvitamin D3 in human plasma using highperformance liquid chromatography. Anal. Biochem., 80, ) Skinner, R. K., and Wills, M. R. (1977): Serum 25-hydroxyvitamin D assay. Evaluation of chromatographic and non-chromatographic procedures. Clin. Chim. Acta, 80, ) Schafter, P. C., and Goldsmith, R. (1978): Quantitation of 25-hydroxycholecalciferol in human serum by high-pressure liquid chromatography. J. Lab. Clin. Med., 91, ) Horst, R. L., Shepard, R. M., Jorgensen, N. A., and DeLuca, H. F. (1979): The determination of the vitamin D metabolites on a single plasma sample: Changes during parturition in dairy cow. Arch. Biochem. Biophys., 192, ) Shepard, R. M., Horst, R. L., Hamstra, A. J., and DeLuca, H. F. (1979): Determination of vitamin D and its metabolites in plasma from normal and anephric man. Biochem. J., 182, ) Delvin, E. E., Glorieux, F. H., Dussault, M., Bourbonnais, R., and Watters, G. (1979): Simultaneous measurement of serum 25-hydroxycholecalciferol and 25-hydroxyergocalciferol. Med. Biol., 57, ) Koshy, K. T., and Vandersilk, A. L. (1977): High-performance liquid chromatographic method for the determination of 25-hydroxycholecalciferol in the bovine liver, kidney and muscle. J. Agric. Food Chem., 25, ) Koshy, K. T., and Vandersilk, A. L. (1978): Improvements in high perfromance liquid chromatographic method for the determination of 25-hydroxycholecalciferol in cow plasma. Anal. Biochem., 85, ) Jones, G. (1978): Assay of vitamin D2 and D3, and 25-hydroxyvitamin D2 and D3 in human plasma by high-performance liquid chromatography. Clin. Chem., 24, ) Matsuyama, N., Okano, T., Takada, K., Takao, T., Terao, Y., Hashimoto, N., and Kobayashi, T. (1979): Assay of 25-hydroxyvitamin D%n human plasma by highperformance liquid chromatography. J. Nutr. Sci. Vitaminol., 25, Vol. 27, No. 1, 1981

12 54 T. OKANO et al. 14) Okano, T., Matsuyama, N., Kobayashi, T., Kuroda, E., Kodama, S., and Matsuo, T. (1979): Isolation and purification of in vivo-generated 25-hydroxyvitamin D2 by preparative high-performance liquid chromatography. J. Nutr. Sci. Vitaminol., 25, ) Bligh, E. G., and Dyer, W. J. (1959): A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol., 37, J. Nutr. Sci. Vitaminol.