首页   按字顺浏览 期刊浏览 卷期浏览 Gas-liquid chromatographic determination of vitamin D in cod-liver oil
Gas-liquid chromatographic determination of vitamin D in cod-liver oil

 

作者: J. G. Bell,  

 

期刊: Analyst  (RSC Available online 1973)
卷期: Volume 98, issue 1165  

页码: 268-273

 

ISSN:0003-2654

 

年代: 1973

 

DOI:10.1039/AN9739800268

 

出版商: RSC

 

数据来源: RSC

 

摘要:

268 Analyst, April, 1973, Vol. 98, pj4. 268-273 Gas - Liquid Chromatographic Determination of Vitamin D in Cod-liver Oil BY J. G. BELL AND A. A. CHRISTIE (Department of Trade and Industry, Laboratory of the Government Chemist, Covnwall House, Stamford Street, London, SE1 9NQ) A gas - liquid chromatographic method for the determination of vitamin D (cholecalciferol) in cod-liver oil is described. It involves saponification of the oil, extraction of the unsaponifiable matter, removal of interferences such as cholesterol and retinol (vitamin A) by precipitation, and colunin chroniatography on Sephadex LH-20 and Florisil ; the final determination of cholecalciferol is carried out by gas - liquid chromatography. A determination can be completed in less than 2 days. EXISTING methods for the determination of vitamin D in foodstuffs and natural products are not entirely satisfactory.Biological assays with rats and chickens are time consuming, expensive and lack precision ; chemical methods are insufficiently sensitive and may be seriously affected by the presence of vitamin A and cholesterol, both of which frequently occur with vitamin D and often in considerable excess. The well known reaction between vitamin D and antimony trichloride suffers also from the disadvantages that the resultant coloured product is highly sensitive to trace amounts of moisture and its intensity varies with time. Of the physicochemical methods, ultraviolet spectroscopy can detect as little as 2 pg of calciferol by absorption measurement at wavelength 265 nm, but many other substances absorb in this region of the spectrum and effectively restrict the value of the technique to the determination of vitamin D in concentrates. Infrared spectrophotonietry has been used to distinguish ergocalciferol from cholecalciferol on the basis of the additional double bond present in the former, but for quantitative work the technique lacks sensitivity, 250 pg of the vitamin being the minimum amount required.In 1960, the application of gas - liquid chromatography to vitamin D determinations gave the first indication that the use of more sensitive methods was possible1 and, in 1966, Murray, Day and Kodicek2 made a major advance by converting ergocalciferol and cholecalciferol into their isovitamins by treatment with antimony trichloride reagent in order to obtain single separate peaks that could be distinguished from those of small amounts of cholesterol.About this time, a need arose in our laboratory for a chemical method of determining vitamin D in products such as cod-liver oil and pharmaceutical preparations. This paper describes the development of a gas - liquid chromatographic method for determining vitamin D in cod-liver oil by saponification and extraction of unsaponifiable matter; removal of vitamin A and other interferences in two stages by partition and adsorption chromatography; precipitation of cholesterol; formation of trimethylsilyl derivatives of the isovitamins ; and gas - liquid chromatography. Ergo- calciferol is used as an internal standard in the method as vitamin I) is present as cholecalci- ferol in cod-liver oil.EXPERIMENTAL RIJAGBNTS AND MATERIALS- All reagents should be of analytical-reagent quality unless otherwise stated. Digitonin B.P.-The material supplied by Koch-Light has been found to be suitable. Sephadcx LH-20-Supplied by Pharmacia, Uppsala, Sweden. Florisil, 60 to 100 mesh. .4 ntimony trichlovide solution, 20 per cent. m/V in chloroform. Tartaric acid solution, 40 per cent. m/V. Bis(trinzethylsi1yl)acetamide-Supplied by Phase Separations, Queensferry, Flintshire. Ergocalcijerol (vitamin D,), pure. Cholecalciferol (vitamin D3), pure. Su-Clzolestane solution-Dissolve 1 mg of pure 5u-cholestane in 100 ml of cliloroform. @ SAC; Crown Copyright Reserved.BELL AND CHRISTIE 269 APPARATUS- Glass columns for chromatography-These were 1-5 and 2 cm in diameter, and of minimum length 40 cm, each fitted with a sintered-glass disc of porosity 2, Quickfit socket and PTFE key.Gas chromatograph-Pye, Model 104, with flame-ionisation detector and 2.7 m x 3 mm i.d. glass columns packed with 3 per cent. OV-17 on Gas-Chrom Q, 100 to 120 mesh. Recorder-Honeywell Electronik 19, 1-mV input, fitted with a disc integrator. PREPARATION OF CHROMATOGRAPHIC COLUMNS- Scphadex LH-20 partition column-A 90 + 10 methanol - water mixture was used as the stationary phase and 2,2,4-trimethylpentane as the mobile phase. Saturate each solvent with the other by shaking equal volumes together for 1 minute and then store them separately. Allow about 100 g of Sephadex LH-20 to swell in 90 per cent. methanol for at least 24 hours before use. During this period decant the supernatant liquor at least twice, re-shaking with fresh 90 per cent.methanol each time. Fit an extension column to the top of a 40 x 2-cm glass column and add the Sephadex slurry to a height of twice that of the column required, q., to 70 cm for a final column height of 35 cm. Open the tap at the bottom of the column and allow the Sephadex to pack by settling. When the flow from the column drops to the rate of 1 ml mi+, close the tap, wash the sides of the extension column with 90 per cent. methanol and allow the Sephadex to settle by gravity until the supernatant liquid is com- pletely clear. After opening the tap, apply pressure to the top of the column extension by means of a blowball and gently compress the Sephadex until no noticeable further reduction in column height is observed.Adjustments to the height of the column can be made by adding or withdrawing slurry but at all times the Sephadex must be kept covered with liquid. Pass at least one bed volume of 2,2,4-trimethylpentane through the Sephadex, after which the column is ready for use. Remove the extension tube for sample addition and replace it for elution. Cool and weigh 2OOg into a clean, dry, screw-capped bottle, add 20ml of water and shake the mixture vigorously until it is thoroughly mixed and the powder is free flowing. Place the bottle on a roller mixer for 2 to 3 hours and allow the bottle to stand overnight. Meanwhile, prepare the eluting agent, a 1 + 4 mixture of diethyl ether and 2,2,4-trimethylpentaneJ add anhydrous sodium sulphate, shake the mixture and allow it to stand overnight in a stoppered flask. Fill a column, 40 x 1.5 cm, about half-full with 2,2,4-trimethylpentane.Weigh 30 g of the prepared Florisil into a 100-ml separating funnel that has been fitted with a cone and con- nected to the top of the column. Open the tap of the separating funnel and allow the Florisil to run smoothly into the solvent in the column, tapping the side of the column occasionally to ensure even packing of the Florisil and the release of any entrained air bubbles. As the packing proceeds it may be necessary to run off some of the 2,2,4-trimethylpentane from the column so as to prevent the liquid level from reaching the stem of the separating funnel. When all of the Florisil has been added, remove the separating funnel and add a 2-cm layer of anhydrous sodium sulphate to the top of the column.Lower the liquid level in the column until it just covers the solid material. The column is now ready for use. FZorisil column-Heat 210 to 220 g of Florisil for 3 hours at 160 "C. CALIBRATION OF COLUMNS- Determine the eluting characteristics of both the Sephadex and Florisil columns before use by transferring approximately 100 pg of vitamin D in the appropriate solvent to the top of each column and collecting fractions of the eluate. For the Sephadex column, collect 5-ml fractions after the appearance of the solvent front by using a flow-rate of approximately 30 to 40 ml h-1; for the Florisil column collect 20-ml fractions at a flow-rate of approximately 60 to 80 ml h-l.All the fractions are scanned on an ultraviolet spectrophotometer at 265 nm to locate the vitamin D peak and thus obtain the eluting position of vitamin D for each column. On Sephadex, vitamin D is normally found in the 20 to 50-ml fraction; on lilorisil, the 140 to 210-ml fraction usually contains the vitamin D. PROCEDURE Carry out all operations in artificial light.270 BELL AND CHRISTIE : GAS - LIQUID CHROMATOGRAPHIC [ANdySt, VOl. 98 SAPONIFICATION AND EXTRACTION- Weigh 20 g of potassium hydroxide pellets into a 250-ml twin-necked round-bottomed flask. Add about 10 ml of water followed by 140 ml of absolute ethanol, shake the mixture and warm it until the solid is completely dissolved. Prepare a solution of ergocalciferol in absolute ethanol, 1 ml of which is equivalent to 50 pg of the vitamin.Add 2 ml of this solution to the flask as an internal standard followed immediately by 5Og of cod-liver oil. Reflux the mixture for 15 minutes while bubbling a slow stream of nitrogen through the liquid, then add 25 ml of water and cool to room temperature as quickly as possible. Transfer the contents of the flask to a 1000-ml separating funnel, dilute to 600 ml with water and add immediately 400ml of a mixture of equal volumes of diethyl ether and light petroleum (boiling range 40 to 60 "C). Allow the phases to separate completely, approximately 10 minutes being required, and run the aqueous layer into a second 1000-ml separating funnel. Repeat the extraction with a further 200 ml of the solvent mixture, discarding the aqueous layer and adding the solvent layer to the first separat- ing funnel.Wash the combined solvent extracts twice by spraying with 100ml of water from a wash-bottle. Do not shake the separating funnel but when the phases have separated discard the aqueous layer. Add 250 ml of water to the separating funnel, shake it gently, allow the phases to separate and discard the washings. Repeat this procedure twice more, shaking the funnel vigorously on the last occasion. At this stage the layers should be clear and completely separated. If not, repeat with a further washing. Run the solvent layer into a large rotary evaporating flask, add 20 ml of absolute ethanol and proceed to remove the solvent under low pressure. At the first sign of cloudiness or precipitation add more absolute ethanol (25 to 50 ml) and evaporate to a volume of about 10 ml, then transfer the solution to a smaller flask and remove all of the solvent.Shake the mixture thoroughly for 2 minutes. PRECIPITATION OF STEROLS- Dissolve the residue in the flask with small amounts of warm methanol and transfer the liquid to a 10-ml graduated centrifuge tube. Cool it at 0 "C for 30 minutes, centrifuge it at 1000 to 2000 r.p.m. for 5 minutes and return the supernatant liquor to the evaporating flask. Remove the solvent, add a few millilitres of 2,2,4-trimethylpentane and evaporate again. Dissolve the residue in 2 ml of 2,2,4-trimethylpentane. SEPHADEX COLUMN CHROMATOGRAPHY- With a pipette, transfer the extract of vitamin D on to the prepared Sephadex column, rinsing the flask with 2 ml of 2,2,4-trimethylpentane, and elute with solvent at the rate of approximately 20 to 40 ml h-l, collecting the fraction that contains vitamin D.Add a few millilitres of ethanol and evaporate to dryness on a rotary evaporator. REMOVAL OF STEROLS AS DIGITONIDES- Dissolve the residue in methanol, transfer the solution to a 10-ml centrifuge tube and dilute it to 9 ml. Add 1 ml of water, shake the mixture vigorously, cool it to 0 "C and allow it to stand for at least 30 minutes. Centrifuge it and pour the supernatant liquor into a 30-ml centrifuge tube containing a digitonin solution prepared by dissolving 0.4 g of digitonin in 10 ml of 90 + 10 methanol - water mixture. Stopper the centrifuge tube, shake it and allow it to stand overnight in a refrigerator.Centrifuge and decant the supernatant liquor through a Whatman No. 541 filter-paper into a 50-ml separating funnel. Shake the precipitate of digitonides with 10 ml of the methanol - water mixture, centrifuge the mixture and use the supernatant liquor to wash the filter-paper. Extract the liquid in the separating funnel with two 15-ml portions of carbon tetrachloride, collecting the extract in an evaporating flask. Remove the solvent by rotary evaporation and repeat the evaporation after the addition of a few millilitres of 2,2,4-trimethylpentane. F FLORISIL COLUMN CHROMATOGRAPHY- the extract to the prepared Florisil column. 60 to 80 ml h-l and collect the fraction containing vitamin D. Dissolve the residue in 2 ml of 1 + 4 diethyl ether - 2,2,4-trimethylpentane and transfer Elute with the solvent mixture at a flow-rate ofApril, 19731 DETERMINATION OF VITAMIN D IN COD-LIVER OIL 271 PREPARATION OF THE ISOVITAMINS AND FORMATION OF THE TRIMETHYLSILYL ETHERS- Remove the solvent on a rotary evaporator and dissolve the residue in 1 ml of the 5a-cholestane solution.Add 4 ml of the antimony trichloride solution, shake the mixture and allow it to stand for exactly 1 minute, then add 6 ml of tartaric acid solution, shake the mixture vigorously and transfer it to a 25-ml separating funnel. Rinse the flask with 10 ml of light petroleum (boiling range 40 to 60 "C), add it to the separating funnel and shake the funnel for about 15 s. Discard the lower layer and wash the light petroleum layer three times with an equal volume of water.Filter the light petroleum through a Whatman No. 541 filter-paper into a small flask, wash the paper with a small volume of solvent and evaporate to dryness. Add 0.1 ml of bis(trimethylsily1)acetamide to the residue and allow it to stand for 10 to 15 minutes. The sample is now ready for injection on to the gas chromato- graph as described below. The operating conditions of the gas chromatograph were: column oven temperature, 235 "C ; detector oven temperature, 250 "C; injection block temperature, 300 "C; carrier gas, nitrogen at the rate of 50 ml min-l, hydrogen at the rate of 50 ml min-l and air at the rate of 500 ml min-I. The recorder chart speed was 2 min cm-l. CALIBRATION OF THE GAS CHROMATOGRAPH- Prepare a standard mixture containing 100 pg of ergocalciferol, 100 pg of cholecalciferol and 10 pg of 5a-cholestane, isomerise the vitamins and prepare teh trimethylsilyl ethers as previously described.Inject between 1 and 2 p1 of this solution on to the gas chromatograph and adjust the sensitivity of the instrument so that responses of the vitamin are between half and full-scale deflection on the recorder. Calculate the areas under the ergocalciferol and cholecalciferol peaks and the retention times of the vitamins relative to 5a-cholestane. The approximate retention times for these conditions are 21 minutes for 5a-cholestane and 46, 61 and 68 minutes for the trimethylsilyl ethers of cholesterol, cholecalciferol isovitamin and ergocalciferol isovitamin, respectively.Inject the sample in the same way, measure the areas under the ergocalciferol and cholecalciferol peaks and calculate the amount of cholecalciferol present from the known addition of ergocalciferol and the ratio of the areas obtained from equal masses of the two forms of the vitamin. CALCULATION- Cholecalciferol (vitamin D,)/pg 8-1 - Peak area of cholecalciferol x ergocalciferol added (pg) Peak area of ergocalciferol x R x mass of sample (g) where R is the ratio of the peak areas of cholecalciferol to ergocalciferol for equal masses of these substances. - RESULTS AND DISCUSSION Saponification losses of vitamin D are minimised by heating the oil with a fairly con- centrated ethanolic solution of potassium hydroxide for a short period and by bubbling nitro- gen gently through the liquid.Extraction of the unsaponifiable matter with a 1 + 1 mixture of diethyl ether and light petroleum gives a cleaner separation than with diethyl ether alone. Most procedures for the removal of interfering substances involve the use of one or more adsorption columns of alumina, magnesium oxide or silicic acid. Celite impregnated with poly(ethy1ene glycol)3 and, more recently, Fluoropak 804 have also been used. Other systems rely on the conversion of retinol (vitamin A) into a derivative that is more readily separated from vitamin D than the parent compound.s~s In the proposed method, vitamin D is well separated from retinol and some other interfering substances on a partition column of Sepha- dex LH-20.7 This technique has the advantage that such columns are easy to handle and can.be readily regenerated. Separation of vitamin D from cholesterol is, however, incomplete and the residual sterol is removed by precipitation with digitonin before a final treatment on a column of Florisil to remove anhydrovitamin A and other unknown interfering compounds. By this means cholesterol present in cod-liver oil can be reduced to manageable proportions but, as shown in the chromatogram of a typical cod-liver oil in Fig. 1, a small amount still remains.272 BELL AND CHRISTIE: GAS - LIQUID CHROMATOGRAPHIC [Analyst, Vol. 98 I Fig. 1. Gas - liquid chromatogram of chole- calciferol in cod-liver oil: peak 1, added 5or-chole- stane ; peak 2, cholesterol trimethylsilyl ether; peak 3, cholecalciferol isovitamin trimethylsilyl ether; and peak 4, added ergocalciferol isovit- amin trimethylsilyl ether In our early work, a sensitivity of 1 pg of vitamin D was obtained on the gas chromato- graph when using glass columns containing Chromosorb G coated with 2-5 per cent.silicone elastomer 30. Later, increased sensitivity and superior separations were obtained with column supports and stationary phases specially designed for steroid analysis, such as 3 per cent. OV-17 on 100 to 120-mesh Gas-Chrom Q. Under these conditions, less than 0.1 pg of vitamin D can easily be determined by using a flame-ionisation detector. TABLE I RESULTS OF GAS - LIQUID CHROMATOGRAPHIC DETERMINATION OF CHOLECALCIFEROL I N REFINED COD-LIVER OILS Sample designation A B C D E F G 13 I J Cholecalciferol/pgg-l 2.8, 2.5, 2-9, 2.7, 3.5, 3.3, 3.3, 3.9, 3.7, 3.4, 3.2 2.8 3.5 3.3 3.5 3-5 3.5 3.9 3.5 3-8 Tables I, I1 and I11 show that the proposed method gives repeatable results that are in reasonable agreement with those given by the standard bioassay method.The procedure, which can be completed in less than 2 days, is suitable for the quality control of the vitamin D content of cod-liver oil. TABLE I1 CHOLECALCIFEROL CONTENT OF SOME COD-LIVER OILS : COMPARISON OF RESULTS BY THE BIOLOGICAL RAT ASSAY AND GAS - LIQUID CHROMATOGRAPHIC METHODS Cholecalciferol content f A > Mean diiTerence, -7 per cent. 10.2 i!i } +6 10-2 410 Fiducial Bioassay Gas - liquid chromatography range/ 1 Type of oil i.u. g-1 t% 8-l i.u. g-l High potency 309 to 466 9.6 382 9.8 Low potency 40 to 68 1.3 52 1.2 49 -6 s’o” ) 0 Veterinary * 56 to 99 1.9 76 1.8 62 to 95 1.9 77 2.0 64 to 93 2.0 78 2.1 83 66 to 102 2.1 83 - - Medicinal 72 to 100 2.1 85 2.3 92 +s *Prepared by adding to the low-potency oil 0-75 pg (30 i.u.) of cholecalciferol per gram.April, 19731 DETERMINATION OF VITAMIN D I N COD-LIVER OIL TABLE IT1 CHOLECALCIFEROL CONTENT OF REFINED COD-LIVER OILS : COMPARISON OF RESULTS BY THE BIOLOGICAL RAT ASSAY AND GAS - LIQUID CHROMATOGRAPHIC METHODS 273 Code number 91684 92485 94787 94787 93687 00588 00788 00788 01889 01889 Fiducial range/ i.u.per fl. oz. 2229 to 3170 3232 to 4322 2377 to 5250 2377 to 5250 2377 to 5250 2212 to 4931 2212 to 4931 2212 to 4931 2207 to 3394 2207 to 3394 Cholecalciferol content A \ Bioassay Gas - liquid chromatography w pg g-1 i.u. per A. oz. 2.6 2733 2.9 3067 3.6 3758 3.0 3090 3.4 3548 3.4 3623 3-4 3548 3.4 3500 3.4 3548 3.6 3772 3.5 3626 3.5 3681 3.5 3626 3.6 3726 3.5 3626 3.3 3480 2.6 2769 3.1 3249 2.6 2769 2.9 2999 Difference, per cent. $- 12 - 18 -1 -1 +6 i - 2 +3 -4 + 17 +9 Our thanks are due to the Marfleet Refining Company for the provision of samples of cod-liver oil and for determining their vitamin D content by the bioassay procedure of the British Pharmacopoeia. This paper is published with the permission of the Government Chemist. REFERENCES 1. 2. 3. 4. 5. 6. 7. Ziffer, H., Vanden Heuvel, W. J. A., Hashti, E. 0. A., and Horning, E. C., J. Amer. Chem. SOG., Murray, T. K., Day, K. C., and Kodicek, E., Biochem. J., 1966, 98, 203. Theivagt, J. G., and Campbell, D. J., Analyt. Chew., 1959, 31, 1375. Chen, P. S., Terepka, A. R., and Remsen, N., Ibid., 1963, 35, 2030. Barua, R. K., and Rao, M. V. K., Analyst, 1964, 89, 534. Said, F., Salah, M. K.. and Girgis, P., Ibid., 1966, 91, 459. Bell, J . G., Chem. G. Ind., 1971, 201. 1960, 82, 6411. Received November 7th, 1972 Accepted December 7th, 1972

 

点击下载:  PDF (587KB)



返 回